Effect on antibiotic prescribing of repeated clinical prompts to use a sore throat score: lessons from a failed community intervention study

Abstract

Objectives: Infections with group A streptococcus (GAS) occur in 10% to 20% of patients with sore throats, whereas antibiotics are prescribed 50% of the time. Clinical scoring rules can more accurately predict the likelihood of GAS infection, but whether family physicians will adopt such approaches is unclear. This study sought to determine whether repeated clinical prompts to use a scoring approach could help family physicians lower antibiotic use in patients with a sore throat.

Study design: Randomized trial in which physicians were assigned to use either (1) chart stickers that prompted them to calculate a score based on clinical findings and provided management recommendations linked to score totals or (2) a clinical checklist.

Population: Ninety-seven family physicians in Ontario, Canada, assessed 621 children and adults with sore throat and obtained a throat swab for culture.

Outcomes measured: (1) Unnecessary antibiotic prescriptions given to patients with a negative throat culture and (2) overall antibiotic use.

Results: There were no differences between the control and intervention group in unnecessary antibiotic prescriptions (16.1% vs 20.4%, respectively, P =.29) or overall antibiotic use (27.9% vs 28.1%, P =.97). However, a number of physicians dropped out of the study; as a result, the characteristics of the physicians in the 2 groups were dissimilar in factors related to prescribing. After adjusting for these differences and patient clustering by physician, the odds ratio for the effect of the intervention on unnecessary antibiotic prescriptions was 0.76 (95% confidence interval [CI] = 0.42, 1.40) and 0.57 for overall antibiotic use (95% CI = 0.27, 1.17).

Conclusions: Chart prompts during clinical encounters to use a clinical score in the assessment of patients with a sore throat did not reduce unnecessary antibiotic prescribing by family physicians. The problems encountered in conducting this community-based intervention trial are discussed in relation to the negative result.