The arthritic patients' perspective of measuring treatment efficacy: Patient Reported Experience Measures (PREMs) as...

We aimed to assess the validity and reliability of a Patient Reported Experience Measures (PREMs) questionnaire which can be used in standard clinical practice to measure self-defined important experiences of inflammatory arthritis patients.

The Patient Reported Experience Measures (PREMs) questionnaire was conceptualised based on frameworks used by the WHO Quality of Life tool, as well as the PRO measurement information system (PROMIS). Cognitive interviews were conducted with 94 inflammatory arthritis patients (diagnosed according to EULAR/ACR criteria 2010), with a range of severity and disease activity to identify item pool of questions. Item selection and reduction was achieved based on patients as well as an interdisciplinary group of physicians, nurses, health educators and occupational therapy (OTs) feedback, in addition to clinometric and psychometric methods. The latter included Rasch and internal consistency reliability analyses. The PREMs questionnaire was developed centered around 5 categories: 1. Journey to diagnosis, 2. Impact of the disease on the patients' everyday life, 3. knowledge about the disease, 4. the care in the hospital, and 5. patient education and aftercare (including what to do in case of exacerbation). After analysis for ordered response options, content analysis and semi structured group discussion to cover these 5 categories, 32 questions were identified as the final item set. The routine clinic was used as a setting for the questionnaire evaluation. 183 patients were asked to complete the PROMs as well as PREMs questionnaires whilst sitting in the waiting area before being examined by the treating physician. Reliability and comprehensibility was assessed using the Test-retest reliability (reproducibility).

The tool was derived from RA patients, therefore establishing its face validity. The PREMs questionnaire showed a high degree of comprehensibility (9.3). It demonstrated a relatively high-standardised alpha (0.892). The questionnaire items correlated significantly (p<0.01) with clinical parameters of disease activity, PROMs, self-helplessness and DAS-28 score supporting its construct validity. The domain of impact of arthritis correlated significantly (p<0.01) with health related quality of life (HRQOL) score as well as disease activity and damage measures, establishing its criterion validity. Patient education and aftercare correlated significantly (p<0.01) with adherence to therapy.

The studied PREMs questionnaire had fair psychometric properties as it was valid, reliable and comprehensible. The patients were able to comprehend varying response options on a categorical scale, and could accurately respond to items using a 7-day recall period. It provides informative measure for the patients' experience with their disease, and in the meantime, facilitates incorporating the patients' feedback into the patients' management algorithm.