Development and validation of a set of patient reported outcome measures to assess effectiveness of asthma prophylaxis

Abstract

Background: In the local setting, asthma control is assessed by symptoms and signs elicited by clinicians because of the limited availability of spirometry. Hence, we intended to develop a patient reported outcome measure (PROM) with more holistic interpretation that could also serve as a tool to measure the asthma control in resource limited settings. Therefore, this study was carried out in Northern Sri Lanka to develop and validate the Asthma Control PROM (AC-PROM) Tamil to measure the effectiveness of asthma prophylaxis based on symptoms, exacerbation and limitation of activity which could also serve as an easy measure of asthma control to the provider. Methods: The AC-PROM Tamil was developed in 3 steps: item generation, item reduction and psychometric evaluation. Items were generated through thematic analysis from focus group discussions among patients with asthma. Items were converted to an interviewer administered questionnaire in Tamil in the format of 5-point Likert scale. Item reduction was done by two rounds of online Delphi surveys among 10 experts and an exploratory factor analysis among 200 patients with asthma. The face and content validity were assessed by a panel of experts during Delphi survey and patients during the pre-test of the tool. Criterion validity of the tool was assessed against the forced expiratory volume in one second of 187 patients with asthma. The cut-off value to assess the asthma control was determined by receiver operating characteristic curve. Reliability was verified by Cronbach’s alpha coefficient. Results: From thematic analysis of focus group discussions 10 items were generated. One item was removed during Delphi survey. Exploratory factor analysis indicated removal of another item with 8 items categorised into two factors. Cronbach’s alpha coefficient of factors 1 and 2 were 0.821 and 0.903 respectively, indicating good reliability. Observations made by experts and responses made by patients were incorporated to improve the clarity and relevance of the items. Criterion validity was demonstrated by significant correlation between the AC-PROM Tamil and forced expiratory volume in one second (r = 0.66, p = 0.001). The cut-off value of the AC-PROM Tamil to detect asthma control was 28.5 with 79% (95% CI 71.3–86.9) sensitivity and 71% (95% CI 61.9–79.6) specificity. The AC-PROM Tamil showed moderate accuracy (the area under the receiver operating characteristic curve = 0.796; 95% CI 0.73–0.86). Response rate of the AC-PROM Tamil was 100% and time taken to complete was 3–4 min. Conclusion: The AC-PROM Tamil is a simple, feasible and reasonably accurate tool to assesses the effectiveness of asthma prophylaxis, particularly in resource limited settings.