The Study of Women's Health Across the Nation (SWAN): The Impact of Midlife and the Menopause Transition on Health and Functioning in Early Old Age

PROJECT SUMMARY
Telephone-based cognition testing lends itself more easily to a broader range of populations than in-person
face-to-face testing, is cost-effective, and because it places a lower burden on study participants, reduces
attrition in longitudinal studies. However, scores obtained from phone-based testing are not directly
comparable to scores obtained from the more traditional face-to-face testing, and some studies have found
systematic differences (offsets) in test scores by mode of administration, with tests of memory and of attention
being particularly susceptible to mode effects. Comparability across modes of test administration is critical for
interpreting scores in research studies obtained using more than one mode of testing across the study sample,
as well as for interpreting longitudinal, within-person change in cognition, in both clinical and research settings,
when the mode of administration switches between visits. Furthermore, whether the comparability of phone-
based scores to scores from face-to-face testing varies across demographic groups and by level of cognition,
is an open question. Some studies suggest that mode effects vary by age and presence of cognitive
impairment. Answers to these questions are critical to longitudinal studies like the SWAN Aging study, which
have tracked cognitive performance serially. In the recently completed SWAN Aging 17th follow up visit,
cognition test batteries were administered both over the telephone and in-person at the study clinic. For
reasons relating to the COVID pandemic and the older ages of participants, nearly one in four V17 participants
chose to complete the cognition testing from home over the telephone. The primary objective of this
administrative supplement to the SWAN Aging U19 is to conduct a cross-calibration study in a new sample of
women, of similar age as V17 participants (age range, 66-76 years) to create test-score adjustment/correction
formulas that incorporate differences by level of cognitive functioning. Our specific aims are to, in a new
calibration sample of 100 women, ages 66-76 years, from across the education spectrum, to administer in
participants' primary languages (English, Spanish, Cantonese, and Japanese) the SWAN V17 phone and in-
person cognition batteries 2-4 weeks apart, in randomized order, and 1) determine between-mode differences
and correlations for individual tests of episodic memory, working memory, fluency, and attention; 2) determine
if race/ethnicity, language of testing, and cognition level modify mode differences; and 3) create calibration
equations to enhance the comparability of scores obtained via telephone testing and face-to-face testing.

Grant Number: 3U19AG063720-04S1
NIH Institute/Center: NIH
Principal Investigator: Maria Brooks