Construction and validation of a predictive model for sleep disorders among rural older adults with mild cognitive impairment.
Yang T, Chi J, et al. • Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology • 2026
A nomogram built from sex, hypertension, self-rated health, and depression predicted sleep disturbance among rural older adults with MCI with good discrimination (AUC = 0.759) and calibration, with internal validation confirming robustness.
Key Findings
Results
More than half of rural older adults with mild cognitive impairment in the sample had sleep disturbance.
The prevalence of sleep disturbance was 54.2%.
The sample was 1,049 rural older adults aged ≥60 years with MCI.
Data came from the 2024 baseline survey of the Active Aging Cohort in Rural Areas, Shandong Province, China.
Sleep quality was assessed with the Pittsburgh Sleep Quality Index (PSQI).
Results
Sex, hypertension, self-rated health, and depression were identified as independent predictors of sleep disturbance in this population.
Logistic regression was used to identify independent predictors from demographic, lifestyle, disease history, and psychological factors.
Data were collected via face-to-face questionnaires and health examinations.
The four identified predictors were used to build the nomogram.
Results
The nomogram showed good discrimination for predicting sleep disturbance.
AUC = 0.759 (95% CI: 0.731-0.778).
Discrimination was evaluated with ROC curve analysis.
Results
The model showed good calibration, and internal validation supported its robustness.
Performance was evaluated with calibration curves, decision curve analysis, and bootstrap validation.
The abstract reports that internal validation confirmed robustness.
Conclusions
The study constructed a risk assessment model intended to support early screening and targeted intervention for sleep disturbance in rural older adults with MCI.
The design was cross-sectional.
The model was presented as a nomogram built from logistic regression predictors.
What This Means
This study looked at sleep problems among older adults with mild cognitive impairment (early memory and thinking difficulties) living in rural Shandong Province, China. Among 1,049 people aged 60 and older, about 54% had poor sleep as measured by a standard questionnaire, the Pittsburgh Sleep Quality Index.
The researchers found that four factors—sex, high blood pressure, self-rated health, and depression—independently helped predict who had sleep problems. They combined these into a simple scoring tool (a nomogram) that estimates an individual's risk. The tool distinguished between people with and without sleep problems reasonably well (AUC of 0.759), and checks of its accuracy within the same dataset supported its reliability.
This research suggests that a short list of easily collected information could help community health workers flag rural older adults with cognitive impairment who may have sleep problems and might benefit from further assessment or support. However, the study was cross-sectional, so it cannot show that these factors cause poor sleep, and the model was validated internally rather than in a separate population.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Yang T, Chi J, Song J, Yan L, Li W, Xu Z, et al.. (2026). Construction and validation of a predictive model for sleep disorders among rural older adults with mild cognitive impairment.. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology. https://doi.org/10.1007/s10072-026-09492-5