Systematic review of clinical prediction models to support the diagnosis of asthma in primary care

Abstract

Diagnosing asthma is challenging. Misdiagnosis can lead to untreated symptoms, incorrect treatment and avoidable deaths. The best combination of clinical features and tests to achieve a diagnosis of asthma is unclear. As asthma is usually diagnosed in non-specialist settings, a clinical prediction model to aid the assessment of the probability of asthma in primary care may improve diagnostic accuracy. We aimed to identify and describe existing prediction models to support the diagnosis of asthma in children and adults in primary care. We searched Medline, Embase, CINAHL, TRIP and US National Guidelines Clearinghouse databases from 1 January 1990 to 23 November 17. We included prediction models designed for use in primary care or equivalent settings to aid the diagnostic decision-making of clinicians assessing patients with symptoms suggesting asthma. Two reviewers independently screened titles, abstracts and full texts for eligibility, extracted data and assessed risk of bias. From 13,798 records, 53 full-text articles were reviewed. We included seven modelling studies; all were at high risk of bias. Model performance varied, and the area under the receiving operating characteristic curve ranged from 0.61 to 0.82. Patient-reported wheeze, symptom variability and history of allergy or allergic rhinitis were associated with asthma. In conclusion, clinical prediction models may support the diagnosis of asthma in primary care, but existing models are at high risk of bias and thus unreliable for informing practice. Future studies should adhere to recognised standards, conduct model validation and include a broader range of clinical data to derive a prediction model of value for clinicians.

Protocol for the derivation and validation of a clinical prediction model to support the diagnosis of asthma in children and young people in primary care

“…Current clinical prediction models for the diagnosis of asthma are at high risk of bias and not recommended for use in clinical practice 12 . This protocol builds on the findings from our systematic review to derive and validate a clinical prediction model for primary healthcare professionals to support their decision making during the assessment of a child or young person with symptoms to suggest asthma.…”

“…Potential candidate predictors were identified based on the results from our systematic review of prediction models for the diagnosis of asthma in primary care 12 and based on clinical usefulness decided after discussion within the research team (including GPs, respiratory paediatricians and statisticians). We will choose the final list of candidate predictors from the following: gender, social class, wheeze, cough, night cough, breathlessness, eczema, hay fever, allergy to food or drink, allergy to substance other than food or drink, maternal asthma, maternal atopy, maternal cigarette smoking during pregnancy, childhood exposure to cigarette smoke, mould in the participants house, lung function indices from spirometry, fractional exhaled nitric oxide (FeNO), skin prick testing results, immunoglobulin E (IgE) serum samples, evidence of lung function or reversibility testing in the patient EHR and prescription of a short-acting beta agonist (SABA).…”

“…A clinical prediction model for asthma diagnosis in primary care could support diagnostic decision making by providing the probability that asthma is present using information from clinical symptoms, signs, medical and family history or tests. A recent systematic review identified seven clinical prediction models for asthma diagnosis in primary care, six models were derived in adults, and one in children 12 . Unfortunately each of the prediction models had methodological limitations and were considered unsuitable for clinical practice 12 .…”

“…A recent systematic review identified seven clinical prediction models for asthma diagnosis in primary care, six models were derived in adults, and one in children 12 . Unfortunately each of the prediction models had methodological limitations and were considered unsuitable for clinical practice 12 . Therefore, this protocol describes plans to derive and internally and externally validate a prediction model intended for use by a primary healthcare professional to aid their diagnostic decision making during the assessment of a child or young person with symptoms suggestive of asthma.…”

Defining high probability when making a diagnosis of asthma in primary care: mixed-methods consensus workshop

“…12 Having a numerically defined threshold could bring clarity over what guidelines describe as 'high probability' of asthma, 12 which would also facilitate greater consistency between clinicians regardless of their levels of experience. Second, in the context of clinical prediction models for asthma diagnosis, 16 it may be possible to input clinical assessment and investigation results into an algorithm and to generate a numerical probability that asthma is present. A consensus on the test-treatment threshold could help clinicians to make sense of the output of such algorithms.…”