Abstract
Introduction.
The use of psychotropic substances has negative short- and long-term health outcomes, including complex direct and indirect effects on sleep and sleep-cardiovascular function. Here, we investigate daily relationships between self-reported substance use and objective measures of sleep and sleep-related heart rate (HR) in community-dwelling young adults.
Methods.
Fifty-five healthy young adults (Mage = 23.1 ± 2.29 y, 30 female) in the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) study completed a 28-day ecological momentary assessment protocol, including remote sleep and HR measurements via Fitbit devices, as well as daily app-based self-reports of alcohol and other substance use.
Results.
A total of 1,459 days of data were collected. Caffeine was the most frequent substance used, followed by alcohol, cannabis, nicotine, and other drugs. The analysis showed that substance use was associated with delays in sleep start and end time, reduced sleep duration and efficiency, and increased wake after sleep onset. Increases in sleep heart rate were associated with prior-day alcohol use.
Discussion.
Substance use negatively influences sleep and sleep HR. These preliminary data highlight the potential value of using remote multimodal data collection to investigate the daily relationships between substance use and sleep in young adults, in an ecological setting.
Keywords: substance use, sleep, cardiovascular, wearable, diary, free-living
2. Introduction
The consumption of psychotropic substances is common among young adults. It involves the use of substances with varying levels of legality across countries, including alcohol, caffeine, nicotine, heroin, cocaine, etc. According to the annual report of the Substance Abuse and Mental Health Services Administration (SAMHSA) from 2022, young adults (aged 18 to 25) represent an at-risk cohort for substance abuse.[1] Besides caffeine, which is by far the most used substance worldwide,[2-4] in 2022 in the U.S., alcohol was the most commonly used substance among young adults (17.5 million, 50.2%), followed by nicotine (10.4 million, 30.0%), cannabis (9.0 million, 25.9%), and other illicit drugs like cocaine (1.3 million, 3.7%), methamphetamine (161.000, 0.5%), and heroin (55.000, 0.2%).[1]
All these substances may have detrimental effects across multiple health domains. For example, heavy alcohol consumption has a profound impact on cardiovascular function and is associated with increased cardiovascular risks.[5] Between 2015 and 2019, almost twenty-five thousand people died of heart complications due to alcohol use in the US.[6] The link between alcohol and cardiovascular health is complex with multiple factors thought to play a role in the relationship, notably including sleep. Alcohol has different effects across distinct sleep processes. A first experimental study conducted by Feige and colleagues [7] clearly showed a strong hypnagogic effect of alcohol, with reduced polysomnographic (PSG) sleep onset latency (i.e., minutes required to get asleep) and increased slow wave sleep (i.e., the so-called ‘deep sleep’) in the first part of the night. However, these initial sleep “improvements” were counter-balanced by a rebound effect characterized by suppression of rapid-eye-movement (REM) sleep (i.e., sleep periods characterized by muscle atonia and physiological activity almost resembling wakefulness) in the first part of the night and increased stage-1 sleep (i.e., the ‘lightest’ phase of non-REM sleep) in the second part. Consistent with Feige et al.,[7] several studies over the years further proved the negative impact of alcohol on both sleep macro- and micro-structure. Indeed, in healthy adults not only has alcohol been proven to reduce total sleep time (i.e., time spent sleeping) and sleep efficiency (i.e., proportion of time spent sleeping over time in bed), but also to reduce REM sleep and increase wakefulness.[8-10] Interestingly, while the above studies showed the possible negative effect of nighttime alcohol consumption on subsequent sleep, an open question is whether this effect changes depending on the time between consumption and sleep onset. The few studies on the topic suggested more pronounced effects when alcohol is taken in the nighttime compared to the afternoon. [11-12] Sleep alterations were also reported after caffeine consumption, consisting of an overall reduction in sleep efficiency and total sleep time (see Gardiner et al.,[13] for a meta-analysis of experimental studies). Similar effects were reported for nicotine.[14-15] Cannabis seems to particularly affect REM sleep by reducing its duration across the night.[16] Evidence of the negative effects of substance consumption has also been reported by actigraphic studies, which detected a negative effect of alcohol [17-18] and cannabis [19] on sleep efficiency. Survey studies also reported an association between worse sleep quality and alcohol, [20] caffeine, [4] and nicotine.[21] Conversely, mixed results were found for cannabis.[20,22] It is important to note that the magnitude of these effects may depend on several factors, including dose and time of substance use, metabolism of individuals, and demographic variables.
Less investigated are the specific effects of alcohol and other substance use on sleep cardiovascular restoration. Sagawa et al.[8] conducted a laboratory study in a sample of 10 young male adults to investigate the effects of alcohol on sleep and heart rate (HR) variability (HRV). Specifically, participants underwent three non-consecutive nights of alcohol-dose administration, followed by sleep PSG recording and HRV monitoring. Results showed a dose-dependent effect of alcohol, with increased HR and decreased HRV associated with higher doses. A similar experiment was conducted by de Zambotti and colleagues[23] in a sample of 26 healthy participants (30-60 y). The authors compared three nights of PSG recordings in which different doses of alcohol were administered before bedtime. Both low (1 standard drink for women and 2 for men) and high (3 standard drinks for women and 4 for men) doses of alcohol strongly affected sleep CV function. Specifically, HR was higher during the first 4 hours of the low-dose sleep night and during the first 6 hours of the high-dose sleep night. Lastly, in women, systolic blood pressure showed increases in the morning hours after high-dose alcohol consumption. Others have reported that nicotine and caffeine also affect sleep CV function by enhancing sympathetic activity during sleep.[24-25] Still largely unknown is the synergistic effect of alcohol and other substances on sleep cardiovascular function.
In the current study, we adopted a 28-day ecological momentary assessment (EMA) design to further investigate the relationship between alcohol and other substance use with sleep in a sample of healthy young adults from the US National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) longitudinal study. A combination of passive wearable acquisition with daily surveys (high-density daily data sampling) is particularly well-suited for studying real-time relationships between biopsychological factors and behaviors.[26] Recent studies have exploited its potential by studying the interactions between daily sleep and adolescent behaviors,[27] and daily HR variability in smokers.[28] However, to our knowledge, the combination of high density wearable and survey data collection has not been used for an analysis of the relationship between substance use and sleep patterns in young adults.
Considering the evidence in the literature, here we hypothesize that alcohol and other substances (i.e., caffeine, nicotine, cannabis, and other drugs) are associated with negative outcomes in sleep behavior (i.e., sleep schedules such as bedtime and wake times), sleep macro-structure (i.e., night-level sleep parameters such as sleep duration and sleep efficiency), and sleep-related HR.
3. Method
3.1. Participants
Fifty-five (55) healthy young adults (Mean age = 23.24 ± 2.42 years, 30 female, 25 male) who were participating in the 8th and 9th year follow-ups of the NCANDA study [29] at the SRI International site, took part in a 28-day Ecological Momentary Assessment (EMA) of sleep, alcohol, and other substance use. Data were collected between January and April 2022. Inclusion criteria for the NCANDA study at baseline included an age range between 12.0 and 21.9 years, absence of medical and psychiatric disorders, absence of developmental disorders, and absence of substance dependence [29]. Inclusion criteria to be involved in the 28-day EMA, were to be enrolled in the main NCANDA study, age > 18 years, absence of medical, psychiatric, and substance use disorders (assessed at the annual NCANDA assessment), and to have a Fitbit-compatible phone.
The Advara Institutional Review Board and the SRI International Institutional Review Board approved the study protocol. Participants in NCANDA completed consent if they were older than 18 years, or assent, with a parent completing consent, if they were younger than 18 years. In order to participate in the EMA and Fitbit portion of the study, they completed an additional specific written consent form and received additional monetary compensation.
See Table 1, for demographics and sample characterization. Data were obtained through the Research Electronic Data Capture (REDCap) data management online platform.[30-31] These data were part of the public data release NCANDA_RELEASE_BASE_REDCAP_MEASUREMENTS_V07, whose collection and distribution were supported by the National Institute on Alcohol Abuse and Alcoholism funding AA021697, AA021695, AA021692, AA021696, AA021681, AA021690, and AA021691.
Table 1.
Sample characterization
| Women | Men | Total | ||
|---|---|---|---|---|
| Sample, n. | 30 | 25 | 55 | |
| Age, years, M (SD) (range = 19-28) | 23.40 (2.49) | 22.80 (2.02) | 23.24 (2.42) | |
| Race | ||||
| Asian, n. | 7 | 3 | 10 | |
| White, n. | 21 | 20 | 41 | |
| African-American/Black | 0 | 2 | 2 | |
| Native American/American Indian | 0 | 1 | 1 | |
| More than one race, n. | 1 | 0 | 1 | |
| Ethnicity | ||||
| Hispanic or Latino, n. | 2 | 5 | 7 | |
| Habitual Sleep and Habitual Alcohol Consumption, M (SD) | ||||
| PSQI (range = 1-10) | 3.82 (2.01) | 4.44 (2.28) | 4.14 (2.14) | |
| Alcohol in the past year, n. of drinks | 68.31 (62.60) | 62.17 (52.31) | 65.53 (57.70) | |
| Body Composition | ||||
| BMI, kg x m-2, M (SD) | 24.19 (4.71) | 24.52 (4.11) | 24. 43 (4.36) | |
| Healthy (BMI: 18.5 to < 25), n. | 20 | 14 | 34 | |
| Overweight (BMI: 25 to < 30), n. | 7 | 8 | 15 | |
| Obese (BMI: 30 or higher), n. | 3 | 3 | 6 | |
| Daily Alcohol and Substances Consumption Over 28-day EMA Assessment, M (SD) | ||||
| Alcohol drinks (n. days/total) | 0.24 (0.43) | 0.19 (0.35) | 0.22 (0.42) | |
| Morning | 0.00 (0.02) | 0.00 (0.01) | 0.00 (0.01) | |
| Afternoon | 0.04 (0.06) | 0.04 (0.10) | 0.04 (0.08) | |
| Evening | 0.22 (0.23) | 0.15 (0.18) | 0.18 (0.21) | |
| Night | 0.14 (0.16) | 0.16 (0.18) | 0.15 (0.17) | |
| Alcohol drinks per day (n. of drink/total days) | 0.85 (0.87) | 0.86 (1.33) | 0.85 (1.09) | |
| Caffeine (n. days/total) | 0.65 (0.48) | 0.57 (0.50) | 0.61 (0.49) | |
| Nicotine (n. days/total) | 0.11 (0.31) | 0.08 (0.28) | 0.10 (0.30) | |
| Cannabis (n. days/total) | 0.12 (0.33) | 0.17 (0.37) | 0.14 (0.35) | |
| Other Drugs (n. days/total) | 0.00 (0.06) | 0.01 (0.33) | 0.01 (0.08) | |
BMI = Body Mass Index; EMA = Ecological Momentary Assessment; PSQI = Pittsburgh Sleep Quality Index [32].
3.2. Procedure
Participants were allowed to freely decide when starting the EMA protocol in accordance with their own schedule. Specifically, most of the participants started the data collection protocol on Friday (27.5%), 20% on Tuesday, and 14.5% on Monday, Wednesday, and Thursday, whereas only 3.5% and 5.5% started on Saturday and Sunday, respectively. The distribution of the days of the week across the 28-day EMA was comparable (Sunday = 14%, Monday = 14.2%, Tuesday = 14.3%, Wednesday = 14.8%, Thursday = 14.1%, Friday = 14.6%, Saturday = 14%). During the EMA protocol, participants were asked to wear a Fitbit Charge 3 (FC3) or a Fitbit Charge 4 (FC4) (Fitbit, LLC., San Francisco, CA) 24/7 on their non-dominant wrist, to synch the device with the Fitbit App as well as to respond to a survey each morning through the mNCANDA mobile phone app.[33] The Fitabase platform (Small Steps Labs LLC., San Diego, CA) was used to support Fitbit data collection and management. Furthermore, participants were guided to turn off the notifications and data feedback of the device and not to run any updates or change the settings. On each morning, participants were asked to sync the Fitbit device with the paired App. The entire EMA procedure is displayed in Figure 1.
Figure 1.
Schematic description of the preliminary phases and the ecological momentary assessment (EMA) protocol of the study.
All instructions and devices were provided during an in-person lab appointment and/or via online meetings. On these occasions, participants were also guided to install Fitbit and mNCANDA apps on their mobile phones using ad-hoc e-mail addresses with no personally identifying information.
3.3. Measures
3.3.1. Fitbit
Sleep, sleep HR, and physical activity measures were obtained via FC3 and FC4 devices (firmware versions: 8.20001.96.29 - 20001.88.11). These consumer-grade devices have been repeatedly evaluated against reference standard PSG.[34-36] Here, we largely followed the recommendations for data management, use, and interpretation of wearable data, as introduced by de Zambotti et al.[37]
Fitbit data (sleep and sleep HR) were exported from Fitabase to be pre-processed before analysis. We first applied the Menghini et al.[27] analytic pipeline to precisely identify nocturnal sleep periods and filter atypical/unreliable cases. Subsequently, we used high-granularity (30-second epoch-by-epoch, EBE) data to derive the outcomes for each identified sleep period starting between 06:00 PM and 06:00 AM, with a minimum sleep duration of ≥ 3 hours. Based on EBE sleep classification, we retrieved the following outcomes (see Menghini et al.[38]): sleep period start (SPS, hh.mm, i.e., the timestamp of the first epoch included in the sleep period), sleep onset (SO, hh.mm, i.e., the timestamp of the first epoch classified as sleep), and end time (SPE, hh:mm, i.e., the timestamp of the last epoch classified as sleep), sleep period duration (SPD, hours, i.e., the time between SPS and SPE), total sleep time (TST, hours, i.e., time spent sleeping between SPS and SPE), wake after sleep onset (WASO, i.e., hours, time spent awake between SO and SPE), sleep efficiency (SE, %, i.e., percentage of time spent asleep between SPS and SPE), the percentages (over TST) of “light,” “deep,” and REM sleep, in addition to sleep midpoint (hh.mm, i.e., the timestamp of the halfway point between SO and SPE). Furthermore, HR per minute was used to compute mean HR (bpm) in non-REM (NREM-HR) and REM sleep (REM-HR).
3.3.2. mNCANDA Survey
The mNCANDA mobile-phone survey consisted of ad-hoc questions to be answered each morning from 07:00 am (after a signal time) via the mNCANDA app.[33] It took around 1-2 minutes to be completed, with no time limit. From the signal time, participants could complete the survey at any time of the day. Specifically, participants were asked to indicate the number of alcoholic drinks consumed the day before (How many alcoholic drinks did you have yesterday?), and the timing (When did you drink? Morning (05:00 – 12:00), Afternoon (12:00 – 17:00), Evening (17:00 – 21:00), Night (21:00 – 05:00)), and if they used cannabis (Did you use marijuana yesterday? Yes/No), other drugs (Did you use another drug yesterday? Yes/No), nicotine/tobacco (Did you use nicotine/tobacco yesterday? Yes/No), and caffeine (Did you drink caffeine yesterday? Yes/No). The last question checked whether the participants synchronized their Fitbit that morning (Did you sync your Fitbit today? Yes/No).
The completion rate before 1:00 pm was 95%, and the remaining 5% completed the interview by 6:00 pm.
After each submission, the corresponding anonymized dataset was automatically loaded into the mNCANDA database and associated with the participant’s identification code.
3.4. Statistical analyses
We investigated the relationships between alcohol and other substance use with Fitbit measures (sleep and sleep HR) by using linear mixed-effect regression models with the restricted maximum likelihood (REML) estimator. This approach is the most suitable for daily observations nested within participants.[39] Specifically, we specified a separate hierarchical set of models for each Fitbit measure (i.e., dependent variables). In each set of models, we firstly inserted demographic (biological sex, age, and body mass index [BMI]) and alcohol use variables as predictors, with the number of alcoholic drinks consumed each day inserted at both within- (i.e., person-mean centered) and between-person level (i.e., person means). In the second step, we included all the other substance use variables (yes/no for nicotine, cannabis, caffeine, other drugs) if they improved the model. Participants were included as random effects. For each predictor, we conducted a stepwise evaluation by considering both the Akaike Information Criterion (AIC) and Schwartz’s Bayesian Information Criterion (BIC), including only the ones generating the lower AIC and BIC. This approach (i.e., stepwise) generates models with a different number of predictors, but greatly increases the goodness-of-fit of the model.
All analyses were conducted by using Jamovi 2.3.21.,[40] a statistical software based on R syntax. The chi-square (χ²) Test was used in the case of categorical measures. Descriptive data were reported as Mean (Standard Deviation), except when otherwise specified. Confidence intervals (CI) were also reported when appropriate. The intraclass correlation coefficient (ICC) was used for each measure as an index of the person- vs day-level variability.
4. Results
4.1. Relationships between substance use, sleep timing, sleep stages, and sleep cardiac activity measures
By considering the whole sample at the within-person level, the linear mixed-effect model analysis yielded several associations between substance use and Fitbit outcomes.
Overall, the sleep variables showed higher within-individual variability (ICC ranging from 0.21 to 0.44) than HR variables (ICC ranging from 0.71 to 0.74). Furthermore, higher within-individual variability was found in alcohol use (ICC = 0.28) than in other substance use (ICC from 0.66 to 0.60 for caffeine, nicotine, and cannabis use).
Regarding sleep timing behaviors, delays in both sleep onset (AIC = 3954, BIC = 4037, R2 = 0.43, ICC = 0.41) and sleep period start (AIC = 3948, BIC = 4057, R2 = 0.43, ICC = 0.42) were predicted by nighttime alcohol and nicotine consumption. Sleep midpoint (AIC = 3583, BIC = 3682, R2 = 0.46, ICC = 0.44) and sleep period end (AIC = 3952, BIC = 4105, R2 = 0.40, ICC = 0.37) were negatively predicted by other drugs consumption (Table 2), indicating an association between other drug use and delayed sleep schedules.
Table 2.
Unstandardized parameters estimated by linear mixed-effect model of the relationship between alcohol and other substance use and Fitbit sleep timing measures (N = 1,459 obs. from 55 participants)
| Sleep Period Start | Sleep Period End | Sleep Onset | Sleep Midpoint | |||||
|---|---|---|---|---|---|---|---|---|
| Predictors | b(SE) | 95% CI | b( SE) | 95% CI | b (SE) | 95% CI | b (SE) | 95% CI |
| Intercept | 24.16 (0.16) | 23.85, 24.48 | 8.03 (0.16) | 7.72, 8.33 | 24.27 (0.16) | 23.96, 24.57 | 28.14 (0.14) | 27.86, 28.43 |
| Sex (F-M) | −0.48 (0.32) | −1.11, 0.16 | −0.14 (0.15) | −0.76, 0.49 | −0.50 (0.32) | −1.13, 0.13 | −0.30 (0.29) | −0.87, 0.27 |
| Age | −0.14 (0.07) | −0.29, 0.00 | −0.08 (0.07) | −0.23, 0.06 | −0.14 (0.07) | −0.29, 0.00 | −0.10 (0.06) | −0.24, 0.02 |
| BMI | 0.02 (0.04) | −0.06, 0.09 | −0.01 (0.04) | −0.08, 0.06 | 0.01 (0.03) | −0.06, 0.09 | −0.00 (0.03) | −0.07, 0.07 |
| Alcoholic Drink/Day (within) | −0.06 (0.04) | −0.15, 0.03 | 0.06 (0.04) | −0.03, 0.14 | −0.03 (0.04) | −0.11, 0.04 | 0.00 (0.03) | −0.06, 0.06 |
| Alcoholic Drink/Day (between) | −0.12 (0.15) | −0.43, 0.18 | −0.24 (0.16) | −0.54, 0.06 | −0.10 (0.15) | −0.41, 0.21 | −0.17 (0.14) | −0.45, 0.10 |
| Morning Alcohol | −0.16 (0.70) | −1.55, 1.23 | −0.25 (0.71) | −1.64, 1.13 | ||||
| Afternoon Alcohol | −0.19 (0.25) | −0.68, 0.29 | −0.39 (0.25) | −0.88, 0.10 | −0.24 (0.25) | −0.73, 0.25 | −0.28 (0.21) | −0.70, 0.13 |
| Evening Alcohol | 0.14 (0.14) | −0.14, 0.41 | −0.05 (0.14) | −0.33, 0.22 | ||||
| Night Alcohol | 0.43 (0.17) | 0.08, 0.77* | 0.10 (0.17) | −0.24, 0.45 | 0.38 (0.17) | 0.05, 0.72* | 0.27 (0.14) | −0.01, 0.55 |
| Cannabis | −0.16 (0.18) | −0.53, 0.19 | ||||||
| Other Drugs | −0.05 (0.27) | −1.05, 0.94 | −1.65 (0.51) | −2,65, −0,66*** | −0.82 (0.43) | −1.66, 0.01* | ||
| Nicotine | 0.69 (0.28) | 0.15, 1.24* | 0.25 (0.29) | −0.08, 0.48 | 0.70 (0.28) | 0.16, 1.25* | 0.42 (0.25) | −0.06, 0.91 |
| Caffeine | 0.19 (0.14) | −0.32, 0.83 | ||||||
p ≤ 0.05.
p ≤ 0.001.
Other drugs consumption was also associated with reduced sleep duration. Indeed, other drugs consumption negatively predicted both sleep period duration (AIC = 13051, BIC = 13055, R2 = 0.27, ICC = 0.23) and total sleep time (AIC = 12783, BIC = 12785, R2 = 0.27, ICC = 0.23) (Table 3). Increased wake after sleep onset (WASO) was predicted by cannabis consumption (AIC = 9826, BIC = 9848, R2 = 0.22, ICC = 0.22), whereas afternoon alcohol predicted lower sleep efficiency (AIC = 5865, BIC = 5962, R2 = 0.29, ICC = 0.31), indicating a more disrupted sleep associated with cannabis and alcohol consumption (Table 3). Light sleep duration was positively predicted by older age (AIC = 10588, BIC = 10501, R2 = 0.43, ICC = 0.37), whereas REM sleep duration (AIC = 9286, BIC = 9308, R2 = 0.25, ICC = 0.26) was predicted by both female gender and younger age. No significant predictor emerged for deep sleep duration (AIC = 8688, BIC = 8728, R2 = 0.22, ICC = 0.21) (Table 4).
Table 3.
Unstandardized parameters estimated by the linear mixed-effect model of the relationship between alcohol and other substance use and Fitbit sleep structure measures (N = 1,459 obs. from 55 participants)
| Sleep Period Duration | Total Sleep Time | Sleep Efficiency | Wake After Sleep Onset | |||||
|---|---|---|---|---|---|---|---|---|
| Predictors | b(SE) | 95% CI | b( SE) | 95% CI | b (SE) | 95% CI | b (SE) | 95% CI |
| Intercept | 471.83 (7.57) | 457.00, 486.66 | 419.38 (6.59) | 406.46, 432.30 | 89.03 (0.30) | 88.44, 89.62 | 46.06 (1.50) | 43.11, 49.00 |
| Sex (F-M) | 21.95 (15.32) | −8.07, 51.97 | 18.06 (13.37) | −8.14, 44.26 | −0.65 (0.59) | −1.82, 0.51 | 4.97 (3.04) | −0.99, 10.95 |
| Age | 4.32 (3.56) | −2.66, 11.30 | 2.79 (3.12) | −3.32, 8.90 | −0.14 (0.14) | −0.41, 0.13 | 1.09 (0.70) | −0.29, 2.48 |
| BMI | −1.69 (1.81) | −5.24, 1.85 | −1.54 (1.57) | −4.62, 1.55 | 0.02 (0.07) | −0.11, 0.16 | −0.07 (0.36) | −0.77, 0.63 |
| Alcoholic Drink/Day (within) | 2.60 (1.80) | −0.92, 6.13 | 3.88 (2.07) | −0.18, 7.93 | 0.01 (0.07) | −0.12, 0.15 | 1.01 (0.54) | −0.05, 2.07 |
| Alcoholic Drink/Day (between) | −6.13 (7.63) | −21.09, 8.82 | −7.19 (6.69) | −20.29, 6.92 | −0.01 (0.29) | −0.58, 0.55 | −1.03 (1.52) | −4.02, 1.95 |
| Morning Alcohol | −2.35 (1.70) | −5.69, 0.98 | 12.22 (10.45) | −8.27, 32.72 | ||||
| Afternoon Alcohol | −1.39 (0.59) | −2.56, −0.23* | ||||||
| Night Alcohol | −11.58 (9.51) | −30.23, 7.06 | −4.77 (2.48) | −9.64, 0.10 | ||||
| Cannabis | 17.70 (10.11) | −2.10, 37.51 | −0.79 (0.43) | −1.64, 0.05 | 6.76 (2.53) | 1.78, 11.73** | ||
| Other Drugs | −99.96 (32.47) | −160.59, −33.32** | −85.36 (28.76) | −141.73, −28.98** | −13.53 (7.50) | −28.24, 1.18 | ||
| Nicotine | −7.97 (16.49) | −40.28, 24.34 | −13.19 (14.71) | −42.03, 15.64 | ||||
| Caffeine | 12.13 (7.78) | −3.11, 27.37 | ||||||
p ≤ 0.05.
p ≤ 0.01.
