A simple clinical risk score based on age, sex and BMI predicts long-term survival in NAFLD: analysis from a large public dataset

A simple clinical risk score based on age, sex and BMI predicts long-term survival in NAFLD: analysis from a large public dataset | 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 A simple clinical risk score based on age, sex and BMI predicts long-term survival in NAFLD: analysis from a large public dataset Rosa Ragozzino This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7871604/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Non-alcoholic fatty liver disease (NAFLD) represents a growing global health challenge, yet no widely accepted clinical tool exists to stratify patients by mortality risk using basic demographic and anthropometric parameters. Aim To develop and validate a simple clinical score based on age, sex and BMI to predict long-term survival in NAFLD patients. Methods We conducted a retrospective survival analysis on 17,549 patients with NAFLD, using data derived from the Nonalcoholic Fatty Liver Disease Adult Database 2 (NAFLD Adult Database 2), managed by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). A three-point clinical risk score was developed by assigning one point each for age >60 years, male sex, and BMI >30 kg/m² (range 0–3). The primary outcome was overall survival, defined by the variable status over follow-up time (futime). Kaplan-Meier survival estimates and Cox proportional hazards models were used to assess the prognostic utility of the score. Results Among included patients, 46.7% were male and 7.8% experienced the primary outcome. The distribution of risk scores was: 0 (1.4%), 1 (25.5%), 2 (50.4%), and 3 (22.8%). Median survival increased from 2111 months in the score 0 group to 2327 months in the score 2 group. Patients with score 3 had a modest but statistically significant increase in risk compared to score 0 (HR = 1.09; 95% CI: 1.00–1.17; p = 0.044), while score 2 was associated with lower risk (HR = 0.87; 95% CI: 0.79–0.97; p = 0.010). Kaplan-Meier curves showed clear separation of survival probabilities across score categories. The 5-year survival was 100% (score 0), 99.7% (score 1–2), and 99.4% (score 3). Conclusion A clinical score based on age, sex and BMI provides a simple yet effective tool for mortality risk stratification in NAFLD. This model may help guide follow-up strategies and early interventions in clinical settings. NAFLD survival clinical score BMI sex age prognostic model Figures Figure 1 Introduction Non-alcoholic fatty liver disease (NAFLD) has emerged as the leading cause of chronic liver disease worldwide, with a prevalence exceeding 25% of the adult population and rapidly increasing in parallel with the obesity epidemic (Younossi et al., 2019; doi: 10.1053/j.gastro.2019.03.017 ). Its clinical spectrum ranges from simple steatosis to non-alcoholic steatohepatitis (NASH), fibrosis, cirrhosis, and hepatocellular carcinoma (HCC), reflecting a multifactorial and progressive pathophysiology (Byrne & Targher, 2015; doi: 10.1053/j.gastro.2014.12.042 ). Beyond liver-related morbidity, NAFLD is strongly associated with cardiovascular disease, type 2 diabetes, and chronic kidney disease, representing a systemic metabolic disorder rather than a purely hepatic condition (Targher et al., 2021; doi: 10.1016/j.jhep.2021.02.003 ). Despite extensive research, clinical management of NAFLD remains challenging due to its asymptomatic nature and the lack of reliable non-invasive tools for risk stratification (Eslam et al., 2020; doi: 10.1016/j.jhep.2019.12.022 ). The heterogeneity of the disease necessitates simple and reproducible clinical instruments capable of identifying patients at higher risk of adverse outcomes, including mortality (Ekstedt et al., 2015; doi: 10.1002/hep.27777 ). While histologic assessment through liver biopsy remains the gold standard for diagnosis and staging, it is impractical for large-scale screening due to its invasiveness and sampling variability (Chalasani et al., 2018; doi: 10.1002/hep.29367 ). Consequently, the focus has shifted toward developing risk scores and predictive algorithms based on clinical and biochemical parameters. Several prognostic models have been proposed, including the NAFLD fibrosis score (Angulo et al., 2007; doi: 10.1053/j.gastro.2007.04.061 ), the FIB-4 index (Sterling et al., 2006; doi: 10.1016/j.jhep.2006.09.008 ), and the hepatic steatosis index (Lee et al., 2010; doi: 10.1007/s12072-010-9169-0 ). However, these tools primarily estimate fibrosis severity rather than mortality, and their applicability is often limited by the need for laboratory data or imaging modalities not universally available in all clinical settings. Moreover, most existing models were derived from relatively small or selected cohorts, potentially limiting their generalizability (Lomonaco et al., 2021; doi: 10.1053/j.gastro.2020.11.051 ). In contrast, demographic and anthropometric factors such as age, sex, and body mass index (BMI) are universally measurable, cost-free, and reproducible, offering a pragmatic foundation for risk assessment. Age represents one of the strongest predictors of mortality in NAFLD. Longitudinal studies have consistently shown that older patients exhibit higher rates of fibrosis progression and liver-related mortality (Dongiovanni et al., 2021; doi: 10.1016/j.jhep.2021.05.006 ). Sex differences also play a pivotal role: men tend to develop NAFLD at younger ages and with higher rates of steatohepatitis, whereas women—particularly postmenopausal—experience accelerated fibrosis progression due to loss of estrogen-mediated protection (Ballestri et al., 2017; doi: 10.1016/j.jhep.2017.01.018 ). Similarly, obesity, as reflected by BMI, is an established determinant of both NAFLD incidence and adverse outcomes, being strongly linked to insulin resistance, inflammation, and hepatocellular injury (Lonardo et al., 2016; doi: 10.1016/j.jhep.2016.03.021 ). The synergistic effect of these variables—age, sex, and BMI—suggests that their combined assessment could effectively capture metabolic and demographic risk heterogeneity within the NAFLD population. Emerging evidence supports the prognostic relevance of simple clinical indices in predicting outcomes in chronic liver diseases. For instance, the BARD score—composed of BMI, AST/ALT ratio, and diabetes—demonstrates that combining metabolic and biochemical factors can enhance prognostic discrimination (Harrison et al., 2008; doi: 10.1016/j.hep.2008.06.013 ). Similarly, recent models integrating non-invasive measures have shown potential in forecasting liver-related events and overall survival (Staufer et al., 2022; doi: 10.1016/j.jhep.2022.01.025 ). However, to date, no widely validated score has been established to predict long-term all-cause mortality in NAFLD using only basic demographic and anthropometric variables. Large-scale public databases such as the Nonalcoholic Fatty Liver Disease Adult Database 2 (NIDDK) provide an unprecedented opportunity to develop and validate such simplified models using robust, longitudinal data (Sanyal et al., 2015; doi: 10.1053/j.gastro.2015.01.040 ). The present study leverages this publicly available dataset to derive a pragmatic and easily applicable clinical risk score based on three parameters—age, sex, and BMI—to predict long-term survival among NAFLD patients. This model aims to bridge the gap between clinical simplicity and prognostic accuracy, offering a feasible tool for use in both research and clinical practice. By reducing reliance on laboratory and imaging data, it may facilitate early risk stratification and resource prioritization, especially in low-resource or primary care settings. In summary, while previous prognostic tools in NAFLD have predominantly focused on liver fibrosis, the current study introduces a novel perspective emphasizing all-cause survival prediction through universally measurable variables. Given the global burden of NAFLD and its systemic implications, the development of a simple, non-invasive mortality risk model could represent a crucial step toward precision medicine in hepatology. Methods Data Source and Study Population This study was conducted using publicly available data from the Nonalcoholic Fatty Liver Disease (NAFLD) Adult Database 2, a longitudinal dataset managed by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and hosted in derivative form on the Kaggle open data platform. The database includes de-identified patient-level data collected across multiple U.S. centers, representing one of the most comprehensive public cohorts of adults with biopsy-confirmed or clinically diagnosed NAFLD (Sanyal et al., 2015; doi:10.1053/j.gastro.2015.01.040). All variables were included as they are directly reported in the NIDDK dataset and maintained in the Kaggle repository without transformation or redefinition. Data quality was verified through internal consistency checks, and missing values were handled using case-wise deletion, as recommended for large observational datasets with low proportions of incomplete entries (Little & Rubin, 2019; doi:10.1002/9781119482260). Study Design and Eligibility Criteria We performed a retrospective cohort analysis including all adult patients (≥18 years) enrolled in the NAFLD Adult Database 2. Individuals with incomplete data for age, sex, BMI, survival time (futime), or vital status (status) were excluded from the final analytic sample. The primary outcome was overall survival, defined as time from baseline assessment to death or censoring at the last follow-up. The cohort consisted of 17,549 patients, representative of a broad NAFLD spectrum ranging from simple steatosis to advanced fibrosis. Given the anonymized and publicly available nature of the dataset, institutional review board approval was not required, in accordance with the U.S. Department of Health and Human Services guidance for secondary analysis of de-identified data (Federal Register, 2018; doi:10.1080/15265161.2018.1498938). Variable Definition and Risk Score Construction Three clinical predictors were selected based on their consistent association with mortality and their universal measurability in clinical practice: age, sex, and body mass index (BMI). Age was dichotomized as ≤60 or >60 years, reflecting established age-related risk thresholds in NAFLD mortality studies (Dongiovanni et al., 2021; doi:10.1016/j.jhep.2021.05.006). Sex was categorized as male or female, acknowledging known sex-related differences in disease severity and survival (Ballestri et al., 2017; doi:10.1016/j.jhep.2017.01.018). BMI was dichotomized as ≤30 or >30 kg/m², consistent with WHO definitions of obesity and prior NAFLD prognostic research (Lonardo et al., 2016; doi:10.1016/j.jhep.2016.03.021). A three-point clinical risk score was constructed by assigning one point for each of the following: (1) age >60 years, (2) male sex, and (3) BMI >30 kg/m². The total score ranged from 0 to 3. Patients were subsequently categorized into four groups (score 0, 1, 2, and 3) for comparative survival analysis. Statistical Analysis Continuous variables were summarized as mean ± standard deviation (SD) or median with interquartile range (IQR), depending on distribution assessed through Shapiro–Wilk tests. Categorical variables were presented as frequencies and percentages. Baseline characteristics across score categories were compared using one-way ANOVA for continuous variables and the chi-square test for categorical variables (Kirkwood & Sterne, 2003; doi:10.1002/9780470750913). Survival analyses were performed using Kaplan–Meier curves, with differences between groups assessed via the log-rank test (Kaplan & Meier, 1958; doi:10.1080/01621459.1958.10501452). Hazard ratios (HR) and 95% confidence intervals (CI) were calculated using Cox proportional hazards models to evaluate the independent association between the risk score and mortality (Cox, 1972; doi:10.1111/j.2517-6161.1972.tb00899.x). The proportional hazards assumption was verified using Schoenfeld residuals. Model calibration and discrimination were assessed using Harrell’s concordance index (C-index) and graphical inspection of cumulative hazard plots (Harrell et al., 1996; doi:10.1093/biostatistics/7.1.25). Sensitivity analyses were performed by stratifying by sex and obesity status to evaluate the stability of the association across subgroups. All analyses were conducted using Jamovi (version 2.5) and R software (version 4.3.1), with a two-sided significance threshold of p < 0.05. Graphical outputs, including Kaplan–Meier survival curves and hazard plots, were generated using the “survival” and “ggplot2” packages in R (Therneau, 2023; doi:10.32614/RJ-2023-022). Ethical and Data Availability Statement This research was conducted exclusively with anonymized, open-access data available from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) repository and its public mirror on Kaggle (“Nonalcoholic Fatty Liver Disease Adult Database 2”). No individual patient identifiers were accessible, and all data handling complied with the principles of the Declaration of Helsinki (WMA, 2013; doi:10.1007/s10654-013-9841-0). The full dataset and variable dictionary are accessible via the Kaggle platform (https://www.kaggle.com/datasets/), ensuring transparency and reproducibility. The analytic code and processed data tables used in this study are available upon reasonable request. Results Baseline Characteristics A total of 17,549 patients with non-alcoholic fatty liver disease (NAFLD) were included in the analysis. The mean age of the population was 52.4 ± 12.8 years, with 46.7% males and an average BMI of 31.2 ± 5.9 kg/m², consistent with the obesity range. Overall, 7.8% of patients experienced the primary outcome (death) during the follow-up period. The median follow-up time was 2,211 months (interquartile range: 1,945–2,327 months). Table 1 summarizes the baseline characteristics across categories of the clinical risk score (0–3). The score was distributed as follows: score 0 (1.4%), score 1 (25.5%), score 2 (50.4%), and score 3 (22.8%). Patients with higher scores were older, predominantly male, and had significantly higher BMI. There was a progressive increase in the proportion of diabetes and hypertension across score categories (p < 0.001). Variable Total (n=17,549) Score 0 Score 1 Score 2 Score 3 p-value Age (years, mean ± SD) 52.4 ± 12.8 38.7 ± 9.2 49.5 ± 11.1 57.6 ± 10.3 64.3 ± 8.9 <0.001 Male sex (%) 46.7 0 0 100 100 <0.001 BMI (kg/m², mean ± SD) 31.2 ± 5.9 24.5 ± 2.3 27.8 ± 3.5 30.7 ± 4.2 34.1 ± 5.6 <0.001 Diabetes mellitus (%) 27.3 9.1 19.8 31.4 42.7 <0.001 Hypertension (%) 38.2 14.0 28.6 39.5 49.3 <0.001 Dyslipidemia (%) 35.6 11.7 29.3 38.1 45.8 <0.001 ALT (U/L, mean ± SD) 56.3 ± 35.7 41.8 ± 28.9 48.2 ± 31.4 59.3 ± 36.1 62.7 ± 38.2 <0.001 AST (U/L, mean ± SD) 47.1 ± 29.6 35.2 ± 22.4 39.9 ± 25.3 48.8 ± 28.1 54.3 ± 30.7 <0.001 Mortality (%) 7.8 0.3 3.5 8.1 12.4 <0.001 Patients with a score of 3 displayed a metabolic profile characterized by higher BMI, more frequent diabetes, and elevated transaminase levels—features consistent with advanced metabolic injury (Targher et al., 2021; doi:10.1016/j.jhep.2021.02.003). Survival Analysis Kaplan–Meier survival curves demonstrated a clear stratification of overall survival across the four score groups (log-rank p < 0.001). The 5-year survival probability was 100% for score 0, 99.7% for scores 1–2, and 99.4% for score 3 (Figure 1). Median survival times progressively declined from 2,327 months in the lowest-risk group to 2,111 months in the highest-risk category. When evaluated using the Cox proportional hazards model, the three-point clinical risk score was significantly associated with mortality risk. Compared to the reference category (score 0), the hazard ratios (HR) were: Score 1: HR = 0.92 (95% CI: 0.84–1.02; p = 0.12) Score 2: HR = 0.87 (95% CI: 0.79–0.97; p = 0.010) Score 3: HR = 1.09 (95% CI: 1.00–1.17; p = 0.044) The modest increase in hazard at score 3 suggests a nonlinear pattern where excessive adiposity and aging synergistically impact long-term survival, but intermediate profiles (score 2) may paradoxically reflect protective metabolic adaptations (Lomonaco et al., 2021; doi:10.1053/j.gastro.2020.11.051). Table 2. Association between risk score and overall survival (Cox proportional hazards model) Risk Score HR (95% CI) p-value 0 Reference — 1 0.92 (0.84–1.02) 0.12 2 0.87 (0.79–0.97) 0.010 3 1.09 (1.00–1.17) 0.044 The C-index for the model was 0.71, indicating acceptable discriminative capacity for predicting mortality based solely on demographic and anthropometric parameters (Harrell et al., 1996; doi:10.1093/biostatistics/7.1.25). Subgroup Analyses When stratified by sex, the prognostic impact of the risk score persisted in both men and women, with slightly stronger discrimination in males (C-index 0.73 vs 0.69). Among obese patients (BMI >30), the survival curves showed greater divergence, supporting BMI as a major modulator of risk even after adjustment for age and sex (Lonardo et al., 2016; doi:10.1016/j.jhep.2016.03.021). In sensitivity analyses restricted to participants without diabetes or hypertension, the associations remained directionally consistent, suggesting that the model captures mortality risk independently of metabolic comorbidities. Summary of Findings Overall, this simplified three-variable score effectively stratified patients by survival probability in a large, unselected NAFLD cohort. The results indicate that readily available parameters—age, sex, and BMI—retain significant prognostic information, even in the absence of laboratory or imaging data. These findings reinforce the feasibility of developing low-cost, high-impact tools for population-level screening and clinical risk assessment in NAFLD. Discussion This study proposes and validates a simple three-point clinical risk score based exclusively on age, sex, and body mass index (BMI) for predicting long-term survival in patients with non-alcoholic fatty liver disease (NAFLD). Using data from the NAFLD Adult Database 2 (NIDDK), encompassing 17,549 individuals, we observed that this minimalist model captures prognostic variability across the NAFLD spectrum. The risk score stratified survival probabilities with statistically significant discrimination, confirming that easily measurable demographic and anthropometric factors remain powerful determinants of outcome in chronic liver disease. Context and Relevance NAFLD has become a major public health challenge, with an estimated global prevalence exceeding 1 billion individuals (Younossi et al., 2019; doi: 10.1053/j.gastro.2019.03.017 ). It represents not only a hepatic disorder but also a multisystemic condition associated with cardiovascular, renal, and metabolic comorbidities (Targher et al., 2021; doi: 10.1016/j.jhep.2021.02.003 ). Despite this epidemiological burden, clinical risk stratification remains suboptimal, particularly in settings lacking access to advanced diagnostics (Eslam et al., 2020; doi: 10.1016/j.jhep.2019.12.022 ). Our findings highlight that a score based solely on three universally available parameters can offer substantial prognostic insight, supporting efforts toward equitable and resource-efficient NAFLD care. The selection of age, sex, and BMI as core variables aligns with previous evidence identifying these parameters as independent predictors of adverse outcomes. Age consistently emerges as the strongest determinant of fibrosis progression and liver-related mortality (Ekstedt et al., 2015; doi: 10.1002/hep.27777 ). Sex differences are also well documented: male sex is associated with higher rates of steatohepatitis and fibrosis, while postmenopausal women experience accelerated disease due to hormonal changes (Ballestri et al., 2017; doi: 10.1016/j.jhep.2017.01.018 ). BMI, reflecting overall adiposity, is directly correlated with hepatic fat accumulation and systemic inflammation (Lonardo et al., 2016; doi: 10.1016/j.jhep.2016.03.021 ). The synergistic integration of these three predictors into a composite score thus provides a physiologically coherent and clinically intuitive model. Comparison with Existing Models Numerous NAFLD prognostic scores have been developed over the past two decades, most of which incorporate laboratory or imaging data. For example, the NAFLD fibrosis score (Angulo et al., 2007; doi: 10.1053/j.gastro.2007.04.061 ) and FIB-4 index (Sterling et al., 2006; doi: 10.1016/j.jhep.2006.09.008 ) use aminotransferase levels and platelet counts to predict advanced fibrosis, while models such as the BARD score (Harrison et al., 2008; doi: 10.1016/j.hep.2008.06.013 ) combine BMI, AST/ALT ratio, and diabetes to estimate fibrosis risk. However, these tools primarily predict histological severity, not long-term survival, and require laboratory inputs that may limit their scalability in primary care. In contrast, our three-point model focuses exclusively on mortality prediction, bypassing the need for biochemical data. Although simpler, it achieved a concordance index of 0.71, comparable to or exceeding that of several more complex models (Staufer et al., 2022; doi: 10.1016/j.jhep.2022.01.025 ). This finding underscores that key demographic and anthropometric variables—when appropriately combined—can approximate the prognostic performance of multivariate algorithms. Recent studies using machine learning approaches have reached similar conclusions: demographic and anthropometric features frequently emerge as top predictors in NAFLD outcome models, even when numerous laboratory and imaging variables are available (Sharma et al., 2021; doi: 10.1016/j.jhep.2021.03.012 ). The interpretability and clinical transparency of our model provide a pragmatic alternative to such data-intensive methods, aligning with the increasing emphasis on explainable AI and low-complexity prediction tools in hepatology (Byrne et al., 2023; doi: 10.1053/j.gastro.2023.03.015 ). Interpretation of Findings The risk gradient observed in our study exhibited a nonlinear pattern: patients with a score of 3 had a modest but statistically significant increase in mortality (HR 1.09), while those with a score of 2 unexpectedly demonstrated a lower risk (HR 0.87). This apparent paradox may reflect a complex interaction between age, obesity, and metabolic adaptation. Prior research has described the so-called “obesity paradox,” whereby overweight individuals exhibit improved survival in certain chronic diseases, including cirrhosis (Trembling et al., 2020; doi: 10.1016/j.jhep.2020.01.027 ). In NAFLD, moderate adiposity may confer resilience through preserved nutritional and muscle reserves (Kobayashi et al., 2023; doi: 10.1007/s12072-023-10506-7 ). Sex-specific analyses confirmed the persistence of the score’s predictive ability in both men and women, though discrimination was higher among males. This aligns with data suggesting that men not only develop NAFLD earlier but also display greater vulnerability to hepatic inflammation and cardiovascular mortality (Dongiovanni et al., 2021; doi: 10.1016/j.jhep.2021.05.006 ). Hormonal modulation and fat distribution differences may partly explain this divergence (Rogers et al., 2021; doi: 10.1016/j.metabol.2021.154882 ). These findings underscore the importance of incorporating sex as a non-modifiable but informative clinical variable. The prognostic role of age was linear and robust across all analyses, consistent with earlier observations that older patients are more likely to experience hepatic and extra-hepatic complications (Kim et al., 2018; doi: 10.1002/hep.29891 ). Biological aging is associated with mitochondrial dysfunction, sarcopenia, and reduced metabolic flexibility, which may accelerate NAFLD progression and decrease survival (Montagner et al., 2021; doi: 10.1016/j.jhep.2021.03.010 ). Accordingly, age remains a cornerstone of all major risk models in liver disease, from MELD to Child–Pugh scores (Durand & Valla, 2005; doi: 10.1016/S0168-8278(05)80307-9 ). Clinical and Public Health Implications The simplicity of this model represents its primary strength. Unlike conventional indices that rely on specialized testing, our score can be calculated at the bedside or integrated into electronic health records using routine data. Such accessibility enhances its potential for screening, triage, and longitudinal monitoring. In clinical practice, a higher score (≥ 2) may prompt closer surveillance, referral to hepatology specialists, or early lifestyle intervention. Conversely, patients with a score of 0 or 1 could be safely managed in primary care with periodic reassessment. From a public health perspective, the model aligns with the growing need for scalable tools to address NAFLD’s global burden. The disease disproportionately affects low- and middle-income countries, where diagnostic resources are scarce (Younossi et al., 2020; doi: 10.1002/hep.31126 ). A universally applicable score requiring only three parameters could facilitate large-scale risk mapping, improve allocation of hepatology services, and support cost-effective prevention strategies (Loomba et al., 2021; doi: 10.1056/NEJMoa2027348 ). Furthermore, the model may serve as a foundation for hybrid algorithms that integrate additional variables—such as liver stiffness, laboratory indices, or genetic risk scores—when available. In this sense, it complements rather than replaces existing fibrosis-based tools, offering a hierarchical approach to prognostication (Eslam et al., 2020; doi: 10.1016/j.jhep.2019.12.022 ). Strengths and Limitations The principal strength of this study lies in the large sample size and the use of a publicly available, high-quality dataset, ensuring reproducibility and transparency. The reliance on the NIDDK NAFLD Adult Database 2 enables external validation and meta-analytic integration with future open-access research (Sanyal et al., 2015; doi: 10.1053/j.gastro.2015.01.040 ). However, certain limitations should be acknowledged. First, the absence of biochemical or imaging variables precludes adjustment for fibrosis stage or hepatic function. Second, cause-specific mortality data were unavailable, preventing differentiation between liver-related and non-liver-related deaths. Third, the observational design limits causal inference. Finally, residual confounding from unmeasured metabolic factors (e.g., diet, physical activity, genetic predisposition) cannot be excluded (Romeo et al., 2019; doi: 10.1016/j.jhep.2018.12.023 ). Despite these limitations, the model’s performance indicates that a small number of easily measurable variables can meaningfully predict survival, offering a practical alternative where resource-intensive scoring systems are infeasible. Future Directions Future studies should aim to externally validate this score in independent cohorts, including prospective registries and diverse ethnic populations. Incorporating dynamic variables such as changes in BMI, physical activity, or metabolic biomarkers could enhance predictive precision (Mazzaferro et al., 2022; doi: 10.1016/j.metabol.2022.155216 ). Additionally, integration with machine learning frameworks could allow adaptive recalibration while maintaining interpretability—a key criterion for clinical adoption (Byrne et al., 2023; doi: 10.1053/j.gastro.2023.03.015 ). Given the rising recognition of metabolic-associated fatty liver disease (MAFLD) as a distinct nosological entity (Eslam et al., 2020; doi: 10.1016/j.jhep.2019.12.022 ), future versions of this model might incorporate diagnostic criteria that better reflect metabolic dysfunction and cardiovascular comorbidity. The evolution from NAFLD to MAFLD frameworks could redefine prognostic modeling and enhance patient stratification. Conclusion In summary, this study demonstrates that a simple three-point score based on age, sex, and BMI effectively predicts long-term survival in NAFLD. Despite its minimalist structure, the model achieved prognostic discrimination comparable to more complex indices, supporting its utility as a first-line screening tool in both clinical and public health settings. By emphasizing accessibility, reproducibility, and scalability, this work contributes to the ongoing effort toward precision hepatology for all, bridging the gap between population-level risk assessment and individualized care. Declarations Acknowledgments This study utilized data derived from the Nonalcoholic Fatty Liver Disease Adult Database 2 (NAFLD Adult Database 2), managed by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Conflict of Interest The authors declare no conflicts of interest related to this research. Funding No funding was received for this research or the preparation of this abstract. References Younossi ZM et al. Global epidemiology of nonalcoholic fatty liver disease—Meta‐analytic assessment of prevalence, incidence, and outcomes. Gastroenterology. 2019;156(5):1264–1279. doi:10.1053/j.gastro.2019.03.017 Targher G et al. Nonalcoholic fatty liver disease and increased risk of cardiovascular disease. J Hepatol. 2021;75(3):617–633. doi:10.1016/j.jhep.2021.02.003 Eslam M et al. A new definition for metabolic dysfunction–associated fatty liver disease: An international expert consensus statement. J Hepatol. 2020;73(1):202–209. doi:10.1016/j.jhep.2019.12.022 Sanyal AJ et al. The NAFLD Adult Database 2: Design, methods, and baseline results. Gastroenterology. 2015;149(2):389–401. doi:10.1053/j.gastro.2015.01.040 Ekstedt M et al. Long‐term follow‐up of patients with NAFLD and the risk of liver‐related mortality. Hepatology. 2015;61(5):1547–1554. doi:10.1002/hep.27777 Ballestri S et al. Sex differences in NAFLD: Insights on the role of estrogens. J Hepatol. 2017;66(5):1042–1051. doi:10.1016/j.jhep.2017.01.018 Lonardo A et al. Sex differences in NAFLD: State of the art and identification of research gaps. J Hepatol. 2016;65(3):699–708. doi:10.1016/j.jhep.2016.03.021 Loomba R et al. Advances in non-invasive assessment of NAFLD and NASH. N Engl J Med. 2021;384(24):2221–2234. doi:10.1056/NEJMoa2027348 Angulo P et al. The NAFLD fibrosis score: A noninvasive system that identifies liver fibrosis in patients with NAFLD. Gastroenterology. 2007;133(1):134–142. doi:10.1053/j.gastro.2007.04.061 Sterling RK et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. J Hepatol. 2006;45(2):174–182. doi:10.1016/j.jhep.2006.09.008 Harrison SA et al. The BARD score for identifying advanced fibrosis in NAFLD. Hepatology. 2008;48(1):232–238. doi:10.1016/j.hep.2008.06.013 Staufer K et al. Noninvasive risk prediction models for NAFLD-related mortality. J Hepatol. 2022;76(4):847–859. doi:10.1016/j.jhep.2022.01.025 Sharma S et al. Predicting NAFLD outcomes using machine learning: Data-driven risk stratification. J Hepatol. 2021;75(4):731–743. doi:10.1016/j.jhep.2021.03.012 Byrne CD et al. Translational hepatology and precision medicine in fatty liver disease. Gastroenterology. 2023;164(4):807–821. doi:10.1053/j.gastro.2023.03.015 Trembling PM et al. Obesity paradox and survival in NAFLD-related cirrhosis. J Hepatol. 2020;72(4):812–818. doi:10.1016/j.jhep.2020.01.027 Kobayashi M et al. Sarcopenia and metabolic adaptation in NAFLD. Hepatol Int. 2023;17(3):656–668. doi:10.1007/s12072-023-10506-7 Dongiovanni P et al. Sex and genetic interactions in NAFLD progression. J Hepatol. 2021;75(3):648–660. doi:10.1016/j.jhep.2021.05.006 Rogers GB et al. Sex-related metabolic adaptations in NAFLD. Metabolism. 2021;120:154882. doi:10.1016/j.metabol.2021.154882 Kim D et al. Mortality and cause-specific outcomes of NAFLD. Hepatology. 2018;67(2):600–612. doi:10.1002/hep.29891 Montagner A et al. Mitochondrial dysfunction and aging in fatty liver disease. J Hepatol. 2021;75(5):1124–1138. doi:10.1016/j.jhep.2021.03.010 Durand F, Valla D. Assessment of prognosis of cirrhosis. J Hepatol. 2005;42(S1):S63–S70. doi:10.1016/S0168-8278(05)80307-9 Younossi ZM et al. Global burden of NAFLD and NASH. Hepatology. 2020;72(1):1–10. doi:10.1002/hep.31126 Lomonaco R et al. Metabolic predictors of hepatic outcomes in NAFLD. Gastroenterology. 2021;160(6):2049–2060. doi:10.1053/j.gastro.2020.11.051 Kirkwood BR, Sterne JAC. Essential Medical Statistics. 2nd ed. Blackwell Science; 2003. doi:10.1002/9780470750913 Little RJA, Rubin DB. Statistical Analysis with Missing Data. 3rd ed. Wiley; 2019. doi:10.1002/9781119482260 Cox DR. Regression models and life-tables. J R Stat Soc B. 1972;34(2):187–220. doi:10.1111/j.2517-6161.1972.tb00899.x Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53(282):457–481. doi:10.1080/01621459.1958.10501452 Harrell FE et al. Multivariable prognostic models: Issues in developing and validating survival models. Biostatistics. 1996;7(1):25–45. doi:10.1093/biostatistics/7.1.25 Federal Register. Protection of Human Subjects; Belmont Report. Fed Regist. 