Florida Public Health Review

Publication Date



Background: With over two million people suffering from opioid abuse disorders, the Centers for Medicare and Medicaid Services has identified opioid abuse as a key priority. Florida is one of eight states labeled as a high-burden opioid abuse and is an “epicenter” for opioid use and misuse.

Purpose: The purpose of this study was to discover potential predictors of opioid abuse in Florida by exploring specific healthcare delivery, geographic, and patient demographic factors.

Methods: A retrospective longitudinal study design was used to examine four years (2014-2017) of Florida inpatient administrative discharge data across 173 hospitals of opioid abuse rate. Main measures included, opioid abuse counts (n=12,804) defined using both ICD-9-CM and ICD-10-CM systems. Negative binomial regression was used to estimate the association between hospital factors, county factor, and opioid abuse hospital rates.

Results: We found a statistically significant association between hospital opioid abuse count and hospital size, location, teaching status, patients’ average age, gender, and race. The estimated probability of opioid abuse for a patient if treated in a large hospital is 0.23 (about 23%), significantly higher than small (8%) and medium (17%) size hospitals. The estimated probability of opioid abuse for a patient if treated in a rural hospital is 0.12 (about 12%), while in an urban hospital is higher at 0.17 (about 17%). The risk ratio is 0.71, which means the risk decreased by two-thirds when treated in rural hospitals. We also found that hospitals with a younger patient population, a higher percent of males and a higher proportion of Caucasian patients, are at a higher risk for an increase in opioid abuse counts.

Discussion: These findings provide policymakers with crucial insight into Florida’s opioid crisis and the identification of predictive factors that contribute to opioid abuse.