Results from a Patient-Based Health Education Intervention in Reducing Antibiotic Use for Acute Upper Respiratory Tract Infections in the Private Sector Primary Care Setting in Singapore

Abstract

We investigated the efficacy of patient-targeted education in reducing antibiotic prescriptions for upper respiratory tract infections (URTIs) among adults in the private primary care setting in Singapore. Our randomized controlled trial enrolled patients aged 21 years and above presenting at general practitioner (GP) clinics with URTI symptoms for 7 days or less. Intervention arm patients were verbally educated via pamphlets about the etiology of URTIs, the role of antibiotics in treating URTIs, and the consequences of inappropriate antibiotic use. Control arm patients were educated on influenza vaccinations. Both arms were compared regarding the proportions prescribed antibiotics and the patients' postconsultation views. A total of 914 patients consulting 35 doctors from 24 clinics completed the study (457 in each arm). The demographics of patients in both arms were similar, and 19.1% were prescribed an antibiotic, but this varied from 0% to 70% for individual GPs. The intervention did not significantly reduce antibiotic prescriptions (odds ratio [OR], 1.20; 95% confidence interval [CI], 0.83-1.73) except in patients of Indian ethnicity (OR, 0.28; 95% CI, 0.09-0.93). Positive associations between the intervention and the view that antibiotics were not needed most of the time for URTIs (P = 0.047) and on being worried about the side effects of antibiotics (P = 0.018) were restricted to the Indian subgroup. GPs in limited liability partnerships or clinic chains prescribed less (OR, 0.36; 95% CI, 0.14 to 0.92), while certain inappropriate patient responses were associated with the receipt of antibiotics. Follow-up studies to investigate differences in responses to educational programs between ethnicities and to explore GP-targeted interventions are recommended.

Keywords: antibiotics; antimicrobial stewardship; intervention; upper respiratory tract infection.

FIG 1

Recruitment of participating GPs and patients.

FIG 2

Effects of intervention on patients' views about URTI and antibiotic use. Black, dark gray, light gray, and white sections of the bars represent strongly disagree, disagree, agree, and strongly agree, respectively, with P values from Mann-Whitney U tests given for the comparison of responses between the control and intervention arms.

FIG 3

Proportion of patients that were prescribed antibiotics in intervention and control arms, stratified by participating GP. Bars (left vertical axis) represent the percentages of control patients (dark gray), all patients (white with bold outline), and intervention patients (light gray) that were prescribed antibiotics. The diamonds coinciding with the middle bars are the differences in proportion with antibiotic prescriptions between the two arms (control − intervention, with error bars denoting the 95% confidence intervals [right vertical axis]). The numbers of interventions and control arm patients for that GP are in the brackets below the bars; two of the GPs are starred because the difference was statistically significant.