Exploring the appropriateness of antibiotic prescribing for common respiratory tract infections in UK primary care

Objective: To use illness severity scores to evaluate the appropriateness of antibiotic prescribing in United Kingdom general practice

Methods: To describe variations in practice prescribing rates, taking account of illness severity. We used three scores in three studies to measure severity: ‘FeverPAIN’ in an adult acute sore throat cohort (n=12,829), the ‘3C score’ in an adult acute lower respiratory infection cohort (n=28,883), and the STARWAVe score in an acute cough and respiratory infection children’s cohort (n=8,394). We calculated median odds ratios to quantify practice-level variation in prescribing rates, adjusted for illness severity.

Results: There was substantial variability in practice prescribing rates (ranges 0%-97%, 7%-100%, and 0%-75% in the three cohorts respectively). There was evidence that higher prescribing practices saw a higher proportion of unwell patients. At the individual level, more unwell patients were more likely to receive a prescription but prescribing levels for those with low scores were still high. The median odds ratio was 2.6 (95% CrI2.3-2.9) in the sore throat dataset, 2.9 (95% CrI 2.6-3.2) in the adult cough dataset and 2.1 (95% CrI 1.8-2.4) in the children’s cough dataset.

Conclusion: Higher prescribing practices may see more unwell patients with high illness severity scores, but the differences in scores accounts for a minority of between practice prescribing variation. There is likely to be scope for further reductions in antibiotic prescribing among patients with low illness severity scores. Further research is needed to explore the additional factors which account for variation in prescribing levels.