Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 29.
Published in final edited form as: J Affect Disord. 2018 Jul 3;239:192–200. doi: 10.1016/j.jad.2018.07.009

Heterogeneity in emotion regulation difficulties among women victims of domestic violence: A latent profile analysis

Nicole H Weiss a, Angela G Darosh a, Ateka A Contractor b, Shannon R Forkus a, Katherine L Dixon-Gordon c, Tami P Sullivan d,*
PMCID: PMC12038377  NIHMSID: NIHMS2073354  PMID: 30014959

Abstract

Background:

Research over the past two decades supports emotion regulation as a transdiagnostic factor related to the etiology, maintenance, and treatment of a wide range of psychiatric difficulties and risky behaviors. However, prior investigations are limited by their focus on difficulties regulating negative (but not positive) emotions. Further, research has not accounted for the heterogeneity in difficulties regulating emotions.

Methods:

Participants were 210 female victims of domestic violence (DV; M age = 36.14, 48.6% African American) who completed measures assessing emotion regulation (Difficulties in Emotion Regulation Scale; Difficulties in Emotion Regulation Scale – Positive), posttraumatic stress disorder (PTSD; Posttraumatic Stress Diagnostic Scale), depression (Center for Epidemiologic Studies-Depression Scale), alcohol misuse (Alcohol Use Disorder Identification Test) and drug misuse (Drug Abuse Screening Test). Latent profile analysis was utilized to identify subgroups of DV-victimized women who were similar in endorsed difficulties in regulating negative and positive emotions. Differences in psychiatric difficulties (i.e., PTSD and depressive symptom severity) and risky behaviors (i.e., alcohol and drug misuse) were examined across these classes.

Results:

Three classes of DV-victimized women differentiated by levels of difficulties regulating negative and positive emotions were identified. Greater psychiatric difficulties were found among classes defined by higher levels of difficulties regulating emotions, regardless of emotion valence. Risky behaviors were more prevalent among the class defined by higher levels of difficulties regulating both negative and positive emotions.

Limitations:

Although results add to the literature on difficulties regulating emotions and their correlates, findings must be interpreted in light of limitations present including use of a cross-sectional and correlation design, reliance on self-report measures, and assessment of a select sample of women victims of DV.

Conclusions:

Results highlight the potential importance of tailoring interventions accounting for the heterogeneity in negative and positive emotion regulation dimensions among DV-victimized women.

Keywords: Difficulties regulating emotions, Latent profile analysis, Posttraumatic stress disorder symptom severity, Depressive symptom severity, Alcohol misuse, Drug misuse

1. Introduction

Research over the past two decades supports emotion regulation as a transdiagnostic factor related to the etiology, maintenance, and treatment of psychopathology (Tull and Aldao, 2015b). Difficulties in emotion regulation have been theoretically and empirically linked to a wide range of psychiatric difficulties (Gratz and Tull, 2010), such as posttraumatic stress disorder (PTSD; Tull et al., 2007; Weiss et al., 2013) and depressive (Dixon-Gordon et al., 2015b; Tull and Gratz, 2008) symptom severity, and risky behaviors (Weiss et al., 2015b,2012b), such as alcohol (Dvorak et al., 2014; Messman-Moore and Ward, 2014) and drug (Bonn-Miller et al., 2008; Tull et al., 2015) misuse. However, existing research is limited by its focus on difficulties regulating negative emotions, despite evidence that individuals also experience difficulties regulating positive emotions (Cyders et al., 2007; Gruber and Moskowitz, 2014; Weiss et al., 2015a). Further, past studies have relied on variable-centered approaches that do not account for the heterogeneity in patterns of emotion regulation difficulties within individuals (Thompson, 1994). Addressing these critical limitations, we aimed to (1) identify subgroups of individuals based on their constellation of endorsed difficulties regulating negative and positive emotions, and (2) examine differences in psychiatric difficulties and risky behaviors across these subgroups.

Although a growing body of research provides support for the clinical utility of examining difficulties regulating positive emotions and their relation to psychopathology, there is limited research in this area (with certain exceptions; e.g., mania; Gruber, 2011; Gruber et al., 2011,2008). Individuals experience difficulties regulating positive emotions that parallel the difficulties observed in negative emotions (Cyders et al., 2007; Weiss et al., 2015a). For example, they may be non-accepting of positive emotional states (Weiss et al., 2015a), judging some positive emotions to be undesirable, unpredictable, or frightening (Beblo et al., 2013; Kissen, 1986; Weiss et al., 2015a), possibly because they induce physiological arousal (Litz et al., 2000) that is experienced as distressing (Roemer et al., 2001). Individuals may also experience difficulties inhibiting impulsive behaviors in the context of positive emotions (Cyders et al., 2007; Weiss et al., 2015a). For instance, intense positive emotions may result in approach-related action urges (Gable and Harmon-Jones, 2008), which, in turn, may increase risk for alcohol and drug use (Berg et al., 2015; Coskunpinar et al., 2013). Finally, positive emotions may interfere with one’s ability to engage in goal-directed behaviors (Weiss et al., 2015a). Indeed, positive emotions increase distractibility (Dreisbach and Goschke, 2004), which may increase the risk for disadvantageous decision-making focused on short-versus long-term goals (Slovic et al., 2004). Recently, measures assessing these difficulties regulating positive emotions (i.e., the Difficulties in Emotion Regulation Scale - Positive Weiss et al., 2015a and the UPPS-P Positive Urgency Cyders et al., 2007) have been developed. Early evidence provides support for the role of these difficulties regulating positive emotions in psychiatric difficulties (e.g., PTSD; Weiss et al., in press-a, 2015c) and risky behaviors (e.g., alcohol and drug misuse; Weiss et al., 2018a, in press-b, 2015c). Moreover, difficulties regulating positive emotions were found to account for a significant amount of variance in psychiatric difficulties (i.e., PTSD and depressive symptoms) above and beyond difficulties regulating negative emotions (Weiss et al., 2018b).

Additionally, little attention has been devoted to examining the heterogeneity in patterns of emotion regulation. Recently, person-centered approaches, such as latent profile/class analysis, have been used to model associations among multiple, discrete emotion regulation strategies. Such approaches classify individuals into distinct and homogeneous subgroups based on patterns of endorsed responses (Nylund et al., 2007). In contrast to variable-centered approaches (e.g., correlations), person-centered approaches function on the premise that samples and their respective populations are not homogeneous, but instead are comprised of meaningful subgroups. Therefore, results that emerge from person-centered analyses may describe subgroups that diverge from the overall sample in important ways (Bergman, 2001; von Eye and Bogat, 2006). Such research is particularly relevant to the examination of emotion regulation, since a given individual typically employs multiple strategies for managing emotions (Aldao and Nolen-Hoeksema, 2013). For instance, Lougheed and Hollenstein (2012) identified six classes characterized by varying levels of five emotion regulation strategies: reappraisal, suppression, concealing, emotional engagement, and adjusting. Classes characterized by the use of fewer emotion regulation strategies were associated with higher psychopathology (e.g., depression, anxiety) relative to those characterized by the use of more emotion regulation strategies (particularly adjusting). Likewise, a study by Dixon-Gordon et al. (2015)a provided support for five classes characterized by varying levels of seven emotion regulation strategies: acceptance, cognitive reappraisal, problem solving, experiential avoidance, expressive suppression, self-criticism, and worry/rumination. Individuals in classes characterized by more emotion regulation strategies and worry/rumination in particular reported greater levels of psychopathology (e.g., depression, borderline personality disorder [BPD]). Finally, Chesney and Gordon (2017) identified four classes characterized by varying levels of six emotion regulation strategies: acceptance, reappraisal, problem solving, avoidance, suppression, and rumination. The highest levels of PTSD symptoms were observed among individuals in classes characterized by (a) low levels of adaptive and high levels of maladaptive emotion regulation strategies, and (b) moderate levels of most adaptive and maladaptive emotion regulation strategies, with the exceptions of low problem solving and high expressive suppression. Taken together, these studies paint an inconclusive picture. Although low “adaptive” strategies and high “maladaptive” strategies were linked to psychiatric difficulties and risky behaviors, so too were consistently low levels of emotion regulation strategies, and consistently high levels of emotion regulation strategies.

The aforementioned investigations suggest the presence of classes of individuals characterized by discrete emotion regulation strategies. However, we are not aware of any studies that have utilized a person-level approach to examine emotion regulation abilities. This is an important limitation given key distinctions between conceptualizations of emotion regulation characterized by the use of emotion regulation strategies versus emotion regulation abilities (Tull and Aldao, 2015a). The former aligns with Gross’s (2015) model of emotion regulation, and suggests that the type and timing of emotion regulation strategies impact emotions and their expression. Conversely, models of ability (Gratz and Roemer, 2004) conceptualize emotion regulation as the dispositional ways in which individuals understand, regard, and respond to their emotions. Tull and Aldao (2015a) proposed that emotion regulation abilities are a higher order process that determines the nature and success of emotion regulation strategies. Therefore, perhaps a focus on emotion regulation abilities will provide a clearer picture of how patterns of emotion regulation link to psychiatric difficulties and risky behaviors.

While not yet empirically studied, there is theoretical reason to believe that heterogeneity in emotion regulation abilities exist (Thompson, 1994). For instance, subgroups of individuals may be particularly prone to elevations on certain domains of emotion regulation difficulties. Linehan (1993) described individuals with BPD as “emotion-phobic” because they commonly fear intense negative affective states. Subsequent work suggests that individuals with BPD attempt to suppress both negative and positive emotions (Beblo et al., 2013), indicating that they may be non-accepting of any intense emotion. Similarly, individuals who fear physiological arousal, such as those with panic disorder (Barlow, 2004) and PTSD (Taylor et al., 1992), have been found to avoid negative and positive emotions (Roemer et al., 2001; Tull, 2006), suggesting that they may be non-accepting of any experience that elicits physiological arousal. Conversely, individuals with an alcohol or drug use disorder may be more likely to report difficulties controlling impulsive behaviors when experiencing intense emotions (Fox et al., 2007; Fox et al., 2008). Indeed, evidence suggests that intense emotions heighten risk for substance use and relapse (Baker et al., 2004; Khantzian, 1997; Marlatt and Donovan, 2005). Lastly, some groups may be particularly prone to distractibility, such as individuals with depression (Lemelin et al., 1997), which may result in difficulties engaging in goal-directed behaviors when experiencing intense emotions.

The above literature highlights the (a) documented associations of psychopathology with difficulties regulating both negative and positive emotions; (b) utility of person-centered approaches, underscoring the importance of examining heterogeneity among samples and their respective populations; (c) observed heterogeneity in emotion regulation strategies use; and (d) theoretical premise for heterogeneity in dispositional difficulties regulating negative and positive emotions. These extant lines of work underscore the need for research that utilizes a person-centered approach to identify patterns that may exist across difficulties regulating negative and positive emotions, and examines whether these patterns relate to differences in psychiatric difficulties and risky behaviors. Understanding the heterogeneity of emotion regulation difficulties and their relation to psychiatric difficulties and risky behaviors may inform the development or refinement of tailored intervention efforts (Bogat et al., 2005). Moreover, it may identify subtypes of individuals at heightened risk for psychiatric difficulties and risky behaviors.

Of note, a relative dearth of research has examined difficulties regulating emotions – or their relation to psychiatric difficulties and risky behaviors – among women victims of domestic violence (DV). DV is often chronic, with most victims reporting revictimization by their partners (Cattaneo and Goodman, 2005), and victims of chronic inter-personal, vs. single-event, traumas display greater difficulties regulating negative emotions (Ehring and Quack, 2010), suggesting that women victims of DV may be particularly susceptible to difficulties regulating emotions. Indeed, traumatic exposure may undermine emotion regulation capacities (Bardeen et al., 2013). For example, traumatic exposure may overwhelm one’s regulatory capacities (Cloitre et al., 2009), making it difficult to modulate intense emotions (Flett et al., 1996). Further, individuals with a history of traumatic exposure may come to rely on maladaptive emotion regulation strategies that reduce emotional distress in the short-term but have para-doxical effects in the long-term (Hayes et al., 1996). Additionally, traumatic violence may hinder a person’s ability to identify and describe emotional states (Kooiman et al., 2004). To date, only two studies have examined the relation between difficulties regulating emotions and psychopathology among women victims of DV (Lilly and Lim, 2013; Weiss et al., in press-a). Consistent with research in other populations, difficulties regulating negative and positive emotions predicted greater psychiatric difficulties. Further research on the relations of difficulties regulating both negative and positive emotions to psychopathology is needed given evidence for elevated psychiatric difficulties and risky behaviors in this population (Golding, 1999).

The goals of the current study were to identify classes characterized by different patterns of difficulties regulating negative and positive emotions and explore between-group differences in psychiatric difficulties (PTSD and depression symptom severity) and risky behaviors (alcohol and drug misuse) across these classes among women victims of DV. Given the absence of research simultaneously examining difficulties regulating negative and positive emotions, no a priori hypotheses were made regarding the nature of the classes, although we expected that classes characterized by higher levels of difficulties regulating negative and positive emotions would have greater PTSD symptom severity (Roemer et al., 2001; Weiss et al., 2013), depression symptom severity (Dixon-Gordon et al., 2015b; Smith et al., 2013), and alcohol and drug misuse (Berg et al., 2015; Coskunpinar et al., 2013).

2. Method

2.1. Participants and procedures

Data were collected as part of a larger study examining the influence of criminal orders of protection on women victims in a DV case. All procedures were reviewed and approved by the last authors’ Institutional Review Board. Women were recruited from two court-houses in an urban and a suburban New England community. Women were eligible to participate if they were a victim in a criminal DV case with a male intimate partner, if their offender was arraigned (i.e., the formal reading of a criminal charge[s] by a judge) approximately 12–15 months prior to study recruitment, and if they spoke English or Spanish. Eligibility criteria were determined via records from the Family Violence Victim Advocates Office or the State of Connecticut Judicial Branch.

Potential participants were sent a letter by our study team inviting them to participate in a confidential two-hour study. Interested participants were asked to call the study phone line in response to the mailed letter. Research assistants followed up on the recruitment letter with a phone call to those who did not respond either because the letter was returned or a call back was not received. If interested, eligible participants were scheduled to participate in an interview.

After providing written informed consent, face-to-face individual interviews were administered by trained masters- or doctoral-level female research associates or postdoctoral fellows in private offices to protect participants’ safety and confidentiality. In addition to the extensive training on DV and study administration all interviewers participated in with the study PI, all interviewers also participated in an additional 24 h of training provided by a DV service provider agency, which certified them as “battered women’s counselors” in the state where the study was conducted. Participants were told that they could take as many breaks as needed and were reminded that they could stop the study at any point without consequence. Participants were remunerated $50 for their participation and provided with a list of community resources relevant to DV, mental health and substance use, social services, employment and economic stability. Additionally, victims were debriefed and offered an opportunity to develop a detailed, individualized safety plan at the conclusion of the interview, which included identification of strategies for coping with distress.

