Dataset Information

Development of Algorithmic Dementia Ascertainment for Racial/Ethnic Disparities Research in the US Health and Retirement Study.

ABSTRACT:

Background

Disparities research in dementia is limited by lack of large, diverse, and representative samples with systematic dementia ascertainment. Algorithmic diagnosis of dementia offers a cost-effective alternate approach. Prior work in the nationally representative Health and Retirement Study has demonstrated that existing algorithms are ill-suited for racial/ethnic disparities work given differences in sensitivity and specificity by race/ethnicity.

Methods

We implemented traditional and machine learning methods to identify an improved algorithm that: (1) had ?5 percentage point difference in sensitivity and specificity across racial/ethnic groups; (2) achieved ?80% overall accuracy across racial/ethnic groups; and (3) achieved ?75% sensitivity and ?90% specificity overall. Final recommendations were based on robustness, accuracy of estimated race/ethnicity-specific prevalence and prevalence ratios compared to those using in-person diagnoses, and ease of use.

Results

We identified six algorithms that met our prespecified criteria. Our three recommended algorithms achieved ?3 percentage point difference in sensitivity and ?5 percentage point difference in specificity across racial/ethnic groups, as well as 77%-83% sensitivity, 92%-94% specificity, and 90%-92% accuracy overall in analyses designed to emulate out-of-sample performance. Pairwise prevalence ratios between non-Hispanic whites, non-Hispanic blacks, and Hispanics estimated by application of these algorithms are within 1%-10% of prevalence ratios estimated based on in-person diagnoses.

Conclusions

We believe these algorithms will be of immense value to dementia researchers interested in racial/ethnic disparities. Our process can be replicated to allow minimally biasing algorithmic classification of dementia for other purposes.

SUBMITTER: Gianattasio KZ

PROVIDER: S-EPMC6888863 | biostudies-literature | 2020 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development of Algorithmic Dementia Ascertainment for Racial/Ethnic Disparities Research in the US Health and Retirement Study.

Gianattasio Kan Z KZ Ciarleglio Adam A Power Melinda C MC

Epidemiology (Cambridge, Mass.) 20200101 1

<h4>Background</h4>Disparities research in dementia is limited by lack of large, diverse, and representative samples with systematic dementia ascertainment. Algorithmic diagnosis of dementia offers a cost-effective alternate approach. Prior work in the nationally representative Health and Retirement Study has demonstrated that existing algorithms are ill-suited for racial/ethnic disparities work given differences in sensitivity and specificity by race/ethnicity.<h4>Methods</h4>We implemented tra ...[more]

PMID: 31567393

Similar Datasets

Project description:Purpose: Equal-access health care systems such as the Veterans Health Administration (VHA) reduce financial and nonfinancial barriers to care. It is unknown if such systems mitigate racial/ethnic mortality disparities, such as those well documented in the broader U.S. population. We examined racial/ethnic mortality disparities among VHA health care users, and compared racial/ethnic disparities in VHA and U.S. general populations. Methods: Linking VHA records for an October 2008 to September 2009 national VHA user cohort, and National Death Index records, we assessed all-cause, cancer, and cardiovascular-related mortality through December 2011. We calculated age-, sex-, and comorbidity-adjusted mortality hazard ratios. We computed sex-stratified, age-standardized mortality risk ratios for VHA and U.S. populations, then compared racial/ethnic disparities between the populations. Results: Among VHA users, American Indian/Alaskan Natives (AI/ANs) had higher adjusted all-cause mortality, whereas non-Hispanic Blacks had higher cause-specific mortality versus non-Hispanic Whites. Asians, Hispanics, and Native Hawaiian/Other Pacific Islanders had similar, or lower all-cause and cause-specific mortality versus non-Hispanic Whites. Mortality disparities were evident in non-Hispanic-Black men compared with non-Hispanic White men in both VHA and U.S. populations for all-cause, cardiovascular, and cancer (cause-specific) mortality, but disparities were smaller in VHA. VHA non-Hispanic Black women did not experience the all-cause and cause-specific mortality disparity present for U.S. non-Hispanic Black women. Disparities in all-cause and cancer mortality existed in VHA but not in U.S. population AI/AN men. Conclusion: Patterns in racial/ethnic disparities differed between VHA and U.S. populations, with fewer disparities within VHAs equal-access system. Equal-access health care may partially address racial/ethnic mortality disparities, but other nonhealth care factors should also be explored.

Dataset Information

Development of Algorithmic Dementia Ascertainment for Racial/Ethnic Disparities Research in the US Health and Retirement Study.

Background

Methods

Results

Conclusions

Publications

Development of Algorithmic Dementia Ascertainment for Racial/Ethnic Disparities Research in the US Health and Retirement Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets