Transdermal sensor features correlate with ecological momentary assessment drinking reports and predict alcohol-related consequences in young adults' natural settings

Examining early adherence measures as predictors of subsequent adherence in an intensive longitudinal study of individuals in mutual help groups: One day at a time

BACKGROUND

Individuals with a substance use disorder complete ecological momentary assessments (EMA) at lower rates than community samples. Previous research in tobacco users indicates that early log-in counts to smoking cessation websites predicted subsequent smoking cessation website usage. We extended this line of research to examine individuals who are seeking to change their drinking behaviors through mutual support groups. We examined whether adherence in the first 7 days (1487 observations) of an intensive longitudinal study design could predict subsequent EMA protocol adherence (50% and 80% adherence separately) at 30 (5700 observations) and 60 days (10,750 observations).

METHODS

Participants (n = 132) attending mutual-help groups for alcohol use completed two assessments per day for 6 months. We trained four classification models (logistic regression, recursive partitioning, support vector machines, and neural networks) using a training dataset (80% of the data) with each of the first 7 days' cumulative EMA assessment completion. We then tested these models to predict the remaining 20% of the data and evaluated model classification accuracy. We also used univariate receiver operating characteristic curves to examine the minimal combination of days and completion percentage to best predict subsequent adherence.

RESULTS

Different modeling techniques can be used with early assessment completion as predictors to accurately classify individuals that will meet minimal and optimal adherence rates later in the study. Models ranged in their performance from poor to outstanding classification, with no single model clearly outperforming other models.

CONCLUSIONS

Traditional and machine learning approaches can be used concurrently to examine several methods of predicting EMA adherence based on early assessment completion. Future studies could investigate the use of several algorithms in real time to help improve participant adherence rates by monitoring early adherence and using early assessment completion as features in predictive modeling.

Subjective intoxication predicts alcohol-related consequences at equivalent alcohol concentrations in young adults using ecological momentary assessment and alcohol sensors

OBJECTIVE

Subjective intoxication (SI) when drinking may serve as an internal barometer of whether to continue drinking or engage in potentially unsafe behavior. Mobile assessments offer the potential to use SI as a prospective risk indicator during drinking episodes; little evidence exists for the validity of real-time SI measures. We test the correspondence of SI with estimated blood alcohol concentration and transdermal alcohol concentration (TAC) in young adults' natural settings. We provide a novel test of whether SI features (peak and mean SI) uniquely predict consequences adjusting for alcohol concentration.

METHOD

Two hundred twenty-two heavy-drinking young adults (Mage = 22.3, 64% female, 79% non-Hispanic White, 84% undergraduates) participated in a 6-day study that used ecological momentary assessment of drinking and TAC sensors. SI was assessed every 30 min during drinking episodes. Multilevel modeling was used to test hypotheses.

RESULTS

Momentary SI and estimated blood alcohol concentration had moderate associations at the moment and day levels (standardized βs = 0.5-0.6); SI was moderately associated with TAC at the day level (βs = 0.5). Associations between SI and alcohol concentration varied widely between persons and across days. Day-level SI features predicted consequences when adjusting for alcohol concentration (incidence rate ratios, IRRs = 1.29-1.70).

CONCLUSIONS

Our two-item SI measure shows evidence of validity in real-world settings with heavy-drinking young adults. SI was significantly correlated with alcohol concentration and was a unique predictor of consequences. The strength of these associations varied greatly across persons and days. Real-time SI measurement may be useful in preventive interventions, but continued research is needed into when and for whom momentary SI is most predictive of risk. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Baseline protective behavioral strategy use predicts more moderate transdermal alcohol concentration dynamics and fewer negative consequences of drinking in young adults' natural settings

OBJECTIVE

Test whether frequent protective behavioral strategies (PBS) users report (a) fewer alcohol-related consequences and (b) less risky alcohol intoxication dynamics (measured via transdermal alcohol concentration [TAC] sensor "features") in daily life.

METHOD

Two hundred twenty-two frequently heavy-drinking young adults (Mage = 22.3 years) wore TAC sensors for 6 consecutive days. TAC features peak (maximum TAC), rise rate (speed of TAC increase), and area under the curve (AUC) were derived for each day. Negative alcohol-related consequences were measured in the morning after each self-reported drinking day. Past-year PBS use was measured at baseline.

