Background Most patients presenting with acute exacerbations of chronic obstructive pulmonary disease (AECOPD) in primary care are prescribed antibiotics, but these may not be beneficial, and they can cause side effects and increase the risk of subsequent resistant infections. Point-of-care tests (POCTs) could safely reduce inappropriate antibiotic prescribing and antimicrobial resistance. Objective To determine whether or not the use of a C-reactive protein (CRP) POCT to guide prescribing decisions for AECOPD reduces antibiotic consumption without having a negative impact on chronic obstructive pulmonary disease (COPD) health status and is cost-effective. Design A multicentre, parallel-arm, randomised controlled open trial with an embedded process, and a health economic evaluation. Setting General practices in Wales and England. A UK NHS perspective was used for the economic analysis. Participants Adults (aged ≥ 40 years) with a primary care diagnosis of COPD, presenting with an AECOPD (with at least one of increased dyspnoea, increased sputum volume and increased sputum purulence) of between 24 hours’ and 21 days’ duration. Intervention CRP POCTs to guide antibiotic prescribing decisions for AECOPD, compared with usual care (no CRP POCT), using remote online randomisation. Main outcome measures Patient-reported antibiotic consumption for AECOPD within 4 weeks post randomisation and COPD health status as measured with the Clinical COPD Questionnaire (CCQ) at 2 weeks. For the economic evaluation, patient-reported resource use and the EuroQol-5 Dimensions were included. Results In total, 653 participants were randomised from 86 general practices. Three withdrew consent and one was randomised in error, leaving 324 participants in the usual-care arm and 325 participants in the CRP POCT arm. Antibiotics were consumed for AECOPD by 212 out of 274 participants (77.4%) and 150 out of 263 participants (57.0%) in the usual-care and CRP POCT arm, respectively [adjusted odds ratio 0.31, 95% confidence interval (CI) 0.20 to 0.47]. The CCQ analysis comprised 282 and 281 participants in the usual-care and CRP POCT arms, respectively, and the adjusted mean CCQ score difference at 2 weeks was 0.19 points (two-sided 90% CI –0.33 to –0.05 points). The upper limit of the CI did not contain the prespecified non-inferiority margin of 0.3. The total cost from a NHS perspective at 4 weeks was £17.59 per patient higher in the CRP POCT arm (95% CI –£34.80 to £69.98; p = 0.408). The mean incremental cost-effectiveness ratios were £222 per 1% reduction in antibiotic consumption compared with usual care at 4 weeks and £15,251 per quality-adjusted life-year gained at 6 months with no significant changes in sensitivity analyses. Patients and clinicians were generally supportive of including CRP POCT in the assessment of AECOPD. Conclusions A CRP POCT diagnostic strategy achieved meaningful reductions in patient-reported antibiotic consumption without impairing COPD health status or increasing costs. There were no associated harms and both patients and clinicians valued the diagnostic strategy. Future work Implementation studies that also build on our qualitative findings could help determine the effect of this intervention over the longer term. Trial registration Current Controlled Trials ISRCTN24346473. Funding This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 24, No. 15. See the NIHR Journals Library website for further project information.