The final sample comprised 298 women. The current study used a subsample of 210 who completed the Difficulties in Emotion Regulation Scale and Difficulties in Emotion Regulation Scale – Positive (which were added to the study after data collection began). Participants ranged in age from 18 to 66 years (M = 36.14, SD = 11.69). In terms of racial/ethnic background, 48.6% of participants (n = 102) self-identified as African American, 29.5% (n = 62) as White, 16.2% (n = 32) as Latina, and 5.7% (n = 12) as another or multiple racial/ethnic backgrounds. Most women were unemployed for over a month prior to the interview (n = 108; 51.4%); 27.1% (n = 57) were employed full-time during the past month and 21.4% (n = 45) were employed part-time during the past month. Women’s monthly household income ranged from $0 to $6400 (M = $1442.71; SD = $1075.27) and their mean level of education was 12.67 years (SD = 2.07). At the time of the study interview (i.e., 12 to 15 months after the arraignment), most women were not dating the offending partner (n = 153; 72.9%). Mean years in a relationship with the offending partner was 5.38, ranging from less than one month to 27 years (SD = 5.29).

2.2. Measures

2.2.1. Emotion regulation indicators

Difficulties regulating positive emotions.

The Difficulties in Emotion Regulation Scale – Positive (DERS-P; Weiss et al., 2015a) is a 13-item self-report measure that assesses individuals’ typical levels of emotion dysregulation across three domains: nonacceptance of positive emotions (DERS-P Nonacceptance), difficulties engaging in goal-directed behaviors when experiencing positive emotions (DERS-P Goals), and difficulties controlling impulsive behaviors when experiencing positive emotions (DERS-P Impulse). Participants rate each item using a 5-point Likert-type scale (1 = almost never, 5 = almost always). Higher scores indicate greater difficulties regulating positive emotions. The DERS-P demonstrates adequate psychometric properties (Weiss et al., 2015a). Cronbach’s α were 0.88 for the 4-item DERS-P Nonacceptance, 0.72 for the 4-item DERS-P Goals, and 0.86 for the 5-item DERS-P Impulse.

Difficulties regulating negative emotions.

The Difficulties in Emotion Regulation Scale (DERS; Gratz and Roemer, 2004) is a 36-item self-report measure that assesses individuals’ typical levels of emotion dysregulation across six domains: nonacceptance of negative emotions (DERS Nonacceptance), difficulties engaging in goal-directed behaviors when experiencing negative emotions (DERS Goals), difficulties controlling impulsive behaviors when experiencing negative emotions (DERS Impulse), limited access to emotion regulation strategies perceived as effective (DERS Strategies), lack of emotional awareness (DERS Aware), and lack of emotional clarity (DERS Clarity). The three DERS domains that correspond with the DERS-P (i.e., DERS Nonacceptance, DERS Goals, and DERS Impulse) were included in the current study. Participants rate each item using a 5-point Likert-type scale (1 = almost never, 5 = almost always). Higher scores indicate greater difficulties regulating negative emotions. The DERS demonstrates adequate psychometric properties (Gratz and Roemer, 2004). Cronbach’s α were 0.87 for the 6-item DERS Nonacceptance, 0.85 for the 5-item DERS Goals, and 0.89 for the 6-item DERS Impulse.

2.2.2. Psychiatric difficulties

Posttraumatic stress disorder (PTSD) symptom severity.

PTSD symptom severity was measured using the Posttraumatic Stress Diagnostic Scale (PDS; Foa et al., 1997). This self-report measure assesses DSM-IV diagnostic criteria for PTSD, including Criterion A traumatic exposure and severity of past 30-day PTSD symptoms (overall and intrusion, avoidance/numbing, hyperarousal clusters). To the extent that it was possible, symptoms were assessed in relation to victimization by the offending partner. Participants rate each of the 17 symptom items using a 6-point Likert-type scale (0 = not at all or only one time, 5 = 5 or more times a week or almost always). Higher scores indicate greater PTSD symptom severity. The PDS also provides a probable PTSD diagnosis based on DSM-IV diagnostic criteria for PTSD (present/absent). The PDS demonstrates adequate psychometric properties (Foa et al., 1997). Cronbach’s α were 0.92 for the 17-item overall PDS, 0.87 for the 5-item PDS intrusion, 0.83 for the 7-item PDS avoidance/numbing, and 0.81 for the 5-item PDS hyperarousal.

Depressive symptom severity.

Past 30-day depression symptom severity was measured using the Center for Epidemiologic Studies-Depression scale (CES-D; Radloff, 1977). Participants rate each item using a 4-point Likert-type scale (0 = rarely or none of the time, 3 = most or all of the time). Higher scores indicate greater depressive symptom severity. A score of 17 or higher indicates clinical depression (Lewinsohn et al., 1997). The CES-D demonstrates adequate psychometric properties (Radloff, 1977). Cronbach’s α was 0.78 for the 20-item CES-D.

2.2.3. Risky behaviors

Alcohol misuse.

The Alcohol Use Disorder Identification Test (AUDIT; Saunders et al., 1993) is a 10-item self-report measure that was used to assess alcohol consumption, drinking behaviors, adverse reactions to drinking, and alcohol-related problems in the past 30 days. Participants rate each item using a 5-point Likert-type scale (0 = never, 4 = daily or almost daily). Higher scores indicate greater alcohol misuse. The AUDIT demonstrates adequate psychometric properties (Saunders et al., 1993). A score of 6 or higher is related to possible alcohol use disorder (Babor et al., 2001). Cronbach’s α was 0.85.

Drug misuse.

The Drug Abuse Screening Test (DAST; Skinner, 1982) is a 10-item self-report measure that was used to assess drug misuse, including occupational or relational problems, illegal activities, or regret in the past 30 days. Responses to each item have 1 (yes) and 0 (no) options. Higher scores indicate greater drug misuse. The DAST demonstrates adequate psychometric properties (Skinner, 1982). A score of 3 or higher indicates possible drug use disorder. Cronbach’s α was 0.83.

2.2.4. Demographic, relationship, and treatment characteristics

Women reported their age, race/ethnicity, income, education, and employment, as well as relationship status (dating: yes/no) and duration with offending partner. Physical victimization was measured by 12 items from the Revised Conflict Tactics Scale (CTS-2; Straus et al., 2003). Psychological victimization was measured by 14 items from the Psychological Maltreatment of Women – Short version (PMWI-S; Tolman, 1999) because this measure assesses psychological victimization more comprehensively than the CTS-2 (e.g., the PMWI-S assesses dominance and isolation whereas the CTS-2 does not). Sexual victimization was measured by the 10-item Sexual Experiences Survey (SES; Koss and Oros, 1982) because this measure assesses sexual victimization more comprehensively than the CTS-2 (e.g., the CTS-2 does not measure sexual coercion using drugs or alcohol). A referent time period of 30 days was used to assess victimization. Response options for all scales were based on CTS-2 response options coded as: 0 (never), 1 (once in the past month), 2 (twice in the past month), 4 (3–5 times in the past month), 8 (6–10 times in the past month), 11 (more than 10 times in the past month). Higher scores on the CTS-2, PMWI-S, and SES indicate greater physical, psychological, and sexual victimization, respectively. Cronbach’s αs were 0.75 for the CTS-2, 0.97 for the PMWI-S, and 0.81 for the SES. Finally, past-month treatment was assessed with two items assessing number of outpatient individual psychotherapy visits and use of psychiatric medication (no/yes).

2.3. Data analysis

Latent profile analysis (LPA), a type of person-centered mixture modeling, was used to identify classes of women with similar patterns of difficulties regulating negative and positive emotions. LPA is a statistical approach to identify the number of homogeneous groups (i.e., classes) based on continuous latent variables. Mixture models, such as LPA, are superior to the more traditional cluster analytic methods in terms of enhanced reliability and ability to examine fit indices (DiStefano and Kamphaus, 2006). Using iterative methods, LPA estimates a series of models that increase by number of classes until all plausible solutions have been exhausted.

In the current study, LPA was conducted using Latent GOLD 5.1 (Vermunt and Magidson, 2016). Models were estimated based on the DERS Nonacceptance, Goals, and Impulse subscales and DERS-P Nonacceptance, Goals, and Impulse subscales. Established recommendations guided model selection (see Berlin et al., 2014) using several criteria for deciding on the number of LPA classes (fit indices, model class sizes, parsimony, and theory).

