Incidence of iatrogenic opioid dependence or abuse in patients with pain who were exposed to opioid analgesic therapy: a systematic review and meta-analysis

Background: The prevalence and incidence of chronic conditions, such as pain and opioid dependence, have implications for policy development, resource allocation, and healthcare delivery. The primary objective of the current review was to estimate the incidence of iatrogenic opioid dependence or abuse after treatment with opioid analgesics. Methods: Systematic electronic searches utilised six research databases (Embase, Medline, PubMed, Cinahl Plus, Web of Science, OpenGrey). A 'grey' literature search and a reference search of included articles were also undertaken. The PICOS framework was used to develop search strategies and the findings are reported in accordance with the PRISMA Statement. Results: After eligibility reviews of 6164 articles, 12 studies (involving 310 408 participants) were retained for inclusion in the meta-analyses. A random effects model (DerSimonian-Laird method) generated a pooled incidence of opioid dependence or abuse of 4.7%. There was little within-study risk of bias and no significant publication bias; however, substantial heterogeneity was found among study effects (99.78%). Sensitivity analyses indicated that the diagnostic criteria selected for identifying opioid dependence or abuse (Diagnostic Statistical Manual (DSM-IV) vs International Classification of Diseases (ICD-9)) accounted for 20% and duration of exposure to opioid analgesics accounted for 18% of variance in study effects. Longer-term opioid analgesic exposure, and prescription of strong rather than weak opioids, were associated with a significantly lower incidence of opioid dependence or abuse. Conclusions: The incidence of iatrogenic opioid dependence or abuse was 4.7% of those prescribed opioids for pain. Further research is required to confirm the potential for our findings to inform prevention of this serious adverse event.