Development and validation of a predictive model combining patient-reported outcome measures, spirometry and exhaled nitric oxide fraction for asthma diagnosis

Abstract

Introduction: Although asthma is a common disease, its diagnosis remains a challenge in clinical practice with both over- and underdiagnosis. Here, we performed a prospective observational study investigating the value of symptom intensity scales alone or combined with spirometry and exhaled nitric oxide fraction (F _ENO) to aid in asthma diagnosis.

Methods: Over a 38-month period we recruited 303 untreated patients complaining of symptoms suggestive of asthma (wheezing, dyspnoea, cough, sputum production and chest tightness). The whole cohort was split into a training cohort (n=166) for patients recruited during odd months and a validation cohort (n=137) for patients recruited during even months. Asthma was diagnosed either by a positive reversibility test (≥12% and ≥200 mL in forced expiratory volume in 1 s (FEV₁)) and/or a positive bronchial challenge test (provocative concentration of methacholine causing a 20% fall in FEV₁ ≤8 mg·mL^-1). In order to assess the diagnostic performance of symptoms, spirometric indices and F _ENO, we performed receiver operating characteristic curve analysis and multivariable logistic regression to identify the independent factors associated with asthma in the training cohort. Then, the derived predictive models were applied to the validation cohort.

Results: 63% of patients in the derivation cohort and 58% of patients in the validation cohort were diagnosed as being asthmatic. After logistic regression, wheezing was the only symptom to be significantly associated with asthma. Similarly, FEV₁ (% pred), FEV₁/forced vital capacity (%) and F _ENO were significantly associated with asthma. A predictive model combining these four parameters yielded an area under the curve of 0.76 (95% CI 0.66-0.84) in the training cohort and 0.73 (95% CI 0.65-0.82) when applied to the validation cohort.

Conclusion: Combining a wheezing intensity scale with spirometry and F _ENO may help in improving asthma diagnosis accuracy in clinical practice.

FIGURE 1

Asthma symptom intensity scales between asthmatic (A: n=105) and nonasthmatic (NA: n=61) subjects in the training cohort (n=166).

FIGURE 2

Receiver operating characteristic curves showing the performance of the predictive models (Models 1–8 in table 5) in the a) training and b) validation cohorts. AUC: area under the curve.