Model solutions were compared using the fit indices of Bayesian Information Criterion (BIC; Schwartz, 1978), Akaike’s Information Criterion (AIC; Akaike, 1987), and entropy (Ramaswamy et al., 1993). BIC and AIC are interpreted in the same way, with lower values indicating better model fit (Hu and Bentler, 1999). Entropy, an indicator of classification accuracy, ranges from 0 to 1. Entropy values closer to one suggest greater probability that participants are successfully placed into the correct latent class (Ramaswamy et al., 1993). The bootstrap likelihood ratio test (BLRT) was also calculated to compare improvement in model fit between neighboring models. Lastly, the size of the smallest class was examined, as classes with fewer than 25 participants may limit power and precision (Lubke and Neale, 2006). To characterize the classes, we examined differences in DERS and DERS-P scores using one-way analyses of variance (ANOVAs) with partial eta square effect sizes. Further, we examined between-class differences in past-month psychotherapy and psychiatric medications using ANOVA and chi-square analyses. Posterior class probabilities were imported into SPSS Version 24 for these analyses.

After the best-fitting LPA model was selected, we examined between-group differences in levels of psychiatric difficulties (PTSD and depression symptom severity) and risky behaviors (alcohol and drug misuse) across the classes of the optimal model. To accomplish this aim, and consistent with Vermunt (2010) and Bakk et al. (2013) we used the three-step approach (multivariate multinomial logistic regression) in Latent GOLD 5.1 (Vermunt and Magidson, 2016) to estimate class membership in relation to auxiliary variables of interest while accounting for misspecification bias.

3. Results

During the 30 days prior to the study interview, 139 women (66.2%) reported being victimized by psychological DV, 21 women (10.0%) reported being victimized by physical DV, and 16 women (7.7%) reported being victimized by sexual DV. Nearly one-quarter of the women (n = 51; 24.3%) met DSM-IV diagnostic criteria for PTSD, with PDS scores ranging from 0 to 48 (M = 12.88, SD = 11.64). Eighty-one women (38.6%) reported depressive symptoms consistent with clinical depression, with CES-D scores ranging from 0 to 53 (M = 14.35, SD = 12.29). Nearly one-fifth of the women reported drug (n = 37; 17.7%) and alcohol (n = 45; 21.4%) misuse consistent with an alcohol and drug use disorder, respectively, with AUDIT scores ranging from 0 to 31 (M = 7.23, SD = 5.01) and DAST scores ranging from 0 to 9 (M = 0.79, SD = 1.65).

3.1. Identification and characteristics of latent classes

LPA models indicated that the three-class solution was the optimal model (see Table 1 and Fig. 1). The BIC and AIC values for the three-class solution were lower than the corresponding values of previous class solutions. In addition, the BIC and AIC values for subsequent solutions exhibited flattening; that is, decreases in fit indices were much smaller than those observed from the one- to two- and two- to three-class solutions compared to the consecutive class comparisons (DiStefano and Kamphaus, 2006). Further, while BLRT values were significant across solutions, the three-class solution was preferred to subsequent solutions in the interest of parsimony. Moreover, the three-class solution was optimal in regards to classification accuracy, power, and precision. The entropy value for the three-class solution was large (0.98), indicating that the latent classes were highly discriminating (Muthén, 2004). In addition, the smallest class in the three-class solution was acceptable in size (See Table 1), while solutions with more classes did not satisfy this criterion; this suggests that the three-class solution is superior in power and precision.

Table 1.

Fit statistics for the latent profile analysis.

Model Fit statistics
Log Likelihood AIC BIC Entropy BLRT 1 2 3 4 5 6
1 −3150.90 6325.79 6365.96 1.00 210
2 −1986.11 4022.22 4105.89 0.99 <0.001 161 49
3 −1692.90 3461.79 3588.98 0.98 <0.001 136 47 27
4 −1589.14 3280.28 3450.98 0.97 <0.001 136 45 22 7
5 −1538.77 3205.54 3419.75 0.98 136 45 22 7 0
6 −1440.92 3035.84 3293.57 0.97 <0.001 136 30 17 14 9 4
Open in a new tab

Note. AIC = Akaike Information Criteria. BIC = Bayesian Information Criteria. BLRT = Bootstrap Likelihood Ratio Test.

Fig. 1.

Fig. 1.

Open in a new tab

Latent profile solution for DERS and DERS-Positive Subscales.

Note. DERS = Difficulties in Emotion Regulation Scale. DERS + = Difficulties in Emotion Regulation Scale-Positive. Latent classes are denoted via plotlines.

To characterize the classes, mean frequencies of difficulties regulating negative and positive emotions for each class were compared to the sample means of those variables to determine relative severity (i.e., high severity > sample mean; low severity < sample mean), an approach that allowed us to examine the conceptual fit of the model (see Table 2). Class 1 (“Lower ER”) was characterized by lower difficulties regulating both negative and positive emotions (65%; n = 136); Class 2 (“Higher Negative ER”) was characterized by higher difficulties regulating negative emotions and lower difficulties regulating positive emotions (with the exception of moderate DERS-P Goals; 22%; n = 47); and Class 3 (“Higher ER”) was characterized by higher difficulties regulating both negative and positive emotions (13%; n = 27). ANOVAs examining between-group differences in DERS and DERS-P scores provided support for these classifications with one exception: moderate DERS Accept was found for Class 2 (see Table 2).

Table 2.

Between-group differences in DERS and DERS-Positive Subscales for the 3-class model and overall sample.

Latent profile analysis classes
Class 1 (n = 136) Class 2 (n = 47) Class 3 (n = 27) Overall Sample (n = 210) F n 2
M (SD) CI M (SD) CI M (SD) CI M (SD) CI
DERS subscales
Nonacceptance 9.88 (4.35)a 9.02–10.73 13.06 (6.03)b 11.60–14.53 16.52 (6.58)c 14.59–18.45 11.44 (5.58) 16.37*** 0.18
Goals 10.65 (4.92)a 9.84–11.46 12.06 (4.54)b 10.68–13.44 16.04 (4.60)bc 14.22–17.86 11.66 (5.10) 14.43*** 0.12
Impulse 9.94 (4.79)a 9.11–10.78 11.40 (4.74)b 9.98–12.83 16.41 (5.96)bc 14.53–18.28 11.10 (5.36) 19.39*** 0.16
DERS-Positive subscales
Nonacceptance 4.00 (0.00)a 3.77–4.23 4.43 (0.65)b 4.03–4.82 7.11 (3.75)bc 6.59–7.63 4.50 (1.70) 58.83*** 0.36
Goals 4.00 (0.00)a 3.80–4.20 5.68 (1.32)b 5.34–6.02 8.04 (2.85)c 7.59–8.49 4.90 (1.82) 143.79*** 0.58
Impulse 5.00 (0.00)a 4.84–5.16 5.00 (0.00)b 4.73–5.27 8.15 (2.67)bc 7.79–8.51 5.40 (1.42) 130.17*** 0.56
Open in a new tab

Note. DERS = Difficulties in Emotion Regulation Scale. DERS-Positive = Difficulties in Emotion Regulation Scale-Positive.

*

p ≤ .05.

**

p ≤ .01.

***

p ≤ .001.

abc

Means that do not share subscripts differ by p < .05 based on post-hoc Bonferroni pairwise comparisons.

Additionally, there were some significant between-group differences in past-month psychotherapy services and psychiatric medication utilization. A significant difference was found for past-month number of outpatient individual psychotherapy visits (F [2, 207] = 3.20, p = .04, η2 = 0.03), with individuals in the Higher ER Class reporting significant more outpatient individual psychotherapy visits (M = 3.11, SD = 6.59) than individuals in the Lower ER class (M = 1.18, SD = 3.64). Further, individuals in the Higher ER class reported the highest rates of psychiatric medication use (55.6%), followed by individuals in the Higher Negative ER class (31.9%) and then individuals in the Lower ER class (21.3%; X2 [2] = 13.51, p = .001).

