Policy threshold for decelerating chronic disease accumulation among older Chinese adults: Evidence from the Healthy China 2030 Initiative

Policy threshold for decelerating chronic disease accumulation among older Chinese adults: Evidence from the Healthy China 2030 Initiative | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Policy threshold for decelerating chronic disease accumulation among older Chinese adults: Evidence from the Healthy China 2030 Initiative Hin-Wa Wong This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8189460/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Health policy evaluations frequently overlook delayed or cumulative effects due to inadequate observation periods, leading to the premature rejection of beneficial interventions. The optimal duration of policy exposure necessary to achieve measurable health impacts is not well established. This study examines the time-dependent effects of China's Healthy China 2030 (HC2030) initiative on chronic disease accumulation among older adults, utilizing staggered provincial implementation as a natural experiment. Methods: Longitudinal data from 7,487 adults aged 60 years and older in the China Health and Retirement Longitudinal Study (CHARLS, 2011–2018) across 28 provinces were analysed. Staggered HC2030 implementation timing (0–21 month variation) was leveraged within a difference-in-differences framework. Provinces were grouped by exposure duration (short-, medium-, or long-term), and group-specific disease progression trajectories were estimated for each group. Overlap weighting, baseline covariate balancing, and wild cluster bootstrapping were employed to ensure robust inference with a limited number of clusters. Results: Prepolicy placebo tests confirmed the validity of the parallel trends assumption. The policy effects differed by exposure duration: short-term (≤15 months) and medium-term (15–18 months) exposures did not yield significant impacts, whereas divergence in disease progression trajectories emerged at approximately 13 months, indicating stabilization. Long-term exposure (≥19 months) significantly decelerated chronic disease accumulation (β = -0.325, 95% , p = 0.005, WCB), offsetting a substantial portion of the age-related disease burden (population mean increase: 0.45). These results indicate that sustained policy implementation is necessary to achieve meaningful health system benefits. Conclusion: Comprehensive health system reforms require extended implementation periods, with approximately 13 months needed to stabilize disease progression and sustained exposure (≥19 months) necessary to achieve significant cumulative effects (β = -0.325, p = 0.005) before measurable population health benefits are observed. Premature evaluations risk dismissing effective interventions. These threshold estimates provide empirical benchmarks for designing evaluation timelines in chronic disease policy research, particularly in low- and middle-income settings. Health Policy Evaluation Chronic Disease Management Difference-in-Differences Policy Implementation Timing Elderly Health China Health Reform Natural Experiments Figures Figure 1 1. Background 1.1 Critical impact of evaluation timing on health policy assessment The temporal dynamics between policy implementation and health outcomes critically shape the evaluation of health system reform. Complex interventions demonstrate delayed effects that emerge only after a sufficient exposure duration [ 1 – 3 ]. Insufficient observation periods risk premature termination of beneficial interventions [ 4 ], yet systematic evidence on optimal evaluation timing remains scarce. Preventive health interventions often exhibit substantial implementation lags, with full effects emerging only after years or decades [ 5 – 8 ]. Chronic disease management programs face particularly long latency periods due to the slow biological processes of disease progression and the time required to establish continuous care relationships [ 2 , 9 ]. However, empirical evidence quantifying these temporal thresholds remains limited. Despite the growing recognition of the importance of long-term evaluation [ 10 , 11 ], few studies have systematically identified critical time thresholds for policy effects. Finland's North Karelia Project required 5 years before measurable cardiovascular benefits were demonstrated [ 12 ], whereas Japan's Health Japan 21 achieved only 17% of predetermined targets after 13 years [ 13 , 14 ]. These cases illustrate the long-term nature of population health interventions but provide limited guidance for evaluation design in other contexts. 1.2 Elderly Chronic Disease Management as an Indicator of Health Service Effectiveness Elderly chronic disease management is a key indicator for assessing health service effectiveness. Chronic diseases are highly prevalent among older adults and are primarily attributable to modifiable risk factors; therefore, policy responses that prioritize long-term outcomes are essential [ 15 , 16 ]. Multimorbidity is common in this population and is closely linked to functional impairment, reduced quality of life, and increased healthcare utilization [ 17 , 18 ]. Effective management of chronic diseases in older adults requires a continuous service chain encompassing prevention, diagnosis, treatment, and rehabilitation [ 19 , 20 ]. The accessibility and continuity of these services are central to the core functions of primary healthcare systems. Service quality influences disease control, complication prevention, and functional independence. Empirical evidence indicates that high-quality continuity of care can reduce hospitalization rates among elderly diabetes patients by approximately 15 percentage points (from 68.2% to 53.5%) and nearly halve mortality (from 18.5% to 8.6%) [ 21 ]. Integrated medical and social care policies also significantly reduce functional dependency risks and care gaps among older populations [ 22 ]. Chronic disease management in older adults is a sensitive indicator for evaluating the effectiveness of "health-centered" integrated service systems [ 23 ]. From a systems science perspective, integrated care constitutes a complex adaptive system whose effectiveness relies on coordination and integration across the clinical, functional, organizational, and systemic levels [ 24 , 25 ]. Disease progression rates reflect the combined influence of medical accessibility (early detection) and service quality (effective management), capturing the cumulative effects of improvements in insurance coverage, provider capacity, and health education programs. 1.3 Healthy China 2030: A Natural Experiment in Health System Reform China's "Healthy China 2030" (HC2030) initiative was launched in October 2016 [ 26 ], providing a unique context for exploring the long-term effects of healthcare service reform [ 27 ]. As China's national-level medium- to long-term health strategy, this policy framework established a macrolevel shift from a "disease-centered" approach to a "health-centered" approach. Provinces successively formulated and launched localized action plans between 2016 and 2018, resulting in variation in the duration of policy exposure (0–21 months) by the time of the 2018 CHARLS survey. This study assumes that these temporal differences are driven primarily by administrative processes and policy diffusion mechanisms rather than by direct responses to local health needs. This assumption is based on typical patterns of policy implementation in China: the diffusion of central policies to the provincial level typically follows experimentation within hierarchical mechanisms [ 28 , 29 ] rather than selective adoption on the basis of local demand. Although HC2030 was released nationwide as a unified strategic framework in 2016 [ 26 , 27 ], the timing of provinces' formulation of specific action plans and supporting policies may reflect this administration-driven diffusion pattern, thereby providing a source of variation in policy exposure duration for quasiexperimental design. HC2030 transformed the healthcare service delivery landscape through a multifaceted approach. This includes expanding insurance coverage to enhance financial accessibility, alongside strengthening primary healthcare infrastructure with standardized management protocols. Complementing these structural changes are efforts to increase medical workforce training—particularly in rural areas—and the development of integrated health information systems to improve care coordination. [ 30 ]. Although preliminary assessments suggest short-term or regional improvements [ 31 ], evidence of medium- to long-term effects on chronic diseases in older adults remains limited. 1.4 Research methods and contributions Traditional binary exposure models, which compare outcomes before and after policy implementation, are unable to capture variations in policy exposure intensity or assess dose-dependent effects. This study utilized a grouped difference-in-differences (DiD) design with dose-specific time trends to compare chronic disease trajectories across three exposure groups: short-term, medium-term, and long-term. This approach tests whether increasing policy exposure duration influences chronic disease accumulation rates by incorporating interaction terms between exposure duration and group membership, estimating group-specific disease progression slopes, and detecting dose-dependent treatment effects. The design is informed by established methods for estimating treatment effect heterogeneity by intensity [ 32 , 33 ] and recent advances in continuous treatment difference-in-differences methodologies [ 34 ]. We focus on three specific research questions: (1) Is extended HC2030 exposure associated with improved chronic disease management outcomes via enhanced healthcare service accessibility? (2) At what time threshold do health system reforms begin to yield measurable health impacts? (3) Do effects differ by baseline health status and sociodemographic characteristics? By identifying critical time points at which policy effects emerge and stabilize, this study offers empirical guidance for designing observation periods in health policy evaluation, particularly in low- and middle-income countries, and advances methodologies in health services research on reform evaluation. 2. Methods 2.1 Data Sources and Sample Selection This study utilized Harmonized CHARLS Version D data, including four waves of nationally representative surveys collected in 2011, 2013, 2015, and 2018. CHARLS employs a multistage, stratified, probability‒proportional‒to-size (PPS) sampling design, ensuring national representativeness of Chinese adults aged 45 years and above [ 35 ]. This data infrastructure enables international comparisons through harmonized variable definitions, making it comparable with international counterpart surveys such as the U.S. Health and Retirement Study (HRS) and the English Longitudinal Study of Aging (ELSA). The main analytical sample includes respondents aged 60 and above in 2015 who provided complete, valid data in both the 2015 and 2018 survey waves. The 2020 survey wave was excluded because COVID-19-related healthcare service disruptions, changes in care-seeking behavior, and interruptions in chronic disease management would introduce systematic biases inconsistent with prepandemic observation periods. The disruptive impact of the pandemic on health data has been well documented in the WHO's multicountry pulse surveys [ 36 ]. All the statistical analyses were conducted via Stata/MP 18.0. The analysis code and detailed data processing scripts are available from the corresponding author upon reasonable request to promote reproducibility and transparency. 2.2 Variable Definitions 2.2.1 Primary outcome variable Chronic disease accumulation was measured as the change in disease count (ΔChronic) between 2015 and 2018. The CHARLS assesses 13 physician-diagnosed chronic conditions, including hypertension, diabetes, and cardiovascular disease (see the CHARLS Harmonized D manual for the full list). Positive values indicate disease accumulation, negative values indicate improvement, and zero values indicate stability. Self-reported disease counts were used as the primary outcome for three reasons: (1) the 2018 CHARLS wave lacked biomarker measurements, precluding longitudinal biomarker analysis; (2) self-reported diagnoses were consistently collected across waves, enabling robust temporal comparisons; and (3) validation studies demonstrated acceptable concordance between self-reports and medical records in similar populations [ 37 ]. Although this approach may conflate true disease incidence with improved detection resulting from increased healthcare access, the 'detection effect' is considered an important early signal of successful policy implementation rather than a limitation. 2.2.2 Secondary outcome variable Changes in activities of daily living (ΔADL) scores were designated secondary outcomes to assess the impact of policy exposure duration on functional status among older adults. This selection is justified primarily by the well-established correlation between chronic disease burden and functional impairment [ 17 ], suggesting that improved disease management may delay functional decline. Furthermore, ADL serves as a validated proxy for quality of life and independence, representing a critical dimension for evaluating the comprehensive benefits of health policies [ 38 ]. 2.2.3 Policy Exposure Measurement The policy exposure duration was operationalized as the cumulative number of months from provincial HC2030 implementation to the 2018 CHARLS follow-up, ranging from 0–21 months across provinces. Provincial implementation dates were defined as the dates when provincial governments officially released "Healthy China 2030" implementation plans or equivalent documents (see the complete provincial policy implementation timeline in Appendix I Table S7). HC2030 implementation enhanced healthcare service accessibility and quality through the following mechanisms: (1) financial accessibility, i.e., expanding insurance coverage and increasing reimbursement rates for outpatient and inpatient services; (2) service capacity, i.e., strengthening primary healthcare infrastructure and establishing standardized chronic disease management protocols; (3) human resources, i.e., increasing medical workforce training and deployment, particularly in underserved rural areas; and (4) system integration, i.e., developing integrated health information systems to improve care coordination and referral networks. Provincial variation in implementation timing (0–21 months) provided natural variation in exposure to these service delivery improvements, enabling identification of when health system strengthening translates into measurable health outcomes. To examine the nonlinear dynamic characteristics of the association between policy exposure duration and chronic disease progression, continuous policy exposure duration (tml) was divided into three mutually exclusive groups on the basis of quartiles: the short-term exposure group (≤ 25th percentile, approximately 0–15 months), the medium-term exposure group (25–75th percentile, approximately 15–18 months), and the long-term exposure group (> 75th percentile, approximately ≥ 19 months). As a single linear trend or simple intergroup mean comparison might mask heterogeneity in disease progression rates within different exposure stages, a piecewise linear regression strategy was employed. Specifically, interaction terms between group dummy variables and continuous time variables (Group × Time) were introduced into the model. This design allows for simultaneous estimation of differences in baseline levels across exposure stages (intercept effects) and differences in rates of change over time (slope effects). Significant interaction terms indicate that chronic disease accumulation trajectories undergo structural changes as the duration of policy exposure increases, thereby validating the marginal benefits of the policy at different stages. This grouping strategy maintains adequate sample sizes for statistical power in each group while enabling the detection of potential nonlinear associations [ 39 , 40 ]. Notably, this grouping design employs a grouped-comparison analytical strategy rather than estimating continuous dose‒response relationships. 2.2.4 Covariates Covariate selection followed the framework of Andersen's behavioral model of health services use [ 41 ], which includes the following: (1) predisposing factors: individual-level demographic characteristics (age, sex, marital status); (2) enabling factors: socioeconomic status (education level, household income or assets), health insurance coverage; (3) need factors: health behaviors (smoking, alcohol consumption), functional and mental health indicators (ADL scores, depression scale scores), baseline chronic disease burden and health service utilization (hospitalization frequency, outpatient visits); and (4) contextual factors: provincial time-varying macroeconomic and health resource indicators extracted from the China Statistical Yearbook, including per capita regional gross domestic product (GDP) and health resource allocation indicators. The physical activity variable exhibited a relatively high proportion of missing values (approximately 50%) in the analytical sample, and missing pattern analysis suggested possible systematic differences. Because including variables with high missing rates can substantially reduce the sample size and introduce selection bias, physical activity control variables were excluded from the main regression model. 2.3 Statistical analysis 2.3.1 Primary Strategy for Evaluating the Timing of Health Service Delivery: Grouped Heterogeneous Treatment Effects with a Difference-in-Differences Design Methodological Positioning The analytical approach leverages variation in HC2030 implementation timing across provinces (0–21 months) to identify policy effects. Rather than treating all provinces identically, provinces are grouped by exposure duration, and the divergence in disease progression trajectories between groups is estimated over time. This strategy tests whether extended policy exposure yields stronger health benefits, which is consistent with a dose–response relationship. Technical approach: We employ a grouped difference-in-differences design combining first-differencing with group-specific time trends, which helps to eliminate time-invariant confounding variables and to detect dose-dependent effects. This builds on recent methods for continuous treatment DiD [ 34 , 42 – 44 ], adapted for discrete exposure groups to improve interpretability and statistical power. While this study draws on core insights from continuous treatment difference-in-differences methods—using variation in treatment intensity to identify the temporal dynamics of policy effects [ 34 ]—the analysis does not estimate a fully continuous dose–response function. Instead, treatment effect heterogeneity is tested by comparing discrete exposure groups (short-term, medium-term, and long-term). This design choice is justified by several considerations: (1) flexibility, as the grouping approach allows policy effects to vary nonparametrically across exposure durations without assuming linear dose‒response relationships; (2) interpretability, since discrete group comparisons are more accessible to policymakers for identifying critical time thresholds; (3) statistical power, as grouped comparisons are more robust with limited sample sizes; and (4) theoretical fit, given that policy effects may exhibit activation thresholds rather than smooth linear changes. Core Research Questions and Model Specification From a health services research perspective, this approach addresses the following key questions: (1) Is extended HC2030 exposure associated with improved chronic disease management outcomes via enhanced healthcare service accessibility? (2) At what time threshold do health system reforms begin to yield measurable health impacts? (3) Do these effects differ by baseline health status and sociodemographic characteristics? The equation for the main analytical model is as follows: where ΔYit represents the cumulative change in chronic disease count for individual i between two follow-ups; TimeMonthsit is the policy exposure duration in months (continuous variable, 0–21 months); TML_Mediumit is a dummy variable for the medium-term exposure group (1 = medium-term exposure, 0 = otherwise); TML_Longit is a dummy variable for the long-term exposure group (1 = long-term exposure, 0 = otherwise); the short-term exposure group serves as the reference category (baseline group); interaction terms estimate the time-dependent differential effects of the medium-term and long-term exposure groups relative to the short-term exposure group; Xitγ is a covariate vector; and εit is the random error term (Table 1 ). Table 1 Key parameters and interpretation. Parameter Meaning Epidemiological Interpretation β₁ Baseline slope Disease accumulation rate in short-term exposure group (reference) β₂M, β₂L Level differences Baseline differences between groups (absorbed by differencing) β₃M, β₃L Slope differences (DiD estimator) Change in accumulation rate for medium/long-term groups relative to baseline Note: Critical insight: β₃L < 0 and statistical significance indicate that long-term exposure significantly decelerates disease accumulation compared with short-term exposure—our core hypothesis. Parallel trends assumption: In the absence of differential policy effects, all groups would follow similar disease trajectories (β₃M = β₃L = 0). Significant negative β₃ coefficients indicate policy-induced divergence. The epidemiological interpretation of model parameters is as follows: β₀ (Intercept): represents the average natural change in chronic diseases at a theoretical zero-exposure time point, or baseline level, for the sample population (background trend due to natural aging). β₁ (baseline slope): represents the dose‒response relationship for the short-term exposure group (reference group). That is, during the short-term exposure stage, there is a marginal increase in chronic disease accumulation per additional month of policy exposure. This constitutes the 'counterfactual' baseline trend for our comparisons. β₂ (Intercept Difference): represents systematic differences in regression line intercepts between medium/long-term exposure groups relative to the short-term exposure group (typically used to absorb inherent level differences between groups). β₃ (Slope Difference/DiD Estimator): This is the core focus of this study. It represents the degree of deviation in the chronic disease accumulation rate (Slope) for the medium/long-term exposure groups relative to the short-term exposure group. If β₃ < 0 and is statistically significant, this indicates that as the duration of policy exposure increases, the cumulative speed of chronic disease is significantly lower than the baseline speed of the short-term group, confirming the marginal protective effect of long-term policy exposure. Parallel Trends Assumption and Testing Under this first-difference framework, the parallel trends assumption transforms into the "parallel growth assumption": we assume that in the absence of causal effects from differential policy exposure duration, the chronic disease accumulation rates (slopes) of the medium/long-term exposure groups would have remained consistent with those of the short-term exposure group (i.e., β₃ should equal 0). Any significant deviation of the β₃ coefficient from zero is attributed to the causal effect of the duration of policy exposure. These assumptions are validated through the following strategies: (1) prepolicy period placebo test (2011–2015), which uses the same grouping strategy and model specification to test whether future exposure group classification is associated with health trajectory differences before policy implementation; (2) covariate balance testing, which involves examining balance across three exposure groups on baseline covariates after overlap weighting; and (3) robustness tests, which include randomized pseudoexposure testing and placebo outcome testing. 2.3.2 Overlap weighting for balancing healthcare service access To construct comparable reference populations across different policy exposure duration groups, we employed the overlap weighting (OW) method for multinomial treatments [ 45 ]. This method reduces selection bias arising from uneven covariate distributions by assigning weights to observations with propensity scores within the common support region, which is particularly important for comparing provinces with different baseline levels of healthcare service accessibility. For three-category treatment (short-term/medium-term/long-term exposure groups), we first estimate generalized propensity scores (GPSs) via multinomial logistic regression: P(TML = k|X) = exp(X'γk)/Σⱼ exp(X'γⱼ), k ∈ {short-term, medium-term, long-term}. where X includes all baseline covariates. On the basis of GPS, overlap weights are calculated as follows: for individual i in exposure group k, w_ik = Π_{j ≠ k} e_ij, which is the product of generalized propensity scores for all groups except k. This weighting scheme has the following advantages: (1) it maximizes the effective sample size: compared with inverse probability weighting (IPW), overlap weights are insensitive to extreme propensity scores; (2) it ensures covariate balance: in weighted pseudopopulation, the three exposure groups have identical covariate distributions; and (3) it focuses on the overlap region: the estimated effects correspond to subpopulations where all three groups are adequately represented. Standardized mean differences (SMDs) are used to assess balance, with SMDs < 0.10 indicating good balance and SMDs < 0.05 indicating excellent balance. 2.3.3 Missing data handling and statistical inference To handle missing data, we used multiple imputation by chained equations (MICE), generating 50 imputed datasets [ 46 ]. The imputation model included all the analytical variables and relevant auxiliary variables to approximate the missing-at-random assumption. The results were combined via Rubin's rules [ 47 ]. Given the limited number of provincial clusters (n = 28), traditional cluster-robust standard errors may underestimate sampling variability and increase Type I error rates. Therefore, we combined wild cluster bootstrapping (WCB) with multiple imputation for statistical inference [ 48 , 49 ]. Specifically, 10,000 WCB repetitions are performed on each imputed dataset via Rademacher weights; point estimates and variances across imputed datasets are combined via Rubin's rules; and 95% confidence intervals and p values are constructed. The simulation results show that WCB methods can still provide reliable inference when the number of clusters is small (e.g., G between 15–20) under certain conditions [ 49 ]. 2.3.4 Identifying Critical Time Points for Healthcare Service Effects To identify critical time points at which healthcare service improvements translate into health outcomes. On the basis of the marginal prediction curves, we calculate the following: Equivalence crossover point The time when long-term and short-term groups reach equal predicted outcomes, calculated as t = -β₂L/β₃L. This represents the point at which health benefits begin to offset initial differences. Zero-Accumulation Point The time when the long-term group's predicted outcome crosses zero, calculated as t = -(β₀+β₂L)/(β₁+β₃L). This marks the transition from disease accumulation to population-level stabilization. These calculations use point estimates from the regression model. We focus on trajectory-level inference (testing whether group slopes differ) rather than precise threshold estimation, as our primary interest is in demonstrating that sustained health system exposure produces measurable benefits within a policy-relevant timeframe. 2.3.5 Robustness and placebo tests To assess the reliability of causal identification [ 50 ], we conducted sensitivity analyses across four dimensions: (1) temporal specificity: prepolicy placebo test (2011–2015), designating 2013 as the "virtual policy start year," testing whether interaction terms between virtual exposure groups and time months are statistically significant in the prepolicy period; (2) treatment specificity: randomized pseudoexposure test, using random number generators to generate provincial random exposure durations (0–21 months), testing whether random grouping produces significant effects; (3) outcome specificity: placebo outcome test, using measured height as a negative control, testing whether policy exposure is associated with height changes; and (4) heterogeneity testing: including three-way interaction terms of the policy exposure group with age, sex, urban‒rural residence, and education level. 3. Results 3.1 Sample characteristics The CHARLS 2015 baseline survey included 25,586 respondents. After applying the inclusion criterion of age 60 and above, 8,337 eligible individuals with a mean age of 67.98 years (SD = 6.67) were included; 48.4% were male, and 90.3% had basic medical insurance. The sample reflects typical characteristics of China's elderly population: 79.5% have a primary education or lower, and 77.5% reside in rural areas. The baseline mean chronic disease count was 2.17 (SD = 1.69); by the 2018 follow-up, chronic diseases had increased on average by 0.45 (SD = 1.29). A total of 17.3% of the respondents reported that the incidence of chronic disease decreased, 40.8% remained unchanged, and 41.9% reported that it increased. The mean policy exposure duration was 16.09 months (SD = 3.73) (Table 2 , Appendix I Table S1 ). After overlap weighting adjustment, the baseline characteristics across the exposure groups achieved good balance, with all SMDs < 0.05 (Table S2). Owing to missing weights in some samples, the final sample size used in the statistical models was 7,487 individuals (Appendix II, Figure S1 ). Table 2 Descriptive Statistics of the Key Variables (N = 8337). Variable N Mean/n (%) SD Missing n(%) Outcome & Core Policy Variables Δ Chronic conditions count 7922 0.45 1.29 415 (5.0%) Time in months (TML) 8337 16.09 3.73 0 (0.0%) TML: Short Duration (≤ 25th pctile) 8337 2211 (26.5%) - - TML: Medium Duration 8337 4972 (59.6%) - - TML: Long Duration (> 75th pctile) 8337 1154 (13.8%) - - Demographic Variables Age (years) 8337 67.98 6.67 0 (0.0%) Male, n (%) 8337 4032 (48.4%) - 0 (0.0%) Low education, n (%) 8337 6629 (79.5%) - 0 (0.0%) Couple Household, n (%) 8337 6552 (78.6%) - 0 (0.0%) Rural residence, n (%) 7700 5967 (77.5%) - 637 (7.6%) Insured, n (%) 7891 7126 (90.3%) - 446 (5.3%) Health and Lifestyle Variables Baseline chronic conditions count 7923 2.17 1.69 414 (5.0%) ADL count 8301 0.54 1.15 36 (0.4%) Hospitalization times (past year) 8290 0.25 0.70 47 (0.6%) Out-of-pocket hospitalization cost In USD (past year) 8147 137.41 755.22 190 (2.3%) Smoking frequency per day 8230 4.14 8.91 107 (1.3%) Ever Drinking last year, n (%) 8300 2671(32.18%) - 61 (0.7%) Note: All the statistics are based on original data before multiple imputation. For continuous variables, the means ± SDs are reported. For binary categorical variables, n (%) of the positive/reference category is reported. See Appendix I Table S1 for the complete table. 3.2 Main analysis and robustness checks 3.2.1 Primary Estimates of Health System Strengthening Effects Main findings: Policy effects depended critically on exposure duration (Table 3 , Appendix I Table S3). Compared with short-term exposure, long-term exposure (≥ 19 months) significantly slowed chronic disease accumulation (β = -0.325, SE = 0.116, p = 0.005, 95% CI: -0.552–0.098). Clinical significance: This effect is substantial. The average population experienced 0.45 additional chronic conditions over the observation period due to natural aging. The estimated policy effect offsets approximately 72% of this age-related burden among individuals with long-term exposure (0.325/0.45 = 0.72), moving from net accumulation toward stabilization. In contrast, neither the main effect of policy exposure duration (β = 0.009, p = 0.879) nor the interaction term for the medium-term exposure group (β=-0.016, p = 0.636) achieved statistical significance. These findings indicate that healthcare service delivery improvements under HC2030 require a longer implementation period (≥ 19 months) to manifest in population-level chronic disease outcomes. Table 3 Difference-in-differences estimates of policy exposure duration on chronic disease accumulation: Group-specific effects with wild cluster bootstrap inference (N = 7,487). Variables (1) Pooled MI (SE) (2) WCB Corrected (SE) Dose Interaction TimeMonths (Continuous, Ref: Low Dose) 0.009 (0.008) 0.009 (0.068) TML: Medium ×TimeMonths -0.016(0.016) -0.008 (0.034) TML: Long ×TimeMonths −0.325*** (0.046) −0.342*** (0.078) Policy Level (DiD Intercept) TML: Long (Ref: Low) 1.323*** (0.226) 1.392*** (0.483) Key Health/SES Controls Baseline chronic conditions count −0.205(0.153)*** −0.213(0.019)*** Rural residence (1 = Rural) −0.230(0.522)*** −0.253(0.066)*** Disability Score PCA 0.096(0.184)*** 0.094(0.019)*** Mental Health Score PCA 0.127(0.021)*** 0.133(0.215)*** Hospitalization times (past year) 0.176(0.028)*** 0.187(0.034)*** Fixed Effects & Inference Individual Fixed Effects Yes Yes Time Fixed Effects Yes Yes Clusters (Provinces) 28 28 VCE/Inference Method Clustered SE Wild Cluster Bootstrap N (Obs) 7487 7487 Note: * * p < 0.01; ** p < 0.05; * p < 0.1. The standard error (STD. err.) is in brackets. See Appendix I Table S3 for the estimated results of the complete model, including the coefficients of all individual characteristics, provincial characteristics, health behavior, and economic variables. Note The intersection of trajectories (approx. 4–13 months) illustrates the critical transition from the initial 'detection effect' (increased diagnosis) to the 'management effect' (decelerated progression). The long-term group (> 75th percentile) demonstrated a clear downwards trajectory after this stabilization period. 3.2.2 Validating the parallel trends assumption: Prepolicy placebo test The placebo test applied the main analytical model to the prepolicy period (2011–2015), using actual future policy exposure duration (measured 2016–2018) as the treatment variable (Table 4 , Appendix I Table S4). The results revealed no significant interaction effects during the placebo period. The critical interaction term for long-term exposure was small and nonsignificant (β = 0.089, p = 0.474, WCB), in contrast with the significant negative effect observed in the main analysis (β=-0.325, p = 0.005). Similarly, the medium-term exposure interaction was not significant (β = 0.010, p = 0.671). These nonsignificant placebo interactions validate the parallel trends assumption, indicating that exposure groups followed similar trajectories before policy initiation. Notably, there was a significant main time effect during the placebo period (β = -0.012, p = 0.033), indicating that provinces that would later have longer exposure periods experienced slightly greater chronic disease deceleration from 2011–2015. This main effect does not violate the parallel trends assumption, which requires parallel group-specific trajectories rather than zero baseline trends. Moreover, this effect reversed during the main analysis period (β = 0.009, p = 0.879), suggesting period-specific secular trends rather than persistent selection bias. Table 4 Comparison of the estimated longitudinal cumulative association coefficients between the main analysis and the placebo period. Indicator Main (2015→2018) Placebo (2011→2015) Difference (M-P) TML Long × TimeMonths_c WCB coef. -0.325 0.089 -0.414 WCB SE 0.116 0.124 — WCB p value 0.005 0.474 — Pooled cofe. -0.325 0.089 -0.414 Pooled p value 0.000 0.254 — Baseline Chronic diseases Count WCB coef. -0.205 -0.003 -0.202 WCB p value < 0.001 0.831 — Hospitalizations WCB coef. 0.176 0.135 0.041 WCB p value < 0.001 0.005 — Note: Post-2016 changes in TML long interactive items and baseline chronic disease effects indicate structural shifts in health management following policy implementation. The prepolicy trends show no significant differences, whereas the postpolicy differences support the parallel trend assumption. 3.2.3 Treatment specificity: Randomized pseudoexposure test To assess whether the significant main time effect during the placebo period reflects spurious correlation with any continuous provincial variable, we conducted a robustness check via randomized pseudoexposure (Table 5 ). Critically, during the main analysis period, randomized pseudoexposure produced a small positive interaction for the long-term exposure group (β = 0.034, p = 0.007), which was in the opposite direction and approximately 10 times smaller in magnitude than the true policy effect (β=-0.325, p = 0.005). The sharp contrast in direction and magnitude provides strong evidence that the observed policy effects reflect true causal impacts rather than spurious correlations with preexisting provincial trends. 3.2.4 Outcome specificity: Placebo outcome test To verify that the observed effects reflect true policy impacts on chronic disease outcomes rather than unmeasured time-varying provincial confounders, we conducted a placebo outcome test using measured height as a negative control. The height of adults aged 60 years and above should not be affected by medical policies for chronic disease management. The results revealed no significant association between policy exposure and changes in height (Appendix I Table S5). The long-term exposure interaction term was small and nonsignificant (β = -0.0083, p = 0.203, WCB), approximately 40 times smaller than the effect on chronic disease count (β = -0.325, p = 0.005, WCB). Table 5 Summary of Robustness Checks: Triple Validation of Causal Inference Validation Dimension Test Design Exposure Variable Key Interaction p value Interpretation (β₃) A. Main Analysis Primary outcome (chronic disease) Main period (2015–2018) True TimeMonths -0.325 0.005 Strong negative effect B. Temporal Specificity Prepolicy placebo Placebo period (2011–2015) True TimeMonths 0.089 0.474 No prepolicy differential trends C. Treatment Specificity Randomized pseudoexposure (placebo) Placebo period (2011–2015) Random TM 0.008 0.422 No spurious pretrends Randomized pseudoexposure (main) Main period (2015–2018) Random TM 0.034 0.007 Opposite direction D. Outcome Specificity Placebo outcome (height) Main period (2015–2018) True TimeMonths -0.008 0.149 No effect on irrelevant outcome Note: β₃ represents the interaction term TML_Long × TimeMonths (or random_tm for randomized tests). Temporal specificity: A significant effect postpolicy but not prepolicy validates the parallel trends assumption. Treatment specificity: True exposure has a strong negative effect; randomized exposure has a weak positive effect (opposite direction, 10-fold smaller magnitude). Outcome specificity: Strong effect on chronic disease; null effect on height (40-fold difference in magnitude). All p values are based on wild cluster bootstrapping with 28 provincial clusters. This finding indicates support for causal interpretation. 3.2.5 Synthesis of Robustness Evidence The convergence of evidence across three independent validation strategies—temporal specificity (significant postpolicy, nonsignificant prepolicy), treatment specificity (true exposure shows a negative effect, random exposure shows the opposite positive effect), and outcome specificity (significant for chronic diseases, null for height)—provides strong support for the causal interpretation of policy effects. This triangulation of evidence substantially enhances confidence in the causal inference that extended HC2030 exposure improves chronic disease management outcomes among elderly populations. 3.3 Temporal Dynamics and Critical Thresholds of Healthcare Service Effects On the basis of marginal prediction analysis, we identified critical time nodes in the association between policy exposure duration and chronic disease progression (Table 6 ): Detection phase (0–13 months) Increased healthcare utilization and disease identification, reflecting improved access rather than management effectiveness. Stabilization threshold (~ 13 months) Disease trajectories cross the zero-accumulation point, marking the transition from detection effects to management benefits. Significant improvement period (≥ 19 months) Statistically significant deceleration of disease accumulation, with progressively widening gaps between the long-term and short-term exposure groups. Table 6 Critical temporal thresholds for policy-induced health benefits: Model-based time points from marginal prediction analysis. Turning Point Estimate (months) Interpretation Equivalence crossover point 4.1 Long = Short burden Zero-Accumulation Point 13.1 Stabilization threshold Sample mean 16.1 Observed average 3.3.1 Prediction-Based Time Thresholds (1) Equivalence crossover point (4.1 months): the times at which the predicted chronic disease counts for the long-term and short-term exposure groups were equal. This point, calculated as t = -β₂L/β₃L, marks the point at which healthcare service improvements begin to produce relative advantages. (2) Zero-accumulation point (13.1 months): the time when the long-term group trajectory shifts from a net increase to stabilization. This threshold, calculated as t = -(β₀ + β₂L)/(β₁ + β₃L), represents population-level stabilization rather than individual-level disease reversal; only 20% of individuals reported net disease reduction. This indicates when health system strengthening begins to produce measurable health benefits. (3) Sustained improvement period (after 13.1 months): After the zero-crossing point, the gap between long-term and short-term groups continues to widen throughout the observation period, with effect sizes increasing monotonically and confidence intervals gradually shifting toward negative values, supporting the cumulative nature of policy effects. 3.3.2 Interpretation of the statistical findings Two complementary statistical findings support our conclusions: (1) Trajectory-level significance: the interaction term between the long-term exposure group and policy duration is statistically significant (β=-0.325, p = 0.005), indicating that chronic disease accumulation rates differ significantly over time across exposure groups. This tests whether trajectories (slopes) diverge—the core assumption of cumulative policy effects. (2) Time point estimates: Marginal predictions at specific time points represent effect sizes rather than hypothesis tests. Confidence intervals reflect estimation uncertainty inherent in quasiexperimental designs with provincial-level policy variation. The converging evidence supports the robustness of the findings: effect sizes increase monotonically with exposure duration, confidence intervals gradually shift toward negative values, and temporal specificity is confirmed by placebo tests. 3.3.3 Trajectory characteristics and clinical significance Further analysis revealed that the long-term exposure group (relative to the short-term group) presented a distinct temporal pattern across the three phases. During the initial period (0–4 months), the long-term group showed greater disease accumulation than did the baseline group, which is consistent with increased case identification following improved healthcare access—a detection effect commonly observed when service barriers are reduced. The transition period (4–13 months) marked a critical shift, with trajectories crossing as health management benefits began to outweigh detection effects. The sustained benefit period (after 13 months) demonstrated consistent protective effects, with the long-term group maintaining lower accumulation rates throughout the remaining observation period, indicating cumulative benefits of prolonged health system exposure. The consistent negative direction at all time points after 13 months further supports the potential clinical relevance of these findings. 3.4 Healthcare Service Utilization and Accessibility Factors The baseline chronic disease count was significantly negatively associated with chronic disease progression (β=-0.205, p < 0.001), and this effect was much stronger during the main analysis period than during the placebo period (β=-0.003, p = 0.831) (Appendix I Table S3, Table S4). This temporal pattern suggests that healthcare service management after policy implementation was particularly effective for high-risk populations (those with higher baseline disease burdens). From a health services research perspective, this "detection effect" reflects an immediate response to improved service accessibility and is an early signal of successful implementation. Hospitalization frequency showed a significant positive correlation during the main analysis period (β = 0.176, p < 0.001), whereas the correlation was weaker during the placebo period (β = 0.135, p = 0.005) (Appendix I Table S3). This strengthening may reflect (1) increased healthcare accessibility and more frequent disease monitoring after policy implementation or (2) increased disease severity requiring hospitalization. Owing to a lack of detailed data on the reasons for hospitalization (planned versus emergency), we cannot definitively distinguish between these two interpretations. Both functional limitations (β = 0.096, p < 0.001) and poor mental health status (β = 0.127, p < 0.001) were associated with faster chronic disease progression, indicating the need for comprehensive healthcare service interventions targeting these vulnerable subgroups. Rural‒urban residential differences also showed a negative correlation (rural: β=-0.230, p < 0.001), potentially reflecting differences in healthcare service capacity between urban and rural areas or relatively comprehensive chronic disease management infrastructure in rural areas (Appendix I Table S3). 3.5 Heterogeneity and Equity Analysis 3.5.1 Healthcare Service Effects across Sociodemographic Groups Heterogeneity analysis aims to generate hypotheses rather than validate predetermined hypotheses. Given that multiple comparisons may increase the risk of Type I error, all interaction term p values should be interpreted cautiously. The policy effects did not significantly differ across gender (p = 0.45), urban‒rural residence (p = 0.31), or education level (p = 0.28) groups, indicating that HC2030 implementation did not exacerbate existing health inequalities (Appendix I Table S6). This finding suggests that healthcare service improvements were distributed relatively equitably across social groups, supporting the policy's universal coverage strategy. 3.5.2 Exogeneity Verification To examine whether policy implementation timing was influenced by regional prior health or economic conditions, we examined correlations between 2015 implementation timing and baseline characteristics. The results revealed no significant correlations between implementation timing and baseline chronic disease count (ρ = 0.215, p = 0.270), ADL limitations (ρ=-0.094, p = 0.631), age (ρ=-0.139, p = 0.477), or per capita GDP (ρ=-0.004, p = 0.982) (Appendix I Table S7). On the basis of policy document records (Appendix I Table S8), we conclude that provincial implementation schedules are driven primarily by external factors, such as administrative processes, rather than by regional initial health status or economic levels. 3.6 Secondary Analysis: Functional Outcomes Analysis of activities of daily living (ADL) as a secondary outcome did not reveal a statistically significant association with policy exposure duration (long-term exposure group interaction term, β = -0.039, p = 0.705; Appendix I Table S9). This result is consistent with the hypothesis that physical function indicators may be less sensitive to short-term policy interventions. Potential explanations include the following: (1) the biological process of functional decline is typically slow, so a 3-year observation window may be insufficient to capture policy impacts on ADLs; (2) ADL limitations are driven primarily by severe chronic diseases and geriatric syndromes, whereas policies implemented during the observation period affected mainly identification and management rather than functional protection; and (3) the measurement of ADLs may be subject to a ceiling effect, as the baseline proportion of ADL-restricted individuals aged 60 years and above was low (mean, 0.54 items; standard deviation (SD), 1.15), limiting the statistical power to detect improvements. This finding aligns with previous literature, which indicates that functional outcomes generally require long-term follow-up (5–10 years) to demonstrate policy benefits [ 51 – 52 ]. 