Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: J Interpers Violence. 2022 Aug 18;38(3-4):3445–3467. doi: 10.1177/08862605221108088

Stability and Change in Men’s Intimate Partner Violence and Substance Use in Early Adulthood

Hyanghee Lee 1
PMCID: PMC9851931  NIHMSID: NIHMS1833933  PMID: 35979878

Abstract

Despite substantial evidence of the role of substance use in intimate partner violence (IPV), little is known about the impact of substance use on stability and change in the experience of IPV as both a perpetrator and a victim. Using an ethnically diverse sample of 232 men in early adulthood (mean age = 29.1, SD = 0.91), this study defined typologies of IPV based on men’s reports of both perpetration and victimization; examined the potential impact of substance use, including alcohol and marijuana use, on IPV typologies over two measurement occasions; and quantified stability and change in these typologies over time. Patterns of IPV were characterized by three classes at each time point: no IPV, psychological aggression, and physical aggression. Men’s regular marijuana use was associated with physical aggression contemporaneously and prospectively. Partner’s problem alcohol use was associated with psychological aggression contemporaneously, suggesting that women’s problem alcohol use could be a risk factor for their own and their partner’s IPV perpetration. IPV appeared to remain somewhat stable over time with 67% of men remaining in the same IPV class. Among those who did transition from one typology to another, it was most often to a less severe IPV typology. Regular marijuana users were more likely to be in the physical aggression typology rather than the no IPV typology, with a higher probability of transitioning to a more severe IPV typology than nonusers. The present study has implications for prevention and intervention efforts by its ability to identify men who are at greatest risk for continued or increased violence and underscores that men’s marijuana use may exacerbate IPV.

Keywords: intimate partner violence, typology, substance use, person-centered approach

Intimate partner violence (IPV) is a major public health concern (Center for Disease Control and Prevention [CDC], 2020). The most recent data from the National Intimate Partner and Sexual Violence Survey (Smith et al., 2018) indicates that 1 in 4 women and 1 in 10 men have experienced some form of psychological, physical, or sexual IPV in their lifetime and that victims of IPV experience numerous negative consequences including physical injuries, fear, anxiety, and post-traumatic stress disorder. A public health approach to the prevention of IPV involves the identification of malleable risk factors. For example, a substantial body of literature has identified substance use as a risk factor for both IPV perpetration and victimization. Much of the work that has examined the relationship between substance use and IPV has linked the individual’s substance use with either IPV perpetration or IPV victimization; however, bidirectional violence in which both partners are involved in perpetration and victimization is the most common pattern of IPV in community-based samples (Kelly & Johnson, 2008). Thus, more work is needed to identify typologies of IPV based on perpetration and victimization, to examine stability and change in an individual’s patterns of IPV over time, and to determine whether substance use plays a role in predicting IPV typologies both contemporaneously and longitudinally. The present study aims to fill these gaps in the literature by examining patterns of IPV perpetration and victimization as well as stability and change in different types of IPV across two time points using an ethnically diverse sample of men. Furthermore, the present study examines the potential role that substance use by the respondent and his intimate partner plays in predicting typologies of IPV.

Associations of Men’s Alcohol and Marijuana Use and IPV

A sizable body of literature has documented the impact of substance use (e.g., alcohol, marijuana) on IPV experiences among men. With regard to alcohol use, for example, a meta-analytic study demonstrated a significant, positive association between men’s alcohol use and the likelihood of IPV perpetration and victimization (Cafferky et al., 2018). Moreover, a review study focusing on the role of substance use in same-day IPV showed that men were more likely to both perpetrate IPV and be a victim of IPV on drinking days (de Bruijn & de Graaf, 2016). While there is a well-documented association between men’s alcohol use and IPV, studies on the association between marijuana use and IPV have produced mixed findings. For example, a recent meta-analysis revealed that men’s marijuana use was positively associated with IPV perpetration and victimization, and that the magnitude of the association with perpetration and victimization was similar (Cafferky et al., 2018). However, another meta-analysis did not find an association between men’s marijuana use and IPV perpetration (Moore et al., 2008). Similarly, in studies which explored the role of marijuana use in same-day IPV, no association between men’s marijuana use and IPV perpetration was detected (Fals-Stewart et al., 2003; Shorey et al., 2014). Findings from a nationally representative study also indicated that marijuana abuse or dependence was not associated with IPV perpetration (Stalans & Ritchie, 2008). As such, there is no definitive empirical consensus regarding whether marijuana use is associated with either IPV perpetration or victimization, or both. Thus, more research on the impact of marijuana use on IPV is needed. This need is perhaps particularly urgent given the growing number of states legalizing recreational marijuana and the likelihood of greater use in the population as a result (Cerdá et al., 2020).

Furthermore, studies that account for both alcohol and marijuana use simultaneously are needed to examine the unique contribution of specific substances. The joint assessment allows for a more complete understanding of the contribution of each substance in predicting IPV and enhances intervention and prevention efforts by focusing on specific substances that are most closely related to IPV. For example, Feingold et al. (2008) found that marijuana use played a more important role in men’s physical and psychological IPV perpetration than did alcohol. These findings were also supported in a study with men arrested for IPV (Shorey et al., 2018), indicating that marijuana use was significantly associated with psychological, physical, and sexual IPV, after controlling for alcohol use. However, to date, most studies have focused on the independent role of alcohol or marijuana use in IPV.

Lastly, many studies of substance use and IPV do not consider both partner’s use of substances—yet the context of IPV in a relationship is likely to be influenced by bother partner’s substance use. Indeed, IPV incidents were shown to be more severe when both partners were drinking alcohol together (Testa et al., 2012). Although IPV researchers recommend considering the substance use of partners (Devries et al., 2014), only a few studies have done so (e.g., Feingold et al., 2014; Gilbert et al., 2013; Low et al., 2017).

A Variable-Centered and Person-Centered Approach to IPV

The CDC (2020) defines IPV as physical violence, sexual violence, stalking, or psychological harm by a current or former partner, suggesting that each individual may experience IPV differently; however, the extant research on IPV is limited by its focus on conceptual and methodological approaches that overlook the heterogeneity of IPV experiences. The IPV literature focuses primarily on presence/absence of IPV as a unitary construct and explores risk factors for IPV perpetration and victimization using a variable-centered approach, such as multiple regression or structural equation modeling, which explores associations among variables (e.g., which variable is the strongest predictor of an outcome) in a population. For example, variable-centered studies have identified several risk factors for IPV perpetration among men, including substance use, mental disorder, gender equity ideologies, and intergenerational transmission of violence and trauma (e.g., Cafferky et al., 2018; Dasgupta et al., 2018; Fleming et al., 2015; Lipsky & Caetano, 2011). While the variable-centered studies are helpful to understand the average effects of a specific risk factor across a study population, they do not assume individual differences in IPV experience; thus, results of the variable-centered studies do not inform researchers about the relationship between a risk factor and IPV at the individual level.

In contrast, a person-centered approach identifies subgroups of individuals who share particular attributes or behaviors using latent class analysis (LCA; Collins & Lanza, 2010). This approach is particularly helpful for identifying subgroups of individuals in a population that show particular patterns of behaviors and exploring differences in covariates among these subgroups. In our setting, person-centered models can identify subgroups of men based on distinctive patterns of IPV behaviors and explore differences in the impact of risk factors (e.g., substance use) among these subgroups. In line with this, the IPV literature explains different types of IPV based on motives, contexts, and patterns (Johnson, 2006; Kelly & Johnson, 2008), and some person-centered studies empirically support the notion of different types of IPV in the population (e.g., Orpinas et al., 2013; Saint-Eloi Cadely et al., 2020). Thus, it is important to recognize the heterogeneity of IPV experiences, rather than a unitary construct of IPV, and to examine how different types of IPV are associated with substance use. Relatedly, some previous variable-centered studies have found the association between substance use and different types of IPV using cross-sectional data. For example, problematic alcohol use was associated with psychological IPV perpetration, but not physical IPV perpetration, among clinical and community samples of men (Grigorian et al., 2020; Juarros-Basterretxea et al., 2022; Lynch & Renzetti, 2020; Stuart et al., 2008). Conversely, drug use including marijuana was associated with physical aggression, but not psychological aggression, among men arrested for domestic violence (Stuart et al., 2008). Taken together, these findings indicate that specific substances may have a differential impact on subtypes of IPV. Building on this, the current study extends previous work by exploring the contemporaneous and prospective impact of substance use on the identified IPV subtypes from a person-centered analysis.

Furthermore, identifying distinct subtypes of IPV is also necessary for examining stability and change in IPV across time. Longitudinal studies on temporal changes in IPV have shown that IPV perpetration tends to be continuous (Capaldi et al., 2003; Shortt et al., 2012) or changes as a function of initial levels of severity such that couples who initially experienced severe aggression showed a decline in aggression over time while couples who initially experienced moderate aggression continued to engage in aggression over time (Lawrence & Bradbury, 2007; Lorber & Daniel O’Leary, 2004). Such findings indicate that there may be subtypes of IPV existing within populations and changes in IPV behaviors vary across subtypes of IPV. For example, a person-centered study revealed an increasing pattern in two distinct psychological IPV subtypes (i.e., minor psychological IPV and severe psychological IPV); however, physical IPV showed a stable pattern with persistent and consistent use of IPV from adolescence to young adulthood (Saint-Eloi Cadely et al., 2020). Thus, exploring multiple patterns of IPV may shed light on our understanding of not only distinct patterns of IPV but stability and change in IPV. In doing so, it could provide insights into targeted interventions by identifying men who are at greatest risk for continued or increased violence.

The Current Study

The present study aims to explore different types of IPV perpetration and victimization across two time points using an ethnically diverse sample of men. The study further examines the potential role that substance use by men and his intimate partner plays in predicting different types of IPV. Specifically, typologies of IPV at each time point is identified using latent class analysis (LCA), which identifies latent subgroups of individuals (i.e., latent classes) in a given population based on similarity of responses to a set of observed categorical items (Collins & Lanza, 2010). Although the literature has considered IPV to be unidimensional, community-based samples show that victims are often perpetrators and perpetrators are often also victims (Anderson, 2002; Johnson & Leone, 2005). Thus, it is advantageous to examine IPV on a continuum and as a qualitative phenomenon, rather than focusing on only presence/absence of victimization or presence/absence of perpetration. Consequently, a person-centered approach is used to identify subgroups, that is latent classes, according to common IPV patterns of victimization and perpetration. It further explores a contemporaneous and prospective impact of substance use, by both the men and their partners, on IPV typologies. Next, the change and stability in different types of IPV over a 2-year period are examined using latent transition analysis (LTA). LTA is a longitudinal extension of LCA that examines transition from one latent subgroup to another over time (Collins & Lanza, 2010). Although IPV frequency and severity can change over time, few researchers have investigated whether IPV is stable or variable using a person-centered approach. To summarize, the current study is designed to address three research questions. First, are there distinct subgroups (i.e., latent classes) of men who experience particular patterns of IPV? Second, does substance use have a contemporaneous and/or prospective potential impact on IPV class membership? Third, is there change in IPV class membership over time, and is stability or change in IPV over time related to substance use?

Methods

Study Design and Participants

Data for the present study came from the Rochester Youth Development Study (RYDS), a two-decade longitudinal investigation of antisocial behaviors in a representative urban community sample. In 1988, 1,000 adolescents were recruited from the seventh and eighth grades in public schools in Rochester, New York. The study was initially funded by the Office of Juvenile Justice and Delinquency Prevention to examine the causes and correlates of adolescent delinquency. To obtain a sufficient number of youths at high risk for serious delinquency and substance use, males living in neighborhoods with high arrest rates were oversampled (i.e., 75% male). The drawn sample included 68% Black, 17% Hispanic, and 15% White participants. Fourteen waves of data were collected over three phases. During Phase 1 (starting in 1988), the adolescents were interviewed 9 times (Waves 1–9) every 6 months. During Phase 2 (starting in 1994), participants were then in their early 20s and were interviewed thrice (Waves 10–12), with approximately 1 year between each interview. During Phase 3 (Waves 13 in 2003 and Wave 14 in 2005), participants were in their late 20s/early 30s and were interviewed twice, with approximately 2 years between interviews. By the end of Phase 3, 80% (n = 803) of the initial sample had been retained and among them, 572 participants were men. In the current study, data from Phase 3 were used to explore the impact of substance use on IPV over two measurement occasions when the participants were adults. Participants who were incarcerated at either wave (n = 39) were excluded, yielding 533 men eligible for the current study. Further, 45 men with missing data on their relationship status were excluded. Of these, men who remained in a relationship with the same partner (n = 232) were included in the current investigation of stability and change in IPV.

Given that IPV data were collected from men who were in a relationship of at least 6 months and being in a relationship is necessary for IPV, men who were not in a relationship at either survey were excluded. Also, as the study explores stability and change in IPV, men who were in a relationship at both waves with each of two different partners were excluded. To describe the bias created as a result of this exclusion, χ2 tests of independence was used to examine whether substance use (i.e., key risk factors) was associated with inclusion (vs. exclusion) based on this criterion. Specifically, the χ2 analysis was used to determine whether substance use differed by five relationship statuses: (1) in a relationship only at Wave 13 (n = 54), (2) in a relationship only at Wave 14 (n = 50), (3) in a relationship at both waves with a new partner at Wave 14 (n = 50), (4) in a relationship with the same partner at both waves (n = 232), and (5) not in a relationship at both waves (n = 102). Results showed that problem alcohol use did not differ across the groups at either wave: χ2(4) = 2.15, p = .71 at Wave 13; χ2(4) = 6.58, p = .16 at Wave 14. Similarly, regular marijuana use did not differ across the groups at either wave: χ2(4) = 8.86, p = .06 at Wave 13; χ2(4) = 5.07, p = .28 at Wave 14).

The demographics of the men in the final sample (n = 232) are as follows. The mean age at Wave 13 was 29.1 years (SD = 0.91). Most men were Black (58%); 21% White and 21% were Hispanic; 19% of men had less than high school education, 43% had high school education, and 38% had some post-secondary education at Wave 13. Most men (94%) were employed at Wave 13.

Measurement

Intimate partner violence.

IPV was measured using 20 items from the conflict tactics scale (CTS; Straus, 1979). Men were asked how frequently they had engaged in the 10 violent behaviors ranging from stomping away from an argument to using a weapon, and then were asked how frequently they had experienced each violent behavior from their partner at Waves 13 and 14 during Phase 3 of RYDS. For example, they were asked, “how many times during the past year have you slapped your partner?” and “how many times during the past year has your partner slapped you?” from 0 (never) to 6 (more than 20 times). To explore types of IPV, binary measures of the IPV behaviors (contrasting any incidence of the IPV behavior versus no incidence) were created because responses to the upper categories were low. Thus, binary indicators were used to indicate whether the IPV behavior had occurred, coding 0 for no and 1 for yes for each item for perpetration and for victimization. The individual binary indicators of the perpetration and victimization items were then used to define a latent class variable of IPV.

Substance use.

At Waves 13 and 14, men were asked whether they drank beer, wine, or liquor or used marijuana during the previous year and, if so, how often. They also reported their partner’s substance use. Based on those reports, two binary measures of substance use were created: problem alcohol use and regular marijuana use for both men and their partners. The problem alcohol use variable was coded 1 if alcohol was consumed regularly (at least once per month) and the respondent reported consuming five or more drinks on at least one occasion during the past year, and 0 if alcohol was consumed less than once a month or if the respondent never drank more than five drinks on a single occasion. At Wave 13, 39.7% reported problem alcohol use; at Wave 14, 32.8% reported problem alcohol use. The same measure was created using the respondent’s reports of their partner’s problematic alcohol use at both time points (17.7% at Wave 13; 12.1% at Wave 14).

Similarly, regular marijuana use was coded 1 if the respondent used at least once a month on average (≥12 times in the past year) and 0 if they infrequently used or refrained from use (<12 times in the past year). At Wave 13, 19% reported at least monthly marijuana use; at Wave 14, 14.7% reported monthly marijuana use. Men reported their partner’s marijuana use with the same measure: at Wave 13, 10.3% of partners used marijuana at least once a month; at Wave 14, 9.1% used marijuana at least once a month.

Control variables.

Several sociodemographic variables were included to account for the association between substance use and IPV class membership as controls after the class enumeration process: race/ethnicity, age, education level, employment status, and community arrest rate at the start of RYDS, as this was a sampling parameter in the study design.

Analysis Plan

First, LCA was fit to identify IPV subtypes at Wave 13 and Wave 14, separately. The LCA cross-sectional model was used to identify unobserved, latent subgroups of men based on patterns of responses to the IPV items. A series of LCA models with one to five classes were conducted at each wave to determine how many classes were needed to best describe the patterns observed. Model selection was based on fit indices: (1) the Bayesian information criterion (BIC), (2) the sample-size adjusted BIC (ABIC), and (3) the Lo-Mendell-Rubin likelihood ratio test (LMR). In general, smaller BIC and ABIC values suggest a better model fit. The LMR test indicates whether a k-class solution fits better than a solution with k−1 classes (Nylund et al., 2007). Each model’s entropy, which indicates classification accuracy, as well as model interpretability based on theoretical and past empirical findings were also considered. After the optimal number of classes based on these criteria was chosen at each wave, the contemporaneous and prospective associations of substance use on IPV class membership were explored, controlling for demographic variables, using the three-step approach. The three-step approach was used to estimate the effects of covariates in mixture models, which included (1) identifying latent classes, (2) assigning individual to latent classes, and (3) estimating the relationship between latent classes and covariates taking into account the measurement error in the class assignment (Asparouhov & Muthén, 2014). All three steps were performed automatically by using the R3STEP option in Mplus (Muthén & Muthén, 1998–2017).

Second, LTA (Collins & Lanza, 2010) was conducted to model change in types of IPV over time (Collins & Lanza, 2010). After finding the optimal number of classes at each time point in the LCA step, measurement invariance was tested to determine whether the classes maintained their meaning across time. Models were evaluated using a likelihood ratio test (LRT) by comparing model fits between a full measurement invariance model and baseline model where item-response probabilities were freely estimated across time (Collins & Lanza, 2010). After an unconditional LTA model was identified, a classify and analyze approach was used to examine the role of substance use in transitions between classes. All analyses were conducted using Mplus, Version 8.4 (Muthén & Muthén, 1998–2017).

Results

Research Question 1: Are There Distinct Types of IPV?

Table 1 shows the fit indices for the two- to five-class solutions at Waves 13 and 14 for the LCA class enumeration exploration. While the BIC was minimized for the three-class model at both waves, the aBIC was minimized for the four-class model at both waves. However, one class in the four-class model contained less than 5% of the sample at both waves and the two classes were not distinguishable from one another in the four-class model. Thus, the three-class solution was chosen as the best fitting LCA model at each wave. Item response probability for the three-class LCA at Waves 13 and 14 is depicted in Figure 1.

Table 1.

Results of Class Enumeration of Intimate Partner Violence Latent Class Analysis (n = 232).