3.2. Between-class differences in psychiatric difficulties and risky behaviors

See Table 3 for between-class differences in psychiatric difficulties and risky behaviors.

Table 3.

Between-group differences in psychiatric difficulties and risky behaviors for the 3-classes and overall sample.

Class 1 (n = 136) Class 2 (n = 47) Class 3 (n = 27) Overall sample (n = 210) Pairwise comparisons
M (SD) M (SD) M (SD) M (SD)
PTSD symptom severity 9.96 (10.59)a 8.15–11.76 14.89 (10.14)b 11.83–17.96 24.11 (11.92)c 20.06–28.16 12.88 (11.64) 1 = 2, 1 < 3, 2 < 3*
Depressive symptom severity 11.22 (10.56)a 9.34–13.11 16.13 (11.61)b 12.92–19.34 27.00 (13.15)c 22.77–31.23 14.35 (12.29) 1 = 2, 1 < 3, 2 < 3
Alcohol misuse 5.74 (5.29)a 4.28–7.20 7.48 (6.95)ab 4.95–10.01 10.71 (8.24)bc 7.47–13.96 6.77 (6.29) 1 = 2, 1 < 3, 2 = 3
Drug misuse* 0.58 (1.43)a 0.30–0.85 0.91 (1.64)ab 0.45–1.38 1.63 (2.34)bc 1.02–2.24 0.79 (1.65) 1 = 2, 1 < 3*, 2 = 3
Open in a new tab

Note. PTSD = posttraumatic stress disorder.

*

p < .08.

PTSD symptom severity.

“Higher ER” had greater PTSD symptom severity compared to “Lower ER” (Wald = 8.90; p = .003); the difference for “Higher Negative ER” approached significance (Wald = 3.71; p = .05).

Depressive symptom severity.

“Higher ER” had greater depression symptom severity compared to “Lower ER” (Wald = 12.19; p = .0004) and “Higher Negative ER” (Wald = 3.99; p = .046).

Alcohol misuse.

Greater alcohol problems were found for “Higher ER” compared to “Lower ER” (Wald = 4.59, p = .03).

Drug misuse.

The difference between “Higher ER” and “Lower ER” approached significance (Wald = 2.99; p = .08).

4. Discussion

The current study extended extant research by identifying classes based on domains of difficulties regulating negative and positive emotions and exploring differences in psychiatric difficulties (PTSD and depressive symptom severity) and risky behaviors (alcohol and drug misuse) across these classes. Results provide support for three emotion regulation classes that were found to demonstrate unique relations with psychiatric difficulties and risky behaviors. These findings suggest the potential value of person-centered approaches for identifying patterns of difficulties regulating emotions and risk for psychiatric difficulties and risky behaviors.

Results provide support for three classes characterized by (1) lower difficulties regulating both negative and positive emotions, (2) higher difficulties regulating negative emotions and lower difficulties regulating positive emotions, and (3) higher difficulties regulating both negative and positive emotions. This finding extends theory on difficulties regulating emotions by suggesting that higher levels of difficulties regulating positive emotions may occur only among individuals who exhibit high levels of difficulties regulating negative emotions. While not hypothesized, this result is not entirely surprising: groups that display difficulties regulating positive emotions (e.g., individuals with BPD, PTSD, and panic disorder; Beblo et al., 2013; Roemer et al., 2001; Tull, 2006) also report difficulties regulating negative emotions (Gratz et al., 2006; Tull, 2006; Weiss et al., 2013). It may be that some mechanisms underlying difficulties regulating positive emotions mirror those for difficulties regulating negative emotions. For instance, individuals may be fearful of intense emotions regardless of valence (as is theorized in BPD; Linehan, 1993) or may experience distress in the context of physiological arousal regardless of its source (e.g., stimuli eliciting negative or positive emotions; as is theorized in PTSD; Taylor et al., 1992). Conversely, higher levels of difficulties regulating negative emotions were found among individuals with both higher and lower levels of difficulties regulating positive emotions. Using the same logic, some of the mechanisms underlying difficulties regulating negative emotions may differ from those for difficulties regulating positive emotions. For example, the personality traits of high neuroticism, low conscientiousness, and low agreeableness – which characterize difficulties regulating negative emotions (Settles et al., 2012) – may be less relevant to our understanding of difficulties regulating positive emotions. Future research is needed to better understand the relations among dimensions of difficulties regulating negative and positive emotions, including their co-occurrence and underlying mechanisms.

Notably, the three class-solution had construct validity wherein levels of psychiatric difficulties and risky behaviors differed significantly across the classes. Regarding psychiatric difficulties, women in classes defined by higher levels of difficulties regulating emotions, regardless of emotion valence, exhibited greater PTSD and depressive symptom severity. This finding is consistent with extant literature on the role of difficulties regulating emotions in PTSD and depressive symptom severity. Specifically, prior studies have found an association between higher levels of difficulties regulating negative emotions and greater PTSD and depressive symptom severity (Dixon-Gordon et al., 2015b; Tull et al., 2007; Tull and Gratz, 2008; Weiss et al., 2012a). Likewise, research provides support for higher levels of difficulties regulating positive emotions and greater PTSD symptom severity (Weiss et al., in press-a, 2015c), and diffiuclties controlling impulsive behaviors when experiencing positive emotions (the one studied dimension of difficulties regulating positive emotions) among individuals with greater depression symptom severity (Smith et al., 2013). The above evidence may suggest that higher levels of difficulties regulating negative or positive emotions relate to PTSD and depressive symptom severity. Alternatively – given evidence here that classes characterized by higher levels of difficulties regulating negative emotions versus difficulties regulating both negative and positive emotions did not significantly differ from one another in terms of PTSD and depressive symptom severity – it may be that higher levels of difficulties regulating negative emotions alone drive the psychiatric difficulties examined here. Future research is needed to explore the relative and unique contributions of difficulties regulating negative and positive emotions to psychiatric difficulties. Such findings would have important implications for treatment, identifying the utility of targeting difficulties regulating negative and positive emotions versus difficulties regulating negative emotions alone (as is current practice) in PTSD and depression treatments. Moreover, additional research in this area would inform theory development and research.

Regarding differences in risky behaviors, results suggest greater alcohol and drug misuse among individuals in classes characterized by higher levels of difficulties regulating both negative and positive emotions. Difficulties regulating negative emotions are speculated to primarily relate to negative reinforcement; individuals who experience difficulties regulating negative emotions may engage in alcohol or drug use in an attempt to alleviate or distract themselves from emotional states perceived as aversive (Baker et al., 2004; Khantzian, 1997). This finding aligns with the self-medication hypothesis (Khantzian, 1997), which suggests that individuals may use alcohol and/or drugs to reduce or manage trauma-related symptoms and distress, demonstrated among women victims of DV (Kaysen et al., 2007). Conversely, difficulties regulating positive emotions are thought to relate to positive reinforcement, such that alcohol and drug use may function to elicit, maintain, or enhance positive emotion states (Cooper et al., 1995; Cooper et al., 2016; Cox and Klinger, 1988). Thus, one explanation for our finding may be that alcohol and drug use are more problematic when driven by negative and positive reinforcement (versus negative reinforcement alone, as may be the case with the class characterized by higher levels of difficulties regulating negative, but not positive, emotions). Future research examining the contributions of negative and positive reinforcement to alcohol and drug misuse may inform interventions for this population. Nevertheless, our findings indicate the clinical importance of assessing difficulties regulating positive emotions, which is not standard in clinical practice. Moreover, they suggest that prevention and intervention efforts targeting alcohol and drug misuse may benefit from a focus on both negative and positive emotion regulation.

