Validation of the Hospital Frailty Risk Score in China

Validation of the Hospital Frailty Risk Score in China | 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 Validation of the Hospital Frailty Risk Score in China Yue Qiu, Weiqing Xiong, Xinyue Fang, Pei Li, Simon Conroy, Laia Maynou, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5551082/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Purpose To validate the Hospital Frailty Risk Score (HFRS) in Chinese hospital settings, describing how patients are allocated to frailty risk groups and how frailty risk is associated with length of stay (LoS) and hospital costs. Design: Retrospective observational study. Setting: 48 hospitals in Lvliang City, Shanxi Province, China. Subjects: Patients aged 75 years or older hospitalised between 1 January 2022 and 31 December 2023 (n = 34,731). Methods A logistic regression model examined the association between long length of stay (LoS) and frailty risk. A generalised linear model assessed the association between hospital costs and frailty risk. Subgroup analyses of age group, sex, and hospital tiers were conducted. Results 22.2% of patients were categorised as having zero risk, 62.4% as low risk, 15.3% as intermediate risk, and 0.08% as high risk. Compared to the zero risk group: for those with low risk the probability of long LoS was 1.92 (95% CI 1.79–2.06) times higher and hospital costs were ¥1,926 (95% CI 1,655-2,197) higher; for those with intermediate risk, the probability of long LoS was 2.7 (95% CI 2.49–2.96) times higher and hospital costs were ¥4,284 (95% CI 3,916-4,653) higher; and for those with high risk the probability of long LoS was 6.7 (95% CI 3.06–14.43) times higher and hospital costs were ¥16,613 (95% CI 12,827 − 20,399) higher. The explanatory power of the HFRS held across subgroups. Conclusions Compared to patients aged 75 + elsewhere, those in China had lower frailty risk scores, likely reflecting a younger age structure and recording of fewer diagnosis codes. Even so, the HFRS is a powerful predictor of long length of stay and hospital costs in China. Hospital Frailty Risk Score HFRS frailty risk assessment length of stay hospital costs Figures Figure 1 Figure 2 Figure 3 Key Summary Points Aim To validate the Hospital Frailty Risk Score (HFRS) in Chinese hospital settings Findings Relatively few older people hospitalised in China are categorised using the HFRS as having high frailty risk The HFRS predicts long length of stay and hospital costs among hospitalised older patients in China Message The HFRS has potential for widespread use in both developed and developing countries that use ICD-10 codes Introduction Health issues among older people are a significant concern in aging societies, impacting individual quality of life and the sustainability of healthcare systems. Frailty is defined as an “age-related, clinically identifiable state of diminished physiologic reserve and increased vulnerability to a broad range of adverse health outcomes” [1]. The associations between frailty and length of stay, hospital costs, and mortality has been confirmed in many studies [2]. Frailty significantly contributes to functional deterioration in older people [3], increasing healthcare demands and burdening families and society. Global recognition of the need to assess and manage frailty has led to the development of various assessment tools [4; 5; 6; 7; 8; 9; 10; 11]. By identifying frailty-related risks, particularly in the first hours of an acute admission [12; 13; 14; 15], it may be possible to mitigate harms such as reduced mobility, which is associated with death or institutionalisation [16], loss of lean muscle mass, pressure ulcer development [17], incontinence, thrombosis, constipation, pain, infections and low mood [18]. However, many tools are too complex for use in acute care settings [19] and require professional evaluations, complicating their integration into clinical care [20; 21]. To address this, Gilbert and colleagues developed the Hospital Frailty Risk Score (HFRS), which uses International Statistical Classification of Diseases and Related Health Problems, tenth revision (ICD-10) codes from electronic health records for hospitalised patients to compute frailty risk scores without additional data gathering [22]. Since its development, the HFRS has garnered extensive attention, with applications to different types of patients to validate its predictive capacity for outcomes such as length of stay, mortality, and treatment costs [23; 24; 25; 26; 27; 28; 29; 30; 31]. These validation studies, however, have largely been conducted in developed countries, with few applications from developing ones. To start addressing this gap, this paper applies the HFRS to older patients admitted to hospital in China. Several studies have measured the prevalence of frailty in the Chinese community, using instruments such as the Frailty Index [32] and a validated physical frailty phenotype (PFP) scale [33]. A meta-analysis showed that 15% of Chinese community dwelling older people aged 75–84 experienced frailty, while this figure rose to 25% for those aged 85 and over [34]. As a rapidly aging developing country, there is a great need to identify and manage frailty in older people in China. But China has few healthcare workers relative to the size of the population (2.4 physicians and 3.3 nurses and midwives per 1000 [35]), which rules out widespread application of many frailty assessment tools that require primary data collection. Using electronic health records to identify frail patients offers a more cost-effective approach in China. This study aimed to assess the applicability of the HFRS for hospitalized patients 75 years and older in China, with a focal emphasis on the effectiveness of frailty identification and the tool’s ability to predict healthcare resource consumption. The findings will provide evidence on whether China can adopt the HFRS or if a new tool needs to be developed based on China’s specific conditions. Methods Setting Health care in China combines a mix of public and privately funded care [36]. Over 95% of the population is insured through either the Urban Employee Basic Medical Insurance (UEBMI) scheme, which provides coverage for urban employees, or through the Urban and Rural Residence Basic Medical Insurance (URRBMI) scheme which covers those without formal employment in urban and rural settings. These insurance schemes offer similar benefits and together account for 46% of total health expenditure, with 26.7% coming from government sources and 27.3% from individuals, mainly as out-of-pocket payments [37]. We conducted a retrospective study of patients aged 75 and older who had been hospitalised between 1 January 2022 and 31 December 2023 in Lvliang City, Shanxi Province, China. Lvliang City has a middle-ranged economic level among the cities of China and, according to the 7th China population census [38], has a permanent population of about 3.3 million, of which 3.9% are aged 75 years and above, and 0.8% are aged 85 years and above. It is a relatively young city as those proportions for the whole country are 4.8% and 1.1% respectively [38]. Data We used a medical database to retrieve anonymised patient-level data. The entire database contained routine administrative medical data, including clinical characteristics and healthcare utilisation, for 357,186 patients admitted to 72 primary hospitals, 45 secondary hospitals, and three tertiary hospitals in Lvliang City from 1 January 2022 to 31 December 2023. In the hospital classification of China, primary hospitals are community facilities with fewer than 100 beds that provide preventive care, primary care and rehabilitation. Secondary hospitals have between 100 and 500 beds, are located in cities and offer broad services and some specialities. Tertiary hospitals are located in major cities and offer a wide range of specialists and have more than 500 beds. In tertiary hospitals, staff tend to be more qualified and equipment more advanced so complexity and quality should be higher in these hospitals [39]. The process of selecting the analytical sample is described in Fig. 1 . We extracted information for all 38,932 patients aged 75 years and over admitted to secondary and tertiary hospitals, dropping patients admitted to primary hospitals for consistency with Gilbert et al [22]. For patients with multiple admissions, only the most recent admission was used for the analysis. After excluding 4,201 (10.8%) patients with missing data, the analytical sample comprised 34,731 patients. We compared key variables before and after excluding patients with missing data for other variables, reported in Appendix A1. This revealed no significant differences except for hospital costs. Ethics approval was obtained from the Institutional Review Board of Tsinghua University (THU01-20240115). Outcome and frailty measures We analysed the relationship between the HFRS and two outcomes, long LoS and hospital costs, controlling for other characteristics: We used the patient’s admission and discharge dates to calculate LoS. Consistent with the original HFRS study by Gilbert et al [22], this was converted into a binary variable where 1 indicated LoS over 10 days, 0 otherwise and analysed using a logistic regression model. As a robustness check we also analysed LoS using a Poisson model, reported in Appendix A5. The cost of the hospital stay is recorded in the patient’s medical record, this being the sum of the patient’s treatment costs for the admission, including the cost of examinations, tests, medicines, consumables, surgical treatment, and medical services. Costs were analysed using a generalised linear model. For each patient the HFRS was calculated by combining a weighted set of 109 3-character ICD-10 diagnostic codes [22]. Given that only two years’ worth of data were available we constructed the HFRS ( a ) form, using diagnostic information from the current admission only [30]. The HFRS takes values from 0 to 173.2. Commonly patients have been categorised as having low (HFRS < 5), intermediate ((5 ≤ HFRS < 15), or high (HFRS ≥ 15) frailty risk [22]. However, in this study a large proportion of patients had a HFRS of zero and, hence, we categorised patients into four groups: zero frailty risk (HFRS = 0), low frailty risk (0 < HFRS < 5), intermediate frailty risk (5 ≤ HFRS < 15), or high frailty risk (HFRS ≥ 15). Those in the zero frailty risk category formed the reference group in the regressions. The analyses controlled for patient age, sex, Charlson comorbidity index (CCI) [40; 41; 42; 43], admission via the emergency department, the number of operations performed, hospital type, and the patient’s insurance scheme. The Appendix details the construction of these control variables and provides specification details about the regression models. Subgroup analyses We conducted subgroup analyses to uncover the relationship between frailty risk and patient outcomes among patients with different characteristics. The subgroups were divided according to age (75–79, 80–84, 85+), sex (male and female), and hospital tiers (secondary and tertiary). The regression models and the control variables in the subgroup analyses were the same as those employed in the analysis of all patients, only excluding the variable used to form the subgroup. Results Descriptive statistics The analytical sample comprised 34,731 patients aged 75 and above. Descriptive statistics are shown in Table 1. Among these patients, the maximum HFRS score was 19.8 points, and 7,715 (22.2%) patients were categorised as having zero frailty risk, 21,667 (62.4%) as low frailty risk, 5,320 (15.3%) as intermediate frailty risk, and only 29 (0.08%) as high frailty risk. Generally, patients with higher frailty risk were more likely to stay in hospital for more than 10 days, had higher costs, were more likely to be admitted through the ED, and had a higher CCI score. Table 1: Descriptive statistics Descriptive Statistics (N = 34731) Variables Zero risk (n=7715,22.21%) Low risk (n=21,667, 62.39%) Intermediate risk (n= 5,320, 15.32%) High risk (n=29, 0.08%) N / Mean Proportion/ SD N / Mean Proportion/ SD N / Mean Proportion/ SD N / Mean Proportion/ SD Long length of stay Yes 1,178 15.27% 5,443 25.12% 1,832 34.44% 18 62.07% No 6,537 84.73% 16,224 74.88% 3,488 65.56% 11 37.93% Hospital costs ¥ 5,931 6,104 7,387 8,766 11,596 21,251 28,324 48,904 Age group 75-79 4,099 53.13% 10,754 49.63% 2,240 42.11% 14 48.28% 80-84 2,326 30.15% 6,796 31.37% 1,821 34.23% 10 34.48% 85+ 1,290 16.72% 4,117 19.00% 1,259 23.67% 5 17.24% Sex Male 3,941 51.08% 10,839 50.03% 2,704 50.83% 15 51.72% Female 3,774 48.92% 10,828 49.97% 2,616 49.17% 14 48.28% Hospital tier Secondary 6,075 78.74% 15,843 73.12% 2,983 56.07% 11 37.93% Tertiary 1640 21.26% 5,824 26.88% 2,337 43.93% 18 62.07% Admission through ED Yes 531 6.88% 2,076 9.58% 772 14.51% 7 24.14% No 7,184 93.12% 19,591 90.42% 4,548 85.49% 22 75.86% Charlson Comorbidity Index (CCI) 0 6,861 88.93% 18,231 84.14% 4,195 78.85% 16 55.17% 1 772 10.01% 3,060 14.12% 973 18.29% 10 34.48% 2+ 82 1.06% 376 1.74% 152 2.86% 3 10.34% Number of operations 0.429 0.707 0.283 0.736 0.526 1.102 1.207 2.007 Basic medical insurance type URBMI UEBMI 6,262 1,453 81.17% 18.83% 17,396 4,271 80.29% 19.71% 4,117 1,203 77.39% 22.61% 19 10 65.52% 34.48% Reimbursement rate 0.69 0.162 0.686 