Mediating role of sleep quality in the relationship between diabetes mellitus and visual function: a cross-sectional study

Abstract

Diabetes mellitus (DM) is a major global health concern associated with complications that significantly impair quality of life. Among these, the impact of DM on visual function and the underlying mechanisms, particularly the role of sleep quality, remain insufficiently understood. This cross-sectional study recruited 629 participants from Henan Provincial People's Hospital to examine the relationships among DM, sleep quality, and visual function. Visual function was assessed using the National Eye Institute Visual Function Questionnaire-25 (NEI-VFQ-25), while sleep quality was evaluated with the Pittsburgh Sleep Quality Index (PSQI). Compared to non-diabetic controls, individuals with DM exhibited significantly poorer visual function and sleep quality. Path analysis further revealed that sleep quality played a significant mediating role in the relationship between DM and visual function, accounting for 42.2% of the total effect. Subgroup analyses indicated that female and elderly patients with DM were more susceptible to the influence of sleep disturbances on visual outcomes. These findings underscore the critical role of sleep quality in diabetes-related visual impairment and highlight the importance of incorporating routine sleep assessments and targeted interventions into diabetes management to help preserve visual function and improve overall quality of life in diabetic populations.

Keywords: Cross-Sectional study; Diabetes mellitus; Mediating effect; Sleep quality; Visual function.

Fig. 1

Comparison of overall VFQ scores and classification between control and DM groups. (A) Comparison of overall VFQ scores between the control and DM groups shows a significant difference, with lower scores in the DM group, indicating impaired visual function. (B) In male participants, overall VFQ scores are significantly lower in the DM group compared to the control group, suggesting a negative impact of DM on visual function in men. (C) In female participants, overall VFQ scores are significantly lower in the DM group, indicating that DM negatively affects visual function in women. (D) In younger participants (< 45 years), overall VFQ scores significantly decline in the DM group compared to the control group, suggesting the possibility of early visual impairment in this age group. (E) In middle-aged participants (45–60 years), overall VFQ scores remain significantly lower in the DM group, indicating the persistent negative impact of DM on vision. (F) In older participants (> 60 years), the DM group continues to show significantly lower overall VFQ scores, emphasizing the exacerbated impact of DM on visual health in the elderly. (G) Distribution of participants with good and poor visual function in the control and DM groups, showing that the proportion of participants with poor visual function is higher in the DM group across all age and gender subgroups. VFQ: visual function questionnaire; DM: diabetes mellitus; Con: control; * P < 0.05; ** P < 0.01; *** P < 0.001.

Fig. 2

Comparison of PSQI scores and sleep quality categories between control and DM groups. (A) Comparison of overall PSQI scores between the control and DM groups, showing that the DM group has significantly higher PSQI scores, indicating poorer sleep quality in DM patients. (B) In male participants, PSQI scores are significantly higher in the DM group compared to the control group, suggesting a negative impact of DM on sleep quality in men. (C) In female participants, the DM group shows significantly higher PSQI scores than the control group, indicating that the negative impact of DM on sleep quality is consistent across genders. (D) In younger participants (< 45 years), PSQI scores are significantly higher in the DM group, suggesting that younger diabetic patients experience poorer sleep quality. (E) Among middle-aged participants (45–60 years), the DM group also has significantly higher PSQI scores compared to the control group, indicating a significant impact of DM on sleep quality in this age group. (F) In older participants (> 60 years), PSQI scores remain significantly higher in the DM group, emphasizing a pronounced negative impact of DM on sleep quality in the elderly. (G) Proportion of participants with good and poor sleep quality based on PSQI scores in the control and DM groups. The results show that the proportion of participants with poor sleep quality is higher in the DM group across all age and gender subgroups. (H) Comparison of PSQI subscale scores between the control and DM groups, including sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, sleeping medications use, and daytime dysfunction. The results show that, except for the sleeping medications use, the DM group scores significantly worse on all subscales, suggesting a broader negative impact of DM on multiple dimensions of sleep quality. PSQI: pittsburgh sleep quality index; DM: diabetes mellitus; Con: control; * P < 0.05; ** P < 0.01; *** P < 0.001.

Fig. 3

Scatter plots of PSQI scores and overall VFQ scores. (A) Scatter plot showing the relationship between PSQI scores and overall VFQ scores across all participants. A significant negative correlation is observed, indicating that higher PSQI scores are associated with lower overall VFQ scores (R² = 0.2607, P < 0.001), suggesting that poorer sleep quality is linked to worse visual function. (B) In the control group, a negative correlation between PSQI scores and overall VFQ scores is also observed (R² = 0.1739, P < 0.001). This suggests that even in non-DM individuals, reduced sleep quality adversely affects visual function. (C) In the DM group, the negative correlation between PSQI scores and overall VFQ scores is more pronounced (R² = 0.2427, P < 0.001). Compared to the control group, the impact of poor sleep quality on visual function is greater in the DM group, indicating that DM patients are more sensitive to the effects of sleep quality on visual function. PSQI: pittsburgh sleep quality index; VFQ: visual function questionnaire; DM: diabetes mellitus; Con: control.

Fig. 4

Standardized path model illustrating the relationships among DM, PSQI scores, and overall VFQ scores. This standardized path model illustrates the relationships among DM, PSQI scores, and overall VFQ scores, while controlling for gender, age, and educational level. Path coefficients are displayed alongside the arrows. The diagram shows that DM has a direct negative effect on overall VFQ scores (β = -0.170) and an indirect negative effect mediated by PSQI scores (β=-0.453 * 0.274 = -0.124). Error terms (e1, e2, e3) are included to account for the unexplained variance in PSQI and VFQ scores. *P < 0.05, ** P < 0.01, *** P < 0.001.