Table 4.
Unstandardized parameters estimated by the linear mixed-effect model of the relationship between alcohol and other substance use and Fitbit sleep stages measures (N = 1,459 obs. from 55 participants)
| Light Sleep Duration | Deep Sleep Duration | REM Sleep Duration | ||||
|---|---|---|---|---|---|---|
| Predictors | b (SE) | 95% CI | b (SE) | 95% CI | b (SE) | 95% CI |
| Intercept | 248.14 (6.31) | 235.77, 260.50 | 79.91 (1.83) | 76.32, 83.51 | 89.13 (2.23) | 84.57, 93.69 |
| Sex (F-M) | 7.43 (12.73) | −17.53, 32.38 | −1.41 (3.71) | −8.68, 5.86 | 10.55 (4.69) | 1.35, 19.74* |
| Age | 7.09 (2.96) | 1.29, 12.90* | −0.05 (0.86) | −1.75, 1.65 | −2.60 (1.10) | −4.76, −0.44* |
| BMI | −1.40 (1.48) | −4.31, 1.51 | 0.26 (0.43) | −0.58, 1.11 | −0.01 (0.55) | −1.08, 1.06 |
| Alcoholic Drink/Day (within) | 0.79 (1.20) | −1.56, 3.15 | 0.93 (0.67) | −0.37, 2.24 | 0.41 (0.65) | −0.86, 1.69 |
| Alcoholic Drink/Day (between) | −7.02 (6.04) | −18.85, 4.80 | 0.51 (1.83) | −3.08, 4.11 | −2.07 (2.31) | −6.61, 2.46 |
| Morning Alcohol | −18.51 (14.80) | −47.52, 10.50 | ||||
| Afternoon Alcohol | ||||||
| Night Alcohol | −4.66 (2.92) | −10.39, 1.07 | ||||
| Other Drugs | −14.58 (9.54) | −33.28, 4.12 | ||||
| Caffeine | 5.74 (2.97) | −0.07, 11.56 | ||||
p ≤ 0.05.
Sleep cardiac activity showed significant associations with alcohol consumption. Specifically, increases in both NREM-HR (AIC = 5962, BIC = 6024, R2 = 0.73, ICC = 0.71) and REM-HR (AIC = 5876, BIC = 5929, R2 = 0.75, ICC = 0.74) were predicted by nighttime alcohol use. Furthermore, increased NREM-HR and REM-HR were also associated with higher BMI (Table 5).
Table 5.
Unstandardized parameters estimated by the linear mixed-effect model of the relationship between alcohol and other substance use and Fitbit sleep cardiac activity measures (N = 1,459 obs. from 55 participants)
| NREM Heart Rate | REM Heart Rate | |||
|---|---|---|---|---|
| Predictors | b(SE) | 95% CI | b(SE) | 95% CI |
| Intercept | 60.49 (1.05) | 58.42, 62.56 | 61.65 (1.09) | 59.51, 63.78 |
| Sex (F-M) | 1.66 (2.11) | −2.47, 5.80 | 1.85 (2.17) | −2.41, 6.11 |
| Age | 0.10 (0.49) | −0.85, 1.07 | 0.09 (0.50) | −0.89, 1.08 |
| BMI | 0.60 (0.25) | 0.12, 1.09* | 0.59 (0.25) | 0.09, 1.08* |
| Alcoholic Drink/Day (within) | 0.07 (0.16) | −0.24, 0.38 | −0.01 (0.15) | −0.31, 0.28 |
| Alcoholic Drink/Day (between) | 1.04 (0.98) | −0.87, 2.96 | 0.58 (1.01) | −1.38, 2.56 |
| Morning Alcohol | −4.03 (2.53) | −8.99, 0.93 | −4.38 (2.42) | −9.13, 0.36 |
| Afternoon Alcohol | −1.48 (0.93) | −3.30, 0.34 | −0.84 (0.89) | −2.58, 0.90 |
| Night Alcohol | 1.44 (0.66) | 0.14, 2.74* | 1.26 (0.63) | 0.02, 2.50* |
| Cannabis | −1.14 (0.72) | −2.55, 0.28 | ||
p ≤ 0.05.
5. Discussion
The current study used a high-density data collection protocol (28 days of remote assessments) via consumer-grade wearable devices and digital surveys, to explore the relationships between alcohol and other substances use and sleep and sleep cardiac function in young healthy adults.
Study outcomes indicated that, in young adults, drinking alcohol up to low/moderate doses (below NIAA definition of heavy drinking, i.e., ≥ 4 drinks/day for women and ≥ 5 drinks/day for men [41]) and, more generally, any substance use was associated with negative outcomes in sleep behaviors, sleep structure, and sleep HR.
Overall participants reported less than one alcoholic drink consumed each day, with greater within-person variability than between-person variability in both alcoholic and sleep measures. As shown by other studies, besides the specific effect of alcohol consumption on sleep, alcohol variability has been related to higher sleep variability [42-43], an index that has been associated with several negative consequences on health, such as increased risk of obesity, metabolic syndrome, diabetes, and mortality.[44-45]
Regarding the relationship between alcohol use and sleep, alcohol use was associated with a delayed sleep period and afternoon alcohol consumption specifically was associated with lower sleep efficiency, denoting a negative effect on objective sleep quality. Considering the young age of our sample, the delayed sleep schedules associated with alcohol use might reflect the tendency to consume these types of substances later in the day, as others have shown [53-54], a tendency associated with other typically youthful nighttime behaviors (e.g., playing video games [55]). Surprisingly, the number of alcoholic drinks consumed or consumption of alcohol at night was not associated with poor sleep outcomes. In contrast, prior studies have linked alcohol consumption at night with poor sleep [7,11-12]. This paucity of evidence linking alcohol and more disrupted sleep should be viewed in light of the following considerations: 1) the effects of alcohol on sleep may be less pronounced in younger age,[7,46] possibly due to their higher total concentration of body water and a more efficient enzymes’ response;[47] 2) differently from PSG studies in which alcohol administration is controlled (as in de Zambotti et al.[23]), here we are not aware of the precise number of drinks consumed right before bedtime; moreover, participants did not report excessive alcohol use, which may have limited the effects on subsequent sleep. Indeed, as reported by PSG studies, negative effects on sleep structure were mainly detected after pre-sleep heavy consumption (e.g., McCullar et al.[48]); 3) The wrist is not the brain: despite some (but not all) controlled laboratory PSG findings indicating that pre-sleep alcohol consumption impacts the subsequent sleep, especially in the first part of the night, sleep estimate from wearables may not have a 1:1 correspondence with PSG-based sleep assessment.[34-36]
When looking at the relation between alcohol and sleep cardiac function, our data showed a positive association between nighttime alcohol use and increased NREM- and REM-HR during sleep. This evidence is consistent with polysomnographic studies that analyzed the impact of nocturnal alcohol on subsequent sleep HR,[8,23] and denotes a possible dose-time-dependent effect of alcohol that future studies should further investigate. Although with lower incidence, even at young ages alcohol disruption of cardiac activity has been associated with several implications, such as cardio-metabolic dysfunction,[49] macro- and microcirculation alterations,[50] and cardiac dysfunctional remodeling.[51] Further studies should evaluate whether sleep cardiovascular restoration is long-term compromised in association with alcohol use, and particularly which specific factors associated with the pattern of alcohol and other substance use (e.g., frequency and amount of drinking, craving processes) are the main determinants of this altered sleep cardiovascular function.
In our study, other than alcohol, we further investigated the relationships between other substances (caffeine, nicotine, cannabis, other drugs), sleep, and sleep HR, but these results should be interpreted cautiously as we only considered whether participants used them over the day (yes/no) but not the specific amount of intake or the timing of use.
Along with alcohol, nicotine use was associated with delayed sleep onset and sleep period start time, in line with its sleep-detrimental stimulant effects.[52] Further studies should better disentangle the effects of substance intake from those of “late activity” on sleep timing, also considering the type of activity the participant engaged in the previous night, which we did not measure. Cannabis also showed a negative effect on sleep, by predicting increased WASO. Studies have shown mixed effects of cannabis on sleep. For example, while Drazdowski and colleagues found an association between cannabis consumption, reduced sleep efficiency, and daytime functioning [22], Graupensperger and colleagues reported that the use of cannabis as a sleep aid was not associated with sleep issues or daytime dysfunctions in young adults. [20] Other drug consumption was generally associated with a reduction of sleep duration, showing negative associations with sleep midpoint, sleep period time, and total sleep time. Although we did not specifically record the type of drug used, these findings depict a general negative effect of drugs on sleep. However, it should be considered that the frequency of other drug use was very low in our sample, and this might have biased the resulting relationships with sleep measures.
A few study limitations should be considered. Given the observational/correlational nature of the study, even though we hypothesized an association between prior substance use and subsequent sleep, we cannot draw a conclusion that these factors are causally linked. Other limitations concern the type of data collected from daily surveys. Specifically, we collected the number of alcoholic drinks consumed each day, which is an approximate measurement unit and does not consider the effective percentage of alcohol intake). Conversely, for other substances (caffeine, nicotine, cannabis, and other drugs) we only derived whether the participants used them or not for each day of data collection. In this regard, future studies conducted with wearable assessment should investigate the effects of other substances, possibly adopting an objective measure of their intake across the day. This could also be useful to evaluate further the hypothesized time-related effects of alcohol and substances on sleep. Future research should also better investigate the potential cumulative effects of consuming alcohol or other substances at multiple time windows (e.g., both afternoon and evening) vs. only once per day (e.g., afternoon only). Indeed, binge drinking has been related to more negative consequences on sleep [56]. That said, our study relies on different strengths. Here we conducted a deeper evaluation by collecting data at high granularity with both passive-wearable (Fitbit) and daily survey measurements. This approach not only enables the calculation of average measures of sleep and HR but also to test the possible variations at the within-person level. Therefore, this type of methodological approach provides a more detailed analysis of the possible effects of substances in the short term and should be preferred in cases where it is necessary to conduct longitudinal investigations, not easily implemented in a laboratory setting.
Conclusions
As one of the first studies to analyze intensive longitudinal data from wearable devices and daily survey administration in young adults, we further demonstrated the importance of monitoring the effects on sleep of alcohol and substance consumption in this age group, as well as the benefits of using daily EMA. Indeed, our data showed that in young adults, substance consumption was negatively associated with delays in sleep timing and poorer sleep structure, with a specific effect of alcohol on sleep HR. Future studies should investigate the possible long-term consequences on sleep and cardiovascular health of alcohol and other substance consumption in young adults.
Funding Source.
This study was supported by the National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) grant AA021696 (FCB, MdZ). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Financial arrangements. None.
Non-financial interests. The authors declare no conflict of interest related to the current work. MdZ and FCB have received research funding unrelated to this work from Noctrix Health Inc. and Verily Life Science LLC. MdZ has received research funding unrelated to this work from Honda Motor Co., Ltd. MdZ is a co-founder and Chief Scientific Officer at Lisa Health Inc. MdZ and FCB have ownership of shares in Lisa Health Inc.
Data Availability.
The analysis is based on the data release NCANDA_RELEASE_BASE_REDCAP_MEASUREMENTS_V07 made accessible to the public according to the NCANDA Data Distribution [X]. X: National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA). Available online: https://www.niaaa.nih.gov/national-consortium-alcohol-and-neurodevelopment-adolescence-ncanda (accessed on 7 July 2022). The data release was distributed according to the NCANDA Data Distribution agreement (https://www-niaaa-nih-gov.laneproxy.stanford.edu/ncanda-data-distribution-agreement), whose collection and distribution were supported by National Institutes of Health, USA funding [National Consortium on Alcohol and Neurodevelopment in Adolescence grant numbers: AA021697, AA021695, AA021692, AA021696, AA021681, AA021690, AA021691].
High-density aggregated measures of Fitbit and mNCANDA App data will be available upon reasonable request to the corresponding author.
References
- 1.SAMHSA: 2022 National Survey on Drug Use and Health (NSDUH) Releases. https://www.samhsa.gov/data/release/2022-national-survey-drug-use-and-health-nsduh-releases (Accessed January 2024). [Google Scholar]
- 2.Mahoney CR, Giles GE, Marriott BP, et al. Intake of caffeine from all sources and reasons for use by college students. Clinical nutrition 2019, 38(2), 668–675. [DOI] [PubMed] [Google Scholar]
- 3.Lone A, Alnawah AK, Hadadi AS, et al. Coffee Consumption Behavior in Young Adults: Exploring Motivations, Frequencies, and Reporting Adverse Effects and Withdrawal Symptoms. Psychology Research and Behavior Management 2023, 3925–3937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Riera-Sampol A, Rodas L, Martínez S, et al. Caffeine intake among undergraduate students: sex differences, sources, motivations, and associations with smoking status and self-reported sleep quality. Nutrients 2022, 14(8), 1661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Biddinger KJ, Emdin CA, Haas ME, et al. Association of Habitual Alcohol Intake With Risk of Cardiovascular Disease. JAMA network open 2022, 5(3), e223849. 10.1001/jamanetworkopen.2022.3849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.CDC Centers for Disease Control and Prevention: Deaths from Excessive Alcohol Use in the United States. https://www.cdc.gov/alcohol/features/excessive-alcohol-deaths.html Accessed December 2023. [Google Scholar]
- 7.Feige B, Gann H, Brueck R, et al. Effects of alcohol on polysomnographically recorded sleep in healthy subjects. Alcoholism: Clinical and Experimental Research 2006, 30(9), 1527–1537. [DOI] [PubMed] [Google Scholar]
- 8.Sagawa Y, Kondo H, Matsubuchi, et al. Alcohol has a dose-related effect on parasympathetic nerve activity during sleep. Alcoholism: Clinical and Experimental Research 2011, 35(11), 2093–2100. [DOI] [PubMed] [Google Scholar]
- 9.Arnedt JT, Rohsenow DJ, Almeida AB, et al. Sleep following alcohol intoxication in healthy, young adults: effects of sex and family history of alcoholism. Alcoholism: Clinical and Experimental Research 2011, 35(5), 870–878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pabon E, Greenlund IM, Carter JR, et al. Effects of alcohol on sleep and nocturnal heart rate: Relationships to intoxication and morning-after effects. Alcoholism: clinical and experimental research 2022, 46(10), 1875–1887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Van Reen E, Tarokh L, Rupp T et al. Does timing of alcohol administration affect sleep?. Sleep 2011, 34(2), 195–205 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yules RB, Freedman DX, Chandler KA. The effect of ethyl alcohol on man's electroencephalographic sleep cycle. Electroencephalogr Clin Neurophysiol 1966, 20: 109–111. [DOI] [PubMed] [Google Scholar]
- 13.Gardiner C, Weakley J, Burke, et al. The effect of caffeine on subsequent sleep: A systematic review and meta-analysis. Sleep Medicine Reviews 2023, 101764. [DOI] [PubMed] [Google Scholar]
- 14.Jaehne A, Loessl B, Bárkai Z, et al. Effects of nicotine on sleep during consumption, withdrawal and replacement therapy. Sleep medicine reviews 2009, 13(5), 363–377. [DOI] [PubMed] [Google Scholar]
- 15.Jaehne A, Unbehaun T, Feige, et al. Sleep changes in smokers before, during and 3 months after nicotine withdrawal. Addiction Biology 2015, 20(4), 747–755. [DOI] [PubMed] [Google Scholar]
- 16.Babson KA, Sottile J, Morabito D. Cannabis, cannabinoids, and sleep: a review of the literature. Current psychiatry reports 2017, 19, 1–12. [DOI] [PubMed] [Google Scholar]
- 17.Fucito LM, Bold KW, Van Reen E, et al. Reciprocal variations in sleep and drinking over time among heavy-drinking young adults. J Abnorm Psychol. 2018, 127(1):92–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Miller MB, Freeman LK, Deroche CB, et al. Sleep and alcohol use among young adult drinkers with insomnia: A daily process model. Addict Behav. 2021, 119:106911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bell KA, Coleman E, Cooke BG, et al. Recreational cannabis use is associated with poorer sleep outcomes in young adult African Americans. Addictive Behaviors. 2022, Nov 1;134:107399. [DOI] [PubMed] [Google Scholar]
- 20.Graupensperger S, Fairlie AM, Vitiello MV, et al. Daily-level effects of alcohol, cannabis, and simultaneous use on young adults’ perceived sleep health. Sleep 2021, 44(12), zsab187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kang SG, Bae SM. The effect of cigarette use and dual-use on depression and sleep quality. Substance Use & Misuse 2021, 56(12), 1869–1873. [DOI] [PubMed] [Google Scholar]
- 22.Drazdowski TK, Kliewer WL, Marzell M. College students’ using cannabis to sleep relates to frequency, problematic use, and sleep problems. Journal of American College Health 2021, 69(1), 103–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.de Zambotti M, Forouzanfar M, Javitz, et al. Impact of evening alcohol consumption on nocturnal autonomic and cardiovascular function in adult men and women: a dose–response laboratory investigation. Sleep 2021, 44(1), zsaa135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Choi JB, Lee YJG, Jeong DU. Transdermal nicotine patch effects on EEG power spectra and heart rate variability during sleep of healthy male adults. Psychiatry investigation 2017, 14(4), 499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bonnet M, Tancer M, Uhde T, et al. Effects of caffeine on heart rate and QT variability during sleep. Depression and anxiety 2005, 22(3), 150–155. [DOI] [PubMed] [Google Scholar]
- 26.Brannon EE, Cushing CC, Crick CJ, et al. The promise of wearable sensors and ecological momentary assessment measures for dynamical systems modeling in adolescents: a feasibility and acceptability study. Translational behavioral medicine 2016, 6(4), 558–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Menghini L, Yuksel D, Prouty D, et al. Wearable and mobile technology to characterize daily patterns of sleep, stress, presleep worry, and mood in adolescent insomnia. Sleep Health 2023, 9(1), 108–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bodin F, McIntyre KM, Schwartz, et al. The association of cigarette smoking with high frequency heart rate variability: An ecological momentary assessment study. Psychosomatic medicine 2017, 79(9), 1045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Brown SA, Brumback T, Tomlinson K, et al. The National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA): A Multisite Study of Adolescent Development and Substance Use. J Stud Alcohol Drugs. 2015, Nov;76(6):895–908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. Journal of biomedical informatics 2009, 42(2), 377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Harris PA, Taylor R, Minor BL, et al. The REDCap consortium: building an international community of software platform partners. Journal of biomedical informatics 2019, 95, 103208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Buysse DJ, Reynolds CF, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry research 1989, 28(2), 193–213. [DOI] [PubMed] [Google Scholar]
- 33.Cummins KM, Brumback T, Chung T, et al. Acceptability, Validity, and Engagement With a Mobile App for Frequent, Continuous Multiyear Assessment of Youth Health Behaviors (mNCANDA): Mixed Methods Study. JMIR Mhealth Uhealth. 2021, Feb 10;9(2):e24472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Haghayegh S, Khoshnevis S, Smolensky MH, et al. Accuracy of wristband Fitbit models in assessing sleep: systematic review and meta-analysis. Journal of medical Internet research 2019, 21(11), e16273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.de Zambotti M, Goldstone A, Claudatos S, et al. A validation study of Fitbit Charge 2™ compared with polysomnography in adults. Chronobiology international 2019, 35(4), 465–476. [DOI] [PubMed] [Google Scholar]
- 36.Lim SE, Kim HS, Lee SW, et al. Validation of Fitbit Inspire 2TM Against Polysomnography in Adults Considering Adaptation for Use. Nature and Science of Sleep 2023, 59–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.de Zambotti M, Goldstein C, Cook J, et al. State of the Science and Recommendations for Using Wearable Technology in Sleep and Circadian Research. Sleep. In press. [DOI] [PubMed] [Google Scholar]
- 38.Menghini L, Cellini N, Goldstone A, et al. A standardized framework for testing the performance of sleep-tracking technology: step-by-step guidelines and open-source code. Sleep 2021, 44(2), zsaa170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Bates D, Mächler M, Bolker B, et al. Fitting linear mixed-effects models using lme4. arXiv preprint 2014, arXiv:1406.5823. [Google Scholar]
- 40.The jamovi project 2023. jamovi (Version 2.3) [Computer Software]. Retrieved from https://www.jamovi.org.
- 41.NIH National Institutes of Health: Alcohol’s effects on health. https://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/moderate-bingedrinking#:~:text=Heavy%20Alcohol%20Use%3A,or%20more%20drinks%20per%20week Accessed January 2024.
- 42.Pasch KE, Latimer LA, Cance JD, et al. Longitudinal bi-directional relationships between sleep and youth substance use. Journal of youth and adolescence 2012, 41, 1184–1196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Heacock RM, Capodilupo ER, Czeisler MÉ, et al. Sleep and alcohol use patterns during federal holidays and daylight saving time transitions in the United States. Frontiers in physiology 2022, 13, 884154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Roenneberg T, Allebrandt KV, Merrow M, et al. Social jetlag and obesity. Current Biology 2012, 22(10), 939–943. [DOI] [PubMed] [Google Scholar]
- 45.Koopman ADM, Rauh SP, van ‘t Riet E, et al. The Association between Social Jetlag, the Metabolic Syndrome, and Type 2 Diabetes Mellitus in the General Population: The New Hoorn Study. J. Biol. Rhythms 2017, 32, 359–368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Van Reen E, Tarokh L, Rupp TL, et al. Does timing of alcohol administration affect sleep?. Sleep 2011, 34(2), 195–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Barry KL, Blow FC. Drinking over the lifespan: Focus on older adults. Alcohol research: current reviews 2016, 38(1), 115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.McCullar KS, Barker DH, McGeary JE, et al. Altered sleep architecture following consecutive nights of pre-sleep alcohol. Sleep 2024, zsae003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Du D, Bruno R, Dwyer, et al. Associations between alcohol consumption and cardio-metabolic risk factors in young adults. European journal of preventive cardiology 2017, 24(18), 1967–1978. [DOI] [PubMed] [Google Scholar]
- 50.Piano MR. Alcohol’s effects on the cardiovascular system. Alcohol research: current reviews 2017, 38(2), 219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Rodrigues P, Santos-Ribeiro S, Teodoro T, et al. Association between alcohol intake and cardiac remodeling. Journal of the American College of Cardiology 2018, 72(13), 1452–1462. [DOI] [PubMed] [Google Scholar]
- 52.Singh N, Wanjari A, Sinha AH. Effects of Nicotine on the Central Nervous System and Sleep Quality in Relation to Other Stimulants: A Narrative Review. Cureus 2023, 15(11). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Singleton RA, & Wolfson AR. Alcohol consumption, sleep, and academic performance among college students. Journal of studies on alcohol and drugs 2009, 70(3), 355–363. [DOI] [PubMed] [Google Scholar]
- 54.Digdon N, & Landry K. University students' motives for drinking alcohol are related to evening preference, poor sleep, and ways of coping with stress. Biological Rhythm Research 2013, 44(1), 1–11. [Google Scholar]
- 55.De Rosa O, Baker FC, Barresi G, et al. Video gaming and sleep in adults: A systematic review. Sleep Medicine 2024. [DOI] [PubMed] [Google Scholar]
- 56.Patrick ME, Griffin J, Huntley ED, & Maggs et al. Energy drinks and binge drinking predict college students’ sleep quantity, quality, and tiredness. Behavioral sleep medicine 2018, 16(1), 92–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The analysis is based on the data release NCANDA_RELEASE_BASE_REDCAP_MEASUREMENTS_V07 made accessible to the public according to the NCANDA Data Distribution [X]. X: National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA). Available online: https://www.niaaa.nih.gov/national-consortium-alcohol-and-neurodevelopment-adolescence-ncanda (accessed on 7 July 2022). The data release was distributed according to the NCANDA Data Distribution agreement (https://www-niaaa-nih-gov.laneproxy.stanford.edu/ncanda-data-distribution-agreement), whose collection and distribution were supported by National Institutes of Health, USA funding [National Consortium on Alcohol and Neurodevelopment in Adolescence grant numbers: AA021697, AA021695, AA021692, AA021696, AA021681, AA021690, AA021691].
High-density aggregated measures of Fitbit and mNCANDA App data will be available upon reasonable request to the corresponding author.