Knowledge, attitudes and practices of Canadian pediatric emergency physicians regarding short-term opioid use: a...

In the midst of the current opioid crisis, physicians are caught between balancing children's optimal pain management and the risks of opioid therapy. This study describes pediatric emergency physicians' practice patterns for prescribing, knowledge and attitudes regarding, and perceived barriers to and facilitators of short-term use of opioids.

We created a survey tool using published methodology guidelines and distributed it from October to December 2017 to all physicians in the Pediatric Emergency Research Canada database using Dillman's tailored design method for mixed-mode surveys. We performed bivariable binomial logistic regressions to ascertain the effects of clinically significant variables (e.g., training, age, sex, degree of worry regarding severe adverse events) on use of opioids as a first-line treatment for moderate pain in the emergency department, and prescription of opioids for moderate or severe pain for at-home use in children.

Of the 224 physicians in the database, 136 (60.7%) completed the survey (60/111 [54.1%] women; median age 44 yr). Of the 136, 74 (54.4%) had subspecialty training. Intranasally administered fentanyl was the most commonly selected opioid for first-line treatment of moderate (47 respondents [34.6%]) and severe (82 [60.3%]) pain due to musculoskeletal injury. On a scale of 0 (not worried) to 100 (extremely worried), physicians' median score for worry regarding physical dependence was 6.0 (25th percentile 0.0, 75th percentile 16.0), for worry regarding addiction 10.0 (25th percentile 2.0, 75th percentile 20.0) and for worry regarding diversion of opioids 24.5 (25th percentile 14.0, 75th percentile 52.0). On a scale of 0 (not at all) to 100 (extremely), the median score for influence of the opioid crisis on willingness to prescribe opioids was 22.0 (25th percentile 8.0, 75th percentile 49.0). The top 3 reported barriers to prescribing opioids were parental reluctance (57 [41.9%]), lack of clear guidelines for pediatric opioid use (35 [25.7%]) and concern about adverse effects (33 [24.3%]). Binomial logistic regression did not identify any statistically significant variables affecting use of opioids in the emergency department or prescribed for use at home.

Emergency department physicians appeared minimally concerned about physical dependence, addiction risk and the current opioid crisis when prescribing opioids to children. Evidence-based development of guidelines and protocols for use of opioids in children may improve physicians' ability to manage pain in children responsibly and adequately.