Executive function associated with sexual risk in young South African women: Findings from the HPTN 068 cohort

Authors

Molly Rosenberg, Indiana University
Audrey Pettifor, University of North Carolina
Mihaela Duta, University of Oxford
Nele Demeyere, University of Oxford
Ryan Wagner, University of The Witwatersrand
Amanda Selin, University of North Carolina
Catherine L. Mac Phail, University of WollongongFollow
Oliver Laeyendecker, Johns Hopkins University
James Hughes, University of Washington
Alan Stein, University of Oxford
Stephen Tollman, University of The Witwatersrand
Kathleen Kahn, University of The Witwatersrand

RIS ID

125956

Publication Details

Rosenberg, M., Pettifor, A., Duta, M., Demeyere, N., Wagner, R. G., Selin, A., MacPhail, C., Laeyendecker, O., Hughes, J. P., Stein, A., Tollman, S. & Kahn, K. (2018). Executive function associated with sexual risk in young South African women: Findings from the HPTN 068 cohort. PLoS One, 13 (4), e0195217-1-e0195217-14.

Abstract

Purpose Heightened sexual risk in adolescence and young adulthood may be partially explained by deficits in executive functioning, the set of cognitive processes used to make reasoned decisions. However, the association between executive function and sexual risk is understudied among adolescent girls and young women, particularly in low- and middle-income countries. Methods In a cohort of 853 young women age 18±25 in rural Mpumalanga province, South Africa, we evaluated executive function with three non-verbal cognitive tests: I. a rule-finding test, II. a trail-making test, and III. a figure drawing test. Using log-binomial regression models, we estimated the association between lower executive function test scores and indicators of sexual risk (unprotected sex acts, concurrent partnerships, transactional sex, and recent HSV-2 infection). Results In general, young women with lower executive function scores reported higher frequencies of sexual risk outcomes, though associations tended to be small with wide confidence intervals. Testing in the lowest quintile of Test I was associated with more unprotected sex [aPR 95% CI): 1.4 (1.0, 1.8)]. Testing in the lowest quintile of Test II was associated with more concurrent relationships and transactional sex [aPR (95% CI): 1.6 (1.1, 2.5) and 1.7 (1.3, 2.4), respectively] , and testing in the lowest four quintiles of Test III was associated with more concurrent relationships [aPR (95% CI): 1.7 (1.0, 2.7)]. Conclusions These results demonstrate an association between low executive function and sexual risk in South African young women. Future work should seek to understand the nature of this association and whether there is promise in developing interventions to enhance executive function to reduce sexual risk.