Although results of the present study add to the literature on difficulties regulating emotions and their correlates, findings must be interpreted in light of limitations present. First, the cross-sectional and correlational nature of the data precludes determination of the precise nature and direction of the relationships examined here. Future research should investigate the nature and direction of these relationships through prospective, longitudinal investigations. Second, this study relied on women’s self-report, which may have been influenced by their ability and/or willingness to report accurately. Third, individuals in classes characterized by Higher ER reported a greater number of past month outpatient individual psychotherapy visits and higher rates of past month psychiatric medication use. Future research would benefit from further exploring the influence of engagement in mental health treatment on the nature and construct validity of emotion regulation typologies. Indeed, as is reviewed in Gratz et al. (2015), a growing number of treatments have been developed to target emotion regulation, and reductions in emotion regulation following treatment are associated with improvements in psychiatric difficulties and risky behaviors. Further, our assessment of psychotherapy did not assess DV services in particular; investigation of the impact of such services on emotion regulation typologies among DV-victimized women is of particular importance. Fourth, although our focus on female victims of DV is a strength of this study, our findings may not generalize to non-DV-victimized women and other DV populations (e.g., men, women in same sex relationships), and thus require replication in these populations. For instance, women victims of DV likely differ in meaningful ways from other individuals who experience psychiatric difficulties and risky behaviors; given the often chronic and ongoing nature of their traumatic experiences (Cattaneo and Goodman, 2005), they may exhibit fear (and at higher levels) related to victimization and suppress their emotional experiences to prevent revictimization, both of which may exacerbate difficulties regulating emotions. Finally, extension of the effect of emotion regulation typologies on other psychiatric difficulties (e.g., BPD, anxiety) and risky behaviors (e.g., nonsuicidal self-injury) is warranted. For instance, while our assessments of PTSD and depression capture some symptoms of anxiety (e.g., irritability, restlessness, sleep difficulties), additional research is needed to explore levels of anxiety among classes characterized by difficulties regulating negative and positive emotions. Indeed, while anxiety has been shown to be associated with greater difficulties regulating negative emotions (Tull et al., 2009), we are not aware of any studies exploring its relation to difficulties regulating positive emotions.

Uniquely, the present study utilized a person-centered approach to extend research on difficulties regulating emotions and their correlates. Results provided support for three classes defined by varying levels of dimensions of difficulties regulating negative and positive emotions. Findings indicate that these classes had differential relations with psychiatric difficulties and risky behaviors. These results highlight the potential importance of tailoring diagnostic assessments and interventions accounting for the heterogeneity in positive and negative emotion regulation dimensions.

Funding

This project was supported by award no. 2012-IJ-CX-0045 by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. Work on this paper by the first author was also supported by National Institute on Drug Abuse grants K23DA039327 and L30DA038349. The opinions, findings, and conclusions or recommendations in this manuscript are those of the authors and do not necessarily reflect those of the Department of Justice or the National Institute on Drug Abuse.

Footnotes

Conflict of interest

The authors have no conflicts of interest to report.