RESULTS

Young adults reporting more frequent baseline PBS use showed (a) fewer alcohol-related consequences and (b) lower intoxication dynamics on average (less AUC, lower peaks, and slower rise rates). Limiting/stopping and manner of drinking PBS showed the same pattern of findings as the total score. Serious harm reduction PBS predicted fewer negative alcohol-related consequences, but not TAC features. Multilevel path models showed that TAC features peak and rise rate partially explained associations between PBS (total, limiting/stopping, and manner of drinking) and consequences. Independent contributions of PBS subscales were small and nonsignificant, suggesting that total PBS use was a more important predictor of risk/protection than the specific types of PBS used.

CONCLUSIONS

Young adults using more total PBS may experience fewer alcohol-related consequences during real-world drinking episodes in part through less risky intoxication dynamics (TAC features). Future research measuring PBS at the daily level is needed to formally test TAC features as day-level mechanisms of protection from acute alcohol-related consequences. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Transdermal alcohol concentration features predict alcohol-induced blackouts in college students

BACKGROUND

Alcohol-induced blackouts (AIBs) are common in college students. Individuals with AIBs also experience acute and chronic alcohol-related consequences. Research suggests that how students drink is an important predictor of AIBs. We used transdermal alcohol concentration (TAC) sensors to measure biomarkers of increasing alcohol intoxication (rise rate, peak, and rise duration) in a sample of college students. We hypothesized that the TAC biomarkers would be positively associated with AIBs.

METHODS

Students were eligible to participate if they were aged 18-22 years, in their second or third year of college, reported drinking 4+ drinks on a typical Friday or Saturday, experienced ≥1 AIB in the past semester, owned an iPhone, and were willing to wear a sensor for 3 days each weekend. Students (N = 79, 55.7% female, 86.1% White, Mage = 20.1) wore TAC sensors and completed daily diaries over four consecutive weekends (89.9% completion rate). AIBs were assessed using the Alcohol-Induced Blackout Measure-2. Logistic multilevel models were conducted to test for main effects.

RESULTS

Days with faster TAC rise rates (OR = 2.69, 95% CI: 1.56, 5.90), higher peak TACs (OR = 2.93, 95% CI: 1.64, 7.11), and longer rise TAC durations (OR = 4.16, 95% CI: 2.08, 10.62) were associated with greater odds of experiencing an AIB.

CONCLUSIONS

In a sample of "risky" drinking college students, three TAC drinking features identified as being related to rising intoxication independently predicted the risk for daily AIBs. Our findings suggest that considering how an individual drinks (assessed using TAC biomarkers), rather than quantity alone, is important for assessing risk and has implications for efforts to reduce risk. Not only is speed of intoxication important for predicting AIBs, but the height of the peak intoxication and the time spent reaching the peak are important predictors, each with different implications for prevention.

Signal processing and machine learning with transdermal alcohol concentration to predict natural environment alcohol consumption

Wrist-worn alcohol biosensors continuously and discreetly record transdermal alcohol concentration (TAC) and may allow alcohol researchers to monitor alcohol consumption in participants' natural environments. However, the field lacks established methods for signal processing and detecting alcohol events using these devices. We developed software that streamlines analysis of raw data (TAC, temperature, and motion) from a wrist-worn alcohol biosensor (BACtrack Skyn) through a signal processing and machine learning pipeline: biologically implausible skin surface temperature readings (< 28°C) were screened for potential device removal and TAC artifacts were corrected, features that describe TAC (e.g., rise duration) were calculated and used to train models (random forest and logistic regression) that predict self-reported alcohol consumption, and model performances were measured and summarized in autogenerated reports. The software was tested using 60 Skyn data sets recorded during 30 alcohol drinking episodes and 30 nonalcohol drinking episodes. Participants (N = 36; 13 with alcohol use disorder) wore the Skyn during one alcohol drinking episode and one nonalcohol drinking episode in their natural environment. In terms of distinguishing alcohol from nonalcohol drinking, correcting artifacts in the data resulted in 10% improvement in model accuracy relative to using raw data. Random forest and logistic regression models were both accurate, correctly predicting 97% (58/60; AUC-ROCs = 0.98, 0.96) of episodes. Area under TAC curve, rise duration of TAC curve, and peak TAC were the most important features for predictive accuracy. With promising model performance, this protocol will enhance the efficiency and reliability of TAC sensors for future alcohol monitoring research. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Integrating Alcohol Biosensors With Ecological Momentary Intervention (EMI) for Alcohol Use: a Synthesis of the Latest Literature and Directions for Future Research

Purpose of Review

Excessive alcohol use is a major public health concern. With increasing access to mobile technology, novel mHealth approaches for alcohol misuse, such as ecological momentary intervention (EMI), can be implemented widely to deliver treatment content in real time to diverse populations. This review summarizes the state of research in this area with an emphasis on the potential role of wearable alcohol biosensors in future EMI/just-in-time adaptive interventions (JITAI) for alcohol use.

Recent Findings

JITAI emerged as an intervention design to optimize the delivery of EMI for various health behaviors including substance use. Alcohol biosensors present an opportunity to augment JITAI/EMI for alcohol use with objective information on drinking behavior captured passively and continuously in participants' daily lives, but no prior published studies have incorporated wearable alcohol biosensors into JITAI for alcohol-related problems. Several methodological advances are needed to accomplish this goal and advance the field. Future research should focus on developing standardized data processing, analysis, and interpretation methods for wrist-worn biosensor data. Machine learning algorithms could be used to identify risk factors (e.g., stress, craving, physical locations) for high-risk drinking and develop decision rules for interpreting biosensor-derived transdermal alcohol concentration (TAC) data. Finally, advanced trial design such as micro-randomized trials (MRT) could facilitate the development of biosensor-augmented JITAI.

Summary

Wrist-worn alcohol biosensors are a promising potential addition to improve mHealth and JITAI for alcohol use. Additional research is needed to improve biosensor data analysis and interpretation, build new machine learning models to facilitate integration of alcohol biosensors into novel intervention strategies, and test and refine biosensor-augmented JITAI using advanced trial design.

College Students' Day-to-Day Maladaptive Drinking Responses to Stress Severity and Stressor-Related Guilt and Anger

BACKGROUND

Stress is a common part of college students' daily lives that may influence their physical activity (PA) and alcohol use. Understanding features of daily stress processes that predict health behaviors could help identify targets for just-in-time interventions.

PURPOSE

This study used intensive longitudinal data to examine whether prior day stress processes predict current day PA or alcohol use.

METHODS

Participants (N=58, Mage=20.5, 59% women, 70% White) were 18-to-25-year-old students who engaged in binge drinking at least twice monthly and used cannabis or tobacco in the past year. They wore activity (activPAL4) and alcohol (Secure Continuous Remote Alcohol Monitor) monitors for 11 days to assess daily PA (e.g., step counts) and alcohol use (e.g., drinking day), and completed daily surveys about yesterday's stress, including number of stressors (i.e., frequency), stressor intensity (i.e., severity), and frequency of affective states (e.g., guilt). Multilevel models examined prior day stress predicting current day PA or alcohol use.

RESULTS

Participants had higher odds of current day drinking (odds ratio=1.21) and greater area under the curve (B=0.08) when they experienced greater than usual stress severity the prior day. Participants had higher current day peak transdermal alcohol concentration (B=0.12) and area under the curve (B=0.11) when they more frequently experienced guilt due to stressors the prior day.

CONCLUSIONS

College students' unhealthy response of increasing alcohol use due to stress could adversely impact health outcomes. There is a critical need for interventions addressing students' ability to effectively manage and respond to the stress-inducing, daily demands of student life.

