Background: Bronchiolitis is a clinical diagnosis, with growing research supporting limited use of diagnostic tests and interventions. Hospitals have reduced use of bronchodilators and imaging, but the national average of respiratory viral testing (RVT) in these children continues to be high at 33%. Over the past three years, RCHSD’s rate has persisted at 31-35%, despite use of a clinical pathway with decision support. OBJECTIVE: Determine key factors affecting physician decisions to obtain RVT among children with bronchiolitis.
Methods: Pediatric hospitalists and nurse practitioners, emergency department physicians, and resident physicians at RCHSD were invited to participate in focus groups assessing factors affecting their decisions to order RVT, as well as risks and benefits of RVT in bronchiolitis. Six sessions (two for each group) were conducted by the authors (MH and KR), audio-recorded, and transcribed verbatim. A preliminary coding scheme was developed; two independent coders reviewed the transcripts, assigned key concepts, and reached consensus on any coding discrepancies.
Results: A total of 34 providers participated, with equal representation between hospitalists, emergency medicine physicians, and resident physicians. Providers frequently reported that they do not obtain RVT in otherwise healthy children with typical presentations of bronchiolitis. However, many felt RVT use was still pervasive in the hospital and reported that they would order RVT if requested by another physician (accepting team, consulting team, or senior physician). The most common reasons cited for not ordering RVT included: it would not change overall management; clinical symptoms are more likely to guide their diagnosis; and high cost. The most common reasons cited for ordering RVT included: other provider requested (“directed”) RVT; provider reassurance after getting test results; diagnosis in chronic patients, critically-ill patients, and those with atypical presentations; help in the diagnosis and management of influenza; aid in cessation of antibiotics in patients thought to have serious bacterial infection; and assumption that testing was required for admission, patient placement, or cohorting.
Conclusions:
Providers continue to obtain RVT for children with bronchiolitis despite recommendations against testing in routine cases and a robust clinical pathway with decision support. Our results suggest there are opportunities to enhance decision support to address provider uncertainty regarding institutional cohorting policies, as well as use this tool to identify directing versus ordering provider for RVT. Further training for less experienced providers in managing diagnostic uncertainty may also improve clinical decision-making skills. These findings may be applicable to settings outside RCHSD, as most large institutions have clinical care pathways but are unable to reduce RVT.