Can a computerized format replace a paper form in PRO and HRQL evaluation? Psychometric testing of the computer-assisted LCSS instrument (eLCSS-QL)

Purpose This trial assessed the ability to enhance healthrelated quality of life (HRQL) and patient-reported outcome (PRO) evaluation in trials and patient management using computer assistance with a handheld device, called a personal digital assistant. The study assessed ease of use and psychometric properties of this approach, comparing the Lung Cancer Symptom Scale (LCSS) paper form with the electronic (eLCSS-QL). Objectives were to: (1) measure completion times; (2) evaluate acceptability by patients, nurses, and physicians; (3) determine the correlation of the eLCSS-QL with the paper version; and (4) determine the feasibility of using a shorter visual analogue scale (VAS) in the electronic version. Patients and methods Patients were entered at 12 COMET clinics. All had: (a) stage III or IV non-small cell lung cancer, (b) Karnofsky performance status (KPS)≥60, (c) no prior chemotherapy, and (d) received initial courses of docetaxel + platinum. Of the148 patients enrolled, characteristics were: men, 57 %; median, KPS 80 %; and median age, 67 years. Of these, 131 patients completed the evaluation form. Results The eLCSS-QL had excellent acceptance by patients, nurses, and physicians. Patients required 2.2 min (mean) to complete the eLCSS-QL. Reliability coefficients using Cronbach’s alpha were high for the paper (0.84) and electronic (0.88) versions. The correlation coefficient between forms was high (0.92). The length of the VAS on the handheld pc (53 mm versus 100 mm on the paper format) resulted in nearly identical scores. Conclusions The high acceptance rate by patients and professionals, the rapid completion time, ease of use, and strong psychometric properties confirm that the electronic LCSS (eLCSS-QL) is practical for use in trials and patient management. This study indicates that computer assistance helps overcome barriers associated with evaluating HRQL and PROs.