A structured interviewer-administered paper questionnaire was developed based on the IMAGES survey tool and Partners for Prevention (P4P), translated into the local Lao language and pre-tested locally. Based on the pretesting results, the questionnaires were revised and finalised. Most of the questions were close-ended and some key variables were included in multiple questions. Specific topics in the questionnaire were: 1) Socio-demographic characteristics such as age, highest education, ethnic group, religion, monthly earning income, employment status, alcohol, and marijuana/Kratom; 2) Childhood perpetration of violence before the age of 18 such as witnessing intimate partner violence (IPV), Sexual Violence Someone touched my buttocks or genitals, sex with someone because I was threatened frightened or forced, perpetration of sexual violence before aged 18, Emotional violence, perpetration of emotional before aged 18, physically punished at school by a teacher, physical punishment in my home, and Physical violence as a child before aged 18; 3) Childhood perpetration of violence in neighbourhoods such as Physical violence neighbourhoods were there bullying, teasing, and harassment in neighborhoods/schools in which you grew up, Girls were mostly treated with respect at my school, Bullied other kids using physical violence at school, friends used drugs at school, Overall perpetration of physical violence in neighbourhoods/school, Sexual Violence in neighborhoods/schools; 4) Types of violence reported against a wife/partner. For the four different types of violence, if the man committed at least one of the listed abusive acts in a particular category, he was considered to perpetrate that specific form of violence. The questionnaire was asked in the respondent’s lifetime and in the 12 months preceding the interview. For each of the lifetime abusive acts, a follow-up question was asked about the frequency with which the violence had happened: once, a few times, or many times. The responses to each item were combined to create a composite index for each type of violence; however, this study focused only on perpetrated physical and sexual violence by male participants.

CommCare and STRATA v.13 were applied for the data encoding and data analysis. The questionnaire was administered face-to-face by the interviewers. Standard procedures were followed to ensure anonymity and confidentiality. There were a total of 6 interviewers in field teams. Since the survey respondents were men, all the data collectors were male because it has been found that male interviewers are likely to get more accurate information on sensitive issues from male respondents. The interviews were conducted at locations convenient for the respondents, usually in a closed room in their homes. Each interview took 60 - 90 minutes. During the field study, core team members visited the study sites to ensure interview quality and respondents’ privacy.

CommCare and STRATA 13 were applied for the data encoding and data analysis. The quantitative component binary logistic regression model, followed by multivariable analysis, was performed to identify the associated factors of GBV. The adjusted odds ratio (AOR) with a 95% confidence interval was computed to assess the strength of the association, and variables with a p-value of <0.05 in multivariable analysis were considered statistically significant.

All variables were significant at the bivariate analysis, and the p-value of 0.05 was entered into the multiple logistic regression. Statistically associated variables were adjusted based on the backward elimination technique by multiple logistic regression; all tested variables with p-values greater than 0.05 were excluded from the analysis model one by one, multiple times, starting from the greatest p-value, in order to achieve the final adjusted model of regression (variables with p < 0.05). The goodness-of-fit (GOF) was also given for each final regression model.

A backward elimination procedure based on the likelihood ratio (or Wald statistic, if backward Wald was used) was applied to obtain the final model. Backward elimination was selected because it begins with a full model containing all candidate predictors, thereby allowing potential confounding variables to be considered simultaneously while progressively removing variables that did not contribute significantly to model fit. This approach resulted in a more parsimonious model while retaining variables independently associated with lifetime physical violence. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported, and statistical significance was defined as p < 0.05.

The study protocol was approved by the Institutional Ethical Committee (IEC) of the University of Health Sciences, No. 491, dated 6/4/2023. Ethical guidelines developed by the WHO on studying sensitive issues were maintained throughout the study and beyond. Participants in the study were fully informed about the nature of the study, the research objectives, and the confidentiality of the data, and gave written consent (a thumbprint for those who could not sign their names) for their participation in the study. Standard procedures were followed to ensure the anonymity and confidentiality of the participants.

The candidate predictor variables included age, education level, occupation, monthly income, marital status, alcohol consumption, substance use, childhood exposure to violence, controlling behaviours by a partner, decision-making power within the household, relationship characteristics, and other covariates relevant to the study objectives.

