A Cross‐Nationally Validated Nomogram for Cardiometabolic Multimorbidity in Overweight/Obese Older Adults: CHARLS and HRS Cohorts

Q: What is the clinical evidence from this study?

Study design: Cross-Sectional. Population: Cardiometabolic multimorbidity (CMM) in overweight or obese middle-aged and older adults (n=12145). Intervention: Eight-variable multidimensional nomogram (residence, ADL score, CESD-10 score, HbA1c, hypertension, arthritis, dyslipidaemia, memory disorder). Primary outcome: Cardiometabolic multimorbidity (coexistence of >=2 of diabetes/high blood sugar, heart disease, and stroke) (AUC 0.871 (development), AUC 0.817 (internal validation), AUC 0.733 (external validation), 95% CI 0.838-0.905).

Yanhan Wei; Siyi He; Huizhen Chen; Yue Wang

doi:10.1111/dom.71077

July 6, 2026Diabetes Obesity and MetabolismOpen Access

A Cross‐Nationally Validated Nomogram for Cardiometabolic Multimorbidity in Overweight/Obese Older Adults: CHARLS and HRS Cohorts

View Full Paper

Ask AI

Bookmark

Share

Key result

Eight-variable nomogram predicts cardiometabolic multimorbidity in overweight older adults with ~0.82 internal validation AUC.

AUC 0.871 (development), AUC 0.817 (internal validation), AUC 0.733 (external validation)
95% CI 0.838-0.905
n=12,145

Why the study?

Can a multidimensional nomogram integrating metabolic, functional, and psychological factors accurately identify cardiometabolic multimorbidity in overweight or obese middle-aged and older adults across Chinese and US cohorts?

Key Points

To develop and validate a multidimensional nomogram incorporating metabolic, functional, and psychological predictors to assess cardiometabolic multimorbidity risk in overweight and obese middle-aged and older adults.
Overweight and obese participants from the China Health and Retirement Longitudinal Study (CHARLS) were randomly split into development (n = 2775) and internal validation (n = 1190) cohorts, alongside an external validation cohort of 8180 participants from the US Health and Retirement Study (HRS).
Predictors were screened using least absolute shrinkage and selection operator (LASSO) regression and multivariable logistic regression, with model discrimination evaluated by the area under the receiver operating characteristic curve (AUC).
Eight independent predictors were retained: rural residence (OR = 0.353, 95% CI: 0.223–0.558, p < 0.001), ADL score (OR = 1.206, 95% CI: 1.002–1.452, p = 0.048), CESD-10 score (OR = 1.050, 95% CI: 1.014–1.087, p = 0.006), HbA1c (OR = 1.565, 95% CI: 1.231–1.991, p < 0.001), hypertension (OR = 4.400, 95% CI: 2.638–7.339, p < 0.001), arthritis (OR = 1.829, 95% CI: 1.172–2.854, p = 0.008), dyslipidaemia (OR = 2.576, 95% CI: 1.636–4.055, p < 0.001), and memory disorder (OR = 2.735, 95% CI: 1.039–7.196, p = 0.042).
Model discrimination yielded an AUC of 0.871 (95% CI: 0.838–0.905) in the development set, 0.817 (95% CI: 0.768–0.867) in the internal validation set, and 0.733 (95% CI: 0.714–0.751) in external validation.
Under baseline CMM prevalence of 3.91%, calibration was acceptable in CHARLS, whereas HRS external validation showed underestimation (O:E ratio = 1.514) that improved to 1.000 following intercept recalibration.

Study Design

Type

Cross-Sectional (n=12,145)

Multicenter

Yes

Structured PICO

Can a multidimensional nomogram integrating metabolic, functional, and psychological factors accurately identify cardiometabolic multimorbidity in overweight or obese middle-aged and older adults across Chinese and US cohorts?

P

Population

12,145 middle-aged and older adults (aged ≥45 years) with overweight or obesity (BMI ≥24.0 kg/m2) from the Chinese CHARLS cohort (n=3,965; median age 57 years, 62.0% female) and the US HRS external validation cohort (n=8,180).

E

Exposure

Multidimensional risk prediction nomogram integrating 8 independent predictors: residence (rural vs urban), Activities of Daily Living (ADL) score, Center for Epidemiologic Studies Depression Scale (CESD-10) score, haemoglobin A1c (HbA1c), hypertension, arthritis, dyslipidaemia, and memory disorder.

