Psychometric properties of outcome measures in non-pharmacological interventions of persons with dementia in low-and middle-income countries: A systematic review

Abstract

Despite high burden of dementia in low-and middle-income countries (LMICs), only a small number of clinical trials of psychosocial interventions for persons with dementia (PwD) have been conducted in these settings. It is essential that such trials use appropriate outcome measures that are methodologically robust and culturally appropriate to evaluate the effectiveness of interventions. We carried out a systematic review to examine the evidence base and psychometric properties of measures employed in these studies in LMICs. A systematic search of published literature on randomised controlled trials (RCT) of psychosocial interventions for PwD in LMICs between 2008 and April 2020 was carried out. Measures employed in each of the eligible studies were identified and through a focused search, we further explored the evidence base and psychometric properties employing Terwee criteria. Data extraction and quality appraisal were conducted by two independent reviewers. The review identified 41 measures from 17 RCTS which fulfilled eligibility criteria and they examined effectiveness across the domains of cognition (n = 16), behaviour and psychological symptoms (n = 11) and quality of life (n = 8). Of these 41, we were able to access relevant literature only for 18 and they were subject to psychometric analysis. Psychometric properties of these 18 instruments were at best modest, with Terwee scores ranging from 3 (low) to 15 (moderate). A majority of the studies were from China (n = 5) and Brazil (n = 6). The evidence base for the routinely employed measures in RCTs of non-pharmacological interventions for PwD in LMICs is limited. The quality of adaptation and validation of these instruments is variable and studies are largely uninformative about their psychometric properties and cultural appropriateness to the study setting. There is an urgent need to develop scientifically robust instruments in LMIC settings that can be confidently employed to measure outcomes in trials of psychosocial interventions for PwD.