Respiratory Testing and Antibiotic Prescribing
In this trial, adding rapid multiplex respiratory microbiological point-of-care testing to primary-care assessment did not improve the overall same-day antibiotic-prescribing outcome, while prespecified subgroup signals require cautious interpretation.
> Research explainer: This briefing examines verified primary research published 50 days before the briefing date. It is not a same-day research update and does not provide medical advice.
A randomized clinical trial examined whether rapid multiplex respiratory microbiological point-of-care testing could alter antibiotic prescribing for acute respiratory tract infections when antibiotic treatment was considered or might be considered in primary care. The intervention tested for 19 respiratory viral pathogens and 4 atypical bacteria from nasal and throat swabs in approximately 45 minutes. [pmid:42149561]
Evidence
The trial enrolled 552 patients aged 12 months or older at 16 general practices in Southwest England. Recruitment and study activity occurred between December 2022 and April 2024. Participants had clinician-diagnosed acute respiratory tract infection of 21 days or less and were randomized 1:1: 276 to testing and 276 to usual care. [pmid:42149561]
The primary outcome was same-day antibiotic prescribing. Antibiotics were prescribed to 124 participants, or 45%, in each group. The reported odds ratio was 1.00, with a 95% confidence interval of 0.71 to 1.41 and P>.99. Thus, the trial’s overall comparison did not identify a reduction in same-day prescribing from use of the test. [pmid:42149561]
The reported safety outcome was patient-reported symptom severity on days 2 through 4. The mean difference was 0.09, with a 95% confidence interval from -0.10 to 0.27 and P=.36. The investigators reported no difference in symptom severity between groups for this outcome. [pmid:42149561]
The protocol included prespecified subgroup analyses. Among participants in whom a virus was detected, the reported odds ratio for antibiotic prescribing was 0.35 (95% CI, 0.20-0.63; P for interaction<.001). For participants with chronic lung disease, the corresponding odds ratio was 0.55 (95% CI, 0.28-1.09; P for interaction=.046). The trial did not report evidence of differential reduction for children younger than 16 years or for cases where patients and clinicians disagreed on antibiotic necessity. [pmid:42149561]
Analysis — Overall prescribing outcome
The central result is straightforward: in the enrolled primary-care population, a rapid multiplex test did not change the overall proportion receiving an antibiotic on the day of consultation. Equal prescribing proportions in the two randomized groups matter more for the broad intervention question than a positive-looking result in a subset, because the primary outcome was designed to assess the intervention across the trial population. [pmid:42149561]
The subgroup results are informative but bounded. A lower prescribing estimate among participants with a detected virus is compatible with clinicians using a viral result differently from an undifferentiated respiratory presentation. The chronic-lung-disease interaction likewise suggests that the intervention’s association with prescribing may have varied across that prespecified characteristic. Neither finding overturns the null overall result or establishes that testing is broadly effective for every respiratory presentation. [pmid:42149561]
Detection should also not be treated as proof that a detected organism caused the illness. The study’s full-text excerpt notes that microbial detection and microbial etiology are not necessarily equivalent for samples from nonsterile sites such as the nose and throat. That interpretive issue is especially relevant when considering why a test result could influence prescribing without demonstrating a population-wide prescribing benefit. [pmid:42149561]
The symptom result adds a narrow safeguard within the measured follow-up window: testing was not associated with a reported worsening in symptom severity on days 2 to 4. It does not, by itself, settle all clinical, implementation, resource, or longer-term outcomes. The appropriate research conclusion is therefore limited to this trial’s measured outcomes and setting: rapid multiplex respiratory testing showed no overall same-day prescribing benefit, with subgroup observations that warrant separate confirmation and contextual evaluation. [pmid:42149561]
Limitations
This is one randomized trial in 16 general practices in Southwest England, so its findings do not establish effectiveness outside its enrolled population, health-care setting, or test workflow. [pmid:42149561]
Although primary-outcome data were available for all 552 participants, safety-outcome data were available for 216 intervention participants (78%) and 203 usual-care participants (74%). Missing symptom data limit how completely the days 2 to 4 comparison represents all randomized participants. [pmid:42149561]
The subgroup analyses were prespecified, but they remain subgroup findings within a single trial. Their estimates should not be generalized as definitive evidence of benefit for all people with viral detection, chronic lung disease, children, or other groups. [pmid:42149561]
This briefing is a research explainer, not medical advice. The supplied evidence does not support patient-specific conclusions, predictions, or claims about outcomes beyond the measured trial comparisons. [pmid:42149561]