0.159 0.672 0.166 0.722 0.153 Regression results The full regression results of for the analyses of long LoS and hospital cost are reported in Appendix Table A2 and summarised as a forest plot in Figure 3. Compared to the zero frailty risk group, the probability of long LoS was: 1.92 (95% confidence interval (CI) 1.79-2.06) times higher for those with low frailty risk; 2.71 (95% CI 2.49-2.96) times higher for those with intermediate frailty risk; and 6.65 (95% CI 3.06-14.43) times higher for those with high frailty risk. The wide CI for the high frailty group reflects the small number of patients in this group. The results of applying the Poisson model to analyse LoS are consistent with the logistic model, as reported in Appendix A5. Hospital costs also increased in line with the frailty risk level. Compared to those with zero frailty risk, costs were ¥1,926 (95% CI 1,655-2,197) higher for those with low frailty risk, ¥4,284 (95%CI 3,916-4,653) higher for those with intermediate frailty risk, and ¥16,613 (95% CI 12,827-20,399) higher for those with high frailty risk. Most control variables revealed significant effects for long LoS or hospital costs. For instance, LoS and hospital costs increased with the number of operations (long LoS OR 1.28 95% CI 1.24-1.32; hospital costs ¥4,955 95% CI 4,817-5,093). In the analysis of long LoS, the odd ratios for CCI were significant but less than 1 (OR 0.91 95% CI 0.85-0.98) for CCI = 1 and insignificant for CCI =2+. For hospital costs, the CCI coefficients were positive (CCI=1 ¥882 95% CI 564-1,202; CCI=2+ ¥786 95% CI -49-1,622). Subgroup analyses The regression results for length of stay and hospital costs across subgroups are summarized in Appendix A6. Across all subgroups, compared to those with zero frailty risk, both LoS and costs increased across the frailty risk categories. For those with low frailty risk, the differences were small; for those in the intermediate frailty risk category the differences are larger and significant; for those in the high frailty risk category there are wide confidence intervals around the point estimates because of the small numbers in this category. These patterns of longer LoS and higher costs as frailty risk increases demonstrate that the HFRS can be applied across age categories, sex, and hospital tiers, thereby underscoring its usefulness as a predictive tool for patients hospitalised in China. Discussion This paper had two objectives. Firstly, to validate the HFRS in Chinese hospital settings population, describing how patients are allocated to frailty risk groups. Secondly to assess how frailty risk is associated with LoS and hospital costs. As regards the first objective, in the Lvliang hospital care setting, 7,715 (22.2%) individuals were identified as having zero frailty risk, 21,667 (62.4%) patients were categorised to the low frailty risk group, 5,320 (15.3%) into the intermediate frailty risk group, and 29 (0.08%) into the high frailty risk group. Notably, very few were identified as being high risk, especially when compared with much higher proportions in the high risk group in two studies from England (20% in [22] and in 21.8% [30]) and of 17% in a study from France [29]. However, other studies of those over 75 years also report small proportions in the high risk group: 2.9% in a study from Switzerland [24], 2.6% in one from Canada [23] and 1.9% in a study from Australia [44]. The HFRS for any particular individual is driven by whether they have one of the 109 ICD-10 codes and the weight attached to that code, these weights ranging from 0.1 to 7.1 (F00 Dementia in Alzheimer’s disease). The proportions of the sample with the 30 ICD-10 codes with a weight > 2.0 are shown in Fig. 2 and Table 2 . These proportions are compared to patients aged 75 and above hospitalised in England between 2013 to 2019 [45]. Table 2 Distribution of ICD-10 codes with HFRS weight > 2.0 ICD-10 code ICD-10 description Weight China sample England sample F00 Dementia in Alzheimer’s disease 7.1 0.04% 3.77% G81 Hemiplegia 4.4 0.61% 2.01% G30 Alzheimer’s disease 4.0 0.20% 5.62% I69 Sequelae of cerebrovascular disease 3.7 11.75% 2.50% R29 Tendency to fall 3.6 0.00% 18.13% F05 Delirium, not by alcohol or psychoactive substances 3.2 0.07% 6.15% N39 Disorders of urinary system, including UTI & urinary incontinence 3.2 4.79% 18.89% S00 Superficial injury of head 3.2 0.23% 3.23% W19 Unspecified fall 3.2 0.07% 8.69% R31 Unspecified haematuria 3.0 0.20% 2.22% B96 Other bacterial agents as the cause of diseases 2.9 0.01% 6.68% R41 Other cognitive functions and awareness symptoms and signs 2.7 0.01% 7.70% I67 Other cerebrovascular diseases 2.6 13.55% 7.91% R26 Abnormalities of gait and mobility 2.6 0.01% 9.01% R56 Convulsions 2.6 0.04% 1.15% R40 Somnolence, stupor and coma 2.5 0.23% 0.80% S06 Intracranial injury 2.4 0.25% 0.88% T83 Complications of genitourinary prosthetic devices, implants, grafts 2.4 0.05% 1.19% E86 Volume depletion 2.3 0.10% 7.47% E87 Other disorders of fluid, electrolyte and acid-base balance 2.3 19.40% 11.69% M25 Other joint disorders 2.3 0.26% 3.58% S42 Fracture of shoulder and upper arm 2.3 0.13% 1.13% R54 Senility 2.2 0.04% 3.50% F03 Unspecified dementia 2.1 0.27% 12.27% W18 Other fall on same level 2.1 0.02% 3.21% Z50 Care involving use of rehabilitation procedures 2.1 0.00% 5.44% F01 Vascular dementia 2.0 0.20% 4.87% L03 Cellulitis 2.0 0.05% 4.66% S80 Superficial injury of lower leg 2.0 0.01% 0.95% Z75 Problems related to medical facilities and health care 2.0 0.00% 2.66% Clear differences are evident. Less than 1% of the Lvliang sample had one of the three ICD-10 codes (F00, G81, G30) with the highest weight; in England 11.4% of the sample had one of these codes. In the Lvliang dataset, the highest proportion is attributed to the ICD-10 code for “E87 Other disorders of fluid, electrolyte and acid-base balance” (19.4%), followed by “I67 Other cerebrovascular disease” (13.6%) and “I69 Sequelae of cerebrovascular disease” (11.8%). In the English dataset, the distribution of the 30 ICD-10 codes is more spread out, with the highest proportions attributed to “N39 Disorders of urinary system including UTI & urinary incontinence” (18.9%) and “R29 Tendency to fall” (18.1%). Comparing these two samples, it is evident that different ICD-10 codes identify the frailty risk of patients in the two countries. The most direct reason for the differences between the Chinese and English data is the lower count of the 109 ICD-10 codes used to construct the HFRS in the Lvliang dataset, particularly for codes with large HFRS weights such as Dementia (F00) [46] and Alzheimer’s disease (G30) [47]. The lower frequency of these codes in China can be attributed partly to the younger age structure and its associated disease spectrum. Compared to 42.5% of the population aged 85 years or older reported in a regional study in England [30], Lvliang’s patients are relatively young, with only 19.2% of the population aged 85 years or older. Due to this younger age structure, diseases like dementia and Alzheimer’s, which are more prevalent in older age groups, have lower incidence rates in Lvliang. Consequently, the frequency of these ICD-10 codes is also lower, resulting in lower risk scores when calculating the HFRS. Besides, the limited availability of well-trained medical records staff might mean that diagnoses are under-coded in China [48; 49]. Disparities in healthcare delivery capacity between countries and regions may also play a role in explaining the differences in the proportions of patients in each frailty risk group [50]. Around 2.5% of patients in China seek care outside the region in which they live [51], the primary reason being to access higher quality care [52]. Patients from Lvliang might seek high quality healthcare in nearby larger cities like Datong or Beijing [53]. These factors will mean that hospitals will be treating quite different patient profiles, implying that the HFRS is influenced by factors other than frailty related diagnostic codes [24] and underscoring a need for further investigation into the underlying causes as well as possible coding differences [44]. Note that, as well as impacting the HFRS, coding differences also have implications for the CCI, where the proportion of patients with a CCI ≥ 2 was 1.76% in our study but 51.6% in the study in the England [30]. Low CCI scores were also reported in a study of older patients admitted to hospital in Beijing [54]. Regarding the second objective, despite differences in patient profiles and diagnostic coding, the HFRS still emerged as a strong explanator or length of stay and costs for patients admitted to hospital in Lvliang city. Patients in low, intermediate, and high HFRS risk groups were significantly and progressively more likely to stay in hospital for over 10 days and have higher treatment costs compared to patients in the zero risk group. If diagnostic coding were improved, frailty risk scores would be higher and the explanatory power of the HFRS would become greater. Even so, the HFRS had greater explantory power for these two outcomes than any of the other variables included in the regression analyses. Subgroup analyses further confirmed the robustness of these findings. For long LoS, our regression results were similar to other HFRS validation studies conducted in different countries [22; 23; 55; 24; 56; 28; 30; 29]. The influence of the HFRS on hospital costs was also consistent with findings from other studies [24; 28]. Our validation study suggests that, despite variations in healthcare system and ICD-10 coding rules across countries, the risk of frailty calculated using the HFRS methodology is a useful predictor of length of stay and hospital costs in China, as has been found elsewhere [57]. Therefore, the HFRS holds great potential for widespread use in countries using ICD-10 codes, both in developed and developing countries, due to its explanatory power, convenience and cost-effectiveness. There are limitations to this study. First, the narrow data window only allowed analysis of the patient’s most recent admission, which will have led to a lower HFRS compared to other studies that also include diagnostic information from the previous two admissions within the last two years, as recommended [30]. This means that, on the one hand, diagnoses recorded in a patient’s previous admissions are not captured but, on the other hand, the HFRS is constructed in a consistent fashion for every patient in the study. Second, diagnoses may have been under-coded. If so, the explanatory power of the HFRS would have been under-estimated. Third, a high proportion of patients were omitted from analysis due to missing data. With the exception of costs, data appeared to be missing at random thereby suggesting that the analytical sample remained representative. Nevertheless the high proportion highlights the scope for improved coding practice, not just of diagnoses but more generally. Fourth, we did not analyse the relationship between the HFRS and in-hospital death because, due to Chinese cultural practices, most older patients prefer to receive end-of-life care at home with family and friends [58]. Reflecting this, only 0.37% died in hospital. Nor did the data allow us to identify those who were discharged to die at home. Fifth, this study utilized regional data from a single city in China, which may not be representative of the entire country though it may be fairly typical of other middle-ranged cities with similar socio-economic characteristics. Future studies of the HFRS using data from other areas in China would be welcome. Conclusions As the population ages, particularly in low-income and middle-income countries, the impact of frailty will escalate [2]. Frailty risk, easily calculated using the HFRS, offers benefits at the micro, meso and macro level of the system. At the micro level, involving clinician-patient interaction, a measure of frailty risk alerts the clinician to the potential prognosis. People with high HFRS scores have an increased risk of dying in hospital – this might prompt the clinician to activate critical care if clinically appropriate and in keeping with the patient’s wishes and preferences. Alternatively, it might lead to a more palliative or supportive paradigm of care being instituted, following assessment of the individual. Fundamentally assessment of frailty risk moves us away from a one-size-fits-all approach, recognising that patients are at different stages on their life trajectory. At the meso level, it allows hospitals to match resources to patient needs. For example, it might be that people with high HFRS scores are found in all parts of the hospital, prompting the development of a geriatric liaison service and evaluating its impact on the available service metrics. At the macro level, it facilitates the creation of registries, which can track the flow of a risk stratified cohort of older people along care pathways and redirect where appropriate. For example, a dynamic frailty risk registry might highlight that a patient with high frailty discharged from hospital has not been referred to community services. This can then be rectified by putting post-discharge support in place. Our study is the first to confirm the predictive effect of HFRS on length of stay and hospital costs in China, a developing country with a growing older population. The HFRS strikes a balance between broad applicability and low cost through its big data driven approach. However, the identification of frailty risk by HFRS in this study was significantly different from the original English population cohorts by Gilbert et al [22], likely due to the differences in age structure, disease spectrum, healthcare delivery capacity and diagnostic coding practices between England and China. In light of these findings, developing countries like China might benefit from employing the HFRS but also from using a similar big data-driven approach to develop localised frailty screening tools, tailored to reflect their particular the demographic and healthcare landscapes [59]. Investment in recruitment and training of coding staff should improve diagnostic coding practices and data quality. In places with the necessary infrastructure, pilot implementation of HFRS in clinical workflows could be a good way to inform how construction of the HFRS might be refined and how it might best be applied to assess frailty risk among hospitalised patients in other developing countries. Declarations Acknowledgements YQ was supported by the Du Shi Special Fund (Grant No. 2024Z11DSZ001) from Tsinghua University, which supports promising faculties dedicated to fundamental research. AS received funding support from LSE’s Global Research Fund. SC, LM, KR and AS received funding support from the National Institute for Health and Care Research (NIHR 203451 - Understanding the Trajectory of Frailty Across the Life Course). The authors thank the journal’s reviewers, Mengxi Pang and Susana Mourato. Data availability statement Chinese data presented in this study may be obtained from a third party on request from the authors and are not publicly available. The English data came from the Hospital Episode Statistics provided by NHS Digital under Data Sharing Agreement NIC-354497-V2J9P. These data may be obtained from a third party and are not publicly available. This paper has been screened to ensure no confidential information is revealed. Conflict of interest statement On behalf of all authors, the corresponding author states that there is no conflict of interest. References Kim DH, Rockwood K. Frailty in older adults. New England Journal of Medicine. 2024;391(6):538-48. Hoogendijk EO, Afilalo J, Ensrud KE, Kowal P, Onder G, Fried LP. Frailty: implications for clinical practice and public health. The Lancet. 2019;394(10206):136575. Dent E, Morley J, Cruz-Jentoft A, Woodhouse L, Rodríguez-Mañas L, Fried L, et al. Physical frailty: ICFSR international clinical practice guidelines for identification and management. The Journal of nutrition, health and aging. 2019;23(9):771-87. Clegg A, Bates C, Young J, Ryan R, Nichols L, Ann Teale E, et al. Development and validation of an electronic frailty index using routine primary care electronic health record data. Age and ageing. 2016;45(3):353-60. De Vries N, Staal J, Van Ravensberg C, Hobbelen J, Rikkert MO, Nijhuis-van der Sanden M. Outcome instruments to measure frailty: a systematic review. Ageing research reviews. 2011;10(1):104-14. Dent E, Kowal P, Hoogendijk EO. Frailty measurement in research and clinical practice: a review. European journal of internal medicine. 2016;31:3-10. Gregorevic KJ, Hubbard RE, Katz B, Lim WK. The clinical frailty scale predicts functional decline and mortality when used by junior medical staff: a prospective cohort study. BMC geriatrics. 2016;16:1-6. Lansbury LN, Roberts HC, Clift E, Herklots A, Robinson N, Sayer AA. Use of the electronic Frailty Index to identify vulnerable patients: a pilot study in primary care. British Journal of General Practice. 2017;67(664):e751-6. Brundle C, Heaven A, Brown L, Teale E, Young J, West R, et al. Convergent validity of the electronic frailty index. Age and ageing. 2019;48(1):152-6. Boyd PJ, Nevard M, Ford JA, Khondoker M, Cross JL, Fox C. The electronic frailty index as an indicator of community healthcare service utilisation in the older population. Age and Ageing. 2019;48(2):273-7. Elliott A, Taub N, Banerjee J, Aijaz F, Jones W, Teece L, et al. Does the clinical frailty scale at triage predict outcomes from emergency care for older people? Annals of emergency medicine. 2021;77(6):620-7. Bernstein SL, Aronsky D, Duseja R, Epstein S, Handel D, Hwang U, et al. The effect of emergency department crowding on clinically oriented outcomes. Academic Emergency Medicine. 2009;16(1):1-10. Pines JM, Pollack Jr CV, Diercks DB, Chang AM, Shofer FS, Hollander JE. The association between emergency department crowding and adverse cardiovascular outcomes in patients with chest pain. Academic Emergency Medicine. 2009;16(7):617-25. Platts-Mills TF, Owens ST, McBride JM. A modern-day purgatory: older adults in the emergency department with nonoperative injuries. Journal of the American Geriatrics Society. 2014;62(3):525-8. Carter EJ, Pouch SM, Larson EL. The relationship between emergency department crowding and patient outcomes: a systematic review. Journal of Nursing Scholarship. 2014;46(2):106-15. Brown CJ, Friedkin RJ, Inouye SK. Prevalence and outcomes of low mobility in hospitalized older patients. Journal of the American Geriatrics Society. 2004;52(8):1263-70. Coleman S, Gorecki C, Nelson EA, Closs SJ, Defloor T, Halfens R, et al. Patient risk factors for pressure ulcer development: systematic review. International journal of nursing studies. 2013;50(7):974-1003. Creditor MC. Hazards of hospitalization of the elderly. Annals of internal medicine. 1993;118(3):219-23. Elliott A, Hull L, Conroy SP. Frailty identification in the emergency department—a systematic review focussing on feasibility. Age and ageing. 2017;46(3):509-13. Rockwood K, Song X, MacKnight C, Bergman H, Hogan DB, McDowell I, et al. A global clinical measure of fitness and frailty in elderly people. Cmaj. 2005;173(5):489-95 Sternberg SA, Schwartz AW, Karunananthan S, Bergman H, Mark Clarfield A. The identification of frailty: a systematic literature review. Journal of the American Geriatrics Society. 2011;59(11):2129-38. Gilbert T, Neuburger J, Kraindler J, Keeble E, Smith P, Ariti C, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. The Lancet. 2018;391(10132):1775-82. McAlister F, van Walraven C. External validation of the Hospital Frailty Risk Score and comparison with the Hospital-patient One-year Mortality Risk Score to predict outcomes in elderly hospitalised patients: a retrospective cohort study. BMJ quality & safety. 2019;28(4):284-8. Eckart A, Hauser SI, Haubitz S, Struja T, Kutz A, Koch D, et al. Validation of the hospital frailty risk score in a tertiary care hospital in Switzerland: results of a prospective, observational study. BMJ open. 2019;9(1):e026923. Kwok CS, Zieroth S, Van Spall HG, Helliwell T, Clarson L, Mohamed M, et al. The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: Frailty and outcomes in heart failure. International Journal of Cardiology. 2020;300:184-90. Bruno RR, Wernly B, Flaatten H, Schölzel F, Kelm M, Jung C. The hospital frailty risk score is of limited value in intensive care unit patients. Critical Care. 2019;23:1-2. Kundi H, Wadhera RK, Strom JB, Valsdottir LR, Shen C, Kazi DS, et al. Association of frailty with 30-day outcomes for acute myocardial infarction, heart failure, and pneumonia among elderly adults. JAMA cardiology. 2019;4(11):1084-91. Bonjour T, Waeber G, Marques-Vidal P. Trends in prevalence and outcomes of frailty in a Swiss university hospital: a retrospective observational study. Age and Ageing. 2021;50(4):1306-13. Gilbert T, Cordier Q, Polazzi S, Bonnefoy M, Keeble E, Street A, et al. External validation of the hospital frailty risk score in France. Age and ageing. 2022;51(1):afab126. Street A, Maynou L, Gilbert T, Stone T, Mason S, Conroy S. The use of linked routine data to optimise calculation of the Hospital Frailty Risk Score on the basis of previous hospital admissions: a retrospective observational cohort study. The Lancet Healthy Longevity. 2021;2(3):e154-62. Turcotte LA, Heckman G, Rockwood K, Vetrano DL, H´ebert P, McIsaac DI, et al. External validation of the hospital frailty risk score among hospitalised home care clients in Canada: a retrospective cohort study. Age and ageing. 2023;52(2):afac334. Ma L, Tang Z, Zhang L, Sun F, Li Y, Chan P. Prevalence of frailty and associated factors in the community-dwelling population of China. Journal of the American Geriatrics Society. 2018;66(3):559-64. Wu C, Smit E, Xue QL, Odden MC. Prevalence and correlates of frailty among community-dwelling Chinese older adults: the China health and retirement longitudinal study. The Journals of Gerontology: Series A. 2018;73(1):102-8. He B, Ma Y, Wang C, Jiang M, Geng C, Chang X, et al. Prevalence and risk factors for frailty among community-dwelling older people in China: a systematic review and meta-analysis. The Journal of nutrition, health and aging. 2019;23(5):442-50. World Bank Group. Physicians, nurses and midwives (per 1000); 2025. Accessed: (7 February 2025). https://data.worldbank.org/indicator/SH.MED.PHYS.ZS and https://data.worldbank.org/indicator/SH.MED.NUMW.P3. Li X, Fan L, Leng SX. The aging tsunami and senior healthcare development in China. Journal of the American Geriatrics Society. 2018;66(8):1462-8. National Health Commission of the People’s Republic of China. Statistical Bulletin on the Development of China’s Health Undertakings in 2023; 2024. Accessed: (7 February 2025). http://www.nhc.gov.cn/ guihuaxxs/s3585u/202408/6c037610b3a54f6c8535c515844fae96/files/ 58c5d1e9876344e5b1aa5aa2b083a51a.pdf. Office of the leading group of the state council for the seventh national population census. China population yearbook 2020. China Statistics press; 2022. Shi H, Fan M, Zhang H, Ma S, Wang W, Yan Z, et al. Perceived health-care quality in China: a comparison of second- and third-tier hospitals. International Journal for Quality in Health Care. 2021 02;33(1):mzab027. Available from: https:// doi.org/10.1093/intqhc/mzab027. Charlson M, Pompei P, Ales K, MacKenzie C. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-83. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Medical Care. 2005;43(11):1130-9. Bannay A, Chaignot C, Blotière P, Basson M, Weill A, Ricordeau P, et al. The Best Use of the Charlson Comorbidity Index With Electronic Health Care Database to Predict Mortality. Medical Care. 2016;54(2):188-94 Toson B, Harvey L, Close J. New ICD-10 version of the Multipurpose Australian Comorbidity Scoring System outperformed Charlson and Elixhauser comorbidities in an older population. Journal of Clinical Epidemiology. 2016;79:62-9. Shebeshi DS, Dolja-Gore X, Byles J. Validation of hospital frailty risk score to predict hospital use in older people: evidence from the Australian Longitudinal Study on Women’s Health. Archives of gerontology and geriatrics. 2021;92:104282. Tsoli S, Blodgett J, Maynou L, Street A, Rockwood K, Davis D, et al. Understanding the trajectory of frailty across the life course: final report. NIHR. 2024. Jia L, Quan M, Fu Y, Zhao T, Li Y, Wei C, et al. Dementia in China: epidemiology, clinical management, and research advances. The Lancet Neurology. 2020;19(1):81-92. Li X, Feng X, Sun X, Hou N, Han F, Liu Y. Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990–2019. Frontiers in Aging Neuroscience. 2022;14:937486. Cao L, Gu D, Ni Y, Xie G. Automatic ICD code assignment based on ICD’s hierarchy structure for Chinese electronic medical records. AMIA Summits on Translational Science Proceedings. 2019;2019:417. Ren R, Qi J, Lin S, Liu X, Yin P, Wang Z, et al. The China alzheimer report 2022. General Psychiatry. 2022;35(1) Dong X, Wang Y. The geography of healthcare: Mapping patient flow and medical resource allocation in China. Economics & Human Biology. 2024;55:101431. National Healthcare Security Administration. Statistical Bulletin on the Development of China’s Health Undertakings in 2023; 2024. Accessed: (7 February 2025). https://www.nhsa.gov.cn/art/2024/7/25/art_7_13340.html. Song T, Ma R, Zhang X, Lv B, Li Z, Guo M, et al. Analysis of the current status and influencing factors of cross-regional hospitalization services utilization by basic medical insurance participants in China- taking a central province as an example. Frontiers in Public Health. 2023;11:1246982. Yan X, Shan L, He S, Zhang J. Cross-city patient mobility and healthcare equity and efficiency: Evidence from Hefei, China. Travel Behaviour and Society. 2022;28:1-12. Li Y, Liu X, Kang L, Li J. Validation and Comparison of Four Mortality Prediction Models in a Geriatric Ward in China. Clinical Interventions in Aging. 2023:200919. McAlister F, Savu A, Ezekowitz J, Armstrong P, Kaul P. The hospital frailty risk score in patients with heart failure is strongly associated with outcomes but less so with pharmacotherapy. Journal of Internal Medicine. 2020;287(3):322-32. Harvey LA, Toson B, Norris C, Harris IA, Gandy RC, Close JJ. Does identifying frailty from ICD-10 coded data on hospital admission improve prediction of adverse outcomes in older surgical patients? A population-based study. Age and ageing. 2021;50(3):802-8. Vermeiren S, Vella-Azzopardi R, Beckwee D, Habbig AK, Scafoglieri A, Jansen B, et al. Frailty and the prediction of negative health outcomes: a meta-analysis. Journal of the American medical directors association. 