Regularly Drinking Alcohol Before Sexual Activity in a Nationally Representative Sample: Prevalence, Sociodemographics, and Associations With Psychiatric and Substance Use Disorders

Abstract

Objectives. We addressed regular drinking before sex and its associated risk factors.

Methods. From the wave 2 National Epidemiologic Survey on Alcohol and Related Conditions, a nationally representative adult US sample (fielded 2004–2005), we determined the 12-month prevalence of regularly drinking alcohol before sexual activity. Among 17 491 sexually active drinkers, we determined the sociodemographic, psychiatric, and substance use correlates of regularly drinking before sex.

Results. Regular presex drinking’s 12-month prevalence was 1.8%. Significant bivariate sociodemographic correlates were age, gender, race/ethnicity, education, family income, marital status, and employment status. Generalized anxiety disorder and alcohol dependence were associated with significantly increased odds of being a regular presex drinker after controlling for covariates.

Conclusions. We estimate that 4.3 million American adults are regular presex drinkers. Future research should examine this public health issue at the population level, with particular focus on pathways that link it to psychopathology.

Alcohol consumption before sexual activity is associated with many negative health outcomes. First, individuals who drink before having sex are more likely to engage in risky sexual behaviors.^1–4 For instance, studies have demonstrated that presex drinkers are 3 to 5 times more likely than are others to engage in unprotected intercourse.^5,6 Condom use becomes inconsistent after consuming alcohol,⁷ with 1 study finding that 16% of individuals used condoms less often after drinking.⁸ As blood alcohol content levels rise, intentions to engage in unprotected sex also increase.⁹ Other risky behaviors associated with drinking alcohol before sex include having more casual sexual partners and a greater number of sexual partners in general.¹⁰ Second, even when individuals attempt to mitigate risk by engaging in safer sex practices, drinking before sex is associated with lowered effectiveness of these practices. Condom failure is more likely among heterosexuals if presex drinking occurs.¹¹ Taken together, these results indicate that the consumption of alcohol before sexual activity is related to (1) increased risky behavior, and (2) fewer and less-effective protective behaviors.

Consistent with these findings, the consumption of alcohol before sexual activity is associated with higher risk of sexually transmitted infections (STIs). In sexual encounters between heterosexual partners and men who have sex with men (MSM), increased rates of HIV seroconversion and the development of STIs have been documented in presex drinkers.^7,12–14 One study found a 1.5 times increased risk of having HIV or an STI among men who visited female sex workers and engaged in presex drinking.⁵ In MSM, alcohol use before sex has been strongly linked to HIV infection, and the attributable fraction of HIV seroincidence for substance use before sex has been estimated at 29.0.¹³

Because of its associations with disease incidence and unsafe sexual practices, regularly drinking before sexual activity is a significant public health concern. However, no nationally representative studies of this phenomenon have been conducted among US adults. Studies of presex drinking tend to focus on more circumscribed issues and groups, most notably HIV/AIDS transmission in sub-Saharan Africa,^7,11,12,15 MSM populations,^{1,2,4,13,14,16} and adolescents and young adults.^{3,8,10,17–19} Although focus on these issues and groups is important, the literature lacks a broad characterization of regularly drinking before sex in the US adult population as a whole.

We focused on 2 important topics for understanding and reducing the risk of regularly drinking before sexual activity. First, an estimate of the prevalence of regularly drinking before sex is needed to indicate the extent of this behavior in the United States. To our knowledge, this topic has yet to be addressed. Second, determining the characteristics of individuals who regularly drink before sex would clarify the pathways that lead individuals to engage in this behavior, eventually informing the development of prevention efforts. In addition to sociodemographic descriptors, the potential associations between regularly drinking alcohol before sexual activity and psychiatric and substance use disorders appear to be key considerations.

For example, some individuals’ regular presex drinking may reflect underlying substance use disorders or the broader range of externalizing disorders, traits, or associated features (e.g., sensation seeking).^20–22 Other individuals with mood or anxiety disorders might regularly drink before sex to overcome fear associated with sexual acts or with their performance, to relax physically, and to alleviate negative emotions.^23,24 Although presex drinking has been documented in individuals living with severe and persistent mental illness (e.g., psychosis),²⁵ we are aware of no studies that have attempted to link psychiatric and substance use disorders to regularly drinking before sex in the general population.

Considering the significant public health implications of regularly consuming alcohol before sexual activity, the lack of answers to fundamental questions relating to this behavior is striking. We examined regular consumption of alcohol before sexual activity in a nationally representative sample of adults. We estimated the prevalence and sociodemographic correlates of regular presex drinking. We then examined the associations between various forms of psychiatric and substance use disorders and regular presex drinking.

METHODS

We analyzed data from wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). The NESARC comprised 2 waves of face-to-face interviews in a US representative sample. NESARC wave 1 (n = 43 093; response rate: 81.0% of those eligible) was fielded from 2001 to 2002, and wave 2 (n = 34 653; 86.7% of eligible original sample; 70.2% cumulative response rate) was fielded from 2004 to 2005. The NESARC sample is representative of the age, racial/ethnic, and gender distribution of the United States as determined by the 2000 US Census.

Fifty-eight percent of the wave 2 participants were women, and ages ranged from 20 years to older than 90 years. Participants self-identified into race/ethnicity categories of non-Hispanic White (70.9%), Hispanic/Latino (11.6%), non-Hispanic Black (11.1%), Asian/Pacific Islander (4.3%), and Native American (2.2%). Black, Hispanic/Latino, and young adults (18–24 years) were oversampled in wave 1, with data adjusted for oversampling and nonresponse.

Assessment

Alcohol consumption before sexual activity.

Drinking before sex was assessed at wave 2 only, via the question “During the last 12 months, how often did you drink alcohol before having sex?” with response options of “never” (n = 7078), “rarely” (n = 5250), “sometimes” (n = 4548), “most of the time” (n = 464), “always” (n = 151), or “don’t know” (n = 86; excluded from all analyses). Because we focused on drinking alcohol before sexual activity on a regular basis, we dichotomized responses into 2 levels: those who regularly drank alcohol before sexual activity (“most of the time” or “always”; n = 615) versus those who did not.²⁶

Multinomial models using sociodemographics, psychiatric disorders, and substance use disorders to predict regular presex drinker frequency when it was treated as ordinal (i.e., all 5 frequency levels) indicated a notable disjunction in odds ratios (ORs) between “sometimes” and “most of the time.” Furthermore, in those analyses, there was a notable similarity of ORs when predicting outcome frequencies of (1) “never,” “rarely,” and “sometimes” and (2) “most of the time” and “always.” Although not reported here, for brevity, these findings empirically supported our dichotomization of the outcome.

Psychiatric and substance use disorders.

We assessed Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV)²⁷ disorders with the Alcohol Use Disorder and Associated Disabilities Interview Schedule—DSM-IV Version (AUDADIS-IV), a structured interview designed for trained lay interviewers.^28,29 We examined lifetime diagnoses of major depressive, dysthymic, generalized anxiety, panic, posttraumatic stress, and antisocial personality disorders; social and specific phobias; and alcohol, cannabis, nicotine, and other drug dependence.

The other drug dependence variable collapsed a variety of relatively infrequent forms of drug dependence (amphetamines, cocaine, inhalants or solvents, hallucinogens, heroin, opioids, sedatives, tranquilizers, and any other drug) into a single variable with adequate variance for analysis, congruent with previous research in this sample.^30,31 AUDADIS-IV diagnostic reliabilities are generally good for psychiatric disorders (κ = 0.42–0.64) and are good to excellent for substance use disorders (κ = 0.67–0.84).^28,29,32 The AUDADIS-IV has notable advantages over some other interviews, such as assessment of clinically significant impairment and distress.³²

Drinking frequency.

Along with various sociodemographic variables (Table 1), we included past-year frequency of drinking alcohol as a covariate in adjusted models. We assessed this as a 10-point Likert-type scale with values indicating past-year drinking frequencies of 1 to 2 times, 3 to 6 times, 7 to 11 times, monthly, 2 to 3 times monthly, weekly, 2 times weekly, 3 to 4 times weekly, nearly every day, and every day.

TABLE 1—

Characteristics of Sexually Active Drinkers by Status of Regularly Drinking Alcohol Before Sexual Activity: National Epidemiologic Survey on Alcohol and Related Conditions, United States, 2004–2005

Variable	Do Not Regularly Drink Alcohol Before Sexual Activity (n = 16 876), % or Mean ±SD	Regularly Drink Alcohol Before Sexual Activity (n = 615), % or Mean ± SD
Main predictors
Psychiatric disorders, lifetime
Major depressive disorder	22.30	30.25
Dysthymic disorder	4.46	7.17
Social phobia	7.17	8.58
Specific phobia	15.48	17.01
Generalized anxiety disorder	7.37	13.64
Panic disorder	7.80	10.37
Posttraumatic stress disorder	9.58	12.15
Antisocial personality disorder	4.65	10.33
Substance use disorders, lifetime
Alcohol dependence	19.73	52.55
Cannabis dependence	2.12	8.14
Nicotine dependence	26.43	44.30
Other drug dependence	2.69	8.91
Covariates
Age, y
18–29	20.73	18.67
30–39	24.23	17.51
40–49	25.23	25.39
≥ 50	29.81	38.41
Race
Non-Hispanic White	73.74	71.60
Non-Hispanic Black	9.44	13.51
Native American	2.05	2.96
Asian/Pacific Islander	3.38	2.77
Hispanic	11.39	9.15
Gender
Male	54.01	69.81
Female	45.99	30.19
US-born
Yes	88.42	91.04
No	11.58	8.96
Urbanicity
Rural	16.21	15.05
Urban	83.79	84.95
Region
Northeast	17.74	19.45
Midwest	18.34	17.70
South	38.57	35.63
West	25.34	27.22
Education
< high school	8.72	12.27
High school graduate	24.45	27.44
Some college	66.82	60.29
Individual income, $
0–19 999	32.66	34.70
20 000–34 999	23.67	22.98
35 000–69 999	29.44	27.88
≥ 70 000	14.23	14.45
Family income, $
0–19 999	11.96	16.52
20 000–34 999	14.94	19.06
35 000–69 999	33.81	32.16
≥ 70 000	39.29	32.26
Marital status
Married	72.34	48.63
Divorced or separated	9.38	23.25
Widowed	1.01	2.51
Never married	17.27	25.61
Employment status
Unemployed	4.43	9.69
Employed	95.57	90.31
Alcohol drinking frequency	5.20 ±2.55	8.52 ±1.58

Note. The sample size was n = 17 491.

We have reported results treating this as a 1 to 10 variable; an alternative scoring, using the number of days drinking per year estimated by the midpoint of each response option range, produced highly similar results.

Statistical Analyses

We conducted analyses using SUDAAN, version 11.0 (RTI International, Research Triangle Park, NC), with Taylor series linearization to account for the complex survey design of the NESARC. We calculated proportions and 95% confidence intervals (CIs) for all covariates. We used bivariate logistic regressions to establish the associations among sexually active drinkers and between all covariates and participant status as a respondent who regularly drank before sex, resulting in ORs. We used multivariable logistic regression to obtain adjusted ORs (AORs). The multivariable model included all variables significant at P < .05 in the bivariate analyses. The AOR values thus represent the unique effect of each covariate on regular presex drinker status, adjusting for all other covariates. There were no missing values.

In addition to the multivariable logistic regression analysis, we examined analogous Poisson and zero-inflated Poisson models. The Akaike information criterion, Bayesian information criterion, and sample size–adjusted Bayesian information criterion all favored the logistic model by far (e.g., Bayesian information criterion values for logistic, Poisson, and zero-inflated Poisson analyses were, respectively, 3926.164, 4050.476, and 4314.247). The 3 models produced the same substantive results. We thus have reported the results of the multivariable logistic regression analysis, but all results are available from N. R. E. on request.

RESULTS

The population-weighted prevalence estimate of regularly consuming alcohol before sexual activity in the full sample of wave 2 respondents (n = 34 653) was 1.8% (2.6% of men; 1.0% of women). To estimate the weighted prevalence in sexually active, non–alcohol-abstaining individuals only, we also excluded participants who had not had sex or consumed alcohol in the past year (n = 17 076) and those who did not answer the question about regular alcohol consumption before sexual activity (n = 86).

This resulted in a final sample of 17 491 sexually active drinkers, of whom 615 individuals regularly consumed alcohol before sexual activity and 16 876 individuals did not (i.e., responses of “never,” “rarely,” and “sometimes”). This yielded a weighted 12-month prevalence estimate of 3.3% (4.2% of men; 2.2% of women) for regularly consuming alcohol before sexual activity among sexually active, alcohol-consuming individuals.

Regular Presex Drinking Correlates

Table 1 presents sociodemographic characteristics and disorder prevalence rates of the sexually active drinker sample stratified by status of regularly drinking before sexual activity. Regular presex drinkers were primarily male (69.8%) and unmarried (51.4%). Bivariate analyses (Table 2) indicated that those who did and did not endorse regularly drinking alcohol before sex differed significantly (P < .05) by age, race/ethnicity, gender, education, family income, marital status, and employment status.

TABLE 2—

Associations Between Regularly Drinking Alcohol Before Sexual Activity and Psychiatric and Sociodemographic Predictors Among Sexually Active Drinkers: National Epidemiologic Survey on Alcohol and Related Conditions, United States, 2004–2005

Variable	OR (95% CI)	AORa (95% CI)
Main predictors
Psychiatric disorders, lifetime
Major depressive disorder	1.51*** (1.19, 1.91)	1.10 (0.81, 1.49)
Dysthymic disorder	1.65* (1.09, 2.51)	0.98 (0.56, 1.72)
Social phobia	1.12 (0.88, 1.43)
Specific phobia	1.12 (0.88, 1.43)
Generalized anxiety disorder	1.98*** (1.49, 2.64)	1.51* (1.07, 2.12)
Panic disorder	1.37 (0.99, 1.89)
Posttraumatic stress disorder	1.30 (0.99, 1.72)
Antisocial personality disorder	2.36*** (1.72, 3.24)	1.10 (0.74, 1.63)
Substance use disorders, lifetime
Alcohol dependence	4.50*** (3.73, 5.44)	2.35*** (1.85, 2.99)
Cannabis dependence	4.08*** (2.61, 6.38)	1.31 (0.71, 2.41)
Nicotine dependence	2.21*** (1.82, 2.70)	1.19 (0.93, 1.52)
Other drug dependence	3.54*** (2.46, 5.11)	1.19 (0.73, 1.96)
Covariates
Age, y
18–29	0.70* (0.51, 0.96)	0.55* (0.34, 0.90)
30–39	0.56*** (0.43, 0.73)	0.64** (0.47, 0.86)
40–49	0.78 (0.60, 1.01)	0.77 (0.59, 1.01)
≥ 50 (Ref)	1.00	1.00
Race
Non-Hispanic White (Ref)	1.00	1.00
Non-Hispanic Black	1.47** (1.17, 1.86)	1.19 (0.92, 1.55)
Native American	1.49 (0.72, 3.06)	1.22 (0.58, 2.57)
Asian/Pacific Islander	0.84 (0.33, 2.19)	1.09 (0.33, 3.56)
Hispanic	0.83 (0.59, 1.16)	1.26 (0.88, 1.82)
Gender
Male	1.97*** (1.60, 2.42)	1.09 (0.88, 1.36)
Female (Ref)	1.00	1.00
US-born
Yes	1.33 (0.86, 2.05)
No (Ref)	1.00
Urbanicity
Urban	1.09 (0.82, 1.45)
Rural (Ref)	1.00	1.00
Region
Northeast	1.02 (0.77, 1.36)
Midwest	0.90 (0.67, 1.21)
South	0.86 (0.66, 1.12)
West (Ref)	1.00	1.00
Education
< high school	1.56* (1.09, 2.23)	1.26 (0.82, 1.92)
High school graduate	1.24 (0.98, 1.58)	1.14 (0.85, 1.53)
Some college (Ref)	1.00	1.00
Individual income, $
0–19 999	1.05 (0.76, 1.44)
20 000–34 999	0.96 (0.67, 1.36)
35 000–69 999	0.93 (0.66, 1.31)
≥ 70 000 (Ref)	1.00	1.00
Family income, $
0–19 999	1.68*** (1.29, 2.20)	0.78 (0.55, 1.13)
20 000–34 999	1.55** (1.18, 2.05)	0.94 (0.68, 1.31)
35 000–69 999	1.16 (0.91, 1.48)	0.99 (0.75, 1.30)
≥ 70 000 (Ref)	1.00	1.00
Marital status
Married (Ref)	1.00	1.00
Divorced or separated	3.69*** (2.92, 4.66)	3.04*** (2.34, 3.96)
Widowed	3.68*** (2.20, 6.16)	2.43** (1.39, 4.27)
Never married	2.21*** (1.76, 2.77)	2.45*** (1.71, 3.50)
Employment status
Unemployed	2.31*** (1.64, 3.26)	1.37 (0.89, 2.10)
Employed (Ref)	1.00	1.00
Alcohol drinking frequency	2.11*** (1.98, 2.25)	2.04*** (1.90, 2.19)

Note. AOR = adjusted odds ratio; CI = confidence interval; OR = odds ratio. The sample size was n = 17 491.

^aThe model controlled for the main predictors and covariates significant at the bivariate level (i.e., major depressive, dysthymic, generalized anxiety, and antisocial personality disorders; alcohol, cannabis, nicotine, and other drug dependence; age, race, gender, education, family income, marital status, and employment status) and past-year alcohol drinking frequency.

*P < .05; **P < .01; ***P < .001.

The odds of being a regular presex drinker were significantly higher for individuals who were aged 50 years and older (vs younger than 40 years), were non-Hispanic Black (vs non-Hispanic White), were male, were not high school graduates, lived in families with lower income, were unmarried, and were unemployed. Prevalence rates for regular presex drinkers were higher for all forms of psychiatric and substance use disorders. More than half of regular presex drinkers were diagnosed with alcohol dependence (52.6% vs 19.7% of individuals who were not regular presex drinkers). Regular presex drinkers also showed higher rates of dependence on other substances—nicotine (44.3% vs 26.4%), cannabis (8.1% vs 2.1%), and other drugs (8.9% vs 2.7%)—than did the rest of the sample. Regular presex drinkers showed higher average drinking frequency (mean = 8.52, falling between 3–4 times weekly and nearly every day) than did the rest of the sample (mean = 5.20, approximately 2–3 times monthly).

Psychiatric and Substance Use Disorder Predictors

Table 2 presents unadjusted bivariate and adjusted multivariable logistic regression models predicting regular presex drinking. Unadjusted bivariate models indicated that significantly increased odds of being a regular presex drinker were associated with major depressive, dysthymic, generalized anxiety, and antisocial personality disorders as well as all 4 substance use disorders.

We then estimated an adjusted multivariable model predicting regular presex drinker status, including all psychiatric disorders, substance use disorders, and covariates that were significant (P < .05) in the bivariate analyses. Generalized anxiety disorder (AOR = 1.51; 95% CI = 1.07, 2.12) and alcohol dependence (AOR = 2.35; 95% CI = 1.85, 2.99) remained significant predictors of regularly drinking alcohol before sexual activity.

DISCUSSION

This was the first study to our knowledge to investigate regularly drinking alcohol before sexual activity in a nationally representative sample. Prevalence rates among sexually active, alcohol-consuming adults indicated that 3.3% (4.2% of men; 2.2% of women) engaged in this behavior regularly in the previous 12 months. Overall, 1.8% (2.6% of men; 1.0% of women) of the sample reported engaging regularly in alcohol consumption before sex. Sociodemographic correlates suggested that older, lower earning, less-educated, male, non-Hispanic Black, unmarried, and unemployed individuals may be at particular risk, as are individuals with generalized anxiety disorder and alcohol dependence.

Because of a current US Census Bureau population estimate of approximately 317 000 000 individuals, of whom approximately 76% are aged 18 years or older,³³ these results suggest that approximately 4 300 000 adults in the United States regularly drink alcohol before sexual activity. There is reason to believe that the true prevalence may be even higher. Social desirability may have influenced some participants to deny, or underestimate the frequency of, regular presex drinking. Furthermore, the NESARC did not assess individuals who were incarcerated or institutionalized for extended periods of time. This may have resulted in a lower than actual population prevalence rate estimate, because of the histories many of these individuals have of substance use, disinhibited behavior, and other externalizing phenomena.

The Roles of Mental Health and Substance Use

This was the first study to our knowledge to investigate the relations between psychiatric and substance use disorders and regular consumption of alcohol before sexual activity in a nationally representative US sample. Various psychiatric and substance use disorders were significantly associated with regular presex drinker status at the bivariate level. Multivariable models that incorporated these disorders and significant covariates clarified these results. Among the substance use disorders, only alcohol dependence was associated with increased odds of being a regular presex drinker, even after controlling for drinking frequency. The other substance disorders were not significant predictors in the adjusted model. For the most part, lifetime psychiatric disorders were unrelated to regularly drinking alcohol before sexual activity as well, with the exception of generalized anxiety disorder, which was robustly associated with 1.5 times the odds of engaging in regular presex drinking in unadjusted and adjusted models.

Future analyses will be necessary to untangle the pathways by which generalized anxiety disorder and alcohol dependence increase the risk of regularly consuming alcohol before sexual activity. In particular, the pathways linking the non–alcohol-related predictor of generalized anxiety disorder to this behavior are unclear and require further study. These pathways may be reciprocal because of previous research indicating that anxiety disorders can both confer risk for problem drinking and be manifestations thereof,³⁴ highlighting the need for longitudinal samples and quantitative methods that can explain these effects.

Our results suggest that treatments addressing generalized anxiety disorder and alcohol dependence may be worthwhile interventions for affecting regular presex drinking behavior. Furthermore, in treatment settings, assessing for regular presex drinking in clinical patient populations is important because of the estimated prevalence rates and because this behavior is unlikely to be a patients’ primary complaint. As a general recommendation, clinicians can inquire about this behavior in high-risk populations—certainly individuals with alcohol dependence, but also patients who are generally anxious, men, unemployed, older, or single.

At least 1 intervention, using motivational interviewing techniques to minimize college student drinking and risky sexual behaviors, has been evaluated. That study’s results suggested that an alcohol risk reduction intervention was effective in lowering alcohol consumption but had no effect on lowering the frequency of unprotected sex.³⁵ Thus, more targeted interventions appear necessary. Such interventions would likely require in-depth functional analysis of the perceived utility of regular presex drinking. Among individuals with generalized anxiety disorder, complex pathways may give rise to this behavior. Some of these individuals may self-medicate with alcohol to “loosen up” or rid themselves of unwanted cognitions. This is a problematic strategy; previous research suggests that drinking alcohol to cope with negative emotions is associated with negative alcohol outcomes.²³

Women with symptoms of worry might use alcohol to reduce related somatic tension, which can be associated with uncomfortable intercourse (e.g., dyspareunia, vaginismus) or inadequate vaginal lubricating response (i.e., female sexual arousal disorder). Men who worry about premature ejaculation might regularly consume alcohol before intercourse because of perceived benefits of delaying orgasm, although such a behavior might actually increase sexual dysfunction (e.g., erectile dysfunction), particularly as the amount of alcohol consumed increases.^36,37 Overall, understanding the pathways leading to regular presex drinking and the functional role this behavior plays will be critical to its reduction. Appropriate intervention will target the behavior’s functions and educate about its dangers while providing less risky alternatives (e.g., relaxation techniques, distress tolerance).

Limitations

Our study has a few limitations. First, we did not address unique pathways by which psychopathology may lead to regular presex drinking. Because of previously documented bidirectional and reciprocal effects that anxiety and problem drinking share,³⁴ the links between generalized anxiety disorder and regularly drinking alcohol before sexual activity are of particular empirical interest. Second, we did not have information on the quantity of alcohol consumed before sexual activity, participants’ level of intoxication, the timing of consumption relative to sexual activities, the nature of the relationship between sexual partners, or contextual factors—all of which should be asked about in future studies.

Third, we derived all information from retrospective self-report, potentially introducing memory and social desirability biases. However, the 12-month timeframe likely mitigated recall problems that might have occurred if a more distal timeframe had been considered. Fourth, clinicians did not make the diagnoses; however, highly trained lay interviewers collected diagnostic information using a reliable and well-validated interview. Finally, the NESARC did not assess negative consequences of participants’ regular alcohol consumption before sexual activity, thus preventing us from distinguishing problematic from nonproblematic regular presex drinking.

Conclusions

At least 4 million adults in the United States regularly drink alcohol before sexual activity. Previous research indicates that such drinking places uninfected individuals at increased risk of becoming infected with HIV and other STIs. Therefore, our findings illustrate that this behavior is a significant public health concern. It is noteworthy that some regular presex drinkers may have already been exposed to STIs. Such possibilities have serious implications for the transmission of disease, particularly because regular presex drinkers engage in more risky sexual behaviors and the effectiveness of the protective behaviors in which they do engage is lessened.

These behavioral patterns place both regular presex drinkers and their sexual partners at risk for numerous serious medical problems. Surprisingly, this phenomenon is understudied at the US general population level, although it has been investigated in multiple studies of specific populations, such as sub-Saharan Africans and MSM. Therefore, our results characterizing the sociodemographic, psychopathological, and substance use factors related to regular presex drinking in the general population take the field a step closer to a broader understanding of this potentially risky behavior.

Not all instances of regularly drinking alcohol before sexual activity are necessarily problematic. For example, individuals in monogamous sexual relationships may regularly consume alcohol before engaging in sexual activity with no discernible negative effects. However, even in monogamous relationships, drinking before sex can be associated with various negative outcomes, including unplanned pregnancy and transmission of existing diseases between discordant partners. Furthermore, whereas regular drinking before sex may be associated with psychopathology, we do not intend to pathologize all regular presex drinking behavior. Instead, our purpose was to gain a better understanding of regular presex drinking and its correlates to broadly inform subsequent investigations because this behavior is often associated with negative outcomes.

Overall, our findings help develop our understanding of regularly drinking alcohol before sexual activity, and they highlight several potential future directions. A better understanding of regular presex drinking at the population level and within various subpopulations (e.g., defined by gender, race/ethnicity, age) is needed. Characteristics of individuals who engage in this behavior should be documented, and their motivations for regular presex drinking should be clarified. Although difficult and resource intensive to implement, critical incident- or multievent-level investigations of specific sexual occasions that were versus those that were not preceded by drinking,^38,39 and their relationships to current versus remitted psychiatric and substance use disorders, would be particularly valuable in elucidating these associations. Finally, interventions for presex drinking can be developed, not only as standalone treatments but also potentially as modules to supplement treatment of substance use disorders and generalized anxiety disorder.