Sage Journals: Discover world-class research

Abstract

Treatment for perpetrators of intimate partner violence (IPV) is a crucial preventive measure for IPV. However, professionals working with IPV perpetrators might be mandated by law to intervene when they suspect future IPV, creating a potential conflict for perpetrators seeking help for their problems. There is limited research on IPV perpetrators’ perspectives on mandatory reporting of intimate partner violence (MR-IPV). We aimed to add to this research by examining knowledge, experiences, and perceptions of MR-IPV among perpetrators who had voluntarily sought help for IPV and comparing them with those of IPV victims recruited from IPV help services. Forty-one IPV perpetrators who voluntarily had sought help from help services answered a questionnaire about MR-IPV. Descriptive statistics were used to examine their knowledge, experiences, and perceptions of MR-IPV. Mann–Whitney U tests, chi-square tests, and multivariate logistic regression were used to compare the IPV perpetrators with 86 victims. Results showed that though the majority of IPV perpetrators reported having been informed about MR-IPV (58.5%), few reported knowing the MR-IPV law (19.5%), though 43.9% reported knowing it to some extent. Only 17.5% had experienced MR-IPV. The IPV perpetrators varied in their perceptions of MR-IPV. More IPV victims had experienced MR-IPV compared to IPV perpetrators, and this difference remained statistically significant when controlling for conceptually informed variables (IPV persistence, IPV severity, and the presence of children), sociodemographic characteristics, and participation mode. Despite a modest sample size, our findings bring new insights concerning voluntarily help-seeking IPV perpetrators’ perspectives on MR-IPV. Our findings suggest that professionals must ensure that information about MR-IPV is adequately understood by IPV perpetrators. Despite the fact that there was a difference between perpetrators and victims in IPV characteristics, this could not account for the difference in MR-IPV experience. This suggests that the MR-IPV law might not be practiced uniformly for IPV perpetrators and victims.

Keywords

Domestic violence domestic violence perpetrator intimate partner violence intimate partner violence perpetrator legal intervention mandatory reporting perpetrator intervention

Introduction

Intimate partner violence (IPV) is a global criminal justice, health, human rights, and social problem that has detrimental effects at both individual and societal levels (Burghart and Backhaus, 2024; Council of Europe, 2014; Lagdon et al., 2014; Pedersen et al., 2023; Stubbs and Szoeke, 2022; World Health Organization [WHO], 2017). Criminologists have studied the nature of IPV (Aguilar Ruiz et al., 2023; Verbruggen et al., 2020) and measures taken in order to prevent and stop IPV from occurring (e.g., Stewart, 2000; Strand et al., 2018; Svalin et al., 2017), including interventions focused on individuals who have perpetrated IPV (hereafter called ‘perpetrators’) and the perpetrators’ perspectives on such interventions (Hilton and Radatz, 2018; Walker et al., 2017). There are several different strategies for IPV prevention. In article 16 of the Istanbul Convention (the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence, 01/08/2014), preventive intervention and treatment programs for perpetrators are specified as important measures to combat and prevent domestic violence (Council of Europe, 2011; Council of Europe, 2014; Hester and Lilley, 2014). Perpetrator programs aim to foster change in perpetrators’ attitudes and behavior and have the benefit of targeting the source of the violence. Lasting change as the result of intervention may also prevent IPV that could occur in future relationships, thus limiting the individual and societal cost of recurring IPV (Council of Europe, 2011; Hester and Lilley, 2014; Pedersen et al., 2023).

When working with perpetrators in interventions such as treatment programs, the Istanbul Convention highlights that ensuring the safety and support of the victim is ‘of primary concern’ (Council of Europe, 2014: 6). Thus, there must be a dual focus on helping the perpetrator while simultaneously monitoring the risk of new violence for the victim. In several countries, laws and regulations set thresholds for when professionals are mandated to act to ensure the safety of the victim. Mandatory reporting of IPV (MR-IPV) laws mandate police reporting or other action when there is suspicion or knowledge about IPV (WHO, 2013; WHO, 2017). Thus, professionals working with perpetrators might be mandated to act on disclosure of IPV perpetration, even though the information was given in a setting where information is usually considered confidential.

A recent systematic review found no studies on perpetrators’ perspectives on MR-IPV (Vatnar et al., 2021), and the topic remains largely unexplored. To the best of our knowledge, only two qualitative studies have examined perpetrators’ perspectives on MR-IPV (Kristiansen et al., 2024; Linge et al., 2025). Both studies originate from the same research project as the present study and examine IPV perpetrators recruited from voluntary treatment services in Norway. Kristiansen et al. (2024) interviewed perpetrators, two of whom had experienced MR-IPV. They described a breach of trust after MR-IPV was used because they perceived it as unnecessary in their situations to use MR-IPV. Linge et al. (2025) examined awareness about MR-IPV among eight IPV perpetrators and found lacking awareness of and only vague understanding of MR-IPV. Over half were not aware of the MR-IPV law. The perpetrators also described withholding information about the IPV from professionals due to shame, guilt, uncertainty about the consequences, and fear of being reported to the police and child welfare services.

Though the two qualitative studies bring important insights regarding perpetrators’ perspectives and experiences, it is uncertain to what extent the findings might generalize to a larger group. Thus, it still remains largely unknown whether perpetrators are aware that MR-IPV laws exist, whether they have experience with MR-IPV and the nature of that experience, and how they view MR-IPV. Additionally, to the best of our knowledge, no studies have compared victims’ and perpetrators’ perspectives on MR-IPV. This is important to investigate, as it might elucidate differences in the effect of MR-IPV for those who seek help for victimization and perpetration. This can, in turn, have important implications for practice. The present study aimed to examine perpetrators’ knowledge, perceptions, and experiences with MR-IPV and to compare them to victims.

The Norwegian mandatory reporting law

The present study was set in Norway, whose MR-IPV law differs from those in other countries in some important aspects. First, the law states that there is a punishable duty to ‘report or by other means avert’ future IPV in cases where an individual believes that persistent or severe IPV will most likely occur (The Penal Code, 2005). Thus, it is not reporting IPV that is the primary object of the law, but rather averting future IPV. However, reporting to the police is mentioned specifically in the law, and police involvement might be necessary to ensure that the IPV is averted. We have chosen to use the term ‘mandatory reporting of intimate partner violence’ (MR-IPV) to refer to the law to be consistent with the research literature. Secondly, the law specifies that the IPV must be persistent or severe for the duty to apply, although the IPV need not be physical to be considered severe. For instance, threats and coercion are specifically mentioned in the law (s. 196; The Penal Code, 2005). Thirdly, the duty concerns averting IPV that likely will occur, not reporting instances of IPV in the past. However, previous IPV is one of the best-documented risk factors for future IPV and thus affects the likelihood that IPV will reoccur in the future (e.g., Kuijpers et al., 2011). This brings us to the fourth point: individuals need not be certain that IPV will occur, but rather believe that it will most likely occur. Thus, there is a discretionary space in the law allowing for tailoring to the specific situation, both concerning how likely it is that the IPV will occur, and how best to avert it. Lastly, MR-IPV applies to any person, including professionals working in help services, and importantly, MR-IPV overrules confidentiality.

Perpetrator interventions, help-seeking, and confidentiality

Despite limited research on perpetrators’ views of MR-IPV, there are some studies on help-seeking and perpetrator interventions that address related topics such as confidentiality. Campbell et al. (2010) examined help-seeking among 73 male perpetrators from a community program for male batterers, 12 of whom also participated in focus group discussions. Focus group interviews and quantitative data revealed that trust and confidentiality were crucial for help-seeking. Additionally, the most highly endorsed source of help was a counselor (52% endorsement), followed by a doctor (37%). The authors noted that the finding could be interpreted as a preference for professionals who are bound by confidentiality. They also remarked that MR-IPV laws might deter help-seeking because confidentiality is no longer assured. This has also been voiced as a concern regarding victims (Jordan and Pritchard, 2021; Lippy et al., 2020; Rodríguez et al., 2002; WHO, 2013). However, it is important to highlight that MR-IPV was not mentioned as a barrier to help-seeking during the focus group discussions in Campbell et al. (2010). Similarly, a systematic synthesis of qualitative studies on male perpetrators’ perspectives on intervention and change did not mention MR-IPV as a barrier to help-seeking or program engagement (McGinn et al., 2020).

Having a strong treatment alliance has been linked to intervention success in treatment for IPV specifically (see Taft and Murphy, 2007) as well as in therapy interventions in general (Horvath et al., 2011; Wampold and Imel, 2015), and trust is an inherent part of the alliance concept (Bordin, 1979). Furthermore, Davis et al. (2021) found that voluntary help-seeking for perpetrators is a process that is constructed over time, and that building trusting relationships is an inherent part of this process. Thus, MR-IPV might be a barrier toward help-seeking or effective treatment if it adversely affects trust in help services. However, whether MR-IPV negatively impacts the trust in help services has, to the best of our knowledge, never been investigated.

Aim and research questions

The aim of the present study was to examine knowledge of, experiences with, and perceptions of MR-IPV among perpetrators who have voluntarily sought help. MR-IPV can lead to very different consequences for perpetrators and victims, which might affect how the groups view MR-IPV and how professionals deal with MR-IPV when working with the different groups. Consequently, we also wished to explore and compare perpetrators’ and victims’ knowledge, experiences, and perceptions. If there were differences, we also sought to examine if the differences would hold when controlling for conceptually important variables. According to the Norwegian MR-IPV law, MR-IPV applies when it is most likely that severe or persistent IPV will occur (The Penal Code, 2005). As previous IPV is among the most important risk factors for future violence (Kuijpers et al., 2011), IPV severity and IPV persistence were important control variables to include. Additionally, previous research has indicated that the presence of children can be a facilitator for MR-IPV compliance among professionals working with both perpetrators and victims (Vatnar et al., 2024). Thus, the presence of children was considered an important control variable. Additionally, we wished to control for other characteristics that might differ between groups, specifically sociodemographic characteristics and participation mode (i.e., digital vs. physical participation and self-report vs. structured interview; see the Methods section for further detail).

Our research questions were as follows:

What knowledge, experience, and perceptions do perpetrators have concerning MR-IPV?

To what extent do perpetrators and victims differ in knowledge, experience, and perceptions of MR-IPV?

If there are differences, do they remain statistically significant when controlling for (a) conceptually informed variables (IPV severity, IPV persistence, and the presence of children) and (b) sociodemographic characteristics and participation mode?

Methods

The present study was part of a research project called MANREPORT-IPV—a mixed-methods study examining knowledge of, attitudes toward, and experience with MR-IPV. The study was approved by Oslo University Hospital's Data Protection Official (22/00221). The Regional Committees for Medical Research Ethics (REK) considered the study to be health service research and thereby not within their mandate (257644). Both perpetrators and victims were recruited for the present study. They were not recruited as couples, but as independent individuals who had sought help from voluntary IPV help services. Perpetrators and victims completed nearly identical questionnaires, modified slightly to accommodate whether the participant was recruited as a perpetrator or a victim.

Participants

A total of 128 participants were recruited through help services for victims and perpetrators: 86 victims and 42 perpetrators. One perpetrator was excluded due to concerns about validity and a large amount of missing data. Thus, the final sample size was 127. Perpetrators were recruited by Alternative to Violence (ATV; n = 23, 56.1%), Anger Management Brøset therapists (n = 14, 34.1%), and the police (n = 4, 9.8%). The ATV and Anger Management are two different voluntary treatment providers for perpetrators (see the Procedures section for more details). Victims were recruited by domestic violence shelters (n = 60, 69.8%), the ATV (n = 16, 18.6%), and the police (n = 10, 11.6%). See Table 1 for the sociodemographic and IPV characteristics of the participants.

Table 1.

Sociodemographic and IPV characteristics of IPV perpetrators (n = 41) and IPV victims (n = 86).

Variable	IPV perpetrators			IPV victims
	n	Mean (range)	SD	n	Mean (range)	SD
Age	41	40.1 (24–81)	12.0	86	43.1 (21–67)	9.1
Years of education (including primary school)	40	14.9 (10–21)	2.7	83	14.6 (6–25)	3.4
Duration of last relationship with IPV (in years)	38	9.8 (0.08–34.9)	7.7	85	11.3 (0.25–33)	8.7
IPV severity—CTS2
Psychological aggression	41	50.4 (0–132)	34.9	82	95.4 (0–200)	43.6
Physical assault	39	7.7 (0–102)	17.1	85	50.8 (0–300)	72.7
Sexual coercion	41	0.2 (0–4)	0.7	85	15.6 (0–75)	24.5
Injury	40	1.9 (0–27)	4.6	85	16.4 (0–127)	27.6

	n	% (n)	n	% (n)
Gender (man)^a	41	82.9 (34)	86	15.1 (13)
Citizenship—native citizen	39	79.5 (31)	85	80.0 (68)
Naturalized citizen^b		10.3 (4)		10.6 (9)
Foreign citizen		10.3 (4)		9.4 (8)
Has children	41	82.9 (34)	86	94.2 (81)
Employment—employed/student	41	65.9 (27)	85	57.6 (49)
Employed and social security benefits		4.9 (2)		11.8 (10)
Social security benefits only/unemployed		29.3 (12)		30.6 (26)
Health—good/very good	41	56.1 (23)	86	37.2 (32)
Neither poor nor good		29.3 (12)		33.7 (29)
Poor/very poor		14.6 (6)		29.1 (25)
IPV persistence—yes	39	35.9 (14)	84	81.0 (68)
To some extent		15.4 (6)		8.3 (7)
No		48.7 (19)		10.7 (9)

Note: Percentages refer to percentage of the n for the item. IPV: intimate partner violence; CTS2: Revised Conflict Tactics Scale.

It was possible to respond ‘other’ and elaborate on gender identity, but no participant used this response option.

Norwegian citizens with immigrant background.

Measures

Sociodemographic variables were drawn from previous IPV studies in Norway (e.g., Vatnar et al., 2017) and included the following variables: gender, age, citizenship, whether the participants had children, employment status, years of education, and a global question asking about the participants’ general health. Possible response options are detailed in Table 1.

Experience with MR-IPV

MR-IPV experience was measured by asking the participants two questions: ‘Has someone from the help services used mandatory reporting regarding you and IPV with your consent?’ and ‘Has someone from the help services used mandatory reporting regarding you and IPV without your consent?’ The response options were no, to some extent, and yes. Yes or to some extent responses to one of the questions was considered indicative of MR-IPV experience. To some extent responses were included to increase the probability that instances where the help services had averted IPV by other means than police reporting were included, and because the responses could indicate MR-IPV with consent secured to some extent.

Knowledge of MR-IPV

Perceived knowledge of MR-IPV was measured by the question, ‘Do you think you know the mandatory reporting law?’ Response options included no, to some extent, and yes. Two questions were used to examine whether participants had been informed about MR-IPV: ‘Has anyone from the help services informed you about mandatory reporting?’ and ‘If you have a care provider, has your care provider informed you about mandatory reporting?’ Response options were no, to some extent, and yes. An overall measure based on the two questions was created. In cases where the responses differed between the two questions, the score indicating the highest extent of having been informed was used for the overall measure.

Perception of MR-IPV

The measure for MR-IPV perception was based on a measure used in a different study in the MANREPORT-IPV research project (Nordby et al., 2024). In that study, a factor analysis was conducted using the responses from 265 professionals who had worked with victims and/or perpetrators. This factor analysis resulted in a factor comprising seven items called skepticism to MR-IPV. Of these, four items were identical or very similar to items in the questionnaire for perpetrators and victims in the present study: (1) Everyone has the right to a physician/therapist/healthcare provider/contact person with absolute client confidentiality; (2) I would lose trust in the doctor/care provider/help service if they disclosed information; (3) I would think they were unprofessional if they disclosed confidential information; (4) The patient's wish matters most; if I don't want to report to the police, the professionals shouldn’t do it. Response options were completely disagree, partly disagree, partly agree, and completely agree. These four items had the highest factor loadings of the seven items in the original study (see Nordby et al., 2024). As the scale comprising the four items had acceptable Cronbach's alpha (α = .805), it was used in the present study (see Table 1 in the Supplementary Materials for the inter-item correlation matrix). It was termed perception of MR-IPV rather than skepticism of MR-IPV, as it is not identical to the factor in Nordby et al.'s study. Scores on the items were summed and averaged, producing a mean score ranging between 1 and 4. Higher scores indicate higher agreement with the statements and thus a more negative perception of MR-IPV.