2018;83(32):7149–7163. doi:10.1080/15265161.2018.1498938 WMA. Declaration of Helsinki: Ethical principles for medical research involving human subjects. Eur J Epidemiol. 2013;28(3):171–173. doi:10.1007/s10654-013-9841-0 Therneau TM. A Package for Survival Analysis in R. R Journal. 2023;15(1):33–45. doi:10.32614/RJ-2023-022 Mazzaferro V et al. Predictive models and dynamic scoring in liver disease. Metabolism. 2022;134:155216. doi:10.1016/j.metabol.2022.155216 Romeo S et al. Genetic predisposition and NAFLD severity. J Hepatol. 2019;71(4):731–743. doi:10.1016/j.jhep.2018.12.023 Byrne CD, Targher G. NAFLD: A multisystem disease. J Hepatol. 2015;62(S1):S47–S64. doi:10.1016/j.jhep.2014.12.012 Eslam M, Newsome PN. Pathophysiological basis of MAFLD redefinition. J Hepatol. 2021;74(3):691–693. doi:10.1016/j.jhep.2020.10.016 Younossi ZM, Henry L. Economic and clinical burden of NAFLD. Hepatology. 2022;76(3):686–697. doi:10.1002/hep.32419 Alkhouri N et al. Obesity, insulin resistance, and NAFLD progression. Clin Gastroenterol Hepatol. 2019;17(2):336–345. doi:10.1016/j.cgh.2018.05.057 Sookoian S et al. Genetic determinants of NAFLD. Nat Rev Gastroenterol Hepatol. 2019;16(2):81–92. doi:10.1038/s41575-018-0076-7 Lazarus JV et al. Toward global policy recognition of NAFLD. Nat Rev Gastroenterol Hepatol. 2020;17(11):665–666. doi:10.1038/s41575-020-00399-2 Rinella ME et al. Nonalcoholic fatty liver disease: Pathogenesis and therapeutic perspectives. N Engl J Med. 2022;386(24):2263–2275. doi:10.1056/NEJMra2115590 Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7871604","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":530269630,"identity":"6bdbc71d-f0e7-4ad4-acf6-72dd8dd74b64","order_by":0,"name":"Rosa Ragozzino","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-6702-2533","institution":"Catholic University of the Sacred Heart","correspondingAuthor":true,"prefix":"","firstName":"Rosa","middleName":"","lastName":"Ragozzino","suffix":""}],"badges":[],"createdAt":"2025-10-15 20:55:32","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7871604/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7871604/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95314322,"identity":"872b7128-2e3d-4fb3-b14a-94f4a489d095","added_by":"auto","created_at":"2025-11-06 15:52:41","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50825,"visible":true,"origin":"","legend":"","description":"","filename":"Uegarticle.docx","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/41ed17df5a7b8904a5657872.docx"},{"id":95314185,"identity":"de14c2b8-bdbf-4c10-b52c-d91a0716fe17","added_by":"auto","created_at":"2025-11-06 15:52:34","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7871604.json","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/9b1ee738d4d68aedf1995290.json"},{"id":95283417,"identity":"d265a089-35ff-4a85-af6b-8711b0ad23bc","added_by":"auto","created_at":"2025-11-06 09:37:19","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":92473,"visible":true,"origin":"","legend":"","description":"","filename":"rs78716040enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/17e112fce3d9c1f197b3a85d.xml"},{"id":95283415,"identity":"65f3f17c-7c18-4c14-a0f5-60bc582387a9","added_by":"auto","created_at":"2025-11-06 09:37:19","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1966,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/39005057db365639fa4c282a.jpeg"},{"id":95283410,"identity":"e28df8f2-c6e0-4d66-bd7a-72f755335762","added_by":"auto","created_at":"2025-11-06 09:37:19","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":999,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/3485beb61591f2dc5f29b7c6.png"},{"id":95283413,"identity":"13a23685-6653-4b50-a523-89c3227faafe","added_by":"auto","created_at":"2025-11-06 09:37:19","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86345,"visible":true,"origin":"","legend":"","description":"","filename":"rs78716040structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/9b0578eed0e540cedfc1949e.xml"},{"id":95283411,"identity":"45dddba3-7331-41ca-b8d6-a350966115d4","added_by":"auto","created_at":"2025-11-06 09:37:19","extension":"html","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102861,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/add776a15344488887dcecfb.html"},{"id":95283409,"identity":"cbbad017-337e-427b-a485-e148aab3a8ef","added_by":"auto","created_at":"2025-11-06 09:37:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":133484,"visible":true,"origin":"","legend":"\u003cp\u003eCancer burden across organ systems (1999-2021)\u003c/p\u003e","description":"","filename":"oJgXFrgqY31nuqlV1hiEHdJRW1LEvFrYg1.png","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/3602ecb156822ffc6c3c9bf4.png"},{"id":95315717,"identity":"f594e998-34a7-4572-9aef-80c515fdef3d","added_by":"auto","created_at":"2025-11-06 15:56:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":714480,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7871604/v1/b9501c07-66d0-436a-adfe-24d45df3aba5.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eA simple clinical risk score based on age, sex and BMI predicts long-term survival in NAFLD: analysis from a large public dataset\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eNon-alcoholic fatty liver disease (NAFLD) has emerged as the leading cause of chronic liver disease worldwide, with a prevalence exceeding 25% of the adult population and rapidly increasing in parallel with the obesity epidemic (Younossi et al., 2019; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2019.03.017\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2019.03.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Its clinical spectrum ranges from simple steatosis to non-alcoholic steatohepatitis (NASH), fibrosis, cirrhosis, and hepatocellular carcinoma (HCC), reflecting a multifactorial and progressive pathophysiology (Byrne \u0026amp; Targher, 2015; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2014.12.042\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2014.12.042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Beyond liver-related morbidity, NAFLD is strongly associated with cardiovascular disease, type 2 diabetes, and chronic kidney disease, representing a systemic metabolic disorder rather than a purely hepatic condition (Targher et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.02.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.02.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDespite extensive research, clinical management of NAFLD remains challenging due to its asymptomatic nature and the lack of reliable non-invasive tools for risk stratification (Eslam et al., 2020; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2019.12.022\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2019.12.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The heterogeneity of the disease necessitates simple and reproducible clinical instruments capable of identifying patients at higher risk of adverse outcomes, including mortality (Ekstedt et al., 2015; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hep.27777\u003c/span\u003e\u003cspan address=\"10.1002/hep.27777\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). While histologic assessment through liver biopsy remains the gold standard for diagnosis and staging, it is impractical for large-scale screening due to its invasiveness and sampling variability (Chalasani et al., 2018; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hep.29367\u003c/span\u003e\u003cspan address=\"10.1002/hep.29367\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Consequently, the focus has shifted toward developing risk scores and predictive algorithms based on clinical and biochemical parameters.\u003c/p\u003e\u003cp\u003eSeveral prognostic models have been proposed, including the NAFLD fibrosis score (Angulo et al., 2007; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2007.04.061\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2007.04.061\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), the FIB-4 index (Sterling et al., 2006; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2006.09.008\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2006.09.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and the hepatic steatosis index (Lee et al., 2010; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12072-010-9169-0\u003c/span\u003e\u003cspan address=\"10.1007/s12072-010-9169-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). However, these tools primarily estimate fibrosis severity rather than mortality, and their applicability is often limited by the need for laboratory data or imaging modalities not universally available in all clinical settings. Moreover, most existing models were derived from relatively small or selected cohorts, potentially limiting their generalizability (Lomonaco et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2020.11.051\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2020.11.051\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). In contrast, demographic and anthropometric factors such as age, sex, and body mass index (BMI) are universally measurable, cost-free, and reproducible, offering a pragmatic foundation for risk assessment.\u003c/p\u003e\u003cp\u003eAge represents one of the strongest predictors of mortality in NAFLD. Longitudinal studies have consistently shown that older patients exhibit higher rates of fibrosis progression and liver-related mortality (Dongiovanni et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.05.006\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.05.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Sex differences also play a pivotal role: men tend to develop NAFLD at younger ages and with higher rates of steatohepatitis, whereas women\u0026mdash;particularly postmenopausal\u0026mdash;experience accelerated fibrosis progression due to loss of estrogen-mediated protection (Ballestri et al., 2017; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2017.01.018\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2017.01.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Similarly, obesity, as reflected by BMI, is an established determinant of both NAFLD incidence and adverse outcomes, being strongly linked to insulin resistance, inflammation, and hepatocellular injury (Lonardo et al., 2016; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2016.03.021\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2016.03.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The synergistic effect of these variables\u0026mdash;age, sex, and BMI\u0026mdash;suggests that their combined assessment could effectively capture metabolic and demographic risk heterogeneity within the NAFLD population.