O

Outcome

Cardiometabolic multimorbidity (CMM), defined as the coexistence of any two or more self-reported physician-diagnosed conditions: (1) diabetes or high blood sugar, (2) heart disease (heart attack, angina, coronary heart disease, heart failure, or other heart problems), and (3) stroke (or transient ischaemic attack in HRS).composite

Main Result

Effect estimate: AUC 0.871 (development), AUC 0.817 (internal validation), AUC 0.733 (external validation) (95% CI 0.838-0.905)

Limitations

CMM and its component comorbidities were self-reported, introducing recall bias and probable under-ascertainment
Complete-case external validation may have introduced selection bias
Cross-sectional design performs concurrent risk stratification rather than prospective prediction, and several predictors (HbA1c, hypertension, dyslipidaemia) overlap conceptually with CMM components
Family history, medication use, physical activity, and diet were not included due to incomplete recording or non-comparable definitions across cohorts
Predictor definitions differed across cohorts (e.g., CESD-8 in HRS vs CESD-10 in CHARLS; self-reported vs laboratory-defined dyslipidaemia with poor agreement, kappa = 0.034)
Relatively low number of events in development and a single random split without bootstrap optimism correction
CMM and component comorbidities were self-reported, introducing recall bias and probable under-ascertainment
Complete-case analysis in external validation may have introduced selection bias
Cross-sectional design performs concurrent risk stratification rather than prospective prediction, and predictors like HbA1c, hypertension, and dyslipidaemia overlap conceptually with CMM outcome components
Excluded established correlates such as family history, medication use, physical activity, and diet due to incomplete or non-comparable data
Predictor measurement differences across cohorts (CESD-10 vs CESD-8; self-reported vs laboratory-defined dyslipidaemia with poor agreement κ = 0.034)
Single random split without bootstrap optimism correction and only 155 total CMM events in CHARLS

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

This cross-nationally validated 8-variable nomogram integrates geriatric functional and psychological factors with metabolic markers to stratify cardiometabolic multimorbidity risk in overweight/obese older adults, offering high negative predictive value (93.6%–99.4%) to rule out low-risk individuals.

Abstract

AIMS: Middle-aged and older adults with overweight or obesity are at increased risk of cardiometabolic multimorbidity (CMM), yet validated prediction tools tailored to this population remain limited. This study aimed to develop and validate a multidimensional nomogram integrating metabolic, functional and psychological predictors. MATERIALS AND METHODS: were randomly split into development (n = 2775) and validation (n = 1190) sets. Candidate predictors spanned socio-demographic, lifestyle, functional, psychological and biochemical factors. External validation included 8180 overweight/obese participants from the US Health and Retirement Study (HRS). Least absolute shrinkage and selection operator and multivariable logistic regression were applied to build a nomogram. Performance was assessed by area under the curve (AUC), calibration and decision curve analysis. RESULTS: CMM prevalence was 3.91%. Eight variables were retained: residence, Activities of daily living (ADL) score, the Center for Epidemiologic Studies Depression Scale (CESD-10) score, haemoglobin A1c (HbA1c), hypertension, arthritis, dyslipidaemia and memory disorder. AUCs were 0.871 (95% CI: 0.838-0.905) (development), 0.817 (95% CI: 0.768-0.867) (validation) and 0.733 (95% CI: 0.714-0.751) in HRS external validation, with acceptable calibration in CHARLS and evidence of underestimation in HRS that improved after intercept recalibration, and clear net benefit. Positive predictive values were additionally reported to support clinical interpretation under the low CMM prevalence. CONCLUSIONS: This nomogram integrating metabolic, functional and psychological predictors demonstrated strong discrimination in the CHARLS cohort and maintained adequate discriminative performance in a cross-national external validation using the US HRS cohort despite differences in population characteristics and measurement instruments, supporting its potential utility for early CMM risk stratification in overweight/obese older adults. The external HRS validation showed fair discrimination, suggesting that model recalibration may be necessary before broad cross-national implementation.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Wei et al. (2026) conducted a cross-sectional in Cardiometabolic multimorbidity (CMM) in overweight or obese middle-aged and older adults (n=12,145). Eight-variable multidimensional nomogram (residence, ADL score, CESD-10 score, HbA1c, hypertension, arthritis, dyslipidaemia, memory disorder) was evaluated on Cardiometabolic multimorbidity (coexistence of >=2 of diabetes/high blood sugar, heart disease, and stroke) (AUC 0.871 (development), AUC 0.817 (internal validation), AUC 0.733 (external validation), 95% CI 0.838-0.905). An 8-variable nomogram predicted cardiometabolic multimorbidity in overweight/obese older adults with AUCs of 0.871 (development), 0.817 (internal validation), and 0.733 (US external validation).

synapsesocial.com/papers/6a4f3c252b81a944af575b0d https://doi.org/10.1111/dom.71077

View Full Paper

Ask AI

Bookmark

Share

A Cross‐Nationally Validated Nomogram for Cardiometabolic Multimorbidity in Overweight/Obese Older Adults: CHARLS and HRS Cohorts

Key Points

Study Design

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

A Cross‐Nationally Validated Nomogram for Cardiometabolic Multimorbidity in Overweight/Obese Older Adults: CHARLS and HRS Cohorts

Key Points

Study Design

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading