Effect of Combination of Point-of-Care C-Reactive Protein Testing and General Practitioner Education and Long-Term Effect of Education on Reducing Antibiotic Prescribing for Children Presenting with Acute Infections in General Practice in Latvia : A Randomized Controlled Intervention Study

Abstract

BACKGROUND: Antibiotics are often overprescribed in children in general practice. We investigated whether the availability of C-reactive protein point-of-care testing (CRP POCT) in daily practice and general practitioner (GP) education reduces antibiotic prescribing for children with acute infections and whether GP education has a long-term effect on antibiotic prescribing. METHODS: This was a randomized controlled intervention study with randomization at the GP practice level. Eligible patients were children aged 1 month to 17 years presenting to general practice with an acute infection. INTERVENTIONS: In the first study period, one GP group received combined interventions (CRP POCT was provided for daily use in combination with a live educational session), while the second GP group continued usual care. During the second study period, the GP groups were switched. During this period, the long-term education effect was evaluated in the GP group, which had previously received both interventions: the CRP POCT was no longer available in their practices in accordance with the study protocol, but education could have a lasting effect. PRIMARY OUTCOME: Antibiotic prescribing at index consultation. RESULTS: GP with combined intervention enrolled 1784 patients, GP with usual care enrolled 886 patients, and GP with long-term education effect enrolled 647 patients. Most of the patients had upper (76.8%) and lower (18.8%) respiratory infections. In total, 29.3% of the study patients received antibiotic prescriptions. Adjusted binary logistic regression analysis showed no differences for the primary outcome between GPs with usual care and GPs with combined intervention (aOR 0.89 (0.74-1.07), p = 0.20), but significantly lower antibiotic prescribing was observed for GPs with long-term education in comparison with GPs with usual care (aOR 0.75 (0.59-0.96), p = 0.02); however, after multilevel analyses, any differences in the antibiotic prescription between intervention groups became non-significant. GPs widely used CRP POCT when it was available in practice (for 69.1% of patients in the combined intervention group), but rarely measured CRP in the laboratory in the usual care group (8.8% (n = 78)) or long-term education group (14.8% (n = 98)). The majority of the tested patients had low CRP levels (below 20 mg/L); despite this, up to 35.4% of them received antibiotic prescriptions. CONCLUSIONS: Our results show that the availability of CRP POCT and educational training for GPs together did not reduce antibiotic prescribing, and one-time education did not have a long-term effect on antibiotic prescribing.