The Impact of Comorbid Sleep-Disordered Breathing on Hospitalization Risk Related to Diabetes and Atherosclerotic Disease: A Retrospective Cohort Analysis

Abstract

Objective: To determine the relationship between comorbid sleep-disordered breathing (SDB) and hospitalization rates related to diabetes mellitus (DM) and atherosclerotic disease (AD). Methods: This study used a retrospective cohort design from a large medical claims database with 5 years of data between 2018 and 2022. The presences of SDB, DM, and AD were identified using International Classification of Diseases (ICD-10) and relevant Current Procedural Terminology (CPT) codes. Hospitalizations related to DM and AD were identified primarily using Place of Service (POS) code 21. Propensity-score matching was first used on data from the entire 5-year period to select matched controls (unadjusted n = 883,910, adjusted n = 888,619) compared to those diagnosed with SDB (n = 519,818) on hospitalization rates during the concurrent 5-year period. A second analysis used propensity-score matching on data from year 1 only to select matched controls (unadjusted n = 248,848, adjusted n = 260,298) compared to those diagnosed with SDB in year 1 (n = 193,671) on hospitalization outcomes in the subsequent 4-year period. Results: Odds ratios (ORs) revealed a significant association between SDB diagnosis and hospitalizations related to DM (OR 1.23-1.71), AD (OR: 1.08-1.34), and either condition (OR 1.17-1.49) in both analyses. Post hoc analysis revealed sex differences in the relationship between SDB and future hospitalizations, with females showing a pattern of significantly elevated risk across all future hospitalization outcomes (OR: 1.25-1.44), whereas males were found to have a significant relationship between SDB diagnosis and future DM hospitalization only (OR 1.10). Conclusions: These findings provide real-world evidence that comorbid SDB increases the risk for hospitalizations related to chronic cardiometabolic conditions. Sex is a potential moderator of this relationship and should be further explored.

Keywords: cardiovascular diseases; diabetes mellitus; hospitalization; sleep apnea syndromes.

Figure 1

Cohort selection for analyses. Flowchart of cohort selection for each analysis. Panel (a) depicts the cohort selection for analysis 1 (5-year matching). Panel (b) depicts the cohort selection for analysis 2 (year 1 matching). In each analysis, the SDB group is compared to the control group (adjusted) in the adjusted model and then compared to the control group (unadjusted) in the unadjusted model. G47.32 = high-altitude periodic breathing, and G47.35 = congenital central alveolar hypoventilation syndrome.

Figure 2

Odds ratios of hospitalization rates for SDB and matched controls. Figure 2 illustrates the odds ratios for hospitalizations due to DM and AD for males, females, and combined groups in the adjusted model in Analysis 1 (a) and in the adjusted model in Analysis 2 (b). There are notable differences in the odds ratios between males and females for both DM and AD hospitalizations, particularly in Analysis 2. OR = odds ratio, CI = confidence interval.