Veronica L Richards1, Yiyang Liu1, Jessica Orr2, Robert F Leeman3,4, Nancy P Barnett5, Kendall Bryant6, Robert L Cook1, Yan Wang1. 1. Department of Epidemiology, University of Florida, Gainesville, Florida, USA. 2. Division of Infectious Diseases, University of Miami, Miami, Florida, USA. 3. Department of Health Education and Behavior, University of Florida, Gainesville, Florida, USA. 4. Yale School of Medicine, New Haven, Connecticut, USA. 5. Brown University School of Public Health, Providence, Rhode Island, USA. 6. National Institute on Alcohol Abuse and Alcoholism, Bethesda, Maryland, USA.
Abstract
BACKGROUND: Transdermal alcohol biosensors can objectively monitor alcohol use by measuring transdermal alcohol concentration (TAC). However, it is unclear how sociodemographic and clinical factors that influence alcohol metabolism are associated with TAC. The main aim of this study was to examine how sociodemographic factors (sex, age, race/ethnicity) and clinical factors (body mass index, liver enzymes: alanine aminotransferase [ALT] and aspartate transaminase [AST]), alcohol use disorder, and HIV status were associated with TAC while controlling for level of alcohol use. METHODS: We analyzed data from a prospective study involving contingency management for alcohol cessation among persons living with and without human immunodeficiency virus (HIV) that used the Secure Continuous Remote Alcohol Monitoring (SCRAM) biosensor. Forty-three participants (Mage = 56.6 years; 63% male; 58% people living with HIV) yielded 183 SCRAM-detected drinking days. Two indices derived from SCRAM: peak TAC (reflecting level of intoxication) and TAC area under the curve (TAC-AUC; reflecting alcohol volume)-were the main outcomes. Self-reported alcohol use (drinks/drinking day) measured by Timeline Followback was the main predictor. To examine whether factors of interest were associated with TAC, we used individual generalized estimating equations (GEE), followed by a multivariate GEE model to include all significant predictors to examine their associations with TAC beyond the effect of self-reported alcohol use. RESULTS: Number of drinks per drinking day (B = 0.29, p < 0.01) and elevated AST (B = 0.50, p = 0.01) were significant predictors of peak TAC. Positive HIV status, female sex, elevated AST, and number of drinks per drinking day were positively associated with TAC-AUC at the bivariate level, whereas only self-reported alcohol use (B = 0.85, p < 0.0001) and female sex (B = 0.67, p < 0.05) were significant predictors of TAC-AUC at the multivariate level. CONCLUSIONS: HIV status was not independently associated with TAC. Future studies should consider the sex and liver function of the participant when using alcohol biosensors to measure alcohol use.
BACKGROUND: Transdermal alcohol biosensors can objectively monitor alcohol use by measuring transdermal alcohol concentration (TAC). However, it is unclear how sociodemographic and clinical factors that influence alcohol metabolism are associated with TAC. The main aim of this study was to examine how sociodemographic factors (sex, age, race/ethnicity) and clinical factors (body mass index, liver enzymes: alanine aminotransferase [ALT] and aspartate transaminase [AST]), alcohol use disorder, and HIV status were associated with TAC while controlling for level of alcohol use. METHODS: We analyzed data from a prospective study involving contingency management for alcohol cessation among persons living with and without human immunodeficiency virus (HIV) that used the Secure Continuous Remote Alcohol Monitoring (SCRAM) biosensor. Forty-three participants (Mage = 56.6 years; 63% male; 58% people living with HIV) yielded 183 SCRAM-detected drinking days. Two indices derived from SCRAM: peak TAC (reflecting level of intoxication) and TAC area under the curve (TAC-AUC; reflecting alcohol volume)-were the main outcomes. Self-reported alcohol use (drinks/drinking day) measured by Timeline Followback was the main predictor. To examine whether factors of interest were associated with TAC, we used individual generalized estimating equations (GEE), followed by a multivariate GEE model to include all significant predictors to examine their associations with TAC beyond the effect of self-reported alcohol use. RESULTS: Number of drinks per drinking day (B = 0.29, p < 0.01) and elevated AST (B = 0.50, p = 0.01) were significant predictors of peak TAC. Positive HIV status, female sex, elevated AST, and number of drinks per drinking day were positively associated with TAC-AUC at the bivariate level, whereas only self-reported alcohol use (B = 0.85, p < 0.0001) and female sex (B = 0.67, p < 0.05) were significant predictors of TAC-AUC at the multivariate level. CONCLUSIONS: HIV status was not independently associated with TAC. Future studies should consider the sex and liver function of the participant when using alcohol biosensors to measure alcohol use.
Authors: Michael Francis Fleming; Matthew J Smith; Erika Oslakovic; Michael R Lucey; Jenny X Vue; Patrice Al-Saden; Josh Levitsky Journal: Alcohol Clin Exp Res Date: 2017-03-20 Impact factor: 3.455
Authors: Scott H Stewart; David G Koch; Ira R Willner; Raymond F Anton; Adrian Reuben Journal: Alcohol Clin Exp Res Date: 2014-05-21 Impact factor: 3.455
Authors: Donald M Dougherty; Tara E Karns; Jillian Mullen; Yuanyuan Liang; Sarah L Lake; John D Roache; Nathalie Hill-Kapturczak Journal: Drug Alcohol Depend Date: 2014-12-31 Impact factor: 4.492
Authors: Nancy P Barnett; Mark A Celio; Jennifer W Tidey; James G Murphy; Suzanne M Colby; Robert M Swift Journal: Addiction Date: 2017-02-22 Impact factor: 6.526
Authors: Donald M Dougherty; Nathalie Hill-Kapturczak; Yuanyuan Liang; Tara E Karns; Sarah L Lake; Sharon E Cates; John D Roache Journal: Addict Disord Their Treat Date: 2015-09