Skip to main content
NCBI home page
Alcohol and Alcoholism (Oxford, Oxfordshire) logoLink to Alcohol and Alcoholism (Oxford, Oxfordshire)
. 2026 Jan 8;61(1):agaf085. doi: 10.1093/alcalc/agaf085

The effects of marital status, partner drinking, and sex differences on the likelihood of remission from alcohol use disorder

Christina Garasky 1,✉, Alexis Duncan 2, Fanghong Dong 3, Gayathri Pandey 4, Jiawen Zhao 5, Jared Balbona 6, Grace Chan 7, Weipeng Kuang 8, Chella Kamarajan 9, Martin Plawecki 10, Victor Hesselbrock 11, Samuel Kuperman 12, Bernice Porjesz 13, Andrey Anokhin 14, Emma Johnson 15, Arpana Agrawal 16, Jessica Salvatore 17, Sally Kuo 18, Kathleen Bucholz 19, Vivia McCutcheon, COGA Collaborators20
PMCID: PMC12779479  PMID: 41501993

Abstract

Objective

Marriage is associated with improved health and alcohol outcomes, yet prior research has often combined diverse non-marital relationships, obscuring whether benefits are unique to marriage or a result of strong interpersonal relationships. This study examines whether marriage, compared to cohabitation, exclusive dating, or being single, is associated with alcohol use disorder (AUD) remission and whether this varies by sex and partner drinking.

Method

Participants were adults from the Collaborative Study on the Genetics of Alcoholism Lifespan Study (n = 1494; 54.4% female) who met criteria for lifetime DSM-5 AUD. Multivariate logistic regression models examined the association between marital status and remission. Subsequent models tested interactions among marital status, partner drinking, and sex. Predicted remission prevalence was estimated using marginal standardization and compared across groups on an additive scale using prevalence differences (PD).

Results

Married individuals had greater predicted prevalence of remission, particularly compared to those who were exclusively dating (PD = 9.2, P = .020). Married females whose partners were average alcohol consumers (PD = 12.20, q = .049) or abstinent (PD = 17.98, q = .017), and married males whose partners had above average alcohol consumption (PD = 25.30, q = .049) had greater predicted prevalence of remission compared to their counterparts who were in non-marital committed relationships.

Conclusion

The findings suggest differences in marital status and relationship factors may indicate increased risk and challenges in sustaining remission and highlight potential caveats of marital status as a protective factor. This emphasizes the need to consider these factors in designing treatment strategies.

Keywords: alcohol use disorder, alcohol use disorder remission, recovery, marital status, sex differences, partner alcohol consumption

Short Summary We examined whether marriage provides unique protective benefits for Alcohol Use Disorder (AUD) remission beyond other marital statuses (cohabitating, exclusively dating, single) and assessed the moderation of sex and partner drinking. Marriage was associated with higher predicted prevalence of remission, particularly compared to those exclusively dating, and effects differed by sex and partner drinking.

Introduction

In 2023, nearly 30 million adults (10.9% of the US population) met criteria for past-year Alcohol Use Disorder (AUD; SAMHSA, 2024), and over 30% of the population is projected to meet criteria for AUD within their lifetime (Grant et al. 2015), costing society billions of dollars a year (Sacks et al. 2015). For many, it is a chronic disorder marked by repeating cycles of remission, defined as the absence of AUD criteria for at least 12 months (American Psychiatric Association. 2013), and followed by a return to symptoms (Grant et al. 2015), with up to 50% achieving remission, but around 68% of those individuals experiencing a return of AUD symptoms (Dawson et al. 2005). AUD is commonly associated with a variety of health and social problems. These include as a greater risk of cancer, liver disease, and pancreatitis, as well as a greater risk for social impairments (Levola et al. 2014). Additionally, AUD is associated with higher comorbidities of other substance use and psychiatric disorders (Bagnardi et al. 2015). These risk factors and repeating cycles of this disorder emphasize the critical need to better understand determinants associated with remission from AUD to ultimately promote enduring remission.

Marital status

Social dynamics, particularly close relationships, are strongly associated with alcohol use. Stressors such as trauma and social isolation can contribute to negative social outcomes (Nolen-Hoeksema 2004) and increase alcohol consumption (Maxwell et al. 2022), especially among females. Conversely, increased social support has been shown to encourage remission (Dawson et al. 2005, McCutcheon et al. 2014). For example, marriage is a protective factor for many health and wellness outcomes (Kiecolt-Glaser and Newton 2001). Married individuals have lower mortality rates (Jia and Lubetkin 2020) and experience lower morbidity from a variety of acute and chronic conditions, such as cancer, heart attacks, and surgery complications (Kiecolt-Glaser and Newton 2001) than unmarried individuals. Moreover, being and staying married are associated with lower rates of substance use and disorders and better recovery outcomes for those with substance use disorders. Specifically, married individuals tend to consume less alcohol than divorced or never married individuals (Simon 2002) and have higher remission rates than those who are single, divorced, or separated (Salvatore et al. 2020).

Although some differences may reflect selection effects, healthier individuals being more likely to marry and remain married (Horwitz et al. 1996), evidence also indicates that marriage can directly and indirectly enhance health (Simon 2002, DeMaris 2018). Marriage may foster social support (Umberson 1992, Seeman 2000, August and Sorkin 2010), which promotes health monitoring, healthy lifestyles, and a greater sense of responsibility and stability (Umberson 1992, August and Sorkin 2010). However, it remains unclear whether these benefits are unique to marriage or extend to other non-marital committed (NMC) relationships, such as cohabiting or exclusively dating partners.

Non-marital committed relationships

Marriage rates have declined in recent decades , with more individuals opting to delay or forgo marriage and instead remain in long term, NMC relationships (Brown et al. 2022). As these relationships become increasingly common, it is important to determine whether they offer similar protections as marriage, particularly for sustaining remission from AUD or whether marriage offers additional benefits beyond strong social relationships. Previous research often focuses on differences between married and unmarried individuals, but overlooks the variety of relational circumstances within the “non-marital” grouping, like those who are cohabitating or dating exclusively. The limited research examining alcohol use outcomes among cohabiting versus married couples has produced mixed findings (Plant et al. 2008, Li et al. 2010, Dinescu et al. 2016), and even less is known about how relationship dynamics influence the maintenance of AUD remission. Strong NMC relationships offer similar benefits to marriage, such as increased social support and emotional closeness (Umberson 1992, Seeman 2000, August and Sorkin 2010), which are associated with successful remission from AUD (Stillman and Sutcliff 2020, Colditz et al. 2023). However, differences associated with the institution of marriage (i.e. legal rights, increased investment, greater societal expectations; Poortman and Mills 2012, Vitali and Fraboni 2022), may ultimately impact remission success (Woods and Avery 2024). Clarifying how marital status relates to remission outcomes can provide deeper insights into the complex interpersonal structural factors that shape recover and help identify contexts linked to risk and resilience in sustaining AUD remission. Thus, the current study investigates the impact that marriage has on remission, by using those in NMC relationships as the comparison group, to see if there are benefits related to remission from AUD that are above and beyond those that one gets from being in a strong high-quality relationship.

Sex and partner behavioral differences

These inconsistencies in the protective benefits in marriage may be a result of sex and partner behavioral difference. The protective benefits of marriage are often more pronounced for males than females (House et al. 1988, Rendall et al. 2011), possibly reflecting females’ greater influence over their partners’ health behaviors (Umberson 1992). Also, males tend to derive a disproportionately higher amount of support from their spouses compared to females (Uhing et al. 2021). Conversely, females are more likely than males to suffer consequences as a result of alcohol use by a spouse or partner (Nayak et al. 2019).

Partner-drinking patterns also play a significant role in shaping an individual’s own drinking patterns. For married individuals, the amount one’s partner drinks is strongly associated with one’s own drinking (Birditt et al. 2019). Individuals, regardless of sex, are less likely to develop AUD if their spouses have no history of AUD (Polenick et al. 2018). Conversely, having a spouse with a lifetime history of AUD significantly increases one’s own risk of developing the disorder (Kendler et al. 2016). Partner-drinking patterns may influence an individual’s own alcohol consumption through shared routines, as spouses, regardless of sex, generally drink together more often than with others (Birditt et al. 2019). Although females tend to have larger social networks, they report drinking more frequently with their spouses than with others, and tend to report having fewer “drinking buddies” compared to males (Leonard & Homish, 2008). This suggests that sex and partner behavioral differences may modify the protective effects of marital status on alcohol consumption outcomes, such as the likelihood of remission from AUD, and highlights the unique role that marriage plays in influencing health and wellness outcomes. Understanding the role of marital status in supporting and sustaining remission from AUD is vital to identifying protective factors that can enhance recovery outcomes.

Current study

The aims of the current study were (i) to determine whether being married, as opposed to being in other types of relationships (i.e. cohabitating, dating exclusively, or single), provides unique protective benefits related to maintaining AUD remission and (ii) among those in committed relationship, to explore how sex and partner drinking moderates the association between marital status and remission. This study extends previous finding of the impact different relationship types have on supporting AUD remission and offers a more comprehensive understanding of the social dynamics that are involved.

Materials and method

Sample

Data were obtained from the most recent wave of Collaborative Study on the Genetics of Alcoholism (COGA) Lifespan Study, which began in 2019 (Agrawal et al. 2023). COGA is a multi-site, longitudinal, high-risk family study designed to assess genetic and environmental influences associated with alcohol use and AUD (Agrawal et al. 2023, Dick et al. 2023). The most recent wave focuses on alcohol use outcomes among midlife and older adults and includes participants from the initial COGA families (Phase I and II) recruited via probands in inpatient and outpatient treatment centers, along with comparison families recruited from the community (e.g. driver’s registries and dental clinics), and their young adult offspring (Prospective Study). Descriptions of these previous studies are provided elsewhere (Dick et al. 2023). For this present study, we used the participants’ most recent responses to the Semi-structured Assessment for the Genetics of Alcoholism, a validated diagnostic interview that captures psychiatric, environmental, and demographic information (Bucholz et al. 1994, Hesselbrock et al. 1999). Participants who met criteria for a lifetime diagnosis of DSM-5 AUD were included in the analysis. Additional analysis explored the influence of partner drinking on remission which included individuals with partner data, excluding single individuals.

Measures

Marital status

Marital status was treated as a categorical independent variable (married, cohabitating, exclusively dating, single). This variable was created using two questions. First, individuals reported being either married, widowed, separated, divorced, never married, or living as married. Then, those who reported being widowed, divorced, or never married were asked if they were currently dating someone exclusively, were engaged, were dating several people, or were not dating. Individuals who reported being married were considered as being married. Individuals who were living together are considered cohabiting. Individuals who reported dating someone exclusively or engaged but not living together were considered to be in an exclusive relationship, and those who reported dating several people or not dating were considered to be single. Those who were separated were excluded from all analysis. Analyses involving partner drinking excluded single individuals, as they did not have a partner from whom partner drinking data could be reported. To increase statistical power and to simplify interpretations of the three-way interaction, those who were exclusively dating and cohabitating were also combined in one analysis (Model 4) to represent those in NMC relationships.

AUD/remission status

Lifetime AUD status was based on “Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition” (DSM-5; American Psychiatric Association [APA] 2013, McCutcheon et al. 2025), defined as two or more criteria occurring in the same 12-month period. We used the 2+ symptom threshold, given that, in addition to being consistent with diagnostic standards, lower severity cases of AUD may be influenced by relationship dynamics and partner alcohol consumption. Remission status was treated as a categorical outcome variable operationalized as having no symptoms other than craving for at least 12 months before the time of interview (APA 2013).

Respondent sex

Sex was treated as a categorical moderator. Individuals reported their sex assigned at birth as male or female. In the current sample, all gender identities aligned with the respondent’s sex assigned at birth (e.g. female-women).

Partner drinking

Partner drinking was treated as a continuous moderator. Respondents were asked “Think about your (spouse’s/partner’s) use of alcohol over the past 6 months. In a TYPICAL week, when (he/she) drank any alcohol in the past 6 months, on how many days do you think (he/she) drank alcohol (including beer, malt liquor, wine, liquor, mixed drinks, etc.)” and “In a TYPICAL week when your (spouse/partner) drank any alcohol in the past 6 months, how many drinks did (he/she) drink on TYPICAL drinking day?” Values from each question were then multiplied to approximate the total number of drinks one’s partner drank in a typical week. For post hoc analyses, partner drinking was estimated at abstinent, the sample mean (average), and one standard deviation above the mean (above average).

Covariates

Sociodemographic factors known to be associated with marriage, partner drinking, and AUD remission were initially included as covariates within the models. These included respondent age, race, ethnicity, education, proband status, AUD lifetime severity, treatment for AUD, childcare, relationship length and satisfaction, partner sex, and history with alcohol. However, to minimize overfitting, only covariates that had main effects that significantly predicted remission were retained in the final model. These included age, race, proband status, AUD lifetime severity, treatment for AUD, childcare, relationship length.

Age was calculated by birth year and confirmed by the respondent’s report. Due to small sample sizes in several racial groups, race was recoded into three categories: White [reference group], Black, and Other Race. The Other race category included Native American, Asian, Pacific Islander, and those who identified as the other race. Participants were identified as a proband [reference group] if they were from the original cohort who were in treatment for alcohol dependence. Lifetime AUD severity was assessed by the reported max number of criteria met throughout one’s lifetime. Treatment for AUD was coded based on responses to the questions “Not counting Alcoholics Anonymous or another self-help group, have you ever been treated for a drinking problem?” and “Did you ever attend a self-help group (like AA, Smart Recovery, Rational Recovery, etc.) for your drinking?” Participants were then grouped based on having received both formal treatment and been a part of a self-help group, having received only formal treatment, having only been a part of self-help groups, or having received neither formal treatment nor been a part of self-help group [reference group]. Child responsibilities were assessed by participants’ responses, yes or no [reference group], to whether they “had ever been responsible for a child for over a year”. Relationship length was assessed by responses to the question “How long have you been with your current partner?”

Data analysis

Data analysis was conducted using STATA 18.5 BE-Basic Edition. Logistic regression analyses were conducted to evaluate associations between marital status and remission from AUD. Model 1 assessed the association of marital status with AUD remission, adjusting for covariates. Subsequent models tested the two-way interactions between marital status and sex (Model 2), and between marital status and partner drinking (Model 3), followed by a final model (Model 4) which examined the three-way interaction among these variables. For all models involving partner drinking (Models 3 & 4), single individuals were excluded from analyses as they do not have a partner of whom to report drinking. We specified the interaction terms between marital status [ref: married], sex [ref: male], and partner drinking, which included all lower order effects, and which were adjusted for the covariates listed previously. All analyses were also clustered by family ID to account for the non-independence of family data.

We estimated predicted probabilities (hereafter, predicted prevalence) and their 95% confidence interval (CI) standardized to the covariate distribution for each relationship status, partner drinking, and sex combination from the logistic regression models, with post hoc tests of pairwise absolute differences in predicted prevalence (predicted prevalence difference [PD]) between each subgroup combination (Muller and MacLehose 2014, Zhong et al. 2022, Duncan et al. 2025). Partner drinking was estimated at abstinent, sample mean (average), and one standard deviation above the mean (above average). To adjust for multiple comparisons, we applied the false discovery rate (FDR) correction to all predicted prevalence differences. The resulting adjusted P-values are represented as q-values.

Results

Marital status and remission

Characteristics of the analytic sample (n = 1494) are shown in Table 1. We first examined whether marital status (married, cohabiting, exclusively dating, single) was associated with sustaining remission from AUD. After adjusting for covariates using logistic regression, predicted remission prevalence differed across relationship groups (see Fig. 1). Married individuals had significantly higher predicted remission prevalence than those who were exclusively dating, with a prevalence difference of 9.2 percentage points (PD = 9.2, 95% CI[1.5, 17.0], P = .020). Although cohabiting (PD = 4.2, P = .374) and single individuals (PD = 3.7, P = .195) also exhibited slightly lower remission prevalence relative to married individuals, these differences were not statistically significant. Additionally, although overall remission rates were similar between those cohabitating and those exclusively dating (see Table 1), a smaller proportion of exclusively dating individuals had remitted after the start of their relationship compared to those cohabitating.

Table 1.

Characteristics of the COGA lifespan study participants who have a lifetime history of AUD by marital status.

N = 1494 Married (n = 723) Living as married (n = 135) Exclusively dating (n = 179) Single (n = 457)
Remission 424 (58.6)a 66 (48.9)b 78 (43.6)b 253 (55.36)a
After (vs before) relationship start 254 (59.8)a 35 (53.03)a 23 (29.48) N/A
AGE 53.11 (13.75) (29–84)a 46.43 (14.37) (27–79)b 46.08 (13.95) (29–83)b 52.69 (14.20) (30–88)a
Length of relationship 23.44 (14.39) 9.58 (8.33) 6.35 (8.55) N/A
Relationship satisfaction (1–5) 4.23 (.65)a 4.17 (.58)a 4.18 (.74)a N/A
Partner drinking (~# of drinks per week) 6.19 (14.29) 6.35 (11.90) 5.7 (11.04) N/A
Partner AUD history 152 (21.02) 35 (25.93) 34 (18.99) N/A
Childcare 548 (75.80) 85 (62.96)a 115 (64.25)a 283 (61.93)a
Max lf AUD sx 6.19 (2.86) 6.69 (2.87)a 6.52 (3.00)a 7.10 (3.00)a
Treatment lf
 None 479 (66.25) 84 (62.22)a 96 (53.63)a 229 (50.11)a
 Self-help only 88 (12.17) 10 (7.41)a 25 (13.97)a 61 (13.35)a
 Prof only 21 (2.90) 5 (3.70)a 10 (5.59)a 21 (4.60)a
 Both 135 (18.67) 36 (26.67)a 48 (26.82)a 146 (31.95)a
Race
 White 625 (86.45) 98 (72.59)a 123 (68.72)a 333 (72.87)a
 Black 65 (8.99) 25 (18.52)a 43 (24.02)a 97 (21.23)a
 Other 33 (4.56) 12 (8.89)a 12 (7.26)a 27 (5.91)a
Sex (% female) 329 (45.50)a 72 (53.33)a 87 (48.60)a 221 (48.36)a
Education 13.83 (2.41) 13.40 (2.57)a 13.29 (2.39)a 13.18 (2.39)a
Case family 610 (84.49) 119 (90.15)a 157 (88.20)a 413 (90.97)a
Marriages
 None N/A 60 (44.44) 0 (0%)a 1 (.22)a
 1 477 (65.98) 32 (23.70) 54 (30.17)a 148 (32.39)a
 2+ 246 (34.02) 42 (31.85) 125 (69.83)a 308 (67.4)a
Divorces
 None 483 (66.8) 66 (48.89) 5 (2.79) 29 (6.35)
 1 170 (23.51) 30 (22.22) 52 (29.05) 136 (29.76)
 2+ 70 (9.68) 39 (28.89) 122 (68.16) 292 (63.89)
Open in a new tab

Notes. Values are presented as percentages (%) or as the mean (M) with standard deviation (SD), where applicable. Education reflects the highest level of schooling completed, with a value of 17 indicating graduate-level education (e.g. M.A., M.S., J.D., M.D., Ph.D.). Matching superscript letters indicate that values that do not differ significantly from each other using chi-square or ANOVA.

Figure 1.

Figure 1

Open in a new tab

Predicted prevalence of remission from AUD by marital status. Notes. Calculated from a multivariable logistic regression model with marital status and covariates. *P < .05.

Marital status, sex, partner drinking, and remission

Two-way interactions

Next, we tested two separate two-way logistic regression model adjusted for covariates, investigating the effects of sex and partner drinking on the probability of sustaining remission from AUD (marital status x sex; marital status x partner drinking). Neither two-way interaction was significant (marital status x sex; marital status x pattern drinking; P > .05). No conditional main effects of sex or marital status were observed (all P’s > .05). However, conditional main effects were observed for partner drinking, such that increased partner drinking reduced the predicted prevalence of remission (P < .001).

Three-way interaction of non-marital committed relationships

We then further tested a three-way interaction among marital status, partner drinking, and sex (Model 4). To increase our power, we compared married individuals to those in NMC relationships, combining those who were cohabiting and those who were exclusively dating. In this logistic regression model that was adjusted for covariates, the three-way interaction was significant (P = .017). The conditional main effect of marital status was not significant (P > .05). However, the conditional main effects of sex and partner drinking were significant, such that being male (P = .038) and increased partner drinking (P = .002) reduced the predicted prevalence of remission. The two-way interactions between marital status and sex, and sex and partner drinking were not significant (all P’s > .05). However, the two-way interaction between marital status and sex was significant (P = .007).

Figure 2 displays the predicted prevalences of remission from AUD by each grouping of marital status, sex, and partner drinking, which varied by subgroup. When partners did not drink, married females had significantly higher remission rates than their unmarried female counterparts, with a predicted prevalence difference of 18.8 percentage points (PD = 18.8; 95%CI[6.6, 31.1], P = .002, q = .017). Conversely, prevalence rates of remission did not significantly differ by marital status for males (P > .05). When partners drank at average levels, married females still had significantly higher remission rates than their unmarried female counterparts, with a predicted prevalence difference of 13.6 percentage points (PD = 13.6; 95%CI[3.3, 23.9], P = .01, q = .049). Again, prevalence rates of remission did not significantly differ by marital status for males (P > .05). When partners drank at above average levels, prevalence rates of remission did not significantly differ by marital status for females (P > .05). However, married males had significantly higher remission rates than their unmarried male counterparts, with a predicted prevalence difference of 23.0 percentage points (PD = 23.0; 95%CI[2.9, 43.2], P = .025, q = .05). When comparing between males and females at similar partner drinking levels, no sex differences were observed (q > .05).

Figure 2.

Figure 2

Open in a new tab

Predicted prevalence of remission from AUD by marital status, sex, and partner drinking. Notes. Calculated from a multivariable logistic regression model with the following independent variables: Marital status, partner drinking, assigned sex, covariates, the two-way interactions between marital status and partner drinking, marital status and sex, partner drinking and sex, and the three-way interaction between marital status, partner drinking, and sex. For post hoc analyses, partner drinking was estimated at abstinent, at the sample mean (average), and one standard deviation above the mean (above average).

Three-way interaction of separated marital status groups

Though low in power, separating, cohabitating and exclusively dating couples showed similar trends to the combined NMC relationship grouping when compared to married couples (Fig. 3). All conditional main effects and lower order interactions were similar to the prior combined marital status three-way interaction model. Predicted prevalence of remission still differed between each marital-sex subgroup. When partners did not drink, married females had significantly higher remission rates than their unmarried female counterparts, with predicted prevalence differences of 18.14 percentage points (PD = 18.14; 95%CI[2.5, 33.8], P = .01, q = .033), compared to those exclusively dating, and a unadjusted significant prevalence difference of 20.1 percentage points (PD = 20.13; 95%CI[4.5, 35.8], P = .032, q = .077), compared to those cohabitating. Conversely, prevalence rates of remission did not significantly differ by marital status for males (P > .05).

Figure 3.

Figure 3

Open in a new tab

Predicted prevalence of remission from AUD by separated marital status, sex, and partner drinking. Notes. Calculated from a multivariable logistic regression model with the following independent variables: Separated marital status, partner drinking, assigned sex, covariates, the two-way interactions between marital status and partner drinking, marital status and sex, partner drinking and sex, and the three-way interaction between marital status, partner drinking, and sex. For post hoc analyses, partner drinking was estimated at abstinent, at the sample mean (average), and one standard deviation above the mean (above average).

When partners drank at average levels, married females still had an unadjusted significantly higher remission rates than their unmarried female counterparts, with a predicted prevalence difference of 16.08 percentage points (PD = 16.08; 95% CI[2.8, 29.4], P = .02, q = .072), compared to those exclusively dating. Prevalence rates of remission did not significantly differ between those married females and those who were cohabitation (P = .053) and by marital status for males (all P’s > .05).

Conversely, but similar to the combined model, when partners drank at above average levels, prevalence rates of remission did not significantly differ by marital status for females (P > .05). However, married males had higher remission rates than those who were exclusively dating, with an unadjusted significant predicted prevalence difference of 24.5 percentage points (PD = 24.5; 95%CI[1.8, 47.1], P = .03, q = .077). Yet, prevalence rates of remission did not significantly differ between those cohabitating (q = .10). When comparing between males and females at similar partner drinking levels, no sex differences were observed (P > .05).

Exploratory analyses

Lastly, we conducted an exploratory analysis to investigate whether relationship satisfaction, relationship length, or relationship dissolution (measured by number of divorces) might be responsible for the association between marital status, partner drinking, and sex. Individually, relationship satisfaction, and relationship dissolution were not significant covariates in the association between marital status and remission (all P’s > .05). Furthermore, when replacing marital status with each explanatory variable as predictors (e.g. relationship satisfaction x partner drinking x sex), no three-way-interactions were found (all P’s > .05). suggesting that marital status may be associated with remission from AUD beyond the influence of having a strong “high-quality” relationship, like those found when exclusively dating one’s partner.

Discussion

The current study aimed to determine whether being married, as opposed to being in other types of relationships (i.e. living as married, dating exclusively or single), provides unique protective benefits related to maintaining AUD remission and explore sex and partner drinking’s moderating effects. In general, marriage was associated with a higher predicted prevalence of remission from AUD when compared to those who are dating exclusively but did not differ from single individuals. Furthermore, sex and partner drinking moderated this relationship.

Among females, predicted remission rates were consistently higher for those who were married compared to those in NMC relationships at abstinent and average levels of partner drinking. However, the difference between marital statuses diminished as partner drinking increased, such that those in marital and non-marital relationships display similarly low rates of remission from AUD. Conversely, among males, remission rates did not differ by marital status when partners were abstinent or average drinkers. However, when partners drank above average, married males showed higher predicted prevalence of remission than males in NMC relationships.

These results are consistent with prior literature that suggests partner drinking levels and motives influence one’s drinking behaviors (Polenick et al. 2018). However, these findings somewhat contradict the claims from previous research suggesting that marriage offers blanket protection against alcohol problems (Rendall et al. 2011). Unlike prior research, the current study compares married individuals and those in NMC relationships, including both those who are cohabitating and those exclusively dating. Past groupings have focused on differences between married and unmarried individuals, overlooking the variety of relational circumstances within the “non-marital” grouping. Current findings suggest that marriage may offer protective advantages over NMC relationships, but only under certain circumstances. Moreover, this buffering effect is not fully explained by relationship quality, as no significant interactions were found between relationship length, relationship satisfaction, or relationship dissolution, on sex, partner drinking, and remission from AUD. This suggests that marital status may be associated with remission from AUD beyond the influence of having a strong “high-quality” relationship, like those found when exclusively dating one’s partner.

Limitations

While this study makes unique contributions to the understanding of the relationship between marital status, partner drinking, sex, and AUD remission, findings should be considered in light of some limitations. First, findings may not generalize to the general population, as the current study uses data from a high-risk sample with heightened exposure to genetic and behavioral influences associated with AUD and remission, which may ultimately impact the relationship between marital status, partner drinking, sex, and AUD remission. Second, partner drinking was not directly obtained from the partner in question but from the study participants’ report of their partner’s drinking behaviors. These estimates may differ from actual drinking levels. Third, the sample predominantly consists of heterosexual couples (98.8%). The associations among marital status, sex, and partner drinking may differ in same-sex couples and should be examined in future research. Fourth, although the current study found that marital status is associated with AUD remission, the results separating the NMC groups are limited by sample size and statistical power. Future research should examine these associations in larger samples. Finally, the current study is cross-sectional, and thus we are unable to make causal determinations regarding the interactions between marriage, sex, and partner drinking on remission from AUD. Therefore, longitudinal studies are necessary to understand the temporality of marriage and AUD remission and how it is associated with these protective benefits.

Conclusion

The current study examines whether marriage has a protective influence on remission exceeding the benefits of being in a NMC relationship, and whether sex and partner drinking moderate this relationship. Although marriage appeared protective under specific circumstances (e.g. married vs exclusively dating; females with lower drinking partners), the observed differences suggest that relationship-related factors may pose additional risks to sustaining remission. These findings emphasize the need for intervention strategies that address diverse relationship contexts and relational dynamics, rather than focusing on general social or spousal support.

Contributor Information

Christina Garasky, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Alexis Duncan, Brown School of Social Work, Washington University in St. Louis, St. Louis, MO 63130, United States.

Fanghong Dong, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Gayathri Pandey, Department of Psychiatry and Behavioral Sciences, SUNY Downstate Health Sciences University, Brooklyn, NY, 11203, United States.

Jiawen Zhao, Department of Psychiatry, Rutgers Robert Wood Johnson Medical School, Piscataway, NJ 08901, United States.

Jared Balbona, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Grace Chan, Department of Psychiatry, University of Connecticut School of Medicine, Farmington, CT 06030, United States.

Weipeng Kuang, Department of Psychiatry and Behavioral Sciences, SUNY Downstate Health Sciences University, Brooklyn, NY, 11203, United States.

Chella Kamarajan, Department of Psychiatry and Behavioral Sciences, SUNY Downstate Health Sciences University, Brooklyn, NY, 11203, United States.

Martin Plawecki, Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN 46202, United States.

Victor Hesselbrock, Department of Psychiatry, University of Connecticut School of Medicine, Farmington, CT 06030, United States.

Samuel Kuperman, Department of Psychiatry, University of Iowa Carver College of Medicine, Iowa City, IA 52242  United States.

Bernice Porjesz, Department of Psychiatry and Behavioral Sciences, SUNY Downstate Health Sciences University, Brooklyn, NY, 11203, United States.

Andrey Anokhin, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Emma Johnson, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Arpana Agrawal, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Jessica Salvatore, Department of Psychiatry, Rutgers Robert Wood Johnson Medical School, Piscataway, NJ 08901, United States.

Sally Kuo, Department of Psychiatry, Rutgers Robert Wood Johnson Medical School, Piscataway, NJ 08901, United States.

Kathleen Bucholz, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Vivia McCutcheon, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO 63110, United States.

Author contributions

Christina Garasky (Conceptualization, Methodology, Formal analysis, Writing—original draft, Writing—review & editing), Alexis Duncan (Conceptualization, Methodology, Formal analysis, Writing—original draft, Writing—review & editing, Supervision), Fanghong Dong (Writing—original draft, Writing—review & editing), Gayathri Pandey (Writing—original draft, Writing—review & editing), Jiawen Zhoa (Writing—original draft, Writing—review & editing), Jared Balbona (Writing—original draft preparation, Writing—review & editing), Grace Chan (Writing—original draft, Writing—review & editing), Weipeng Kuang (Writing—original draft, Writing—review & editing), Chella Kamarajan (Writing—original draft, Writing—review & editing), Martin Plawecki (Writing—original draft, Writing—review & editing), Victor Hesselbrock (Writing—original draft, Writing—review & editing), Samuel Kuperman (Writing—original draft, Writing—review & editing), Bernice Porjesz (Writing—original draft, Writing—review & editing), Andrey Anokhin (Writing—original draft, Writing—review & editing), Emma Johnson (Writing—original draft, Writing—review & editing), Arpana Agrawal (Writing—original draft, Writing—review & editing, Supervision), Jessica Salvatore (Writing—original draft, Writing—review & editing), Sally Kuo (Writing—original draft, Writing—review & editing), Kathleen Bucholz (Writing—original draft, Writing—review & editing), and Vivia McCutcheon (Conceptualization, Methodology, Formal analysis, Writing—original draft, Writing—review & editing, Supervision)

Conflict of interest: All authors have no disclosures or conflict of interests.

Funding

This work was supported by the National Institutes of Health [T32 DA015025, R01 AA030563, U10 AA00840].

Data availability

COGA data access is restricted but can be accessed via an application to the National Institute on Alcohol Abuse and Alcoholism or in collaboration with a COGA investigator as a sponsor for a secondary analysis proposal. Details regarding access at https://cogastudy.org/resources-for-researchers/#accessing-coga-data.

References

  1. Agrawal  A, Brislin  SJ, Bucholz  KK. et al.  The collaborative study on the genetics of alcoholism: overview. Genes Brain Behav. 2023;22:e12864. 10.1111/gbb.12864 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. American Psychiatric Association . Diagnostic and Statistical Manual of Mental Disorders (DSM-5), Fifth edn. Arlington, VA: American Psychiatric Association, 2013. [Google Scholar]
  3. August  KJ, Sorkin  DH. Marital status and gender differences in managing a chronic illness: the function of health-related social control. Soc Sci Med. 2010;71:1831–8. 10.1016/j.socscimed.2010.08.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bagnardi  V, Rota  M, Botteri  E. et al.  Alcohol consumption and site-specific cancer risk: a comprehensive dose–response meta-analysis. Br J Cancer. 2015;112:580–93. 10.1038/bjc.2014.579 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Birditt  KS, Polenick  CA, Antonucci  TC. Drinking together: implications of drinking Partners for Negative Marital Quality. J Stud Alcohol Drugs. 2019;80:167–76. 10.15288/jsad.2019.80.167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Brown  SL, Lin  I-F, Mellencamp  KA. The rising midlife first marriage rate in the U.S. J Marriage Fam. 2022;84:1220–33. 10.1111/jomf.12861 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bucholz  KK, Cadoret  R, Cloninger  CR. et al.  A new, semi-structured psychiatric interview for use in genetic linkage studies: a report on the reliability of the SSAGA. J Stud Alcohol. 1994;55:149–58. 10.15288/jsa.1994.55.149 [DOI] [PubMed] [Google Scholar]
  8. Colditz  JB, Chu  KH, Hsiao  L. et al.  Characterizing online social support for alcohol use disorder: a mixed-methods approach. Alcohol Clin Exp Res. 2023;47:2110–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dawson  DA, Grant  BF, Stinson  FS. et al.  Recovery from DSM-IV alcohol dependence: United States, 2001–2002. Addiction. 2005;100:281–92. 10.1111/j.1360-0443.2004.00964.x [DOI] [PubMed] [Google Scholar]
  10. DeMaris  A. Marriage advantage in subjective well-being: causal effect or unmeasured heterogeneity?  Marriage Fam Rev. 2018;54:335–50. 10.1080/01494929.2017.1359812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dick  DM, Balcke  E, McCutcheon  V. et al.  The collaborative study on the genetics of alcoholism: sample and clinical data. Genes Brain Behav. 2023;22:e12860. 10.1111/gbb.12860 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dinescu  D, Turkheimer  E, Beam  CR. et al.  Is marriage a buzzkill? A twin study of marital status and alcohol consumption. Journal of Family Psychology. 2016;30:698–707. 10.1037/fam0000221 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Duncan  AE, McGuire  FH, Garasky  C. et al.  Gender identity moderates the association between alcohol use and eating disorder risk in US college and university students. Journal of Studies on Alcohol and Drugs. jsad-25. 2025. [DOI] [PubMed] [Google Scholar]
  14. Grant  BF, Goldstein  RB, Saha  TD. et al.  Epidemiology of DSM-5 alcohol use disorder: results from the National Epidemiologic Survey on alcohol and related conditions III. JAMA Psychiatry. 2015;72:757–66. 10.1001/jamapsychiatry.2015.0584 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hesselbrock  M, Easton  C, Bucholz  KK. et al.  A validity study of the SSAGA-a comparison with the SCAN. Addiction. 1999;94:1361–70. 10.1046/j.1360-0443.1999.94913618.x [DOI] [PubMed] [Google Scholar]
  16. Horwitz  AV, White  HR, Howell-White  S. Becoming married and mental health: a longitudinal study of a cohort of young adults. J Marriage Fam. 1996;58:895. 10.2307/353978 [DOI] [Google Scholar]
  17. House  JS, Landis  KR, Umberson  D. Social relationships and health. Science. 1988;241:540–5. 10.1126/science.3399889 [DOI] [PubMed] [Google Scholar]
  18. Jia  H, Lubetkin  EI. Life expectancy and active life expectancy by marital status among older U.S. adults: results from the U.S. Medicare health outcome survey (HOS). SSM - Population Health. 2020;12:100642. 10.1016/j.ssmph.2020.100642 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kendler  KS, Lönn  SL, Salvatore  J. et al.  Effect of marriage on risk for onset of alcohol use disorder: a longitudinal and Co-relative analysis in a Swedish National Sample. Am J Psychiatry. 2016;173:911–8. 10.1176/appi.ajp.2016.15111373 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kiecolt-Glaser  JK, Newton  TL. Marriage and health: his and hers. Psychol Bull. 2001;127:472–503. 10.1037/0033-2909.127.4.472 [DOI] [PubMed] [Google Scholar]
  21. Leonard  KE, Homish  GG. Predictors of heavy drinking and drinking problems over the first 4 years of marriage. Psychol Addict Behav. 2008;22:25–35. 10.1037/0893-164X.22.1.25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Levola  J, Kaskela  T, Holopainen  A. et al.  Psychosocial difficulties in alcohol dependence: a systematic review of activity limitations and participation restrictions. Disabil Rehabil. 2014;36:1227–39. 10.3109/09638288.2013.837104 [DOI] [PubMed] [Google Scholar]
  23. Li  Q, Wilsnack  R, Wilsnack  S. et al.  Cohabitation, gender, and alcohol consumption in 19 countries: A multilevel analysis. Substance Use & Misuse. 2010;45:2481–2502. 10.3109/10826081003692106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Maxwell  AM, Harrison  K, Rawls  E. et al.  Gender differences in the psychosocial determinants underlying the onset and maintenance of alcohol use disorder. Front Neurosci. 2022;16:808776.1–9. 10.3389/fnins.2022.808776 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. McCutcheon VV, Chun Kuo S, Smith RL. et al. Parental Remission from Alcohol Use Disorder and Offspring Alcohol Use Initiation, AUD, and Remission in a High-Risk Family Sample. Journal of Studies on Alcohol and Drugs, jsad-24. 2025. [DOI] [PMC free article] [PubMed]
  26. McCutcheon  VV, Kramer  JR, Edenberg  HJ. et al.  Social contexts of remission from DSM-5 alcohol use disorder in a high-risk sample. Alcohol Clin Exp Res. 2014;38:2015–23. 10.1111/acer.12434 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Muller  CJ, MacLehose  RF. Estimating predicted probabilities from logistic regression: different methods correspond to different target populations. Int J Epidemiol. 2014;43:962–70. 10.1093/ije/dyu029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nayak  MB, Patterson  D, Wilsnack  SC. et al.  Alcohol’s Secondhand harms in the United States: new data on prevalence and risk factors. J Stud Alcohol Drugs. 2019;80:273–81. 10.15288/jsad.2019.80.273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Nolen-Hoeksema  S. Gender differences in risk factors and consequences for alcohol use and problems. Clin Psychol Rev. 2004;24:981–1010. 10.1016/j.cpr.2004.08.003 [DOI] [PubMed] [Google Scholar]
  30. Plant  M, Miller  P, Plant  M. et al.  Marriage, cohabitation and alcohol consumption in young adults: An international exploration. Journal of Substance Use. 2008;13:83–98. 10.1080/14659890701820028 [DOI] [Google Scholar]
  31. Polenick  CA, Birditt  KS, Blow  FC. Couples’ alcohol use in middle and later life: stability and mutual influence. J Stud Alcohol Drugs. 2018;79:111–8. 10.15288/jsad.2018.79.111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Poortman  AR, Mills  M. Investments in marriage and cohabitation: the role of legal and interpersonal commitment. J Marriage Fam. 2012;74:357–76. 10.1111/j.1741-3737.2011.00954.x [DOI] [Google Scholar]
  33. Rendall  MS, Weden  MM, Favreault  MM. et al.  The protective effect of marriage for survival: a review and update. Demography. 2011;48:481–506. 10.1007/s13524-011-0032-5 [DOI] [PubMed] [Google Scholar]
  34. Sacks  JJ, Gonzales  KR, Bouchery  EE. et al.  2010 national and state costs of excessive alcohol consumption. Am J Prev Med. 2015;49:e73–9. 10.1016/j.amepre.2015.05.031 [DOI] [PubMed] [Google Scholar]
  35. Salvatore  JE, Gardner  CO, Kendler  KS. Marriage and reductions in men’s alcohol, tobacco, and cannabis use. Psychol Med. 2020;50:2634–40. 10.1017/S0033291719002964 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Seeman  TE. Health promoting effects of friends and family on health outcomes in older adults. Am J Health Promot. 2000;14:362–70. 10.4278/0890-1171-14.6.362 [DOI] [PubMed] [Google Scholar]
  37. Simon  RW. Revisiting the relationships among gender, marital status, and mental health. Am J Sociol. 2002;107:1065–96. 10.1086/339225 [DOI] [PubMed] [Google Scholar]
  38. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. (2024) . 2023.  NSDUH Detailed Tables. SAMHSA. https://www.samhsa.gov/data/report/2023-nsduh-detailed-tables [Google Scholar]
  39. Stillman  MA, Sutcliff  J. Predictors of relapse in alcohol use disorder: identifying individuals most vulnerable to relapse. Addiction and Substance Abuse. 2020;1:3–8. [Google Scholar]
  40. Uhing  A, Williams  JS, Garacci  E. et al.  Gender differences in the relationship between social support and strain and mortality among a national sample of adults. J Behav Med. 2021;44:673–81. 10.1007/s10865-021-00221-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Umberson  D. Gender, marital status and the social control of health behavior. Soc Sci Med. 1992;34:907–17. 10.1016/0277-9536(92)90259-S [DOI] [PubMed] [Google Scholar]
  42. Vitali  A, Fraboni  R. Pooling of wealth in marriage: the role of premarital cohabitation. Eur J Popul. 2022;38:721–54. 10.1007/s10680-022-09627-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Woods  M, Avery  J. Stigma and Alcohol Use Disorder: Overcoming Societal Attitudes. In: Khan, M., Avery, J. (eds) From Stigma to Support. Psychiatry Update, 2024. vol 4. Springer, Cham. 153-62   10.1007/978-3-031-73553-0_12 [DOI] [Google Scholar]
  44. Zhong  Y, McGuire  FH, Duncan  AE. Who is trying to lose weight? Trends and prevalence in past-year weight loss attempts among US adults 1999–2018 at the intersection of race/ethnicity, gender, and weight status. Eat Behav. 2022;47:101682. 10.1016/j.eatbeh.2022.101682 [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

COGA data access is restricted but can be accessed via an application to the National Institute on Alcohol Abuse and Alcoholism or in collaboration with a COGA investigator as a sponsor for a secondary analysis proposal. Details regarding access at https://cogastudy.org/resources-for-researchers/#accessing-coga-data.

Articles from Alcohol and Alcoholism (Oxford, Oxfordshire) are provided here courtesy of Oxford University Press

RESOURCES