References

  1. Akaike H, 1987. Factor analysis and AIC. Psychometrika 52, 317–332. [Google Scholar]
  2. Aldao A, Nolen-Hoeksema S, 2013. One versus many: capturing the use of multiple emotion regulation strategies in response to an emotion-eliciting stimulus. Cognit. Emotion 27, 753–760. [DOI] [PubMed] [Google Scholar]
  3. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG, World Health Organization, 2001. AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Health Care.
  4. Baker TB, Piper ME, McCarthy DE, Majeskie MR, Fiore MC, 2004. Addiction motivation reformulated: an affective processing model of negative reinforcement. Psychol. Rev 111, 33–51. [DOI] [PubMed] [Google Scholar]
  5. Bakk Z, Tekle FB, Vermunt JK, 2013. Estimating the association between latent class membership and external variables using bias-adjusted three-step approaches. Sociological Method. 43, 272–311. [Google Scholar]
  6. Bardeen JR, Kumpula MJ, Orcutt HK, 2013. Emotion regulation difficulties as a prospective predictor of posttraumatic stress symptoms following a mass shooting. J. Anxiety Disord 27, 188–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Barlow DH, 2004. Anxiety and its disorders: The Nature and Treatment of Anxiety and Panic. Guilford Press. [Google Scholar]
  8. Beblo T, Fernando S, Kamper P, Griepenstroh J, Aschenbrenner S, Pastuszak A, …, Driessen M, 2013. Increased attempts to suppress negative and positive emotions in borderline personality disorder. Psychiatry Res. 210, 505–509. [DOI] [PubMed] [Google Scholar]
  9. Berg JM, Latzman RD, Bliwise NG, Lilienfeld SO, 2015. Parsing the heterogeneity of impulsivity: a meta-analytic review of the behavioral implications of the UPPS for psychopathology. Psychol. Assess 27, 1129–1146. [DOI] [PubMed] [Google Scholar]
  10. Bergman LR, 2001. A person approach in research on adolescence: some methodological challenges. J. Adolesc. Res 16, 28–53. [Google Scholar]
  11. Berlin KS, Parra GR, Williams NA, 2014. An introduction to latent variable mixture modeling (part 2): longitudinal latent class growth analysis and growth mixture models. J. Pediatr. Psychol 39, 188–203. [DOI] [PubMed] [Google Scholar]
  12. Bogat GA, Levendosky AA, Von Eye A, 2005. The future of research on intimate partner violence: person-oriented and variable-oriented perspectives. Am. J. Community. Psychol 36, 49–70. [DOI] [PubMed] [Google Scholar]
  13. Bonn-Miller MO, Vujanovic AA, Zvolensky MJ, 2008. Emotional dysregulation: association with coping-oriented marijuana use motives among current marijuana users. Subst. Use Misuse 43, 1653–1665. [DOI] [PubMed] [Google Scholar]
  14. Cattaneo LB, Goodman LA, 2005. Risk factors for reabuse in intimate partner violence: a cross-disciplinary critical review. Trauma Violence Abuse 6, 141–175. [DOI] [PubMed] [Google Scholar]
  15. Chesney SA, Gordon NS, 2017. Profiles of emotion regulation: understanding regulatory patterns and the implications for posttraumatic stress. Cognit. Emotion 31, 598–606. [DOI] [PubMed] [Google Scholar]
  16. Cloitre M, Stolbach BC, Herman JL, Kolk BVD, Pynoos R, Wang J, Petkova E, 2009. A developmental approach to complex PTSD: childhood and adult cumulative trauma as predictors of symptom complexity. J. Trauma Stress 22, 399–408. [DOI] [PubMed] [Google Scholar]
  17. Cooper ML, Frone MR, Russell M, Mudar P, 1995. Drinking to regulate positive and negative emotions: a motivational model of alcohol use. J. Pers. Soc. Psychol 69, 990–1005. [DOI] [PubMed] [Google Scholar]
  18. Cooper ML, Kuntsche E, Levitt A, Barber LL, Wolf S, 2016. Motivational Models of Substance Use: A Review of Theory and Research on Motives for Using Alcohol, Marijuana, and Tobacco. Oxford University Press, New York, NY. [Google Scholar]
  19. Coskunpinar A, Dir AL, Cyders MA, 2013. Multidimensionality in impulsivity and alcohol use: a meta-analysis using the UPPS model of impulsivity. Alcohol. Clin. Exp. Res 37, 1441–1450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cox WM, Klinger E, 1988. A motivational model of alcohol use. J. Abnorm. Psychol 97, 168–180. [DOI] [PubMed] [Google Scholar]
  21. Cyders MA, Smith GT, Spillane NS, Fischer S, Annus AM, Peterson C, 2007. Integration of impulsivity and positive mood to predict risky behavior: development and validation of a measure of positive urgency. Psychol. Assess 19, 107–118. [DOI] [PubMed] [Google Scholar]
  22. DiStefano C, Kamphaus RW, 2006. Investigating subtypes of child development: a comparison of cluster analysis and latent class cluster analysis in typology creation. Educ. Psychol. Meas 66, 778–794. [Google Scholar]
  23. Dixon-Gordon KL, Aldao A, De Los Reyes A, 2015a. Repertoires of emotion regulation: a person-centered approach to assessing emotion regulation strategies and links to psychopathology. Cognit. Emotion 29, 1314–1325. [DOI] [PubMed] [Google Scholar]
  24. Dixon-Gordon KL, Weiss NH, Tull MT, DiLillo D, Messman-Moore T, Gratz KL, 2015b. Characterizing emotional dysfunction in borderline personality, major depression, and their co-occurrence. Compr. Psychiatry 62, 187–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Dreisbach G, Goschke T, 2004. How positive affect modulates cognitive control: reduced perseveration at the cost of increased distractibility. J. Exp. Psychol. Learn. Memory. Cognit 30, 343–353. [DOI] [PubMed] [Google Scholar]
  26. Dvorak RD, Sargent EM, Kilwein TM, Stevenson BL, Kuvaas NJ, Williams TJ, 2014. Alcohol use and alcohol-related consequences: associations with emotion regulation difficulties. Am. J. Drug Alcohol Abuse 40, 125–130. [DOI] [PubMed] [Google Scholar]
  27. Ehring T, Quack D, 2010. Emotion regulation difficulties in trauma survivors: the role of trauma type and PTSD symptom severity. Behav. Therapy 41, 587–598. [DOI] [PubMed] [Google Scholar]
  28. Flett GL, Blankstein KR, Obertynski M, 1996. Affect intensity, coping styles, mood regulation expectancies, and depressive symptoms. Pers. Individ. Differences 20, 221–228. [Google Scholar]
  29. Foa EB, Cashman L, Jaycox L, Perry K, 1997. The validation of a self-report measure of posttraumatic stress disorder: the posttraumatic diagnostic scale. Psychol. Assess 9, 445–451. [Google Scholar]
  30. Fox HC, Axelrod SR, Paliwal P, Sleeper J, Sinha R, 2007. Difficulties in emotion regulation and impulse control during cocaine abstinence. Drug Alcohol Depend. 89, 298–301. [DOI] [PubMed] [Google Scholar]
  31. Fox HC, Hong KA, Sinha R, 2008. Difficulties in emotion regulation and impulse control in recently abstinent alcoholics compared with social drinkers. Addict. Behav 33, 388–394. [DOI] [PubMed] [Google Scholar]
  32. Gable PA, Harmon-Jones E, 2008. Approach-motivated positive affect reduces breadth of attention. Psychol. Sci 19, 476–482. [DOI] [PubMed] [Google Scholar]
  33. Golding JM, 1999. Intimate partner violence as a risk factor for mental disorders: a meta-analysis. J. Family Violence 14, 99–132. [Google Scholar]
  34. Gratz KL, Roemer L, 2004. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difficulties in emotion regulation scale. J. Psychopathol. Behav. Assess 26, 41–54. [Google Scholar]
  35. Gratz KL, Rosenthal MZ, Tull MT, Lejuez CW, Gunderson JG, 2006. An experimental investigation of emotion dysregulation in borderline personality disorder. J. Abnorm. Psychol 115, 850–855. [DOI] [PubMed] [Google Scholar]
  36. Gratz KL, Tull MT, 2010. Emotion Regulation as a Mechanism of Change in Acceptance-and Mindfulness-Based Treatments. In: Baer RA (Ed.), Assessing Mindfulness and Acceptance: Illuminating the Theory and Practice of Change. New Harbinger Publications, Oakland, CA, pp. 105–133. [Google Scholar]
  37. Gratz KL, Weiss NH, Tull MT, 2015. Examining emotion regulation as an outcome, mechanism, or target of psychological treatments. Curr. Opin. Psychol 3, 85–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Gross JJ, 2015. Emotion regulation: current status and future prospects. Psychol. Inquiry 26, 1–26. [Google Scholar]
  39. Gruber J, 2011. Can feeling too good be bad? Positive emotion persistence (PEP) in bipolar disorder. Curr. Dir. Psychol. Sci 20, 217–221. [Google Scholar]
  40. Gruber J, Eidelman P, Johnson SL, Smith B, Harvey AG, 2011. Hooked on a feeling: rumination about positive and negative emotion in inter-episode bipolar disorder. J. Abnorm. Psychol 120, 956–961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Gruber J, Johnson SL, Oveis C, Keltner D, 2008. Risk for mania and positive emotional responding: too much of a good thing? Emotion 8, 23–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Gruber J, Moskowitz JT, 2014. Positive Emotion: Integrating the Light Sides and Dark Sides. Oxford University Press, Oxford, United Kingdom. [Google Scholar]
  43. Hayes SC, Wilson KG, Gifford EV, Follette VM, Strosahl K, 1996. Experiential avoidance and behavioral disorders: a functional dimensional approach to diagnosis and treatment. J. Consult. Clin. Psychol 64, 1152–1168. [DOI] [PubMed] [Google Scholar]
  44. Hu L, Bentler PM, 1999. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct. Equ. Model 6, 1–55. [Google Scholar]
  45. Kaysen D, Dillworth TM, Simpson T, Waldrop A, Larimer ME, Resick PA, 2007. Domestic violence and alcohol use: trauma-related symptoms and motives for drinking. Addict. Behav 32, 1272–1283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Khantzian EJ, 1997. The self-medication hypothesis of substance use disorders: a re-consideration and recent applications. Harv. Rev. Psychiatry 4, 231–244. [DOI] [PubMed] [Google Scholar]