4. Discussion 4.1 Main findings Using staggered provincial implementation of China's Healthy China 2030 initiative as a natural experiment, we demonstrate that comprehensive health system reforms require extended exposure periods—approximately 13 months for stabilization and long-term sustained exposure—which produces a significant dose‒response gradient (interaction term β=-0.325, p = 0.005) before measurable population‒level benefits are produced. These findings address a critical gap in health policy evaluation: empirically identifying the temporal thresholds necessary to avoid premature dismissal of effective interventions. The identified critical time points—4.1 months (initiation) and 13.1 months (stabilization)—provide empirical anchors for timing health policy evaluations. These findings indicate that comprehensive health system reforms require longer time windows for service delivery changes to translate into measurable population health improvements, which is consistent with evaluations of quality assurance programs that show effects gradually appearing and tending to stabilize in the first two years [ 53 ]. Multiple lines of evidence support causal interpretation: temporal specificity (no prepolicy trends), treatment specificity (opposite effects for randomized pseudoexposure), and outcome specificity (null effects on height, a biologically implausible outcome). This convergent evidence substantially mitigates concerns about unmeasured confounding (detailed validation in Tables 3 – 5 ). 4.2 Service delivery mechanisms and temporal dynamics Our findings reveal two sequential lag mechanisms explaining why health benefits emerge only after extended exposure: Phase 1: Administrative Implementation Lag (0–13 months) During this initial period, policy directives translate into operational changes: resource allocation, protocol development, workforce training, and service reorganization. The 'detection effect'—increased healthcare utilization and disease diagnosis—dominates during this phase. This explains the initial increase in disease counts among long-term exposure groups (months 0–4, Fig. 1 ): improved access identifies previously undiagnosed conditions. Phase 2: Physiological response lag (13–19 months) As service delivery stabilizes, sustained clinical management produces biological responses. Blood pressure control, medication adherence, and lifestyle modifications require 3–6 months to manifest clinically [ 54 ]. Our 13-month stabilization threshold aligns with these biological timescales, whereas the 19-month threshold for significant effects reflects the cumulative nature of chronic disease management—preventing complications and slowing disease progression rather than reversing existing conditions. Mechanistic support: The strengthening effect of baseline disease burden in the main period (β = -0.205, p < 0.001) versus the placebo period (β = -0.003, p = 0.831) supports this interpretation. HC2030 particularly benefits high-risk populations through improved detection and management [ 55 ], which is consistent with the dual-lag framework. 4.3 Urban‒Rural Service Capacity Differences and Health Equity Urban–rural differences in temporal effects may reflect persistent differences in healthcare service capacity and accessibility. Despite decades of healthcare reform, urban areas maintain substantially higher medical professional density and service quality [ 56 ], potentially accelerating policy implementation and effect manifestation. However, this study revealed that policy effects were relatively evenly distributed across gender, urban‒rural, and education level groups, with no significant signs of inequality expansion. Challenging the Inverse Care Law through catch-up effects Rural residents exhibited slower chronic disease progression (β = -0.230, p < 0.001), an unexpected finding that contradicts Tudor Hart's (1971) "Inverse Care Law" [ 57 ]. This classic theory predicts that resources are concentrated in affluent areas with lower health needs rather than flowing to impoverished areas with more urgent needs—a pattern confirmed even in modern health systems [ 58 ]. We interpret this through a health economics lens: in historically underserved rural areas with low baseline physician density [ 56 ], HC2030's supply-side investments (infrastructure and workforce) likely generated high marginal utility—a 'catch-up effect.' Conversely, urban areas with higher baseline service levels may face 'diminishing marginal returns', where additional investments yield smaller immediate health gains. This suggests that HC2030's strategy effectively acted to narrow the structural health inequities predicted by the Inverse Care Law, ensuring that the policy benefits were most pronounced where the need was greatest [ 57 – 58 ]. Historical evidence supports the primacy of supply-side reform. A critical comparison shows why HC2030 succeeded when previous reforms failed. The new cooperative medical scheme (NCMS) rapidly expanded from its 2003 launch to cover over 800 million rural residents by 2008 but had no significant impact on rural residents' health status or out-of-pocket medical expenditures [ 59 ]. This null finding reveals the importance of supply-side constraints—simply expanding health insurance coverage without simultaneously improving rural healthcare service capacity fails to translate into substantial health improvements. HC2030 differs fundamentally by simultaneously addressing both demand-side factors (insurance coverage) and supply-side factors (service capacity, workforce, infrastructure), which may explain its greater observable effects on chronic disease outcomes than the NCMS does. This provides a crucial context for understanding why HC2030 generated measurable benefits in rural areas: universal coverage is necessary but insufficient, and service capacity determines whether expanded access translates into health gains. Caveats and alternative explanations We cannot rule out alternative explanations, including differential health behaviors, differences in disease surveillance quality, or selection bias in care seeking. However, the absence of significant interaction effects by urban/rural residence in our heterogeneity analysis (Appendix I Table S6) suggests that policy effects are distributed relatively equally across urban and rural areas. Future policies should prioritize rural healthcare service capacity building alongside, rather than covering, expansion. 4.4 Comparison with International Experience The temporal evolution patterns found in this study share similarities and differences with international experience. The observed time thresholds differ from those reported for comprehensive health promotion programs such as Finland's North Karelia Project and Japan's Healthy Japan 21, which require several years to demonstrate systematic improvements [ 12 , 13 ]. This difference may reflect the distinct nature of interventions: while those programs focused primarily on behavioral change and disease prevention, HC2030's effects in this study largely reflect expanded access to medical services and health system strengthening. Service expansion interventions may demonstrate measurable impacts more rapidly than behavioral interventions do. The Singapore primary healthcare project also exhibited similar medium-term improvement patterns [ 60 ]. However, international experience also highlights the necessity and complexity of long-term evaluation. The first phase of Japan's Health Japan 21 (2000–2012) achieved only 17% of predetermined targets after 13 years of implementation [ 14 ], revealing the long-term nature and multifactor constraints of chronic disease prevention and control. Even in developed countries with relatively complete policy implementation and evaluation systems, achieving the expected goals still faces enormous challenges. This further indicates that the time thresholds identified in this study (13 months for effect manifestation, 19 months for significant enhancement) are early signals of policy effects rather than complete reflections of final health benefits. It is important to distinguish between early detectable effects and ultimate program impacts. The time thresholds we identified (13-month stabilization, 19-month significant enhancement) represent the earliest points at which policy effects become statistically measurable, not the full realization of long-term health benefits. Our findings should be interpreted as evidence of early policy traction—demonstrating that health system improvements are beginning to influence disease trajectories—rather than as a complete assessment of HC2030's ultimate impacts. 4.5 Implications for Health Services Research and Policy Evaluation These findings indicate the need to align evaluation timing with the temporal dynamics of policy effects. Health services research has long recognized the critical impact of evaluation timing on conclusions about intervention effectiveness [ 1 , 2 ]. Premature evaluation may lead to erroneous rejection of beneficial interventions, and insufficient observation periods may fail to detect lag effects or lead to premature termination of beneficial projects [ 2 – 4 ]. This longitudinal study revealed that meaningful deceleration in chronic disease accumulation is detectable only after a minimum exposure period, with significant improvements stabilizing after 13 months. This directly informs a stratified evaluation framework: process indicators (0–4 months) focusing on coverage and service utilization; intermediate indicators (4–13 months) evaluating behavior change and service integration; and outcome indicators (≥ 13 months) measuring health improvements and disease progression control. This empirically anchored framework ensures that evaluations align with the actual trajectory of policy impacts, reducing the likelihood of false negative conclusions and improving evaluation precision. The most recent evaluation of Chinese hospital reform revealed that while structural improvements occurred rapidly, changes in final health outcomes required longer observation periods [ 61 ], which reinforces the importance of multistage evaluation strategies that assess both implementation process indicators and final outcomes at appropriate time intervals. 4.6 Practical Recommendations for Health System Planners On the basis of these findings, we offer the following recommendations for health system planners and policymakers: Minimum Implementation Period: Healthcare service reforms require at least 13 months of implementation before expecting measurable population health improvements. Governments should avoid hasty policy adjustments or termination decisions on the basis of data from the first year to ensure policy continuity and stability. Correct Interpretation of Early Utilization Increases: Utilization rate increases during the early stage (0–4 months) reflect improved accessibility and detection, rather than policy failure, and should be anticipated and planned accordingly. Health management departments should be prepared for demand surges during this stage, including temporary allocation of medical resources and patient flow management. Sustained Investment Commitment: Service delivery infrastructure investments should be maintained for at least 18–24 months before a comprehensive evaluation. Policymakers must resist pressure to prematurely reallocate resources on the basis of short-term indicators. Staged evaluation framework: Implement stratified monitoring strategies, with early focus (0–4 months) on process indicators and coverage, medium-term (4–13 months) evaluation of intermediate outcomes and service integration, and later-stage (≥ 13 months) measurement of substantial health outcome improvements. Equity Monitoring Mechanisms: While this study revealed no significant expansion of health inequalities, continuous monitoring of policy effects across different social groups remains crucial. Routine health equity monitoring systems should be established to promptly identify and correct potential disparities. Special attention should be given to vulnerable subgroups facing multiple barriers to healthcare access. Research examining digital health adoption among patients with chronic diseases has revealed significant gaps between awareness and actual service utilization among older adults [ 62 ]. Policy implementation strategies should incorporate targeted interventions to accelerate the management of these vulnerable groups. 4.7 Limitations Our study has several important limitations: Quasiexperimental design . While we validate parallel trends through placebo tests and employ robust inference methods, residual confounding from unmeasured time-varying provincial factors cannot be ruled out entirely. However, the convergence of evidence across three validation strategies (temporal, treatment, and outcome specificity) strengthens causal inference. Self-reported disease outcomes . Chronic disease counts rely on self-reports, potentially conflating true incidence with improved detection ('detection effect'). However, (a) validation studies demonstrate acceptable concordance with medical records [ 37 ]; (b) the detection effect represents an important early implementation signal; and (c) our outcome-specific placebo test (null effects on height) suggests that findings are not mere artifacts of increased healthcare interaction. Short observation window. The 3-year follow-up captures early-to-intermediate effects but not long-term sustainability. Critical unanswered questions include persistence beyond three years, impacts on mortality and disability, and differential effects across specific chronic conditions. Our identified thresholds (7-month stabilization, 19-month improvement) represent when effects become statistically detectable, not ultimate program impacts. Measurement limitations . Physical activity data were excluded because of substantial missing data (~ 50%), although the included ADL measures serve as functional proxies. Additionally, we cannot separate detection effects from true disease progression—future research should leverage biomarker data and healthcare utilization records to clarify the mechanisms involved. Generalizability . Our threshold estimates reflect China's unique context: near-universal insurance coverage (90.3%), extensive primary care networks, and specific implementation patterns. Settings with lower coverage or weaker infrastructure may require longer exposure periods for comparable effects. 5. Conclusion Evaluating health system reforms requires aligning assessment timelines with the latent periods of intervention effectiveness. Using the staggered implementation of China's Healthy China 2030 initiative as a natural experiment, we demonstrate that comprehensive chronic disease interventions require approximately 13 months to stabilize disease progression and 19 months to achieve significant population-level improvements. These empirically derived thresholds provide concrete benchmarks for evaluation design, challenging the practice of premature assessment that risks dismissing effective interventions. Our findings have three critical implications for policymakers and researchers: (1) Resist premature abandonment: Policy effects undergo a predictable temporal evolution—from initial detection effects (0–4 months) through stabilization (13 months) to significant improvement (19 + months). Evaluations conducted before these thresholds systematically underestimate program benefits. (2) Supply-side reforms matter: HC2030's success derived from simultaneously strengthening service capacity (infrastructure, workforce) alongside insurance expansion—a lesson from previous reforms that achieved high coverage but minimal health gains owing to supply constraints [ 59 ]. (3) Context-dependent thresholds: Our estimates reflect China's specific institutional context. Generalization requires understanding that systems with weaker baseline infrastructures may need even longer exposure periods. Ultimately, effective health policy evaluation requires aligning assessment timelines with the biological and administrative realities of population health improvement. Only through this combination of patience and rigor can we build health systems that truly improve population health rather than merely expanding coverage. Abbreviations CHARLS The China Health and Retirement Longitudinal Study DiD Difference-in-Differences HC2030 Healthy China 2030 MI Multiple Imputation WCB Wild Cluster Bootstrap Declarations Ethics approval and consent to participate This study complies with the Declaration of Helsinki. Ethical approval for the China Health and Retirement Longitudinal Study (CHARLS) was obtained from the Institutional Review Board of Peking University (IRB00001052-11015 for the main household survey; IRB00001052-11014 for biomarker collection). All participants provided written informed consent at the time of data collection. This study involved secondary analysis of deidentified CHARLS data obtained under a data-use licence. In accordance with institutional guidelines for research using publicly available anonymized data, no additional ethical approval was required for this secondary analysis. Consent for publication Informed consent was obtained from all participants by the CHARLS research team prior to data collection. This study is based on anonymized secondary data. Competing Interests The authors declare that they have no competing interests. Funding This research was funded by the Guangdong Planning Office of Philosophy and Social Science, grant number GD23XSH18. The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Author Contribution As the sole author, I was responsible for the conception and design of the study, data analysis and interpretation, and manuscript preparation. I have read and approved the final manuscript. Acknowledgement The authors acknowledge the China Health and Retirement Longitudinal Study (CHARLS) team for providing access to the data used in this study. Data Availability The data supporting the findings of this study are available from the China Health and Retirement Longitudinal Study (CHARLS) at https://charls.pku.edu.cn. Access to CHARLS data is open to researchers upon completion of a data use agreement and registration process, in accordance with CHARLS data sharing policies. Provincial policy implementation dates were compiled from official government documents and are provided in Appendix I Table S8 with full source citations. The analytical code used in this study is available from the corresponding author upon reasonable request. References Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ. 2008;a1655. https://doi.org/10.1136/bmj.a1655 . Hill KG, Woodward D, Woelfel T, Hawkins JD, Green S. Planning for Long-Term Follow-Up: Strategies Learned from Longitudinal Studies. Prev Sci. 2016;17:806–18. https://doi.org/10.1007/s11121-015-0610-7 . White IR, Horton NJ, Carpenter J, Statistics RIMAS, Pocock SJ. Strategy for intention to treat analysis in randomised trials with missing outcome data. BMJ. 2011;342(1):d40–40. https://doi.org/10.1136/bmj.d40 . Gaglio B, Shoup JA, Glasgow RE. The RE-AIM Framework: A Systematic Review of Use Over Time. Am J Public Health. 2013;103:e38–46. https://doi.org/10.2105/AJPH.2013.301299 . Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011;104:510–20. https://doi.org/10.1258/jrsm.2011.110180 . Shediac-Rizkallah MC, Bone LR. Planning for the sustainability of community-based health programs: conceptual frameworks and future directions for research, practice and policy. Health Educ Res. 1998;13:87–108. https://doi.org/10.1093/her/13.1.87 . Scheirer MA. Is Sustainability Possible? A Review and Commentary on Empirical Studies of Program Sustainability. Am J Evaluation. 2005;26:320–47. https://doi.org/10.1177/1098214005278752 . Sandler I, Wolchik SA, Cruden G, Mahrer NE, Ahn S, Brincks A, et al. Overview of Meta-Analyses of the Prevention of Mental Health, Substance Use, and Conduct Problems. Annu Rev Clin Psychol. 2014;10:243–73. https://doi.org/10.1146/annurev-clinpsy-050212-185524 . Lopez Bernal J, Cummins S, Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2016. https://doi.org/10.1093/ije/dyw098 . :dyw098. Proctor E, Luke D, Calhoun A, McMillen C, Brownson R, McCrary S, et al. Sustainability of evidence-based healthcare: research agenda, methodological advances, and infrastructure support. Implement Sci. 2015;10:88. https://doi.org/10.1186/s13012-015-0274-5 . Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of complex interventions: Medical Research Council guidance. BMJ. 2015;350:h1258–1258. https://doi.org/10.1136/bmj.h1258 . mar19 6. Puska P, Jaini P. The North Karelia Project: Prevention of Cardiovascular Disease in Finland Through Population-Based Lifestyle Interventions. Am J Lifestyle Med. 2020;14:495–9. https://doi.org/10.1177/1559827620910981 . Tsuji I. Current status and issues concerning Health Japan 21 (second term). Nutrition Reviews. 2020;78 Supplement_3:14–7. https://doi.org/10.1093/nutrit/nuaa079 Nomura S, Sakamoto H, Ghaznavi C, Inoue M. Toward a third term of Health Japan 21 – implications from the rise in non-communicable disease burden and highly preventable risk factors. Lancet Reg Health - Western Pac. 2022;21:100377. https://doi.org/10.1016/j.lanwpc.2021.100377 . Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396:1204–22. https://doi.org/10.1016/S0140-6736(20)30925-9 . Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396:1204–22. https://doi.org/10.1016/S0140-6736(20)30925-9 . Marengoni A, Angleman S, Melis R, Mangialasche F, Karp A, Garmen A, et al. Aging with multimorbidity: A systematic review of the literature. Ageing Res Rev. 2011;10:430–9. https://doi.org/10.1016/j.arr.2011.03.003 . Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. 2012;380:37–43. https://doi.org/10.1016/S0140-6736(12)60240-2 . Ontario H. Continuity of care to optimize chronic disease management in the community setting: an evidence-based analysis. Ont Health Technol Assess Ser. 2013;13:1–41. Organization WH, editor. others. Integrated care for older people: guidelines on community-level interventions to manage declines in intrinsic capacity. In: Integrated care for older people: guidelines on community-level interventions to manage declines in intrinsic capacity. 2017. Worrall G, Knight J. Continuity of care is good for elderly people with diabetes: retrospective cohort study of mortality and hospitalization. Can Fam Physician. 2011;57:e16–20. Zhou W, Guo M, Hu B, Jiang Y, Yao Y. The effect of China’s Integrated Medical and Social Care Policy on functional dependency and care deficits in older adults: a nationwide quasi-experimental study. Lancet Healthy Longev. 2025;6:100697. https://doi.org/10.1016/j.lanhl.2025.100697 . De Araujo I, Epping-Jordan J, Pot AM, Kelley E, Toro N, Thiyagarajan JA, et al. Organizing integrated health-care services to meet older people’s needs. Bull World Health Organ. 2017;95:756–63. https://doi.org/10.2471/BLT.16.187617 . Edgren L. The meaning of integrated care: a systems approach. Int J Integr Care. 2008;8. https://doi.org/10.5334/ijic.293 . Valentijn PP, Schepman SM, Opheij W, Bruijnzeels MA. Understanding integrated care: a comprehensive conceptual framework based on the integrative functions of primary care. Int J Integr Care. 2013;13. https://doi.org/10.5334/ijic.886 . The State Council of the People’s Republic of China. Healthy China 2030 planning outline. 2016. http://www.gov.cn/zhengce/2016-10/25/content_5124174.htm . Accessed 11 Aug 2025. Tan X, Liu X, Shao H. Healthy China 2030: A Vision for Health Care. Value Health Reg Issues. 2017;12:112–4. https://doi.org/10.1016/j.vhri.2017.04.001 . Heilmann S. Policy Experimentation in China’s Economic Rise. St Comp Int Dev. 2008;43:1–26. https://doi.org/10.1007/s12116-007-9014-4 . Xu C. The Fundamental Institutions of China’s Reforms and Development. J Econ Lit. 2011;49:1076–151. https://doi.org/10.1257/jel.49.4.1076 . Wang Y, Wang R, Jiang M, Ying X. Provincial inequality of China’s progress towards universal health coverage: An empirical analysis in 2016–21. J Glob Health. 2024;14:04122. https://doi.org/10.7189/jogh.14.04122 . Ning C, Pei H, Huang Y, Li S, Shao Y. Does the Healthy China 2030 Policy Improve People’s Health? Empirical Evidence Based on the Difference-in-Differences Approach. RMHP 2024;Volume 17:65–77. https://doi.org/10.2147/RMHP.S439581 Duflo E. Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual Policy Experiment. Am Econ Rev. 2001;91:795–813. https://doi.org/10.1257/aer.91.4.795 . Bleakley H. Disease and Development: Evidence from Hookworm Eradication in the American South. Q J Econ. 2007;122:73–117. https://doi.org/10.1162/qjec.121.1.73 . Callaway B, Sant’Anna PHC. Difference-in-Differences with multiple time periods. J Econ. 2021;225:200–30. https://doi.org/10.1016/j.jeconom.2020.12.001 . Zhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort Profile: The China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43:61–8. https://doi.org/10.1093/ije/dys203 . Organization WH. others. Second round of the national pulse survey on continuity of essential health services during the COVID-19 pandemic: January-March 2021: interim report, 22 April 2021. In: Second round of the national pulse survey on continuity of essential health services during the COVID-19 pandemic: January-March 2021: interim report, 22 April 2021. 2021. Liu H, Zhao Y, Qiao L, Yang C, Yang Y, Zhang T, et al. Consistency between self-reported disease diagnosis and clinical assessment and under-reporting for chronic conditions: data from a community-based study in Xi’an, China. Front Public Health. 2024;12:1296939. https://doi.org/10.3389/fpubh.2024.1296939 . Chan KS, Kasper JD, Brandt J, Pezzin LE, IADL Difficulty Across International Surveys of Aging. Measurement Equivalence in ADL and : Findings From the HRS, SHARE, and ELSA. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2012;67B:121–32. https://doi.org/10.1093/geronb/gbr133 Imbens G. The role of the propensity score in estimating dose-response functions. Biometrika. 2000;87:706–10. https://doi.org/10.1093/biomet/87.3.706 . VanderWeele T. Explanation in causal inference: methods for mediation and interaction. Oxford University Press; 2015. Andersen RM. Revisiting the Behavioral Model and Access to Medical Care: Does it Matter? J Health Soc Behav. 1995;36:1. https://doi.org/10.2307/2137284 . Callaway B, Goodman-Bacon A, Sant’Anna PH. Difference-in-differences with a Continuous Treatment. Cambridge, MA: National Bureau of Economic Research; 2024. https://doi.org/10.3386/w32117 . de Chaisemartin C, D’Haultfœuille X. Two-way fixed effects and differences-in-differences with heterogeneous treatment effects: a survey. Econometrics J. 2023;26:C1–30. https://doi.org/10.1093/ectj/utac017 . Goodman-Bacon A. Difference-in-differences with variation in treatment timing. J Econ. 2021;225:254–77. https://doi.org/10.1016/j.jeconom.2021.03.014 . Li F, Li F. Propensity score weighting for causal inference with multiple treatments. Ann Appl Stat. 2019;13. https://doi.org/10.1214/19-AOAS1282 . Van Buuren S. Flexible imputation of missing data. CRC; 2018. Little R, Rubin D. Multiple imputation for nonresponse in surveys. Wiley. 1987;10:9780470316696. Cameron AC, Gelbach JB, Miller DL. Bootstrap-Based Improvements for Inference with Clustered Errors. Rev Econ Stat. 2008;90:414–27. https://doi.org/10.1162/rest.90.3.414 . MacKinnon JG, Webb MD. The wild bootstrap for few (treated) clusters. Econometrics J. 2018;21:114–35. https://doi.org/10.1111/ectj.12107 . Angrist JD, Pischke J-S. Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton University Press; 2009. https://doi.org/10.1515/9781400829828 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J et al. Frailty in Older Adults: Evidence for a Phenotype. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2001;56:M146–57. https://doi.org/10.1093/gerona/56.3.M146 Stuck AE, Egger M, Hammer A, Minder CE, Beck JC. Home Visits to Prevent Nursing Home Admission and Functional Decline in Elderly People: Systematic Review and Meta-regression Analysis. JAMA. 2002;287. https://doi.org/10.1001/jama.287.8.1022 . Gehrig S, Zander-Jentsch B, Gutzeit M, Klein S, Rauh J. Estimating the causal effect of a quality assurance program on quality of care in Germany. BMC Health Serv Res. 2025;25:815. https://doi.org/10.1186/s12913-025-12939-8 . Wagner EH, Austin BT, Davis C, Hindmarsh M, Schaefer J, Bonomi A. Improving Chronic Illness Care: Translating Evidence Into Action. Health Aff. 2001;20:64–78. https://doi.org/10.1377/hlthaff.20.6.64 . Hu X, Huang J, Lv Y, Li G, Peng X. Status of prevalence study on multimorbidity of chronic disease in China: Systematic review. Geriatr Gerontol Int. 2015;15:1–10. https://doi.org/10.1111/ggi.12340 . Zhao N, Chen K. Equity and efficiency of medical and health service system in China. BMC Health Serv Res. 2023;23:33. https://doi.org/10.1186/s12913-023-09025-2 . Tudor Hart J, THE INVERSE CARE LAW. Lancet. 1971;297:405–12. https://doi.org/10.1016/S0140-6736(71)92410-X . The Lancet. 50 years of the inverse care law. Lancet. 2021;397:767. https://doi.org/10.1016/S0140-6736(21)00505-5 . Lei X, Lin W. The New Cooperative Medical Scheme in rural China: does more coverage mean more service and better health? Health Econ. 2009;18. https://doi.org/10.1002/hec.1501 . Goh LH, Siah CJR, Szücs A, Tai ES, Valderas JM, Young D. Integrated patient-centred care for type 2 diabetes in Singapore Primary Care Networks: a mixed-methods study. BMJ Open. 2024;14:e083992. https://doi.org/10.1136/bmjopen-2024-083992 . Jiang Y, Han Z, Nicholas S, Yang W, Maitland E, Shi X, et al. Did China’s hospital reforms improve curative care expenditures? Evidence from Beijing hospitals. BMC Health Serv Res. 2025;25:627. https://doi.org/10.1186/s12913-025-12785-8 . Qiu R, Song R, Wu X, Feng J, Yang Y, Pan Y, et al. From awareness to adoption: a panoramic perspective on the utilization of Internet Medical Services among Chinese patients with chronic disease. BMC Health Serv Res. 2025;25:1415. https://doi.org/10.1186/s12913-025-13601-z . Additional Declarations No competing interests reported. Supplementary Files supplementary1125.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8189460","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":562796123,"identity":"f8ded73a-f20c-45f3-b330-e10ca8a075ae","order_by":0,"name":"Hin-Wa Wong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0UlEQVRIiWNgGAWjYBACAwh1QI6B+QCQZiNBizEDWwKJWhIbiNZiLpH87OHXtjvp29l4DBg+lB1m4J/dgF+L5Yw0c2PZtme5O9t4DBhnnDvMIHHnAAGH3Ugwk5ZsO5y74X6PATNv22EGA4kEQlrSv4G0pBsc4zFg/kuclhwzyY9thxPAWhiJ0nLmTZk0w7nDhhuOsRUc7DmXziNxg5CW4+nbJH+UHZY3OMa88cGPMms5/hkEtIAAMy80Og4AMQ9h9UDA+OMPUepGwSgYBaNgpAIABDpGjBbJrQ0AAAAASUVORK5CYII=","orcid":"","institution":"Suzhou University","correspondingAuthor":true,"prefix":"","firstName":"Hin-Wa","middleName":"","lastName":"Wong","suffix":""}],"badges":[],"createdAt":"2025-11-24 05:38:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8189460/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8189460/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98729075,"identity":"7dfe5fad-3e81-4e67-a0c3-beff7e866fc7","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129211,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript1129.docx","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/c333aa67818a8c61e8cd73f0.docx"},{"id":98729074,"identity":"9cda903c-9d56-443f-8c85-3b37190d5a06","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5077,"visible":true,"origin":"","legend":"","description":"","filename":"06d559f8aa5743348cae13c15345a6c7.json","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/e81d1ba0957efb03cb0346c0.json"},{"id":98777129,"identity":"8f41023f-3661-47a8-9250-2121d99bc4c7","added_by":"auto","created_at":"2025-12-22 12:25:27","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231832,"visible":true,"origin":"","legend":"","description":"","filename":"supplementary1125.docx","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/82df7e94f31db2f1f8cc2e0e.docx"},{"id":98729077,"identity":"5a0cf8bb-caab-4d97-a272-702bdcd600d0","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":192587,"visible":true,"origin":"","legend":"","description":"","filename":"06d559f8aa5743348cae13c15345a6c71enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/77bc3ecd6d54cf4a53f38102.xml"},{"id":98777427,"identity":"52164c0b-3546-4b8a-b2d8-43e179b5f887","added_by":"auto","created_at":"2025-12-22 12:27:04","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56324,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/cf6dc8ff37422e30b8395875.jpeg"},{"id":98729080,"identity":"18f90024-aaf9-476f-bcdc-06044c51dd29","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50043,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/897da4b7f9e1f6caf2ffa71c.png"},{"id":98729078,"identity":"3e06c137-0390-4e21-8168-a1d643a3a842","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"xml","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":190715,"visible":true,"origin":"","legend":"","description":"","filename":"06d559f8aa5743348cae13c15345a6c71structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/8db075b855dc9fe88e80ec44.xml"},{"id":98729082,"identity":"5bc484a0-6654-4d2f-9888-c80f677d0e5b","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"html","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":204838,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/721511184b4b430cc48d1647.html"},{"id":98729073,"identity":"6dee3dd9-5051-4326-8c96-8928eb1e9265","added_by":"auto","created_at":"2025-12-22 04:27:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":215272,"visible":true,"origin":"","legend":"\u003cp\u003ePolicy Effects Emerge only After Extended Exposure: Temporal Trajectories of Chronic Disease Accumulation by Exposure Duration Group.\u003c/p\u003e\n\u003cp\u003eNote: The intersection of trajectories (approx. 4–13 months) illustrates the critical transition from the initial 'detection effect' (increased diagnosis) to the 'management effect' (decelerated progression). The long-term group (\u0026gt;75th percentile) demonstrated a clear downwards trajectory after this stabilization period.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/a3ef047a53b824810da1fee5.png"},{"id":105788050,"identity":"da077813-1d6a-4e4c-bcd0-793d9537a020","added_by":"auto","created_at":"2026-03-31 06:58:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2143465,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/f0e986fc-dadd-49f7-a1e6-099c944492e3.pdf"},{"id":98729079,"identity":"03f418a3-eb30-42f5-8a8c-958a7f3f0c67","added_by":"auto","created_at":"2025-12-22 04:27:33","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":231832,"visible":true,"origin":"","legend":"","description":"","filename":"supplementary1125.docx","url":"https://assets-eu.researchsquare.com/files/rs-8189460/v1/c8524681a958a56716e96719.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Policy threshold for decelerating chronic disease accumulation among older Chinese adults: Evidence from the Healthy China 2030 Initiative","fulltext":[{"header":"1. Background","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e1.1 Critical impact of evaluation timing on health policy assessment\u003c/h2\u003e \u003cp\u003eThe temporal dynamics between policy implementation and health outcomes critically shape the evaluation of health system reform. Complex interventions demonstrate delayed effects that emerge only after a sufficient exposure duration [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Insufficient observation periods risk premature termination of beneficial interventions [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], yet systematic evidence on optimal evaluation timing remains scarce. Preventive health interventions often exhibit substantial implementation lags, with full effects emerging only after years or decades [\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Chronic disease management programs face particularly long latency periods due to the slow biological processes of disease progression and the time required to establish continuous care relationships [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. However, empirical evidence quantifying these temporal thresholds remains limited.\u003c/p\u003e \u003cp\u003eDespite the growing recognition of the importance of long-term evaluation [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], few studies have systematically identified critical time thresholds for policy effects. Finland's North Karelia Project required 5 years before measurable cardiovascular benefits were demonstrated [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], whereas Japan's Health Japan 21 achieved only 17% of predetermined targets after 13 years [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. These cases illustrate the long-term nature of population health interventions but provide limited guidance for evaluation design in other contexts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e1.2 Elderly Chronic Disease Management as an Indicator of Health Service Effectiveness\u003c/h2\u003e \u003cp\u003eElderly chronic disease management is a key indicator for assessing health service effectiveness. Chronic diseases are highly prevalent among older adults and are primarily attributable to modifiable risk factors; therefore, policy responses that prioritize long-term outcomes are essential [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Multimorbidity is common in this population and is closely linked to functional impairment, reduced quality of life, and increased healthcare utilization [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEffective management of chronic diseases in older adults requires a continuous service chain encompassing prevention, diagnosis, treatment, and rehabilitation [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. The accessibility and continuity of these services are central to the core functions of primary healthcare systems. Service quality influences disease control, complication prevention, and functional independence. Empirical evidence indicates that high-quality continuity of care can reduce hospitalization rates among elderly diabetes patients by approximately 15 percentage points (from 68.2% to 53.5%) and nearly halve mortality (from 18.5% to 8.6%) [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Integrated medical and social care policies also significantly reduce functional dependency risks and care gaps among older populations [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eChronic disease management in older adults is a sensitive indicator for evaluating the effectiveness of \"health-centered\" integrated service systems [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. From a systems science perspective, integrated care constitutes a complex adaptive system whose effectiveness relies on coordination and integration across the clinical, functional, organizational, and systemic levels [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Disease progression rates reflect the combined influence of medical accessibility (early detection) and service quality (effective management), capturing the cumulative effects of improvements in insurance coverage, provider capacity, and health education programs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e1.3 Healthy China 2030: A Natural Experiment in Health System Reform\u003c/h2\u003e \u003cp\u003eChina's \"Healthy China 2030\" (HC2030) initiative was launched in October 2016 [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], providing a unique context for exploring the long-term effects of healthcare service reform [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. As China's national-level medium- to long-term health strategy, this policy framework established a macrolevel shift from a \"disease-centered\" approach to a \"health-centered\" approach. Provinces successively formulated and launched localized action plans between 2016 and 2018, resulting in variation in the duration of policy exposure (0\u0026ndash;21 months) by the time of the 2018 CHARLS survey.\u003c/p\u003e \u003cp\u003eThis study assumes that these temporal differences are driven primarily by administrative processes and policy diffusion mechanisms rather than by direct responses to local health needs. This assumption is based on typical patterns of policy implementation in China: the diffusion of central policies to the provincial level typically follows experimentation within hierarchical mechanisms [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] rather than selective adoption on the basis of local demand. Although HC2030 was released nationwide as a unified strategic framework in 2016 [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], the timing of provinces' formulation of specific action plans and supporting policies may reflect this administration-driven diffusion pattern, thereby providing a source of variation in policy exposure duration for quasiexperimental design.\u003c/p\u003e \u003cp\u003eHC2030 transformed the healthcare service delivery landscape through a multifaceted approach. This includes expanding insurance coverage to enhance financial accessibility, alongside strengthening primary healthcare infrastructure with standardized management protocols. Complementing these structural changes are efforts to increase medical workforce training\u0026mdash;particularly in rural areas\u0026mdash;and the development of integrated health information systems to improve care coordination. [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Although preliminary assessments suggest short-term or regional improvements [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], evidence of medium- to long-term effects on chronic diseases in older adults remains limited.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e1.4 Research methods and contributions\u003c/h2\u003e \u003cp\u003eTraditional binary exposure models, which compare outcomes before and after policy implementation, are unable to capture variations in policy exposure intensity or assess dose-dependent effects. This study utilized a grouped difference-in-differences (DiD) design with dose-specific time trends to compare chronic disease trajectories across three exposure groups: short-term, medium-term, and long-term. This approach tests whether increasing policy exposure duration influences chronic disease accumulation rates by incorporating interaction terms between exposure duration and group membership, estimating group-specific disease progression slopes, and detecting dose-dependent treatment effects. The design is informed by established methods for estimating treatment effect heterogeneity by intensity [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] and recent advances in continuous treatment difference-in-differences methodologies [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe focus on three specific research questions: (1) Is extended HC2030 exposure associated with improved chronic disease management outcomes via enhanced healthcare service accessibility? (2) At what time threshold do health system reforms begin to yield measurable health impacts? (3) Do effects differ by baseline health status and sociodemographic characteristics? By identifying critical time points at which policy effects emerge and stabilize, this study offers empirical guidance for designing observation periods in health policy evaluation, particularly in low- and middle-income countries, and advances methodologies in health services research on reform evaluation.\u003c/p\u003e \u003c/div\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Data Sources and Sample Selection\u003c/h2\u003e \u003cp\u003eThis study utilized Harmonized CHARLS Version D data, including four waves of nationally representative surveys collected in 2011, 2013, 2015, and 2018. CHARLS employs a multistage, stratified, probability‒proportional‒to-size (PPS) sampling design, ensuring national representativeness of Chinese adults aged 45 years and above [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. This data infrastructure enables international comparisons through harmonized variable definitions, making it comparable with international counterpart surveys such as the U.S. Health and Retirement Study (HRS) and the English Longitudinal Study of Aging (ELSA).\u003c/p\u003e \u003cp\u003eThe main analytical sample includes respondents aged 60 and above in 2015 who provided complete, valid data in both the 2015 and 2018 survey waves. The 2020 survey wave was excluded because COVID-19-related healthcare service disruptions, changes in care-seeking behavior, and interruptions in chronic disease management would introduce systematic biases inconsistent with prepandemic observation periods. The disruptive impact of the pandemic on health data has been well documented in the WHO's multicountry pulse surveys [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAll the statistical analyses were conducted via Stata/MP 18.0. The analysis code and detailed data processing scripts are available from the corresponding author upon reasonable request to promote reproducibility and transparency.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Variable Definitions\u003c/h2\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1 Primary outcome variable\u003c/h2\u003e \u003cp\u003eChronic disease accumulation was measured as the change in disease count (ΔChronic) between 2015 and 2018. The CHARLS assesses 13 physician-diagnosed chronic conditions, including hypertension, diabetes, and cardiovascular disease (see the CHARLS Harmonized D manual for the full list). Positive values indicate disease accumulation, negative values indicate improvement, and zero values indicate stability.\u003c/p\u003e \u003cp\u003eSelf-reported disease counts were used as the primary outcome for three reasons: (1) the 2018 CHARLS wave lacked biomarker measurements, precluding longitudinal biomarker analysis; (2) self-reported diagnoses were consistently collected across waves, enabling robust temporal comparisons; and (3) validation studies demonstrated acceptable concordance between self-reports and medical records in similar populations [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Although this approach may conflate true disease incidence with improved detection resulting from increased healthcare access, the 'detection effect' is considered an important early signal of successful policy implementation rather than a limitation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2 Secondary outcome variable\u003c/h2\u003e \u003cp\u003eChanges in activities of daily living (ΔADL) scores were designated secondary outcomes to assess the impact of policy exposure duration on functional status among older adults. This selection is justified primarily by the well-established correlation between chronic disease burden and functional impairment [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], suggesting that improved disease management may delay functional decline. Furthermore, ADL serves as a validated proxy for quality of life and independence, representing a critical dimension for evaluating the comprehensive benefits of health policies [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e2.2.3 Policy Exposure Measurement\u003c/h2\u003e \u003cp\u003eThe policy exposure duration was operationalized as the cumulative number of months from provincial HC2030 implementation to the 2018 CHARLS follow-up, ranging from 0\u0026ndash;21 months across provinces. Provincial implementation dates were defined as the dates when provincial governments officially released \"Healthy China 2030\" implementation plans or equivalent documents (see the complete provincial policy implementation timeline in Appendix I Table S7).\u003c/p\u003e \u003cp\u003eHC2030 implementation enhanced healthcare service accessibility and quality through the following mechanisms: (1) financial accessibility, i.e., expanding insurance coverage and increasing reimbursement rates for outpatient and inpatient services; (2) service capacity, i.e., strengthening primary healthcare infrastructure and establishing standardized chronic disease management protocols; (3) human resources, i.e., increasing medical workforce training and deployment, particularly in underserved rural areas; and (4) system integration, i.e., developing integrated health information systems to improve care coordination and referral networks.\u003c/p\u003e \u003cp\u003eProvincial variation in implementation timing (0\u0026ndash;21 months) provided natural variation in exposure to these service delivery improvements, enabling identification of when health system strengthening translates into measurable health outcomes. To examine the nonlinear dynamic characteristics of the association between policy exposure duration and chronic disease progression, continuous policy exposure duration (tml) was divided into three mutually exclusive groups on the basis of quartiles: the short-term exposure group (\u0026le;\u0026thinsp;25th percentile, approximately 0\u0026ndash;15 months), the medium-term exposure group (25\u0026ndash;75th percentile, approximately 15\u0026ndash;18 months), and the long-term exposure group (\u0026gt;\u0026thinsp;75th percentile, approximately\u0026thinsp;\u0026ge;\u0026thinsp;19 months). As a single linear trend or simple intergroup mean comparison might mask heterogeneity in disease progression rates within different exposure stages, a piecewise linear regression strategy was employed. Specifically, interaction terms between group dummy variables and continuous time variables (Group \u0026times; Time) were introduced into the model. This design allows for simultaneous estimation of differences in baseline levels across exposure stages (intercept effects) and differences in rates of change over time (slope effects). Significant interaction terms indicate that chronic disease accumulation trajectories undergo structural changes as the duration of policy exposure increases, thereby validating the marginal benefits of the policy at different stages. This grouping strategy maintains adequate sample sizes for statistical power in each group while enabling the detection of potential nonlinear associations [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Notably, this grouping design employs a grouped-comparison analytical strategy rather than estimating continuous dose‒response relationships.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e2.2.4 Covariates\u003c/h2\u003e \u003cp\u003eCovariate selection followed the framework of Andersen's behavioral model of health services use [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e], which includes the following: (1) predisposing factors: individual-level demographic characteristics (age, sex, marital status); (2) enabling factors: socioeconomic status (education level, household income or assets), health insurance coverage; (3) need factors: health behaviors (smoking, alcohol consumption), functional and mental health indicators (ADL scores, depression scale scores), baseline chronic disease burden and health service utilization (hospitalization frequency, outpatient visits); and (4) contextual factors: provincial time-varying macroeconomic and health resource indicators extracted from the China Statistical Yearbook, including per capita regional gross domestic product (GDP) and health resource allocation indicators.\u003c/p\u003e \u003cp\u003eThe physical activity variable exhibited a relatively high proportion of missing values (approximately 50%) in the analytical sample, and missing pattern analysis suggested possible systematic differences. Because including variables with high missing rates can substantially reduce the sample size and introduce selection bias, physical activity control variables were excluded from the main regression model.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Statistical analysis\u003c/h2\u003e \u003cp\u003e \u003cem\u003e2.3.1 Primary Strategy for Evaluating the Timing of Health Service Delivery: Grouped Heterogeneous Treatment Effects with a Difference-in-Differences Design\u003c/em\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMethodological Positioning\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe analytical approach leverages variation in HC2030 implementation timing across provinces (0\u0026ndash;21 months) to identify policy effects. Rather than treating all provinces identically, provinces are grouped by exposure duration, and the divergence in disease progression trajectories between groups is estimated over time. This strategy tests whether extended policy exposure yields stronger health benefits, which is consistent with a dose\u0026ndash;response relationship.\u003c/p\u003e \u003cp\u003eTechnical approach: We employ a grouped difference-in-differences design combining first-differencing with group-specific time trends, which helps to eliminate time-invariant confounding variables and to detect dose-dependent effects. This builds on recent methods for continuous treatment DiD [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan additionalcitationids=\"CR43\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e], adapted for discrete exposure groups to improve interpretability and statistical power.\u003c/p\u003e \u003cp\u003eWhile this study draws on core insights from continuous treatment difference-in-differences methods\u0026mdash;using variation in treatment intensity to identify the temporal dynamics of policy effects [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u0026mdash;the analysis does not estimate a fully continuous dose\u0026ndash;response function. Instead, treatment effect heterogeneity is tested by comparing discrete exposure groups (short-term, medium-term, and long-term). This design choice is justified by several considerations: (1) flexibility, as the grouping approach allows policy effects to vary nonparametrically across exposure durations without assuming linear dose‒response relationships; (2) interpretability, since discrete group comparisons are more accessible to policymakers for identifying critical time thresholds; (3) statistical power, as grouped comparisons are more robust with limited sample sizes; and (4) theoretical fit, given that policy effects may exhibit activation thresholds rather than smooth linear changes.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCore Research Questions and Model Specification\u003c/b\u003e \u003c/p\u003e \u003cp\u003eFrom a health services research perspective, this approach addresses the following key questions: (1) Is extended HC2030 exposure associated with improved chronic disease management outcomes via enhanced healthcare service accessibility? (2) At what time threshold do health system reforms begin to yield measurable health impacts? (3) Do these effects differ by baseline health status and sociodemographic characteristics?\u003c/p\u003e \u003cp\u003eThe equation for the main analytical model is as follows:\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"609\" height=\"154\"\u003e\u003c/p\u003e\u003c/p\u003e \u003cp\u003ewhere ΔYit represents the cumulative change in chronic disease count for individual i between two follow-ups; TimeMonthsit is the policy exposure duration in months (continuous variable, 0\u0026ndash;21 months); TML_Mediumit is a dummy variable for the medium-term exposure group (1\u0026thinsp;=\u0026thinsp;medium-term exposure, 0\u0026thinsp;=\u0026thinsp;otherwise); TML_Longit is a dummy variable for the long-term exposure group (1\u0026thinsp;=\u0026thinsp;long-term exposure, 0\u0026thinsp;=\u0026thinsp;otherwise); the short-term exposure group serves as the reference category (baseline group); interaction terms estimate the time-dependent differential effects of the medium-term and long-term exposure groups relative to the short-term exposure group; Xitγ is a covariate vector; and εit is the random error term (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eKey parameters and interpretation.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameter\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeaning\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEpidemiological Interpretation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eβ₁\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBaseline slope\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDisease accumulation rate in short-term exposure group (reference)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eβ₂M, β₂L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLevel differences\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBaseline differences between groups (absorbed by differencing)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eβ₃M, β₃L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSlope differences (DiD estimator)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChange in accumulation rate for medium/long-term groups relative to baseline\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003eNote: Critical insight: β₃L\u0026thinsp;\u0026lt;\u0026thinsp;0 and statistical significance indicate that long-term exposure significantly decelerates disease accumulation compared with short-term exposure\u0026mdash;our core hypothesis.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eParallel trends assumption: In the absence of differential policy effects, all groups would follow similar disease trajectories (β₃M\u0026thinsp;=\u0026thinsp;β₃L\u0026thinsp;=\u0026thinsp;0). Significant negative β₃ coefficients indicate policy-induced divergence.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe epidemiological interpretation of model parameters is as follows: β₀ (Intercept): represents the average natural change in chronic diseases at a theoretical zero-exposure time point, or baseline level, for the sample population (background trend due to natural aging). β₁ (baseline slope): represents the dose‒response relationship for the short-term exposure group (reference group). That is, during the short-term exposure stage, there is a marginal increase in chronic disease accumulation per additional month of policy exposure. This constitutes the 'counterfactual' baseline trend for our comparisons. β₂ (Intercept Difference): represents systematic differences in regression line intercepts between medium/long-term exposure groups relative to the short-term exposure group (typically used to absorb inherent level differences between groups). β₃ (Slope Difference/DiD Estimator): This is the core focus of this study. It represents the degree of deviation in the chronic disease accumulation rate (Slope) for the medium/long-term exposure groups relative to the short-term exposure group. If β₃ \u0026lt; 0 and is statistically significant, this indicates that as the duration of policy exposure increases, the cumulative speed of chronic disease is significantly lower than the baseline speed of the short-term group, confirming the marginal protective effect of long-term policy exposure.\u003c/p\u003e \u003cp\u003e \u003cb\u003eParallel Trends Assumption and Testing\u003c/b\u003e \u003c/p\u003e \u003cp\u003eUnder this first-difference framework, the parallel trends assumption transforms into the \"parallel growth assumption\": we assume that in the absence of causal effects from differential policy exposure duration, the chronic disease accumulation rates (slopes) of the medium/long-term exposure groups would have remained consistent with those of the short-term exposure group (i.e., β₃ should equal 0). Any significant deviation of the β₃ coefficient from zero is attributed to the causal effect of the duration of policy exposure.\u003c/p\u003e \u003cp\u003eThese assumptions are validated through the following strategies: (1) prepolicy period placebo test (2011\u0026ndash;2015), which uses the same grouping strategy and model specification to test whether future exposure group classification is associated with health trajectory differences before policy implementation; (2) covariate balance testing, which involves examining balance across three exposure groups on baseline covariates after overlap weighting; and (3) robustness tests, which include randomized pseudoexposure testing and placebo outcome testing.\u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Overlap weighting for balancing healthcare service access\u003c/h2\u003e \u003cp\u003eTo construct comparable reference populations across different policy exposure duration groups, we employed the overlap weighting (OW) method for multinomial treatments [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. This method reduces selection bias arising from uneven covariate distributions by assigning weights to observations with propensity scores within the common support region, which is particularly important for comparing provinces with different baseline levels of healthcare service accessibility.\u003c/p\u003e \u003cp\u003eFor three-category treatment (short-term/medium-term/long-term exposure groups), we first estimate generalized propensity scores (GPSs) via multinomial logistic regression: P(TML\u0026thinsp;=\u0026thinsp;k|X)\u0026thinsp;=\u0026thinsp;exp(X'γk)/Σⱼ exp(X'γⱼ), k \u0026isin; {short-term, medium-term, long-term}. where X includes all baseline covariates. On the basis of GPS, overlap weights are calculated as follows: for individual i in exposure group k, w_ik\u0026thinsp;=\u0026thinsp;Π_{j\u0026thinsp;\u0026ne;\u0026thinsp;k} e_ij, which is the product of generalized propensity scores for all groups except k.\u003c/p\u003e \u003cp\u003eThis weighting scheme has the following advantages: (1) it maximizes the effective sample size: compared with inverse probability weighting (IPW), overlap weights are insensitive to extreme propensity scores; (2) it ensures covariate balance: in weighted pseudopopulation, the three exposure groups have identical covariate distributions; and (3) it focuses on the overlap region: the estimated effects correspond to subpopulations where all three groups are adequately represented. Standardized mean differences (SMDs) are used to assess balance, with SMDs\u0026thinsp;\u0026lt;\u0026thinsp;0.10 indicating good balance and SMDs\u0026thinsp;\u0026lt;\u0026thinsp;0.05 indicating excellent balance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 Missing data handling and statistical inference\u003c/h2\u003e \u003cp\u003eTo handle missing data, we used multiple imputation by chained equations (MICE), generating 50 imputed datasets [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. The imputation model included all the analytical variables and relevant auxiliary variables to approximate the missing-at-random assumption. The results were combined via Rubin's rules [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eGiven the limited number of provincial clusters (n\u0026thinsp;=\u0026thinsp;28), traditional cluster-robust standard errors may underestimate sampling variability and increase Type I error rates. Therefore, we combined wild cluster bootstrapping (WCB) with multiple imputation for statistical inference [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. Specifically, 10,000 WCB repetitions are performed on each imputed dataset via Rademacher weights; point estimates and variances across imputed datasets are combined via Rubin's rules; and 95% confidence intervals and p values are constructed. The simulation results show that WCB methods can still provide reliable inference when the number of clusters is small (e.g., G between 15\u0026ndash;20) under certain conditions [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e \u003ch2\u003e2.3.4 Identifying Critical Time Points for Healthcare Service Effects\u003c/h2\u003e \u003cp\u003eTo identify critical time points at which healthcare service improvements translate into health outcomes. On the basis of the marginal prediction curves, we calculate the following:\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eEquivalence crossover point\u003c/strong\u003e \u003cp\u003eThe time when long-term and short-term groups reach equal predicted outcomes, calculated as t = -β₂L/β₃L. This represents the point at which health benefits begin to offset initial differences.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eZero-Accumulation Point\u003c/strong\u003e \u003cp\u003eThe time when the long-term group's predicted outcome crosses zero, calculated as t = -(β₀+β₂L)/(β₁+β₃L). This marks the transition from disease accumulation to population-level stabilization.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eThese calculations use point estimates from the regression model. We focus on trajectory-level inference (testing whether group slopes differ) rather than precise threshold estimation, as our primary interest is in demonstrating that sustained health system exposure produces measurable benefits within a policy-relevant timeframe.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e \u003ch2\u003e2.3.5 Robustness and placebo tests\u003c/h2\u003e \u003cp\u003eTo assess the reliability of causal identification [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e], we conducted sensitivity analyses across four dimensions: (1) temporal specificity: prepolicy placebo test (2011\u0026ndash;2015), designating 2013 as the \"virtual policy start year,\" testing whether interaction terms between virtual exposure groups and time months are statistically significant in the prepolicy period; (2) treatment specificity: randomized pseudoexposure test, using random number generators to generate provincial random exposure durations (0\u0026ndash;21 months), testing whether random grouping produces significant effects; (3) outcome specificity: placebo outcome test, using measured height as a negative control, testing whether policy exposure is associated with height changes; and (4) heterogeneity testing: including three-way interaction terms of the policy exposure group with age, sex, urban‒rural residence, and education level.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Sample characteristics\u003c/h2\u003e \u003cp\u003eThe CHARLS 2015 baseline survey included 25,586 respondents. After applying the inclusion criterion of age 60 and above, 8,337 eligible individuals with a mean age of 67.98 years (SD\u0026thinsp;=\u0026thinsp;6.67) were included; 48.4% were male, and 90.3% had basic medical insurance. The sample reflects typical characteristics of China's elderly population: 79.5% have a primary education or lower, and 77.5% reside in rural areas. The baseline mean chronic disease count was 2.17 (SD\u0026thinsp;=\u0026thinsp;1.69); by the 2018 follow-up, chronic diseases had increased on average by 0.45 (SD\u0026thinsp;=\u0026thinsp;1.29). A total of 17.3% of the respondents reported that the incidence of chronic disease decreased, 40.8% remained unchanged, and 41.9% reported that it increased. The mean policy exposure duration was 16.09 months (SD\u0026thinsp;=\u0026thinsp;3.73) (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Appendix I Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAfter overlap weighting adjustment, the baseline characteristics across the exposure groups achieved good balance, with all SMDs\u0026thinsp;\u0026lt;\u0026thinsp;0.05 (Table S2). Owing to missing weights in some samples, the final sample size used in the statistical models was 7,487 individuals (Appendix II, Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive Statistics of the Key Variables (N\u0026thinsp;=\u0026thinsp;8337).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMean/n (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMissing n(%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOutcome \u0026amp; Core Policy Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eΔ Chronic conditions count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7922\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e415 (5.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime in months (TML)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e16.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML: Short Duration (\u0026le;\u0026thinsp;25th pctile)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2211 (26.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML: Medium Duration\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4972 (59.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML: Long Duration (\u0026gt;\u0026thinsp;75th pctile)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1154 (13.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDemographic Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e67.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4032 (48.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLow education, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6629 (79.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCouple Household, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6552 (78.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRural residence, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7700\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5967 (77.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e637 (7.6%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInsured, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7891\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7126 (90.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e446 (5.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth and Lifestyle Variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline chronic conditions count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7923\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e414 (5.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eADL count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8301\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36 (0.4%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospitalization times (past year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8290\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e47 (0.6%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOut-of-pocket hospitalization cost In USD (past year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8147\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e137.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e755.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e190 (2.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking frequency per day\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8230\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e107 (1.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEver Drinking last year, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8300\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2671(32.18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e61 (0.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eNote: All the statistics are based on original data before multiple imputation. For continuous variables, the means\u0026thinsp;\u0026plusmn;\u0026thinsp;SDs are reported. For binary categorical variables, n (%) of the positive/reference category is reported. See Appendix I Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e for the complete table.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Main analysis and robustness checks\u003c/h2\u003e \u003cdiv id=\"Sec21\" class=\"Section3\"\u003e \u003ch2\u003e3.2.1 Primary Estimates of Health System Strengthening Effects\u003c/h2\u003e \u003cp\u003eMain findings: Policy effects depended critically on exposure duration (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Appendix I Table S3). Compared with short-term exposure, long-term exposure (\u0026ge;\u0026thinsp;19 months) significantly slowed chronic disease accumulation (β = -0.325, SE\u0026thinsp;=\u0026thinsp;0.116, p\u0026thinsp;=\u0026thinsp;0.005, 95% CI: -0.552\u0026ndash;0.098).\u003c/p\u003e \u003cp\u003eClinical significance: This effect is substantial. The average population experienced 0.45 additional chronic conditions over the observation period due to natural aging. The estimated policy effect offsets approximately 72% of this age-related burden among individuals with long-term exposure (0.325/0.45\u0026thinsp;=\u0026thinsp;0.72), moving from net accumulation toward stabilization.\u003c/p\u003e \u003cp\u003eIn contrast, neither the main effect of policy exposure duration (β\u0026thinsp;=\u0026thinsp;0.009, p\u0026thinsp;=\u0026thinsp;0.879) nor the interaction term for the medium-term exposure group (β=-0.016, p\u0026thinsp;=\u0026thinsp;0.636) achieved statistical significance. These findings indicate that healthcare service delivery improvements under HC2030 require a longer implementation period (\u0026ge;\u0026thinsp;19 months) to manifest in population-level chronic disease outcomes.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDifference-in-differences estimates of policy exposure duration on chronic disease accumulation: Group-specific effects with wild cluster bootstrap inference (N\u0026thinsp;=\u0026thinsp;7,487).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1) Pooled MI (SE)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2) WCB Corrected (SE)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDose Interaction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTimeMonths (Continuous, Ref: Low Dose)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.009 (0.008)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.009 (0.068)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML: Medium \u0026times;TimeMonths\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.016(0.016)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.008 (0.034)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML: Long \u0026times;TimeMonths\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.325*** (0.046)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;0.342*** (0.078)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePolicy Level (DiD Intercept)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML: Long (Ref: Low)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.323*** (0.226)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.392*** (0.483)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKey Health/SES Controls\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline chronic conditions count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.205(0.153)***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;0.213(0.019)***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRural residence (1\u0026thinsp;=\u0026thinsp;Rural)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.230(0.522)***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;0.253(0.066)***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDisability Score PCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.096(0.184)***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.094(0.019)***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMental Health Score PCA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.127(0.021)***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.133(0.215)***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospitalization times (past year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.176(0.028)***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.187(0.034)***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFixed Effects \u0026amp; Inference\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual Fixed Effects\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime Fixed Effects\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClusters (Provinces)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVCE/Inference Method\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClustered SE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWild Cluster Bootstrap\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN (Obs)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7487\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7487\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003eNote: * * p\u0026thinsp;\u0026lt;\u0026thinsp;0.01; ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; * p\u0026thinsp;\u0026lt;\u0026thinsp;0.1. The standard error (STD. err.) is in brackets. See Appendix I Table S3 for the estimated results of the complete model, including the coefficients of all individual characteristics, provincial characteristics, health behavior, and economic variables.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eNote\u003c/strong\u003e \u003cp\u003eThe intersection of trajectories (approx. 4\u0026ndash;13 months) illustrates the critical transition from the initial 'detection effect' (increased diagnosis) to the 'management effect' (decelerated progression). The long-term group (\u0026gt;\u0026thinsp;75th percentile) demonstrated a clear downwards trajectory after this stabilization period.\u003c/p\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section3\"\u003e \u003ch2\u003e3.2.2 Validating the parallel trends assumption: Prepolicy placebo test\u003c/h2\u003e \u003cp\u003eThe placebo test applied the main analytical model to the prepolicy period (2011\u0026ndash;2015), using actual future policy exposure duration (measured 2016\u0026ndash;2018) as the treatment variable (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Appendix I Table S4). The results revealed no significant interaction effects during the placebo period. The critical interaction term for long-term exposure was small and nonsignificant (β\u0026thinsp;=\u0026thinsp;0.089, p\u0026thinsp;=\u0026thinsp;0.474, WCB), in contrast with the significant negative effect observed in the main analysis (β=-0.325, p\u0026thinsp;=\u0026thinsp;0.005). Similarly, the medium-term exposure interaction was not significant (β\u0026thinsp;=\u0026thinsp;0.010, p\u0026thinsp;=\u0026thinsp;0.671). These nonsignificant placebo interactions validate the parallel trends assumption, indicating that exposure groups followed similar trajectories before policy initiation.\u003c/p\u003e \u003cp\u003eNotably, there was a significant main time effect during the placebo period (β = -0.012, p\u0026thinsp;=\u0026thinsp;0.033), indicating that provinces that would later have longer exposure periods experienced slightly greater chronic disease deceleration from 2011\u0026ndash;2015. This main effect does not violate the parallel trends assumption, which requires parallel group-specific trajectories rather than zero baseline trends. Moreover, this effect reversed during the main analysis period (β\u0026thinsp;=\u0026thinsp;0.009, p\u0026thinsp;=\u0026thinsp;0.879), suggesting period-specific secular trends rather than persistent selection bias.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of the estimated longitudinal cumulative association coefficients between the main analysis and the placebo period.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndicator\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMain\u003c/p\u003e \u003cp\u003e(2015\u0026rarr;2018)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePlacebo\u003c/p\u003e \u003cp\u003e(2011\u0026rarr;2015)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDifference\u003c/p\u003e \u003cp\u003e(M-P)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTML Long \u0026times; TimeMonths_c\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB coef.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.325\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.414\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB SE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.116\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB p value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.474\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePooled cofe.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.325\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.414\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePooled p value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.254\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline Chronic diseases Count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB coef.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.205\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.202\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB p value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.831\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospitalizations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB coef.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.176\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.135\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWCB p value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eNote: Post-2016 changes in TML long interactive items and baseline chronic disease effects indicate structural shifts in health management following policy implementation. The prepolicy trends show no significant differences, whereas the postpolicy differences support the parallel trend assumption.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e3.2.3 Treatment specificity: Randomized pseudoexposure test\u003c/h2\u003e \u003cp\u003eTo assess whether the significant main time effect during the placebo period reflects spurious correlation with any continuous provincial variable, we conducted a robustness check via randomized pseudoexposure (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Critically, during the main analysis period, randomized pseudoexposure produced a small positive interaction for the long-term exposure group (β\u0026thinsp;=\u0026thinsp;0.034, p\u0026thinsp;=\u0026thinsp;0.007), which was in the opposite direction and approximately 10 times smaller in magnitude than the true policy effect (β=-0.325, p\u0026thinsp;=\u0026thinsp;0.005). The sharp contrast in direction and magnitude provides strong evidence that the observed policy effects reflect true causal impacts rather than spurious correlations with preexisting provincial trends.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e \u003ch2\u003e3.2.4 Outcome specificity: Placebo outcome test\u003c/h2\u003e \u003cp\u003eTo verify that the observed effects reflect true policy impacts on chronic disease outcomes rather than unmeasured time-varying provincial confounders, we conducted a placebo outcome test using measured height as a negative control. The height of adults aged 60 years and above should not be affected by medical policies for chronic disease management. The results revealed no significant association between policy exposure and changes in height (Appendix I Table S5). The long-term exposure interaction term was small and nonsignificant (β = -0.0083, p\u0026thinsp;=\u0026thinsp;0.203, WCB), approximately 40 times smaller than the effect on chronic disease count (β = -0.325, p\u0026thinsp;=\u0026thinsp;0.005, WCB).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of Robustness Checks: Triple Validation of Causal Inference\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eValidation Dimension\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eTest Design\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eExposure Variable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKey Interaction\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eInterpretation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(β₃)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eA. Main Analysis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrimary outcome (chronic disease)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMain period (2015\u0026ndash;2018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTrue TimeMonths\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.325\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eStrong negative effect\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eB. Temporal Specificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrepolicy placebo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePlacebo period (2011\u0026ndash;2015)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTrue TimeMonths\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.474\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo prepolicy differential trends\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eC. Treatment Specificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRandomized pseudoexposure (placebo)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePlacebo period (2011\u0026ndash;2015)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRandom TM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.422\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo spurious pretrends\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRandomized pseudoexposure (main)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMain period (2015\u0026ndash;2018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRandom TM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eOpposite direction\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD. Outcome Specificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlacebo outcome (height)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMain period (2015\u0026ndash;2018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTrue TimeMonths\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.149\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo effect on irrelevant outcome\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003eNote: β₃ represents the interaction term TML_Long \u0026times; TimeMonths (or random_tm for randomized tests). Temporal specificity: A significant effect postpolicy but not prepolicy validates the parallel trends assumption. Treatment specificity: True exposure has a strong negative effect; randomized exposure has a weak positive effect (opposite direction, 10-fold smaller magnitude). Outcome specificity: Strong effect on chronic disease; null effect on height (40-fold difference in magnitude). All p values are based on wild cluster bootstrapping with 28 provincial clusters. This finding indicates support for causal interpretation.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003e3.2.5 Synthesis of Robustness Evidence\u003c/h2\u003e \u003cp\u003eThe convergence of evidence across three independent validation strategies\u0026mdash;temporal specificity (significant postpolicy, nonsignificant prepolicy), treatment specificity (true exposure shows a negative effect, random exposure shows the opposite positive effect), and outcome specificity (significant for chronic diseases, null for height)\u0026mdash;provides strong support for the causal interpretation of policy effects. This triangulation of evidence substantially enhances confidence in the causal inference that extended HC2030 exposure improves chronic disease management outcomes among elderly populations.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Temporal Dynamics and Critical Thresholds of Healthcare Service Effects\u003c/h2\u003e \u003cp\u003eOn the basis of marginal prediction analysis, we identified critical time nodes in the association between policy exposure duration and chronic disease progression (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e):\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eDetection phase (0\u0026ndash;13 months)\u003c/strong\u003e \u003cp\u003eIncreased healthcare utilization and disease identification, reflecting improved access rather than management effectiveness.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eStabilization threshold (~\u0026thinsp;13 months)\u003c/strong\u003e \u003cp\u003eDisease trajectories cross the zero-accumulation point, marking the transition from detection effects to management benefits.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSignificant improvement period (\u0026ge;\u0026thinsp;19 months)\u003c/strong\u003e \u003cp\u003eStatistically significant deceleration of disease accumulation, with progressively widening gaps between the long-term and short-term exposure groups.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCritical temporal thresholds for policy-induced health benefits: Model-based time points from marginal prediction analysis.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTurning Point\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEstimate (months)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInterpretation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEquivalence crossover point\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLong\u0026thinsp;=\u0026thinsp;Short burden\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZero-Accumulation Point\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStabilization threshold\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSample mean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eObserved average\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003e3.3.1 Prediction-Based Time Thresholds\u003c/h2\u003e \u003cp\u003e(1) Equivalence crossover point (4.1 months): the times at which the predicted chronic disease counts for the long-term and short-term exposure groups were equal. This point, calculated as t = -β₂L/β₃L, marks the point at which healthcare service improvements begin to produce relative advantages.\u003c/p\u003e \u003cp\u003e(2) Zero-accumulation point (13.1 months): the time when the long-term group trajectory shifts from a net increase to stabilization. This threshold, calculated as t = -(β₀ + β₂L)/(β₁ + β₃L), represents population-level stabilization rather than individual-level disease reversal; only 20% of individuals reported net disease reduction. This indicates when health system strengthening begins to produce measurable health benefits.\u003c/p\u003e \u003cp\u003e(3) Sustained improvement period (after 13.1 months): After the zero-crossing point, the gap between long-term and short-term groups continues to widen throughout the observation period, with effect sizes increasing monotonically and confidence intervals gradually shifting toward negative values, supporting the cumulative nature of policy effects.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section3\"\u003e \u003ch2\u003e3.3.2 Interpretation of the statistical findings\u003c/h2\u003e \u003cp\u003eTwo complementary statistical findings support our conclusions: (1) Trajectory-level significance: the interaction term between the long-term exposure group and policy duration is statistically significant (β=-0.325, p\u0026thinsp;=\u0026thinsp;0.005), indicating that chronic disease accumulation rates differ significantly over time across exposure groups. This tests whether trajectories (slopes) diverge\u0026mdash;the core assumption of cumulative policy effects. (2) Time point estimates: Marginal predictions at specific time points represent effect sizes rather than hypothesis tests. Confidence intervals reflect estimation uncertainty inherent in quasiexperimental designs with provincial-level policy variation.\u003c/p\u003e \u003cp\u003eThe converging evidence supports the robustness of the findings: effect sizes increase monotonically with exposure duration, confidence intervals gradually shift toward negative values, and temporal specificity is confirmed by placebo tests.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section3\"\u003e \u003ch2\u003e3.3.3 Trajectory characteristics and clinical significance\u003c/h2\u003e \u003cp\u003eFurther analysis revealed that the long-term exposure group (relative to the short-term group) presented a distinct temporal pattern across the three phases. During the initial period (0\u0026ndash;4 months), the long-term group showed greater disease accumulation than did the baseline group, which is consistent with increased case identification following improved healthcare access\u0026mdash;a detection effect commonly observed when service barriers are reduced. The transition period (4\u0026ndash;13 months) marked a critical shift, with trajectories crossing as health management benefits began to outweigh detection effects. The sustained benefit period (after 13 months) demonstrated consistent protective effects, with the long-term group maintaining lower accumulation rates throughout the remaining observation period, indicating cumulative benefits of prolonged health system exposure.\u003c/p\u003e \u003cp\u003eThe consistent negative direction at all time points after 13 months further supports the potential clinical relevance of these findings.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec30\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Healthcare Service Utilization and Accessibility Factors\u003c/h2\u003e \u003cp\u003eThe baseline chronic disease count was significantly negatively associated with chronic disease progression (β=-0.205, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and this effect was much stronger during the main analysis period than during the placebo period (β=-0.003, p\u0026thinsp;=\u0026thinsp;0.831) (Appendix I Table S3, Table S4). This temporal pattern suggests that healthcare service management after policy implementation was particularly effective for high-risk populations (those with higher baseline disease burdens). From a health services research perspective, this \"detection effect\" reflects an immediate response to improved service accessibility and is an early signal of successful implementation.\u003c/p\u003e \u003cp\u003eHospitalization frequency showed a significant positive correlation during the main analysis period (β\u0026thinsp;=\u0026thinsp;0.176, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), whereas the correlation was weaker during the placebo period (β\u0026thinsp;=\u0026thinsp;0.135, p\u0026thinsp;=\u0026thinsp;0.005) (Appendix I Table S3). This strengthening may reflect (1) increased healthcare accessibility and more frequent disease monitoring after policy implementation or (2) increased disease severity requiring hospitalization. Owing to a lack of detailed data on the reasons for hospitalization (planned versus emergency), we cannot definitively distinguish between these two interpretations.\u003c/p\u003e \u003cp\u003eBoth functional limitations (β\u0026thinsp;=\u0026thinsp;0.096, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and poor mental health status (β\u0026thinsp;=\u0026thinsp;0.127, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were associated with faster chronic disease progression, indicating the need for comprehensive healthcare service interventions targeting these vulnerable subgroups. Rural‒urban residential differences also showed a negative correlation (rural: β=-0.230, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), potentially reflecting differences in healthcare service capacity between urban and rural areas or relatively comprehensive chronic disease management infrastructure in rural areas (Appendix I Table S3).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Heterogeneity and Equity Analysis\u003c/h2\u003e \u003cdiv id=\"Sec32\" class=\"Section3\"\u003e \u003ch2\u003e3.5.1 Healthcare Service Effects across Sociodemographic Groups\u003c/h2\u003e \u003cp\u003eHeterogeneity analysis aims to generate hypotheses rather than validate predetermined hypotheses. Given that multiple comparisons may increase the risk of Type I error, all interaction term p values should be interpreted cautiously. The policy effects did not significantly differ across gender (p\u0026thinsp;=\u0026thinsp;0.45), urban‒rural residence (p\u0026thinsp;=\u0026thinsp;0.31), or education level (p\u0026thinsp;=\u0026thinsp;0.28) groups, indicating that HC2030 implementation did not exacerbate existing health inequalities (Appendix I Table S6). This finding suggests that healthcare service improvements were distributed relatively equitably across social groups, supporting the policy's universal coverage strategy.