2016;17(12):1163-e1. Weng L, Hu Y, Sun Z, Yu C, Guo Y, Pei P, et al. Place of death and phenomenon of going home to die in Chinese adults: a prospective cohort study. The Lancet Regional Health–Western Pacific. 2022;18. Cesari M, Prince M, Thiyagarajan JA, De Carvalho IA, Bernabei R, Chan P, et al. Frailty: an emerging public health priority. Journal of the American Medical Directors Association. 2016;17(3):188-92. Gilbert T, Cordier Q, Polazzi S, Street A, Conroy S, Duclos A. Combining the hospital frailty risk score with the Charlson and Elixhauser multimorbidity indices to identify older patients at risk of poor outcomes in acute care. Medical Care. 2024;62(2):117-24. Supplementary Files HFRSChinaR1appendixAnon2.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 07 Apr, 2025 Reviewers invited by journal 05 Apr, 2025 Editor invited by journal 05 Apr, 2025 First submitted to journal 04 Apr, 2025 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-5551082","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":438927871,"identity":"1c9b765d-9f08-45f3-ae30-bd0d58164685","order_by":0,"name":"Yue Qiu","email":"","orcid":"https://orcid.org/0009-0001-1524-5583","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Qiu","suffix":""},{"id":438927872,"identity":"c53d2122-847f-48db-a528-053054f5d84e","order_by":1,"name":"Weiqing Xiong","email":"","orcid":"https://orcid.org/0009-0000-4662-9128","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Weiqing","middleName":"","lastName":"Xiong","suffix":""},{"id":438927873,"identity":"8f311a39-93cf-435e-b46c-750a2dcb63ce","order_by":2,"name":"Xinyue Fang","email":"","orcid":"https://orcid.org/0009-0007-8870-4000","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Xinyue","middleName":"","lastName":"Fang","suffix":""},{"id":438927874,"identity":"01218c8d-ce18-4e40-8363-35d0cac1f0e7","order_by":3,"name":"Pei Li","email":"","orcid":"https://orcid.org/0009-0006-6667-0017","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Pei","middleName":"","lastName":"Li","suffix":""},{"id":438927875,"identity":"e341a851-ea04-4c6c-9179-25198fb392b9","order_by":4,"name":"Simon Conroy","email":"","orcid":"https://orcid.org/0000-0002-4306-6064","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Simon","middleName":"","lastName":"Conroy","suffix":""},{"id":438927876,"identity":"f2c3c75b-fa34-44c8-b685-fa83c99faf12","order_by":5,"name":"Laia Maynou","email":"","orcid":"https://orcid.org/0000-0002-0447-2959","institution":"University of Barcelona: Universitat de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Laia","middleName":"","lastName":"Maynou","suffix":""},{"id":438927877,"identity":"16788b7d-2555-4079-8e09-48393d693c7b","order_by":6,"name":"Kenneth Rockwood","email":"","orcid":"https://orcid.org/0000-0002-6674-995X","institution":"Dalhousie University","correspondingAuthor":false,"prefix":"","firstName":"Kenneth","middleName":"","lastName":"Rockwood","suffix":""},{"id":438927878,"identity":"697fb66e-c470-49b2-813e-026caae87101","order_by":7,"name":"Xien Liu","email":"","orcid":"https://orcid.org/0000-0002-9054-0523","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Xien","middleName":"","lastName":"Liu","suffix":""},{"id":438927879,"identity":"6a21206d-91bc-457a-b59b-616daf484bcb","order_by":8,"name":"Ji Wu","email":"","orcid":"https://orcid.org/0000-0001-6170-726X","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Ji","middleName":"","lastName":"Wu","suffix":""},{"id":438927880,"identity":"8888b77a-e754-478e-8b16-a9e48ab1be9c","order_by":9,"name":"Andrew David Street","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzUlEQVRIiWNgGAWjYDADfnYYi4ewYsYGICEh2UyyFoPDxGrhb2B//riiwqbO+DD7wwcMNXYMBmcO4NcicYDHsPHMmTQJs8M8xgYMx5IZDM42ELDmAA9jY2PbYZAWNgkGtgMMBucJ6JA/wP6wsfHffwnjZvbnPxj+EaHF4ACDYWNjwwEJA2YGMwbGtgOEHWZ4mMdwZsOxZMkZQL9IJPYl80gS8r7c8fYHHxtq7Pj529sffvjwzU6O70wCAZcxI3MSiIrIUTAKRsEoGAUEAQB1pD6L02TmKQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-2540-0364","institution":"LSE: The London School of Economics and Political Science","correspondingAuthor":true,"prefix":"","firstName":"Andrew","middleName":"David","lastName":"Street","suffix":""}],"badges":[],"createdAt":"2024-11-29 18:56:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5551082/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5551082/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":80276878,"identity":"e55962d4-b32a-4f2f-b2e0-528619d24e23","added_by":"auto","created_at":"2025-04-10 05:08:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":96659,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of sample selection\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5551082/v1/ca12572bb9c9f9f0462597e1.png"},{"id":80279125,"identity":"170d9872-6779-4b0c-8ff7-54e146b0aae8","added_by":"auto","created_at":"2025-04-10 05:32:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":96060,"visible":true,"origin":"","legend":"\u003cp\u003eProportion of sample with each ICD-10 diagnosis\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5551082/v1/04acf86dfcc09aeec5e0027d.png"},{"id":80276876,"identity":"87fa932b-01fc-4d83-8dda-07995dd09dc2","added_by":"auto","created_at":"2025-04-10 05:08:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":184965,"visible":true,"origin":"","legend":"\u003cp\u003eForest plot of regression results for the full analytical sample\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5551082/v1/a38f91fa1aa76dc3322b8cc3.png"},{"id":80279212,"identity":"38daa808-a9c7-4972-9a22-a979c8cc8b87","added_by":"auto","created_at":"2025-04-10 05:33:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1242706,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5551082/v1/93cf3389-1613-49b7-acfc-3ef5318be702.pdf"},{"id":80279190,"identity":"8a49ab8b-b3aa-42e9-9799-7e7a9219add8","added_by":"auto","created_at":"2025-04-10 05:33:07","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1542392,"visible":true,"origin":"","legend":"","description":"","filename":"HFRSChinaR1appendixAnon2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5551082/v1/1b00766eb5d6b7bbefcf40c8.pdf"}],"financialInterests":"","formattedTitle":"Validation of the Hospital Frailty Risk Score in China","fulltext":[{"header":"Key Summary Points","content":"\u003cp\u003e\u003cstrong\u003eAim\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo validate the Hospital Frailty Risk Score (HFRS) in Chinese hospital settings\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFindings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRelatively few older people hospitalised in China are categorised using the HFRS as having high frailty risk\u003c/p\u003e\n\u003cp\u003eThe HFRS predicts long length of stay and hospital costs among hospitalised older patients in China\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMessage\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe HFRS has potential for widespread use in both developed and developing countries that use ICD-10 codes\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eHealth issues among older people are a significant concern in aging societies, impacting individual quality of life and the sustainability of healthcare systems. Frailty is defined as an \u0026ldquo;age-related, clinically identifiable state of diminished physiologic reserve and increased vulnerability to a broad range of adverse health outcomes\u0026rdquo; [1]. The associations between frailty and length of stay, hospital costs, and mortality has been confirmed in many studies [2]. Frailty significantly contributes to functional deterioration in older people [3], increasing healthcare demands and burdening families and society.\u003c/p\u003e \u003cp\u003eGlobal recognition of the need to assess and manage frailty has led to the development of various assessment tools [4; 5; 6; 7; 8; 9; 10; 11]. By identifying frailty-related risks, particularly in the first hours of an acute admission [12; 13; 14; 15], it may be possible to mitigate harms such as reduced mobility, which is associated with death or institutionalisation [16], loss of lean muscle mass, pressure ulcer development [17], incontinence, thrombosis, constipation, pain, infections and low mood [18]. However, many tools are too complex for use in acute care settings [19] and require professional evaluations, complicating their integration into clinical care [20; 21]. To address this, Gilbert and colleagues developed the Hospital Frailty Risk Score (HFRS), which uses International Statistical Classification of Diseases and Related Health Problems, tenth revision (ICD-10) codes from electronic health records for hospitalised patients to compute frailty risk scores without additional data gathering [22].\u003c/p\u003e \u003cp\u003eSince its development, the HFRS has garnered extensive attention, with applications to different types of patients to validate its predictive capacity for outcomes such as length of stay, mortality, and treatment costs [23; 24; 25; 26; 27; 28; 29; 30; 31]. These validation studies, however, have largely been conducted in developed countries, with few applications from developing ones. To start addressing this gap, this paper applies the HFRS to older patients admitted to hospital in China.\u003c/p\u003e \u003cp\u003eSeveral studies have measured the prevalence of frailty in the Chinese community, using instruments such as the Frailty Index [32] and a validated physical frailty phenotype (PFP) scale [33]. A meta-analysis showed that 15% of Chinese community dwelling older people aged 75\u0026ndash;84 experienced frailty, while this figure rose to 25% for those aged 85 and over [34]. As a rapidly aging developing country, there is a great need to identify and manage frailty in older people in China. But China has few healthcare workers relative to the size of the population (2.4 physicians and 3.3 nurses and midwives per 1000 [35]), which rules out widespread application of many frailty assessment tools that require primary data collection. Using electronic health records to identify frail patients offers a more cost-effective approach in China.\u003c/p\u003e \u003cp\u003eThis study aimed to assess the applicability of the HFRS for hospitalized patients 75 years and older in China, with a focal emphasis on the effectiveness of frailty identification and the tool\u0026rsquo;s ability to predict healthcare resource consumption. The findings will provide evidence on whether China can adopt the HFRS or if a new tool needs to be developed based on China\u0026rsquo;s specific conditions.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSetting\u003c/h2\u003e \u003cp\u003eHealth care in China combines a mix of public and privately funded care [36]. Over 95% of the population is insured through either the Urban Employee Basic Medical Insurance (UEBMI) scheme, which provides coverage for urban employees, or through the Urban and Rural Residence Basic Medical Insurance (URRBMI) scheme which covers those without formal employment in urban and rural settings. These insurance schemes offer similar benefits and together account for 46% of total health expenditure, with 26.7% coming from government sources and 27.3% from individuals, mainly as out-of-pocket payments [37].\u003c/p\u003e \u003cp\u003eWe conducted a retrospective study of patients aged 75 and older who had been hospitalised between 1 January 2022 and 31 December 2023 in Lvliang City, Shanxi Province, China. Lvliang City has a middle-ranged economic level among the cities of China and, according to the 7th China population census [38], has a permanent population of about 3.3\u0026nbsp;million, of which 3.9% are aged 75 years and above, and 0.8% are aged 85 years and above. It is a relatively young city as those proportions for the whole country are 4.8% and 1.1% respectively [38].\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eData\u003c/h3\u003e\n\u003cp\u003eWe used a medical database to retrieve anonymised patient-level data. The entire database contained routine administrative medical data, including clinical characteristics and healthcare utilisation, for 357,186 patients admitted to 72 primary hospitals, 45 secondary hospitals, and three tertiary hospitals in Lvliang City from 1 January 2022 to 31 December 2023. In the hospital classification of China, primary hospitals are community facilities with fewer than 100 beds that provide preventive care, primary care and rehabilitation. Secondary hospitals have between 100 and 500 beds, are located in cities and offer broad services and some specialities. Tertiary hospitals are located in major cities and offer a wide range of specialists and have more than 500 beds. In tertiary hospitals, staff tend to be more qualified and equipment more advanced so complexity and quality should be higher in these hospitals [39].\u003c/p\u003e \u003cp\u003eThe process of selecting the analytical sample is described in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. We extracted information for all 38,932 patients aged 75 years and over admitted to secondary and tertiary hospitals, dropping patients admitted to primary hospitals for consistency with Gilbert et al [22]. For patients with multiple admissions, only the most recent admission was used for the analysis. After excluding 4,201 (10.8%) patients with missing data, the analytical sample comprised 34,731 patients. We compared key variables before and after excluding patients with missing data for other variables, reported in Appendix A1. This revealed no significant differences except for hospital costs. Ethics approval was obtained from the Institutional Review Board of Tsinghua University (THU01-20240115).\u003c/p\u003e\n\u003ch3\u003eOutcome and frailty measures\u003c/h3\u003e\n\u003cp\u003eWe analysed the relationship between the HFRS and two outcomes, long LoS and hospital costs, controlling for other characteristics:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eWe used the patient\u0026rsquo;s admission and discharge dates to calculate LoS. Consistent with the original HFRS study by Gilbert et al [22], this was converted into a binary variable where 1 indicated LoS over 10 days, 0 otherwise and analysed using a logistic regression model. As a robustness check we also analysed LoS using a Poisson model, reported in Appendix A5.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe cost of the hospital stay is recorded in the patient\u0026rsquo;s medical record, this being the sum of the patient\u0026rsquo;s treatment costs for the admission, including the cost of examinations, tests, medicines, consumables, surgical treatment, and medical services. Costs were analysed using a generalised linear model.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eFor each patient the HFRS was calculated by combining a weighted set of 109 3-character ICD-10 diagnostic codes [22]. Given that only two years\u0026rsquo; worth of data were available we constructed the \u003cem\u003eHFRS\u003c/em\u003e(\u003cem\u003ea\u003c/em\u003e) form, using diagnostic information from the current admission only [30]. The HFRS takes values from 0 to 173.2. Commonly patients have been categorised as having low (HFRS\u0026thinsp;\u003cem\u003e\u0026lt;\u003c/em\u003e\u0026thinsp;5), intermediate ((5\u0026thinsp;\u0026le;\u0026thinsp;HFRS\u0026thinsp;\u003cem\u003e\u0026lt;\u003c/em\u003e\u0026thinsp;15), or high (HFRS\u0026thinsp;\u0026ge;\u0026thinsp;15) frailty risk [22]. However, in this study a large proportion of patients had a HFRS of zero and, hence, we categorised patients into four groups: zero frailty risk (HFRS\u0026thinsp;=\u0026thinsp;0), low frailty risk (0\u0026thinsp;\u003cem\u003e\u0026lt;\u003c/em\u003e\u0026thinsp;HFRS\u0026thinsp;\u003cem\u003e\u0026lt;\u003c/em\u003e\u0026thinsp;5), intermediate frailty risk (5\u0026thinsp;\u0026le;\u0026thinsp;HFRS\u0026thinsp;\u003cem\u003e\u0026lt;\u003c/em\u003e\u0026thinsp;15), or high frailty risk (HFRS\u0026thinsp;\u0026ge;\u0026thinsp;15). Those in the zero frailty risk category formed the reference group in the regressions.\u003c/p\u003e \u003cp\u003eThe analyses controlled for patient age, sex, Charlson comorbidity index (CCI) [40; 41; 42; 43], admission via the emergency department, the number of operations performed, hospital type, and the patient\u0026rsquo;s insurance scheme. The Appendix details the construction of these control variables and provides specification details about the regression models.\u003c/p\u003e\n\u003ch3\u003eSubgroup analyses\u003c/h3\u003e\n\u003cp\u003eWe conducted subgroup analyses to uncover the relationship between frailty risk and patient outcomes among patients with different characteristics. The subgroups were divided according to age (75\u0026ndash;79, 80\u0026ndash;84, 85+), sex (male and female), and hospital tiers (secondary and tertiary). The regression models and the control variables in the subgroup analyses were the same as those employed in the analysis of all patients, only excluding the variable used to form the subgroup.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eDescriptive statistics\u003c/h2\u003e\n\u003cp\u003eThe analytical sample comprised 34,731 patients aged 75 and above. Descriptive statistics are shown in Table 1. Among these patients, the maximum HFRS score was 19.8 points, and 7,715 (22.2%) patients were categorised as having zero frailty risk, 21,667 (62.4%) as low frailty risk, 5,320 (15.3%) as intermediate frailty risk, and only 29 (0.08%) as high frailty risk. Generally, patients with higher frailty risk were more likely to stay in hospital for more than 10 days, had higher costs, were more likely to be admitted through the ED, and had a higher CCI score.\u003c/p\u003e\n\u003cp\u003eTable 1: Descriptive statistics\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"578\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDescriptive Statistics (N = 34731)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"bottom\" style=\"width: 56px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eZero risk (n=7715,22.21%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLow risk (n=21,667, 62.39%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIntermediate risk (n= 5,320, 15.32%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHigh risk (n=29, 0.08%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003eN / Mean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003eProportion/ SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003eN / Mean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003eProportion/ SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003eN / Mean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003eProportion/ SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003eN / Mean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003eProportion/ SD\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLong length of stay\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1,178\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e15.27%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e5,443\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e25.12%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1,832\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e34.44%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e62.07%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e6,537\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e84.73%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e16,224\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e74.88%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e3,488\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e65.56%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e37.93%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHospital costs\u0026nbsp;\u003c/strong\u003e\u0026yen;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e5,931\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e6,104\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e7,387\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e8,766\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e11,596\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e21,251\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e28,324\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e48,904\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge group\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e75-79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e4,099\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e53.13%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e10,754\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e49.63%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,240\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e42.11%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e48.28%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e80-84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,326\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e30.15%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e6,796\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e31.37%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1,821\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e34.23%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e34.48%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e85+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1,290\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e16.72%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e4,117\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e19.00%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1,259\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e23.67%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e17.24%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e3,941\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e51.08%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e10,839\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e50.03%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,704\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e50.83%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e51.72%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e3,774\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e48.92%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e10,828\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e49.97%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,616\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e49.17%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e48.28%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHospital tier\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eSecondary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e6,075\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e78.74%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e15,843\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e73.12%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,983\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e56.07%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e37.93%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eTertiary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1640\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e21.26%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e5,824\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e26.88%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,337\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e43.93%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e62.07%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdmission through ED\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e531\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e6.88%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e2,076\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e9.58%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e772\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e14.51%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e24.14%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e7,184\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e93.12%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e19,591\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e90.42%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e4,548\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e85.49%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e75.86%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharlson Comorbidity Index (CCI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e6,861\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e88.93%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e18,231\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e84.14%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e4,195\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e78.85%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e55.17%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e772\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e10.01%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e3,060\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e14.12%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e973\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e18.29%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e34.48%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e2+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1.06%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e376\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1.74%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e152\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2.86%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e10.34%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of operations\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.429\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e0.707\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.283\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e0.736\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.526\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1.102\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e1.207\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2.007\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBasic medical insurance type\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eURBMI\u003c/p\u003e\n \u003cp\u003eUEBMI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e6,262\u003c/p\u003e\n \u003cp\u003e1,453\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e81.17%\u003c/p\u003e\n \u003cp\u003e18.83%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e17,396\u003c/p\u003e\n \u003cp\u003e4,271\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e80.29%\u003c/p\u003e\n \u003cp\u003e19.71%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e4,117\u003c/p\u003e\n \u003cp\u003e1,203\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e77.39%\u003c/p\u003e\n \u003cp\u003e22.61%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e65.52%\u003c/p\u003e\n \u003cp\u003e34.48%\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eReimbursement rate\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e0.162\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.686\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e0.159\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.672\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e0.166\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e0.722\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 78px;\"\u003e\n \u003cp\u003e0.153\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\u003c/br\u003e\n\u003ch2\u003eRegression results\u003c/h2\u003e\n\u003cp\u003eThe full regression results of for the analyses of long LoS and hospital cost are reported in Appendix Table A2 and summarised as a forest plot in Figure 3. Compared to the zero frailty risk group, the probability of long LoS was: 1.92 (95% confidence interval (CI) 1.79-2.06) times higher for those with low frailty risk; 2.71 (95% CI 2.49-2.96) times higher for those with intermediate frailty risk; and 6.65 (95% CI 3.06-14.43) times higher for those with high frailty risk. The wide CI for the high frailty group reflects the small number of patients in this group. The results of applying the Poisson model to analyse LoS are consistent with the logistic model, as reported in Appendix A5.