Point-of-Care C-Reactive Protein Testing to Reduce Antibiotic Prescribing for Respiratory Tract Infections in Primary Care: Systematic Review and Meta-Analysis of Randomised Controlled Trials

Abstract

C-reactive protein (CRP) point-of-care testing (POCT) is increasingly being promoted to reduce diagnostic uncertainty and enhance antibiotic stewardship. In primary care, respiratory tract infections (RTIs) are the most common reason for inappropriate antibiotic prescribing, which is a major driver for antibiotic resistance. We systematically reviewed the available evidence on the impact of CRP-POCT on antibiotic prescribing for RTIs in primary care. Thirteen moderate to high-quality studies comprising 9844 participants met our inclusion criteria. Meta-analyses showed that CRP-POCT significantly reduced immediate antibiotic prescribing at the index consultation compared with usual care (RR 0.79, 95%CI 0.70 to 0.90, p = 0.0003, I² = 76%) but not during 28-day (n = 7) follow-up. The immediate effect was sustained at 12 months (n = 1). In children, CRP-POCT reduced antibiotic prescribing when CRP (cut-off) guidance was provided (n = 2). Meta-analyses showed significantly higher rates of re-consultation within 30 days (n = 8, 1 significant). Clinical recovery, resolution of symptoms, and hospital admissions were not significantly different between CRP-POCT and usual care. CRP-POCT can reduce immediate antibiotic prescribing for RTIs in primary care (number needed to (NNT) for benefit = 8) at the expense of increased re-consultations (NNT for harm = 27). The increase in re-consultations and longer-term effects of CRP-POCT need further evaluation. Overall, the benefits of CRP-POCT outweigh the potential harms (NNTnet = 11).

Keywords: antibiotic prescribing; antibiotic stewardship; antibiotic use; c-reactive protein; diagnostics; meta-analysis; point-of-care testing; primary care; respiratory tract infection; systematic review.

Figure 1

Study identification and process for selection of studies included in the review.

Figure 2

Comparison: CRP-POCT versus usual care. Overall antibiotic prescribing at index consultations. CRP, C-Reactive Protein Point Of Care Test; RCTs, Randomised Controlled Trials; M-H, Mantel-Haenszel; CI, Confidence Interval; df, degrees of freedom; I², heterogeneity between trials.

Figure 3

Comparison: CRP-POCT versus usual care. Antibiotic prescribing at index consultations by type of RTI: (a) Upper RTI; (b) Lower RTI. CRP, C-Reactive Protein Point Of Care Test; RCTs, Randomised Controlled Trials; M-H, Mantel-Haenszel; CI, Confidence Interval; df, degrees of freedom; I², heterogeneity between trials.

Figure 4

Comparison: CRP-POCT versus usual care. Antibiotic prescribing at index consultations by age group: (a) Adults; (b) Children. CRP, C-Reactive Protein Point Of Care Test; RCTs, Randomised Controlled Trials; M-H, Mantel-Haenszel; CI, Confidence Interval; df, degrees of freedom; I², heterogeneity between trials.

Figure 5

Comparison: CRP-POCT versus usual care. Antibiotic prescribing at index consultations by age group with CRP (cut-off) guidance to withhold antibiotics: (a) Adults; (b) Children. CRP values recommended to withhold antibiotics are presented within brackets. CRP, C-Reactive Protein (Point Of Care Test); RCTs, Randomised Controlled Trials; M-H, Mantel-Haenszel; CI, Confidence Interval; df, degrees of freedom; I², heterogeneity between trials.