Grant: Travel Awards - Institute Community Support

Our main research objective was to test the effect of missing data and supplementary variables on the precision of longitudinal change in patient-reported outcomes (PROs) in clinical registries. Clinical registry is an electronic database that contains information about the health and healthcare use of patients defined by a particular disease, condition or treatment. PROs are information provided by patients on aspects of their health status that are relevant to their health-related quality of life, including physical and mental health functioning. The secondary objective was to test association of multiple chronic diseases on estimates of change in health-related quality of life over time, in the presence of missing data, amongst older people undergoing knee replacement surgery. Study data were from a population-based joint replacement registry for Manitoba, Canada. PRO information was collected using the 12-item Short Form Survey (SF-12) one month before surgery at a patient pre-assessment clinic and one-year after surgery via a mailed questionnaire. We used a statistical model to estimate the average change in PRO scores, after adjusting for confounding demographic and comorbid chronic condition covariates. Methods suitable to address missing data problems was adopted. The study cohort was comprised of 5631 patients that underwent joint replacement surgery, 58.7% were females and the average age of patients was 66.7 years. Our results revealed that chronic conditions such as back pain, diabetes and heart disease reduced the amount of improvement on physical health summary scores and depression had a negative impact on mental health summary scores in patients undergoing joint replacement. The choice of methods to address missing data had a substantial impact on the precision of these estimates. In anticipation of unavoidable missing data in clinical registries, collection of supplemental information is beneficial to increase precision in estimates of change.