---
title: "Nonlinear association of sleep quality (PSQI) with type 2 diabetes prevalence in community chronic"
id: "plos-one-18-nonlinear-relationship-between-sleep-quality-and-prevalence-of-type-2-diabetes"
canonical_url: "https://medichelpline.com/clinical-feed/plos-one-18-nonlinear-relationship-between-sleep-quality-and-prevalence-of-type-2-diabetes"
content_type: "clinical_feed_article"
specialty: "Endocrinology"
source_name: "PLOS ONE (Medicine)"
source_url: "https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0356250"
published_at: "2026-08-21T14:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Nonlinear association of sleep quality (PSQI) with type 2 diabetes prevalence in community chronic
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/plos-one-18-nonlinear-relationship-between-sleep-quality-and-prevalence-of-type-2-diabetes
- **Specialty:** [Endocrinology](https://medichelpline.com/clinical-feed/endocrinology.md)
- **Primary Source:** PLOS ONE (Medicine)
- **Source URL:** [Original Journal Publication](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0356250)
- **Published At:** 2026-08-21T14:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- This cross-sectional study enrolled community-dwelling patients with chronic diseases in Jiangsu and Shanghai between 5 January and 30 March 2025 to examine the relationship between **sleep quality** and prevalence of **type 2 diabetes (T2D)**. - A total of 902 valid participants were analyzed; 565 (62.64%) had T2D. The mean Pittsburgh Sleep Quality Index (**PSQI**) score was 11.11. - Sleep quality was assessed with the Chinese-validated **PSQI** (Cronbach’s α = 0.84), and demographic and clinical variables were collected by questionnaire. - Analyses included descriptive statistics, logistic regression, and restricted cubic spline (**RCS**) regression to evaluate potential nonlinear dose–response relationships between PSQI scores and T2D prevalence. - Higher PSQI scores were positively associated with T2D prevalence across multiple regression models. - RCS modeling identified a nonlinear relationship with two inflection points at PSQI scores of **12** and **13**: when PSQI > 12 the odds ratio (OR) for prevalent T2D rose with increasing PSQI, and at PSQI = 13 the OR curve plateaued, indicating a slowing in the rate of risk increase. - The authors conclude sleep quality demonstrates a threshold-range effect on T2D prevalence and suggest community practitioners consider sleep assessment and targeted sleep interventions for high-risk patients, particularly those with **PSQI≥12**. - Data availability, funding, and competing-interest statements were reported: supporting data are in the article’s supporting information; funded by Shanghai Jianqiao University School-level Key Curriculum Project; authors declared no competing interests.
## Clinical Analysis & Structured Key Points
Nonlinear relationship between sleep quality and prevalence of type 2 diabetes: A cross-sectional study of community patients with chronic diseases | PLOS One Browse Subject Areas ? Click through the PLOS taxonomy to find articles in your field. For more information about PLOS Subject Areas, click here . Article Authors Metrics Comments Media Coverage Reader Comments Figures Figures Abstract Objective To delve into the intricate relationship between sleep quality and prevalence of type 2 diabetes (T2D) among community-dwelling chronic disease patients, and to provide insights that can inform clinical practice. Methods Data on basic information and sleep quality of patients with chronic diseases were collected via questionnaire surveys. Following descriptive and logistic regression analyses, restricted cubic spline regression was further applied to explore a possible nonlinear dose-response relationship between sleep quality and prevalence of T2D among these patients. Results A total of 902 valid samples were included in this study, among which 565 were cases of T2D, accounting for 62.64%. The average score of the Pittsburgh Sleep Quality Index (PSQI) was 11.11. In all three models, a positive association was observed between higher PSQI scores and prevalence of T2D. Subsequent analysis employing RCS regression substantiates this nonlinear association and discerns two inflection points at 12 and 13 When the PSQI score was greater than 12, the odds ratio (OR) increased with higher PSQI scores, suggesting a significant increase in the association with the prevalence of T2D. When the PSQI score reaches 13, the OR stabilises, indicating that the rate of increase in risk has slowed. Conclusion Sleep quality is an important factor for the prevalence of T2D, exhibiting a threshold range effect. Community health practitioners may consider sleep quality assessment as a supplementary tool, particularly in patients with existing chronic conditions. Implementing targeted sleep improvement interventions in this high-risk subgroup (PSQI≥12) may holds substantial potential to be associated with reduced T2D prevalence. Citation: Ge J, Lu P, Peng S, Huang H (2026) Nonlinear relationship between sleep quality and prevalence of type 2 diabetes: A cross-sectional study of community patients with chronic diseases. PLoS One 21(8): e0356250. https://doi.org/10.1371/journal.pone.0356250 Editor: Aleksandra Klisic, University of Montenegro-Faculty of Medicine, MONTENEGRO Received: March 26, 2026; Accepted: August 2, 2026; Published: August 21, 2026 Copyright: © 2026 Ge et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data is included in the Supporting information file. Funding: This work was supported by the Shanghai Jianqiao University School-level Key Curriculum Project (Grant No. JXGG202564), awarded to Juan Ge. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: T2D, type 2 diabetes; COPD, Chronic Obstructive Pulmonary Disease; RCS, Restricted Cubic Spline; PSQI, Pittsburgh Sleep Quality Scale; BMI, Body Mass Index; SE, standard error; OR, Odds Ratio; CI, Confidence Intervals; OSA, Obstructive Sleep Apnea 1. Introduction Type 2 diabetes (T2D) is the predominant form of diabetes on a global scale, and its prevalence continues to increase in all countries. The data published by the International Diabetes Federation (IDF) shows that more than 580 million people worldwide live with diabetes, with the number expected to surge to 853 million by 2050 [ 1 ]. It is noteworthy that patients with T2D account for over 90% of these cases. China has the highest global burden of diabetes, and rapidly evolving lifestyle with an increasingly aging population continues to drive the rising prevalence of T2D [ 2 ]. Diabetes is characterized by persistent elevated blood glucose levels and a range of severe systemic complications, including diabetic nephropathy, retinopathy and cardiovascular disease. These collectively have a detrimental effect on patients’ quality of life and life expectancy [ 3 ]. Furthermore, the management of T2D necessitates protracted medical interventions and an ongoing financial commitment, thereby imposing a substantial socioeconomic burden on healthcare systems and affected families [ 4 , 5 ]. Parker et al.[ 4 ] revealed that the healthcare expenditures for patients with diagnosed diabetes were 2.6 times higher than the anticipated expenses for non-diabetic patients. These significant implications underscore the necessity for in-depth research on the pathogenesis and modifiable risk factors of T2D, which are imperative for the development of effective prevention strategies and the improvement of disease management programs. A mounting body of evidence indicates a significant clinical association between sleep quality and the development of T2D [ 6 , 7 ]. Sleep is an important physiological process that plays a key role in maintaining normal metabolic function and endocrine homeostasis [ 8 ]. Contemporary research suggests that sleep-related disturbances, including short sleep duration, impaired sleep architecture and circadian rhythm dysregulation, may increase the risk of T2D through a variety of pathophysiological mechanisms [ 7 ]. The mechanisms implicated in this process encompass reduced insulin sensitivity, dysregulated glucose metabolism, and chronic low-grade inflammation [ 9 ]. Research indicated that individuals who sleep for duration less than 7 hours or more than 8 hours per night are predisposed to a higher risk of developing T2D. Furthermore, it has been demonstrated that poor sleep quality and chronic nighttime patterns are significantly associated with increased susceptibility to T2D. It is noteworthy that daytime naps lasting more than 30 minutes have been associated with a 7% to 20% increased risk of developing diabetes compared to naps of shorter duration or none at all [ 7 ]. Notwithstanding these advances, there are manifest limitations to the current research paradigm. The majority of studies have used linear regression models founded upon population-level assumptions, a practice which may result in an oversimplification of the complex, non-linear dynamics that exist between sleep parameters and metabolic outcomes. This methodological limitation underscores the necessity for advanced computational methodologies, including machine learning algorithms and spline regression models, to more accurately delineate the dose-response relationship and ascertain critical thresholds for the pathogenesis of sleep-related diabetes. Individuals with chronic disease constitute a critical demographic for diabetes prevention and management strategies, yet the association between sleep quality and diabetes risk in this population remains inadequately explored [ 10 ]. Pathophysiological analyses reveal that the inherent biological predisposition toward metabolic dysregulation in patients with chronic conditions amplifies their susceptibility to diabetes development [ 11 , 12 ]. Wang [ 13 ] and colleagues demonstrated that, following adjustment for pertinent variables, hypertensive patients exhibited a 2.956-fold increased risk of diabetes compared with non-hypertensive subjects.Implementing a bidirectional Mendelian randomization analysis, Wang and colleagues [ 14 ] established causal evidence that chronic obstructive pulmonary disease (COPD) confers a 6% incremental risk for T2D development (OR 1.06, 95% CI 1.01–1.11; P = 0.006). Concurrently, patients with chronic diseases often experience significant disruption of their sleep architecture due to the dual impact of disease symptoms (such as pain and dyspnea) and therapeutic medications (such as glucocorticoids), thereby falling into a vicious cycle of “disease—sleep disturbance—metabolic deterioration” [ 15 , 16 ]. Clinically, sleep