Accuracy of transdermal alcohol monitoring devices in a laboratory setting

The development of transdermal alcohol sensors (TASs) presents a new method to monitor alcohol consumption with the ability to objectively measure data 24/7. We aimed to evaluate the accuracy of two TASs (BACtrack Skyn and Smart Start BARE) in a laboratory setting. Thirty-two adults received a dose of ethanol 0.56 g/kg body weight as a 20% solution while wearing the two TASs and provided Breath Alcohol Concentration (BrAC) measurements for 3.5 h postalcohol consumption. Pearson's correlations and repeated measures analysis of variance tests were conducted on the peak, time-to-peak, and area under the curve data. Bland-Altman plots were derived. A time series analysis and cross-correlations were conducted to adjust for time lag. Both TASs were able to detect alcohol and increase within 20 min. BrAC peaked significantly quicker than Skyn and BARE. BrAC and Skyn peaks were negatively significantly correlated (r = -0.381, P = .035, n = 31), while Skyn and BARE peaks were positively significantly correlated (r = 0.380, P = .038, n = 30). Repeated measures analysis of variance found a significant difference between BrAC, Skyn, and BARE (F(1.946, 852.301) = 459.873, P < .001)). A time series analysis found when BrAC-Skyn and BrAC-BARE were adjusted for the delay to peak, and there was still a significant difference. Failure rates: 1.7% (Skyn) and 4.8% (BARE). Some evidence was obtained for TAS validity as both consistently detected alcohol. Failure rates and time lag show improvements in older device generations. However, neither TAS presented strong equivalence to the breathalyser even when the lag time was adjusted. With further testing and technology advancements, TAS could be a potential alcohol monitoring tool. Two of the newest TAS devices were worn in laboratory conditions for one afternoon to compare their accuracy of alcohol monitoring to a breathalyser. Findings suggest that the two TASs (BACtrack Skyn and SmartStart BARE) recorded significantly similar data postalcohol consumption, but not with the breathalyser.

Using alcohol biosensors and biomarkers to measure changes in drinking: Associations between transdermal alcohol concentration, phosphatidylethanol, and self-report in a contingency management study of persons with and without HIV

Alcohol use can be measured in many ways, including objectively through transdermal alcohol biosensors (e.g., transdermal alcohol concentration; TAC) or blood biomarkers (e.g., phosphatidylethanol; PEth), or subjectively through self-report (e.g., with the timeline followback; TLFB). However, it is unclear which measures best indicate changes in alcohol use within individuals following intervention, and if they have concurrent validity. In the context of contingency management (CM) with a goal of 30-day abstinence (n = 45, 60% male, 80% Black; Mage = 56.7; 58% with HIV), we examined relationships among changes in TAC-AUC (area under the curve, reflecting volume consumed), PEth, and self-reported number of drinks. The Secure Continuous Remote Alcohol Monitor Continuous Alcohol Monitoring (SCRAM-CAM) biosensor was used to collect TAC-AUC during a pre-CM period (∼7 days) and over a 30-day CM period. PEth was collected at baseline and 30-day follow-up. Number of drinks was self-reported through a 30-day TLFB at baseline and follow-up. Daily TAC-AUC and number of self-reported drinks were calculated for the pre-CM period and for the last 7 days of the CM period. Linear regression models controlling for baseline values revealed that change in TAC-AUC was significantly associated with change in PEth (β = 0.33, p < .0001) and with change in number of self-reported drinks (β = 0.34, p < .0001). Change in PEth was significantly associated with change in number of self-reported drinks (β = 0.85, p < .0001). We conclude that all three measures may be appropriate for measuring within-person change in alcohol use, while controlling for baseline values, in the context of a study testing an intervention such as CM. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Detection of Alcohol Intoxication Using Voice Features: A Controlled Laboratory Study

OBJECTIVE

Devices such as mobile phones and smart speakers could be useful to remotely identify voice alterations associated with alcohol intoxication that could be used to deliver just-in-time interventions, but data to support such approaches for the English language are lacking. In this controlled laboratory study, we compare how well English spectrographic voice features identify alcohol intoxication.

METHOD

A total of 18 participants (72% male, ages 21-62 years) read a randomly assigned tongue twister before drinking and each hour for up to 7 hours after drinking a weight-based dose of alcohol. Vocal segments were cleaned and split into 1-second windows. We built support vector machine models for detecting alcohol intoxication, defined as breath alcohol concentration > .08%, comparing the baseline voice spectrographic signature to each subsequent timepoint and examined accuracy with 95% confidence intervals (CIs).

RESULTS

Alcohol intoxication was predicted with an accuracy of 98% (95% CI [97.1, 98.6]); mean sensitivity = .98; specificity = .97; positive predictive value = .97; and negative predictive value = .98.

CONCLUSIONS

In this small, controlled laboratory study, voice spectrographic signatures collected from brief recorded English segments were useful in identifying alcohol intoxication. Larger studies using varied voice samples are needed to validate and expand models.

Acceptability and anklet user experience with the SCRAM-CAM transdermal alcohol concentration sensor in regularly drinking young adults' natural environments

Recent developments in wearable technologies have allowed for device-based capture of alcohol concentration among participants in their natural environments. Currently, the Continuous Alcohol Monitor from SCRAM systems (SCRAM-CAM) is the most extensively studied and validated transdermal alcohol concentration (TAC) sensor. However, there has been relatively little work focusing on its acceptability from the participants' perspective. In the current study, we assess the user experience of the SCRAM-CAM anklet in a sample of 222 regularly heavy drinking young adults (mean age = 22.3) who wore the anklet in their natural environments for five 24-h periods spanning 6 consecutive days. Differences in the anklet user experience were measured along a number of dimensions, and differences were tested by sex at birth, white/non-white racial/ethnic group membership, and alcohol use disorder (AUD) risk (measured through Alcohol Use Disorder Identification Test [AUDIT] scores). Males and females differed significantly on six of the eight acceptability items, with males showing more positive responses toward the anklet than females. No differences were found by white/non-white racial/ethnic groups nor AUD risk. Results suggest positive levels of acceptability toward the device overall while in natural environments, further validating the usage of the device in studies that measure alcohol consumption among different groups, including those with high levels of alcohol consumption. Researchers should take into consideration the different levels of burden or discomfort in male versus female participants when using the device.

Does it really matter that I do not remember my night? Consequences related to blacking out among college student drinkers

BACKGROUND

Alcohol-induced blackouts (AIBs) are experienced frequently by college student drinkers and are more likely to occur on days with high-intensity drinking (HID; 8+ for females/10+ for males) than non-HID days. Research suggests that AIBs are associated with experiencing other alcohol-related consequences (ARCs), including more serious ARCs (SARCs; e.g., legal and sexual consequences), but we do not know whether individuals experience more ARCs and more SARCs on occasions when they black out than when they do not black out. This study examines the associations between AIBs and the total number of both ARCs and SARCs.

METHODS

Students (N = 462, 51.7% female, 87.7% White, Mage  = 20.1) were assessed across 6 weekends via e-surveys (80%-97% response rate). Multilevel models were used to test for main effects, controlling for drinking (HID or estimated blood alcohol concentration; eBAC) and sex.

RESULTS

Drinking days when an AIB was experienced were associated with more total ARCs (b = 3.54, 95% CI: 3.10, 3.99) and more SARCs (b = 0.77, 95% CI: 0.60, 0.95) per day than non-AIB days. The more frequently a person experienced an AIB, the more total ARCs (b = 5.33, 95% CI: 4.40, 6.25) and SARCs (1.05, 95% CI: 0.80, 1.30) they reported on average.

CONCLUSION

Alcohol-induced blackout days were associated with higher levels of harm than non-AIB days, even at the same levels of drinking. Interventions that focus on reducing the occurrence of AIBs and factors that contribute to them, in addition to reducing alcohol consumption, may help reduce total harm associated with drinking among college students.

Daily Positive Alcohol Expectancies and Alcohol Use Outcomes among College Students

Expectancy theory supports alcohol expectancies, or the expected effects of drinking, as an important factor in alcohol use behaviors. Recent research supports that alcohol expectancies fluctuate daily, but scant research has examined specific types of expectancies and their associations with alcohol use at the daily level. Consequently, the present study examined (1) the daily association between select expectancies (i.e., sociability, tension reduction, liquid courage) and likelihood of drinking, and (2) whether daily expectancies predict alcohol use outcomes (i.e., alcohol-related problems, heavy drinking) on drinking days. Participants (N = 221; 79.2% women) were primarily White (48.4%) and Black (29.7%) university students. Participants completed daily measures of expectancies, alcohol use, and alcohol-related problems each day for 14 days. Results from multilevel models revealed daily sociability as the only expectancy subscale linked with increased odds of drinking. Daily expectancies did not predict heavy drinking or alcohol-related problems on drinking days. At the between-person level, average levels of sociability and tension reduction expectancies were associated with a greater likelihood of engaging in heavy drinking and experiencing a problem. Findings highlight that fluctuations in sociability expectancies may be proximally linked to likelihood of alcohol use, which may inform just-in-time intervention efforts targeting alcohol expectancies.

Alcohol-induced blackouts among college student drinkers: A multilevel analysis

OBJECTIVE

To identify factors (manner of drinking, combined alcohol and other substance use, physiology) that are associated with alcohol-induced blackouts (AIBs) over and above estimated blood alcohol concentration (eBAC).

METHODS

Students (N = 462, 51.7 % female, 87.7 % White, M_age = 20.1) were assessed across 6 weekends via e-surveys (80-97 % response rate). eBAC was calculated using standard number of drinks, drinking duration, sex, and weight. Three-level multilevel models (days, weeks, persons) were conducted to test for main effects, controlling for eBAC.

RESULTS

Protective behavioral strategies (PBS) were associated with decreased odds of AIBs on the daily (OR = 0.64, 95 % CI: 0.53, 0.77), weekly (OR = 0.84, 95 % CI: 0.72, 0.98), and person-levels (OR = 0.62, 95 % CI: 0.51, 0.74). Combined cannabis with alcohol was associated with increased odds of AIBs on the weekly (OR = 2.13, 95 % CI 1.13, 4.07) and person-levels (OR = 3.56, 95 % CI 1.60, 7.93). People who more frequently played drinking games (OR = 1.41, 95 % CI: 1.12, 1.77), pregamed (OR = 1.55, 95 % CI: 1.19, 2.03), and showed higher tolerance (OR = 1.22, 95 % CI: 1.08, 1.37) showed increased risk of AIBs, over and above eBAC levels.

CONCLUSION

We identified a number of daily-, weekly-, and person-level factors that uniquely contribute to the prediction of AIBs even at equivalent eBACs. Many of these factors were behavioral, suggesting that they may serve as malleable prevention targets for AIBs in college student drinkers.

Psychological distance intervention reminders reduce alcohol consumption frequency in daily life

Modifying behaviors, such as alcohol consumption, is difficult. Creating psychological distance between unhealthy triggers and one's present experience can encourage change. Using two multisite, randomized experiments, we examine whether theory-driven strategies to create psychological distance-mindfulness and perspective-taking-can change drinking behaviors among young adults without alcohol dependence via a 28-day smartphone intervention (Study 1, N = 108 participants, 5492 observations; Study 2, N = 218 participants, 9994 observations). Study 2 presents a close replication with a fully remote delivery during the COVID-19 pandemic. During weeks when they received twice-a-day intervention reminders, individuals in the distancing interventions reported drinking less frequently than on control weeks-directionally in Study 1, and significantly in Study 2. Intervention reminders reduced drinking frequency but did not impact amount. We find that smartphone-based mindfulness and perspective-taking interventions, aimed to create psychological distance, can change behavior. This approach requires repeated reminders, which can be delivered via smartphones.

Resting state connectivity in people living with HIV before and after stopping heavy drinking

Background

Heavy alcohol use in people living with HIV (PLWH) has widespread negative effects on neural functioning. It remains unclear whether experimentally-induced reduction in alcohol use could reverse these effects. We sought to determine the effects of 30-days drinking cessation/reduction on resting state functional connectivity in people with and without HIV.

Methods

Thirty-five participants (48.6% PLWH) demonstrating heavy alcohol use attempted to stop drinking for 30 days via contingency management (CM). MRI was acquired at baseline and after thirty days, and functional connectivity across five resting-state fMRI (rsfMRI) networks was calculated with the Conn toolbox for Matlab and examined in relation to transdermal alcohol concentration (TAC) recorded by the ankle-worn secure continuous remote alcohol monitor (SCRAM) and self-reported alcohol use (timeline follow-back; TLFB). Associations between alcohol use and reduction, HIV status, functional connectivity, and change in functional connectivity across five major rsfMRI networks were determined relative to the pre- and post-CM timepoints.

Results

Baseline resting-state functional connectivity was not significantly associated with average TAC-AUC during the pre-CM period, though higher self-reported alcohol use over the preceding 30 days was significantly associated with higher baseline connectivity within the Dorsal Attention Network (DAN; p-FDR < 0.05). Baseline connectivity within the Salience network was significantly negatively related to objective drinking reduction after intervention (DAN; p-FDR < 0.05), whereas baseline connectivity within the Limbic network was positively associated with self-reported drinking reduction (p-FDR < 0.05). Change in between-networks functional connectivity after intervention was significantly positively associated with biosensor-confirmed drinking reduction such that higher reduction was associated with stronger connectivity between the limbic and fronto-parietal control networks (p-FDR < 0.05). PLWH with lower DAN connectivity at baseline demonstrated poorer alcohol reduction than those with higher DAN connectivity at baseline.

Discussion

Lower resting-state functional connectivity of the Salience network significantly predicted stronger drinking reduction across all participants, suggesting a potential biomarker for reduced susceptibility to the environmental and social cues that often make alcohol use reduction attempts unsuccessful. Increased between-networks connectivity was observed in participants with higher alcohol reduction after CM, suggesting a positive benefit to brain connectivity associated with reduced drinking. PLWH with lower baseline DAN connectivity may not benefit as greatly from CM for alcohol reduction.

Acceptability and validity of using the BACtrack skyn wrist-worn transdermal alcohol concentration sensor to capture alcohol use across 28 days under naturalistic conditions - A pilot study

Wrist-worn transdermal alcohol concentration (TAC) sensors have the potential to provide detailed information about day-level features of alcohol use but have rarely been used in field-based research or in early adulthood (i.e., 26-40 years) alcohol users. This pilot study assessed the acceptability, user burden, and validity of using the BACtrack Skyn across 28 days in individuals' natural settings. Adults aged 26-37 (N = 11, Mage = 31.2, 55% female, 73% non-Hispanic white) participated in a study including retrospective surveys, a 28-day field protocol wearing Skyn and SCRAM sensors and completing ecological momentary assessments (EMA) of alcohol use and duration (daily morning reports and participant-initiated start/stop drinking EMAs), and follow-up interviews. Day-level features of alcohol use extracted from self-reports and/or sensors included drinks consumed, estimated Blood Alcohol Concentration (eBAC), drinking duration, peak TAC, area under the curve (AUC), rise rate, and fall rate. Repeated-measures correlations (rrm) tested within-person associations between day-level features of alcohol use from the Skyn versus self-report or the SCRAM. Participants preferred wearing the Skyn over the SCRAM [t (10) = -6.79, p < .001, d = 2.74]. Skyn data were available for 5614 (74.2%) out of 7566 h, with 20.7% of data lost due to syncing/charging issues and 5.1% lost due to device removal. Skyn agreement for detecting drinking days was 55.5% and 70.3% when compared to self-report and the SCRAM, respectively. Correlations for drinking intensity between self-report and the Skyn were 0.35 for peak TAC, 0.52 for AUC, and 0.30 for eBAC, which were smaller than correlations between self-report and SCRAM, at 0.78 for peak TAC, 0.79 for AUC, and 0.61 for eBAC. Correlations for drinking duration were larger when comparing self-report to the Skyn (rrm = 0.36) versus comparing self-report to the SCRAM (rrm = 0.31). The Skyn showed moderate-to-large, significant correlations with the SCRAM for peak TAC (rrm = 0.54), AUC (rrm = 0.80), and drinking duration (rrm = 0.63). Our findings support the acceptability and validity of using the Skyn for assessing alcohol use across an extended time frame (i.e., 28 days) in individuals' natural settings, and for providing useful information about day-level features of alcohol use.

Correspondence between alcohol use measured by a wrist-worn alcohol biosensor and self-report via ecological momentary assessment over a 2-week period

BACKGROUND

Transdermal alcohol biosensors measure alcohol use continuously, passively, and non-invasively. There is little field research on the Skyn biosensor, a new-generation, wrist-worn transdermal alcohol biosensor, and little evaluation of its sensitivity and specificity and the day-level correspondence between transdermal alcohol concentration (TAC) and number of self-reported drinks.

METHODS

Participants (N = 36; 61% male, M age  = 34.3) wore the Skyn biosensor and completed ecological momentary assessment (EMA) surveys about their alcohol use over 2 weeks. A total of 497 days of biosensor and EMA data were collected. Skyn-measured drinking episodes were defined by TAC > 5 μg/L. Skyn data were compared to self-reported drinking to calculate sensitivity and specificity (for drinking day vs. nondrinking day). Generalized estimating equations models were used to evaluate the correspondence between TAC features (peak TAC and TAC-area under the curve (AUC)) and number of drinks. Individual-level factors (sex, age, race/ethnicity, body mass index, human immunodeficiency virus status, and hazardous drinking) were examined to explore associations with TAC controlling for number of drinks.

RESULTS

Using a minimum TAC threshold of 5 μg/L plus coder review, the biosensor had sensitivity of 54.7% and specificity of 94.6% for distinguishing drinking from nondrinking days. Without coder review, the sensitivity was 78.1% and the specificity was 55.2%. Peak TAC (β = 0.92, p < 0.0001) and TAC-AUC (β = 1.60, p < 0.0001) were significantly associated with number of drinks. Females had significantly higher TAC levels than males for the same number of drinks.

CONCLUSIONS

Skyn-derived TAC can be used to measure alcohol use under naturalistic drinking conditions, additional research is needed to accurately identify drinking episodes based on Skyn TAC readings.

To Drink or Not to Drink: Is That the Question? Examining Correspondence and Predictive Validity of Morning Drinking Intentions for Young Adults' Drinking Behaviors and Consequences

The Theory of Planned Behavior suggests that intentions are significant and proximal determinants of behavior. The purpose of this study was to test the predictive validity of drinking intentions for subsequent same-day drinking behaviors and negative consequences. Regularly drinking young adults (N = 222, 21-29 years, 84% undergraduates) completed an ecological momentary assessment (EMA) protocol and wore an alcohol monitor for five consecutive 24-h periods spanning 6 days (Wednesday-Monday). Each morning, participants reported their drinking intentions for the day and their previous day's alcohol consumption and the number of negative drinking consequences. Multilevel models showed that, at the within-person level, on days when people reported intending to drink, to get drunk, or to drink more than usual, they had higher odds of drinking, consumed more drinks, and had higher peak transdermal alcohol concentrations later that day. However, drinking occurred on 28% of days without drinking intentions, suggesting intentions were an imperfect signal for future drinking behavior. Morning drinking intentions also predicted experiencing more negative consequences, even after controlling for alcohol consumption. On average, young adults' morning-reported drinking-related intentions predicted increased odds of same-day drinking behavior and alcohol-related consequences. However, drinking frequently occurred on days participants did not intend to drink, suggesting that focusing only on drinking intention days will result in many missed prevention opportunities. Together, these results suggest the need for additional research to increase the predictive value of drinking intention assessments and for prevention interventions aimed at helping individuals follow through on their intentions not to drink.

Examining new-generation transdermal alcohol biosensor performance across laboratory and field contexts

BACKGROUND

Wrist-worn transdermal alcohol sensors have the potential to change how alcohol consumption is measured. However, hardware and data analytic challenges associated with transdermal sensor data have kept these devices from widespread use. Given recent technological and analytic advances, this study provides an updated account of the performance of a new-generation wrist-worn transdermal sensor in both laboratory and field settings.

METHODS

This work leverages machine learning models to convert transdermal alcohol concentration data into estimates of Breath Alcohol Concentration (BrAC) in a large-scale laboratory sample (N = 256, study 1) and a pilot field sample (N = 27, study 2). Specifically, in both studies, the accuracy of the translation is evaluated by comparing BAC estimates yielded by BACtrack Skyn to real-time breathalyzer measurements collected in the laboratory and in the field.

RESULTS

The newest version of the Skyn device demonstrates a substantially lower error rate than older hand-assembled prototypes (0% to 7% vs. 29% to 53%, respectively). On average, real-time estimates of BrAC yielded by these transdermal sensors are within 0.007 of true BAC readings in the laboratory context and within 0.019 of true BrAC readings in the field. In both contexts, the distance between true and estimated BrAC was larger when only alcohol episodes were examined (laboratory = 0.017; field = 0.041). Finally, results of power-law-curve projections indicate that, given their accuracy, transdermal BrAC estimates in real-world contexts have the potential to improve markedly (>25%) with adequately sized datasets for model training.

CONCLUSION

Findings from this study indicate that the latest version of the transdermal wrist sensor holds promise for the accurate assessment of alcohol consumption in field contexts. A great deal of additional work is needed to provide a full picture of the utility of these devices, including research with large participant samples in field contexts.