The development and first validation of the Manchester Early Morning Symptoms Index (MEMSI) for patients with COPD

Aim Early morning symptoms (EMS) in people with COPD are associated with poor health, impaired activities and increased exacerbation risk. We describe the development and preliminary validation of the Manchester Early Morning Symptom Index (MEMSI) to quantify EMS in COPD. Methods Focus groups and cognitive debriefing with patients with COPD were used to develop the potential item list, followed by a cross-sectional study to finalise the items for inclusion. In addition to test-retest reliability, comparisons with the St George's Respiratory Questionnaire-C (SGRQ-C), modified Medical Research Council Dyspnoea Scale, Functional Assessment of Chronic Illness Therapy Fatigue Scale (FACIT-F) and Hospital Anxiety and Depression Scale (HADS) evaluated construct validity. Hierarchical methods informed item deletion and Rasch analysis was applied to assess scale unidimensionality. Results 23 items were identified from the focus groups and debriefings. The cross-sectional study involved 203 patients with COPD (mean age 64.7 SD 7.5 years, male 63%, Global Initiative for Chronic Obstructive Lung Disease (GOLD): 1:14% 2:41% 3:25% 4:7%). 13 items were removed during item reduction. MEMSI contains 10 items, demonstrates good overall fit to the Rasch model (chi(2) p=0.26) and item score distribution; excellent reliability (Person Separation Index: 0.91) and good test-retest repeatability (r=0.82). It correlates with the SGRQ-C (r=0.73), FACIT-F (r=-0.65) and HADS (r=0.53-0.54) indicating good construct validity. Conclusions MEMSI is a reliable and valid unidimensional measure of EMS for patients with COPD. It is simple to use and score supporting its suitability for research and clinical use. Work is underway to determine the minimal clinical important difference and cross-cultural validity.