Development and validation of a nomogram to predict depression in older adults with heart disease: a national survey in China

Abstract

Background: Comorbid depression, frequently observed in heart disease patients, has detrimental effects on mental health and may exacerbate cardiac conditions. The objective of this study was to create and validate a risk prediction nomogram specifically for comorbid depression in older adult patients suffering from heart disease.

Methods: The 2018 data from the Chinese Longitudinal Healthy Longevity Survey (CLHLS) was analyzed and 2,110 older adult patients with heart disease aged 60 and above were included in the study. They were randomly divided in a 7:3 ratio into a training set (n = 1,477) and a validation set (n = 633). Depression symptoms were assessed using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) and the participants were categorized into depressed (n = 687) and non-depressed (n = 1,423) groups. We collected information regarding general demographics, lifestyle habits, and medical history of the included patients. LASSO regression and binary logistic regression analyses were performed to identify independent risk factors and construct the depression prediction nomogram. Receiver operating characteristic curve analysis and the Hosmer-Lemeshow test were used to assess the model's discrimination and calibration. Decision curve analysis helped evaluate the clinical utility of the predictive nomogram.

Results: Based on the LASSO and multivariable regression analyses, education level, quality of life, sleep quality, frequency of watching TV, and history of arthritis were identified as independent risk factors for comorbid depression in the older adult heart disease patients. A nomogram model was constructed with these five independent risk factors. The nomogram showed good clinical performance with an area under the curve (AUC) value of 0.816 (95% CI: 0.793 to 0.839). The calibration curves and Hosmer-Lemeshow goodness-of-fit test (training set ${\chi }_t^2$ = 4.796, p = 0.760; validation set ${\chi }_v^2$ = 7.236, p = 0.511) showed its satisfactory. Clinical usefulness of the nomogram was confirmed by decision curve analysis.

Conclusions: A five-parameter nomogram for predicting depression in older adult heart disease patients was developed and validated. The nomogram demonstrated high accuracy, discrimination ability, and clinical utility in assessing the risk of depression in the older adult patients with heart disease.

Keywords: depression; heart disease; nomogram; predictive model; risk factors.

Figure 1

Flow diagram of the study design.

Figure 2

Variable selection using the LASSO regression model. (A) LASSO coefficient pathway diagram of predictive variables, each distinct colored line represents the trajectory of a feature's coefficient, as it evolves with the increase in the regularization parameter λ. As λ increases, certain coefficients gradually diminish and, ultimately, approach zero, signifying that these features are effectively pruned from the model by the LASSO regression. (B) A tenfold cross-validation was performed to select the optimal penalty parameter (lambda) for the model. The cross-validation identified log(lambda.1se) = −3.392 (lambda.1se = 0.03364981) as the point where the model error was within one standard deviation of the minimum.

Figure 3

Nomogram for predicting the risk of depression in older adult patients with heart disease.

Figure 4

Receiver operating characteristic (ROC) curve evaluates the risk prediction nomogram for depression in older adult patients with heart disease. The Y-axis represents the true-positive rate of the risk prediction, and the X-axis represents the false-positive rate of risk prediction. (A) ROC curve of the training set; (B) ROC curve of the validation set.

Figure 5

Calibration curves for the risk nomogram of depression in older adult patients with heart disease. The Y-axis represents the actual number of diagnosed depression cases, and the X-axis represents the predictive risk of depression. The diagonal dotted line represents perfect prediction by an ideal model, and a closer fit to the diagonal dotted line indicates better prediction. (A) Calibration curve of the training set; (B) Calibration curve of the validation set.

Figure 6

Decision curve analysis for the risk nomogram of depression in older adult patients with heart disease. The Y-axis represents the net benefit. The thick solid line represents the assumption that none of the patients have depression, the thin solid line represents the assumption that all patients have depression, and the red line represents the risk nomogram. (A) Decision curve of the training set; (B) Decision curve of the validation set.