We enjoyed the article1 on new technologies in general practice and are excited by their potential; however, it is vitally important that these are appropriately researched. Recently, NHS England and Boots introduced point-of-care throat swab tests into Boots pharmacies2 and following a small feasibility evaluation3 (designed and funded by Boots) they now plan to roll this out nationally.
Pharmacy staff identified patients with a sore throat who had a history of fever and/or the absence of cough, and a trained pharmacist examined the tonsils for exudate and palpated for tender cervical lymphadenopathy. Three hundred and sixty-seven patients were recruited; 40% were positive for 3 of 4 of the CENTOR clinical scoring system (these patients were offered a throat swab test).3 Patients were asked their hypothetical course of action had they not accessed the service, and data were available on 60% of patients. From this, the number of GP consultations prevented and a reduction in antibiotic prescribing were estimated. The authors did not present any statistical data.3
A study such as this is at high risk of selection bias and is likely to overestimate any health service benefit. It omits the vital step of a control group in which the new service was not available, to calculate clinical effectiveness, cost-effectiveness, and impact. For example, CENTOR was developed and validated in patients attending A&E4 and examined by clinicians. People self-presenting to a pharmacy are different from those seen in clinical settings; they are likely to be healthier, so fewer need antibiotics, limiting the potential for antibiotic reduction. Moreover, the skills of clinicians and pharmacists are likely to differ. It is therefore possible that this service may actually increase antibiotic usage.
As an NIHR Diagnostic Evidence Cooperative we are excited that NHS England is seeking innovative ways to improve patient experience and workload. However, we urge NHS England to consider the evidence (NICE does not recommend this test),5 possible harms (including asymptomatic streptococcal carriage in low-risk populations),6 and ethics (patients paid for this test and subsequent antibiotic treatment yet this obvious financial conflict of interest remains unaddressed).
For national-level changes, speed should not substitute science.
- © British Journal of General Practice 2017