\u003c/p\u003e\n\u003cp\u003eHospital costs also increased in line with the frailty risk level. Compared to those with zero frailty risk, costs were\u0026nbsp;\u0026yen;1,926 (95% CI 1,655-2,197) higher for those with low frailty risk,\u0026nbsp;\u0026yen;4,284 (95%CI 3,916-4,653) higher for those with intermediate frailty risk, and\u0026nbsp;\u0026yen;16,613 (95% CI 12,827-20,399) higher for those with high frailty risk.\u003c/p\u003e\n\u003cp\u003eMost control variables revealed significant effects for long LoS or hospital costs. For instance, LoS and hospital costs increased with the number of operations (long LoS OR 1.28 95% CI 1.24-1.32; hospital costs \u0026yen;4,955 95% CI 4,817-5,093). In the analysis of long LoS, the odd ratios for CCI were significant but less than 1 (OR 0.91 95% CI 0.85-0.98) for CCI = 1 and insignificant for CCI =2+. For hospital costs, the CCI coefficients were positive (CCI=1 \u0026yen;882 95% CI 564-1,202; CCI=2+ \u0026yen;786 95% CI -49-1,622).\u003c/p\u003e\n\u003ch2\u003eSubgroup analyses\u003c/h2\u003e\n\u003cp\u003eThe regression results for length of stay and hospital costs across subgroups are summarized in Appendix A6.\u003c/p\u003e\n\u003cp\u003eAcross all subgroups, compared to those with zero frailty risk, both LoS and costs increased across the frailty risk categories. For those with low frailty risk, the differences were small; for those in the intermediate frailty risk category the differences are larger and significant; for those in the high frailty risk category there are wide confidence intervals around the point estimates because of the small numbers in this category.\u003c/p\u003e\n\u003cp\u003eThese patterns of longer LoS and higher costs as frailty risk increases demonstrate that the HFRS can be applied across age categories, sex, and hospital tiers, thereby underscoring its usefulness as a predictive tool for patients hospitalised in China.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis paper had two objectives. Firstly, to validate the HFRS in Chinese hospital settings population, describing how patients are allocated to frailty risk groups. Secondly to assess how frailty risk is associated with LoS and hospital costs.\u003c/p\u003e \u003cp\u003eAs regards the first objective, in the Lvliang hospital care setting, 7,715 (22.2%) individuals were identified as having zero frailty risk, 21,667 (62.4%) patients were categorised to the low frailty risk group, 5,320 (15.3%) into the intermediate frailty risk group, and 29 (0.08%) into the high frailty risk group. Notably, very few were identified as being high risk, especially when compared with much higher proportions in the high risk group in two studies from England (20% in [22] and in 21.8% [30]) and of 17% in a study from France [29]. However, other studies of those over 75 years also report small proportions in the high risk group: 2.9% in a study from Switzerland [24], 2.6% in one from Canada [23] and 1.9% in a study from Australia [44].\u003c/p\u003e \u003cp\u003eThe HFRS for any particular individual is driven by whether they have one of the 109 ICD-10 codes and the weight attached to that code, these weights ranging from 0.1 to 7.1 (F00 Dementia in Alzheimer\u0026rsquo;s disease). The proportions of the sample with the 30 ICD-10 codes with a weight\u0026thinsp;\u003cem\u003e\u0026gt;\u003c/em\u003e\u0026thinsp;2.0 are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. These proportions are compared to patients aged 75 and above hospitalised in England between 2013 to 2019 [45].\u003c/p\u003e \u003cp\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\u003eDistribution of ICD-10 codes with HFRS weight\u0026thinsp;\u003cem\u003e\u0026gt;\u003c/em\u003e\u0026thinsp;2.0\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=\"left\" 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=\"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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eICD-10 code\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eICD-10 description\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWeight\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChina sample\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEngland sample\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDementia in Alzheimer\u0026rsquo;s disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.77%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eG81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHemiplegia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.61%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.01%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eG30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlzheimer\u0026rsquo;s disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.62%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSequelae of cerebrovascular disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11.75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTendency to fall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e18.13%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDelirium, not by alcohol or psychoactive substances\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.15%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDisorders of urinary system, including UTI \u0026amp; urinary incontinence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.79%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e18.89%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSuperficial injury of head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.23%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnspecified fall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.07%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e8.69%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnspecified haematuria\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.22%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eB96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOther bacterial agents as the cause of diseases\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.68%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOther cognitive functions and awareness symptoms and signs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7.70%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOther cerebrovascular diseases\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.55%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7.91%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAbnormalities of gait and mobility\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e9.01%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eConvulsions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.15%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSomnolence, stupor and coma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.23%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.80%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIntracranial injury\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.25%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.88%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eComplications of genitourinary prosthetic devices, implants, grafts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.19%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eE86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVolume depletion\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7.47%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eE87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOther disorders of fluid, electrolyte and acid-base balance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e11.69%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOther joint disorders\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.26%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.58%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFracture of shoulder and upper arm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.13%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.13%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSenility\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.04%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.50%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnspecified dementia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.27%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e12.27%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOther fall on same level\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.21%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCare involving use of rehabilitation procedures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.44%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVascular dementia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.87%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eL03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCellulitis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.05%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.66%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSuperficial injury of lower leg\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.95%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProblems related to medical facilities and health care\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.66%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eClear differences are evident. Less than 1% of the Lvliang sample had one of the three ICD-10 codes (F00, G81, G30) with the highest weight; in England 11.4% of the sample had one of these codes. In the Lvliang dataset, the highest proportion is attributed to the ICD-10 code for \u0026ldquo;E87 Other disorders of fluid, electrolyte and acid-base balance\u0026rdquo; (19.4%), followed by \u0026ldquo;I67 Other cerebrovascular disease\u0026rdquo; (13.6%) and \u0026ldquo;I69 Sequelae of cerebrovascular disease\u0026rdquo; (11.8%). In the English dataset, the distribution of the 30 ICD-10 codes is more spread out, with the highest proportions attributed to \u0026ldquo;N39 Disorders of urinary system including UTI \u0026amp; urinary incontinence\u0026rdquo; (18.9%) and \u0026ldquo;R29 Tendency to fall\u0026rdquo; (18.1%). Comparing these two samples, it is evident that different ICD-10 codes identify the frailty risk of patients in the two countries.\u003c/p\u003e \u003cp\u003eThe most direct reason for the differences between the Chinese and English data is the lower count of the 109 ICD-10 codes used to construct the HFRS in the Lvliang dataset, particularly for codes with large HFRS weights such as Dementia (F00) [46] and Alzheimer\u0026rsquo;s disease (G30) [47]. The lower frequency of these codes in China can be attributed partly to the younger age structure and its associated disease spectrum. Compared to 42.5% of the population aged 85 years or older reported in a regional study in England [30], Lvliang\u0026rsquo;s patients are relatively young, with only 19.2% of the population aged 85 years or older. Due to this younger age structure, diseases like dementia and Alzheimer\u0026rsquo;s, which are more prevalent in older age groups, have lower incidence rates in Lvliang. Consequently, the frequency of these ICD-10 codes is also lower, resulting in lower risk scores when calculating the HFRS. Besides, the limited availability of well-trained medical records staff might mean that diagnoses are under-coded in China [48; 49].