Model BIC ABIC LMR p value Entropy
Wave 13
 Two-class model 2,237.71 2,114.10 .00 0.92
 Three-class model 2,160.61 1,973.62 .13 0.93
 Four-class model 2,192.10 1,941.70 .05 0.89
 Five-class model Model did not converge
Wave 14
 Two-class model 2,051.10 1933.83 .00 0.91
 Three-class model 1,996.16 1818.67 .00 0.89
 Four-class model 2,020.73 1783.02 .00 0.92
 Five-class model Model did not converge
Open in a new tab

Note. BIC = Bayesian Information Criterion; aBIC = sample-size adjusted BIC; LMR = Lo-Mendell-Rubin likelihood ratio test.

Figure 1.

Figure 1.

Open in a new tab

Item response probability for the three-class Latent Class Analysis (n = 232).

Note. (P) = IPV perpetration item; (V) = IPV victimization item.

The classes were labeled no IPV, psychological aggression, and physical aggression based on item-response probabilities. Men in the psychological aggression class reported high likelihoods of perpetrating psychological forms of IPV including insulting and stomping out of a room; however, they showed low likelihoods for physical violence. This was the most prevalent class (56% at Wave 13; 45% at Wave 14) and included psychological forms of IPV perpetrated by both men and their partners. In contrast, the physical aggression class was defined by a greater probability of physical aggressive behaviors: 8% at Wave 13 and 13% at Wave 14. For example, they were more likely to be perpetrators in pushing their partner (0.80) or be victims in being pushed by their partner (0.86) at Wave 13. While minor assaults (e.g., throwing something, pushing) were reported by both partners at a similar rate, men reported that their partner used higher levels of physical violence (e.g., hitting with an object, kicking) than they used against their partners. For example, men were about 3 times more likely to be hit with an object than they were to hit their partners at both waves. This class also involved generally high likelihoods of psychological aggression perpetrated by both men and their partners. Lastly, the no IPV class included 36% and 42% of the men at Waves 13 and 14, respectively.

Research Question 2: Does Substance Use Have a Contemporaneous and/or Prospective Potential Impact on IPV Class Membership?

Next, substance use and the control variables were added as predictors of IPV latent class membership. Two models were fit. First, a contemporaneous model was fit in which IPV latent class membership at Wave 13 was regressed on the substance use measures at Wave 13 and the control variables. Second, a prospective model was fit in which IPV latent class membership at Wave 14 was regressed on the substance use measures at Wave 13 and the control variables. The results are presented in Table 2.

Table 2.

Odds Ratio for the Covariates on Latent Class Membership at Each Wave (n = 232)

Covariates Psychological aggression Physical aggression
OR 95% CI OR 95% CI
Wave 13 IPV
 Age 1.15 [0.79, 1.67] 1.04 [0.50, 2.19]
 Black 0.91 [0.37, 2.20] - -
 Hispanic 0.58 [0.20, 1.69] - -
 Employment statusa 0.78 [0.15, 4.09] 0.06** [0.01, 0.45]
 Education 1.03 [0.86, 1.23] 0.61 [0.35, 1.04]
 Community arrest rate 0.86 [0.71, 1.05] 0.94 [0.67, 1.33]
 Men’s problem alcohol use at wave 13 2.07* [1.03, 4.18] 1.10 [0.15, 7.55]
 Men’s regular marijuana use at wave 13 2.24 [0.69, 7.28] 12.00** [2.28, 63.14]
 Partner’s problem alcohol use at wave 13 1.36 [0.52, 3.52] 6.71† [0.83, 54.48]
 Partner’s regular marijuana use at wave 13 1.37 [0.33, 5.69] 1.11 [0.16, 8.02]
Wave 14 IPV
 Age 1.09 [0.75, 1.57] 1.82 [1.00, 3.34]
 Black 0.85 [0.36, 2.05] 6.98* [1.27, 38.34]
 Hispanic 0.53 [0.18, 1.53] 6.95* [1.12, 43.23]
 Employment statusa 0.95 [0.20, 4.63] 0.19* [0.05, 0.76]
 Education 0.95 [0.79, 1.16] 1.04 [0.77, 1.40]
 Community arrest rate 0.92 [0.77, 1.10] 0.81 [0.63, 1.03]
 Men’s problem alcohol use at wave 13 1.18 [0.58, 2.40] 1.92 [0.67, 5.47]
 Men’s regular marijuana use at wave 13 1.01 [0.33, 3.09] 10.91*** [3.12, 38.22]
 Partner’s problem alcohol use at wave 13 1.92 [0.79, 4.64] 1.07 [0.23, 4.87]
 Partner’s regular marijuana use at wave 13 1.73 [0.46, 6.49] 0.67 [0.15, 3.04]
Open in a new tab

Note. Coefficients for Black and Hispanic for the physical aggression class at wave 13 were not determined due to the low number of participants assigned to this class.

Reference group = No IPV class

a

binary variable (1= employed; 0= not employed); OR = odds ratio, CI = confidence interval

†

p < .1

*

p < .05

**

p <. 01

***

p < .001

With regard to the contemporaneous impact of substance use on IPV latent class membership, results showed that men’s problem alcohol use at Wave 13 were significantly associated with the odds of being in the psychological aggression class relative to the no IPV class at Wave 13 (OR = 2.07, 95% CI [1.03, 4.18]). The effects of partner’s substance use on the odds of membership in the psychological aggression (as compared to the no IPV class) were substantially smaller and not statistically significant. Partner’s problem alcohol use at Wave 13 were associated with the odds of being in the physical aggression class relative to the no IPV class (OR = 6.71, 95% CI [.83, 54.48]) although the effect was just outside of the boundary for significance. Men’s regular marijuana use at Wave 13 was significantly associated with the odds of being in the physical aggression class relative to the no IPV class at Wave 13 (OR = 12, 95% CI [2.28, 63.14]).

With regard to the prospective potential impact of substance use at Wave 13 on IPV at Wave 14, men’s prior marijuana use was associated with the odds of being in the physical aggression class versus being in the no violence class at Wave 14 (OR = 10.91, 95% CI [3.12, 38.22]). Partner’s substance use was not significantly associated with latent class membership at Wave 14. Taken together, the results suggest that men’s problem alcohol use was associated with psychological aggression contemporaneously and men’s regular marijuana use was associated with physical aggression contemporaneously and prospectively. Although large confidence intervals indicated a high degree of uncertainty in the estimate of the parameter, perhaps because of the small sample size, the findings are consistent with previous studies indicating that marijuana use is associated with physical IPV but not psychological IPV perpetration (Stuart et al., 2008). Partner’s problem alcohol use appeared to only have a contemporaneous impact on physical aggression.