  47. Kissen Morton, 1986. Characterological aspects of depression in borderline patients. Curr. Issues Psychoanal. Pract 2, 45–63. [Google Scholar]
  48. Kooiman CG, van Rees Vellinga S, Spinhoven P, Draijer N, Trijsburg RW, Rooijmans HG, 2004. Childhood adversities as risk factors for alexithymia and other aspects of affect dysregulation in adulthood. Psychother. Psychosom 73, 107–116. [DOI] [PubMed] [Google Scholar]
  49. Koss MP, Oros CJ, 1982. Sexual experiences survey: a research instrument investigating sexual aggression and victimization. J. Consult. Clin. Psychol 50, 455–457. [DOI] [PubMed] [Google Scholar]
  50. Lemelin S, Baruch P, Vincent A, Everett J, Vincent P, 1997. Distractibility and processing resource deficit in major depression: evidence for two deficient attentional processing models. J. Nerv. Ment. Disord 185, 542–548. [DOI] [PubMed] [Google Scholar]
  51. Lewinsohn PM, Seeley JR, Roberts RE, Allen NB, 1997. Center for epidemiologic studies depression scale (CES-D) as a screening instrument for depression among community-residing older adults. Psychol. Aging 12, 277–287. [DOI] [PubMed] [Google Scholar]
  52. Lilly MM, Hong Phylice Lim B, 2013. Shared pathogeneses of posttrauma pathologies: attachment, emotion regulation, and cognitions. J. Clin. Psychol 69, 737–748. [DOI] [PubMed] [Google Scholar]
  53. Linehan MM, 1993. Cognitive Behavioral Treatment of Borderline Personality Disorder. Guilford Press, New York, NY. [Google Scholar]
  54. Litz BT, Orsillo SM, Kaloupek D, Weathers F, 2000. Emotional processing in posttraumatic stress disorder. J. Abnorm. Psychol 109, 26–39. [DOI] [PubMed] [Google Scholar]
  55. Lougheed JP, Hollenstein T, 2012. A limited repertoire of emotion regulation strategies is associated with internalizing problems in adolescence. Soc. Dev 21, 704–721. [Google Scholar]
  56. Lubke G, Neale MC, 2006. Distinguishing between latent classes and continuous factors: resolution by maximum likelihood? Multivariate Behav. Res 41, 499–532. [DOI] [PubMed] [Google Scholar]
  57. Marlatt GA, Donovan DM, 2005. Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors. Guilford Press. [Google Scholar]
  58. Messman-Moore TL, Ward RM, 2014. Emotion dysregulation and coping drinking motives in college women. Am. J. Health Behav 38, 553–559. [DOI] [PubMed] [Google Scholar]
  59. Muthén DH, 2004. Latent variable analysis: Growth mixture modeling and related techniques for longitudinal data. In: Kaplan D (Ed.), Handbook of quantitative methodology for the social sciences. Sage, Newbury Park, CA, pp. 345–368. [Google Scholar]
  60. Nylund K, Bellmore A, Nishina A, Graham S, 2007. Subtypes, severity, and structural stability of peer victimization: what does latent class analysis say? Child Dev. 78, 1706–1722. [DOI] [PubMed] [Google Scholar]
  61. Radloff LS, 1977. The CES-D scale a self-report depression scale for research in the general population. Appl. Psychol. Meas 1, 385–401. [Google Scholar]
  62. Ramaswamy V, DeSarbo WS, Reibstein DJ, Robinson WT, 1993. An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Market. Sci 12, 103–124. [Google Scholar]
  63. Roemer L, Litz BT, Orsillo SM, Wagner AW, 2001. A preliminary investigation of the role of strategic withholding of emotions in PTSD. J. Trauma Stress 14, 149–156. [Google Scholar]
  64. Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M, 1993. Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction 88, 791–804. [DOI] [PubMed] [Google Scholar]
  65. Schwartz G, 1978. Estimating the dimension of a mode. Ann. Stat 6, 461–464. [Google Scholar]
  66. Settles RE, Fischer S, Cyders MA, Combs JL, Gunn RL, Smith GT, 2012. Negative urgency: a personality predictor of externalizing behavior characterized by neuroticism, low conscientiousness, and disagreeableness. J. Abnorm. Psychol 121, 160–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Skinner HA, 1982. The drug abuse screening test. Addict. Behav 7, 363–371. [DOI] [PubMed] [Google Scholar]
  68. Slovic P, Finucane ML, Peters E, MacGregor DG, 2004. Risk as analysis and risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk Anal. 24, 311–322. [DOI] [PubMed] [Google Scholar]
  69. Smith GT, Guller L, Zapolski TCB, 2013. A comparison of two models of urgency: urgency predicts both rash action and depression in youth. Clin. Psychol. Sci 3, 266–275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Straus MA, Hamby SL, Warren WL, 2003. The Conflict Tactics Scales Handbook. Western Psychological Services, Los Angeles, CA. [Google Scholar]
  71. Taylor S, Koch WJ, McNally RJ, 1992. How does anxiety sensitivity vary across the anxiety disorders? J. Anxiety Disord 6, 249–259. [Google Scholar]
  72. Thompson RA, 1994. Emotion regulation: a theme in search of definition. Monogr. Soc. Res. Child Dev 59, 25–52. [PubMed] [Google Scholar]
  73. Tolman RM, 1999. The validation of the psychological maltreatment of women inventory. Violence Vict. 14, 25–37. [PubMed] [Google Scholar]
  74. Tull MT, 2006. Extending an anxiety sensitivity model of uncued panic attack frequency and symptom severity: the role of emotion dysregulation. Cognit. Therapy Res 30, 177–184. [Google Scholar]
  75. Tull MT, Aldao A, 2015a. Editorial overview: new directions in the science of emotion regulation. Curr. Opin. Psychol 3, 4–10. [Google Scholar]
  76. Tull MT, Aldao A, 2015b. Emotion regulation. Curr. Opin. Psychol 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Tull MT, Bardeen JR, DiLillo D, Messman-Moore T, Gratz KL, 2015. A prospective investigation of emotion dysregulation as a moderator of the relation between posttraumatic stress symptoms and substance use severity. J. Anxiety Disord 29, 52–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Tull MT, Barrett HM, McMillan ES, Roemer L, 2007. A preliminary investigation of the relationship between emotion regulation difficulties and posttraumatic stress symptoms. Behav. Therapy 38, 303–313. [DOI] [PubMed] [Google Scholar]
  79. Tull MT, Gratz KL, 2008. Further examination of the relationship between anxiety sensitivity and depression: the mediating role of experiential avoidance and difficulties engaging in goal-directed behavior when distressed. J. Anxiety Disord 22, 199–210. [DOI] [PubMed] [Google Scholar]
  80. Tull MT, Stipelman BA, Salters-Pedneault K, Gratz KL, 2009. An examination of recent non-clinical panic attacks, panic disorder, anxiety sensitivity, and emotion regulation difficulties in the prediction of generalized anxiety disorder in an analogue sample. J. Anxiety Disord 23, 275–282. [DOI] [PubMed] [Google Scholar]
  81. Vermunt JK, 2010. Latent class modeling with covariates: Two improved three-step approaches. Political Anal. 18, 450–469. [Google Scholar]
  82. von Eye A, Bogat GA, 2006. Person-oriented and variable-oriented research: concepts, results, and development. Merrill Palmer Q. 52, 390–420. [Google Scholar]
  83. Weiss NH, Dixon-Gordon KL, Peasant C, Sullivan TP, in press-a. An examination of the role of difficulties regulating positive emotions in posttraumatic stress disorder. J. Trauma. Stress [DOI] [PMC free article] [PubMed] [Google Scholar]
  84. Weiss NH, Forkus SR, Contractor AA, Schick MR, 2018a. Difficulties regulating positive emotions and alcohol and drug misuse: A path analysis. Addict. Behav 84, 45–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  85. Weiss NH, Gratz KL, Lavender J, 2015a. Factor structure and initial validation of a multidimensional measure of difficulties in the regulation of positive emotions: the DERS-positive. Behav. Modif 39, 431–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Weiss NH, Nelson R, Contractor AA, Sullivan TP, 2018b. Emotion dysregulation and posttraumatic stress disorder severity: A test of the incremental role of difficulties regulating positive emotions. Manuscript submitted for publication. [DOI] [PMC free article] [PubMed]
  87. Weiss NH, Sullivan TP, Tull MT, 2015b. Explicating the role of emotion dysregulation in risky behaviors: a review and synthesis of the literature with directions for future research and clinical practice. Curr. Opin. Psychol 3, 22–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  88. Weiss NH, Tull MT, Anestis MD, Gratz KL, 2013. The relative and unique contributions of emotion dysregulation and impulsivity to posttraumatic stress disorder among substance dependent inpatients. Drug Alcohol Depend. 128, 45–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Weiss NH, Tull MT, Davis LT, Dehon EE, Fulton JJ, Gratz KL, 2012a. Examining the association between emotion regulation difficulties and probable posttraumatic stress disorder within a sample of African Americans. Cogn. Behav. Ther 41, 5–14. [DOI] [PubMed] [Google Scholar]
  90. Weiss NH, Tull MT, Dixon-Gordon KL, Gratz KL, in press-b. Assessing the negative and positive emotion-dependent nature of risky behaviors among substance dependent patients. Assessment. [DOI] [PMC free article] [PubMed] [Google Scholar]
  91. Weiss NH, Tull MT, Sullivan TP, Dixon-Gordon KL, Gratz KL, 2015c. Posttraumatic stress disorder symptoms and risky behaviors among trauma-exposed inpatients with substance dependence: the influence of negative and positive urgency. Drug Alcohol Depend. 155, 147–153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  92. Weiss NH, Tull MT, Viana AG, Anestis MD, Gratz KL, 2012b. Impulsive behaviors as an emotion regulation strategy: examining associations between PTSD, emotion dysregulation, and impulsive behaviors among substance dependent inpatients. J. Anxiety Disord 26, 453–458. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES