SIGNIFICANT NATIONWIDE VARIATION EXISTS IN OPIOID PRESCRIPTION RATES AMONGST PEDIATRIC PATIENTS WITH ACL INJURIES: A RETROSPECTIVE COHORT STUDY

Abstract

Background:

Prescription opioid rates have been found to vary on a county and state level. We aim to find if significant hospital-level and regional-level variation in opioid prescription rates exists in a pediatric orthopedic setting.

Methods:

We conducted a retrospective cohort study of children ages 10-18 who sustained an ACL injury between 2012-2016, from 39 hospitals throughout the United States, found through the Pediatric Health Information System. Patients were divided into four regions according to the regional census of hospital admission, and demographics were recorded. Primary outcomes were hospital-level variation in opioid prescription rates using a mixed effects linear regression model to compare mean differences in average opioid prescription rate.

Results:

11,452 pediatric patients were analyzed for regional-level variation in opioid prescription rates. Of those patients 7,879 were analyzed for hospital-level variation in opioid prescription rates based on if the hospital of incidence had a sample size of at least 250 patients over the five-year span (16 of the 43 hospitals fit this criteria). The most common demographics were patients ages 16-18 (n=5,613; 49.01%), white (n=6,588; 57.53%), and male (n=5,905; 51.56%). The 11,452 patients were divided into four regions: (1) Northeast (n=2,409; 21.04%), (2) Midwest (n=2,499; 21.82%), (3) South (n=3,869; 33.78%), and (4) West (n=2,675; 23.36%). After adjustment, significant variation in opioid prescription rates was found to exist at both a hospital-level using a random intercept and random slope model (-2LL=17882.381, p<.05) and at a regional-level by comparing mean differences in average opioid prescription rate (p<.05).

Conclusions:

Significant variation of opioid prescription rates exists at both a hospital-level and a regional-level over the last five years in a pediatric orthopedic setting. The findings of this study build off previous findings from the Centers for Disease Control, which found significant variation existed in opioid prescription rates on a state-wide level. Given the recent guidelines proposed by the CDC to control and limit prescription opioids (as of March 2016), future studies finding the impact of the guidelines on variation of prescription rates in a pediatric orthopedic setting could be beneficial.

Table 1:.

Mean Difference in Opioid Prescription Rate (Regional Comparison by Year)

Region 1	Region 2	2012	2013	2014	2015	2016
Midwest	Northeast	0.608*	0.738*	0.829*	0.512*	0.634*
Midwest	South	-0.109	-0.165*	0.020	-0.153*	-0.014
Midwest	West	0.113	0.133	0.417*	0.247*	0.335*
Northeast	South	-0.717*	-0.904*	-0.809*	-0.665*	-0.648*
Northeast	West	-0.495*	-0.606*	-0.411*	-0.266*	-0.299*
South	West	0.223*	0.298*	0.397*	0.400*	0.350*

*p<.05

Table 1: A comparison of mean differences in opioid prescription rates. Table 1 compares the mean differences between opioid prescription rates for five years, on a regional level. An asterisk (*) indicates that at p<.05, the opioid prescription rate varied significantly for the regions compared within that year. A negative value indicates that region 1 prescribed fewer opioids than region 2. A positive value indicates that region 2 prescribed more opioids than region 1.

Figure 1:.

Total opioid variation across years describing regional variation