Is C-reactive protein the answer to antibiotics over-prescribing in primary care? A behavioural experiment in two...

Antibiotic overprescribing fuels antimicrobial resistance. C-reactive protein (CRP) is widely promoted as a point-of-care test for targeted prescribing. While routinely used in some European countries such as Norway, it is not part of standard practice in UK primary care. Understanding how CRP testing might impact general practitioners' (GP) judgement is critical for informed implementation. We explored this with UK and Norwegian GPs.

One hundred and one GPs per country responded to four scenarios, describing patients with a range of infectious or inflammatory conditions. After reading a patient description, GPs could request further information, including CRP. The CRP result was either high or low, randomly assigned per GP and scenario. Each time GPs obtained new information, they indicated whether antibiotics were needed (Likert scale). At the end, they selected a management option.

Norwegian GPs requested CRP more often than UK GPs [77% versus 28%, odds ratio (OR) = 18.09 (95% CI 9.86-33.20), P < 0.001], but prescribed antibiotics at similar rates [39% versus 42%, OR = 1.36 (95% CI 0.96-1.94), P = 0.085]. When CRP was requested, elevated values increased prescribing odds [OR = 1.84 (95% CI 1.15-2.97), P = 0.012], while low values had no impact. If GPs believed antibiotics were needed, a low CRP did not reassure them. Higher uncertainty led to more CRP requests, but these did not reduce uncertainty.

Despite marked differences in CRP requests, we detected no between-country differences in prescribing. Elevated CRP increased prescribing, whereas low CRP was not sufficiently reassuring. The results do not support routine use of CRP as an antimicrobial stewardship tool in primary care.