Substance use profiles predict health outcomes among U.S. adults with diabetes: Implications for primary care screening and intervention.

AIMS: Substance use is common among adults with diabetes yet remains poorly characterized in terms of co-occurring patterns. This study sought to identify distinct substance use profiles among U.S. adults with diabetes, characterize factors associated with profile membership, and examine their clinical relevance.

METHODS

Data were from the 2022-2023 National Survey on Drug Use and Health (n = 89,889; 6805 with diabetes). Latent class analysis identified substance use profiles. Multinomial logistic regression examined demographic, socioeconomic, and mental health characteristics associated with profile membership. Weighted regression models evaluated associations between membership and health indicators.

RESULTS

Three profiles emerged: Low/Non-Use (67.6%), Cannabis and Binge Drinking (24.4%), and Polysubstance Use (8.0%). Higher-risk membership was associated with male sex, mental health burden, and urban residence; marriage and college education were protective. Both higher-risk classes were more than twice as likely to report poor self-rated health. Polysubstance Use class members had significantly higher odds of hospitalization (OR = 1.46).

CONCLUSIONS

Substance use among adults with diabetes is heterogeneous and linked to measurably worse health status outcomes including poorer self-rated health and higher hospitalization risk. These findings support routine, risk-stratified screening for substance use and mental health in diabetes care and may help inform targeted interventions.