Back

Development and Psychometric Evaluation of a Bilingual Instrument for Assessing Beliefs Affecting Health-Seeking Intentions in Cognitive Decline Among Latino Populations

Mora Pinzon, M. C.; Sayegh, P.; Fernandez de Cordova, S.; Brown, R.; Barrett, B.

2025-04-25 public and global health
10.1101/2025.04.24.25326381 medRxiv
Show abstract

BackgroundThe Latino community in the United States is disproportionately affected by Alzheimers disease and related dementias (ADRD). However, there are no instruments to assess health seeking behaviors in this population. This study describes the development and validation process of a new 35-item instrument, "Beliefs affecting health Seeking Intentions in Cognitive decline (BESIC)", available in both English and Spanish. MethodsThe psychometric analysis involved assessing the goodness of fit of three measurement models; congeneric model, tau-equivalency, and the parallel model. Multigroup Confirmatory Factor Analysis (MGCFA) was used to check if the same factors are being measured in the same way across different groups. ResultsThe partial tau-equivalent model provided the best fit for our data, suggesting that all items in the instrument measure the same underlying concept construct with different degrees of precision and error. The instrument demonstrated good reliability for all sub-domains in the total sample, as well as for the two language surveys. While average scores in the two language groups were somewhat different, MGCFA demonstrated that the instrument works well and similarly in both English and Spanish. ConclusionThese psychometric validation findings suggest that BESIC is a useful tool for measuring beliefs affecting health-seeking intentions during cognitive decline in both English and Spanish-speaking populations in the United States. However, while the overall structure of the instrument was equivalent across languages, the strength of these relationships and the average scores on the items were not. This suggests that the way individuals from different language groups respond to the items may vary, requiring further investigation. This will help refine the instrument further and ensure its accuracy and usefulness in future research and practice.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.