\u003c/p\u003e \u003cp\u003eDisparities in healthcare delivery capacity between countries and regions may also play a role in explaining the differences in the proportions of patients in each frailty risk group [50]. Around 2.5% of patients in China seek care outside the region in which they live [51], the primary reason being to access higher quality care [52]. Patients from Lvliang might seek high quality healthcare in nearby larger cities like Datong or Beijing [53]. These factors will mean that hospitals will be treating quite different patient profiles, implying that the HFRS is influenced by factors other than frailty related diagnostic codes [24] and underscoring a need for further investigation into the underlying causes as well as possible coding differences [44]. Note that, as well as impacting the HFRS, coding differences also have implications for the CCI, where the proportion of patients with a CCI\u0026thinsp;\u0026ge;\u0026thinsp;2 was 1.76% in our study but 51.6% in the study in the England [30]. Low CCI scores were also reported in a study of older patients admitted to hospital in Beijing [54].\u003c/p\u003e \u003cp\u003eRegarding the second objective, despite differences in patient profiles and diagnostic coding, the HFRS still emerged as a strong explanator or length of stay and costs for patients admitted to hospital in Lvliang city. Patients in low, intermediate, and high HFRS risk groups were significantly and progressively more likely to stay in hospital for over 10 days and have higher treatment costs compared to patients in the zero risk group. If diagnostic coding were improved, frailty risk scores would be higher and the explanatory power of the HFRS would become greater. Even so, the HFRS had greater explantory power for these two outcomes than any of the other variables included in the regression analyses. Subgroup analyses further confirmed the robustness of these findings.\u003c/p\u003e \u003cp\u003eFor long LoS, our regression results were similar to other HFRS validation studies conducted in different countries [22; 23; 55; 24; 56; 28; 30; 29]. The influence of the HFRS on hospital costs was also consistent with findings from other studies [24; 28]. Our validation study suggests that, despite variations in healthcare system and ICD-10 coding rules across countries, the risk of frailty calculated using the HFRS methodology is a useful predictor of length of stay and hospital costs in China, as has been found elsewhere [57]. Therefore, the HFRS holds great potential for widespread use in countries using ICD-10 codes, both in developed and developing countries, due to its explanatory power, convenience and cost-effectiveness.\u003c/p\u003e \u003cp\u003eThere are limitations to this study. First, the narrow data window only allowed analysis of the patient\u0026rsquo;s most recent admission, which will have led to a lower HFRS compared to other studies that also include diagnostic information from the previous two admissions within the last two years, as recommended [30]. This means that, on the one hand, diagnoses recorded in a patient\u0026rsquo;s previous admissions are not captured but, on the other hand, the HFRS is constructed in a consistent fashion for every patient in the study. Second, diagnoses may have been under-coded. If so, the explanatory power of the HFRS would have been under-estimated. Third, a high proportion of patients were omitted from analysis due to missing data. With the exception of costs, data appeared to be missing at random thereby suggesting that the analytical sample remained representative. Nevertheless the high proportion highlights the scope for improved coding practice, not just of diagnoses but more generally. Fourth, we did not analyse the relationship between the HFRS and in-hospital death because, due to Chinese cultural practices, most older patients prefer to receive end-of-life care at home with family and friends [58]. Reflecting this, only 0.37% died in hospital. Nor did the data allow us to identify those who were discharged to die at home. Fifth, this study utilized regional data from a single city in China, which may not be representative of the entire country though it may be fairly typical of other middle-ranged cities with similar socio-economic characteristics. Future studies of the HFRS using data from other areas in China would be welcome.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eAs the population ages, particularly in low-income and middle-income countries, the impact of frailty will escalate [2]. Frailty risk, easily calculated using the HFRS, offers benefits at the micro, meso and macro level of the system. At the micro level, involving clinician-patient interaction, a measure of frailty risk alerts the clinician to the potential prognosis. People with high HFRS scores have an increased risk of dying in hospital \u0026ndash; this might prompt the clinician to activate critical care if clinically appropriate and in keeping with the patient\u0026rsquo;s wishes and preferences. Alternatively, it might lead to a more palliative or supportive paradigm of care being instituted, following assessment of the individual. Fundamentally assessment of frailty risk moves us away from a one-size-fits-all approach, recognising that patients are at different stages on their life trajectory.\u003c/p\u003e \u003cp\u003eAt the meso level, it allows hospitals to match resources to patient needs. For example, it might be that people with high HFRS scores are found in all parts of the hospital, prompting the development of a geriatric liaison service and evaluating its impact on the available service metrics.\u003c/p\u003e \u003cp\u003eAt the macro level, it facilitates the creation of registries, which can track the flow of a risk stratified cohort of older people along care pathways and redirect where appropriate. For example, a dynamic frailty risk registry might highlight that a patient with high frailty discharged from hospital has not been referred to community services. This can then be rectified by putting post-discharge support in place.\u003c/p\u003e \u003cp\u003eOur study is the first to confirm the predictive effect of HFRS on length of stay and hospital costs in China, a developing country with a growing older population. The HFRS strikes a balance between broad applicability and low cost through its big data driven approach. However, the identification of frailty risk by HFRS in this study was significantly different from the original English population cohorts by Gilbert et al [22], likely due to the differences in age structure, disease spectrum, healthcare delivery capacity and diagnostic coding practices between England and China. In light of these findings, developing countries like China might benefit from employing the HFRS but also from using a similar big data-driven approach to develop localised frailty screening tools, tailored to reflect their particular the demographic and healthcare landscapes [59]. Investment in recruitment and training of coding staff should improve diagnostic coding practices and data quality. In places with the necessary infrastructure, pilot implementation of HFRS in clinical workflows could be a good way to inform how construction of the HFRS might be refined and how it might best be applied to assess frailty risk among hospitalised patients in other developing countries.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYQ was supported by the Du Shi Special Fund (Grant No. 2024Z11DSZ001) from Tsinghua University, which supports promising faculties dedicated to fundamental research. AS received funding support from LSE\u0026rsquo;s Global Research Fund. SC, LM, KR and AS received funding support from the National Institute for Health and Care Research (NIHR 203451 - Understanding the Trajectory of Frailty Across the Life Course). The authors thank the journal\u0026rsquo;s reviewers, Mengxi Pang and Susana Mourato.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eChinese data presented in this study may be obtained from a third party on request from the authors and are not publicly available. The English data came from the Hospital Episode Statistics provided by NHS Digital under Data Sharing Agreement NIC-354497-V2J9P. These data may be obtained from a third party and are not publicly available. This paper has been screened to ensure no confidential information is revealed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOn behalf of all authors, the corresponding author states that there is no conflict of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eKim DH, Rockwood K. Frailty in older adults. New England Journal of Medicine. 2024;391(6):538-48.\u003c/li\u003e\n \u003cli\u003eHoogendijk EO, Afilalo J, Ensrud KE, Kowal P, Onder G, Fried LP. Frailty: implications for clinical practice and public health. The Lancet. 2019;394(10206):136575.\u003c/li\u003e\n \u003cli\u003eDent E, Morley J, Cruz-Jentoft A, Woodhouse L, Rodr\u0026iacute;guez-Ma\u0026ntilde;as L, Fried L, et al. Physical frailty: ICFSR international clinical practice guidelines for identification and management. The Journal of nutrition, health and aging. 2019;23(9):771-87.\u003c/li\u003e\n \u003cli\u003eClegg A, Bates C, Young J, Ryan R, Nichols L, Ann Teale E, et al. Development and validation of an electronic frailty index using routine primary care electronic health record data. Age and ageing. 2016;45(3):353-60.\u003c/li\u003e\n \u003cli\u003eDe Vries N, Staal J, Van Ravensberg C, Hobbelen J, Rikkert MO, Nijhuis-van der Sanden M. Outcome instruments to measure frailty: a systematic review. Ageing research reviews. 2011;10(1):104-14.\u003c/li\u003e\n \u003cli\u003eDent E, Kowal P, Hoogendijk EO. Frailty measurement in research and clinical practice: a review. European journal of internal medicine. 2016;31:3-10.\u003c/li\u003e\n \u003cli\u003eGregorevic KJ, Hubbard RE, Katz B, Lim WK. The clinical frailty scale predicts functional decline and mortality when used by junior medical staff: a prospective cohort study. BMC geriatrics. 2016;16:1-6.\u003c/li\u003e\n \u003cli\u003eLansbury LN, Roberts HC, Clift E, Herklots A, Robinson N, Sayer AA. Use of the electronic Frailty Index to identify vulnerable patients: a pilot study in primary care. British Journal of General Practice. 2017;67(664):e751-6.\u003c/li\u003e\n \u003cli\u003eBrundle C, Heaven A, Brown L, Teale E, Young J, West R, et al. Convergent validity of the electronic frailty index. Age and ageing. 2019;48(1):152-6.\u003c/li\u003e\n \u003cli\u003eBoyd PJ, Nevard M, Ford JA, Khondoker M, Cross JL, Fox C. The electronic frailty index as an indicator of community healthcare service utilisation in the older population. Age and Ageing. 2019;48(2):273-7.\u003c/li\u003e\n \u003cli\u003eElliott A, Taub N, Banerjee J, Aijaz F, Jones W, Teece L, et al. Does the clinical frailty scale at triage predict outcomes from emergency care for older people? Annals of emergency medicine. 2021;77(6):620-7.\u003c/li\u003e\n \u003cli\u003eBernstein SL, Aronsky D, Duseja R, Epstein S, Handel D, Hwang U, et al. The effect of emergency department crowding on clinically oriented outcomes. Academic Emergency Medicine. 2009;16(1):1-10.\u003c/li\u003e\n \u003cli\u003ePines JM, Pollack Jr CV, Diercks DB, Chang AM, Shofer FS, Hollander JE. The association between emergency department crowding and adverse cardiovascular outcomes in patients with chest pain. Academic Emergency Medicine. 2009;16(7):617-25.\u003c/li\u003e\n \u003cli\u003ePlatts-Mills TF, Owens ST, McBride JM. A modern-day purgatory: older adults in the emergency department with nonoperative injuries. Journal of the American Geriatrics Society. 2014;62(3):525-8.\u003c/li\u003e\n \u003cli\u003eCarter EJ, Pouch SM, Larson EL. The relationship between emergency department crowding and patient outcomes: a systematic review. Journal of Nursing Scholarship. 2014;46(2):106-15.\u003c/li\u003e\n \u003cli\u003eBrown CJ, Friedkin RJ, Inouye SK. Prevalence and outcomes of low mobility in hospitalized older patients. Journal of the American Geriatrics Society. 2004;52(8):1263-70.\u003c/li\u003e\n \u003cli\u003eColeman S, Gorecki C, Nelson EA, Closs SJ, Defloor T, Halfens R, et al. Patient risk factors for pressure ulcer development: systematic review. International journal of nursing studies. 2013;50(7):974-1003.\u003c/li\u003e\n \u003cli\u003eCreditor MC. Hazards of hospitalization of the elderly. Annals of internal medicine. 1993;118(3):219-23.\u003c/li\u003e\n \u003cli\u003eElliott A, Hull L, Conroy SP. Frailty identification in the emergency department\u0026mdash;a systematic review focussing on feasibility. Age and ageing. 2017;46(3):509-13.\u003c/li\u003e\n \u003cli\u003eRockwood K, Song X, MacKnight C, Bergman H, Hogan DB, McDowell I, et al. A global clinical measure of fitness and frailty in elderly people. Cmaj. \u0026nbsp;2005;173(5):489-95\u003c/li\u003e\n \u003cli\u003eSternberg SA, Schwartz AW, Karunananthan S, Bergman H, Mark Clarfield A. The identification of frailty: a systematic literature review. Journal of the American Geriatrics Society. 2011;59(11):2129-38.\u003c/li\u003e\n \u003cli\u003eGilbert T, Neuburger J, Kraindler J, Keeble E, Smith P, Ariti C, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. The Lancet. 2018;391(10132):1775-82.