Many HCAP studies include derived estimates of general cognitive functioning and specific domains of cognitive functioning (memory, executive functioning, language/fluency, orientation, visuospatial functioning), based on confirmatory factor analysis models.

Models use state-of-the-art advanced psychometric techniques to model the theoretical structure underlying cognition and account for measurement error and inter-correlations between items from the same task (e.g., immediate and delayed word recall).

Studies to assess the structure and measurement of cognition using these methods have been conducted for:

• HRS-HCAP
• ELSA-HCAP
• LASI-DAD

The harmonization of cognitive functioning across studies requires more detailed, careful examination of differences between studies in the administration and scoring of cognitive test items. Once harmonized items are identified, psychometric methods rooted in item response theory can enable statistical harmonization leveraging information from both harmonized and unharmonized items to obtain estimates of cognitive function that are both harmonized and which have maximal precision. General steps in the process are shown below.

• LASI-DAD
• Mex-Cog
• ELSA-HCAP
• HRS-HCAP

• Pre-statistical harmonization
• Statistical harmonization