Table 19iSummary of traditional meta-analysis results for HUI cognition from CCHS-CLSA and Buschke Free from CSHA, and NuAge Using Hedges’ g

Method of Statistical HarmonizationAdjustmentSummary Effect Estimate (95% CI)I2Q*p-value
Raw DataUnadjusted0.15 (−0.06, 0.36)0.9020.62<0.001
Unadjusted (Including Participants with Complete Data for Covariates)0.15 (−0.05, 0.35)0.8816.54<0.001
Study-Specific Statistically Significant Covariates0.11 (−0.08, 0.30)0.8816.56<0.001
Common Set of Covariates0.12 (−0.07, 0.31)0.8614.530.001
T-ScoreUnadjusted0.13 (−0.05, 0.31)0.8614.260.001
Unadjusted (Including Participants with Complete Data for Covariates)0.13 (−0.05, 0.32)0.8412.560.002
Common Set of Covariates0.12 (−0.08, 0.32)0.8714.880.001
C-ScoreUnadjusted0.15 (−0.06, 0.36)0.9020.61<0.001
Unadjusted (Including Participants with Complete Data for Covariates)0.15 (−0.05, 0.35)0.8816.54<0.001
Common Set of Covariates0.12 (−0.07, 0.31)0.8614.530.001

CCHS = Canadian Community Health Survey; CI = confidence interval; CLSA = Canadian Longitudinal Study on Aging; CSHA = Canadian Study of Health and Aging; NuAge = Quebec Longitudinal Study on Nutrition and Aging

*

All Q Statistics have 2 degrees of freedom.

T-Scores minimally adjusted for age, sex and education level.

From: Methods and Results: Implementing and Evaluating Three Methods of Statistical Harmonization Applied to Cognitive Measures (Objective 3)

Cover of Harmonization of Cognitive Measures in Individual Participant Data and Aggregate Data Meta-Analysis
Harmonization of Cognitive Measures in Individual Participant Data and Aggregate Data Meta-Analysis [Internet].
Griffith L, van den Heuvel E, Fortier I, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.