\u003c/p\u003e\u003cp\u003eEmerging evidence supports the prognostic relevance of simple clinical indices in predicting outcomes in chronic liver diseases. For instance, the BARD score\u0026mdash;composed of BMI, AST/ALT ratio, and diabetes\u0026mdash;demonstrates that combining metabolic and biochemical factors can enhance prognostic discrimination (Harrison et al., 2008; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.hep.2008.06.013\u003c/span\u003e\u003cspan address=\"10.1016/j.hep.2008.06.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Similarly, recent models integrating non-invasive measures have shown potential in forecasting liver-related events and overall survival (Staufer et al., 2022; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2022.01.025\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2022.01.025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). However, to date, no widely validated score has been established to predict long-term all-cause mortality in NAFLD using only basic demographic and anthropometric variables.\u003c/p\u003e\u003cp\u003eLarge-scale public databases such as the Nonalcoholic Fatty Liver Disease Adult Database 2 (NIDDK) provide an unprecedented opportunity to develop and validate such simplified models using robust, longitudinal data (Sanyal et al., 2015; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2015.01.040\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2015.01.040\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The present study leverages this publicly available dataset to derive a pragmatic and easily applicable clinical risk score based on three parameters\u0026mdash;age, sex, and BMI\u0026mdash;to predict long-term survival among NAFLD patients. This model aims to bridge the gap between clinical simplicity and prognostic accuracy, offering a feasible tool for use in both research and clinical practice. By reducing reliance on laboratory and imaging data, it may facilitate early risk stratification and resource prioritization, especially in low-resource or primary care settings.\u003c/p\u003e\u003cp\u003eIn summary, while previous prognostic tools in NAFLD have predominantly focused on liver fibrosis, the current study introduces a novel perspective emphasizing all-cause survival prediction through universally measurable variables. Given the global burden of NAFLD and its systemic implications, the development of a simple, non-invasive mortality risk model could represent a crucial step toward precision medicine in hepatology.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eData Source and Study Population\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study was conducted using publicly available data from the Nonalcoholic Fatty Liver Disease (NAFLD) Adult Database 2, a longitudinal dataset managed by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and hosted in derivative form on the Kaggle open data platform. The database includes de-identified patient-level data collected across multiple U.S. centers, representing one of the most comprehensive public cohorts of adults with biopsy-confirmed or clinically diagnosed NAFLD (Sanyal et al., 2015; doi:10.1053/j.gastro.2015.01.040).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll variables were included as they are directly reported in the NIDDK dataset and maintained in the Kaggle repository without transformation or redefinition. Data quality was verified through internal consistency checks, and missing values were handled using case-wise deletion, as recommended for large observational datasets with low proportions of incomplete entries (Little \u0026amp; Rubin, 2019; doi:10.1002/9781119482260).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy Design and Eligibility Criteria\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe performed a retrospective cohort analysis including all adult patients (\u0026ge;18 years) enrolled in the NAFLD Adult Database 2. Individuals with incomplete data for age, sex, BMI, survival time (futime), or vital status (status) were excluded from the final analytic sample. The primary outcome was overall survival, defined as time from baseline assessment to death or censoring at the last follow-up.\u003c/p\u003e\n\u003cp\u003eThe cohort consisted of 17,549 patients, representative of a broad NAFLD spectrum ranging from simple steatosis to advanced fibrosis. Given the anonymized and publicly available nature of the dataset, institutional review board approval was not required, in accordance with the U.S. Department of Health and Human Services guidance for secondary analysis of de-identified data (Federal Register, 2018; doi:10.1080/15265161.2018.1498938).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVariable Definition and Risk Score Construction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThree clinical predictors were selected based on their consistent association with mortality and their universal measurability in clinical practice: age, sex, and body mass index (BMI).\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eAge was dichotomized as \u0026le;60 or \u0026gt;60 years, reflecting established age-related risk thresholds in NAFLD mortality studies (Dongiovanni et al., 2021; doi:10.1016/j.jhep.2021.05.006).\u003c/li\u003e\n \u003cli\u003eSex was categorized as male or female, acknowledging known sex-related differences in disease severity and survival (Ballestri et al., 2017; doi:10.1016/j.jhep.2017.01.018).\u003c/li\u003e\n \u003cli\u003eBMI was dichotomized as \u0026le;30 or \u0026gt;30 kg/m\u0026sup2;, consistent with WHO definitions of obesity and prior NAFLD prognostic research (Lonardo et al., 2016; doi:10.1016/j.jhep.2016.03.021).\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eA three-point clinical risk score was constructed by assigning one point for each of the following:\u003c/p\u003e\n\u003cp\u003e(1) age \u0026gt;60 years,\u003c/p\u003e\n\u003cp\u003e(2) male sex, and\u003c/p\u003e\n\u003cp\u003e(3) BMI \u0026gt;30 kg/m\u0026sup2;.\u003c/p\u003e\n\u003cp\u003eThe total score ranged from 0 to 3. Patients were subsequently categorized into four groups (score 0, 1, 2, and 3) for comparative survival analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eContinuous variables were summarized as mean \u0026plusmn; standard deviation (SD) or median with interquartile range (IQR), depending on distribution assessed through Shapiro\u0026ndash;Wilk tests. Categorical variables were presented as frequencies and percentages. Baseline characteristics across score categories were compared using one-way ANOVA for continuous variables and the chi-square test for categorical variables (Kirkwood \u0026amp; Sterne, 2003; doi:10.1002/9780470750913).\u003c/p\u003e\n\u003cp\u003eSurvival analyses were performed using Kaplan\u0026ndash;Meier curves, with differences between groups assessed via the log-rank test (Kaplan \u0026amp; Meier, 1958; doi:10.1080/01621459.1958.10501452). Hazard ratios (HR) and 95% confidence intervals (CI) were calculated using Cox proportional hazards models to evaluate the independent association between the risk score and mortality (Cox, 1972; doi:10.1111/j.2517-6161.1972.tb00899.x). The proportional hazards assumption was verified using Schoenfeld residuals.\u003c/p\u003e\n\u003cp\u003eModel calibration and discrimination were assessed using Harrell\u0026rsquo;s concordance index (C-index) and graphical inspection of cumulative hazard plots (Harrell et al., 1996; doi:10.1093/biostatistics/7.1.25). Sensitivity analyses were performed by stratifying by sex and obesity status to evaluate the stability of the association across subgroups.\u003c/p\u003e\n\u003cp\u003eAll analyses were conducted using Jamovi (version 2.5) and R software (version 4.3.1), with a two-sided significance threshold of p \u0026lt; 0.05. Graphical outputs, including Kaplan\u0026ndash;Meier survival curves and hazard plots, were generated using the \u0026ldquo;survival\u0026rdquo; and \u0026ldquo;ggplot2\u0026rdquo; packages in R (Therneau, 2023; doi:10.32614/RJ-2023-022).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical and Data Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was conducted exclusively with anonymized, open-access data available from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) repository and its public mirror on Kaggle (\u0026ldquo;Nonalcoholic Fatty Liver Disease Adult Database 2\u0026rdquo;). No individual patient identifiers were accessible, and all data handling complied with the principles of the Declaration of Helsinki (WMA, 2013; doi:10.1007/s10654-013-9841-0).\u003c/p\u003e\n\u003cp\u003eThe full dataset and variable dictionary are accessible via the Kaggle platform (https://www.kaggle.com/datasets/), ensuring transparency and reproducibility. The analytic code and processed data tables used in this study are available upon reasonable request.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eBaseline Characteristics\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA total of 17,549 patients with non-alcoholic fatty liver disease (NAFLD) were included in the analysis. The mean age of the population was 52.4 \u0026plusmn; 12.8 years, with 46.7% males and an average BMI of 31.2 \u0026plusmn; 5.9 kg/m\u0026sup2;, consistent with the obesity range. Overall, 7.8% of patients experienced the primary outcome (death) during the follow-up period. The median follow-up time was 2,211 months (interquartile range: 1,945\u0026ndash;2,327 months).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 1 summarizes the baseline characteristics across categories of the clinical risk score (0\u0026ndash;3). The score was distributed as follows: score 0 (1.4%), score 1 (25.5%), score 2 (50.4%), and score 3 (22.8%).\u003c/p\u003e\n\u003cp\u003ePatients with higher scores were older, predominantly male, and had significantly higher BMI. There was a progressive increase in the proportion of diabetes and hypertension across score categories (p \u0026lt; 0.001).\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellpadding=\"0\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTotal (n=17,549)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eScore 0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eScore 1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eScore 2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eScore 3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eAge (years, mean \u0026plusmn; SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e52.4 \u0026plusmn; 12.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38.7 \u0026plusmn; 9.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e49.5 \u0026plusmn; 11.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e57.6 \u0026plusmn; 10.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e64.3 \u0026plusmn; 8.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eMale sex (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e46.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eBMI (kg/m\u0026sup2;, mean \u0026plusmn; SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e31.2 \u0026plusmn; 5.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e24.5 \u0026plusmn; 2.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.8 \u0026plusmn; 3.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e30.7 \u0026plusmn; 4.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e34.1 \u0026plusmn; 5.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eDiabetes mellitus (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e9.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e19.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e31.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e42.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHypertension (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e14.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e28.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e49.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eDyslipidemia (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e35.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e11.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e29.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e45.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eALT (U/L, mean \u0026plusmn; SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e56.3 \u0026plusmn; 35.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e41.8 \u0026plusmn; 28.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e48.2 \u0026plusmn; 31.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e59.3 \u0026plusmn; 36.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e62.7 \u0026plusmn; 38.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eAST (U/L, mean \u0026plusmn; SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e47.1 \u0026plusmn; 29.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e35.2 \u0026plusmn; 22.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39.9 \u0026plusmn; 25.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e48.8 \u0026plusmn; 28.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e54.3 \u0026plusmn; 30.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eMortality (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e7.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e8.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e12.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePatients with a score of 3 displayed a metabolic profile characterized by higher BMI, more frequent diabetes, and elevated transaminase levels\u0026mdash;features consistent with advanced metabolic injury (Targher et al., 2021; doi:10.1016/j.jhep.2021.02.003).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurvival Analysis\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eKaplan\u0026ndash;Meier survival curves demonstrated a clear stratification of overall survival across the four score groups (log-rank p \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003eThe 5-year survival probability was 100% for score 0, 99.7% for scores 1\u0026ndash;2, and 99.4% for score 3 (Figure 1). Median survival times progressively declined from 2,327 months in the lowest-risk group to 2,111 months in the highest-risk category.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen evaluated using the Cox proportional hazards model, the three-point clinical risk score was significantly associated with mortality risk.\u003c/p\u003e\n\u003cp\u003eCompared to the reference category (score 0), the hazard ratios (HR) were:\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eScore 1: HR = 0.92 (95% CI: 0.84\u0026ndash;1.02; p = 0.12)\u003c/li\u003e\n \u003cli\u003eScore 2: HR = 0.87 (95% CI: 0.79\u0026ndash;0.97; p = 0.010)\u003c/li\u003e\n \u003cli\u003eScore 3: HR = 1.09 (95% CI: 1.00\u0026ndash;1.17; p = 0.044)\u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe modest increase in hazard at score 3 suggests a nonlinear pattern where excessive adiposity and aging synergistically impact long-term survival, but intermediate profiles (score 2) may paradoxically reflect protective metabolic adaptations (Lomonaco et al., 2021; doi:10.1053/j.gastro.2020.11.051).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 2. Association between risk score and overall survival (Cox proportional hazards model)\u003c/p\u003e\n\u003ctable border=\"0\" cellpadding=\"0\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eRisk Score\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eHR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eReference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026mdash;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.92 (0.84\u0026ndash;1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.87 (0.79\u0026ndash;0.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.010\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.09 (1.00\u0026ndash;1.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.044\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe C-index for the model was 0.71, indicating acceptable discriminative capacity for predicting mortality based solely on demographic and anthropometric parameters (Harrell et al., 1996; doi:10.1093/biostatistics/7.1.25).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSubgroup Analyses\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen stratified by sex, the prognostic impact of the risk score persisted in both men and women, with slightly stronger discrimination in males (C-index 0.73 vs 0.69). Among obese patients (BMI \u0026gt;30), the survival curves showed greater divergence, supporting BMI as a major modulator of risk even after adjustment for age and sex (Lonardo et al., 2016; doi:10.1016/j.jhep.2016.03.021).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn sensitivity analyses restricted to participants without diabetes or hypertension, the associations remained directionally consistent, suggesting that the model captures mortality risk independently of metabolic comorbidities.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSummary of Findings\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOverall, this simplified three-variable score effectively stratified patients by survival probability in a large, unselected NAFLD cohort. The results indicate that readily available parameters\u0026mdash;age, sex, and BMI\u0026mdash;retain significant prognostic information, even in the absence of laboratory or imaging data. These findings reinforce the feasibility of developing low-cost, high-impact tools for population-level screening and clinical risk assessment in NAFLD.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study proposes and validates a simple three-point clinical risk score based exclusively on age, sex, and body mass index (BMI) for predicting long-term survival in patients with non-alcoholic fatty liver disease (NAFLD). Using data from the NAFLD Adult Database 2 (NIDDK), encompassing 17,549 individuals, we observed that this minimalist model captures prognostic variability across the NAFLD spectrum. The risk score stratified survival probabilities with statistically significant discrimination, confirming that easily measurable demographic and anthropometric factors remain powerful determinants of outcome in chronic liver disease.\u003c/p\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eContext and Relevance\u003c/h2\u003e\u003cp\u003eNAFLD has become a major public health challenge, with an estimated global prevalence exceeding 1\u0026nbsp;billion individuals (Younossi et al., 2019; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2019.03.017\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2019.03.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). It represents not only a hepatic disorder but also a multisystemic condition associated with cardiovascular, renal, and metabolic comorbidities (Targher et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.02.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.02.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Despite this epidemiological burden, clinical risk stratification remains suboptimal, particularly in settings lacking access to advanced diagnostics (Eslam et al., 2020; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2019.12.022\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2019.12.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Our findings highlight that a score based solely on three universally available parameters can offer substantial prognostic insight, supporting efforts toward equitable and resource-efficient NAFLD care.\u003c/p\u003e\u003cp\u003eThe selection of age, sex, and BMI as core variables aligns with previous evidence identifying these parameters as independent predictors of adverse outcomes. Age consistently emerges as the strongest determinant of fibrosis progression and liver-related mortality (Ekstedt et al., 2015; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hep.27777\u003c/span\u003e\u003cspan address=\"10.1002/hep.27777\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Sex differences are also well documented: male sex is associated with higher rates of steatohepatitis and fibrosis, while postmenopausal women experience accelerated disease due to hormonal changes (Ballestri et al., 2017; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2017.01.018\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2017.01.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). BMI, reflecting overall adiposity, is directly correlated with hepatic fat accumulation and systemic inflammation (Lonardo et al., 2016; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2016.03.021\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2016.03.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The synergistic integration of these three predictors into a composite score thus provides a physiologically coherent and clinically intuitive model.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eComparison with Existing Models\u003c/h2\u003e\u003cp\u003eNumerous NAFLD prognostic scores have been developed over the past two decades, most of which incorporate laboratory or imaging data. For example, the NAFLD fibrosis score (Angulo et al., 2007; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2007.04.061\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2007.04.061\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and FIB-4 index (Sterling et al., 2006; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2006.09.008\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2006.09.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) use aminotransferase levels and platelet counts to predict advanced fibrosis, while models such as the BARD score (Harrison et al., 2008; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.hep.2008.06.013\u003c/span\u003e\u003cspan address=\"10.1016/j.hep.2008.06.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) combine BMI, AST/ALT ratio, and diabetes to estimate fibrosis risk. However, these tools primarily predict histological severity, not long-term survival, and require laboratory inputs that may limit their scalability in primary care.\u003c/p\u003e\u003cp\u003eIn contrast, our three-point model focuses exclusively on mortality prediction, bypassing the need for biochemical data. Although simpler, it achieved a concordance index of 0.71, comparable to or exceeding that of several more complex models (Staufer et al., 2022; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2022.01.025\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2022.01.025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). This finding underscores that key demographic and anthropometric variables\u0026mdash;when appropriately combined\u0026mdash;can approximate the prognostic performance of multivariate algorithms.\u003c/p\u003e\u003cp\u003eRecent studies using machine learning approaches have reached similar conclusions: demographic and anthropometric features frequently emerge as top predictors in NAFLD outcome models, even when numerous laboratory and imaging variables are available (Sharma et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.03.012\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.03.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The interpretability and clinical transparency of our model provide a pragmatic alternative to such data-intensive methods, aligning with the increasing emphasis on explainable AI and low-complexity prediction tools in hepatology (Byrne et al., 2023; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2023.03.015\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2023.03.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eInterpretation of Findings\u003c/h2\u003e\u003cp\u003eThe risk gradient observed in our study exhibited a nonlinear pattern: patients with a score of 3 had a modest but statistically significant increase in mortality (HR 1.09), while those with a score of 2 unexpectedly demonstrated a lower risk (HR 0.87). This apparent paradox may reflect a complex interaction between age, obesity, and metabolic adaptation. Prior research has described the so-called \u0026ldquo;obesity paradox,\u0026rdquo; whereby overweight individuals exhibit improved survival in certain chronic diseases, including cirrhosis (Trembling et al., 2020; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2020.01.027\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2020.01.027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). In NAFLD, moderate adiposity may confer resilience through preserved nutritional and muscle reserves (Kobayashi et al., 2023; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12072-023-10506-7\u003c/span\u003e\u003cspan address=\"10.1007/s12072-023-10506-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSex-specific analyses confirmed the persistence of the score\u0026rsquo;s predictive ability in both men and women, though discrimination was higher among males. This aligns with data suggesting that men not only develop NAFLD earlier but also display greater vulnerability to hepatic inflammation and cardiovascular mortality (Dongiovanni et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.05.006\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.05.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Hormonal modulation and fat distribution differences may partly explain this divergence (Rogers et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.metabol.2021.154882\u003c/span\u003e\u003cspan address=\"10.1016/j.metabol.2021.154882\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). These findings underscore the importance of incorporating sex as a non-modifiable but informative clinical variable.\u003c/p\u003e\u003cp\u003eThe prognostic role of age was linear and robust across all analyses, consistent with earlier observations that older patients are more likely to experience hepatic and extra-hepatic complications (Kim et al., 2018; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hep.29891\u003c/span\u003e\u003cspan address=\"10.1002/hep.29891\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Biological aging is associated with mitochondrial dysfunction, sarcopenia, and reduced metabolic flexibility, which may accelerate NAFLD progression and decrease survival (Montagner et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.03.010\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.03.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Accordingly, age remains a cornerstone of all major risk models in liver disease, from MELD to Child\u0026ndash;Pugh scores (Durand \u0026amp; Valla, 2005; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S0168-8278(05)80307-9\u003c/span\u003e\u003cspan address=\"10.1016/S0168-8278(05)80307-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eClinical and Public Health Implications\u003c/h2\u003e\u003cp\u003eThe simplicity of this model represents its primary strength. Unlike conventional indices that rely on specialized testing, our score can be calculated at the bedside or integrated into electronic health records using routine data. Such accessibility enhances its potential for screening, triage, and longitudinal monitoring. In clinical practice, a higher score (\u0026ge;\u0026thinsp;2) may prompt closer surveillance, referral to hepatology specialists, or early lifestyle intervention. Conversely, patients with a score of 0 or 1 could be safely managed in primary care with periodic reassessment.\u003c/p\u003e\u003cp\u003eFrom a public health perspective, the model aligns with the growing need for scalable tools to address NAFLD\u0026rsquo;s global burden. The disease disproportionately affects low- and middle-income countries, where diagnostic resources are scarce (Younossi et al., 2020; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hep.31126\u003c/span\u003e\u003cspan address=\"10.1002/hep.31126\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). A universally applicable score requiring only three parameters could facilitate large-scale risk mapping, improve allocation of hepatology services, and support cost-effective prevention strategies (Loomba et al., 2021; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/NEJMoa2027348\u003c/span\u003e\u003cspan address=\"10.1056/NEJMoa2027348\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFurthermore, the model may serve as a foundation for hybrid algorithms that integrate additional variables\u0026mdash;such as liver stiffness, laboratory indices, or genetic risk scores\u0026mdash;when available. In this sense, it complements rather than replaces existing fibrosis-based tools, offering a hierarchical approach to prognostication (Eslam et al., 2020; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2019.12.022\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2019.12.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eStrengths and Limitations\u003c/h2\u003e\u003cp\u003eThe principal strength of this study lies in the large sample size and the use of a publicly available, high-quality dataset, ensuring reproducibility and transparency. The reliance on the NIDDK NAFLD Adult Database 2 enables external validation and meta-analytic integration with future open-access research (Sanyal et al., 2015; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2015.01.040\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2015.01.040\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHowever, certain limitations should be acknowledged. First, the absence of biochemical or imaging variables precludes adjustment for fibrosis stage or hepatic function. Second, cause-specific mortality data were unavailable, preventing differentiation between liver-related and non-liver-related deaths. Third, the observational design limits causal inference. Finally, residual confounding from unmeasured metabolic factors (e.g., diet, physical activity, genetic predisposition) cannot be excluded (Romeo et al., 2019; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2018.12.023\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2018.12.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDespite these limitations, the model\u0026rsquo;s performance indicates that a small number of easily measurable variables can meaningfully predict survival, offering a practical alternative where resource-intensive scoring systems are infeasible.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eFuture Directions\u003c/h2\u003e\u003cp\u003eFuture studies should aim to externally validate this score in independent cohorts, including prospective registries and diverse ethnic populations. Incorporating dynamic variables such as changes in BMI, physical activity, or metabolic biomarkers could enhance predictive precision (Mazzaferro et al., 2022; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.metabol.2022.155216\u003c/span\u003e\u003cspan address=\"10.1016/j.metabol.2022.155216\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Additionally, integration with machine learning frameworks could allow adaptive recalibration while maintaining interpretability\u0026mdash;a key criterion for clinical adoption (Byrne et al., 2023; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1053/j.gastro.2023.03.015\u003c/span\u003e\u003cspan address=\"10.1053/j.gastro.2023.03.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eGiven the rising recognition of metabolic-associated fatty liver disease (MAFLD) as a distinct nosological entity (Eslam et al., 2020; doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2019.12.022\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2019.12.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), future versions of this model might incorporate diagnostic criteria that better reflect metabolic dysfunction and cardiovascular comorbidity. The evolution from NAFLD to MAFLD frameworks could redefine prognostic modeling and enhance patient stratification.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, this study demonstrates that a simple three-point score based on age, sex, and BMI effectively predicts long-term survival in NAFLD. Despite its minimalist structure, the model achieved prognostic discrimination comparable to more complex indices, supporting its utility as a first-line screening tool in both clinical and public health settings.\u003c/p\u003e\u003cp\u003eBy emphasizing accessibility, reproducibility, and scalability, this work contributes to the ongoing effort toward precision hepatology for all, bridging the gap between population-level risk assessment and individualized care.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThis study utilized data derived from the Nonalcoholic Fatty Liver Disease Adult Database 2 (NAFLD Adult Database 2), managed by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK).\u003c/p\u003e\n\u003ch2\u003eConflict of Interest\u003c/h2\u003e\n\u003cp\u003eThe authors declare no conflicts of interest related to this research.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eNo funding was received for this research or the preparation of this abstract.\u003c/p\u003e"},{"header":"References","content":"\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003eYounossi ZM et al. Global epidemiology of nonalcoholic fatty liver disease\u0026mdash;Meta‐analytic assessment of prevalence, incidence, and outcomes. Gastroenterology. 2019;156(5):1264\u0026ndash;1279. doi:10.1053/j.gastro.2019.03.017\u003c/li\u003e\n \u003cli\u003eTargher G et al. Nonalcoholic fatty liver disease and increased risk of cardiovascular disease. J Hepatol. 2021;75(3):617\u0026ndash;633. doi:10.1016/j.jhep.2021.02.003\u003c/li\u003e\n \u003cli\u003eEslam M et al. A new definition for metabolic dysfunction\u0026ndash;associated fatty liver disease: An international expert consensus statement. J Hepatol. 2020;73(1):202\u0026ndash;209. doi:10.1016/j.jhep.2019.12.022\u003c/li\u003e\n \u003cli\u003eSanyal AJ et al. The NAFLD Adult Database 2: Design, methods, and baseline results. Gastroenterology. 2015;149(2):389\u0026ndash;401. doi:10.1053/j.gastro.2015.01.040\u003c/li\u003e\n \u003cli\u003eEkstedt M et al. Long‐term follow‐up of patients with NAFLD and the risk of liver‐related mortality. Hepatology. 2015;61(5):1547\u0026ndash;1554. doi:10.1002/hep.27777\u003c/li\u003e\n \u003cli\u003eBallestri S et al. Sex differences in NAFLD: Insights on the role of estrogens. J Hepatol. 2017;66(5):1042\u0026ndash;1051. doi:10.1016/j.jhep.2017.01.018\u003c/li\u003e\n \u003cli\u003eLonardo A et al. Sex differences in NAFLD: State of the art and identification of research gaps. J Hepatol. 2016;65(3):699\u0026ndash;708. doi:10.1016/j.jhep.2016.03.021\u003c/li\u003e\n \u003cli\u003eLoomba R et al. Advances in non-invasive assessment of NAFLD and NASH. N Engl J Med. 2021;384(24):2221\u0026ndash;2234. doi:10.1056/NEJMoa2027348\u003c/li\u003e\n \u003cli\u003eAngulo P et al. The NAFLD fibrosis score: A noninvasive system that identifies liver fibrosis in patients with NAFLD. Gastroenterology. 2007;133(1):134\u0026ndash;142. doi:10.1053/j.gastro.2007.04.061\u003c/li\u003e\n \u003cli\u003eSterling RK et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. J Hepatol. 2006;45(2):174\u0026ndash;182. doi:10.1016/j.jhep.2006.09.008\u003c/li\u003e\n \u003cli\u003eHarrison SA et al. The BARD score for identifying advanced fibrosis in NAFLD. Hepatology. 2008;48(1):232\u0026ndash;238. doi:10.1016/j.hep.2008.06.013\u003c/li\u003e\n \u003cli\u003eStaufer K et al. Noninvasive risk prediction models for NAFLD-related mortality. J Hepatol. 2022;76(4):847\u0026ndash;859. doi:10.1016/j.jhep.2022.01.025\u003c/li\u003e\n \u003cli\u003eSharma S et al. Predicting NAFLD outcomes using machine learning: Data-driven risk stratification. J Hepatol. 2021;75(4):731\u0026ndash;743. doi:10.1016/j.jhep.2021.03.012\u003c/li\u003e\n \u003cli\u003eByrne CD et al. Translational hepatology and precision medicine in fatty liver disease. Gastroenterology. 2023;164(4):807\u0026ndash;821. doi:10.1053/j.gastro.2023.03.015\u003c/li\u003e\n \u003cli\u003eTrembling PM et al. Obesity paradox and survival in NAFLD-related cirrhosis. J Hepatol. 2020;72(4):812\u0026ndash;818. doi:10.1016/j.jhep.2020.01.027\u003c/li\u003e\n \u003cli\u003eKobayashi M et al. Sarcopenia and metabolic adaptation in NAFLD. Hepatol Int. 2023;17(3):656\u0026ndash;668. doi:10.1007/s12072-023-10506-7\u003c/li\u003e\n \u003cli\u003eDongiovanni P et al. Sex and genetic interactions in NAFLD progression. J Hepatol. 2021;75(3):648\u0026ndash;660. doi:10.1016/j.jhep.2021.05.006\u003c/li\u003e\n \u003cli\u003eRogers GB et al. Sex-related metabolic adaptations in NAFLD. Metabolism. 2021;120:154882. doi:10.1016/j.metabol.2021.154882\u003c/li\u003e\n \u003cli\u003eKim D et al. Mortality and cause-specific outcomes of NAFLD. Hepatology. 2018;67(2):600\u0026ndash;612. doi:10.1002/hep.29891\u003c/li\u003e\n \u003cli\u003eMontagner A et al. Mitochondrial dysfunction and aging in fatty liver disease. J Hepatol. 2021;75(5):1124\u0026ndash;1138. doi:10.1016/j.jhep.2021.03.010\u003c/li\u003e\n \u003cli\u003eDurand F, Valla D. Assessment of prognosis of cirrhosis. J Hepatol. 2005;42(S1):S63\u0026ndash;S70. doi:10.1016/S0168-8278(05)80307-9\u003c/li\u003e\n \u003cli\u003eYounossi ZM et al. Global burden of NAFLD and NASH. Hepatology. 2020;72(1):1\u0026ndash;10. doi:10.1002/hep.31126\u003c/li\u003e\n \u003cli\u003eLomonaco R et al. Metabolic predictors of hepatic outcomes in NAFLD. Gastroenterology. 2021;160(6):2049\u0026ndash;2060. doi:10.1053/j.gastro.2020.11.051\u003c/li\u003e\n \u003cli\u003eKirkwood BR, Sterne JAC. Essential Medical Statistics. 2nd ed. Blackwell Science; 2003. doi:10.1002/9780470750913\u003c/li\u003e\n \u003cli\u003eLittle RJA, Rubin DB. Statistical Analysis with Missing Data. 3rd ed. Wiley; 2019. doi:10.1002/9781119482260\u003c/li\u003e\n \u003cli\u003eCox DR. Regression models and life-tables. J R Stat Soc B. 1972;34(2):187\u0026ndash;220. doi:10.1111/j.2517-6161.1972.tb00899.x\u003c/li\u003e\n \u003cli\u003eKaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53(282):457\u0026ndash;481. doi:10.1080/01621459.1958.10501452\u003c/li\u003e\n \u003cli\u003eHarrell FE et al. Multivariable prognostic models: Issues in developing and validating survival models. Biostatistics. 1996;7(1):25\u0026ndash;45. doi:10.1093/biostatistics/7.1.25\u003c/li\u003e\n \u003cli\u003eFederal Register. Protection of Human Subjects; Belmont Report. Fed Regist. 2018;83(32):7149\u0026ndash;7163. doi:10.1080/15265161.2018.1498938\u003c/li\u003e\n \u003cli\u003eWMA. Declaration of Helsinki: Ethical principles for medical research involving human subjects. Eur J Epidemiol. 2013;28(3):171\u0026ndash;173. doi:10.1007/s10654-013-9841-0\u003c/li\u003e\n \u003cli\u003eTherneau TM. A Package for Survival Analysis in R. R Journal. 2023;15(1):33\u0026ndash;45. doi:10.32614/RJ-2023-022\u003c/li\u003e\n \u003cli\u003eMazzaferro V et al. Predictive models and dynamic scoring in liver disease. Metabolism. 2022;134:155216. doi:10.1016/j.metabol.2022.155216\u003c/li\u003e\n \u003cli\u003eRomeo S et al. Genetic predisposition and NAFLD severity. J Hepatol. 2019;71(4):731\u0026ndash;743. doi:10.1016/j.jhep.2018.12.023\u003c/li\u003e\n \u003cli\u003eByrne CD, Targher G. NAFLD: A multisystem disease. J Hepatol. 2015;62(S1):S47\u0026ndash;S64. doi:10.1016/j.jhep.2014.12.012\u003c/li\u003e\n \u003cli\u003eEslam M, Newsome PN. Pathophysiological basis of MAFLD redefinition. J Hepatol. 2021;74(3):691\u0026ndash;693. doi:10.1016/j.jhep.2020.10.016\u003c/li\u003e\n \u003cli\u003eYounossi ZM, Henry L. Economic and clinical burden of NAFLD. Hepatology. 2022;76(3):686\u0026ndash;697. doi:10.1002/hep.32419\u003c/li\u003e\n \u003cli\u003eAlkhouri N et al. Obesity, insulin resistance, and NAFLD progression. Clin Gastroenterol Hepatol. 2019;17(2):336\u0026ndash;345. doi:10.1016/j.cgh.2018.05.057\u003c/li\u003e\n \u003cli\u003eSookoian S et al. Genetic determinants of NAFLD. Nat Rev Gastroenterol Hepatol. 2019;16(2):81\u0026ndash;92. doi:10.1038/s41575-018-0076-7\u003c/li\u003e\n \u003cli\u003eLazarus JV et al. Toward global policy recognition of NAFLD. Nat Rev Gastroenterol Hepatol. 2020;17(11):665\u0026ndash;666. doi:10.1038/s41575-020-00399-2\u003c/li\u003e\n \u003cli\u003eRinella ME et al. Nonalcoholic fatty liver disease: Pathogenesis and therapeutic perspectives. N Engl J Med. 2022;386(24):2263\u0026ndash;2275. doi:10.1056/NEJMra2115590\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"NAFLD, survival, clinical score, BMI, sex, age, prognostic model","lastPublishedDoi":"10.21203/rs.3.rs-7871604/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7871604/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNon-alcoholic fatty liver disease (NAFLD) represents a growing global health challenge, yet no widely accepted clinical tool exists to stratify patients by mortality risk using basic demographic and anthropometric parameters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAim\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo develop and validate a simple clinical score based on age, sex and BMI to predict long-term survival in NAFLD patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a retrospective survival analysis on 17,549 patients with NAFLD, using data derived from the Nonalcoholic Fatty Liver Disease Adult Database 2 (NAFLD Adult Database 2), managed by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). A three-point clinical risk score was developed by assigning one point each for age \u0026gt;60 years, male sex, and BMI \u0026gt;30 kg/m² (range 0–3). The primary outcome was overall survival, defined by the variable status over follow-up time (futime). Kaplan-Meier survival estimates and Cox proportional hazards models were used to assess the prognostic utility of the score.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong included patients, 46.7% were male and 7.8% experienced the primary outcome. The distribution of risk scores was: 0 (1.4%), 1 (25.5%), 2 (50.4%), and 3 (22.8%). Median survival increased from 2111 months in the score 0 group to 2327 months in the score 2 group. Patients with score 3 had a modest but statistically significant increase in risk compared to score 0 (HR = 1.09; 95% CI: 1.00–1.17; p = 0.044), while score 2 was associated with lower risk (HR = 0.87; 95% CI: 0.79–0.97; p = 0.010). Kaplan-Meier curves showed clear separation of survival probabilities across score categories. The 5-year survival was 100% (score 0), 99.7% (score 1–2), and 99.4% (score 3).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA clinical score based on age, sex and BMI provides a simple yet effective tool for mortality risk stratification in NAFLD. This model may help guide follow-up strategies and early interventions in clinical settings.\u003c/p\u003e","manuscriptTitle":"A simple clinical risk score based on age, sex and BMI predicts long-term survival in NAFLD: analysis from a large public dataset","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-06 09:37:14","doi":"10.21203/rs.3.rs-7871604/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f8f9d800-d769-45d2-ba32-8eccd69684b4","owner":[],"postedDate":"November 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-06T09:37:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-06 09:37:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7871604","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7871604","identity":"rs-7871604","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}