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec33\" class=\"Section3\"\u003e \u003ch2\u003e3.5.2 Exogeneity Verification\u003c/h2\u003e \u003cp\u003eTo examine whether policy implementation timing was influenced by regional prior health or economic conditions, we examined correlations between 2015 implementation timing and baseline characteristics. The results revealed no significant correlations between implementation timing and baseline chronic disease count (ρ\u0026thinsp;=\u0026thinsp;0.215, p\u0026thinsp;=\u0026thinsp;0.270), ADL limitations (ρ=-0.094, p\u0026thinsp;=\u0026thinsp;0.631), age (ρ=-0.139, p\u0026thinsp;=\u0026thinsp;0.477), or per capita GDP (ρ=-0.004, p\u0026thinsp;=\u0026thinsp;0.982) (Appendix I Table S7). On the basis of policy document records (Appendix I Table S8), we conclude that provincial implementation schedules are driven primarily by external factors, such as administrative processes, rather than by regional initial health status or economic levels.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec34\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Secondary Analysis: Functional Outcomes\u003c/h2\u003e \u003cp\u003eAnalysis of activities of daily living (ADL) as a secondary outcome did not reveal a statistically significant association with policy exposure duration (long-term exposure group interaction term, β = -0.039, p\u0026thinsp;=\u0026thinsp;0.705; Appendix I Table S9). This result is consistent with the hypothesis that physical function indicators may be less sensitive to short-term policy interventions.\u003c/p\u003e \u003cp\u003ePotential explanations include the following: (1) the biological process of functional decline is typically slow, so a 3-year observation window may be insufficient to capture policy impacts on ADLs; (2) ADL limitations are driven primarily by severe chronic diseases and geriatric syndromes, whereas policies implemented during the observation period affected mainly identification and management rather than functional protection; and (3) the measurement of ADLs may be subject to a ceiling effect, as the baseline proportion of ADL-restricted individuals aged 60 years and above was low (mean, 0.54 items; standard deviation (SD), 1.15), limiting the statistical power to detect improvements. This finding aligns with previous literature, which indicates that functional outcomes generally require long-term follow-up (5\u0026ndash;10 years) to demonstrate policy benefits [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec36\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Main findings\u003c/h2\u003e \u003cp\u003eUsing staggered provincial implementation of China's Healthy China 2030 initiative as a natural experiment, we demonstrate that comprehensive health system reforms require extended exposure periods\u0026mdash;approximately 13 months for stabilization and long-term sustained exposure\u0026mdash;which produces a significant dose‒response gradient (interaction term β=-0.325, p\u0026thinsp;=\u0026thinsp;0.005) before measurable population‒level benefits are produced. These findings address a critical gap in health policy evaluation: empirically identifying the temporal thresholds necessary to avoid premature dismissal of effective interventions.\u003c/p\u003e \u003cp\u003eThe identified critical time points\u0026mdash;4.1 months (initiation) and 13.1 months (stabilization)\u0026mdash;provide empirical anchors for timing health policy evaluations. These findings indicate that comprehensive health system reforms require longer time windows for service delivery changes to translate into measurable population health improvements, which is consistent with evaluations of quality assurance programs that show effects gradually appearing and tending to stabilize in the first two years [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMultiple lines of evidence support causal interpretation: temporal specificity (no prepolicy trends), treatment specificity (opposite effects for randomized pseudoexposure), and outcome specificity (null effects on height, a biologically implausible outcome). This convergent evidence substantially mitigates concerns about unmeasured confounding (detailed validation in Tables\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec37\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Service delivery mechanisms and temporal dynamics\u003c/h2\u003e \u003cp\u003eOur findings reveal two sequential lag mechanisms explaining why health benefits emerge only after extended exposure:\u003c/p\u003e \u003cp\u003e \u003cem\u003ePhase 1: Administrative Implementation Lag (0\u0026ndash;13 months)\u003c/em\u003e \u003c/p\u003e \u003cp\u003eDuring this initial period, policy directives translate into operational changes: resource allocation, protocol development, workforce training, and service reorganization. The 'detection effect'\u0026mdash;increased healthcare utilization and disease diagnosis\u0026mdash;dominates during this phase. This explains the initial increase in disease counts among long-term exposure groups (months 0\u0026ndash;4, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e): improved access identifies previously undiagnosed conditions.\u003c/p\u003e \u003cp\u003e \u003cem\u003ePhase 2: Physiological response lag (13\u0026ndash;19 months)\u003c/em\u003e \u003c/p\u003e \u003cp\u003eAs service delivery stabilizes, sustained clinical management produces biological responses. Blood pressure control, medication adherence, and lifestyle modifications require 3\u0026ndash;6 months to manifest clinically [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. Our 13-month stabilization threshold aligns with these biological timescales, whereas the 19-month threshold for significant effects reflects the cumulative nature of chronic disease management\u0026mdash;preventing complications and slowing disease progression rather than reversing existing conditions.\u003c/p\u003e \u003cp\u003eMechanistic support: The strengthening effect of baseline disease burden in the main period (β = -0.205, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) versus the placebo period (β = -0.003, p\u0026thinsp;=\u0026thinsp;0.831) supports this interpretation. HC2030 particularly benefits high-risk populations through improved detection and management [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e], which is consistent with the dual-lag framework.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec38\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Urban‒Rural Service Capacity Differences and Health Equity\u003c/h2\u003e \u003cp\u003eUrban\u0026ndash;rural differences in temporal effects may reflect persistent differences in healthcare service capacity and accessibility. Despite decades of healthcare reform, urban areas maintain substantially higher medical professional density and service quality [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], potentially accelerating policy implementation and effect manifestation. However, this study revealed that policy effects were relatively evenly distributed across gender, urban‒rural, and education level groups, with no significant signs of inequality expansion.\u003c/p\u003e \u003cp\u003e \u003cem\u003eChallenging the Inverse Care Law through catch-up effects\u003c/em\u003e \u003c/p\u003e \u003cp\u003eRural residents exhibited slower chronic disease progression (β = -0.230, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), an unexpected finding that contradicts Tudor Hart's (1971) \"Inverse Care Law\" [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. This classic theory predicts that resources are concentrated in affluent areas with lower health needs rather than flowing to impoverished areas with more urgent needs\u0026mdash;a pattern confirmed even in modern health systems [\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe interpret this through a health economics lens: in historically underserved rural areas with low baseline physician density [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], HC2030's supply-side investments (infrastructure and workforce) likely generated high marginal utility\u0026mdash;a 'catch-up effect.' Conversely, urban areas with higher baseline service levels may face 'diminishing marginal returns', where additional investments yield smaller immediate health gains. This suggests that HC2030's strategy effectively acted to narrow the structural health inequities predicted by the Inverse Care Law, ensuring that the policy benefits were most pronounced where the need was greatest [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eHistorical evidence supports the primacy of supply-side reform.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eA critical comparison shows why HC2030 succeeded when previous reforms failed. The new cooperative medical scheme (NCMS) rapidly expanded from its 2003 launch to cover over 800\u0026nbsp;million rural residents by 2008 but had no significant impact on rural residents' health status or out-of-pocket medical expenditures [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. This null finding reveals the importance of supply-side constraints\u0026mdash;simply expanding health insurance coverage without simultaneously improving rural healthcare service capacity fails to translate into substantial health improvements.\u003c/p\u003e \u003cp\u003eHC2030 differs fundamentally by simultaneously addressing both demand-side factors (insurance coverage) and supply-side factors (service capacity, workforce, infrastructure), which may explain its greater observable effects on chronic disease outcomes than the NCMS does. This provides a crucial context for understanding why HC2030 generated measurable benefits in rural areas: universal coverage is necessary but insufficient, and service capacity determines whether expanded access translates into health gains.\u003c/p\u003e \u003cp\u003e \u003cem\u003eCaveats and alternative explanations\u003c/em\u003e \u003c/p\u003e \u003cp\u003eWe cannot rule out alternative explanations, including differential health behaviors, differences in disease surveillance quality, or selection bias in care seeking. However, the absence of significant interaction effects by urban/rural residence in our heterogeneity analysis (Appendix I Table S6) suggests that policy effects are distributed relatively equally across urban and rural areas. Future policies should prioritize rural healthcare service capacity building alongside, rather than covering, expansion.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec39\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Comparison with International Experience\u003c/h2\u003e \u003cp\u003eThe temporal evolution patterns found in this study share similarities and differences with international experience. The observed time thresholds differ from those reported for comprehensive health promotion programs such as Finland's North Karelia Project and Japan's Healthy Japan 21, which require several years to demonstrate systematic improvements [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. This difference may reflect the distinct nature of interventions: while those programs focused primarily on behavioral change and disease prevention, HC2030's effects in this study largely reflect expanded access to medical services and health system strengthening. Service expansion interventions may demonstrate measurable impacts more rapidly than behavioral interventions do. The Singapore primary healthcare project also exhibited similar medium-term improvement patterns [\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHowever, international experience also highlights the necessity and complexity of long-term evaluation. The first phase of Japan's Health Japan 21 (2000\u0026ndash;2012) achieved only 17% of predetermined targets after 13 years of implementation [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], revealing the long-term nature and multifactor constraints of chronic disease prevention and control. Even in developed countries with relatively complete policy implementation and evaluation systems, achieving the expected goals still faces enormous challenges. This further indicates that the time thresholds identified in this study (13 months for effect manifestation, 19 months for significant enhancement) are early signals of policy effects rather than complete reflections of final health benefits. It is important to distinguish between early detectable effects and ultimate program impacts. The time thresholds we identified (13-month stabilization, 19-month significant enhancement) represent the earliest points at which policy effects become statistically measurable, not the full realization of long-term health benefits. Our findings should be interpreted as evidence of early policy traction\u0026mdash;demonstrating that health system improvements are beginning to influence disease trajectories\u0026mdash;rather than as a complete assessment of HC2030's ultimate impacts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec40\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Implications for Health Services Research and Policy Evaluation\u003c/h2\u003e \u003cp\u003eThese findings indicate the need to align evaluation timing with the temporal dynamics of policy effects. Health services research has long recognized the critical impact of evaluation timing on conclusions about intervention effectiveness [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Premature evaluation may lead to erroneous rejection of beneficial interventions, and insufficient observation periods may fail to detect lag effects or lead to premature termination of beneficial projects [\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. This longitudinal study revealed that meaningful deceleration in chronic disease accumulation is detectable only after a minimum exposure period, with significant improvements stabilizing after 13 months. This directly informs a stratified evaluation framework: process indicators (0\u0026ndash;4 months) focusing on coverage and service utilization; intermediate indicators (4\u0026ndash;13 months) evaluating behavior change and service integration; and outcome indicators (\u0026ge;\u0026thinsp;13 months) measuring health improvements and disease progression control.\u003c/p\u003e \u003cp\u003eThis empirically anchored framework ensures that evaluations align with the actual trajectory of policy impacts, reducing the likelihood of false negative conclusions and improving evaluation precision. The most recent evaluation of Chinese hospital reform revealed that while structural improvements occurred rapidly, changes in final health outcomes required longer observation periods [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e], which reinforces the importance of multistage evaluation strategies that assess both implementation process indicators and final outcomes at appropriate time intervals.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec41\" class=\"Section2\"\u003e \u003ch2\u003e4.6 Practical Recommendations for Health System Planners\u003c/h2\u003e \u003cp\u003eOn the basis of these findings, we offer the following recommendations for health system planners and policymakers:\u003c/p\u003e \u003cp\u003eMinimum Implementation Period: Healthcare service reforms require at least 13 months of implementation before expecting measurable population health improvements. Governments should avoid hasty policy adjustments or termination decisions on the basis of data from the first year to ensure policy continuity and stability.\u003c/p\u003e \u003cp\u003eCorrect Interpretation of Early Utilization Increases: Utilization rate increases during the early stage (0\u0026ndash;4 months) reflect improved accessibility and detection, rather than policy failure, and should be anticipated and planned accordingly. Health management departments should be prepared for demand surges during this stage, including temporary allocation of medical resources and patient flow management.\u003c/p\u003e \u003cp\u003eSustained Investment Commitment: Service delivery infrastructure investments should be maintained for at least 18\u0026ndash;24 months before a comprehensive evaluation. Policymakers must resist pressure to prematurely reallocate resources on the basis of short-term indicators.\u003c/p\u003e \u003cp\u003eStaged evaluation framework: Implement stratified monitoring strategies, with early focus (0\u0026ndash;4 months) on process indicators and coverage, medium-term (4\u0026ndash;13 months) evaluation of intermediate outcomes and service integration, and later-stage (\u0026ge;\u0026thinsp;13 months) measurement of substantial health outcome improvements.\u003c/p\u003e \u003cp\u003eEquity Monitoring Mechanisms: While this study revealed no significant expansion of health inequalities, continuous monitoring of policy effects across different social groups remains crucial. Routine health equity monitoring systems should be established to promptly identify and correct potential disparities. Special attention should be given to vulnerable subgroups facing multiple barriers to healthcare access. Research examining digital health adoption among patients with chronic diseases has revealed significant gaps between awareness and actual service utilization among older adults [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. Policy implementation strategies should incorporate targeted interventions to accelerate the management of these vulnerable groups.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec42\" class=\"Section2\"\u003e \u003ch2\u003e4.7 Limitations\u003c/h2\u003e \u003cp\u003eOur study has several important limitations:\u003c/p\u003e \u003cp\u003e \u003cem\u003eQuasiexperimental design\u003c/em\u003e. While we validate parallel trends through placebo tests and employ robust inference methods, residual confounding from unmeasured time-varying provincial factors cannot be ruled out entirely. However, the convergence of evidence across three validation strategies (temporal, treatment, and outcome specificity) strengthens causal inference.\u003c/p\u003e \u003cp\u003e \u003cem\u003eSelf-reported disease outcomes\u003c/em\u003e. Chronic disease counts rely on self-reports, potentially conflating true incidence with improved detection ('detection effect'). However, (a) validation studies demonstrate acceptable concordance with medical records [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]; (b) the detection effect represents an important early implementation signal; and (c) our outcome-specific placebo test (null effects on height) suggests that findings are not mere artifacts of increased healthcare interaction.\u003c/p\u003e \u003cp\u003eShort observation window. The 3-year follow-up captures early-to-intermediate effects but not long-term sustainability. Critical unanswered questions include persistence beyond three years, impacts on mortality and disability, and differential effects across specific chronic conditions. Our identified thresholds (7-month stabilization, 19-month improvement) represent when effects become statistically detectable, not ultimate program impacts.\u003c/p\u003e \u003cp\u003e \u003cem\u003eMeasurement limitations\u003c/em\u003e. Physical activity data were excluded because of substantial missing data (~\u0026thinsp;50%), although the included ADL measures serve as functional proxies. Additionally, we cannot separate detection effects from true disease progression\u0026mdash;future research should leverage biomarker data and healthcare utilization records to clarify the mechanisms involved.\u003c/p\u003e \u003cp\u003e \u003cem\u003eGeneralizability\u003c/em\u003e. Our threshold estimates reflect China's unique context: near-universal insurance coverage (90.3%), extensive primary care networks, and specific implementation patterns. Settings with lower coverage or weaker infrastructure may require longer exposure periods for comparable effects.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eEvaluating health system reforms requires aligning assessment timelines with the latent periods of intervention effectiveness. Using the staggered implementation of China's Healthy China 2030 initiative as a natural experiment, we demonstrate that comprehensive chronic disease interventions require approximately 13 months to stabilize disease progression and 19 months to achieve significant population-level improvements. These empirically derived thresholds provide concrete benchmarks for evaluation design, challenging the practice of premature assessment that risks dismissing effective interventions.\u003c/p\u003e \u003cp\u003eOur findings have three critical implications for policymakers and researchers:\u003c/p\u003e \u003cp\u003e(1) Resist premature abandonment: Policy effects undergo a predictable temporal evolution\u0026mdash;from initial detection effects (0\u0026ndash;4 months) through stabilization (13 months) to significant improvement (19\u0026thinsp;+\u0026thinsp;months). Evaluations conducted before these thresholds systematically underestimate program benefits.\u003c/p\u003e \u003cp\u003e(2) Supply-side reforms matter: HC2030's success derived from simultaneously strengthening service capacity (infrastructure, workforce) alongside insurance expansion\u0026mdash;a lesson from previous reforms that achieved high coverage but minimal health gains owing to supply constraints [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e(3) Context-dependent thresholds: Our estimates reflect China's specific institutional context. Generalization requires understanding that systems with weaker baseline infrastructures may need even longer exposure periods.\u003c/p\u003e \u003cp\u003eUltimately, effective health policy evaluation requires aligning assessment timelines with the biological and administrative realities of population health improvement. Only through this combination of patience and rigor can we build health systems that truly improve population health rather than merely expanding coverage.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCHARLS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eThe China Health and Retirement Longitudinal Study\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDiD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDifference-in-Differences\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eHC2030\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eHealthy China 2030\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMultiple Imputation\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eWCB\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eWild Cluster Bootstrap\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e \u003cp\u003eThis study complies with the Declaration of Helsinki. Ethical approval for the China Health and Retirement Longitudinal Study (CHARLS) was obtained from the Institutional Review Board of Peking University (IRB00001052-11015 for the main household survey; IRB00001052-11014 for biomarker collection). All participants provided written informed consent at the time of data collection. This study involved secondary analysis of deidentified CHARLS data obtained under a data-use licence. In accordance with institutional guidelines for research using publicly available anonymized data, no additional ethical approval was required for this secondary analysis.\u003c/p\u003e\u003ch2\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eInformed consent was obtained from all participants by the CHARLS research team prior to data collection. This study is based on anonymized secondary data.\u003c/h2\u003e\u003ch2\u003eCompeting Interests\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis research was funded by the Guangdong Planning Office of Philosophy and Social Science, grant number GD23XSH18. The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAs the sole author, I was responsible for the conception and design of the study, data analysis and interpretation, and manuscript preparation. I have read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors acknowledge the China Health and Retirement Longitudinal Study (CHARLS) team for providing access to the data used in this study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data supporting the findings of this study are available from the China Health and Retirement Longitudinal Study (CHARLS) at https://charls.pku.edu.cn. Access to CHARLS data is open to researchers upon completion of a data use agreement and registration process, in accordance with CHARLS data sharing policies. Provincial policy implementation dates were compiled from official government documents and are provided in Appendix I Table S8 with full source citations. The analytical code used in this study is available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCraig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ. 2008;a1655. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.a1655\u003c/span\u003e\u003cspan address=\"10.1136/bmj.a1655\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHill KG, Woodward D, Woelfel T, Hawkins JD, Green S. Planning for Long-Term Follow-Up: Strategies Learned from Longitudinal Studies. Prev Sci. 2016;17:806\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11121-015-0610-7\u003c/span\u003e\u003cspan address=\"10.1007/s11121-015-0610-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhite IR, Horton NJ, Carpenter J, Statistics RIMAS, Pocock SJ. Strategy for intention to treat analysis in randomised trials with missing outcome data. BMJ. 2011;342(1):d40\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.d40\u003c/span\u003e\u003cspan address=\"10.1136/bmj.d40\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaglio B, Shoup JA, Glasgow RE. The RE-AIM Framework: A Systematic Review of Use Over Time. Am J Public Health. 2013;103:e38\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2105/AJPH.2013.301299\u003c/span\u003e\u003cspan address=\"10.2105/AJPH.2013.301299\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011;104:510\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1258/jrsm.2011.110180\u003c/span\u003e\u003cspan address=\"10.1258/jrsm.2011.110180\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShediac-Rizkallah MC, Bone LR. Planning for the sustainability of community-based health programs: conceptual frameworks and future directions for research, practice and policy. Health Educ Res. 1998;13:87\u0026ndash;108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/her/13.1.87\u003c/span\u003e\u003cspan address=\"10.1093/her/13.1.87\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheirer MA. Is Sustainability Possible? A Review and Commentary on Empirical Studies of Program Sustainability. Am J Evaluation. 2005;26:320\u0026ndash;47. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/1098214005278752\u003c/span\u003e\u003cspan address=\"10.1177/1098214005278752\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandler I, Wolchik SA, Cruden G, Mahrer NE, Ahn S, Brincks A, et al. Overview of Meta-Analyses of the Prevention of Mental Health, Substance Use, and Conduct Problems. Annu Rev Clin Psychol. 2014;10:243\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev-clinpsy-050212-185524\u003c/span\u003e\u003cspan address=\"10.1146/annurev-clinpsy-050212-185524\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLopez Bernal J, Cummins S, Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2016. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/ije/dyw098\u003c/span\u003e\u003cspan address=\"10.1093/ije/dyw098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. :dyw098.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProctor E, Luke D, Calhoun A, McMillen C, Brownson R, McCrary S, et al. Sustainability of evidence-based healthcare: research agenda, methodological advances, and infrastructure support. Implement Sci. 2015;10:88. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13012-015-0274-5\u003c/span\u003e\u003cspan address=\"10.1186/s13012-015-0274-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of complex interventions: Medical Research Council guidance. BMJ. 2015;350:h1258\u0026ndash;1258. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.h1258\u003c/span\u003e\u003cspan address=\"10.1136/bmj.h1258\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. mar19 6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePuska P, Jaini P. The North Karelia Project: Prevention of Cardiovascular Disease in Finland Through Population-Based Lifestyle Interventions. Am J Lifestyle Med. 2020;14:495\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/1559827620910981\u003c/span\u003e\u003cspan address=\"10.1177/1559827620910981\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTsuji I. Current status and issues concerning Health Japan 21 (second term). Nutrition Reviews. 2020;78 Supplement_3:14\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/nutrit/nuaa079\u003c/span\u003e\u003cspan address=\"10.1093/nutrit/nuaa079\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNomura S, Sakamoto H, Ghaznavi C, Inoue M. Toward a third term of Health Japan 21 \u0026ndash; implications from the rise in non-communicable disease burden and highly preventable risk factors. Lancet Reg Health - Western Pac. 2022;21:100377. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lanwpc.2021.100377\u003c/span\u003e\u003cspan address=\"10.1016/j.lanwpc.2021.100377\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990\u0026ndash;2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396:1204\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(20)30925-9\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(20)30925-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990\u0026ndash;2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396:1204\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(20)30925-9\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(20)30925-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarengoni A, Angleman S, Melis R, Mangialasche F, Karp A, Garmen A, et al. Aging with multimorbidity: A systematic review of the literature. Ageing Res Rev. 2011;10:430\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.arr.2011.03.003\u003c/span\u003e\u003cspan address=\"10.1016/j.arr.2011.03.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. 2012;380:37\u0026ndash;43. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(12)60240-2\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(12)60240-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOntario H. Continuity of care to optimize chronic disease management in the community setting: an evidence-based analysis. Ont Health Technol Assess Ser. 2013;13:1\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOrganization WH, editor. others. Integrated care for older people: guidelines on community-level interventions to manage declines in intrinsic capacity. In: Integrated care for older people: guidelines on community-level interventions to manage declines in intrinsic capacity. 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorrall G, Knight J. Continuity of care is good for elderly people with diabetes: retrospective cohort study of mortality and hospitalization. Can Fam Physician. 2011;57:e16\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou W, Guo M, Hu B, Jiang Y, Yao Y. The effect of China\u0026rsquo;s Integrated Medical and Social Care Policy on functional dependency and care deficits in older adults: a nationwide quasi-experimental study. Lancet Healthy Longev. 2025;6:100697. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lanhl.2025.100697\u003c/span\u003e\u003cspan address=\"10.1016/j.lanhl.2025.100697\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Araujo I, Epping-Jordan J, Pot AM, Kelley E, Toro N, Thiyagarajan JA, et al. Organizing integrated health-care services to meet older people\u0026rsquo;s needs. Bull World Health Organ. 2017;95:756\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2471/BLT.16.187617\u003c/span\u003e\u003cspan address=\"10.2471/BLT.16.187617\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdgren L. The meaning of integrated care: a systems approach. Int J Integr Care. 2008;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5334/ijic.293\u003c/span\u003e\u003cspan address=\"10.5334/ijic.293\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValentijn PP, Schepman SM, Opheij W, Bruijnzeels MA. Understanding integrated care: a comprehensive conceptual framework based on the integrative functions of primary care. Int J Integr Care. 2013;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5334/ijic.886\u003c/span\u003e\u003cspan address=\"10.5334/ijic.886\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe State Council of the People\u0026rsquo;s Republic of China. Healthy China 2030 planning outline. 2016. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.gov.cn/zhengce/2016-10/25/content_5124174.htm\u003c/span\u003e\u003cspan address=\"http://www.gov.cn/zhengce/2016-10/25/content_5124174.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 11 Aug 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan X, Liu X, Shao H. Healthy China 2030: A Vision for Health Care. Value Health Reg Issues. 2017;12:112\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.vhri.2017.04.001\u003c/span\u003e\u003cspan address=\"10.1016/j.vhri.2017.04.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeilmann S. Policy Experimentation in China\u0026rsquo;s Economic Rise. St Comp Int Dev. 2008;43:1\u0026ndash;26. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12116-007-9014-4\u003c/span\u003e\u003cspan address=\"10.1007/s12116-007-9014-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu C. The Fundamental Institutions of China\u0026rsquo;s Reforms and Development. J Econ Lit. 2011;49:1076\u0026ndash;151. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/jel.49.4.1076\u003c/span\u003e\u003cspan address=\"10.1257/jel.49.4.1076\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y, Wang R, Jiang M, Ying X. Provincial inequality of China\u0026rsquo;s progress towards universal health coverage: An empirical analysis in 2016\u0026ndash;21. J Glob Health. 2024;14:04122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7189/jogh.14.04122\u003c/span\u003e\u003cspan address=\"10.7189/jogh.14.04122\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNing C, Pei H, Huang Y, Li S, Shao Y. Does the Healthy China 2030 Policy Improve People\u0026rsquo;s Health? Empirical Evidence Based on the Difference-in-Differences Approach. RMHP 2024;Volume 17:65\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2147/RMHP.S439581\u003c/span\u003e\u003cspan address=\"10.2147/RMHP.S439581\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuflo E. Schooling and Labor Market Consequences of School Construction in Indonesia: Evidence from an Unusual Policy Experiment. Am Econ Rev. 2001;91:795\u0026ndash;813. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/aer.91.4.795\u003c/span\u003e\u003cspan address=\"10.1257/aer.91.4.795\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBleakley H. Disease and Development: Evidence from Hookworm Eradication in the American South. Q J Econ. 2007;122:73\u0026ndash;117. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1162/qjec.121.1.73\u003c/span\u003e\u003cspan address=\"10.1162/qjec.121.1.73\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCallaway B, Sant\u0026rsquo;Anna PHC. Difference-in-Differences with multiple time periods. J Econ. 2021;225:200\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jeconom.2020.12.001\u003c/span\u003e\u003cspan address=\"10.1016/j.jeconom.2020.12.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y, Hu Y, Smith JP, Strauss J, Yang G. Cohort Profile: The China Health and Retirement Longitudinal Study (CHARLS). Int J Epidemiol. 2014;43:61\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/ije/dys203\u003c/span\u003e\u003cspan address=\"10.1093/ije/dys203\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOrganization WH. others. Second round of the national pulse survey on continuity of essential health services during the COVID-19 pandemic: January-March 2021: interim report, 22 April 2021. In: Second round of the national pulse survey on continuity of essential health services during the COVID-19 pandemic: January-March 2021: interim report, 22 April 2021. 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu H, Zhao Y, Qiao L, Yang C, Yang Y, Zhang T, et al. Consistency between self-reported disease diagnosis and clinical assessment and under-reporting for chronic conditions: data from a community-based study in Xi\u0026rsquo;an, China. Front Public Health. 2024;12:1296939. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpubh.2024.1296939\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2024.1296939\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan KS, Kasper JD, Brandt J, Pezzin LE, IADL Difficulty Across International Surveys of Aging. Measurement Equivalence in ADL and : Findings From the HRS, SHARE, and ELSA. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2012;67B:121\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/geronb/gbr133\u003c/span\u003e\u003cspan address=\"10.1093/geronb/gbr133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eImbens G. The role of the propensity score in estimating dose-response functions. Biometrika. 2000;87:706\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/biomet/87.3.706\u003c/span\u003e\u003cspan address=\"10.1093/biomet/87.3.706\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanderWeele T. Explanation in causal inference: methods for mediation and interaction. Oxford University Press; 2015.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndersen RM. Revisiting the Behavioral Model and Access to Medical Care: Does it Matter? J Health Soc Behav. 1995;36:1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2137284\u003c/span\u003e\u003cspan address=\"10.2307/2137284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCallaway B, Goodman-Bacon A, Sant\u0026rsquo;Anna PH. Difference-in-differences with a Continuous Treatment. Cambridge, MA: National Bureau of Economic Research; 2024. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3386/w32117\u003c/span\u003e\u003cspan address=\"10.3386/w32117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Chaisemartin C, D\u0026rsquo;Haultfœuille X. Two-way fixed effects and differences-in-differences with heterogeneous treatment effects: a survey. Econometrics J. 2023;26:C1\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/ectj/utac017\u003c/span\u003e\u003cspan address=\"10.1093/ectj/utac017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodman-Bacon A. Difference-in-differences with variation in treatment timing. J Econ. 2021;225:254\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jeconom.2021.03.014\u003c/span\u003e\u003cspan address=\"10.1016/j.jeconom.2021.03.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi F, Li F. Propensity score weighting for causal inference with multiple treatments. Ann Appl Stat. 2019;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1214/19-AOAS1282\u003c/span\u003e\u003cspan address=\"10.1214/19-AOAS1282\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Buuren S. Flexible imputation of missing data. CRC; 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLittle R, Rubin D. Multiple imputation for nonresponse in surveys. Wiley. 1987;10:9780470316696.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCameron AC, Gelbach JB, Miller DL. Bootstrap-Based Improvements for Inference with Clustered Errors. Rev Econ Stat. 2008;90:414\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1162/rest.90.3.414\u003c/span\u003e\u003cspan address=\"10.1162/rest.90.3.414\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacKinnon JG, Webb MD. The wild bootstrap for few (treated) clusters. Econometrics J. 2018;21:114\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ectj.12107\u003c/span\u003e\u003cspan address=\"10.1111/ectj.12107\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAngrist JD, Pischke J-S. Mostly Harmless Econometrics: An Empiricist\u0026rsquo;s Companion. Princeton University Press; 2009. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1515/9781400829828\u003c/span\u003e\u003cspan address=\"10.1515/9781400829828\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J et al. Frailty in Older Adults: Evidence for a Phenotype. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2001;56:M146\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/gerona/56.3.M146\u003c/span\u003e\u003cspan address=\"10.1093/gerona/56.3.M146\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStuck AE, Egger M, Hammer A, Minder CE, Beck JC. Home Visits to Prevent Nursing Home Admission and Functional Decline in Elderly People: Systematic Review and Meta-regression Analysis. JAMA. 2002;287. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1001/jama.287.8.1022\u003c/span\u003e\u003cspan address=\"10.1001/jama.287.8.1022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGehrig S, Zander-Jentsch B, Gutzeit M, Klein S, Rauh J. Estimating the causal effect of a quality assurance program on quality of care in Germany. BMC Health Serv Res. 2025;25:815. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12913-025-12939-8\u003c/span\u003e\u003cspan address=\"10.1186/s12913-025-12939-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagner EH, Austin BT, Davis C, Hindmarsh M, Schaefer J, Bonomi A. Improving Chronic Illness Care: Translating Evidence Into Action. Health Aff. 2001;20:64\u0026ndash;78. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1377/hlthaff.20.6.64\u003c/span\u003e\u003cspan address=\"10.1377/hlthaff.20.6.64\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu X, Huang J, Lv Y, Li G, Peng X. Status of prevalence study on multimorbidity of chronic disease in China: Systematic review. Geriatr Gerontol Int. 2015;15:1\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ggi.12340\u003c/span\u003e\u003cspan address=\"10.1111/ggi.12340\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao N, Chen K. Equity and efficiency of medical and health service system in China. BMC Health Serv Res. 2023;23:33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12913-023-09025-2\u003c/span\u003e\u003cspan address=\"10.1186/s12913-023-09025-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTudor Hart J, THE INVERSE CARE LAW. Lancet. 1971;297:405\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(71)92410-X\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(71)92410-X\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Lancet. 50 years of the inverse care law. Lancet. 2021;397:767. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(21)00505-5\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(21)00505-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLei X, Lin W. The New Cooperative Medical Scheme in rural China: does more coverage mean more service and better health? Health Econ. 2009;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/hec.1501\u003c/span\u003e\u003cspan address=\"10.1002/hec.1501\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoh LH, Siah CJR, Sz\u0026uuml;cs A, Tai ES, Valderas JM, Young D. Integrated patient-centred care for type 2 diabetes in Singapore Primary Care Networks: a mixed-methods study. BMJ Open. 2024;14:e083992. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmjopen-2024-083992\u003c/span\u003e\u003cspan address=\"10.1136/bmjopen-2024-083992\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang Y, Han Z, Nicholas S, Yang W, Maitland E, Shi X, et al. Did China\u0026rsquo;s hospital reforms improve curative care expenditures? Evidence from Beijing hospitals. BMC Health Serv Res. 2025;25:627. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12913-025-12785-8\u003c/span\u003e\u003cspan address=\"10.1186/s12913-025-12785-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiu R, Song R, Wu X, Feng J, Yang Y, Pan Y, et al. From awareness to adoption: a panoramic perspective on the utilization of Internet Medical Services among Chinese patients with chronic disease. BMC Health Serv Res. 2025;25:1415. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12913-025-13601-z\u003c/span\u003e\u003cspan address=\"10.1186/s12913-025-13601-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Health Policy Evaluation, Chronic Disease Management, Difference-in-Differences, Policy Implementation Timing, Elderly Health, China Health Reform, Natural Experiments","lastPublishedDoi":"10.21203/rs.3.rs-8189460/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8189460/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003eHealth policy evaluations frequently overlook delayed or cumulative effects due to inadequate observation periods, leading to the premature rejection of beneficial interventions. The optimal duration of policy exposure necessary to achieve measurable health impacts is not well established. This study examines the time-dependent effects of China's Healthy China 2030 (HC2030) initiative on chronic disease accumulation among older adults, utilizing staggered provincial implementation as a natural experiment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003eLongitudinal data from 7,487 adults aged 60 years and older in the China Health and Retirement Longitudinal Study (CHARLS, 2011–2018) across 28 provinces were analysed. Staggered HC2030 implementation timing (0–21 month variation) was leveraged within a difference-in-differences framework. Provinces were grouped by exposure duration (short-, medium-, or long-term), and group-specific disease progression trajectories were estimated for each group. Overlap weighting, baseline covariate balancing, and wild cluster bootstrapping were employed to ensure robust inference with a limited number of clusters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003ePrepolicy placebo tests confirmed the validity of the parallel trends assumption. The policy effects differed by exposure duration: short-term (≤15 months) and medium-term (15–18 months) exposures did not yield significant impacts, whereas divergence in disease progression trajectories emerged at approximately 13 months, indicating stabilization. Long-term exposure (≥19 months) significantly decelerated chronic disease accumulation (β = -0.325, 95% , p = 0.005, WCB), offsetting a substantial portion of the age-related disease burden (population mean increase: 0.45). These results indicate that sustained policy implementation is necessary to achieve meaningful health system benefits.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003eComprehensive health system reforms require extended implementation periods, with approximately 13 months needed to stabilize disease progression and sustained exposure (≥19 months) necessary to achieve significant cumulative effects (β = -0.325, p = 0.005) before measurable population health benefits are observed. Premature evaluations risk dismissing effective interventions. These threshold estimates provide empirical benchmarks for designing evaluation timelines in chronic disease policy research, particularly in low- and middle-income settings.\u003c/p\u003e","manuscriptTitle":"Policy threshold for decelerating chronic disease accumulation among older Chinese adults: Evidence from the Healthy China 2030 Initiative","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-22 04:27:28","doi":"10.21203/rs.3.rs-8189460/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"656473c1-340c-4993-99ca-50466e1c9d1a","owner":[],"postedDate":"December 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-31T06:56:56+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-22 04:27:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8189460","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8189460","identity":"rs-8189460","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}