Analysis of Hospitalization for Co-Morbid Mental and Substance Use Disorders

Monday, June 5, 2017: 4:50 PM
420A, Boise Centre
Luigi F Garcia Saavedra , New Mexico Department of Health, Santa Fe, NM
Jessica R Reno , University of New Mexico, Albuquerque, NM

BACKGROUND: Drug overdose is a major public health concern in New Mexico (NM) and the United States (US). Suicide is also a persistent and increasingly important public health issue in NM. Mortality resulting from drug overdose and suicide stem from underlying morbidity related to substance use (SUD) and mental disorders (MD). This study describes the rate of hospitalizations and estimates the cost of Medicaid reimbursements for healthcare claims due to SUD/MD comorbidity.

METHODS: The 2014-2015 NM Hospital Inpatient Discharge Dataset (HIDD) was used to calculate the hospitalization rates. The 2014-2015 Medicaid Medical Assistance Division (MAD) claims dataset was used to calculate rates of healthcare utilization. HIDD and MAD records related to SUD or MD were identified using ICD-9 coding recommendations from the Council of State and Territorial Epidemiologists Workgroup for Substance Abuse and Mental Health Surveillance Indicators, with some modifications which included an additional mental health sub-indicator to quantify the number of non-drug-induced mental disorders diagnosed, cross-walking ICD-10 codes with the original ICD-9 codes used to identify substance use disorders, and secondary diagnoses to capture all co-morbid conditions diagnosed in one visit.

RESULTS: Between 2014-2015, 2,907 Medicaid clients were diagnosed with co-morbid SUD/MD. Preliminary analysis reveal that 2.4% of clients with primary diagnosis of MD also had SUD diagnosis, and 99.6% of those with primary diagnosis of SUD also had MD diagnosis. The most commonly diagnosed SUD among clients with primary MD diagnosis was prescription opioid poisoning. 99% of clients with co-morbid SUD/MD were hospitalized, compared to 75% of clients with only MD or SUD. Similarly, 82% of clients with co-morbid SUD/MD were admitted to an emergency room, compared to 56% of clients with only MD or SUD.

CONCLUSIONS: Preliminary results demonstrate the high prevalence of SUD among individuals with MD, and the relatively high utilization of costly healthcare services for those with co-morbid diagnoses. Efforts should be made to increase the utilization of outpatient behavioral health care for those with co-morbid SUD/MD, as only 75% of these clients had seen a behavioral health clinician. Increasing this lower cost, preventive healthcare service for high-risk clients could decrease inpatient hospitalizations and emergency room visits.