Alcohol‐free and low‐alcohol beverage consumption and concurrent and subsequent alcohol outcomes among young adults

Abstract

Background and aims

Alcohol‐Free and Low‐Alcohol Beverages (AFLABs) have grown in popularity, yet predictors and longitudinal outcomes of their use remain unclear. We aimed to measure the association between AFLAB consumption and (1) concurrent and (2) subsequent alcohol consumption, risk and change measures (e.g. consequences, quit attempts, abstention) and drinking motives among young adults.

Design

Observational study using two survey waves (W15 and W16; 2022–2024) from a longitudinal cohort study (STRATA).

Setting

California, USA (surveys fielded 2022–2024).

Participants

Cross sectional analysis of young adults (n = 877 individuals, of which 849 were followed in W16) who report past‐month alcohol consumption in W15 from a California‐based cohort study, and analysis of those with past‐month alcohol consumption at both W15 and W16 (n = 693).

Measurements

Individuals completed survey items on past‐month AFLAB consumption, alcohol consumption (past‐month heavy episodic drinking and past‐year abstention), alcohol risk and change measures (alcohol‐related consequences, indicators of a change effort and the alcohol contemplation ladder) and drinking motives (from the drinking motives questionnaire).

Findings

Twenty‐four percent of participants reported past‐month AFLAB use at baseline. Adjusted cross‐sectional models using W15 data indicated that AFLAB use was associated with lower social drinking motives [coefficient = −0.005; 95% confidence interval (CI) = −0.011 to 0.0001] but was not statistically significantly associated with other alcohol consumption, risk or change measures. In adjusted prospective models, baseline AFLAB use was associated with a 5.1 percentage point (pp) higher probability of alcohol abstention at W16 (95% CI = 1.1–9.1) than lack of AFLAB use. In adjusted prospective models, baseline AFLAB use was not statistically significantly associated with heavy episodic drinking, alcohol‐related consequences or most change‐related measures, although there was suggestive evidence of a higher likelihood of reporting a quit attempt in the past 3 months.

Conclusions

Use of alcohol‐free and low‐alcohol beverages appears to be common among young adults in California, USA, and appears to be concurrently associated with lower social drinking motives and prospectively associated with subsequent alcohol abstention.

INTRODUCTION

Young adulthood is a developmental stage marked by identity exploration and changing social roles. Young adults experience higher rates of heavy episodic drinking and elevated risk for alcohol‐related harms, but have low treatment uptake for alcohol problems [1, 2]. Evidence suggests drinking norms among young adults are shifting, with rising abstinence and a growing normalization of non‐drinking [3, 4].

Against this backdrop, alcohol‐free [i.e. beverages that contain no or a minute amount of alcohol by volume (ABV); typically 0% ABV to 0.5% ABV] and low‐alcohol beverages (typically 0.5% to 1.2% ABV) (AFLABs) have expanded rapidly in availability. Studies document a rising uptake of AFLABs and highlight individual, product and environmental drivers of use, including health motivations, taste expectations and social context [5, 6]. Studies also suggest that some people use AFLABs when attempting to cut down or to experiment with non‐drinking within ‘sober curious’ practices [6, 7, 8]. However, there is mixed evidence on whether AFLAB use substitutes for, complements or leaves unchanged standard‐strength alcohol consumption.

Despite the growing popularity of AFLABs, critical evidence gaps remain. Most studies are cross‐sectional or focus on general adult samples, leaving limited evidence on AFLAB use among contemporary United States (US) young adults, a cohort with distinct drinking patterns and shifting alcohol‐related norms. Longitudinal evidence is also limited on whether AFLAB use predicts subsequent alcohol consumption, alcohol‐related consequences, intentions or attempts to reduce drinking, abstention and drinking motives after accounting for prior levels of these outcomes. To address these gaps, the present study uses multi‐wave data from a cohort of US young adults to answer two research questions: (1) is AFLAB use associated with concurrent alcohol use, consequences, motives and change‐related intentions or behaviors? (2) Does AFLAB use prospectively predict changes in alcohol outcomes and drinking motives across adjacent waves?

METHODS

Study design

Data come from a longitudinal cohort of young adults originally recruited into a middle‐school prevention trial in 2008 that took place in 16 Southern California schools. They were followed‐up annually from 2008 to 2024. Intervention effects dissipated after the first year [9]. Cohort procedures, retention and measures have been described in detail elsewhere [10]. Surveys were web‐based with monetary compensation. All participants provided informed consent, and study procedures were approved by the Human Subjects Protection Committee at RAND.

Analyses used waves 15 [W15 (the first year that questions related to AFLAB consumption were included; data collected 2022–2023)] and 16 [W16 (data collected 2023–2024)] to (1) identify correlates of AFLAB use; and (2) estimate prospective associations between AFLAB use at W15 and alcohol related outcomes at W16.

Measures

AFLAB was measured as past‐month use of alcohol‐free or low‐alcohol beverages; defined for participants as ‘At least one drink of no‐alcohol or low‐alcohol (<1.2% alcohol by volume) beer, wine or liquor (like Kaliber or Heineken 0.0%, Surely Wines, Ritual Zero Proof)’ (binary).

Demographics included age, race and ethnicity (non‐Hispanic White, non‐Hispanic Black, Hispanic, non‐Hispanic Asian and non‐Hispanic other), sex (male, female), sexual identity (heterosexual, gay/lesbian, bisexual and other), educational attainment (a high school diploma or less, GED or some college, associate's degree, bachelor's degree or certificate; graduate degree).and pregnancy status (pregnant, not pregnant).

Alcohol consumption measures included past month heavy episodic drinking (binary; 5+ drinks on one occasion) and alcohol abstention (past year reporting 0 standard strength alcohol drinks).

Alcohol risk and change measures included nine alcohol‐related consequences (number of times a participant experienced negative consequences in the past year, e.g. a fight or argument because of drinking alcohol) [11], indicators of a change effort (a binary measure, which took the value 1 if someone had an alcohol quit attempt in the past 3 months) and binary self‐reported measures of reduced drinking in the past month, plans to quit and thinking about quitting, derived from the alcohol contemplation ladder [12].

Drinking motives were assessed with social (α = 0.93) and coping (α = 0.91) subscales from the Drinking Motives Questionnaire (continuous measure) [13].

Analytic samples

We restricted the sample to past‐month drinkers because several outcomes were administered only to respondents who reported alcohol use. We also excluded 50 individuals with missing pregnancy data and four with missing motive data. Our resulting analytic cross‐sectional sample consists of 877 individuals in W15. We used two analytic samples in the prospective models. First, in the models using abstention as an outcome, we included 849 individuals who responded at both waves. Second, in models using other outcomes, which were contingent on drinking at W16, we included 693 individuals who reported past‐month alcohol consumption at W16.

Statistical analysis

We used descriptive statistics to compare baseline W15 characteristics by W15 past‐month AFLAB use. These comparisons were conducted in the baseline W15 sample, the prospective abstention sample and the prospective non‐abstention outcome sample. We, then, estimated cross‐sectional associations between AFLAB use and W15 outcomes. We fit separate models for each outcome. Both unadjusted and demographic adjusted models were estimated. Finally, we estimated prospective associations between W15 AFLAB use and W16 outcomes. For each W16 outcome, we estimated sequential models: (1) unadjusted models including W15 AFLAB use; (2) demographic‐adjusted models including W15 AFLAB use; (3) demographic‐adjusted models excluding W15 AFLAB use; (4) fully adjusted models including demographics, W15 AFLAB use and the baseline value of the outcome; and (5) an additional abstention model adjusting for demographics and baseline change/motivation measures. Because the abstention analysis was restricted to participants who reported past‐month drinking at W15, models for abstention could not adjust for baseline abstention. For binary outcomes, we estimated logistic regression models and report average marginal effects. For continuous outcomes, we estimated linear regression models and report unstandardized coefficients. The analysis was not pre‐registered and results should, therefore, be considered exploratory. Analyses were conducted in Stata/MP 18.

RESULTS

Table 1 compares baseline W15 characteristics by past‐month AFLAB use. Non‐Hispanic White individuals were less likely to report AFLAB use, (15.9% versus 24.1%), as were gay or lesbian individuals (2.3% versus 5.4%), whereas women were more likely (65.4% versus 56.4%). Similar patterns were observed in the prospective samples.

TABLE 1.

Sample demographics and alcohol consumption, risk and change measures and motives, by AFLAB consumption.

	Baseline W15 sample			Prospective sample for abstention analysis			Prospective (non‐abstention outcome) sample
	AFLAB consumption			AFLAB consumption			AFLAB consumption
	No	Yes	P‐value	No	Yes	P‐value	No	Yes	P‐value
Age	25.7	25.7	0.766	25.7	25.7	0.877	25.7	25.7	0.899
Race/ethnicity
Non‐Hispanic White	24.1%	15.9%	0.006	23.7%	15.9%	0.012	25.3%	15.5%	0.005
Non‐Hispanic Black	2.0%	2.3%	0.748	2.0%	2.4%	0.746	1.9%	1.9%	0.951
Hispanic	40.0%	44.4%	0.258	39.7%	44.0%	0.285	38.7%	43.2%	0.313
Non‐Hispanic Asian	22.3%	25.7%	0.322	22.9%	26.1%	0.360	21.4%	27.7%	0.114
Non‐Hispanic other	11.6%	11.7%	0.978	11.7%	11.6%	0.973	12.8%	11.6%	0.682
Sex
Female	56.4%	65.4%	0.018	56.4%	66.7%	0.008	57.2%	68.4%	0.010
Male	43.6%	34.6%	0.018	43.6%	33.3%	0.008	42.8%	31.6%	0.010
Educational attainment
HS/GED or less	19.3%	15.9%	0.245	19.6%	16.4%	0.290	19.3%	15.5%	0.256
College degree	70.0%	72.0%	0.578	69.9%	72.0%	0.572	69.5%	71.6%	0.613
Graduate degree	10.7%	12.1%	0.571	10.4%	11.6%	0.648	11.2%	12.9%	0.563
Sexual orientation
Heterosexual	79.2%	81.8%	0.401	79.3%	82.6%	0.282	78.1%	85.8%	0.021
Gay/lesbian	5.4%	2.3%	0.023	5.3%	1.4%	0.002	5.0%	0.6%	0.000
Bisexual	13.3%	14.0%	0.784	13.6%	14.0%	0.869	15.4%	11.6%	0.207
Other	2.1%	1.9%	0.823	1.9%	1.9%	0.954	1.5%	1.9%	0.715
Pregnant
No	99.5%	98.6%	0.263	99.5%	98.6%	0.263	99.4%	98.7%	0.448
Yes	0.5%	1.4%	0.263	0.5%	1.4%	0.263	0.6%	1.3%	0.448
Heavy episodic drinking, past month
No	45.6%	51.4%	0.138	46.0%	51.2%	0.190	43.9%	47.1%	0.479
Yes	54.4%	48.6%	0.138	54.0%	48.8%	0.190	56.1%	52.9%	0.479
Reported quit attempt, past 3 months
No	78.1%	77.1%	0.756	77.6%	77.3%	0.935	75.8%	78.7%	0.448
Yes	21.9%	22.9%	0.756	22.4%	22.7%	0.935	24.2%	21.3%	0.448
Reported quitting or reducing alcohol consumption
No	93.4%	91.6%	0.406	93.3%	91.8%	0.482	93.9%	92.3%	0.502
Yes	6.6%	8.4%	0.406	6.7%	8.2%	0.482	6.1%	7.7%	0.502
Plans of quitting or reducing in the future
No	87.3%	88.3%	0.699	87.4%	88.9%	0.556	86.1%	89.7%	0.209
Yes	12.7%	11.7%	0.699	12.6%	11.1%	0.556	13.9%	10.3%	0.209
Thinking of quitting or reducing alcohol in the future
No	86.7%	86.4%	0.918	86.8%	86.0%	0.781	85.5%	84.5%	0.765
Yes	13.3%	13.6%	0.918	13.2%	14.0%	0.781	14.5%	15.5%	0.765
Alcohol consequences	4.8	5.5	0.594	4.9	5.5	0.596	5.1	6.4	0.378
Social motives for alcohol consumption	13.6	12.6	0.041	13.6	12.6	0.040	13.9	13.4	0.355
Coping motives for alcohol consumption	8.5	8.3	0.710	8.4	8.4	0.789	8.5	8.8	0.362
n	663	214	877	642	207	849	538	155	693

Note: Data come from wave 15 of the STRATA cohort study. The baseline W15 sample includes participants who reported past‐month alcohol consumption at W15 and had non‐missing demographic and outcome data. The prospective abstention sample includes participants from the baseline W15 sample who were followed at W16 and had non‐missing demographic data; this sample is used for analyses of transition to past‐year abstention at W16. The prospective non‐abstention outcome sample is smaller because several alcohol risk, change and motive outcomes were only asked among participants who reported past‐month alcohol use at both W15 and W16 and had non‐missing values for all model covariates and outcomes. Reported values refer to W15 characteristics. P‐values compare participants who did and did not report past‐month AFLAB use within each analytic sample.

Abbreviations: AFLAB, alcohol‐free and low‐alcohol beverage; GED, General Educational Development; HS, high school; wave 15, W15; wave 16, W16;

In the baseline W15 sample, people reporting AFLAB use were not significantly different from those not reporting in terms of past‐month heavy episodic drinking (48.6% versus 54.4%), recent quit attempts (22.9% versus 21.9%), reported quitting or reducing alcohol consumption (8.4% versus 6.6%), plans to quit or reduce drinking (11.7% versus 12.7%), thinking about quitting or reducing drinking (13.6% versus 13.3%), alcohol‐related consequences (means = 5.5 versus 4.8), or coping motives (8.3 versus 8.5). However, those reporting AFLAB use reported lower social drinking motives (12.6 versus 13.6).

Cross sectional associations

Table 2 reports cross‐sectional associations using W15 data. In unadjusted and adjusted models, AFLAB use was associated with lower social drinking motives (unadjusted coefficient = −0.005; 95% CI = −0.011 to −0.0001; adjusted coefficient = −0.005; 95% CI = −0.011 to 0.0001). All other associations were statistically insignificant.

TABLE 2.

Cross‐sectional associations between AFLAB consumption and alcohol consumption, risk and change measures and motives, wave 15.

Alcohol consumption, risk and change measures and motives	Unadjusted marginal effect [95% CI]	Adjusted marginal effect [95% CI]
Heavy episodic drinking, past month	−0.043 [−0.100 to 0.013]	−0.035 [−0.092 to 0.022]
Reported quit attempt, past 3 months	0.011 [−0.057 to 0.079]	0.019 [−0.049 to 0.087]
Reported quitting or reducing alcohol consumption	0.047 [−0058 to 0.152]	0.058 [−0.048 to 0.165]
Reported plans of quitting or reducing in the future	−0.017 [−0.105 to 0.071]	−0.031 [−0.120 to 0.058]
Reported thinking of quitting or reducing alcohol in the future	0.004 [−0.079 to 0.088]	0.007 [−0.075 to 0.090]
Alcohol consequences	0.001 [−0.002 to 0.003]	0.001 [−0.001 to 0.003]
Social motives for alcohol consumption	−0.005 [−0.011 to −0.0001]	−0.005 [−0.011 to 0.0001]
Coping motives for alcohol consumption	−0.001 [−0.009 to 0.006]	−0.0006 [−0.008 to 0.006]

Note: Data come from wave 15 of the STRATA cohort study. Models use the baseline W15 analytic sample of participants who reported past‐month alcohol consumption and had non‐missing data. Each row presents results from a separate regression model estimating the cross‐sectional association between past‐month AFLAB use and the listed alcohol consumption, risk and change measure or drinking motive. Adjusted models control for age, race/ethnicity, sex, educational attainment, sexual orientation and pregnancy status. For binary outcomes, logistic regression models were estimated and average marginal effects are reported. For continuous outcomes, linear regression models were estimated and unstandardized coefficients are reported. Robust standard errors were used. Bold = P < 0.05. Italicized = P < 0.10.

Abbreviation: AFLAB, alcohol‐free and low‐alcohol beverage; CI, confidence interval; W15, Wave 15.

Longitudinal associations

Table 3 reports prospective associations between W15 AFLAB use and W16 outcomes. For the abstention model, we restricted the sample to participants who reported past‐month alcohol consumption at W15 so that all participants were at risk of transitioning to abstention at follow‐up and because of this restriction, the model could not adjust for baseline abstention. Baseline AFLAB use was associated with a higher probability of past‐year abstention at follow‐up in the unadjusted model [4.9 percentage point (pp); 95% CI = 1.0–8.7] and the demographic‐adjusted model (5.1 pp; 95% CI = 1.1–9.1). This association was similar after additional adjustment for baseline indicators of recent quit attempts, reduced drinking, plans to quit or reduce drinking, thinking about quitting or reducing drinking and social and coping drinking motives (4.8 pp; 95% CI = 1.0–8.5). In the fully adjusted model, there was suggestive evidence (P < 0.10) that AFLAB use was associated with a higher likelihood of reporting a quit attempt in the past 3 months (6 pp; 95% CI = −0.9 to 12.8). All other prospective associations were statistically insignificant. In most models, adding baseline AFLAB consumption modestly improved model fit, as reflected by higher R ² or pseudo‐R ² values compared with otherwise similar models that excluded AFLAB use.

TABLE 3.

Longitudinal associations between alcohol consumption, risk and change measures and motives and baseline AFLAB consumption.

Outcomes	(1)	(2)	(3)	(4)	(5)
	Unadjusted, incl. baseline AFLAB	Demographic adjusted, incl. baseline AFLAB (no baseline outcome adjustment)	Demographic adjusted model excl. baseline AFLAB	Fully adjusted, incl. baseline AFLAB	Fully adjusted abstention model with additional adjustment for demographics, baseline change and motivation
Alcohol Abstention	0.049 [0.010–0.087] Pseudo R 2 = 0.013	0.051 [0.011–0.091] Pseudo R 2  = 0.041	Pseudo R 2  = 0.027	N/A	0.048 [0.010–0.085] Pseudo R 2  = 0.093
Heavy episodic drinking, past month	−0.052 [−0.140 to 0.036] Pseudo R 2 = 0.001	−0.040 [−0.128 to 0.049] Pseudo R 2 = 0.039	Pseudo R2 = 0.027	−0.035 [−0.113 to 0.042] Pseudo R 2 = 0.179	N/A
Reported quit attempt, past 3 months	0.050 [−0.025 to 0.126] Pseudo R 2 = 0.002	0.053 [−0.024 to 0.130] Pseudo R 2 = 0.018	Pseudo R 2 = 0.011	0.060 [−0.009 to 0.128] Pseudo R 2  = 0.186	N/A
Reported quitting or reducing alcohol consumption	−0.005 [−0.053 to 0.043] Pseudo R2 = 0.000	−0.009 [−0.057 to 0.038] Pseudo R 2 = 0.030	Pseudo R 2 = 0.023	−0.016 [−0.060 to 0.028] Pseudo R 2 = 0.108	N/A
Reported plans of quitting or reducing in the future	0.027 [−0.036 to 0.090] Pseudo R 2 = 0.001	0.026 [−0.038 to 0.090] Pseudo R 2 = 0.025	Pseudo R 2 = 0.024	0.038 [−0.023 to 0.098] Pseudo R 2 = 0.079	N/A
Reported thinking of quitting or reducing alcohol in the future	−0.025 [−0.090 to 0.040] Pseudo R 2 = 0.001	−0.024 [−0.091 to 0.043] Pseudo R 2 = 0.029	Pseudo R 2 = 0.018	−0.026 [−0.089 to 0.038] Pseudo R 2 = 0.084	N/A
Alcohol consequences	−0.137 [−2.485 to 2.210] R 2 = 0.000	0.369 [−2.181 to 2.835] R 2 = 0.030	R 2 = 0.023	−0.427 [−2.415 to 1.561] R 2 = 0.221	N/A
Social motives for alcohol consumption	−0.655 [−1.692 to 0.383] R 2 = 0.002	−0.515 [−1.557 to 0.527] R 2 = 0.029	R 2 = 0.031	−0.270 [−1.057 to 0.518] R 2 = 0.420	N/A
Coping motives for alcohol consumption	−0.195 [−0.950 to 0.560] R 2 = 0.000	−0.066 [−0.854 to 0.723] R 2 = 0.028	R 2 = 0.026	−0.407 [−1.000 to 0.186] R 2 = 0.435	N/A

Note: Data come from waves 15 and 16 of the STRATA cohort study. Each row presents a separate prospective regression model estimating the association between baseline W15 AFLAB use and the listed W16 outcome. The abstention models use the prospective abstention sample: participants who reported past‐month alcohol consumption at W15, were followed at W16 and had non‐missing data. This restriction ensured that all participants were active drinkers when AFLAB use was measured and therefore ‘at risk’ of transitioning to abstention at follow‐up. Models for all other outcomes use the prospective non‐abstention outcome sample, which is smaller because several alcohol risk, change and motive outcomes were only asked among participants who reported past‐month alcohol use at both W15 and W16 and had non‐missing values for all model covariates and outcomes. Column 1 reports unadjusted models. Column 2 adjusts for demographic characteristics and AFLAB, but not baseline outcomes. Column 3 reports the R ² or pseudo‐R ² from otherwise comparable demographic‐adjusted models that exclude AFLAB use. Column 4 reports fully adjusted models that include demographic characteristics and the baseline value of the outcome. Column 5 reports an additional abstention model that adjusts for demographics and baseline indicators of recent quit attempts, reduced drinking, plans to quit or reduce drinking, thinking about quitting or reducing drinking and social and coping drinking motives; because all participants in the abstention sample reported past‐month drinking at W15, this model could not adjust for baseline abstention. For binary outcomes, logistic regression models were estimated and average marginal effects are reported. For continuous outcomes, linear regression models were estimated and unstandardized coefficients are reported. Robust standard errors were used. Bold = P < 0.05. Italicized = P < 0.10.

Abbreviation: AFLAB, alcohol‐free and low‐alcohol beverage; incl., including; NA, not applicable; W15, wave 15; W16, wave 16.

DISCUSSION

To date, few studies have investigated predictors of AFLAB use specifically among contemporary US young adults, and few have used longitudinal data to test whether AFLAB use predicts subsequent alcohol consumption, risk and change measures and drinking motives after accounting for prior levels of use.

We examined demographic differences in AFLAB use and whether AFLAB use was associated with concurrent and subsequent alcohol consumption, risk and change measures and drinking motives. Cross‐sectionally, AFLAB use was more common among young adult women, less common among non‐Hispanic White participants and less common among those identifying as gay or lesbian. AFLAB use was not consistently associated with indicators of alcohol risk or change in baseline comparisons, although AFLAB users reported slightly lower social drinking motives. After adjusting for demographics, AFLAB use remained associated with lower social drinking motives. Longitudinally, baseline AFLAB use was associated with a higher likelihood of alcohol abstention in the subsequent year, with suggestive evidence of a higher likelihood of reporting a past‐3‐month quit attempt.

Our findings extend a literature that has documented rising product availability, interest and mixed motivations for AFLAB use [5, 6]. In baseline comparisons, AFLAB use was not consistently associated with alcohol risk or change measures, suggesting that AFLAB use among young adults may reflect heterogeneous motivations rather than a single risk profile. However, baseline AFLAB use was prospectively associated with subsequent abstention, with suggestive evidence of an association with quit attempts, even though it was not associated with heavy episodic drinking or alcohol‐related consequences. The persistence of the abstention association after adjustment for baseline change and motivation measures suggests that AFLAB use may capture drinking‐change processes not reflected in quit attempts or plans to reduce drinking. This pattern, combined with the concurrent association with social motives, suggests AFLABs may support or accompany transitions into abstention for some young adults, including lighter or infrequent drinkers or those using AFLABs for lower‐risk substitution [6, 7, 8]. At the same time, the absence of associations with heavy episodic drinking or alcohol‐related consequences suggests that AFLAB use should not be interpreted as a broader marker of reduced alcohol risk or harm over the follow‐up period.

From a prevention and clinical perspective, findings suggest that AFLABs may be relevant to harm‐reduction approaches that emphasize feasible, lower‐risk substitutions. The association between AFLABs and subsequent abstention suggests they may be useful for some young adults experimenting with reduced drinking or non‐drinking. Campus health and clinical programs could discuss AFLABs within brief interventions, such as motivational interviewing, as one option for planned non‐drinking days or beverage substitution. However, because AFLAB use was not associated with reductions in heavy episodic drinking or alcohol‐related consequences, AFLABs should be presented as one possible tool within a broader behavior‐change strategy rather than as a standalone approach to reducing alcohol‐related harms. Public health messaging could also present AFLAB use as a socially acceptable strategy for cutting back [14, 15].

Our study is not without limitations. Outcomes were self‐reported and may be affected by recall, social‐desirability bias or seasonality. Past‐12‐month abstention may partially overlap with the pre‐baseline period for some participants, attenuating temporal interpretation. Missing data may bias estimates if related to alcohol behaviors or AFLAB use. Bounded and right‐skewed continuous outcomes should be interpreted as conditional mean differences. Although models adjusted for demographic covariates and baseline values of the modelled outcome, unmeasured confounding, bidirectional processes and motivation to cut down remain possible. Baseline AFLAB use may partly capture motivation, particularly given our single binary measure. Estimates should not be interpreted causally, and some findings may reflect chance because of multiple hypothesis testing. As such, additional research is needed to confirm the findings presented in this study in other samples.

In conclusion, AFLAB use among young adults is common and was prospectively associated with alcohol abstention, with suggestive evidence of an association with quit attempts. Findings suggest that AFLABs may be a useful tool within broader self‐regulation strategies as drinking norms continue to evolve in this developmental period. Future research should consider how findings vary in different samples and heterogeneously across demographic factors.

AUTHOR CONTRIBUTIONS

Samuel Mann: Conceptualization; validation; formal analysis; writing—review and editing; writing—original draft; investigation. Daniel Siconolfi: Supervision; writing—original draft; writing—review and editing. Anthony Rodriguez: Writing—review and editing. Rachana Seelam: Data curation; writing—review and editing. Eric R. Pedersen: Writing—review and editing. Michael Dunbar: Writing—review and editing. Elizabeth J. D'Amico: Writing—review and editing; funding acquisition; supervision.

DECLARATION OF INTERESTS

None.

Mann S, Siconolfi D, Rodriguez A, Seelam R, Pedersen ER, Dunbar M, et al. Alcohol‐free and low‐alcohol beverage consumption and concurrent and subsequent alcohol outcomes among young adults. Addiction. 2026;121(10):2845–2851. 10.1111/add.70514

DATA AVAILABILITY STATEMENT

Data are not yet available as the parent study is still in progress.