Improving accuracy in the estimation of probable dementia in racially and ethnically diverse groups with penalized regression and transfer learning

Algorithmic estimations of dementia status are widely used in public health and epidemiological research, however, inadequate algorithm performance across racial/ethnic groups has been a barrier. We present improvements in the accuracy of group-specific "probable dementia" estimation using a transfer learning approach. Transfer learning involves combining models trained on a large "source" dataset with imprecise outcome assessments, alongside models trained on a smaller "target" dataset with high-quality outcome assessments. Transfer learning improves model accuracy by leveraging large source data while refining estimations with smaller, target data. We illustrate with data from the Health and Retirement Study (source data: N=6,630) and the Harmonized Cognitive Assessment Protocol (target data: N=2,388). Models for dementia status estimation were evaluated through overall accuracy (Brier score), calibration (intercept, slope), and discriminative ability (area under the receiver operating characteristic curve, AUR; area under the precision-recall curve, AUPRC). The transfer-learned algorithm showed higher accuracy compared to the best previously reported algorithm among both non-Hispanic Black participants (Brier 0.049 vs. 0.061; AUC 0.84 vs. 0.81; AUPRC 0.52 vs. 0.39) and Hispanic participants (Brier 0.052 vs. 0.056; AUC 0.89 vs. 0.87; AUPRC 0.61 vs. 0.56). Transfer learning can improve dementia status estimation for groups historically underrepresented in research.