\u003c/li\u003e\n \u003cli\u003eMcAlister F, van Walraven C. External validation of the Hospital Frailty Risk Score and comparison with the Hospital-patient One-year Mortality Risk Score to predict outcomes in elderly hospitalised patients: a retrospective cohort study. BMJ quality \u0026amp; safety. 2019;28(4):284-8.\u003c/li\u003e\n \u003cli\u003eEckart A, Hauser SI, Haubitz S, Struja T, Kutz A, Koch D, et al. Validation of the hospital frailty risk score in a tertiary care hospital in Switzerland: results of a prospective, observational study. BMJ open. 2019;9(1):e026923.\u003c/li\u003e\n \u003cli\u003eKwok CS, Zieroth S, Van Spall HG, Helliwell T, Clarson L, Mohamed M, et al. The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure short running title: Frailty and outcomes in heart failure. International Journal of Cardiology. 2020;300:184-90.\u003c/li\u003e\n \u003cli\u003eBruno RR, Wernly B, Flaatten H, Sch\u0026ouml;lzel F, Kelm M, Jung C. The hospital frailty risk score is of limited value in intensive care unit patients. Critical Care. 2019;23:1-2.\u003c/li\u003e\n \u003cli\u003eKundi H, Wadhera RK, Strom JB, Valsdottir LR, Shen C, Kazi DS, et al. Association of frailty with 30-day outcomes for acute myocardial infarction, heart failure, and pneumonia among elderly adults. JAMA cardiology. 2019;4(11):1084-91.\u003c/li\u003e\n \u003cli\u003eBonjour T, Waeber G, Marques-Vidal P. Trends in prevalence and outcomes of frailty in a Swiss university hospital: a retrospective observational study. Age and Ageing. 2021;50(4):1306-13.\u003c/li\u003e\n \u003cli\u003eGilbert T, Cordier Q, Polazzi S, Bonnefoy M, Keeble E, Street A, et al. External validation of the hospital frailty risk score in France. Age and ageing. 2022;51(1):afab126.\u003c/li\u003e\n \u003cli\u003eStreet A, Maynou L, Gilbert T, Stone T, Mason S, Conroy S. The use of linked routine data to optimise calculation of the Hospital Frailty Risk Score on the basis of previous hospital admissions: a retrospective observational cohort study. The Lancet Healthy Longevity. 2021;2(3):e154-62.\u003c/li\u003e\n \u003cli\u003eTurcotte LA, Heckman G, Rockwood K, Vetrano DL, H\u0026acute;ebert P, McIsaac DI, et al. External validation of the hospital frailty risk score among hospitalised home care clients in Canada: a retrospective cohort study. Age and ageing. 2023;52(2):afac334.\u003c/li\u003e\n \u003cli\u003eMa L, Tang Z, Zhang L, Sun F, Li Y, Chan P. Prevalence of frailty and associated factors in the community-dwelling population of China. Journal of the American Geriatrics Society. 2018;66(3):559-64.\u003c/li\u003e\n \u003cli\u003eWu C, Smit E, Xue QL, Odden MC. Prevalence and correlates of frailty among community-dwelling Chinese older adults: the China health and retirement longitudinal study. The Journals of Gerontology: Series A. 2018;73(1):102-8.\u003c/li\u003e\n \u003cli\u003eHe B, Ma Y, Wang C, Jiang M, Geng C, Chang X, et al. Prevalence and risk factors for frailty among community-dwelling older people in China: a systematic review and meta-analysis. The Journal of nutrition, health and aging. 2019;23(5):442-50.\u003c/li\u003e\n \u003cli\u003eWorld Bank Group. Physicians, nurses and midwives (per 1000); 2025. Accessed: (7 February 2025).\u0026nbsp;https://data.worldbank.org/indicator/SH.MED.PHYS.ZS\u0026nbsp;and\u0026nbsp;https://data.worldbank.org/indicator/SH.MED.NUMW.P3.\u003c/li\u003e\n \u003cli\u003eLi X, Fan L, Leng SX. The aging tsunami and senior healthcare development in China. Journal of the American Geriatrics Society. 2018;66(8):1462-8.\u003c/li\u003e\n \u003cli\u003eNational Health Commission of the People\u0026rsquo;s Republic of China. Statistical Bulletin on the Development of China\u0026rsquo;s Health Undertakings in 2023; 2024. Accessed: (7 February 2025).\u0026nbsp;http://www.nhc.gov.cn/ guihuaxxs/s3585u/202408/6c037610b3a54f6c8535c515844fae96/files/ 58c5d1e9876344e5b1aa5aa2b083a51a.pdf.\u003c/li\u003e\n \u003cli\u003eOffice of the leading group of the state council for the seventh national population census. China population yearbook 2020. China Statistics press; 2022.\u003c/li\u003e\n \u003cli\u003eShi H, Fan M, Zhang H, Ma S, Wang W, Yan Z, et al. Perceived health-care quality in China: a comparison of second- and third-tier hospitals. International Journal for Quality in Health Care. 2021 02;33(1):mzab027. Available from:\u0026nbsp;https:// doi.org/10.1093/intqhc/mzab027.\u003c/li\u003e\n \u003cli\u003eCharlson M, Pompei P, Ales K, MacKenzie C. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-83.\u003c/li\u003e\n \u003cli\u003eQuan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Medical Care. 2005;43(11):1130-9.\u003c/li\u003e\n \u003cli\u003eBannay A, Chaignot C, Bloti\u0026egrave;re P, Basson M, Weill A, Ricordeau P, et al. The Best Use of the Charlson Comorbidity Index With Electronic Health Care Database to Predict Mortality. Medical Care. 2016;54(2):188-94\u003c/li\u003e\n \u003cli\u003eToson B, Harvey L, Close J. New ICD-10 version of the Multipurpose Australian Comorbidity Scoring System outperformed Charlson and Elixhauser comorbidities in an older population. Journal of Clinical Epidemiology. 2016;79:62-9.\u003c/li\u003e\n \u003cli\u003eShebeshi DS, Dolja-Gore X, Byles J. Validation of hospital frailty risk score to predict hospital use in older people: evidence from the Australian Longitudinal Study on Women\u0026rsquo;s Health. Archives of gerontology and geriatrics. 2021;92:104282.\u003c/li\u003e\n \u003cli\u003eTsoli S, Blodgett J, Maynou L, Street A, Rockwood K, Davis D, et al. Understanding the trajectory of frailty across the life course: final report. NIHR. 2024.\u003c/li\u003e\n \u003cli\u003eJia L, Quan M, Fu Y, Zhao T, Li Y, Wei C, et al. Dementia in China: epidemiology, clinical management, and research advances. The Lancet Neurology. 2020;19(1):81-92.\u003c/li\u003e\n \u003cli\u003eLi X, Feng X, Sun X, Hou N, Han F, Liu Y. Global, regional, and national burden of Alzheimer\u0026rsquo;s disease and other dementias, 1990\u0026ndash;2019. Frontiers in Aging Neuroscience. 2022;14:937486.\u003c/li\u003e\n \u003cli\u003eCao L, Gu D, Ni Y, Xie G. Automatic ICD code assignment based on ICD\u0026rsquo;s hierarchy structure for Chinese electronic medical records. AMIA Summits on Translational Science Proceedings. 2019;2019:417.\u003c/li\u003e\n \u003cli\u003eRen R, Qi J, Lin S, Liu X, Yin P, Wang Z, et al. The China alzheimer report 2022. General Psychiatry. 2022;35(1)\u003c/li\u003e\n \u003cli\u003eDong X, Wang Y. The geography of healthcare: Mapping patient flow and medical resource allocation in China. Economics \u0026amp; Human Biology. 2024;55:101431.\u003c/li\u003e\n \u003cli\u003eNational Healthcare Security Administration. Statistical Bulletin on the Development of China\u0026rsquo;s Health Undertakings in 2023; 2024. Accessed: (7 February 2025).\u0026nbsp;https://www.nhsa.gov.cn/art/2024/7/25/art_7_13340.html.\u003c/li\u003e\n \u003cli\u003eSong T, Ma R, Zhang X, Lv B, Li Z, Guo M, et al. Analysis of the current status and influencing factors of cross-regional hospitalization services utilization by basic medical insurance participants in China- taking a central province as an example. Frontiers in Public Health. 2023;11:1246982.\u003c/li\u003e\n \u003cli\u003eYan X, Shan L, He S, Zhang J. Cross-city patient mobility and healthcare equity and efficiency: Evidence from Hefei, China. Travel Behaviour and Society. 2022;28:1-12.\u003c/li\u003e\n \u003cli\u003eLi Y, Liu X, Kang L, Li J. Validation and Comparison of Four Mortality Prediction Models in a Geriatric Ward in China. Clinical Interventions in Aging. 2023:200919.\u003c/li\u003e\n \u003cli\u003eMcAlister F, Savu A, Ezekowitz J, Armstrong P, Kaul P. The hospital frailty risk score in patients with heart failure is strongly associated with outcomes but less so with pharmacotherapy. Journal of Internal Medicine. 2020;287(3):322-32.\u003c/li\u003e\n \u003cli\u003eHarvey LA, Toson B, Norris C, Harris IA, Gandy RC, Close JJ. Does identifying frailty from ICD-10 coded data on hospital admission improve prediction of adverse outcomes in older surgical patients? A population-based study. Age and ageing. 2021;50(3):802-8.\u003c/li\u003e\n \u003cli\u003eVermeiren S, Vella-Azzopardi R, Beckwee D, Habbig AK, Scafoglieri A, Jansen B, et al. Frailty and the prediction of negative health outcomes: a meta-analysis. Journal of the American medical directors association. 2016;17(12):1163-e1.\u003c/li\u003e\n \u003cli\u003eWeng L, Hu Y, Sun Z, Yu C, Guo Y, Pei P, et al. Place of death and phenomenon of going home to die in Chinese adults: a prospective cohort study. The Lancet Regional Health\u0026ndash;Western Pacific. 2022;18.\u003c/li\u003e\n \u003cli\u003eCesari M, Prince M, Thiyagarajan JA, De Carvalho IA, Bernabei R, Chan P, et al. Frailty: an emerging public health priority. Journal of the American Medical Directors Association. 2016;17(3):188-92.\u003c/li\u003e\n \u003cli\u003eGilbert T, Cordier Q, Polazzi S, Street A, Conroy S, Duclos A. Combining the hospital frailty risk score with the Charlson and Elixhauser multimorbidity indices to identify older patients at risk of poor outcomes in acute care. Medical Care. 2024;62(2):117-24.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"european-geriatric-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"EGEM","sideBox":"Learn more about [European Geriatric Medicine](https://www.springer.com/journal/41999)","snPcode":"41999","submissionUrl":"https://www.editorialmanager.com/egem/default2.aspx","title":"European Geriatric Medicine","twitterHandle":"","acdcEnabled":false,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Hospital Frailty Risk Score, HFRS, frailty risk assessment, length of stay, hospital costs","lastPublishedDoi":"10.21203/rs.3.rs-5551082/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5551082/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e \u003cp\u003eTo validate the Hospital Frailty Risk Score (HFRS) in Chinese hospital settings, describing how patients are allocated to frailty risk groups and how frailty risk is associated with length of stay (LoS) and hospital costs.\u003c/p\u003e\u003ch2\u003eDesign:\u003c/h2\u003e \u003cp\u003eRetrospective observational study.\u003c/p\u003e\u003ch2\u003eSetting:\u003c/h2\u003e \u003cp\u003e48 hospitals in Lvliang City, Shanxi Province, China.\u003c/p\u003e\u003ch2\u003eSubjects:\u003c/h2\u003e \u003cp\u003ePatients aged 75 years or older hospitalised between 1 January 2022 and 31 December 2023 (n\u0026thinsp;=\u0026thinsp;34,731).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA logistic regression model examined the association between long length of stay (LoS) and frailty risk. A generalised linear model assessed the association between hospital costs and frailty risk. Subgroup analyses of age group, sex, and hospital tiers were conducted.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003e22.2% of patients were categorised as having zero risk, 62.4% as low risk, 15.3% as intermediate risk, and 0.08% as high risk. Compared to the zero risk group: for those with low risk the probability of long LoS was 1.92 (95% CI 1.79\u0026ndash;2.06) times higher and hospital costs were \u0026yen;1,926 (95% CI 1,655-2,197) higher; for those with intermediate risk, the probability of long LoS was 2.7 (95% CI 2.49\u0026ndash;2.96) times higher and hospital costs were \u0026yen;4,284 (95% CI 3,916-4,653) higher; and for those with high risk the probability of long LoS was 6.7 (95% CI 3.06\u0026ndash;14.43) times higher and hospital costs were \u0026yen;16,613 (95% CI 12,827\u0026thinsp;\u0026minus;\u0026thinsp;20,399) higher. The explanatory power of the HFRS held across subgroups.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eCompared to patients aged 75\u0026thinsp;+\u0026thinsp;elsewhere, those in China had lower frailty risk scores, likely reflecting a younger age structure and recording of fewer diagnosis codes. Even so, the HFRS is a powerful predictor of long length of stay and hospital costs in China.\u003c/p\u003e","manuscriptTitle":"Validation of the Hospital Frailty Risk Score in China","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-10 05:07:57","doi":"10.21203/rs.3.rs-5551082/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2025-04-07T04:08:53+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-06T00:57:54+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"European Geriatric Medicine","date":"2025-04-05T21:55:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"European Geriatric Medicine","date":"2025-04-04T08:55:23+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"european-geriatric-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"EGEM","sideBox":"Learn more about [European Geriatric Medicine](https://www.springer.com/journal/41999)","snPcode":"41999","submissionUrl":"https://www.editorialmanager.com/egem/default2.aspx","title":"European Geriatric Medicine","twitterHandle":"","acdcEnabled":false,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"2e9c0383-2a3a-4823-b0f1-37554b39b429","owner":[],"postedDate":"April 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-04-10T05:07:57+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-10 05:07:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5551082","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5551082","identity":"rs-5551082","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}