disturbances in these patients are often simplistically attributed to the symptoms of their underlying diseases, which results in the underestimation of the significance of sleep disturbances as an independent risk factor. The existing studies, mostly based on healthy populations, may not accurately reflect the unique characteristics of individuals with chronic diseases. Consequently, there is an urgent need for in-depth studies in chronically ill populations to more accurately assess the impact of sleep quality on diabetes risk in this group. The restricted cubic spline (RCS) approach provides a flexible regression framework for modeling potential nonlinear associations between continuous predictors and outcome variables, overcoming limitations of linearity assumptions [ 17 , 18 ]. In the context of an investigation into the association between sleep quality and the prevalence of T2D in patients suffering from chronic diseases, the proposed model has the capacity to accurately reveal the complex non-linear relationship between the two, thereby circumventing the potential bias that may be introduced by assuming a linear relationship. The RCS model facilitates a more precise analysis of the impact of sleep quality on T2D risk across a range of values, thereby providing a valuable instrument for gaining insight into the underlying relationship. Furthermore, the model has the capacity to adjust for potential confounders, thereby enhancing the reliability and accuracy of the study results. Consequently, the primary objectives are to investigate the association between sleep quality and the prevalence of T2D in patients with chronic diseases using the RCS approach. The secondary objective is to determine the critical sleep parameter thresholds that may affect the prevalence of T2D, and to provide a more scientific theoretical basis for clinical intervention and disease prevention. 2. Materials and methods 2.1. Participants From 5 January to 30 March 2025, a cohort of patients was recruited from community hospitals in Jiangsu and Shanghai. The inclusion criteria were as follows: (1) age of 18 years or older; (2) presence of chronic diseases such as hypertension, diabetes, cardiovascular diseases, and chronic obstructive pulmonary disease (COPD). The exclusion criteria included: (1) withdrawal during the study period; (2) incomplete key information (e.g., age, gender, height, weight); (3) severe mental disorders; (4) severe visual or hearing impairments; (5) central nervous system diseases (e.g., stroke, Parkinson’s disease). All participants are required to sign a written informed consent form. 2.2. Instruments and measurements A validated demographic survey instrument was administered to document baseline characteristics, encompassing biological parameters (age, sex, BMI-derived metrics), educational background, and additional population-relevant variables. This study employed the Pittsburgh Sleep Quality Index (PSQI), a version validated in Chinese, for assessment. The Cronbach’s α coefficient of this index was 0.84 [ 19 ]. This tool quantifies sleep quality via seven psychometrically derived subscales assessing: sleep initiation latency, nightly sleep duration, sleep maintenance efficiency, sleep fragmentation events, self-rated sleep satisfaction, pharmacological sleep aid consumption, and diurnal functional impairment. Component scores (0–3 points each) aggregate to produce a composite index (0–21 points), where higher scores reflect poorer sleep qualitywhere higher scores reflect poorer sleep quality. Globally, a total score of ≤5 is commonly considered indicative of good sleep quality, whereas scores >5 suggest poor sleep quality [ 20 ]. The diagnosis of T2D in this study was determined primarily on the basis of standardized medical records from community health centers, and all participants were documented patients with chronic disease management whose diagnostic information was based on the WHO 1999 classification criteria and clearly documented [ 21 ]. 2.3. Sampling method As illustrated in Fig 1 , a total of 1,109 participants were initially approached, of whom 902 completed the survey (response rate: 81.3%). Exclusions included age 0.05). Continuous variables were expressed as mean ± standard deviation (SD), whilst categorical variables were expressed as percentage. Between-group differences in baseline variables were assessed by means of independent samples t-tests for continuous variables and chi-square tests for categorical variables. Between-group differences in baseline variables were assessed by means of t-tests and chi-square tests. Multiple logistic regression models were developed to express the findings as odds ratio (OR) and 95% confidence interval (CI). Three continuous models were used: model 1 (unadjusted); model 2 (adjusted for age and education level); and model 3 (further adjusted for body mass index, physical activity, smoking status, and alcohol consumption in addition to the covariates in model 2). To analyze the potentially nonlinear association between sleep quality and T2D prevalence, we employed the RCS regression model, a flexible statistical method that allows for the detection of threshold effects and complex dose-response patterns without imposing linearity assumptions [ 24 ]. All statistical analyses were performed using R software (version 4.3.3) with a two-tailed p-value threshold of 0.05 for statistical significance. 2.5. Ethical approval The process of obtaining ethical approval and consent to participate is a prerequisite for the initiation of any research study.This study was conducted in accordance with the Helsinki Declaration and received academic ethics review from the College of Health Management of Shanghai Jian Qiao University. The ethics review number is (No: 2024-12-236). 3. Results 3.1. Basic characteristics of the object of study Table 1 presents the baseline characteristics of all participants according to the T2D subgroups. In the present study, a total of 902 valid samples were analyzed, of which 565 (62.64%) were found to have T2D. The mean age of the study participants was found to be 59.38 years (±15.09) and 43.02% of them were female. The diabetic group exhibited higher levels of PSQI (11.83 ± 2.02 vs. 9.90 ± 2.72) and BMI (24.51 ± 2.15 vs. 23.96 ± 2.14), and were of a greater mean age in comparison to the non-diabetic group. Furthermore, non-diabetic participants were more likely to have attained higher levels of education, to engage in more frequent exercise, and to currently smoke and consume alcohol. (See Table 1 for details). The schematic representation of specimen acquisition methodology is presented in Fig 1 . Download: PNG larger image TIFF original image Table 1. Demographic characteristics and disease-related factors of the study population. https://doi.org/10.1371/journal.pone.0356250.t001 3.2. Associations between sleep quality and prevalence of T2D Multiple logistic regression analyses were employed to explore the relationship between PSQI and T2D, and the results are presented in Table 2 . A positive association between PSQI (continuous variable) and the occurrence of DM was found in all models, both unadjusted and adjusted. The odds of developing DM increased by 41.1%, 38.3%, and 35.4% for each unit increase in T2D, respectively, from Model 1 to Model 3. Kindly direct your attention to Table 2 for further details. Download: PNG larger image TIFF original image Table 2. Associations between sleep quality and diabetes. https://doi.org/10.1371/journal.pone.0356250.t002 3.3. Restricted cubic spline regression findings The covariate-adjusted RCS model revealed a statistically significant nonlinear association between PSQI scores and T2D prevalence ( P -non linearity 13. Multiple logistic regression models, with PSQI as a categorical exposure variable, showed consistent cross-sectional associations in both crude and adjusted analyses ( Table 3 ). Participants in both higher PSQI categories (12–13 and >13) had statistically significantly higher odds of prevalent T2D compared with the reference group (all P < 0.05; P for trend <0.05). These findings support a dose–response pattern between poorer sleep quality and higher odds of prevalence of T2D in this cross-sectional study. Download: PNG larger image TIFF original image Table 3. Relationship between categorized sleep quality measures and prevalence of T2D using RCS-derived thresholds. https://doi.org/10.1371/journal.pone.0356250.t003 Download: PNG larger image TIFF original image Fig 2. The dose-response relationship between PSQI score and prevalence of T2D. https://doi.org/10.1371/journal.pone.0356250.g002 4. Discussions In this study, the PSQI total score was used as a composite measure of sleep quality. The findings of the present study demonstrated a positive association between PSQI scores and the OR of prevalence of T2D. Furthermore, it was observed that patients diagnosed with T2D exhibited significantly higher sleep quality scores in comparison to those without T2D. Higher PSQI scores were found to be associated with greater odds of T2D, even after adjusting for relevant confounding factors.This is consistent with the research results of Li [ 25 ] and others. Poor sleep quality may elevate T2D risk through dual physiological pathways: primarily by diminishing insulin sensitivity and, secondarily, impairing glucose tolerance. These combined metabolic disturbances result in sustained hyperglycemia, consequently increasing susceptibi
## Related Clinical Research

- [Moving Forward: The Future of GLP-1 Therapies](https://medichelpline.com/clinical-feed/endocrine-news-18-moving-forward-the-future-of-glp-1-therapies.md)
- [Waist circumference predicts mortality more than BMI in adults 65+: study findings](https://medichelpline.com/clinical-feed/medical-news-today-0-bmi-vs-waist-size-study-finds-belly-fat-matters-more-than-weight-for-death-risk.md)
- [Diet and Alzheimer’s risk: Why no single best diet fits everyone](https://medichelpline.com/clinical-feed/medical-news-today-0-is-there-a-best-diet-to-help-lower-alzheimer-s-risk.md)
- [Caffeinated Coffee and Heart Health: How Much Coffee Is Safe Daily?](https://medichelpline.com/clinical-feed/aha-news-0-coffee-and-heart-health-how-many-cups-of-caffeinated-coffee-are-safe-to-drink.md)
- [Single-cell scTWAS identifies immune cell–specific genes linked to gestational diabetes](https://medichelpline.com/clinical-feed/plos-one-0-leveraging-expression-quantitative-trait-loci-information-in-single-cell.md)

## Navigation
- [← Back to Endocrinology Feed](https://medichelpline.com/clinical-feed/endocrinology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.