Examining the construct and known-group validity of a composite endpoint for The Older Persons and Informal Caregivers Survey Minimum Data Set (TOPICS-MDS); A large-scale data sharing initiative

by Cynthia S. Hofman, Jennifer E. Lutomski, Han Boter, Bianca M. Buurman, Anton J. M. de Craen, Rogier Donders, Marcel G. M. Olde Rikkert, Peter Makai, René J. F. Melis, TOPICS-MDS research consortium

Background

Preference-weighted multi-faceted endpoints have the potential to facilitate comparative effectiveness research that incorporates patient preferences. The Older Persons and Informal Caregivers Survey—Composite endpoint (TOPICS-CEP) is potentially a valuable outcome measure for evaluating interventions in geriatric care as it combines multiple outcomes relevant to older persons in a single metric. The objective of this study was to validate TOPICS-CEP across different study settings (general population, primary care and hospital).

Methods

Data were extracted from TOPICS Minimum Dataset (MDS), a pooled public-access national database with information on older persons throughout the Netherlands. Data of 17,603 older persons were used. Meta-correlations were performed between TOPICS-CEP indexed scores, EuroQol5-D utility scores and Cantril's ladder life satisfaction scores. Mixed linear regression analyses were performed to compare TOPICS-CEP indexed scores between known groups, e.g. persons with versus without depression.

Results

In the complete sample and when stratified by study setting TOPICS-CEP and Cantril's ladder were moderately correlated, whereas TOPICS-CEP and EQ-5D were highly correlated. Higher mean TOPICS-CEP scores were found in persons who were: married, lived independently and had an education at university level. Moreover, higher mean TOPICS-CEP scores were found in persons without dementia, depression, and dizziness with falls, respectively. Similar results were found when stratified by subgroup.

Conclusion

This study supports that TOPICS-CEP is a robust measure which can potentially be used in broad settings to identify the effect of intervention or of prevention in elderly care.

http://ift.tt/2mOKs4S