Online Supplement
Running Head: HYPERTENSION AND COGNITION
Untreated Hypertension Decreases Heritability of Cognition in Late Middle Age
Terrie Vasilopoulos1; William S. Kremen2,3; Kathleen Kim2,3; Matthew S. Panizzon2; Phyllis K. Stein4; Hong Xian5; Michael D. Grant6; Michael J. Lyons6; Rosemary Toomey6; Lindon J. Eaves7; Carol E. Franz2,8; Kristen C. Jacobson1,8
1Department of Psychiatry & Behavioral Neuroscience, University of Chicago, Chicago, IL
2Department of Psychiatry, University of California, San Diego, La Jolla, CA
3VA San Diego Healthcare System, La Jolla, CA
4Cardiovascular Division, Washington University School of Medicine, St. Louis, MO
5Department of Psychiatry, Washington University School of Medicine, St. Louis, MO
6Department of Psychology, Boston University, Boston, MA
7Virginia Institute for Psychiatric & Behavioral Genetics, Virginia Commonwealth University, Richmond, VA
8Drs. Franz and Jacobson are joint senior authors on this manuscript
Author Correspondence: Terrie Vasilopoulos
5841 S Maryland Ave, MC 3077, rm 603
University of Chicago
Chicago, IL 60637
Phone: 773-834-4645; Fax: 773-834-4536
Email:
Supplemental Methods
Detailed Description of Cognitive Domains
Verbal Ability. This domain was indexed by the Vocabulary subtest from Wechsler Abbreviated Scale of Intelligence (WASI; (Wechsler 1997a).
Visual-Spatial Ability. Visual-spatial ability was measured by two tests: the Hidden Figures test (Thurstone 1944) and Card Rotation test (Ekstrom et al. 1976). The Hidden Figures test required the identification of a simple figure hidden within a complex figure and was measured by the percentage of correct responses out of a five total sections. In the Card Rotation test participants were shown a figure and then judged which one of five subsequent figures was a planar rotation of original. Cronbach’s alpha for visual-spatial ability was α = .64.
Episodic Memory. Episodic memory was indexed by four tests indexing verbal and visual-spatial memory tasks. Verbal memory was assessed by the Logical Memory subtest of the Wechsler Memory Scale-III (WMS-III; (Wechsler 1997b), where participants were presented with two stories that were later recalled with both immediate and delayed free recall conditions. In our administration, each story was read only one time. Visual-spatial memory was assessed by the WMS-III Visual Reproduction subtest, where participants were presented with five designs and then had to draw the designs later in both an immediate and delayed recall condition. Episodic memory was indexed by total number of correct responses in both the immediate and delayed recall tasks in each test as well as scores on the delayed tasks adjusted for their associated immediate recall score (Cronbach’s α = .73).
Short-Term Memory. Short-Term Memory was measured by WMS-III Digit Span (forward) and Spatial Span (forward) (Wechsler 1997b). Cronbach’s alpha for short term memory was α = .34.
Working Memory. Working memory was assessed by three tests: WMS-III Digit Span (backward), Spatial Span (backward) tests and Letter-Number Sequencing. The backward span scores were adjusted for performance on the associated forward condition. Letter-Number Sequencing required participants to put unordered numbers and letters in the correct order (Wechsler 1997b). The Letter-Number Sequencing score was adjusted for performance on the Digit Span forward condition. Working memory was calculated as the average score across all three tests (Cronbach’s α = .45).
Processing Speed. Processing speed was assessed by two tests (Trails 2 and Trails 3) from the Delis-Kaplan Executive Function System (D-KEFS) Trail Making Test (Delis et al. 2001) and the word reading condition of the Stroop Color-Word Test (Golden 1978; Stroop 1935). All conditions of the Trail Making Test were presented on double-sized sheets of paper. Trails 2 and 3 required participants to draw lines to connect numbers and letters, respectively, in sequential order. For the Stroop Word test, participants must correctly read color names, written in black ink, in a given time limit. Scores for all three tests represent time in seconds needed for completion; therefore, when these scores were averaged to create the total domain score (Cronbach’s α = .65), they were multiplied by -1 so that higher scores represented better performance.
Verbal Fluency. Verbal Fluency was measured by two Category Fluency conditions from the D-KEFS (Delis et al. 2001): animals and boys’ names. Cronbach’s alpha for verbal fluency was α = .61.
Executive Function. This domain specifically assessed executive function switching processes using three measures created from the D-KEFS Trail Making Test and the D-KEFS Category Switching test. Condition 4 of the D-KEFS Trail Making Test (Trails 4) is a number-letter switching condition analogous to the traditional Trails B subtest (Delis et al. 2001). For creation of the executive function domain, we included a Trails 4 score adjusted for Trails 2 and a Trails 4 score adjusted for Trails 3. After adjustment, both Trails 4 scores were multiplied by -1 so that higher score represented better performance. In the D-KEFS Verbal Fluency Category Switching task, participants had to alternate saying fruit words and furniture words. This score was adjusted for Category Fluency (animal category) scores. Cronbach’s alpha for executive function was α = .62.
Abstract Reasoning. Abstract Reasoning was indexed by the WASI Matrix Reasoning subtest (Wechsler, 1997a).
Supplemental Results
Correlations among domains and correlations between each domain and general cognitive ability.
Table S1 reports the correlations among general cognitive ability (as measured by AFQT at midlife) and the nine cognitive domains described above. Inter-domain correlations ranged from r = .14 to r = .47 (all p-values < .01). Correlations between individual cognitive domains with general cognitive ability were also statistically significant (p < .01) and ranged from r = .33 to r = .60.
Differences between Non-Hypertensive and Medicated Hypertensive Groups
Table S2 shows the heritability estimates (h2) for all cognitive variables for the non-hypertensive and medicated hypertensive groups, as well as the chi-square tests for significant difference in unstandardized genetic (A) and nonshared environmental (E) influences across group. For all cognitive variables, unstandardized parameters could be equated across the two groups without a significant change in fit.
Online References
Delis D, Kaplan E, Kramer J (2001) The Delis-Kaplan Executive Function System. Psychological Corporation, San Antonio, TX
Ekstrom R, French J, Harman H (1976) Manual for kit of factor-referenced cognitive tests. Educational Testing Service, Princetown, NJ
Golden C (1978) Stroop Color and Word Test: A Manual for Clinical and Experimental Uses. Skoelting, Chicago
Stroop J (1935) Studies of interference in serial verbal reactions. Journal of Experimental Psychology 643-662
Thurstone L (1944) A factorial study of perception. University of Chicago Press, Chicago
Wechsler D (1997a) Manual for the Wechsler Adult Intelligence Scale - Third Edition. Psychological Corporation, San Antonio, TX
Wechsler D (1997b) Manual for the Wechsler Memory Scale-Third Edition. Psychological Corporation, San Antonio, TX
1
Table S1. Correlations among cognitive domains and with AFQT
AFQT / VA / VSA / EM / SM / WM / PS / VF / EF / ARGeneral Cognitive Ability (AFQT) / 1.0 / - / - / - / - / - / - / - / - / -
Verbal Ability (VA) / 0.55 / 1.0 / - / - / - / - / - / - / - / -
Visual Spatial Ability (VSA) / 0.60 / 0.35 / 1.0 / - / - / - / - / - / - / -
Episodic Memory (EM) / 0.42 / 0.40 / 0.33 / 1.0 / - / - / - / - / - / -
Short-Term Memory (SM) / 0.35 / 0.35 / 0.35 / 0.26 / 1.0 / - / - / - / - / -
Working Memory (WM) / 0.39 / 0.28 / 0.36 / 0.30 / 0.14 / 1.0 / - / - / - / -
Processing Speed (PS) / 0.41 / 0.35 / 0.47 / 0.30 / 0.38 / 0.34 / 1.0 / - / - / -
Verbal Fluency (VF) / 0.33 / 0.43 / 0.35 / 0.32 / 0.34 / 0.24 / 0.42 / 1.0 / - / -
Executive Function (EF) / 0.36 / 0.27 / 0.39 / 0.27 / 0.31 / 0.35 / 0.29 / 0.30 / 1.0 / -
Abstract Reasoning (AR) / 0.59 / 0.43 / 0.46 / 0.36 / 0.32 / 0.36 / 0.38 / 0.32 / 0.32 / 1.0
Table S1 notes: All correlations were statistically significant at p < 0.01
1
Table S2. Heritability (h2) of cognitive domains for the non-hypertensive and medicated hypertensive groups.
HeritabilityNon-hypertensive / Medicated Hypertensive / χ2 Testa / p-value
Verbal Ability / 0.65 / 0.59 / 1.18 / 0.56
Visual-Spatial Ability / 0.73 / 0.76 / 0.23 / 0.89
Episodic Memory / 0.62 / 0.59 / 0.50 / 0.78
Abstract Reasoning / 0.62 / 0.58 / 2.10 / 0.35
Processing Speed / 0.52 / 0.66 / 2.99 / 0.22
Working Memory / 0.41 / 0.36 / 0.21 / 0.90
Executive Function / 0.44 / 0.44 / 2.85 / 0.15
Short-Term Memory / 0.57 / 0.60 / 0.18 / 0.92
Verbal Fluency / 0.51 / 0.56 / 1.43 / 0.49
General Cognitive Ability / 0.77 / 0.74 / 1.66 / 0.44
aChi-square tests based on comparison of unstandardized parameter estimates for genetic (A) and nonshared environmental (E) parameters across groups. Df = 2 for all chi-square tests.
1