Defining high probability of an asthma diagnosis in primary care: results from an expert consensus workshop

Background: Making a diagnosis of asthma can be challenging. Guidelines recommend identification of patients at high probability of asthma, but this is not defined numerically. Aim: We sought consensus on what constitutes high probability of asthma: defined as the probability at which there is enough information to add a diagnostic code of asthma and a subsequent negative test would not alter clinical opinion. Methods: A consensus workshop (modified Nominal Group Technique) was held with international health professionals with primary care knowledge/experience. After facilitated discussion, participants voted over three rounds, answering: At what numerical probability would you consider someone to be at high probability of asthma diagnosis? The result was summarised using mean/SD. The workshop discussion was recorded, transcribed and analysed qualitatively. Results: 10 participants (9 GPs, 1 asthma/allergy physician) from nine countries took part. Using final votes, the mean for a high probability of asthma was 75% (SD 7.6), indicating a trade-off between limiting false positives (more likely if a lower threshold was used) and pragmatism (on the grounds that first line preventive treatment with inhaled corticosteroids is relatively low risk). The need to review treatment response was strongly emphasised to check the diagnosis and detect non-responders. Conclusion: 75% was the probability at which participants felt confident to establish a diagnosis of asthma in primary care, albeit with the caveat that a review of treatment response was essential. Contextual factors such as availability of tests and the ease that patients could be reviewed influenced participants threshold choice.