The initial sample consisted of 496 patients seeking treatment for SUD in the “Proyecto Hombre Navarra and ANTOX” addiction programmes in Navarra, Spain between February 2017 and June 2019. These programs have a cognitive behavioural basis and are geared towards abstinence. Study inclusion criteria were as follows: a) meeting the diagnostic criteria for SUD based on the DSM-5 (American Psychiatric Association, 2013); b) having experienced physical and/or sexual abuse; c) being older than 18 years old; d) receiving SUD treatment; and e) giving consent to participate in the study. The exclusion criteria were a) presenting serious mental illness that made it inadvisable to participate in the study (e.g., psychotic disorders); and b) not having knowledge of the Spanish language. Based on these criteria, 57 patients were selected for the study (11.5% of the initial sample). Of these subjects, 54.4% (n = 31) were men, and 45.6% (n = 26) were women. The average age of the final sample was 41.5 years (SD = 11.8). About half of the participants were single (47.4%, n = 27), had secondary studies (50.9%, n = 29), and were unemployed (56.1%, n = 32). The main substance that motivated treatment was alcohol in 45.1% (n = 25) of the final sample, followed by cocaine (22.8%, n = 13), poly consumption (21.1%, n = 12), and other substances (12.3%, n = 7) as heroin, cannabis, etc. The remaining patients were excluded for not having experienced lifetime abuse (n = 439). Two previous studies were done with part of this sample (Haro et al., 2021; López-Goñi et al., 2021).

The EuropASI (Kokkevi & Hartgers, 1995; Spanish version by Bobes et al., 1996) is the European version of the Addiction Severity Index (ASI; McLellan et al., 1980). This tool assesses the need for treatment in patients with SUD in seven different areas: (a) general medical condition, (b) employment and financial situation, (c) alcohol consumption, (d) use of other drugs, (e) legal problems, (f) family and social relationships, and (g) psychological state; scores ranging from 0 (no problem) to 9 (extreme problems). Interviewer Severity Rates (ISRs), which are directly related to the severity of consumption (López-Goñi et al., 2012), were used. Moreover, some items of the EuropASI were used to obtain specific information about the presence of psychopathological problems (psychiatric scale: items 3, 4, 6, 7, 8, 9, 10, and 10A). Thus, the variables derived from the interview considered in this study were: substance motivating treatment, the seven areas of the EuropASI, number of previous SUD treatments, depressive problems, anxiety problems, hallucinations, violence problems, suicidal ideation and attempts, and previous psychopharmacological treatments. For the assessment of lifetime physical and/or sexual abuse, two specific items from the family/social area of the EuropASI were used: 18B (“Has anyone ever physically abused you?”) and 18C (“Has anyone ever sexually abused you?”). The short-term test-retest reliabilities of the ASI severity ratings have been reported to be greater than or equal to .92 for all domains in McLellan et al. (1985).

The Symptom Checklist (SCL-90-R) (Derogatis, 1992) is a self-report tool that assesses psychopathological symptoms in the last two weeks. It is composed of 90 items that are answered on a five-point Likert scale, from 0 (never) to 4 (more than seven days). This tool aims to assess the symptoms of psychological distress. It consists of nine primary symptom dimensions: somatization (e.g., headaches), obsession-compulsion (e.g., unwanted thoughts, words, or ideas that won’t leave your mind), interpersonal sensitivity (e.g., seeing people in a negative way), depression (e.g., feeling low in energy or slowed down), anxiety (e.g., suddenly scared for no reason), hostility (e.g., feeling easily annoyed or irritated), phobic anxiety (e.g., feeling afraid to go out of your house alone), paranoid ideation (e.g., feeling uneasy when people are watching or talking about you), and psychoticism (e.g., the idea that someone else can control your thoughts); and three global indices of severity: the Global Severity Index (GSI), which reflects overall symptom severity; the Positive Symptom Distress Index (PSDI), which indicates symptom intensity; and the Positive Symptom Total (PST), which includes the number of items answered with a score different from 0. In this sample the internal consistency ranges from .70 to .90.

The Maladjustment Scale (Echeburúa et al., 2000) reveals how each patient is currently affected in six different areas of everyday life: labour, social, leisure, partner, family and general (e.g., “because of my current problems, my usual social life has been affected...”). This scale evaluates maladjustment in each area and it ranges from 0 (nothing) to 5 (extremely) on a six-point Likert scale. The total scale range is 0-30. Higher scores indicate higher levels of maladjustment. In this sample the internal consistency is .94.

The Severity Posttraumatic Stress Disorder Scale - Revised (EGS-R; Echeburúa et al., 2016) is a Spanish semi structured interview that evaluates the presence of PTSD and the severity of the symptoms. It is based on the diagnostic criteria of the DSM-5. It is composed of two parts: a) exposure to a traumatic event and b) severity and frequency of PTSD symptoms. It consists of 21 items scored from 0 (nothing/never) to 3 (extremely/5 or more times a week) on a four-point Likert scale. The variables evaluated were type of abuse, frequency of the traumatic event, previous treatments, exposure during childhood/adolescence, revictimization, physical sequalae, PTSD symptoms severity, and PTSD diagnosis. In addition, six items valorise the interference of trauma in different daily life areas. In this sample the internal consistency is .91.

The ethics committees of the Universidad Pública de Navarra (code: PI-006/16), Proyecto Hombre Navarra (code: PHN2016-01), and ANTOX (code: A2017-01) approved the protocol for this study. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. All participants signed informed consent forms. Informed consent was obtained from all patients for being included in the study.

All patients were interviewed by clinical psychologists specialized in the assessment and treatment of addictions and trauma. The assessment of the sample was carried out in two sessions at the beginning of the addiction treatment. Self-reports were completed in the presence and with the support of the interviewers. During the first session, sociodemographic data and physical and/or sexual abuse variables were screened as well as the EuropASI (Kokkevi & Hartgers, 1995; Spanish version by Bobes et al., 1996). The procedure to determine the EuropASI scores was the two-step methodology suggested by Bobes et al. (2008). In the first step, the scores of each area were calculated based on the interviewer’s judgement considering the patient’s answers. In the second step, once the interview with the patient had finished, the interviewer’s scores were reviewed by the intervention team, blind to the participant’s identity. If a discrepancy was observed, the intervention team discussed it until an agreement was reached.

In the second session, the psychopathological and maladjustment consequences were assessed. An in-depth interview was carried out with patients who reported a history of physical and/or sexual abuse to confirm the traumatic event and evaluate the corresponding variables with the EGS-R (Echeburúa et al., 2016). The same researcher was in charge of these in-depth interviews. For individuals experiencing more than one traumatic event, the event with the most severe consequences on the patient was considered the principal type of abuse. The clinical judgment about the PTSD diagnosis was made considering the scores of the EGS-R (Echeburúa et al., 2016). When a doubt appeared, the researcher team discussed the symptomatology of the patient until reaching an agreement.

Descriptive analyses were conducted for all variables. Bivariate analyses were performed using χ² tests or Student’s t-tests for independent samples, depending on the nature of the variables analysed. The variance homogeneity was assessed in the case of the Student’s t-tests. Depending on the result of Levene’s test, the corresponding correction was used (Glass & Stanley, 1970). The t-test was used because data presented a random distribution (assessed by Runs test for randomness), and the size of the sample was higher than 40. The statistical analyses were two-sided. Effect sizes (Cohen’s d or Cramer’s V) for the analyses were provided. The interpretation of Cohen’s d was made following the next considerations: d < 0.20 (no effect size), 0.2 ≤ d < 0.5 (small effect size), 0.5 ≤ d < 0.80 (medium effect size) and d ≥ 0.80 (large effect size; Cohen, 1998). The interpretation of Cramer’s V was made according to Rea and Parker’s (1992) conventions: V ≤ .20 (small effect), .20 < V ≤ .60 (medium effect), and V > .60 (large effect). Statistical power post-hoc was calculated as a function of α = .05, the population effect size, and N. Scores above .80 are considered adequate for accepting the statistically significant (p < .05) differences. Regarding multivariate analyses, a logistic regression analysis (forward stepwise entry method) was conducted to determine which specific factors were the most important to differentiate between men and women. In this analysis, all the variables were included. Due the preliminary character of the study, the variable entry criterion was set to 0.05, and the variable retention criterion was set to 0.10. The proportion of the variance explained was calculated as Nagelkerke’s R². Moreover, the Hosmer-Lemeshow test was used to assess the goodness of fit of this model (Hosmer & Lemeshow, 2013). A difference of p < .05 was considered significant. Statistical analysis were performed with the SPSS program (25.0) and G*Power 3 software (Faul et al., 2007).

There were no gender differences in sociodemographic characteristics (Table 1). Comparisons between men and women in severity of consumption and psychopathological characteristics are presented in Table 2. Statistically significant differences in the substance that motivated treatment were observed (χ² = 6.0, p = .014), with a low effect size. Alcohol was the substance that motivated treatment in 43.9% (n = 25) of patients: 61.5% (n = 16) of women and 29% (n = 9) of men. Regarding addiction severity assessed by the EuropASI, men showed higher scores in drug use (t = 2.2, p < .033) and legal problem area (t = 2.6, p < .012) than women, with moderate effect sizes. Moreover, women presented a higher prevalence of previous psychopharmacological treatments than men (χ² = 6.4, p = .014), with a low effect size. Regarding psychopathological variables assessed by the SCL-90-R, men obtained significantly higher scores on the hostility dimension (t = 3.3, p = .0002), with a large effect size.

Table 1

Note. ¹In the χ² analysis of marital status, the categories “separated/divorced” and “widower” have been joined. ²In the χ² analysis of education level, the categories “no studies” and “primary”, as well as “secondary” and “university”, have been joined. ³In the χ² analysis of employment situation, the categories “unemployed” and “retired” have been joined.

Table 2

Note. SCL-90-R = Symptom Checklist; ISR = Interviewer Severity Rate; GSI = Global Severity Index; PSDI = Positive Symptom Distress Index; PST = Positive Symptom Distress.

Most patients had been victims of more than one traumatic event (n = 47, 82.5%) and had experienced the traumatic event during childhood or adolescence (n = 43, 75.4%), without gender differences. Moreover, 26.3% (n = 15) of the sample presented a diagnosis of PTSD. The results showed a differential profile in trauma-related variables between men and women (Table 3). Regarding the type of abuse, more women (n = 20, 76.9%) than men (n = 5, 16.1%) had experienced at least one event of sexual abuse, with a significant gender difference (χ² = 21.2, p < .001) and a moderate effect size. Most of men had been victims of physical abuse (n = 26, 83.9%). On the other hand, women presented a higher frequency of revictimization to another traumatic event than men (χ² = 15.0, p < .001), with a moderate effect size, and a higher presence of avoidance symptoms as a consequence of having experienced the traumatic event (t = 2.1, p < .037), with a moderate effect size.

Regarding violence perpetration against people, men (80.6%, n = 25) presented a higher percentage than women (38.5%, n = 10), which was significantly different (χ² = 10.6, p = .001) with a low effect size.

Table 3

The logistic regression analysis showed that the three variables that best differentiated women from men were having experienced sexual abuse, alcohol as the substance that motivated treatment, higher scores in depression and lower paranoid ideation scores (SCL-90-R). These three variables explained 79.6% of the variance and correctly classified 87.3% of the patients (Table 4).

Table 4

Note. SCL-90-R = Symptom Checklist.