Research Question 3: Is There Any Change in IPV Class Membership Over Time?

Before fitting an LTA model, measurement invariance was examined to determine whether the same number and type of classes emerge at each wave. If measurement invariance holds, then a straightforward comparison of classes and transitions across time is facilitated. The full measurement invariance model was fit by constraining all item-response probabilities to be equal for each class over time. The LRT comparing the fixed and freed models was not significant (χ2diff = 25.85, df = 54, p = .99), indicating that the meanings of the IPV classes are not significantly different across time (i.e., measurement invariance). Thus, item response probabilities were constrained to be equal in the subsequent analyses.

Overall, men tended to remain in the same class: 67% of the men were stayers (i.e., those who remained in the same IPV class) and 33% were movers (i.e., those who transitioned among the IPV classes). Among the stayers, 40% were in the no IPV class, 46% in the psychological aggression class, and 14% were in the physical aggression class. Among the movers, 63% moved to a less severe class (e.g., physical aggression class → psychological aggression or no IPV class, psychological aggression → No IPV class) and 37% moved to a more severe class (e.g., no IPV class → physical or psychological aggression class, psychological aggression class → physical aggression class).

Next, the association between substance use and transitions in latent class membership over time was examined. Although ideally, predictors of the transitions would be directly added to the LTA model, the small sample size precluded this approach. More complex latent transition models (i.e., LTA with predictors of transitions) require relatively large samples and subgroup sizes (Ryoo et al., 2018). Instead, a classify and analyze approach was used by assigning each participant to their most likely class at each wave based on the unconditional model, and then produced descriptive statistics to examine the latent transition matrix as a function of substance use at Wave 13. The results are presented in Table 3. The boldface numbers represent the probability of membership in the same latent class over time (i.e., stayers) and the off-diagonal numbers represent the probability of changing to a different latent class at Wave 14 (i.e., movers).

Table 3.

Latent Transition Probabilities (n = 232).

Wave 14
No IPV Psychological Aggression Physical Aggression
Wave 13
Full sample
 No IPV 0.78 0.21 0.01
 Psychological aggression 0.26 0.66 0.08
 Physical aggression 0.09 0.33 0.58
Latent transition probabilities as a function of substance usea
 Nonalcohol users at Wave 13 (n = 140)
  No IPV 0.73 0.23 0.05
  Psychological aggression 0.31 0.62 0.07
  Physical aggression 0.13 0.43 0.43
 Problem alcohol users at Wave 13 (n = 92)
  No IPV 0.73 0.27 0
  Psychological aggression 0.27 0.62 0.12
  Physical aggression 0.11 0.28 0.61
 Non-marijuana users at Wave 13 (n =188)
  No IPV 0.73 0.23 0.04
  Psychological aggression 0.29 0.63 0.08
  Physical aggression 0.16 0.48 0.36
 Regular marijuana users at Wave 13 (n = 44)
  No IPV 0.71 0.29 0
  Psychological aggression 0.29 0.57 0.14
  Physical aggression 0.06 0.19 0.75
Open in a new tab

Note. IPV = intimate partner violence. Boldface numbers represent the probability of membership in the same latent class over 2 years (i.e., stayers).

a

Transition probabilities are based on each participant’s most likely class membership (i.e., “hard” classification of people using modal class assignment).

First, problem alcohol users were more likely to stay in the physical aggression class (0.61 for problem alcohol users vs. 0.43 for nonalcohol users). Second, similar to problem alcohol use, men who used marijuana regularly were more likely to stay in the physical aggression class (.75 for regular marijuana users vs. 0.36 for non-marijuana users). If they were not in the physical aggression class at Wave 13, they were more likely to transition to a more severe class (e.g., psychological aggression → physical aggression: 0.14) compared to non-marijuana users (e.g., psychological aggression → physical aggression: 0.08). Third, nonalcohol and non-marijuana users had relatively similar transition patterns; they were highly likely to stay in the no IPV class (0.73 for both nonalcohol and non-marijuana users) or transition to a less severe IPV class. For example, nonalcohol users in the physical aggression class had a 0.43 probability of transitioning into the psychological aggression class. However, it was very unlikely that a nonalcohol user in the psychological aggression class would transition into the physical aggression class (0.07). Similarly, non-marijuana users in the physical aggression class had a 0.48 probability of transitioning into the psychological aggression class. And non-marijuana users in the psychological aggression class were not likely to transition into the physical aggression class (0.08).

Discussion

The present study reported the results of a two-wave assessment of substance use and IPV in a community sample of men. First, three distinct types of IPV at each time point were identified: (1) no IPV, (2) psychological aggression, and (3) physical aggression. The majority of the participants (56% at Wave 13 and 45% at Wave 14) experienced psychological aggression, which often occurs in the context of arguments and conflicts between partners (Kelly & Johnson, 2008). While the current findings are based on men’s reports of partner’s IPV, the findings are consistent with previous studies indicating the bidirectional nature of IPV in a community sample (Langhinrichsen-Rohling et al., 2012). Those in the physical aggression class were at greatest risk; men reported that both partners perpetrated physical violence, such as pushing or slapping. Interestingly, men in this class reported that their partner used more severe levels of physical violence (e.g., hitting with an object) than they used against their partner. While this finding does not provide the context of violence or primary perpetrator, for example, women’s violence might occur to stop their violent partners, it may indicate a difference in the types and severity of the violence experienced and committed by men and women. In fact, previous studies have shown a gender difference in violent behaviors such that female perpetrators are more likely to use a weapon or object while male perpetrators tend to use their bodies to injure their partners (Busch & Rosenberg, 2004; Henning & Feder, 2004). Thus, the finding may suggest that physical aggressive behaviors are not always gender symmetrical, in other words, the severity and frequency of violence that two partners are perpetrating vary across gender, though it is critical to acknowledge that our results are based on men’s report of both their own and their partners’ IPV. Taken together, the results are consistent with the notion that IPV is a bidirectional construct, not a unitary construct with a primary perpetrator and victim.

Second, the results showed some contemporaneous as well as prospective associations of substance use on IPV types. Specifically, men’s own marijuana use and partner’s alcohol use were associated with physical aggression contemporaneously. The contemporaneous association of substance use by both men and their partners implies the reciprocal influence between partners, suggesting that IPV is shaped by the other partner’s substance use. Thus, when considering the role of men’s substance use in IPV, it may be important to understand the dyadic context of substance use—it may occur in the context of partner drinking and/or using marijuana. Also, the prospective model showed that that men’s prior marijuana use was associated with later physical aggression, indicating a prospective prediction of physical aggression based on prior substance use. This finding provides important implications for prevention planning. For example, a history of problematic substance use could be used to screen individuals into IPV prevention programs or tailor the content of prevention programs for those who have a history of problematic substance use. It is worth noting that partner’s substance use was not associated with IPV in the prospective model, suggesting that the degree of contemporaneous and prospective impacts of substance use from each partner may differ over time. A potential explanation for this finding is that substance use plays a more important role in IPV in men than women and could have a lingering impact on conflict and aggression 2 years later. In sum, the findings reiterate the importance of examining each partner’s substance use and its impacts on IPV, rather than focusing on individual partner factors alone.

Furthermore, the findings suggest that the potential impact of specific substance on IPV varies. Specifically, marijuana use seems to be more related to severe IPV than alcohol use. Men’s problem alcohol use was associated with psychological aggression, while marijuana use was associated with physical aggression, which was a severe form of IPV in the current typologies. These findings may be explained in part by the race/ethnic makeup of the current study sample (i.e., 58% Black, 21% White, 21% Hispanic) and differences in substance use among racial/ethnic groups. Problem alcohol use (e.g., binge drinking) is less common among Black men (age 26–34) than among Hispanic and White men (age 26–34), while marijuana use is more common among Black men than White and Hispanic men, according to the 2018 National Survey on Drug Use and Health (Substance Abuse and Mental Health Services Administration, 2019). Thus, it is possible that the overrepresentation of Black individuals may drive these findings; however, the findings are consistent with previous studies indicating that marijuana use plays a bigger role in IPV perpetration than alcohol use among men (e.g., Feingold et al., 2008; Shorey et al., 2018). Given the racial/ethnic and socioeconomic disparities related to IPV, the use of an ethnically/racially diverse sample may have the benefit of improving minority representation in IPV research.

Third, transitions between three IPV types over time were examined. Overall, IPV appears to remain somewhat stable over time with 67% of men remaining in the same IPV class, which is consistent with previous findings on stability in IPV among young men (Capaldi et al., 2003; Shortt et al., 2012). However, a positive shift also emerged: among movers, men tended to transition into a less severe IPV class (63%). It may be in part, as they mature, they grow into more stable relationships (Barry et al., 2009). Given that the data captures short-term changes in IPV (i.e., over a 2-year period), future studies should explore temporal changes in IPV using longitudinal data collected over a longer period of time. Although testing the effect of substance use on the transition probabilities was not possible because of the small sample size, the descriptive statistics suggested that regular marijuana users were more likely to transition to a more severe IPV class (e.g., no IPV → physical aggression class; psychological aggression → physical aggression class), compared to non-marijuana users, indicating that regular marijuana use appears to be a risk factor for IPV. The finding corroborates previous research showing a significant association between men’s marijuana use and IPV perpetration (Shorey et al., 2018).

Limitations and Future Research

The current research findings must be interpreted in light of some limitations. First, the use of the CTS (Straus, 1979) to assess IPV could limit interpretations. Although many studies have used this scale, it fails to indicate contexts in which aggression occurs (Ackerman, 2017), such as whether the physical aggression class includes a primary perpetrator with a self-defending victim or whether both partners battle for control. Thus, future studies should consider using assessments that measure contextual factors and motivations for aggression. Furthermore, the original CTS did not include items on sexual violence, which may have limited the ability to detect different patterns of IPV. Second, the study sample was comprised of mostly minority men living in one urban jurisdiction in New York. Thus, the generalizability of these findings may be somewhat limited. However, the benefit of an ethnically/racially diverse, urban sample may offset this limitation. Third, by relying exclusively on men’s reports of partner’s IPV and substance use, this study may inaccurately assess partner’s behaviors. Given poor agreement between partners about the presence of different types of IPV (Freeman et al., 2015), future studies should collect dyadic data from both partners to accurately capture bidirectional IPV and the effects of substance use on relationship dynamics. Also, substance use measures were based on self-report, which are prone to reporting error. Fourth, there are surely a number of confounding variables that lead to both substance use and IPV. Future research is needed to identify, measure, and control for these confounding variables so that our identification of salient associations can be tested as potential causal effects. Related to this, the associations between substance use and IPV can be bidirectional. Although the current research explored the prospective impact of substance use at Wave 13 on IPV at Wave 14, it is possible that IPV impacts substance use (e.g., Simmons et al., 2015). Thus, future work would benefit from more intensive assessments over time to determine the causal associations between substance use and IPV. Lastly, the data used in this study were collected in the mid-2000s. Changing sociocultural contexts and recent legalization of marijuana in some states may affect the associations between substance use and IPV.

Despite the limitations, the present study adds new insights to IPV research. IPV involves complex partner dynamics, contexts, and consequences. This study used a person-centered approach to understand IPV patterns and tested their associations with substance use over time. Findings of the present study may help researchers and clinicians to identify IPV patterns that could escalate into more severe violence, with adverse mental and physical outcomes. Furthermore, the current study underscores that marijuana use may exacerbate IPV. Although several states have recently legalized marijuana for recreational use based on arguments that it is one of the least dangerous illicit drugs (Lachenmeier & Rehm, 2015), the findings showed its potential deleterious impact on IPV. Thus, it may be important for clinicians to screen for marijuana use among mutually violent couples and to explore the couple dynamics around marijuana use. Continued investigation of the association between IPV and substance use would greatly benefit public health and prevention/intervention efforts to reduce IPV.

Acknowledgment

I thank Dr. Kimberly L. Henry for her help in writing the manuscript.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for the Rochester Youth Development Study has been provided by the National Institute on Drug Abuse (R01DA020195 and R01DA005512), the Office of Juvenile Justice and Delinquency Prevention (86-JN-CX-0007, 96-MU-FX-0014, and 2004-MU-FX-0062), the National Science Foundation (SBR-9123299), and the National Institute of Mental Health (R01MH56486 and R01MH63386). Technical assistance for this project was also provided by an NICHD grant (R24HD044943) to The Center for Social and Demographic Analysis at the University at Albany. Dr. Lee’s time to work on this study was funded in part by NIDA (R01DA020195). Points of view or opinions in this document are those of the authors and do not necessarily represent the official position or policies of the funding agencies.

Biography

Hyanghee Lee is a postdoctoral scholar at the Edna Bennett Pierce Prevention Research Center at Pennsylvania State University. She will be joining the Educational Psychology department at the University of North Texas as an assistant professor in the fall, 2022. Her research interests include individual and family development and program evaluation.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

  1. Ackerman J (2017). Assessing conflict tactics scale validity by examining intimate partner violence overreporting. Psychology of Violence, 8(2), 207–217. 10.1037/vio0000112 [DOI] [Google Scholar]
  2. Anderson KL (2002). Perpetrator or victim? Relationships between intimate partner violence and well-being. Journal of Marriage and Family, 64(4), 851–863. 10.1111/j.1741-3737.2002.00851.x [DOI] [Google Scholar]
  3. Asparouhov T, & Muthén B (2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 329–341. 10.1080/10705511.2014.915181 [DOI] [Google Scholar]
  4. Barry CMN, Madsen SD, Nelson LJ, Carroll JS, & Badger S (2009). Friendship and romantic relationship qualities in emerging adulthood: Differential associations with identity development and achieved adulthood criteria. Journal of Adult Development, 16(4), 209–222. 10.1007/s10804-009-9067-x [DOI] [Google Scholar]
  5. Busch AL, & Rosenberg MS (2004). Comparing women and men arrested for domestic violence: A preliminary report. Journal of Family Violence, 19(1), 49–57. 10.1023/B:JOFV.0000011582.05558.2e [DOI] [Google Scholar]
  6. Cafferky BM, Mendez M, Anderson JR, & Stith SM (2018). Substance use and intimate partner violence: A meta-analytic review. Psychology of Violence, 8(1), 110–131. 10.1037/vio0000074 [DOI] [Google Scholar]
  7. Capaldi DM, Shortt JW, & Crosby L (2003). Physical and psychological aggression in at-risk young couples: Stability and change in young adulthood. Merrill-Palmer Quarterly, 49(1), 1–27. 10.1353/mpq.2003.0001 [DOI] [Google Scholar]
  8. Center for Disease Control for Prevention. (2020, October 9). Intimate Partner Violence. https://www.cdc.gov/violenceprevention/intimatepartnerviolence/index.html
  9. Cerdá M, Mauro C, Hamilton A, Levy NS, Santaella-Tenorio J, Hasin D, Wall MM, Keyes KM, & Martins SS (2020). Association between recreational marijuana legalization in the United States and changes in marijuana use and cannabis use disorder from 2008 to 2016. JAMA Psychiatry, 77(2), 165–171. 10.1001/jamapsychiatry.2019.3254 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Collins LM, & Lanza ST (2010). Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences. Wiley. [Google Scholar]
  11. Dasgupta A, Silverman J, Saggurti N, Ghule M, Donta B, Battala M, Nair S, Gajanan V, & Raj A (2018). Understanding men’s elevated alcohol use, gender equity ideologies, and intimate partner violence among married couples in rural India. American Journal of Men’s Health, 12(4), 1084–1093. 10.1177/1557988318775844 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. de Bruijn DM, & de Graaf IM (2016). The role of substance use in same-day intimate partner violence: A review of the literature. Aggression and Violent Behavior, 27, 142–151. 10.1016/j.avb.2016.02.010 [DOI] [Google Scholar]
  13. Devries KM, Child JC, Bacchus LJ, Mak J, Falder G, Graham K, Watts C, & Heise L (2014). Intimate partner violence victimization and alcohol consumption in women: A systematic review and meta-analysis. Addiction, 109(3), 379–391. 10.1111/add.12393 [DOI] [PubMed] [Google Scholar]
  14. Fals-Stewart W, Golden J, & Schumacher JA (2003). Intimate partner violence and substance use: A longitudinal day-to-day examination. Addictive Behaviors, 28(9), 1555–1574. 10.1016/j.addbeh.2003.08.035 [DOI] [PubMed] [Google Scholar]
  15. Feingold A, Kerr DCR, & Capaldi DM (2008). Associations of substance use problems with intimate partner violence for at-risk men in long-term relationships. Journal of Family Psychology, 22(3), 429–438. 10.1037/0893-3200.22.3.429 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Feingold A, Washburn IJ, Tiberio SS, & Capaldi DM (2014). Changes in the associations of heavy drinking and drug use with intimate partner violence in early adulthood. Journal of Family Violence, 30(1), 27–34. 10.1007/s10896-014-9658-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Fleming PJ, McCleary-Sills J, Morton M, Levtov R, Heilman B, & Barker G (2015). Risk factors for men’s lifetime perpetration of physical violence against intimate partners: Results from the International Men and Gender Equality Survey (IMAGES) in eight countries. PLoS One, 10(3), e0118639. 10.1371/journal.pone.0118639 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Freeman AJ, Schumacher JA, & Coffey SF (2015). Social desirability and partner agreement of men’s reporting of intimate partner violence in substance abuse treatment settings. Journal of Interpersonal Violence, 30(4), 565–579. 10.1177/0886260514535263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gilbert L, El-Bassel N, Chang M, Shaw SA, Wu E, & Roy L (2013). Risk and protective factors for drug use and partner violence among women in emergency care. Journal of Community Psychology, 41(5), 565–581. 10.1002/jcop.21557 [DOI] [Google Scholar]
  20. Grigorian HL, Brem MJ, Garner A, Florimbio AR, Wolford-Clevenger C, & Stuart GL (2020). Alcohol use and problems as a potential mediator of the relationship between emotion dysregulation and IPV perpetration. Psychology of Violence, 10(1), 91. 10.1037/VIO0000237 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Henning K, & Feder L (2004). A comparison of men and women arrested for domestic violence: Who presents the greater threat? Journal of Family Violence, 19(2), 69–80. 10.1023/B:JOFV.0000019838.01126.7c [DOI] [Google Scholar]
  22. Johnson MP (2006). Conflict and control: Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12(11), 1003–1018. 10.1177/1077801206293328 [DOI] [PubMed] [Google Scholar]
  23. Johnson MP, & Leone JM (2005). The differential effects of intimate terrorism and situational couple violence: Findings from the national violence against women survey. Journal of Family Issues, 26(3), 322–349. 10.1177/0192513X04270345 [DOI] [Google Scholar]
  24. Juarros-Basterretxea J, Herrero J, Escoda-Menéndez P, & Rodríguez-Díaz FJ (2022). Cluster B personality traits and psychological intimate partner violence: Considering the mediational role of alcohol. Journal of Interpersonal Violence, 37(3–4), NP1566–NP1587. 10.1177/0886260520922351 [DOI] [PubMed] [Google Scholar]
  25. Kelly JB, & Johnson MP (2008). Differentiation among types of intimate partner violence: Research update and implications for interventions. Family Court Review, 46(3), 476–499. 10.1111/j.1744-1617.2008.00215.x [DOI] [Google Scholar]
  26. Lachenmeier DW, & Rehm J (2015). Comparative risk assessment of alcohol, tobacco, cannabis and other illicit drugs using the margin of exposure approach. Scientific Reports, 5, 8126. 10.1038/srep08126 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Langhinrichsen-Rohling J, Misra TA, Selwyn C, & Rohling ML (2012). Rates of bidirectional versus unidirectional intimate partner violence across samples, sexual orientations, and race/ethnicities: A comprehensive review. Partner Abuse, 3(2), 199–230. 10.1891/1946-6560.3.2.199 [DOI] [Google Scholar]
  28. Lawrence E, & Bradbury TN (2007). Trajectories of change in physical aggression and marital satisfaction. Journal of Family Psychology, 21(2), 236–247. 10.1037/0893-3200.21.2.236 [DOI] [PubMed] [Google Scholar]
  29. Lipsky S, & Caetano R (2011). Intimate partner violence perpetration among men and emergency department use. Journal of Emergency Medicine, 40(6), 696–703. 10.1016/j.jemermed.2008.04.043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lorber MF, & Daniel O’Leary K (2004). Predictors of the persistence of male aggression in early marriage. Journal of Family Violence, 19(6), 329–338. 10.1007/s10896-004-0678-5 [DOI] [Google Scholar]
  31. Low S, Tiberio SS, Shortt JW, Capaldi DM, & Eddy JM (2017). Associations of couples’ intimate partner violence in young adulthood and substance use: A dyadic approach. Psychology of Violence, 7(1), 120–127. 10.1037/vio0000038 [DOI] [Google Scholar]
  32. Lynch KR, & Renzetti CM (2020). Alcohol use, hostile sexism, and religious self-regulation: Investigating risk and protective factors of IPV perpetration. Journal of Interpersonal Violence, 35(17–18), 3237–3263. 10.1177/0886260517708758 [DOI] [PubMed] [Google Scholar]
  33. Moore TM, Stuart GL, Meehan JC, Rhatigan DL, Hellmuth JC, & Keen SM (2008). Drug abuse and aggression between intimate partners: A meta-analytic review. Clinical Psychology Review, 28(2), 247–274. 10.1016/j.cpr.2007.05.003 [DOI] [PubMed] [Google Scholar]
  34. Muthén LK, & Muthén BO (1998–2017). Mplus User’s Guide. Eighth Edition. Los Angeles, CA: Muthén & Muthén. [Google Scholar]
  35. Orpinas P, Hsieh HL, Song X, Holland K, & Nahapetyan L (2013). Trajectories of physical dating violence from middle to high school: Association with relationship quality and acceptability of aggression. Journal of Youth and Adolescence, 42(4), 551–565. 10.1007/s10964-012-9881-5 [DOI] [PubMed] [Google Scholar]
  36. Ryoo JH, Wang C, Swearer SM, Hull M, & Shi D (2018). Longitudinal model building using latent transition analysis: An example using school bullying data. Frontiers in Psychology, 9(May), 675. 10.3389/fpsyg.2018.00675 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Saint-Eloi Cadely H, Pittman JF, Pettit GS, Lansford JE, Bates JE, Dodge KA, & Holtzworth-Munroe A (2020). Classes of intimate partner violence from late adolescence to young adulthood. Journal of Interpersonal Violence, 35(21–22), 4419–4443. 10.1177/0886260517715601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shorey RC, Haynes E, Brem M, Florimbio AR, Grigorian H, & Stuart GL (2018). Marijuana use is associated with intimate partner violence perpetration among men arrested for domestic violence. Translational Issues in Psychological Science, 4(1), 108–118. 10.1037/tps0000140 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Shorey RC, Stuart GL, McNulty JK, & Moore TM (2014). Acute alcohol use temporally increases the odds of male perpetrated dating violence: A 90-day diary analysis. Addictive Behaviors, 39(1), 365–368. 10.1016/j.addbeh.2013.10.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shortt JW, Capaldi DM, Kim HK, Kerr DCR, Owen LD, & Feingold A (2012). Stability of intimate partner violence by men across 12 years in young adulthood: Effects of relationship transitions. Prevention Science, 13(4), 360–369. 10.1007/s11121-011-0202-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Simmons SB, Knight KE, & Menard S (2015). Consequences of intimate partner violence on substance use and depression for women and men. Journal of Family Violence, 30(3), 351–361. 10.1007/s10896-015-9691-0 [DOI] [Google Scholar]
  42. Smith SG, Zhang X, Basile KC, Merrick MT, Wang J, Kresnow M, & Chen J (2018). The National Intimate Partner and Sexual Violence Survey (NISVS): 2015 data brief – Updated Release. National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. [Google Scholar]
  43. Stalans LJ, & Ritchie J (2008). Relationship of substance use/abuse with psychological and physical intimate partner violence: Variations across living situations. Journal of Family Violence, 23(1), 9–24. 10.1007/s10896-007-9125-8 [DOI] [Google Scholar]
  44. Straus MA (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and the Family, 41(1), 75–88. 10.2307/351733 [DOI] [Google Scholar]
  45. Stuart GL, Temple JR, Follansbee KW, Bucossi MM, Hellmuth JC, & Moore TM (2008). The role of drug use in a conceptual model of intimate partner violence in men and women arrested for domestic violence. Psychology of Addictive Behaviors, 22(1), 12–24. 10.1037/0893-164X.22.1.12 [DOI] [PubMed] [Google Scholar]
  46. Substance Abuse and Mental Health Services Administration. (2019). 2018 NSDUH Detailed Tables. https://www.samhsa.gov/data/report/2018-nsduh-detailed-tables
  47. Testa M, Kubiak A, Quigley BM, Houston RJ, Derrick JL, Levitt A, Homish GG, & Leonard KE (2012). Husband and wife alcohol use as independent or interactive predictors of intimate partner violence. Journal of Studies on Alcohol and Drugs, 73(2), 268–276. 10.15288/jsad.2012.73.268 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES