Inpatient Opioid Use Disorder and Social Determinants of Health: A Nationwide Analysis of the National Inpatient Sample (2012-2014 and 2016-2017).

Objective To evaluate the trends and relationship of inpatient presentations of Social Determinant of Health (SDOH) with superimposed Opioid Use Disorder (OUD), comparing 2012-2014 (ICD-9) and 2016-2017(ICD-10). Methods We used the Nationwide Inpatient Sample (NIS) from the Healthcare Cost and Utilization Project (HCUP). We identified OUD among patients with any record of SDOH as the primary or secondary diagnosis using the International Classification of Diseases (ICD)-9/10 codes. A weighted SDOH sample size of 3,002,558 (2.8%) and 1,254,899(1.8%) was included for 2012-2014 and 2016-2017, respectively. The main predictors include census division, race, gender, and covariates, including age, income, disposition, payer, rural-urban classification, and combined SDOH indicator, which was used as a control variable in the regression analysis. The study provides a descriptive analysis of the social determinant of health in relation to OUD. We also evaluated the rate of the presentation by age group and race. Results A sample of 367,960 (12%) and 153,535 (12%) OUD presentations with SDOH indicators were identified for 2012-2014 and 2016-2017, respectively. An increase in housing difficulties between 2016-2017 (45%) as compared to 2012-2014 (20%) was observed. A statistically significant higher odds of presentation among black and other races were observed. There was significant variance in the presentation by region with Middle Atlantic with OR-2.04, 95 C. I (1.98, 2.10) and East North Central with OR-1.94, 95 C. I (1.89, 1.99). Higher admissions were observed for both 2012-2014 and 2016-2017 among patients without any payment method classified as no charge. Statistically significant relationships (p<0.05) were also observed among low income <$54,000 OR-1.01, 95 C. I (0.99, 1.04), Housing OR-1.49, 95 C. I (1.47, 1.52), Primary Support OR-0.93, 95 C. I (0.90, 0.95), Employment OR-1.37, 95 C. I (1.35, 1.39), Psychosocial OR-1.63, 95 C. I (1.59, 1.67), Age group 25-34 OR-15.64, 95 C. I (14.20, 17.22), 35-44 OR-29.07, 95 C. I (26.45, 31.95) in both 2012-2014 and 2016-2017. Conclusion SDOH has a direct impact on inpatient OUD presentations. Socio-economic disparities exist in all census regions, race, sex, and rural-urban demographics. Interventions aimed at reducing the incidence and risk of OUD should focus on specific local dynamics using a multidisciplinary, data-driven quality improvement (QI) approach to address the root cause of presentations effectively. A community-based approach to addressing SDOH through collaboration with care providers could play a substantial role in decreasing length of stay (LOS), cost, and potential readmission among these populations.