Low-value pharmaceutical care among Dutch GPs: a retrospective cohort study

Abstract

Background: Low-value pharmaceutical care exists in general practice. However, the extent among Dutch GPs remains unknown.

Aim: To assess the prevalence of low-value pharmaceutical care among Dutch GPs.

Design and setting: Retrospective cohort study using data from patient records.

Method: The prevalence of three types of pharmaceutical care prescribed by GPs between 2016 and 2019 were examined: topical antibiotics for conjunctivitis, benzodiazepines for non-specific lower back pain, and chronic acid-reducing medication (ARM) prescriptions. Multilevel logistic regression analysis was performed to assess prescribing variation and the influence of patient characteristics on receiving a low-value prescription.

Results: Large variation in prevalence as well as practice variation was observed among the types of low-value pharmaceutical GP care examined. Between 53% and 61% of patients received an inappropriate antibiotics prescription for conjunctivitis, around 3% of patients with lower back pain received an inappropriate benzodiazepine prescription, and 88% received an inappropriate chronic ARM prescription during the years examined. The odds of receiving an inappropriate antibiotic or benzodiazepine prescription increased with age (P<0.001), but decreased for chronic inappropriate ARM prescriptions (P<0.001). Sex affected only the odds of receiving a non-indicated chronic ARM, with males being at higher risk (P<0.001). The odds of receiving an inappropriate ARM increased with increasing neighbourhood socioeconomic status (P<0.05). Increasing practice size decreased the odds of inappropriate antibiotic and benzodiazepine prescriptions (P<0.001).

Conclusion: The results show that the prevalence of low-value pharmaceutical GP care varies among these three clinical problems. Significant variation in inappropriate prescribing exists between different types of pharmaceutical care - and GP practices.

Keywords: general practice; general practitioners; inappropriate prescribing; medical overuse.

Figure 1.

Estimates of the prevalence of patients receiving one of three types of low-value care. Associated with a) conjunctivitis; b) non-specific lower back pain; and c) acid-reducing medication.

Figure 2.

Proportion of patients in each practice who received each of the low-value prescriptions for care at least once during 2019. Associated with a) conjunctivitis; b) lower back pain; and c) acid-reducing medication. (The practice numbers do not directly correlate to the practice numbers as provided in Supplementary Table S3.)

Figure 3.

Adjusted odds ratios and confidence intervals associated with patient characteristics for all three types of low-value GP care. Associated with a) conjunctivitis; b) lower back pain; and c) acid-reducing medication. ^aReference categories. SES = socioeconomic status.