Determinants of Inappropriate opioid prescribing: A cross sectional study in a South London population

Problem

Opioid prescribing in the UK has been increasing in recent years. While opioids are known to be effective for treating cancer related pain, their role in non-cancer pain is less clear. There is emerging evidence that long-term use of opioids for non-cancer pain can lead to adverse health outcomes, including dependency, falls and premature mortality.Opioid prescribing has been noted to be more likely in deprived groups; this may be due to confounding by other health factors. The aim of this study was to examine determinants of opioid PIP, and the relation to deprivation, adjusted for potential confounders.

Approach

Study design: Retrospective cross-sectional study of patients reported according to the RECORD-PE criteria. Prescribing and clinical records for all patients aged ≥18 years registered at a Lambeth GP for the period 03/06/19-30/11/19 were extracted (N=323,980; Practices N=41). Exclusions: Cancer diagnoses, substance abuse disorders prescribed oral methadone or buprenorphine. Potentially inappropriate prescribing (PIP) of opioids was defined using the Guidelines of the Faculty of Pain Medicine; PIP being greater than 90 days of continuous opioid use for non-cancer pain or a morphine daily dose equivalent of >120mg. Both partially and fully adjusted multilevel (adjusted for practice) logistic regression models were developed to analyse the association between PIP and patient deprivation (Indices of Multiple Deprivation), as well as other relevant covariates (age, sex and co-morbidities).

Findings

Opioid prescribing rates were 2% in Lambeth, of those patients prescribed an opioid in the last 3 months, more than half met the criteria for PIP. Determinants of inappropriate prescribing in the fully adjusted model were, increasing multi-morbidity, increasing age, relative deprivation and female sex. Following adjustment, relative deprivation was associated with inappropriate prescribing of an opioid with those in the most deprived quintile in Lambeth at increased risk of being inappropriately prescribed an opioid compared with the least (AOR 1.59 , 95%CI 1.41-1.79,p<0.001).Patients also had increasing odds of being inappropriately prescribed opioids if female (AOR 1.33, 95%CI 1.24-1.43,p<0.001), increasing age (p<0.001) and increasing multi-morbidity (p<0.001).

Consequences

We found lower prescribing rates of opioids in Lambeth significantly lower than national figures. This would be in line with previous studies that have shown lower prescribing rates for opioids in London. Key determinants of inappropriate opioid prescribing were multi-morbidity, increasing age, relative deprivation and female sex. This study has identified inequalities in opioid prescribing, with potential for avoidable drug dependencies and undesirable side effects. Identifying and addressing reasons for this variation will help improve prescribing quality. I think this would be worthy of further exploration looking at PIP trends overtime and validating these findings in a national database.