Harmonization of Later-Life Cognitive Function Across National Contexts: Results from the Harmonized Cognitive Assessment Protocols (HCAPs)
Gross, A. L.; Kobayashi, L. C.; Li, C.; Briceno, E. M.; Renteria, M. A.; Jones, R. N.; Langa, K. M.; Manly, J. J.; Nichols, E. L.; Weir, D.; Wong, R.; Berkman, L.; Lee, J.
Show abstract
BackgroundThe Harmonized Cognitive Assessment Protocol (HCAP) is an innovative instrument for cross-national comparisons of later-life cognitive function, yet its suitability across diverse populations is unknown. We aimed to harmonize general and domain-specific cognitive scores from HCAPs across six countries, and evaluate precision and criterion validity of the resulting harmonized scores. MethodsWe statistically harmonized general and domain-specific cognitive function across the six publicly available HCAP partner studies in the United States, England, India, Mexico, China, and South Africa (N=21,141). We used an item banking approach that leveraged common cognitive test items across studies and tests that were unique to studies, as identified by a multidisciplinary expert panel. We generated harmonized factor scores for general and domain- specific cognitive function using serially estimated graded-response item response theory (IRT) models. We evaluated precision of the factor scores using test information plots and criterion validity using age, gender, and educational attainment. FindingsIRT models of cognitive function in each country fit well. We compared measurement reliability of the harmonized general cognitive function factor across each cohort using test information plots; marginal reliability was high (r> 0{middle dot}90) for 93% of respondents across six countries. In each country, general cognitive function scores were lower with older ages and higher with greater levels of educational attainment. InterpretationWe statistically harmonized cognitive function measures across six large, population-based studies of cognitive aging in the US, England, India, Mexico, China, and South Africa. Precision of the estimated scores was excellent. This work provides a foundation for international networks of researchers to make stronger inferences and direct comparisons of cross-national associations of risk factors for cognitive outcomes. FundingNational Institute on Aging (R01 AG070953, R01 AG030153, R01 AG051125, U01 AG058499; U24 AG065182; R01AG051158)
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Self-reported health behaviors and longitudinal cognitive performance: Results from the Wisconsin Registry for Alzheimer’s Prevention 94%
- Prevalence of DSM-5 mild and major neurocognitive disorder in India: Results from the LASI-DAD 93%
- The Healthy Brain Initiative (HBI): A prospective cohort study protocol 93%
Similar papers in this journal
- The New Therapeutics in Alzheimer’s Disease Longitudinal Cohort study (NTAD): study protocol 93%
- The Longitudinal Aging Study Amsterdam COVID-19 exposure index: a cross-sectional analysis of the impact of the pandemic on daily functioning of older adults 92%
- Developing a core outcome set for interventions in people with mild cognitive impairment: study protocol 92%
Similar papers in this journal
- Psychological frailty in older adults: a systematic scoping review 90%
- Analyses of academician cohorts generate biased pandemic excess death estimates 89%
- Large-scale validation of the Prediction model Risk Of Bias ASsessment Tool (PROBAST) using a short form: high risk of bias models show poorer discrimination 89%
Similar papers in this journal
- Self-reported Memory Problems Eight Months after Non-Hospitalized COVID-19 in a Large Cohort 89%
- Disparities in COVID-19 Reported Incidence, Knowledge, and Behavior 88%
- Unhealthy lifestyle mediates the adverse effect of childhood traumas on acceleration of aging: analysis of 110,596 UK Biobank participants 88%
"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.