IPV severity: The Revised Conflict Tactics Scales (CTS2)

The Norwegian version of the Revised Conflict Tactics Scales (Bendixen, 2005; CTS2; Straus et al., 1996) was used to measure IPV severity. The CTS2 has demonstrated good validity and reliability and is among the most widely used measures of IPV (Chapman and Gillespie, 2019; Costa and Barros, 2016). It measures the number of times in the past year the participant and their partner engaged in different behaviors to deal with conflicts. There are scores for negotiation (not used in this study), psychological aggression, physical assault, sexual coercion, and injury. In this study, past-year chronicity scores were used, indicating the number of incidents of the behaviors in the scales in the last year of the relationship.

Usually, the CTS2 is scored in a way that yields separate victimization and perpetration scores for each subscale (Straus et al., 1996). However, when examining MR-IPV, the IPV severity in the relationship as a whole is the most important aspect, rather than whether the participant was subjected to or perpetrated the IPV. Using the separate perpetration/victimization scores, the most relevant information regarding IPV severity would likely be present in the victimization scores for the victims and in the perpetration scores for the perpetrators, making it difficult to compare the groups. Thus, we created a combined measure indicating the violence in the relationship as a whole. Variables for each of the CTS2 subscales were computed by using the participants’ highest score among the perpetration and victimization scores for the relevant subscale. For instance, if a participant had a score of 5 on psychological aggression victimization and 60 on perpetration, their psychological aggression score would be 60. To the best of our knowledge, such an approach has not been used in previous studies, but a similar approach has been used by researchers aiming to create one measure based on the CTS2 scores of both members of a couple, termed maximum dyadic report (Cuenca et al., 2015; O'Leary and Williams, 2006). In these studies, maximum dyadic report scores were created by using the highest score reported by either member of the couple. In the present study, the highest score among perpetration and victimization scores was used to indicate the IPV in the relationship as a whole. Cronbach's alpha for the combined CTS2 scales were .834 for Psychological Aggression, .951 for Physical Assault, .902 for Sexual Coercion, and .827 for Injury. The distribution for some of the CTS2 scores was skewed. However, as normally distributed data is not a requirement for the statistical analyses used in the study (logistic regression), no transformation of the scores was performed.

IPV persistence

IPV persistence was measured using two variables. One item measured relationship duration with the question, ‘How long has your (most recent) relationship with violence lasted?’ Participants filled out the number of years and months. Because a long duration of the IPV relationship does not guarantee that the IPV has persisted over time, an additional item was used to examine IPV persistence, modeled after a domain in the Spousal Assault Risk Assessment Guide version 3 (SARA-V3) risk assessment guide (Kropp and Hart, 2015): ‘Has the intimate partner violence persisted over a long period of time? Such as several months or years, started early in the relationship, also violence in previous relationships?’ Response options were no, to some extent, yes, and unknown. Unknown answers were treated as missing (n = 2). Participants were asked about the persistence of IPV in the past year (past year score) as well as the time before (previous score). For participants whose IPV relationship had ended more than 1 year ago, their previous score was used. For participants still in the IPV relationship or whose IPV relationship had ended within the past year, the past year score was used.

Procedures

Professionals from four different help services were approached to recruit perpetrators and victims. The first, ATV, is a foundation offering voluntary individual- and group-based psychological treatment for individuals who have perpetrated violence (IPV included). They also offer services for victims, and thus, we recruited both perpetrators and victims from ATV. The ATV has 15 offices located throughout the country, which all agreed to help recruit. Two police districts working with IPV prevention also agreed to help recruit perpetrators and victims. Additionally, we approached professionals who offered a treatment program called Anger Management (Brøset model), which offers group therapy for individuals who voluntarily seek help for anger issues (IPV included). IPV perpetrators are among those seeking help from such groups. Professionals offering Anger Management groups work in several parts of the help service system, but typically in public family welfare services. Anger Management therapists only recruited perpetrators. Lastly, 15 domestic violence shelters also recruited victims.

Inclusion criteria were having been subjected to and/or having perpetrated IPV, being age 18 or above, and having been in contact with an IPV help service. All professionals made individual assessments about whether the potential participants had perpetrated or been subjected to IPV before asking about research participation. The professionals were informed about the study and given an information letter detailing the research project that could be distributed to potential participants. If an individual was interested in participating or wanted further information, their contact information was forwarded to the first author. The participant was subsequently contacted by one of the researchers, who ensured that the participant had received and understood the information about the research project. They enquired about the participants’ preferred mode of participation and scheduled a time and place for participation. Written informed consent was obtained from all participants.

Participation in the project could be organized in several different ways, according to the participants’ preferences. Participants could choose self-report (n = 30, 73.2% perpetrators; n = 51, 59.3% victims) or having the questionnaire presented as a structured interview (n = 11, 26.8% perpetrators; n = 35, 40.7% victims). Furthermore, they could choose between digital (n = 31, 75.6% perpetrators; n = 57, 66.3% victims) and in-person (n = 10, 24.4% perpetrators; n = 29, 33.7% victims) participation. For participation in person, the participant and researcher met in an office of the help service that recruited the participant, unless the participant wished for a different location, or the help service did not have an available room. All participants met a researcher who was available for questions during participation. Three perpetrators and five victims required a professional interpreter. The average completion time was 57 (self-report) and 112 (structured interview) minutes. Data was collected between April 2022 and November 2024.

Analyses

Descriptive statistics were used to examine perpetrators’ knowledge, experiences, and perceptions concerning MR-IPV. There were four stages in the analysis process comparing perpetrators and victims, depicted in Figure 1. The analysis strategy was guided by the purposeful selection approach recommended by Hosmer and Lemeshow (2012). In the first stage, we examined differences on the MR-IPV variables using univariate tests. We performed Mann–Whitney U tests to examine MR-IPV knowledge, MR-IPV perception, and having been informed about MR-IPV. The chi-square test was used to examine differences in MR-IPV experience between perpetrators and victims. As there was a statistically significant difference in MR-IPV experience, we also performed a univariate logistic regression with MR-IPV experience, predicting belonging to the victim group. This was done in order to better compare the results of the univariate tests with multivariate tests in the later stages of the analysis process.

Figure 1.

Illustration of the four stages of the analysis process comparing IPV perpetrators and victims. Note: All logistic regression models predicted belonging to the IPV victim group. Variables in bold text reached the threshold for inclusion in the next stage in the analysis process (i.e., variables with p < .05 in Stage 1, Step 2b, Step 3b, and Stage 4; variables with p < .20 in Steps 2a and 3a). IPV: intimate partner violence; MR-IPV: mandatory reporting of intimate partner violence; CTS2: Revised Conflict Tactics Scale.

In the second stage of the analysis process, we wished to examine if the differences in the first stage would hold when controlling for conceptually informed control variables: IPV severity, IPV persistence, and the presence of children. The presence of children was included in the group of conceptually informed variables rather than among the other sociodemographic variables because previous research has indicated that the presence of children might be a facilitator for MR-IPV compliance among professionals (Vatnar et al., 2024). Thus, there were empirical grounds for including it among the variables of particular interest in this study. We created separate multivariate logistic regression models with MR-IPV experience and each of the conceptually informed control variables, predicting belonging to the victim group (Step 2a in Figure 1). First, we examined IPV persistence, creating a logistic regression model including MR-IPV experience, IPV persistence, and IPV relationship duration. Then we examined IPV severity, measured by the CTS2 scales Psychological Aggression, Physical Assault, Sexual Coercion, and Injury. As there was high multicollinearity between the CTS2 scales Physical Assault and Injury, two parallel logistic regression models were created. The first included MR-IPV experience and the CTS2 scales Psychological Aggression, Sexual Coercion, and Physical Assault. In the second model, the CTS2 scale Injury replaced the Physical Assault scale.

Furthermore, we created a model including MR-IPV experience and whether the participants had children. Finally, a combined multivariate model was created (Step 2b in Figure 1), incorporating MR-IPV experience and the variables from the other models with p < .20: IPV persistence, relationship duration, the CTS2 scales for Psychological Aggression and Sexual Coercion, and the presence of children. The threshold of p < .20 is recommended by Hosmer and Lemeshow (2012) and was chosen to ensure that all potentially confounding variables were included and to reduce the chance of type II errors.

In the third stage of the analysis process, we wished to examine if the difference in MR-IPV experience would hold when controlling for other potential differences between the groups, including both sociodemographic characteristics and participation mode (physical vs. digital participation; self-report vs. structured interview; see Figure 1). The descriptive data showed a marked gender difference between perpetrators and victims, with the sample consisting of mostly male (82.9%) perpetrators and mostly female (84.9%) victims. Consequently, we expected the effect of gender to be very large, with the risk of obscuring other relevant effects. We wished to control for gender, but also make sure that other relevant variables were detected. Therefore, we decided to run the analyses in Stage 3 in two separate sets of analyses: one including and one excluding gender. Due to our limited sample size, it was not possible to include all variables of interest in a multivariate model. Thus, we employed an analysis strategy to reduce the number of independent variables. First, we performed univariate logistic regression analyses with the potential sociodemographic control variables and participation mode (Step 3a in Figure 1). All variables with p < .20 were included in multivariate models along with MR-IPV experience. There were two models created—one including and one excluding gender (Step 3b in Figure 1).

In the fourth and final stage of the analysis process, a final multivariate logistic regression model was created (see Figure 1). MR-IPV experience was included as an independent variable, along with the variables with p < .05 from the final models in Stages 2 and 3. In Stage 3, the two multivariate models had slightly different results. As the model including gender had overlapping but a greater number of statistically significant variables than the model excluding gender, the statistically significant variables in this model were included in the final model. The threshold of p < .05 was chosen rather than p < .20 in order to limit the number of variables in the final model. The final multivariate model included the variables MR-IPV experience, the CTS2 scale Psychological Aggression, relationship duration, gender, and age.

Results

The first research question was, ‘What knowledge, experience, and perceptions do perpetrators have concerning MR-IPV?’ Table 2 presents the descriptive results regarding this research question. Around one in five of the perpetrators reported knowing the MR-IPV law (19.5%). However, the most common response was knowing the law to some extent (43.9%). The remaining 36.6% reported not knowing the law. Furthermore, 58.5% reported having been informed about MR-IPV by someone from the help services. A small number (14.6%) reported having been informed to some extent, while 26.8% reported not receiving information about MR-IPV. Only 17.5% reported having experienced MR-IPV. Regarding MR-IPV perception, the mean score was 2.3 (SD = 0.15), indicating that perpetrators as a group do not lean particularly strongly toward either a positive or a negative perception of MR-IPV.

Table 2.

IPV perpetrators’ knowledge, experience, and perception of MR-IPV.

Item	n	No	To some extent	Yes
Perceived knowledge	41	36.6% (15)	43.9% (18)	19.5% (8)
Informed about MR-IPV	41	26.8% (11)	14.6% (6)	58.5% (24)
Experienced MR-IPV, total score (dichotomous)	40	82.5% (33)	–	17.5% (7)
Experienced MR-IPV with consent	41	90.2% (37)	0% (0)	9.8% (4)
Experienced MR-IPV without consent	40	90.0% (36)	7.5% (3)	2.5% (1)
		M	95% CI	SD
MR-IPV perception	39	2.3	2.0–2.6	1.0

		Completely disagree	Partly disagree	Partly agree	Completely agree
Everyone has the right to a physician/therapist/healthcare provider/contact person with absolute client confidentiality	40	27.5% (11)	12.5% (5)	27.5% (11)	32.5% (13)
I would lose trust in the doctor/care provider/help service if they disclosed information	39	38.5% (15)	20.5% (8)	12.8% (5)	28.2% (11)
I would think they were unprofessional if they disclosed confidential information	40	52.5% (21)	12.5% (5)	10.0% (4)	25.0% (10)
The patient's wish matters most; if I don’t want to report to the police, the professionals shouldn’t do it	40	37.5% (15)	22.5% (9)	22.5% (9)	17.5% (7)

Note: Percentages refer to the percentage of the n for the item. Count for each response category presented in parentheses. IPV: intimate partner violence; MR-IPV: mandatory reporting of intimate partner violence; CI: confidence interval.

The second and third research questions were ‘To what extent do perpetrators and victims differ in knowledge, experience, and perceptions of MR-IPV? If there are differences, do they remain statistically significant when controlling for (a) conceptually informed variables (IPV severity, IPV persistence, and the presence of children) and (b) sociodemographic characteristics and participation mode?’ Of the MR-IPV-related variables, perpetrators and victims were only different regarding MR-IPV experience (see Table 3). More victims had experienced MR-IPV (45.3%) compared to perpetrators (17.5%; p = .003 for chi-square test; OR = 3.912, CI = 1.560–9.811, p = .004 for univariate logistic regression).

Table 3.

Differences between IPV perpetrators and victims on MR-IPV measures (univariate analyses).

Item	IPV perpetrators				IPV victims				Sig.	OR (95% CI)
	n	No	To some extent	Yes	n	No	To some extent	Yes
Perceived knowledge	41	36.6% (15)	43.9% (18)	19.5% (8)	86	31.4% (27)	50.0% (43)	18.6% (16)	.722
Informed about MR-IPV	41	26.8% (11)	14.6% (6)	58.5% (24)	84	40.5% (34)	20.2% (17)	39.3% (33)	.053
Experienced MR-IPV	40	82.5% (33)	–	17.5% (7)	86	54.7% (47)	–	45.3% (39)	. 003	3.912 (1.560–9.811)^a
	n	M (95% CI)		SD	n	M (95% CI)		SD	Sig.
MR-IPV perception	39	2.3 (2.0–2.6)		1.0	82	2.2 (2.0–2.4)		0.8	.610

Note: Chi-square test and univariate logistic regression used for MR-IPV experience. Mann–Whitney U tests used for the remaining comparisons. Bold text indicates statistically significant findings (p < .05). Percentages refer to the percentage of the n for the item. Count for each response category presented in parentheses. IPV: intimate partner violence; MR-IPV: mandatory reporting of intimate partner violence; OR: odds ratio; CI: confidence interval.

Univariate logistic regression predicting 1 = IPV victim, 0 = IPV perpetrator, p = .004.

Next, we examined whether the difference in MR-IPV experience could be explained by conceptually informed control variables, namely IPV severity, IPV persistence, and the presence of children. Table 4 shows the results of the analyses in Stage 2. In the multivariate analyses with MR-IPV and each of the conceptually informed variables (Step 2a), victims had higher odds of reporting more IPV persistence and psychological aggression. The difference in MR-IPV experience remained statistically significant in all analyses. In the multivariate model incorporating several of the conceptually informed variables, victims had higher odds of reporting more psychological aggression and a longer duration of the IPV relationship. The difference in MR-IPV remained statistically significant.

Table 4.

Multivariate logistic regression models in Stage 2 of the analyses: testing differences between IPV perpetrators and victims on MR-IPV and conceptually informed control variables.

Variable		Step 2a: Multivariate models with MR-IPV conceptually informed control variables		Step 2b: Multivariate model with variables with p < .20 in Step 2a, n = 116
Variable	n	OR (95% CI)	p-value	OR (95% CI)	p-value
IPV persistence	120
MR-IPV experience		3.797 (1.243–11.594)	. 019	4.856 (1.289–18.299)	. 020
IPV persistence (no)			< . 001		.268
To some extent		3.280 (.738–14.568)	.118	2.594 (.374–17.996)	.335
Yes		9.622 (3.334–27.774)	<.001	2.827 (.782–10.210)	.113
Relationship duration		1.044 (.985–1.106)	.146	1.087 (1.007–1.173)	. 032
IPV Severity	120/120^a
MR-IPV experience^a		3.783 (1.204–11.890)/3.809 (1.198–12.110)	.023/.023
CTS2 psychological aggression^a		1.024 (1.008–1.040)/1.025 (1.010–1.039)	.002/<.001	1.026 (1.011–1.041)	< . 001
CTS2 physical assault		1.004 (.981–1.027)	.738
CTS2 sexual coercion^a		1.658 (.962–2.856)/1.648 (.953–2.848)	.069/.074	1.565 (.888–2.757)	.122
CTS2 injury		1.011 (.953–1.072)	.728
The presence of children	126
MR-IPV experience		3.642 (1.439–9.217)	. 006
Children (yes)		2.834 (.807–9.947)	.104	1.233 (.202–7.523)	.821

Note: Multivariate models predicting IPV victim = 1, IPV perpetrator = 0. Bold text indicates statistically significant findings (p < .05). In Step 2a, four multivariate models were created, all including MR-IPV experience and one type of conceptually informed control variable (see Note a). In Step 2b, variables with p < .20 from analyses in Step 2a were included in one multivariate logistic regression model. Multivariate model in Step 2b: Cox and Snell R² = .449. Nagelkerke R² = .629. Hosmer–Lemeshow test non-significant. IPV: intimate partner violence; MR-IPV: mandatory reporting of intimate partner violence; OR: odds ratio; CI: confidence interval; CTS2: Revised Conflict Tactics Scale.

Due to high multicollinearity between Physical Assault and Injury, two models were created. Results from the first model, including the Physical Assault scale, are presented before the solidus mark. Results from the second model, including the Injury scale, are presented after the solidus mark.

Table 5 shows the results of the analyses in Stage 3, examining sociodemographic characteristics, participation mode, and MR-IPV experience. In univariate analyses (Step 3a), victims had higher odds of being female, reporting higher age, and reporting that their health was poor or very poor, compared to good or very good. In the multivariate analyses, the difference in MR-IPV experience remained statistically significant. In the multivariate model including gender, victims had higher odds of being female and reporting higher age. In the multivariate model excluding gender, only MR-IPV experience was statistically significant.

Table 5.

Univariate and multivariate logistic regression models in Stage 3 of the analyses: testing differences between IPV perpetrators and victims on MR-IPV, sociodemographic characteristics, and response mode.

Variable	Step 3a: Univariate logistic regression			Step 3b: Multivariate models with variables with p < .20 in Step 3a
	Step 3a: Univariate logistic regression			Model excluding gender, n = 126		Model including gender, n = 126
	n	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value
MR-IPV experience				3.866 (1.492–10.016)	.005	4.657 (1.237–17.526)	.023
Gender (male)	127	.037 (.013–.100)	<.001			.019 (.005−.071)	<.001
Age	127	1.031 (.992–1.072)	.121	1.029 (.986–1.074)	.186	1.075 (1.016–1.136)	.011
Health (good/very good)	127		.097		.415		.630
Neither good nor poor		1.737 (.735–4.105)	.339	1.623 (.637–4.135)	.310	1.844 (.530–6.412)	.336
Poor/very poor		2.995 (1.059–8.470)	.039	1.876 (.612–5.747)	.271	1.286 (.237–6.970)	.770
Citizenship (native citizen)	124		.988
Naturalized citizen		1.026 (.293–3.587)	.968
Foreign citizen		.912 (.255–3.257)	.887
Education (years)	123	.975 (.866–1.098)	.678
Employment (employed/student)	124		.448
Employed and benefits		2.755 (.562–13.499)	.211
Benefits/unemployed		1.194 (.521–2737)	.676
Interview (vs. self-report)	127	1.872 (.829–4.224)	.131	1.644 (.390–6.928)	.498	1.018 (.132–7.869)	.986
Physical participation (vs. digital)	127	1.809 (.762–4.293)	.179	1.169 (.247–5.528)	.844	4.514 (.465–43.817)	.194

Note: All logistic regression models predicting IPV victim = 1, IPV perpetrator = 0. Bold text indicates statistically significant findings (p < .05). Column with Step 3a shows the results of univariate logistic regression models. In Step 3b, variables with p < .20 from analyses in Step 3a were included in multivariate logistic regression models. One model included gender; the other did not. Multivariate model in Step 3b excluding gender: Cox and Snell R² = .122. Nagelkerke R² = .171. Multivariate model in Step 3b, including gender: Cox and Snell R² = .446. Nagelkerke R² = .625. Hosmer–Lemeshow test non-significant in both models. IPV: intimate partner violence; MR-IPV: mandatory reporting of intimate partner violence; OR: odds ratio; CI: confidence interval.

In the fourth and final stage of the analyses, the difference in MR-IPV experience was still statistically significant (see Table 6). Victims had higher odds of reporting MR-IPV experience, more psychological aggression in the relationship, and female gender.

Table 6.

Final multivariate regression model testing differences between IPV perpetrators and victims on MR-IPV (Stage 4), n = 119.

Variable	OR (95% CI)	p-value
MR-IPV experience	4.677 (1.020–21.443)	. 047
IPV severity—CTS2 psychological aggression	1.035 (1.017–1.053)	< . 001
IPV persistence—relationship duration	1.030 (.942–1.126)	.517
Gender	.029 (.007–.123)	< . 001
Age	1.072 (1.000–1.150)	.051

Note: Logistic regression models predicting IPV victim = 1, IPV perpetrator = 0. Bold text indicates statistically significant findings (p < .05). Cox and Snell R² = .512. Nagelkerke R² = .717. Hosmer–Lemeshow test non-significant in both models. IPV: intimate partner violence; MR-IPV: mandatory reporting of intimate partner violence; OR: odds ratio; CI: confidence interval; CTS2: Revised Conflict Tactics Scale.

Discussion

To the best of our knowledge, the present study is the first to examine IPV perpetrators’ perspectives on MR-IPV, using a quantitative approach. As the perpetrator sample size is modest (n = 41), our findings may have limited generalizability but nevertheless represent a first glimpse into a largely unexplored topic. Our main findings were that the majority of perpetrators who had sought help had received information about MR-IPV. However, few reported having experienced MR-IPV (17.5%), and few reported that they knew the law (19.5%). The most common was knowing the law to some extent (43.9%), and around a third reported no knowledge (36.6%). The participants varied in their perceptions of MR-IPV. More victims reported MR-IPV experience than perpetrators. This difference remained statistically significant even when controlling for conceptually informed control variables (IPV severity, IPV persistence, and the presence of children), sociodemographic characteristics, and participation mode.

IPV perpetrators’ perspectives on and experiences with MR-IPV

In the present study, we found that the majority (58.5%) of perpetrators who had sought help were informed about MR-IPV. However, considerably fewer participants reported that they knew the law—only 19.5%. Approximately one in three reported not knowing the law. The combination of these findings suggests that although the majority of perpetrators receive information about the MR-IPV law, professionals might not provide sufficient information to ensure that the perpetrators are confident in their knowledge about MR-IPV. The findings are similar to those by Linge et al. (2025), who found that IPV perpetrators in voluntary treatment in Norway had lacking awareness and a vague understanding of MR-IPV. High prevalence of a vague understanding might be reflected in the high prevalence of to some extent answers in the present study. Clinicians working with perpetrators should bear in mind that simply giving information about MR-IPV might not be enough; professionals must ensure that the perpetrators understand what MR-IPV entails and what consequences it might have for them.

The perpetrators in the present study varied in their perceptions of MR-IPV. For all four items, there was an approximate 40/60 split between the partly/completely agree and partly/completely disagree response options. Sixty percent partly or completely agreed that everyone has the right to a physician/therapist/healthcare provider/contact person with absolute client confidentiality. For the other three items, approximately 60% disagreed that they would lose trust in or think professionals unprofessional, or prioritize their own choice over a mandated duty to report to the police. Thus, we found no clear agreement among perpetrators about issues concerning the perception of MR-IPV.

In contrast, Kristiansen et al. (2024) found that the two perpetrators who had experienced MR-IPV described a breach of trust and were thus unified in their perception of MR-IPV. However, the professionals in Kristiansen et al.'s study reported that even when there was a breach of trust, they could work with the perpetrator or victim to repair the trust. This highlights that MR-IPV is a complex issue, and the way MR-IPV is complied with might be just as important as whether it occurs. The variation in the responses in the present study might reflect that MR-IPV can take various forms, and that the collaboration between the perpetrator and professionals both before, during, and after may play important roles in shaping the perception.

Another finding was that few of the perpetrators (17.5%) reported having experienced MR-IPV. The present study was not population-based, and we cannot draw conclusions about the prevalence of MR-IPV experiences from the data of 41 perpetrators. Future studies should examine whether similar rates are found in populations of perpetrators in other countries, as well as perpetrators attending mandatory IPV treatment. One might have expected a higher number of perpetrators with MR-IPV experience. However, there are several plausible explanations for the results. First, perpetrators who voluntarily seek help for IPV might, at a group level, perpetrate less severe IPV compared with other perpetrators, and there might thus be fewer cases where MR-IPV applies. Caman et al. (2024) interviewed both self-referred and mandatory attendees in a perpetrator program and found differences in the reasons why the different groups used aggression. Though the authors noted that there were more similarities than differences, self-referred perpetrators tended to describe reactive violence after feeling pressured and vulnerable, while court-mandated perpetrators used aggression proactively to exert power and dominance over their partners.

Another possible explanation for the low number of perpetrators with MR-IPV experience is that professionals might have used the discretionary space in the Norwegian MR-IPV law and averted IPV by other means than reporting to police. The perpetrators in treatment might not have been informed that professionals acted on a duty or recognized the actions taken by professionals as something that fulfills MR-IPV requirements, but rather viewed it as an integrated part of their treatment. The previously mentioned study by Kristiansen et al. (2024) could support such an interpretation. They found that professionals often used the discretionary space allowed in the Norwegian MR-IPV law to avert ‘by other means’. This included seeking collaboration with the perpetrator to develop safety plans and being honest and direct when they feared that there was elevated risk for new IPV.

A third possible explanation of the low occurrence of MR-IPV experience is the possibility that the professionals did not act because of limited knowledge about MR-IPV or because they feared the perpetrator would experience it as a breach of trust and terminate the treatment. This concern has been raised by researchers who found that trust and confidentiality were considered crucial for perpetrator treatment (Campbell et al., 2010), as well as concerning victims who seek help (Jordan and Pritchard, 2021; Lippy et al., 2020; Rodriguez et al., 1999; Rodríguez et al., 2002). Some of the professionals in the study by Kristiansen et al. (2024) also voiced this fear. However, Kristiansen and colleagues found that the professionals rarely experienced that victims or perpetrators cut contact after MR-IPV, but rather that the most common was to remain in contact with the help services and work on rebuilding any potential breach in trust.

Differences between IPV perpetrators and victims in MR-IPV experience

Our main finding concerning the second and third research questions was that more victims than perpetrators had experienced MR-IPV, and the difference remained statistically significant when controlling for IPV characteristics, the presence of children, sociodemographic characteristics, and participation mode. The analyses examining each of the conceptually informed control variables indicated that even though victims scored higher on IPV persistence and severity, these differences could not explain the difference in MR-IPV experience. In the initial univariate analyses, victims had higher odds of having experienced MR-IPV (OR = 3.912). In the variable-type-specific models in Stage 2a of the analyses, the OR for the difference in MR-IPV experience was lower than in the univariate analysis (see Table 4). However, the confidence intervals were also larger, indicating more uncertain estimates. Thus, our results indicate that a difference in the conceptually informed variables might explain a small portion of the variance in MR-IPV, but far from all of it. IPV severity and persistence are important factors deciding whether MR-IPV applies and are important risk factors for future IPV. It was therefore surprising that even though there was a difference between victims and perpetrators on IPV characteristics, it did not account for the difference in MR-IPV experience. This suggests that MR-IPV practices might differ for victims and perpetrators because of other factors than those outlined as important in the MR-IPV law, which in turn implies that the MR-IPV law might not be practiced as intended.

Next, we examined whether differences in sociodemographic characteristics and participation mode might explain the difference in MR-IPV experience. Victims continued to report more MR-IPV experience, both when controlling for sociodemographic characteristics and participation mode in a separate model and in the final multivariate model. Only gender remained statistically significant in the final multivariate model, though age was trend significant (p = .051). The effect of gender was very large, with perpetrators having 33.981 times higher odds of being male compared to victims in the final multivariate model. This high effect was not surprising given the highly skewed distribution of gender in our sample, with 85% female victims and 83% male perpetrators. The differences are similar to those in the population of individuals seeking help for IPV. For instance, Norwegian shelters reported that in 2023, 88.9% of the individuals staying at a domestic violence shelter and 91.1% of those receiving help from day-based services were women (Bufdir, 2025a; Bufdir, 2025b). Conversely, the ATV reported that 80.7% of perpetrators in treatment in 2023 were men (Alternative to Violence, 2024). Even though there was a large difference in gender, our results indicate that the difference in MR-IPV experience could not be explained by a difference in gender. However, we note that though our sample mirrors the population of help-seeking individuals, it is not optimal for examining the effect of gender due to the large differences in gender distribution between the groups. The role of gender should be investigated further in samples with smaller differences in gender distribution.

There might also be other differences between perpetrators and victims that are important concerning MR-IPV that were not measured in the current study. MR-IPV can have very different consequences for perpetrators and victims. For the victims, MR-IPV is a measure intended to safeguard and help. For perpetrators, on the other hand, MR-IPV might lead to police involvement, investigation of the crime, and safety measures that are experienced as negative for the perpetrator. Professionals working with victims and perpetrators might be affected by this difference in the consequences for their clients, and therefore act differently. There might also be differences in the professionals’ knowledge and attitudes toward MR-IPV laws. Nordby et al. (2024) examined differences in attitudes among professionals in Norway and found that professionals were generally supportive of MR-IPV. Individuals with more experience with IPV cases (both perpetrators and victims) held more positive attitudes. Both shelter workers, anger management therapists, and ATV employees were less skeptical compared with emergency department workers, but there was no comparison between professionals working primarily with victims or perpetrators.

Limitations

There are several limitations in the present study that should be noted. First, the sample consisted of 41 perpetrators and 86 victims. The 41 perpetrators constrained the power in the analyses, and there is a higher likelihood of a type II error due to the limited sample size in the perpetrator group. This is particularly important concerning the multivariate models with several predictors, and the trend significant findings, such as the difference in having been informed about MR-IPV (p = .053). Future research should further examine whether there is a difference in the information given to perpetrators and victims. The small perpetrator sample size also limits the generalizability of the results and the certainty of the estimates in analyses, and future research should examine the topic of perpetrators’ experiences and perspectives on MR-IPV further. Moreover, there was a notable gender difference between the perpetrators and victims in our sample, as well as limited variance concerning having children. The role of gender and the presence of children should be investigated further.

Secondly, only perpetrators who had voluntarily sought help were included in the study, and the results are not necessarily generalizable to perpetrators in mandatory treatment or a broader perpetrator population. Because the recruitment process involved many different help services and individual professionals, it was not feasible to get a precise estimation of the response rate. The strength in this approach is that the participants better represent the group of perpetrators who seek help and are not limited to help-seeking from a specific help service or a few professionals. However, it is difficult to know if those who agreed to participate differed systematically in some way from those who declined, and the representativeness is difficult to gauge without an estimate of response rate. The Norwegian legislation, help services, and culture also differ from other countries, which might impact the results and limit the generalizability to other contexts.

Thirdly, when comparing perpetrators and victims on IPV experiences, there are reporting issues to consider. Previous studies have found discrepancies in IPV reporting among couples, with women reporting higher IPV levels than men (see e.g., Armstrong et al., 2002). This discrepancy has also been found among couples where the man is in voluntary IPV treatment (including the ATV; Strandmoen et al., 2015), particularly concerning psychological/emotional violence (Strandmoen et al., 2015; Vall et al., 2021). A difference in reporting of IPV between the predominantly male perpetrators and predominantly female victims in the current study might have resulted in inflated variance in the variables measuring IPV severity and persistence. However, as our results indicate that a difference in IPV characteristics could not explain the difference in MR-IPV experience, the conclusion would likely be the same had there been a smaller difference in IPV characteristics. Additionally, the combined CTS2 measures in the current study used the highest score among the perpetration and victimization scores for the participants. This might alleviate the problem of under-reporting among perpetrators in cases where perpetrators reported higher victimization rates than perpetration rates.

Fourthly, MR-IPV perception was only measured with four items, and none of them addressed directly whether MR-IPV was a barrier to help-seeking or IPV disclosure. Additionally, no items asked specifically about MR-IPV actions beyond police reporting, although the Norwegian MR-IPV law allows averting IPV ‘by other means’. The items were identical for victims and perpetrators, with the intent to examine perception of mandatory reporting as a general concept. However, it is possible that the participants’ responses do not reflect their perceptions of MR-IPV in general, but rather MR-IPV in the context they were recruited from. In other words, perpetrators might have responded based on their perceptions of MR-IPV in the context of perpetrator interventions, whereas the victims might have pictured a scenario with MR-IPV when a victim seeks help. Future studies should examine MR-IPV perception in more detail, using both quantitative and qualitative approaches. It would be beneficial to distinguish more clearly between MR-IPV in cases where a victim and a perpetrator seek help, to elucidate whether there are different opinions among perpetrators in these two scenarios.

Finally, two of the items measuring MR-IPV perception asked whether the participants would lose trust in or think professionals were unprofessional if they forwarded information. The items did not specify whether the information was forwarded in order to comply with MR-IPV, though the items were presented alongside other items phrased to indicate that the topic was MR-IPV. If the participants asked, it was also clarified that the items pertained to MR-IPV. However, there is a possibility that the participants could have interpreted the items as inquiring about professionals forwarding information in general, including cases without a legal mandate to do so.

Future research directions

The present study was explorative, as there is limited prior research on perpetrators’ experiences and perspectives on MR-IPV. As such, future research is needed to replicate all the results in a new sample of voluntarily help-seeking perpetrators. Additionally, research should examine the perspectives of perpetrators in mandatory treatment, as well as perpetrators who have not sought help. The latter is of particular importance to elucidate whether MR-IPV is a barrier to help-seeking among perpetrators. Whether gender and the presence of children are associated with MR-IPV experience should be examined further, as well as whether there is a difference between perpetrators and victims in the information received about MR-IPV. Our study indicated that there is a difference between victims and perpetrators in MR-IPV experience, but future research is needed to replicate the results and investigate why such a difference might exist. Investigating professionals’ perspectives on their practices might be a fruitful avenue in addition to more research comparing the experiences of victims and perpetrators. As the current state of knowledge is limited, qualitative research might be particularly important to generate hypotheses and guide quantitative research.

Clinical and policy implications

The present study found that although most perpetrators report receiving information about MR-IPV, few felt confident in their knowledge about the law. This indicates that the information given might not be enough to fully inform perpetrators about MR-IPV and what it entails. Professionals should ensure that they inform perpetrators about MR-IPV in a manner that makes it possible to understand what MR-IPV entails and how it might apply to their specific situation. Professionals should also take extra care to ensure that the information is understood by perpetrators. Additionally, the concern has been raised that MR-IPV might be a barrier to IPV disclosure or might affect the relationship of trust between perpetrator and professionals. Our findings indicate that this is a concern among some perpetrators, but that the majority did not believe they would lose trust in professionals. The findings thus lend preliminary support for MR-IPV, but we note that the moderate sample size makes estimates uncertain, and future studies are needed to examine this more thoroughly.

Conclusion

The findings in the present study shed light on the mostly unexplored topic of voluntarily help-seeking IPV perpetrators’ perspectives on MR-IPV. Our findings suggest that although the majority of perpetrators reported receiving information about MR-IPV, professionals must ensure that information about MR-IPV is adequately understood by perpetrators. Few perpetrators had experienced MR-IPV, and they varied in their perceptions of the duty. More victims had experienced MR-IPV compared to perpetrators, and the difference remained statistically significant even when controlling for IPV characteristics, sociodemographic characteristics, and participation mode. The study had a small sample size, and future research is needed to examine the topic in further detail.

Supplemental Material

sj-docx-1-euc-10.1177_14773708251365065 - Supplemental material for Perpetrators of intimate partner violence and mandatory reporting laws: The perspectives and experiences of perpetrators who seek help voluntarily compared with those of victims of intimate partner violence

Supplemental material, sj-docx-1-euc-10.1177_14773708251365065 for Perpetrators of intimate partner violence and mandatory reporting laws: The perspectives and experiences of perpetrators who seek help voluntarily compared with those of victims of intimate partner violence by Astrid Gravdal Vølstad, Kevin S Douglas, Thea Beate Brevik, Christine Nordby and Solveig Karin Bø Vatnar in European Journal of Criminology

Footnotes

Acknowledgements

We wish to thank Petter Laake and Stål Kapstø Bjørkly for invaluable help and advice. We also wish to thank the participants and the help services who helped us recruit.

ORCID iDs

Astrid Gravdal Vølstad

Kevin S Douglas

Thea Beate Brevik

Christine Nordby

Solveig Karin Bø Vatnar

Ethical approval

Regional Committees for Medical Research Ethics (REK) considered the study to be health service research and thereby not within their mandate (Ref. No. 257644).

Consent to participate

Written informed consent was obtained from all participants.

Author contributions

All authors contributed to the design of the study. Participant recruitment and data collection: A.G.V., T.B.B., and C.N. Statistical analyses: A.G.V., K.S.D., and S.K.B.V. All authors contributed to the writing process for the present article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Research Council of Norway, project number 313902.

Declaration of conflicting interests

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

Data availability

Due to their sensitive nature, the data for this research is not publicly available.

Supplemental material

Supplemental material for this article is available online.

References

Aguilar Ruiz

González Calderón

González García

(2023) Severe versus less severe intimate partner violence: Aggressors and victims. European Journal of Criminology 20(4): 1211–1230.

Alternative to Violence (2024) Alternativ til Vold Årsrapport 2023 [Alternative to Violence Annual Report 2023]. Alternativ til vold. https://atv-stiftelsen.no/wp-content/uploads/2024/03/atv_2023_O_oppslag_hoyopplost.pdf.

Armstrong

Wernke

Medina

, et al. (2002) Do partners agree about the occurrence of intimate partner violence? A review of the current literature. Trauma, Violence, & Abuse 3(3): 181–193.

Bendixen

(2005) Konflikter og konflikthåndtering blant studenter i parforhold - Resultatrapport 1 [Conflicts and conflict management among students in intimate partner relationships. Result report 1]. Trondheim: NTNU. ISBN 8280940049.

Bordin

(1979) The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research & Practice 16(3): 252–260.

Bufdir (2025a) Om beboerne [About the residents]. Available at: https://www.bufdir.no/statistikk-og-analyse/krisesentrene/om-dagbrukere (accessed 20 January 2025).

Bufdir (2025b) Om dagbrukerne [About the users of day-based services]. Available at: https://www.bufdir.no/statistikk-og-analyse/krisesentrene/om-dagbrukere (accessed 20 January 2025).

Burghart

Backhaus

(2024) The long-term consequences of family violence victimization: An umbrella review of longitudinal meta-analyses on child maltreatment and intimate partner violence. Journal of Family Violence: 1–11.

Caman

Alm

Simonsson

(2024) ‘Everyone is against me’: A qualitative study of intimate partner violence perpetration with both convicted and voluntary help-seeking men. Journal of Forensic Psychology Research and Practice: 1–27. DOI: https://doi.org/10.1080/24732850.2024.2400287.

10.

Campbell

Neil

Jaffe

, et al. (2010) Engaging abusive men in seeking community intervention: A critical research & practice priority. Journal of Family Violence 25(4): 413–422.

11.

Chapman

Gillespie

(2019) The revised conflict tactics scales (CTS2): A review of the properties, reliability, and validity of the CTS2 as a measure of partner abuse in community and clinical samples. Aggression and Violent Behavior 44: 27–35.

12.

Costa

Barros

(2016) Instruments to assess intimate partner violence: A scoping review of the literature. Violence and Victims 31(4): 591–621.

13.

Council of Europe (2011) Explanatory Report to the Council of Europe Convention on preventing and combating violence against women and domestic violence. CETS No. 210, 11/05/2011. Council of Europe.

14.

Council of Europe (2014) The Council of Europe Convention on preventing and combating violence against women and domestic violence. CETS No. 210, 11/05/2011. Council of Europe.

15.

Cuenca

Graña

Redondo

(2015) Differences in prevalence of partner aggression as the revised Conflict Tactics Scale: Individual and dyadic report. Behavioral Psychology/Psicología Conductual 23(1): 127–140.

16.

Davis

Fernandez

Jonson-Reid

, et al. (2021) Pathways to seeking help from a partner abuse intervention program: A qualitative study of voluntary and non-court-mandated Latino men’s experiences. Journal of Interpersonal Violence 36(21–22): 10454–10478.

17.

Hester

Lilley

S-J

(2014) Domestic and sexual violence perpetrator programmes: Article 16 of the Istanbul Convention: A collection of papers on the Council of Europe Convention on preventing and combating violence against women and domestic violence. Council of Europe. Available at: https://rm.coe.int/168046e1f2.

18.

Hilton

Radatz

(2018) The criminogenic and noncriminogenic treatment needs of intimate partner violence offenders. International Journal of Offender Therapy and Comparative Criminology 62(11): 3247–3259.

19.

Horvath

Del Re

Flückiger

, et al. (2011) Alliance in individual psychotherapy. Psychotherapy 48(1): 9.

20.

Hosmer

Lemeshow

(2012) Logistic regression. In: Veierød

Lydersen

Laake

(eds) Medical Statistics in Clinical and Epidemiologic Research. Oslo: Gyldendal Norsk Forlag, 90–126.

21.

Jordan

Pritchard

(2021) Mandatory reporting of domestic violence: What do abuse survivors think and what variables influence those opinions? Journal of Interpersonal Violence 36(7–8): NP4170–NP4190.

22.

Kristiansen

STT

Leer-Salvesen

Dahl

, et al. (2024) Experience with mandatory reporting of intimate partner violence: An explorative qualitative study of service providers and help-seeking IPV victims and IPV perpetrators. Journal of Family Violence. DOI: https://doi.org/10.1007/s10896-024-00778-w.

23.

Kropp

Hart

(2015) The Spousal Assault Risk Assessment Guide Version 3 (SARA-V3). Vancouver, BC: ProActive Resolutions Inc.

24.

Kuijpers

van der Knaap

Lodewijks

IAJ

(2011) Victims’ influence on intimate partner violence revictimization: A systematic review of prospective evidence. Trauma, Violence, & Abuse 12(4): 198–219.

25.

Lagdon

Armour

Stringer

(2014) Adult experience of mental health outcomes as a result of intimate partner violence victimisation: A systematic review. European Journal of Psychotraumatology 5(1): 24794.

26.

Linge

Leer-Salvesen

Dahl

, et al. (2025) Awareness of mandatory reporting of intimate-partner violence: A thematic analysis of the perspectives of help-seeking victims and perpetrators and service providers in Norway. Journal of Family Violence. DOI: https://doi.org/10.1007/s10896-025-00916-y.

27.

Lippy

Jumarali

Nnawulezi

, et al. (2020) The impact of mandatory reporting laws on survivors of intimate partner violence: Intersectionality, help-seeking and the need for change. Journal of Family Violence 35(3): 255–267.

28.

McGinn

McColgan

Taylor

(2020) Male IPV perpetrator’s perspectives on intervention and change: A systematic synthesis of qualitative studies. Trauma, Violence, & Abuse 21(1): 97–112.

29.

Nordby

Douglas

Vatnar

SKB

(2024) Service providers’ attitudes toward mandatory reporting of intimate partner violence: The impact of professional experience. Journal of Interpersonal Violence 0(0): 08862605241305145. DOI: https://doi.org/10.1177/08862605241305145

30.

O'Leary

Williams

(2006) Agreement about acts of aggression in marriage. Journal of Family Psychology 20(4): 656–662.

31.

Pedersen

Johnsen

PFF

von Hanno

, et al. (2023) Samfunnskostnader av vold i nære relasjoner [Social costs of domestic violence]. Menon-publikasjon 15/2023. Menon Economics.

32.

Rodríguez

Sheldon

Rao

(2002) Abused patient’s attitudes about mandatory reporting of intimate partner abuse injuries to police. Women & Health 35(2–3): 135–147.

33.

Rodriguez

McLoughlin

Bauer

, et al. (1999) Mandatory reporting of intimate partner violence to police: Views of physicians in California. American Journal of Public Health 89(4): 575–578.

34.

Stewart

(2000) Who are the respondents of domestic violence protection orders? Australian & New Zealand Journal of Criminology 33(1): 77–90.

35.

Strand

SJM

Fröberg

Storey

(2018) Protecting victims of intimate partner violence: Swedish prosecutors’ experiences of decision-making regarding restraining orders. Journal of Scandinavian Studies in Criminology and Crime Prevention 19(2): 170–186.

36.

Strandmoen

J-F

Askeland

Tjersland

O-A

, et al. (2015) Intimate partner violence in men voluntarily attending treatment: A study of couple agreement. Violence and Victims 31(1): 124–134.

37.

Straus

Hamby

Boney-McCoy

, et al. (1996) The revised conflict tactics scales (CTS2) development and preliminary psychometric data. Journal of Family Issues 17(3): 283–316.

38.

Stubbs

Szoeke

(2022) The effect of intimate partner violence on the physical health and health-related behaviors of women: A systematic review of the literature. Trauma, Violence, & Abuse 23(4): 1157–1172.

39.

Svalin

Mellgren

Torstensson Levander

, et al. (2017) Assessing and managing risk for intimate partner violence: Police employees’ use of the Police Screening Tool for Violent Crimes in Scania. Journal of Scandinavian Studies in Criminology and Crime Prevention 18(1): 84–92.

40.

Taft

Murphy

(2007) The working alliance in intervention for partner violence perpetrators: Recent research and theory. Journal of Family Violence 22: 11–18.

41.

The Penal Code (2005) The Penal Code [Lov om straff] (LOV-2005-05-20-28). Norway: Ministry of Justice and Public Security.

42.

Vall

Sala-Bubaré

Hester

, et al. (2021) Evaluating the impact of intimate partner violence: A comparison of men in treatment and their (ex-)partners accounts. International Journal of Environmental Research and Public Health 18(11): 5859.

43.

Vatnar

SKB

Dahl

Kristiansen

STT

, et al. (2024) What difference does it make if they have children? A mixed methods study of the impact of children on service providers’ perception of mandatory reporting of intimate partner violence. Journal of Family Violence. DOI: https://doi.org/10.1007/s10896-024-00773-1.

44.

Vatnar

SKB

Friestad

Bjørkly

(2017) Intimate partner homicide in Norway 1990–2012: Identifying risk factors through structured risk assessment, court documents, and interviews with bereaved. Psychology of Violence 7(3): 395–405.

45.

Vatnar

SKB

Leer-Salvesen

Bjorkly

(2021) Mandatory reporting of intimate partner violence: A mixed methods systematic review. Trauma, Violence, & Abuse 22(4): 635–655.

46.

Verbruggen

Blokland

Robinson

, et al. (2020) The relationship between criminal behaviour over the life-course and intimate partner violence perpetration in later life. European Journal of Criminology 17(6): 784–805.

47.

Walker

Bowen

Brown

, et al. (2017) Subjective accounts of the turning points that facilitate desistance from intimate partner violence. International Journal of Offender Therapy and Comparative Criminology 61(4): 371–396.

48.

Wampold

Imel

(2015) The Great Psychotherapy Debate: The Evidence for What Makes Psychotherapy Work. New York: Routledge.

49.

World Health Organization (2013) Responding to Intimate Partner Violence and Sexual Violence against Women: WHO Clinical and Policy Guidelines. Geneva: World Health Organization.

50.

World Health Organization (2017) Strengthening Health Systems to Respond to Women Subjected to Intimate Partner Violence or Sexual Violence: A Manual for Health Managers. Geneva: World Health Organization.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB