National U.S. time-trends in opioid use disorder hospitalizations and associated healthcare utilization and mortality.

backgroundThe opioid epidemic is a major public health crisis in the U.S. Contemporary data on opioid use disorder (OUD) related hospitalizations are needed. Our objective was to assess whether OUD hospitalizations and associated mortality are increasing over time and examine the factors associated healthcare utilization and mortality. METHODS AND

findingsWe examined the rates of OUD hospitalizations and associated mortality using the U.S. National Inpatient Sample (NIS) data from 1998-2016. Multivariable-adjusted logistic regression assessed the association of demographic, clinical and hospital characteristics with inpatient mortality and healthcare utilization (total hospital charges, discharge to a rehabilitation facility, length of hospital stay) during the index hospitalization for opioid use disorder. We calculated the odds ratio (OR) and 95% confidence intervals (CI). We estimated 781,767 OUD hospitalizations. The rate of OUD hospitalization and associated mortality (/100,000 overall NIS hospitalizations) increased from 59.8 and 1.2 in 1998-2000 to 190.7 and 5.9 in 2015-16, respectively. In the multivariable-adjusted analysis, the following factors were associated with worse outcomes; compared to age <34 years, older age was associated with higher risk of hospital charges above the median and length of stay >3 days, slightly higher risk of discharge to a rehabilitation facility. Higher Deyo-Charlson score was associated with higher hospital charges, length of hospital stay, and inpatient mortality. Women had lower odds of inpatient mortality than men and blacks had lower odds of mortality than whites.

conclusionsRising OUD hospitalizations from 1998 to 2016 and increasing associated inpatient mortality are concerning. Certain groups are at higher risk of poor utilization outcomes and inpatient mortality. Resources and healthcare policies need to focus on the high-risk group to reduce mortality and associated utilization.