Can combined biomarker testing shape the future of primary care antibiotic prescribing? Authors: Carl Llor, Niels Kristian Kjær Published: 2026 Antibiotic overuse is a major global public health concern, particularly in the management of respiratory tract infections (RTIs), which represent one of the most common reasons for consultations in primary care. Although the majority of RTIs, especially upper RTIs, are viral in origin and self-limiting, antibiotics continue to be prescribed at high rates. The problem is especially pronounced in China, where human antibiotic consumption accounts for approximately one quarter of worldwide antibiotic use.1 In Chinese primary care settings, antibiotics are prescribed in more than half of consultations for acute upper RTIs, and many patients receive multiple antibacterial agents, some of which are administered intravenously.2 However, antibiotics are not justified to reduce the risk of serious complications.3 Antibiotic prescribing rates are even higher for predominantly viral lower RTIs, with approximately 70% of acute bronchitis cases being treated with antibiotics in China.2 The widespread overprescribing of antibiotics is a major contributor to antimicrobial resistance, now recognised as one of the most serious global health threats. Antimicrobial resistance reduces treatment effectiveness and increases healthcare costs, morbidity, and mortality.4 Consequently, reducing unnecessary antibiotic use in RTIs has become a key objective of antimicrobial stewardship programmes, with increasing emphasis on improving diagnostic accuracy and supporting evidence-based prescribing. Combined acute biomarker point-of-care testing, such as the combination of C-reactive protein (CRP) and serum amyloid A (SAA), is emerging as a promising approach to improve antimicrobial stewardship. CRP is a well-established inflammatory biomarker that is commonly more elevated in bacterial than in viral infections and high in severe bacterial infections with a history of more than 6–8 h. The use of this rapid test in acute RTIs by general practitioners has been associated with a mean reduction in antibiotic prescribing of 23%, according to the latest Cochrane review.5 SAA is another acute-phase inflammatory protein produced predominantly by the liver in response to infection and inflammation; however, its concentration rises rapidly within a few hours of inflammatory stimulation, often earlier than CRP, showing its utility in patients with septicaemia.6 Viral infections may produce elevated SAA levels, whereas bacterial infections are typically associated with increases in both SAA and other inflammatory biomarkers such as CRP or procalcitonin. A cluster-randomised controlled trial evaluated a comprehensive point-of-care testing package combining CRP and SAA in 40 village clinics in rural China to reduce inappropriate antibiotic use for acute RTIs.7 Compared with usual care, the intervention significantly reduced antibiotic prescribing (50⋅7% vs 67⋅5%; adjusted relative risk 0⋅80) without compromising patient safety. These findings are encouraging and support the growing interest in biomarker-guided antimicrobial stewardship strategies in primary care. Rapid point-of-care testing may be particularly valuable in busy clinical settings, where treatment decisions must often be made quickly and with limited diagnostic resources. Combined with educational initiatives highlighting that most viral infections are self-limiting and do not require antibiotics, CRP testing can support clinicians in safely withholding unnecessary antibiotic treatment by providing immediate objective inflammatory information. Despite these promising findings, uptake of combined CRP and SAA point-of-care testing in this trial remained low (7⋅4%), highlighting the need to better understand barriers to implementation in routine practice. One possible explanation is that 97% of included cases were upper RTIs, where CRP and SAA testing is generally unnecessary when clinical assessment already strongly indicates a viral, self-limiting illness. The clinical usefulness of CRP and SAA testing has primarily been demonstrated in lower RTIs, particularly in distinguishing pneumonia from other self-limiting infections. An effective point-of-care test should be rapid, accurate, easy to use, affordable, minimally invasive, and reproducible for the conditions in which it is intended to be used. The combined rapid test evaluated in this trial meets some, but not all, of these criteria. Moreover, for widespread adoption in primary care, such tests must reliably distinguish between bacterial and viral infections and provide clear interpretation guidance to support clinical decision-making.8 Combining different biomarkers may increase specificity, although often at the expense of reduced sensitivity, when distinguishing viral from bacterial infections compared with the use of a single biomarker, as demonstrated in a recent study of childhood pneumonia.9 Educational initiatives on rational antibiotic use, including guidance on CRP interpretation, reduce antibiotic prescribing more effectively than providing CRP devices alone.10 Although CRP point-of-care test interpretation is relatively straightforward, the combined biomarker package evaluated in this study is more complex, which is summarised in Table S1 of the paper. Qualitative research involving clinicians and patients is needed to identify barriers to rapid biomarker test use in settings with high antibiotic expectations. Raising public awareness about antibiotic side effects may reduce demand for unnecessary prescriptions. Successful implementation also requires simpler tests, clearer interpretation guidance, and improved clinician education. Before widespread adoption, pragmatic real-world studies should confirm that these tests sustainably reduce antibiotic prescribing without increasing complexity in primary care. Contributors All authors conceptualised the commentary and reviewed and edited the manuscript. Declaration of interests None declared. Acknowledgements Role of the funding source: No funding. References