OBJECTIVES:To evaluate the expenditure and resources associated with narcotic users in the US, 2008-2017.

METHODS:We used a pooled serial cross-sectional design (2008-2017) based on a nationally representative sample from the Medical Expenditures Panel Survey. We compared annual healthcare expenditures (total, inpatient, outpatient, emergency department, prescription) and resource utilization (number of prescriptions, office-based visits, emergency department visits, and number of inpatient night stays) between narcotic and non-narcotic users. A propensity score algorithm was used to match non-narcotic respondents to narcotic respondents. Two-part generalized linear models were used to estimate the annual healthcare expenditures and resource utilization adjusting for multiple covariates. All costs were adjusted to 2019 US dollars using the gross domestic product deflator.

RESULTS:A total of 31,696,671 weighted narcotic users were matched to 31,536,639 weighted non-narcotic users after propensity score matching. Among narcotic users, healthcare expenditures and resource utilization were significantly higher compared to non-narcotic users. After adjustments, these differences remained significantly different. The adjusted annual total expenditures for narcotic and non-narcotic users were $16,542 and $7,067, respectively (P<0.001). Similarly, adjusted prescription ($3,067 vs $2,293), outpatient ($1,804 vs $650), emergency department ($703 vs $249), and inpatient expenditures ($5,610 vs $1,640) were significantly higher for the narcotic group compared to the non-narcotic group. Average resource utilization was also significantly higher in the narcotic group compared to the non-narcotic group adjusting for confounders (number of prescriptions filled: 29.5 vs 19.0; number of office-based visits: 12.4 vs. 8,1; number of emergency department visits: 0.59 vs 0.25; and number of inpatient night stays: 1.61 vs 0.58).

CONCLUSIONS:Narcotic users had higher healthcare expenditures and resource utilization compared to non-narcotic users from 2008-2017. Understanding the causes of these differences will be vital to optimizing treatment quality and effectiveness. These results may inform health policy initiatives that seek to reduce unnecessary wastage associated with narcotic users.