Among the 680 men who were enrolled in the study, the mean age of participants was 40.05, with a minimum age of 19 and a maximum age of 55. More than two-thirds of participants (65.6%) are in the age group 30 - 45 years old. Slightly less than half of the participants (48%) had lower and upper secondary schools. The majority of them are the Lao Tai ethnic group (89.3%). The mean household size was 5.93, with a minimum size of 2 and a maximum size of 13. One-third of them (67.8%) lived with a partner/spouse. The majority of them (80.7%) earned income for their families. The median monthly income was 3,000,000 LAK, with a minimum of 30,000 LAK and a maximum of 40,000,000 LAK. About four-fifths of them (82.4%) were employed, and among these employed male participants, 64% were employed informally (Table 1). Regarding their health risk behaviour, 63.8% reported drinking alcohol 2 - 3 times a week, followed by having 5 or more drinks on one occasion monthly (42.5%). A few per cent of participants (9.7% and 7.5%) used marijuana and kratom last year, respectively (Table 1).

Table 1. Socio-demographic characteristics and health risk behaviuors of participants.

Table 2 presents data on respondents’ perpetration of violence until the age of 18. Rates of emotional violence were moderate (50% - 58.4%), while rates of physical violence and witnessing of intimate partner violence were 69.4%. Of the respondents, 78.3% of men reported having been slapped or spanked by their parents when they were children. About 44.6% of men, in particular, suffered more severe forms of physical violence. More than two-thirds of men (69.4%) reported witnessing their mother being beaten by her husband or partner. They also suffered from sexual violence, as someone touched their buttocks or genitals or made them touch them on the genitals when they did not want to (58.4%). Regarding emotional violence, they were insulted or humiliated by someone in their family in front of other people (79.7%). For sexual violence, 39.4% of them experienced having sex with someone because I was threatened, frightened or forced.

Overall, 25.4% of participants experienced moderate childhood physical violence, and 69.4% witnessed childhood physical violence; 43.4% and 78.8% experienced emotional and sexual violence, respectively.

Table 3 presents the Childhood perpetration of violence in neighbourhoods. Most male participants experienced physical violence, as 83.1% said they were bullied or teased and harassed; 29.9% were themselves teased and harassed; 78.2% of them teased and harassed others. Regarding sexual violence, 75.9% of them suggested friends, and they would touch girls or say sexual things to them to tease

Table 2. Experience of particular forms of violence before the age of 18.

them, and 44.7% said they and their school friends were a group and would arrange to have sex with girls after school. Overall, 52.9% of male participants experienced moderate physical violence, and 50.7% had experienced sexual violence.

Table 3. Childhood perpetration of Violence in neighbourhoods.

Overall, 67.1% of male participants had positive attitudes towards GBV. For example, 72.5% of participants thought it was really and sort of wrong for a man to hit his partner/wife if HE says he is sorry afterwards. Seven in ten male participants (70.1%) suggested they were really and sort of wrong for a man to hit his partner/wife if HE thinks SHE deserves it. 84.4% stated really and sort of wrong for a man to hit his partner/wife if HE is drunk (Table 4).

Table 4. Attitudes towards GBV.

The most common form of intimate partner violence (IPV) was physical violence, which almost half the sampled men reported having enacted at some time. The second most common form was physical violence – more than three in four of the male participants (76%) reported being physically violent. Two in five of the men stated that they had enacted sexual violence; economic violence was lowest among Lao men (only around 10%).

The responses to violence against women in the past year showed that more than 40% of men had committed some form of violence in the past year. The most common form of violence in the last year, as is the case with ever use, was physical violence, followed by sexual violence.

Figure 1Figure 1 outlines the relationships and Violence during the lifetime and last year (Physical violence). About 43.4% had slapped a partner or thrown something at her that could hurt her more than 1 time, followed by pushing or shoving a partner more than 1 time (38.2%); hitting a partner with a fist or with something else that could hurt her more than 1 time (26.2%). Overall, 69.4% of male participants reported physical violence in their lifetime, while 76% reported having physical violence during the last year. About one-fourth of participants (20.3%) had experienced sexual violence during the last year (see Figure 2Figure 2).

The primary outcome variable for the multivariable logistic regression analysis was lifetime physical violence. The relationships observed in bivariate analysis were reassessed by using multivariate analysis to identify important determinants, adjusting for the confounding effects of other factors. Variables with p less than 0.05

Figure 1. Frequency of each statement of each type of GBV.

Figure 2.Type of GBV perpetrated by men.

included sociodemographic characteristics (age, education, ethnicity, employment status, income), perpetration of childhood violence, perpetration of physical violence in the neighbourhood, knowledge of GBV, attitude towards GBV, frequency of drinking alcohol, and substance use. The results of the logistic regression where the dependent variable is “perpetration of physical violence (ever)” and the odds ratios corresponding to the associated factors are presented in Table 4 along with their significance levels.

Most of these indicators are significant predictors of men’s perpetration of physical violence against their intimate partners, some more strongly than others. Being informal employment status (AOR: 2.8; 95% CI: 1.2 - 6.4); perpetration of being witnesses and perpetration of GBV during childhood (AOR: 7.6; 95% CI: 3 - 19.4); perpetration of physical violence in the neighbourhood during childhood before aged 18 years (AOR: 7.3; 95% CI: 2.3 - 22.5); attitudes towards GBV (AOR: 1.9; 95% CI: 1.2 - 3.1) and drug used in the last 12 months (AOR: 6.5; 95% CI: 3.7 - 11.3) are significantly associated with Perpetration of Physical Violence by men (See Table 5).

Table 5. Factors associated with Male perpetration of Physical violence during the last year.

The study revealed that a few male participants (27.5%) agreed that physical abuse, such as hitting, is still acceptable among men in Lao PDR and not necessarily viewed as violent behaviour that needs to be changed. About two-thirds of participants (67.1%) of our study hold a positive attitude towards GBV, while the LSIS 2017 also showed less accepting attitudes toward GBV, with just 29.5% of women and 16.2% of men believing that violence against women who defy gender norms is acceptable. Another study showed that 22.9% agreed that a male should demonstrate his authority, while 29.4% agreed that a lady must have sex with her husband. Screening men for their attitudes towards GBV could be helpful. A systematic review of GBV reported that men who expressed attitudes that favoured violence toward women and male sexual entitlement were more likely to perpetrate GBV [11]. Interventions that fostered belief in gender equality among men reduced violent behaviour [12]. This could be due to different settings, sociocultural contexts, and time periods.

Employment status was positively associated with male perpetrators of physical violence. In other words, when the male individual is employed, it may be linked to a higher likelihood of the male perpetrating physical violence. A study in Vietnam and Nepal found that businessmen and those working in or having a shop are far more likely (2.47 times in Nepal and 1.67 times in Vietnam) to perpetrate violence than those in a profession. The likelihood is also high among manual labourers in Vietnam, who are more than twice as likely (2.1 times) as professionals to use violence against their partners [13]. In contrast, this finding was in contrast with previous studies, which found that better job opportunities, increased disposable income, enhanced access to knowledge, more equitable gender attitudes, and greater control over intimate partnerships mediate the association and reduce women’s IPV risk [1][2][5].

Another important finding was the association between perpetration of childhood violence and being a male perpetrator of physical violence. Childhood violence can have a lasting impact on individuals, affecting their emotional, psychological, and social development. Higher scores reflect rejection of GBV. In addition, the study also revealed there is an association between perpetration of physical violence in the neighbourhood and being perpetrated by men. Exposure to violence in the neighbourhood can have a significant impact on individuals’ social and behavioural development. It can shape their perceptions of conflict resolution, influence their attitudes toward violence, and potentially contribute to a higher risk of engaging in aggressive behaviour, particularly among males. There have been findings that men who experienced physical violence and abuse in the community and schools were more likely to commit abuse in their marriage [14][15]. This suggests that environmental factors such as neighbourhood and school perpetration play a significant role in shaping the behaviour of male individuals, potentially leading to a greater likelihood of perpetrating physical violence. Understanding this association is crucial for developing effective interventions and support systems to address and prevent violence in communities and schools, as well as to provide necessary assistance to individuals who have experienced violence.

The study found that attitude towards GBV was associated with male perpetrators of physical violence. Attitudes towards GBV are by a variety of factors, including cultural norms, socialization, and exposure to violence. These attitudes can influence individuals’ beliefs about power and gender roles, and can contribute to a higher risk of perpetrating physical violence, particularly among males.

Substance use was significantly related to perpetrated physical violence by male participants. Substance abuse, including alcohol and drug use, can lower inhibitions, impair judgment, and increase the likelihood of engaging in aggressive or violent behaviour. It also exacerbates existing conflicts and contributes to a higher risk of perpetrating physical violence, particularly in intimate relationships. This finding was in line with previous studies that found an association between substance use and IPV [16].

In Lao PDR, the study was carried out in only two provinces, one in the Central and one in the south, so the data are not representative for the country as a whole. The analysis, however, does indicate some significant patterns and trends with respect to gender-based violence that deserve further policy attention. This current study applied measures to maximise participation, minimise bias, and enhance data quality: Questionnaires were administered in Lao languages; gender-matched interviewers received training on asking sensitive questions and building rapport; GBV was measured with a validated research instrument; and respondents were asked about a wide range of violent acts, providing respondents with multiple reliable opportunities to recall and disclose GBV experience. The fieldwork coincided with the peak monsoon, so there were several barriers to physically accessing the target populations. This is a cross-sectional study design, so cause and effect could not be determined; however, the study findings provided valuable insight into male perpetrators and factors associated with physical violence by male perpetrators.