Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2021 Mar 15;16(3):e0248587. doi: 10.1371/journal.pone.0248587

Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study

Nicharuch Panjaphothiwat 1, Ratipark Tamornpark 1,2, Tawatchai Apidechkul 1,2,*, Prapamon Seeprasert 1, Onnalin Singkhorn 3, Panupong Upala 2, Phitnaree Thutsanti 4, Fartima Yeemard 1, Rachanee Sunsern 1
Editor: Siyan Yi5
PMCID: PMC7959343  PMID: 33720967

Abstract

Background

Domestic violence significantly affects physical and mental health, particularly among children, women, and the elderly. Living in certain family environments could lead to victimization by domestic violence, especially among families with a poor socioeconomic status, such as the Lahu hill tribe people in Thailand. This study aimed to estimate the prevalence of and determine the factors associated with domestic violence among Lahu children, women, and the elderly.

Methods

A cross-sectional study was conducted of participants who belonged to the Lahu hill tribe and lived in 20 selected villages in Chiang Rai Province, Thailand. A validated questionnaire was used to collect personal information and information regarding experiences related to domestic violence in the past year from children (aged 5–15 years), women (aged 16–59 years), and the elderly (aged 60 years and over). A binary logistic regression was used to detect associations between the variables.

Results

A total of 646 participants were recruited into the study, specifically, 98 children aged 5–15 years, 430 women aged 16–59 years, and 118 elderly people. Children who smoked (AOR = 8.70; 95%CI = 1.27–59.45) had greater odds of experiencing domestic violence than children who did not smoke. Women who had a role as a family member (AOR = 1.59; 95%CI = 1.02–2.50), used alcohol (AOR = 3.36; 95%CI = 2.27–4.99), lived in a family with financial problems (AOR = 4.01; 95%CI = 2.52–7.66), and lived with a family member who uses alcohol (AOR = 2.87; 95%CI = 2.20–5.63) had greater odds of suffering domestic violence than women who did not share these characteristics. The elderly who used alcohol (AOR = 3.25, 95%CI = 1.08–9.81), lived with a family member who uses alcohol (AOR = 3.31; 95%CI = 1.26–7.34), or lived in a family with financial problems in the past year (AOR = 2.16; 95%CI = 1.16–8.77) had greater odds of facing domestic violence than the elderly who did not have these characteristics.

Conclusion

Family financial problems and substance use are associated with domestic violence in Lahu families in Thailand. Health interventions to reduce the use of substances, including training programs to respond to domestic violence, should be promoted for Lahu children, women, and the elderly.

Introduction

Domestic violence is widely recognized as a global concern and leads to public health and other social problems [1]. Children, women, and the elderly are the most vulnerable to domestic violence or intimate partner violence (IPV) [2]. There are four main types of violence, namely, physical violence, sexual violence, stalking, and psychological aggression [2]. The World Health Organization (WHO) has reported that more than one billion children aged 2–17 years have experienced physical, sexual, or emotional violence or neglect in the past year, which impacts lifelong health and well-being [3]. Moreover, target no. 16.2 of the Sustainable Development Goals (SDG) of the United Nations (UN) clearly states that all forms of violence against children, including abuse, exploitation, trafficking and torture, should be minimized in the next few years [4]. The WHO also confirms that violence against women, especially IPV and sexual violence, is a global problem [5]. It has been indicated that 35.0% of women worldwide have experienced either physical and/or sexual IPV at least once in their lives, which can eventually negatively affect women’s physical, mental, sexual, and reproductive health [5]. One in six individuals among the elderly (16.6%) around the world has experienced some form of abuse from people in their community in the past year. Elder abuse leads to severe physical and psychological problems [6]. In 2019, the UN estimated that the total number of people aged 60 years and over will be more than 2 billion by 2050 [7]. They become vulnerable domestically, particularly the elderly who live in developing countries, including Thailand [1].

Thailand is a developing country and has been reported to have a high prevalence of domestic violence, particularly among children [8], women [9] and the elderly [10]. Most cases of domestic violence have been reported in populations that live in poor socioeconomic environments and have low levels of education [10–14]. The WHO has reported that children and women in Thailand living with people who use substances such as amphetamines and alcohol have a greater chance of being victimized by some form of domestic violence [15]. Alcohol use in Thai society is very common [16]. The integration between economic constraints and substance use has become a significant factor that contributes to domestic violence [17]. Adult men have been indicated as the contributors to domestic violence in Thai society who can make people in their families vulnerable to physical, mental, and sexual violence [18].

The hill tribes are a group of people who migrated from South China over the past two centuries and have settled in the northern region of Thailand [19]. There are six main groups, namely, the Akha, Lahu, Hmong, Yao, Karen, and Lisu, with approximately 3.5 million people of the hill tribes living in Thailand in 2019 [19]. The Lahu people are the second-largest hill tribe in Thailand [19, 20]. Lahu villages are settled along the border of Thailand and Myanmar [21]. The Lahu people have their own language, lifestyle and perception with respect to using substances such as alcohol [22, 23] and amphetamines [24], particularly the young adult and adult populations. A large proportion of the Lahu people in Thailand live under the national poverty line, which is less than 2,500 baht (US$84) per month per person [25]. Under the current situation of the Lahu people in Thailand, information relevant to the prevalence of domestic violence against children, women, and the elderly, including information on the factors contributing to violence, is not available. This information is needed for use in public health policy development and implementation.

With several conditions related to the Lahu people, this study aimed to estimate the prevalence of and determine the factors associated with domestic violence, which occurs in children aged ≤ 15 years, women, and the elderly aged ≥60 years among the Lahu people in Thailand.

Materials and methods

Study design and settings

A cross-sectional study was conducted in 20 villages randomly selected out of 216 Lahu villages in Chiang Rai Province, Thailand [20], using a random number generation method. In the study process, only 20 Lahu villages were randomly selected from the list.

Study populations

The study population comprised all Lahu people who live in the 20 selected villages. The targeted population was classified into the 3 groups: children aged 5–15 years, women aged 16–59 years, and older people aged 60 years and over. The entire targeted population lived in the selected villages at the date of data collection and met the inclusion criteria, while the people who could not provide essential information relevant to the study protocol were excluded from the study.

Study sample and sample size calculation

The study sample size was calculated based on the formula for a cross-sectional study, that is, n = [Z2α/2*P*Q]/e2, where Z is the value from the standard normal distribution that corresponds to the desired confidence level (Z = 1.96 for 95% CI), P is the expected true proportion (which is considered based on the previous study), and e is the desired precision, which is 0.05. Based on a study conducted in Thailand in 2018, the prevalence of domestic violence was 15.0% [9]. Therefore, the minimum required sample size for the study was 196 participants for each sub-population (children aged 5–15, women aged 16–59, and the elderly aged 60). When 10.0% was added to account for any possible errors in the study process, 646 individuals were required for the analyses.

The research tool and its development

A questionnaire was developed for the study. The information used for questionnaire development was acquired from different sources, such as a literature review, discussions with psychologists who worked in the field, and discussions with Lahu village leaders. The validity and reliability of the questionnaire were improved before use. The validity of the questionnaire was detected and improved by an item-objective congruence (IOC) method that asked three external experts in the field to provide scores and comments. The questions that scored lower than 0.5 were deleted from the set of questionnaires, while the questions that scored 0.5–0.7 were considered to require revision before inclusion in the questionnaire. The questions that scored higher than 0.7 were included in the questionnaire.

The questionnaire was piloted with 20 Lahu people who lived in two villages in Mae Chan District, Chiang Rai Province, between February 1 and 20, 2019. The purpose was to improve the validation and reliability before using the questionnaire in the field. Moreover, it was also important to confirm the forms of domestic violence in Lahu people.

Ultimately, the questionnaire comprised five parts. In part one, 16 questions were used to collect data on the general characteristics of the participants, such as their age, sex, education and marital status. In part two, 7 questions were used to collect information on substance use, such as smoking, alcohol use, marijuana use, amphetamine use, glue use, opium use, and family members who used alcohol. In part three, 7 questions were used to collect information on women’s experiences of harm related to sex, such as experiences with sexual harassment and being forced to have sex. In part four, 5 questions were used to collect information on domestic violence among children aged ≤15 years and the elderly aged ≥ 60 years, such as being left to live alone and being ignored regarding financial support and caregiving when they have had health problems. In part five, 12 questions were used to collect information on different kinds of domestic violence among children aged ≤ 15 years, women, and the elderly aged≥60 years, such as being forced to use substances, being insulted or addressed rudely and being addressed with anger to provoke a fearful reaction. The information from part 5 was used as the dependent variable in the study. The overall Cronbach’s alpha [26] on the questions related to domestic violence among children, women, and the elderly were 0.70, 0.71, and 0.71. Domestic violence is defined as any form of violence or abuse that occurs in any relationship within a family [27]. In the study, the participants were asked about their experience of domestic violence through different questions, such as “have you experienced getting kicked out of the house by a family member in the past year?”, “have you experienced being forced to drink alcohol, smoke, or use substances in the past year by a family member?” and “have you experienced being forced to ask for money or borrow items from others?”.

Steps of data collection

Access to all 20 selected villages was granted by district officers. Village headmen were contacted and provided with brief information regarding the study, and appointments were made to visit the village 5 days prior. At the time of data collection, the people who were willing to participate in the study were provided essential information and asked to provide an informed consent form before starting the interview. Participants aged less than 18 years and their parents were also asked to provide informed consent. For the children (aged 15 years and below), the information was obtained from their parents, except for the questions that asked about harms received. For these questions, the children were carefully questioned in a private room by a trained psychiatric nurse. People who were not able to provide essential information due to personal health problems were excluded from the study. Participants who could read Thai were asked to complete the questionnaire by themselves. However, for those who could not read Thai, one of the researchers who spoke Thai and Lahu conducted the interviews. The village health volunteers were asked for their help in obtaining information from the participants who could not communicate in Thai. The interviews were conducted in a private and confidential room that was provided in a community hall in the village and lasted for 30 minutes each. Data were collected from June-October 2019.

Statistical analyses

Data were entered in duplicate into Excel sheets, and errors were checked before the data were transferred into SPSS version 24, 2020 (SPSS, Chicago, IL) for the analyses. Descriptive and inferential analyses were performed. Descriptive statistics were used to describe the characteristics of the participants. Categorical data are presented as percentages, while continuous data with normal distributions are presented as the means and standard deviations (SDs). A binary logistic regression was used to detect the associations between the variables at a significance of α = 0.05. The “Enter” mode was used in the step involving the selection of the independent variables into the statistical model. The pseudo R2 of the Cox-Snell R2 and Nagelkerke R2 and the Hosmer-Lemshow chi-square were used to determine the fit of the model in all steps. In the univariable logistic model, all independent variables detected the association with the dependent variable one-by-one in each step. These variables found to be significant in the univariable logistic model were considered to be put into the multivariable model. Before fitting the final multivariable logistic model, some variables were controlled as the confounder factors before interpretation.

Ethical considerations

The study’s proposal and protocol were reviewed and approved by the Mae Fah Luang University Research Ethics Committee on Human Research (No. REH-60107). Before starting the interviews, all participants were provided all relevant and essential information. Informed consent was obtained on a voluntary basis before starting the interview. The interviews were closely monitored by a psychiatric nurse, and when they appeared to have a negative impact on the participants, it was recommended that the question be skipped or the interview be stopped. Afterward, all questionnaires were properly destroyed after coding and entering their information into an Excel spreadsheet. The data files were stored confidentially with a specific code for access.

Results

A total of 646 participants were included in the analyses, namely, 98 children aged 5–15 years, 430 women aged 16–59 years, and 118 people aged 60 years and over. Among the children, 52.0% were male, and 70.4% were attending school. Among the women, 77.0% were married, 51.9% were Buddhist, 52.0% were illiterate, and 44.2% were farmers. Among the elderly, 53.4% were female, 68.5% were married, 61.2% had no income, and 26.3% had a medical illness (Table 1).

Table 1. Characteristics of the participants.

Characteristics Total Children aged 5–15 years Women aged 16–59 years Elderly aged ≥ 60 years
n (%) n % n % n %
Total 646 (100.0) 98 100.0 430 100.0 118 100.0
Sex
    Female 540 (83.6) 47 48.0 430 100.0 63 53.4
    Male 106 (16.4) 51 52.0 N/A N/A 55 46.6
Marital status
    Single 157 (24.3) 96 98.0 58 13.5 3 2.5
    Married 414 (64.1) 2 2.0 331 77.0 81 68.6
    Ever married 75 (11.6) 0 0.0 41 9.5 34 28.8
Religion
    Buddhism 348 (53.8) 61 62.2 223 51.9 64 53.8
    Christianity 298 (46.2) 37 37.8 207 48.1 54 46.2
Education
    Illiterate 336 (52.0) 4 4.1 217 50.5 115 97.5
    Primary school 175 (27.0) 68 69.4 105 24.4 2 1.7
    Secondary school and higher 37 (21.0) 26 26.5 108 25.1 1 0.8
Occupation
    Student 101 (15.6) 69 70.4 28 6.5 4 3.4
    Unemployed 122 (18.8) 6 6.1 71 16.5 45 38.1
    Farmer 253 (39.2) 0 0.00 190 44.2 63 53.4
    Other 170 (26.4) 23 23.5 141 32.8 6 5.1
Having income
    Yes 391 (60.5) 5 5.1 352 81.9 34 28.8
    No 255 (39.5) 93 94.9 78 18.1 84 71.2
Having family financial problems in the past year
    Yes 289 (46.1) 22 22.4 201 46.7 66 55.9
    No 348 (53.9) 76 77.6 229 53.3 52 44.1
Medical illness
    Yes 71 (11.0) 0 0.00 40 9.3 31 26.3
    No 575 (89.0) 98 100.0 390 90.7 87 73.7
Role in family
    Head 145 (22.4) 0 0.00 76 17.7 69 58.5
    Member 501 (77.6) 98 100.0 354 82.3 49 41.5
Open in a new tab

Different kinds of substance use were reported by the participants: 15.3% and 7.1% of children aged 5–15 years reported alcohol use and smoking, respectively. Moreover, 37.7% and 11.9% of the women and 15.3% and 49.2% of the elderly aged 60 years and over reported alcohol use and smoking, respectively (Table 2).

Table 2. Substance use behaviors among Lahu children, women, and the elderly.

Characteristic Children aged 5–15 years Women aged 16–59 years Elderly aged ≥ 60 years
n % n % n %
Smoking
    Yes 7 7.1 51 11.9 58 49.2
    No 91 92.9 379 88.1 60 50.8
Alcohol use
    Yes 15 15.3 162 37.7 15 15.3
    No 83 84.7 268 62.3 83 84.7
Marijuana use
    Yes 1 1.0 3 0.7 1 0.8
    No 97 99.0 427 99.3 117 99.2
Amphetamine use
    Yes 0 0.0 5 1.2 2 1.7
    No 98 100.0 425 98.8 116 98.3
Glue use
    Yes 0 0.0 2 0.5 0 0.0
    No 98 100.0 428 99.5 118 100.0
Opium use
    Yes 0 0.0 2 0.5 1 0.8
    No 98 100.0 428 99.5 117 99.2
Family member who uses alcohol
    Yes 28 28.6 136 31.6 36 30.5
    No 70 71.4 294 68.4 82 69.5
Open in a new tab

Six out of 98 children (6.1%) had experienced some form of domestic violence; 4.1% had been insulted or addressed rudely, 4.1% had been addressed with anger, 3.1% had been forced to ask for money from others, etc. Seventy-four women (17.2%) had experienced some form of domestic violence; 13.3% had experienced sexual abuse, 6.3% had been insulted or addressed rudely, 6.0% had been forced to buy food or drink, etc. Twenty-three of the 118 elderly (19.5%) had experienced some form of domestic violence; 4.2% had been forced to ask for money from others, 3.4% had been kicked out of the house, 3.4% had been forced to buy food or drink or forced to work, etc. (Table 3).

Table 3. Characteristics of domestic violence among Lahu children, women, and the elderly in the past year.

Characteristic Children aged 5–15 years Women aged 16–59 years Elderly aged ≥ 60 years
n % n % n %
Total participants 98 100.0 430 100.0 118 100.0
Total experiencing domestic violence (persons)
    Yes 6 6.1 74 17.2 23 19.5
    No 92 93.9 356 82.8 95 80.5
Physical abuse
        Getting kicked out of the house
        Yes 0 0.0 10 2.3 4 3.4
        No 98 100.0 420 97.7 114 96.6
  Being pushed, pulled or scratched or having items thrown at them
    Yes 0 0.0 10 2.3 2 1.7
    No 98 100.0 420 97.7 116 98.3
  Being slapped, hit, kicked or strangled
    Yes 0 0.0 4 0.9 2 1.7
    No 98 100.0 426 99.1 116 98.3
  Being threatened with a weapon
    Yes 0 0.0 2 0.5 0 0
    No 98 100.0 428 99.5 118 100.0
  Being forced to drink alcohol, smoke, or use substances
    Yes 2 2.0 19 4.4 2 1.7
    No 96 98.0 411 95.6 116 98.3
Mental abuse
  Being insulted or addressed rudely
    Yes 4 4.1 27 6.3 2 1.7
    No 94 95.9 403 93.7 116 98.3
  Being addressed with anger
    Yes 4 4.1 25 5.8 2 1.7
    No 94 95.9 405 94.2 116 98.3
  Being told that another person in the household will kill himself or herself
    Yes 0 0.0 5 1.2 3 2.5
    No 98 100.0 425 98.8 115 97.5
  Being asked to kill others
    Yes 1 1.0 2 0.5 3 2.5
    No 97 99.0 428 99.5 115 97.5
  Being forced to buy food or drink or being forced to work
    Yes 2 2.0 26 6.0 4 3.4
    No 96 98.0 404 94.0 114 96.6
  Being forced to ask for money or borrow items from others
    Yes 3 3.1 22 5.1 5 4.2
    No 95 96.9 408 94.9 113 95.8
Sexual abuse **(only women)
    Yes 1 2.0 57 13.3 1 1.6
    No 48 98.0 373 86.7 62 98.4
Open in a new tab

In the univariate analysis, smoking was found to be associated with experiencing domestic violence in Lahu children. In the multiple logistic regression analysis, children who smoked had 8.70 times (95%CI = 1.27–59.45) greater odds to suffer domestic violence than children who did not smoke (Table 4).

Table 4. Univariate and multivariate analyses of the factors associated with domestic violence among Lahu children.

Characteristic Domestic violence OR 95%CI p-value AOR 95%CI p-value
Yes No
n % n %
Total 6 6.1 92 93.9 NA NA NA NA NA NA
Age (years)
    5–10 1 2.6 37 97.4 1.00
    11–15 5 8.3 55 91.7 3.36 0.38–29.97 0.227
Marital status
    Single 6 6.3 90 93.8 NA
    Married 0 0.0 2 100.0 NA NA NA
Religion
    Buddhism 2 3.3 59 96.7 1.00
    Christianity 4 10.8 33 89.2 3.58 0.62–20.58 0.154
Education
    Illiterate 1 25.0 3 75.0 2.56 0.20–33.16 0.473
    Primary school 2 2.9 66 97.1 0.23 0.04–1.48 0.122
    Secondary school and higher 3 11.5 23 88.5 1.00
Occupation
    Student 4 5.8 65 94.2 1.00
    Employed and unemployed 2 6.9 27 93.1 1.20 0.20–6.96 0.836
Having income
    No 5 5.4 88 94.6 1.00
    Yes 1 20.0 4 80.0 4.40 0.41–47.04 0.220
Having family financial problems in the past year
    No 2 9.1 20 90.9 1.00
    Yes 4 5.3 72 94.7 0.55 0.09–3.05 0.514
Health illness
    No 6 6.1 92 93.9 NA
    Yes 0 0.0 0 0.0 NA NA NA
Role in family
    Head 6 6.1 92 93.9 NA
    Member 0 0.0 0 0.0 NA NA NA
Smoking
    No 4 4.4 87 95.6 1.00 1.00
    Yes 2 28.6 5 71.4 8.70 1.27–59.45 0.027* 8.70 1.27–59.45 0.027*
Alcohol use
    No 4 4.8 79 95.2 1.00
    Yes 2 13.3 13 86.7 3.04 0.51–18.31 0.225
Having a family member who uses alcohol
    Yes 3 10.7 25 89.3 2.68 0.50–14.16 0.245
    No 3 4.3 67 95.7 1.00
Marijuana use
    No 6 6.2 91 93.8 NA
    Yes 0 0.0 1 100.0 NA NA NA
Amphetamine use
    No 6 6.1 92 93.9 NA
    Yes 0 0.0 0 0.0 NA NA NA
Glue use
    No 6 6.1 92 93.9 NA
    Yes 0 0.0 0 0.0 NA NA NA
Opium use
    No 6 6.1 92 93.9 NA
    Yes 0 0.0 0 0.0 NA NA NA
Open in a new tab

* Significance level at α = 0.05

** Significance level at α = 0.05 after controlling for religion and education.

In the univariate analysis, it was found that four variables were associated with domestic violence in Lahu women, specifically, having family financial problems in the past year, having a role as a family member, using alcohol, and having a family member who uses alcohol.

In the multiple logistic regression analysis, four variables were found to be associated with domestic violence in Lahu women, namely, having a role as a family member, using alcohol, having family financial problems, and having a family member who uses alcohol. The women who had a role as a family member had 1.59 times (95%CI = 1.02–2.50) greater odds of experiencing domestic violence than the women who were the head of the family. The women who used alcohol had 3.36 times (95%CI = 2.27–4.99) greater odds of suffering domestic violence than the women who did not use alcohol. The women who had family financial problems in the past year had 4.01 times (95%CI = 2.52–7.66) greater odds of enduring domestic violence than the women who did not have family financial problems. The women who had a family member who used alcohol had 2.87 times (95%CI = 2.20–5.63) greater odds of experiencing domestic violence than the women who did not have a family member who used alcohol (Table 5).

Table 5. Univariate and multivariate analyses of the factors associated with domestic violence among Lahu women.

Characteristic Domestic violence OR 95%CI p-value AOR 95%CI p-value
Yes No
n % n %
Total 74 17.2 356 82.8 NA NA NA NA NA NA
Age (years)
    16–30 35 20.0 140 80.0 1.56 0.81–3.03 0.182
    31–45 24 16.4 122 83.6 1.23 0.61–2.48 0.557
46–59 15 13.8 94 86.2 1.00
Marital status
    Single 8 13.8 50 86.2 1.00
    Married 64 19.3 267 80.7 1.50 0.68–3.32 0.319
    Ever married 2 4.9 39 95.1 0.32 0.06–1.60 0.165
Religion
    Buddhism 34 15.2 189 84.8 1.00
    Christianity 40 19.3 167 80.7 1.33 0.81–2.20 0.264
Education
    Illiterate 35 16.1 182 83.9 1.11 0.58–2.10 0.759
    Primary school 23 21.9 82 78.1 1.61 0.80–3.26 0.183
    Secondary school and higher 16 14.8 92 85.2 1.00
Occupation
    Student 2 7.1 26 92.9 1.00
    Unemployed 17 23.9 54 76.1 4.09 0.88–19.05 0.073
    Farmer 37 19.5 153 80.5 3.14 0.71–13.84 0.130
    Employed 18 12.8 123 87.2 1.90 0.42–8.71 0.407
Having income
    No 16 20.5 62 79.5 1.31 0.71–2.43 0.394
    Yes 58 16.5 294 83.5 1.00
Having family financial problems in the past year
    No 18 62.1 211 37.9 1.00
    Yes 56 27.9 145 72.1 4.52 2.55–8.01 <0.001* 4.01 2.52–7.66 <0.001*
Health illness
    No 68 17.4 322 82.6 1.00
    Yes 6 15.0 34 85.0 0.83 0.33–2.06 0.697
Role in family
    Head 7 9.2 69 90.8 1.00
    Member 67 18.9 287 81.1 2.30 1.01–5.23 0.047* 1.59 1.02–2.50 0.042*
Smoking
    No 65 17.2 314 82.8 1.00
    Yes 9 17.6 42 82.4 1.04 0.48–2.23 0.930
Alcohol use
    No 36 13.4 232 86.6 1.00
    Yes 38 23.5 124 76.5 1.98 1.19–3.27 0.008* 3.36 2.27–4.99 <0.001*
Having a family member who uses alcohol
    Yes 43 31.6 93 68.4 3.92 2.33–6.59 <0.001* 2.87 2.20–5.63 <0.001*
    No 31 10.5 263 89.5 1.00
Marijuana use
    No 72 16.9 355 83.1 1.00
    Yes 2 66.7 1 33.3 9.86 0.88–110.21 0.063
Amphetamine use
    No 72 16.9 353 83.1 1.00
    Yes 2 40.0 3 60.0 3.27 0.54–19.91 0.199
Glue use
    No 73 17.1 355 82.9 1.00
    Yes 1 50.0 1 50.0 4.86 0.30–78.64 0.265
Opium use
    No 74 17.3 354 82.7 NA
    Yes 0 0.0 2 100.0 NA NA NA
Open in a new tab

* Significance level at α = 0.05

** Significance level at α = 0.05 after controlling for age, religion and education.

In the univariate analyses, the following four variables were found to be associated with domestic violence in the Lahu elderly: having family financial problems in the past year; smoking; using alcohol; and having a family member who uses alcohol.

In the multiple logistic regression analysis, three variables were found to be associated with domestic violence in the Lahu elderly as follows: having family financial problems in the past year; using alcohol; and having a family member who uses alcohol. The elderly who used alcohol had 3.25 times (95%CI = 1.08–9.81) greater odds of suffering domestic violence than the elderly who did not use alcohol. The elderly who had a family member who uses alcohol had 3.31 times (95%CI = 1.26–7.34) greater odds of experiencing domestic violence than the elderly who did not have a family member who uses alcohol. The elderly who had family financial problems in the past year had 2.16 times (95%CI = 1.16–8.77) greater odds of undergoing domestic violence than the elderly who did not have family financial problems (95%CI = 1.16–8.77) (Table 6).

Table 6. Univariate and multivariate analyses of the factors associated with domestic violence among the Lahu elderly.

Characteristic Domestic violence OR 95%CI p-value AOR 95%CI p-value
Yes No
n % n %
Total 23 19.5 95 80.5 NA NA NA NA NA NA
Age (years)
    60–70 19 21.8 68 78.2 1.88 0.59–6.06 0.287
    >70 4 12.9 27 87.1 1.00
Marital status
    Single 1 33.3 2 66.7 2.33 0.18–30.10 0.516
    Married 16 19.8 65 80.2 1.15 0.41–3.24 0.793
    Ever married 6 17.6 28 82.4 1.00
Religion
    Buddhism 9 14.1 55 85.9 1.00
    Christianity 14 25.9 40 74.1 2.14 0.84–5.43 0.110
Attended school
    No 21 18.3 94 81.7 1.00
    Yes 2 66.7 1 33.3 8.95 0.78–103.39 0.079
Having income
    No 3 8.8 31 91.2 3.23 0.89–11.70 0.074
    Yes 20 23.8 64 76.2 1.00
Having family financial problems in the past year
    No 5 9.6 47 90.4 1.00
    Yes 18 27.3 48 72.7 3.50 1.20–10.27 0.020* 2.61 1.16–8.77 0.018*
Health illness
    No 19 21.8 68 78.2 1.00
    Yes 4 12.9 27 87.1 0.53 0.16–1.70 0.287
Role in family
    Head 15 78.3 54 78.3 1.42 0.55–3.68 0.466
    Member 8 16.3 41 83.7 1.00
Smoking
    No 5 8.3 55 91.7 1.00
    Yes 18 31.0 40 69.0 4.95 1.70–14.45 0.003*
Alcohol use
    No 9 11.3 71 88.8 1.00 1.00
    Yes 14 36.8 24 63.2 4.60 1.77–11.98 0.002* 3.25 1.08–9.81 0.036*
Having a family member who uses alcohol
    Yes 13 36.1 23 63.9 4.06 1.57–10.50 0.003* 3.31 1.26–7.34 0.001*
    No 10 12.2 72 87.8 1.00
Marijuana use
    No 23 19.7 94 80.3 NA
    Yes 0 0.0 1 100.0 NA NA NA
Amphetamine use
    No 23 19.8 93 80.2 NA
    Yes 0 0.0 2 100.0 NA NA NA
Opium use
    No 22 18.8 95 81.2 NA
    Yes 1 100.0 0 0.0 NA NA NA
Open in a new tab

* Significance level at α = 0.05

** Significance level at α = 0.05 after controlling for age, religion and education.

Discussion

The Lahu people in Thailand have a poor socioeconomic status and work as agriculturalists. With respect to substances, using alcohol and smoking are the most common practices. Children, women, and the elderly are vulnerable to domestic violence, and one-fifth of them had experienced domestic violence in the past year. A wide range of physical and mental abuse was reported, from low to heavy levels of harm, such as being asked to kill the vulnerable individuals in Lahu families. Some characteristics were detected as the factors associated with the presence of domestic violence in Lahu families, such as using alcohol, having a family member who uses alcohol, having family financial problems in the past year, and smoking.

In the Lahu family context, it was found that the prevalence of domestic violence in the three major victim groups of children (aged 5–15 years), women (aged 16–59 years), and the elderly (aged ≥ 60 years) in the past year was 6.1%, 17.2%, and 19.5%, respectively. The WHO [28] has reported that the rate of domestic violence against women is 30.0%, which is higher than the estimates in the Lahu community in Thailand. A very interesting meta-analysis of the prevalence and health outcomes of domestic violence among the people attended to in hospitals of Arab countries found that 73.3% of women had been exposed to some form of domestic violence in their lifetimes [29]. Regarding domestic violence against children, the Centers for Disease Control and Prevention (CDC) reported that 11.0% of children have been exposed to some form of family violence in the past year in American families [30], while a prevalence of 34.0% was reported in Canada [31]. The WHO has also presented that the problem of child abuse is related to family economics, parents’ education, parenting styles, and other environmental factors [28]. Regarding the elderly, a systematic review study reported that the prevalence of domestic violence for the elderly (aged ≥ 60 years) was 5.6–14.1%, including neglecting family members’ essential needs for daily life in the past year [32]. Thailand is reported to have a 14.0% prevalence of domestic violence among the elderly [10]. This problem increases according to family economics and chronic health problems in the elderly [10]. Accordingly, Lahu children and women are living with less domestic violence than other communities, while the Lahu elderly have a greater chance of experiencing domestic violence than the elderly in other communities.

Among Lahu children, it was found that smoking is associated with them suffering domestic violence. This finding is supported by a study in Thailand [33], which reported that any form of family member smoking behavior is related to experiencing domestic violence. Sharma, et al. [34] clearly demonstrated the relationship between family member smoking and family financial problems. Moreover, smoking and amphetamine use are closely related, particularly among the Lahu people, which is supported by Apidechkul et al. [24]. Parents do not desire to see their child involved in amphetamine use, which is illegal in Thailand, and this results in domestic violence toward children. Many studies have reported and confirmed the association between substance use and domestic violence in both straight and reverse effects [15, 23, 24, 34, 35]. Moreover, on explanation is that parents do not want to see their child smoking; therefore, domestic violence could be occurring to stop children from smoking [23].

Lahu women who are defined as a family member with a role in in their family are at a greater risk of experiencing domestic violence than Lahu women who are the head of the family. This is supported by a study in Jordan that reported that women who lived with an extended family and who were not the family head were at risk of suffering domestic violence more than women who were the family head [36]. Moreover, the WHO has reported that women in many societies and families are commonly identified as one of the key victims of domestic violence due to a lack of power to protect themselves from their husband due to their role in their family [28]. With less power in the family, Lahu women are disadvantaged by any family decision, and when they disagree with their husband’s decision, violence can occur.

Interestingly, having financial problems in the past year in a Lahu family was found as one of the factors associated with domestic violence against women aged 16–59 years and against the elderly aged 60 years and over. This is supported by a study conducted by Stylianou [37]. Another demonstration of the relationship between family financial problems and domestic violence was reported by the Queensland Domestic and Family Violence Research Centre in 2017 [38]. The WHO has also reported that family financial problems are one of the contributors to domestic violence, particularly toward women and the elderly [39]. Additionally, in this study, it was found that the Lahu people in Thailand have a poor family economic status. This poor family economic status can easily contribute to domestic violence against women and the elderly.

In our study, it was found that Lahu women and the elderly were more likely to experience domestic violence while living with family members who use alcohol. This is confirmed by an official report by the WHO [40] that showed that alcohol is related to the occurrence of domestic violence in any society. A study in the United Kingdom (UK) also showed that alcohol was detected as a major cause of domestic violence [41]. Chan [42] and Yates [43] clearly demonstrated the association between alcohol use and domestic violence in the Australian context. In Thailand, a retrospective study conducted by Tongsamsi [16] reported that alcohol was found to be the major cause of domestic violence in Thailand. Moreover, from the national survey of smoking and alcohol in Thailand in 2017, it was found that families with members who use alcohol show significantly greater evidence of domestic violence than families with members who do not use alcohol [44]. According to the nature of the Lahu people, particularly young adults, they are very familiar with alcohol use [23], and after drinking alcohol, it would be easy for the elderly in the family to experience violence in some form.

Some participants could not understand the questions well, and the community health volunteers were asked to help as translators to improve the understanding of the context of the question asked before the response. Given the methodology used, the responses to some sensitive questions may not have been accurate, such as the pattern of experiencing some type of harm and amphetamine use. As per their norms, the Lahu people do not typically tell family stories to the public, particularly to individuals of a different gender. To minimize this problem, we used gender-matched interviewers and participants to collect the data. Moreover, as per the Lahu culture, the participants were not familiar with talking about domestic conflicts or conflicts within the family with people outside the family.

Conclusion

The Lahu people in Thailand live with a low education level and work as agriculturists. Having three generations (grandparents, parents, and children) live together in the Lahu family is common. Smoking in children leads to them experiencing domestic violence, while alcohol use and family financial problems lead to the elderly suffering domestic violence. Regarding Lahu women experiencing domestic violence, using alcohol, having a family member who uses alcohol, having a role as a family member in their family and family financial problems are detected as the contributing factors. For recommendations to stop domestic violence at the individual level, the Lahu people should be encouraged to reduce or stop alcohol consumption and smoking. Improving personal skills to detect and prevent violence perpetrated by family members is also important. At the community level, all stakeholders in a community, including community leaders and health workers, have to be trained in family violence detection and reduction, particularly in conducting a campaign to minimize substance use among villagers. At the policy maker level, there should be concern, and policies should be formed to reduce the substance use problem and improve working skills for the Lahu people.

Supporting information

S1 Appendix. Questionnaire used in the study (Thai version).

(PDF)

Click here for additional data file. (88.8KB, pdf)
S2 Appendix. Questionnaire used in the study (English version).

(PDF)

Click here for additional data file. (91.5KB, pdf)
S3 Appendix. Data file of the study.

(SAV)

Click here for additional data file. (493.1KB, sav)

Acknowledgments

We would like to thank all the village headmen and participants for providing all the essential information.

Data Availability

All relevant data are within the manuscript and its Supporting Information files.

Funding Statement

The project was supported by The Center for Alcohol Studies, Thailand (Grant No 60-A-0011). Dr. Tawatchai Apidechkul was the principle investigator and received the grant. However, the grant funder did not involve in all steps of the study.

References

Decision Letter 0

Siyan Yi

29 Oct 2020

PONE-D-20-27497

Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study

PLOS ONE

Dear Dr. Apidechkul,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Dec 13 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Siyan Yi, MD, MHSc, PhD

Academic Editor

PLOS ONE

Additional Editor Comments:

We have received comments from three reviewers. One reviewer recommended rejection, one recommended minor revisions, and the other one recommended major revisions. The primary concerns are around the sampling and data analysis methods. I agree with two reviewers that stratified analyses should have been performed to explore the differences in domestic violence prevalence and its risk factors in different sub-populations. The sub-populations included girls and boys under 15 years, women and men, and female and male elderly over 60 years. The authors expressed an interest in studying the subgroups and should have reflected this in the analyses. The lack of sub-group results introduced a challenge in discussing and comparing this study's findings against other studies. We will consider this manuscript for further review if the authors could address these concerns.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.  

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services.  If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

  • The name of the colleague or the details of the professional service that edited your manuscript

  • A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

  • A clean copy of the edited manuscript (uploaded as the new *manuscript* file)

3. Please include a copy of the questionnaire, in the original language,as Supporting Information.

4. In statistical methods, please clarify whether you corrected for multiple comparisons.

5. Cross-sectional study design cannot be used to determine causation. In this light, please revise your conclusions to indicate that your study presents associations between exposure variables and domestic violence. For example, your statement that "smoking and alcohol and amphetamine use, leads to domestic violence" implies causation which is cannot be determined using your study design.

6. Please amend your list of authors on the manuscript to ensure that each author is linked to an affiliation. Authors’ affiliations should reflect the institution where the work was done (if authors moved subsequently, you can also list the new affiliation stating “current affiliation:….” as necessary).

7. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please ensure that your ethics statement is included in your manuscript, as the ethics statement entered into the online submission form will not be published alongside your manuscript.

8. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for the opportunity to review “Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study.” The authors have tackled an important public health issue ie violence against the 3 vulnerable groups. However, I have several pertinent issues of concern with the study/paper, which I will outline below.

i) One of the main concern the study methodology and analysis procedure. The author indicated in their methods section of the paper that the target populations of interest were children (<=15 yrs), women and the elderly. They have also outlined the questionnaire items used to measure violence in each of the 3 groups. The authors then went on to derived a measure for domestic violence from these items. Considering that the drivers of violence in the 3 groups are likely to be different, I would have preferred if the authors conducted a stratified analysis instead so they can look at these groups separately. The type of violence experienced by the 3 groups are different and most likely perpetrated by different members of the HH. Possibly the factors associated with violence experienced by the 3 groups are also different.

ii) Another concern is the study methodology. The authors have not clearly outlined how participants from households were recruited. Did they include everyone in the household (children, mother & father and the elderly)? If they did, did they account for clustering by household in their analysis?

If they did not, then they need to clearly explain in their methodology how HH members were selected.

iii) The authors have defined children as <=15yrs. They need to clearly indicate the minimum age for this group in the methods section (as part of inclusion criteria definition) and how data was collected from them ( eg was it interviewer administered?). I understand in Table 1 they have outlined that the minimum age in the sample was 5yrs. But it is not clear whether 5yrs was a cut off or just minimum age for those interviewed. I am also rather concerned about the integrity of information on violence corrected from children (10yrs or less)

iv) The authors also indicated in the methods section that the target population of interest were children (<=15yrs), women(16-59yrs) and the elderly (>=60yrs). However, their data shows that they also have men (n=382) aged between 16-59yrs in the sample. I do not understand the rationale for including men (16-59yrs) in the study.

v) The general norm for violence studies is to analyze data for male and females separately as they tend to have different risk factors. I would prefer that they present their analysis that way.

Other general comments are:

• They can make their tables more succinct by reporting one category for all binary variables (‘yes”) rather both ‘yes’ and ‘no’

• The manuscript can do with some proof reading and editing.

• The author should consider having prevalence estimates for each type of violence (eg physical violence).

• With the study having 3 groups, it is not clear if participants were reporting their own behavior ( eg alcohol use) and wondering what could be the impact of these on the estimates.

• Sample size formula (if it is necessary to show it) should be explained in the context of the study, e.g what is p in this study?

• They could have a separate table for the multivariate analysis.

Reviewer #2: Background

1.Please clarify the scope of this study because there is different definition between domestic violence and intimate partner violence so the reader might more clear and understand (Domestic violence is violence in the family, the perpetrator is family member, whereas intimate partner violence is violence between partner)

2. The authors should mention research gaps and why have to conduct this study in Lahu tribe.

Method

1. How the researchers randomly selected the village?

2. The sample size calculation is n=826 but the researcher collected the data from n=1,028. Please calrify and give the explanation

3. What is the reliability (pre-test) value from this study?

Result

1. Heading of Table 3 show n= 647 but in detail in the table showed n= 1,028. However, some item "n" is lower than 1,028. Please check the consistency of the data

2. Recoomendation pls consider to give recommendation to 3 level; individual, community, and policy level

Reviewer #3: General comments: Authors have made a good attempt to analyse the extent of domestic violence and factors associated with domestic among children, women and the elderly in a poor community in rural Thailand. This paper will contribute to understanding of domestic violence in low and middle income countries, and provide data for understanding subgroups. However, the authors have introduced interest in studying subgroups but have not done so in the analyses, and this presents a challenge in the analysis and discussion of the generalize results as they often seek to make comparisons or justifications of their results against subgroups elsewhere without actual analyses of their own to back up the comparisons. My recommendations are that the authors must revise their analyses and present the overall study population analysis and present children: girls and boys under 15 years, women and men (by whichever is legal age limit for marriage perhaps), elderly women and men over 60 years. The lack of data on perpetrators is a major limitation but inferences may be safely drawn with supporting study as to who the common perpetrators of DV usually are in such contexts in Thailand. The subgroup analyses will help authors make a case for the understanding of how substance abuse increases risk of violence for specific subgroups to some extent even without understanding the perpetrators data.

Abstract:

Revise according to the recommendations from the content of the manuscript

Background:

“Lead to”: Authors may not reach conclusion that certain variable ‘lead to’ certain risks with cross-sectional data including their own study as such studies only allows for determination of risk factors. We do speak of associations unless experimental studies were done.

“Substance use” versus abuse: authors have described substance ‘use’ and not abuse this makes it difficult to determine risk outright without measuring the extent of misuse.

The background starts well with the global picture but needs to locate the domestic violence problem and risk factors in the context of Thailand, and do so indicating the extent to which other studies have addressed this issue.

- Go into some detail about the Thai social life, income, earnings, religion, culture and describing who Thai people are, what other tribes or social groupings exist in Thailand and the extent of violence there as well as any risk factors or protective factors against DV.

- Also contrast with other neighbouring countries where possible, at least in terms of the prevalence.

- Any cross-border issues that make the hill tribes different from other Thai society?

- Describe the context of substance use and substance abuse in Thailand. How are amphetamines accessed? Natural or other synthetic stimulants? History of substance abuse. What kind of alcohol is used or abused? How is it sourced in communities which have been described as poor? How do they access amphetamines?

- I also propose that some effort into explaining violence against children younger than 15 years be made, and explaining how such studies were done in Thailand or neighbouring countries. What are the risk factors?

Page 3:

Paragraph 1:

- the mention of the SDGs: This is a goal and not a fact, so the authors need to talk about this SDG in the correct context. What are authors trying to say about this goal?

- at the end of paragraph: What do the authors mean ‘domestically’?

Paragraph 2:

- at the end of the paragraph, authors mention contributors to violence: Do authors mean ‘perpetrators’? how do they contribute? If as the perpetrators then authors should state that explicitly. And mention other contributors if they exist. In some context, we have instigators and some Asian studies refer to instigators.

Last paragraph describing hill tribes needs to go into the Setting/study site information in the methods section and explain the following: What is the population size relative to the largest and smallest tribe? Describe the lifestyle that distinguishes Lahu people from others?

Last paragraph of the Background section: Provide a justification for the paper explaining the conceptual framework used for this study as mentioned in the methods section. Thus explaining why a paper on prevalence and associated factors. Why was the study conducted? And why its important for the paper to be published?

Methods:

Paragraph 1 - Explain size of the Lahu hill area. Sources of living such as access to water, income, who is working or not, what is the family structure?

Paragraph 2 - Study sample recruitment needs to be explained. The study aim suggests data was collected for children, women and older persons. It is not clear though what the sampling unit was for the study. Was it families or just individuals in the community?

Paragraph 4 regarding the questionnaire:

- Which questions were inappropriate for the Lahu people due to culture and beliefs?

- Last sentence about pilot is incomplete. Explaining the pilot is good. So finish the sentence please.

Paragraph 5 about the questionnaire:

- Authors need to present a table of each set of questions measured, describing what the variable is made up of, and explaining the source of the variables especially if used in other studies or new including those verified and changed with inputs from the Lahu people.

- Please name that dependent variable and clarify what it comprised of it is a combination of variables. What is the outcome variable made up of?

- Domestic violence questions need to be explained in more detail as an outcome variable. The explanation given here is different compared to what is presented in the results. Clearly explain all variables here first, what they entail including any variation that may be presented in the results. Also provide the time frame that is reported for each form of DV, lifetime experience or in the past 12 months?

Paragraph 6:

- Informed consent forms are meant to remain with the authors particularly to demonstrate that consent was sought and obtained from participants. Informed consent is also a process, it is not clear how this was done. Rather explain how information about the study was provided to participants, what was were these essential aspects that were explained, and how consent was given by particpants.

- Sentence: “People who were not able to provide essential information due to personal health problems were excluded from the study” does not explain the eligibility criteria for sample to participate in the study. Please explain the criteria used, was it one factor or multiple factors that were considered?

- Issue of language used by participants: Move this to the questionnaire section. Was this a paper-based questionnaire? Explain how data was collected in more precisely, like language, who collected the data, how they were trained, their level of qualification. What the role of the village health volunteers including what they usually do.

- How was a confidential room determined? Maybe explain this in a different way? Merge the note about ethical clearance at the end of this paragraph.

Data analysis:

Logistic regression modelling is a process, explain that process. What was the outcome variable, candidate variable, how they were determined first, were bivariate analyses performed first, how were candidate variables selected and fitted into the model, any process of elimination engaged? This is crucial information to determine what was done and whether it can be replicated and reach the same conclusions the authors did.

Results section:

- Table 1 and Table 2 and Table 3 are difficult to decipher because the study is looking at 3 sets of participants: children under 15, women, older persons older than 60.

- Authors need to demonstrate the same characteristics for the study population alongside these characteristics for the study population sets. Including data from men and boys and classifying elders by difference between women, men, girls, boys, older women, older men. The socio-demographics do not make sense without a clear understanding of what the univariate data looks like for all sets of individuals. This is important as data also needs to account for any child marriage which is possible under 15 years, number of family members living in the same household, etc. Vulnerability to violence is also determined by gender. Cross tabulation of the differences by gender is also key in this analysis and table.

- In Table 2 there are other drugs that have been measured which are not explained in the methods section. This justifies why a table including types of substances is needed.

- Table 3: Authors can provide an overall estimation of DV for the population of interest, but still need to extent the analysis to demonstrate the differences between children, women and elderly persons, so that there is clarity on the extent of vulnerability in the population including their subgroups, this is most useful to provide guidance to which groups health interventions should prioritise and address and what actual interventions could be considered for which subgroups or to address which factors associated with vulnerability of which subgroup. Lumping all children, women and elderly persons is not justified as an end in itself, but would be more valuable if authors are able to single out the most vulnerable, even at a comparative level. The analysis is insufficient at this level and needs to demonstrate the complexities that are entailed in domestic violence as a multi-layered construct that can be differentiated for women into intimate partner violence or domestic violence by family members other than husband if women are married, or child abuse by parents or other family members, or even strangers if the latter was measured, etc.

- In Table 3, for each type of domestic violence need to provide a composite estimate as done with domestic violence. This will explain the extent of physical or emotional abuse overall.

- Authors need to explain the risk of smoking and how it links to perpetration of violence, as this variable doesn’t make sense in terms of how it increases vulnerability to DV. Do this in the background section.

- Table 4: The analysis appears to have skipped a stage as not all variables need to go into the multivariate analysis, so authors need to demonstrate candidate variables through bivariate analysis and look at some degree of association of each variable with the outcome variable. So bivariate analysis is required. Authors much explain Why are some analyses adjusted for and others are not?

Discussion:

- The authors are evidently struggling to explain the study prevalence in it entirety without a clear indication per children, women, elderly women or men subsets analysed. The data does not provide data to indicate the extent of violence against children. The story of substance use in general is also not telling a clear story without understanding the extent to which increased risk and perhaps authors should also report on who is using, and whether combinations of substances are used and determine risk from that angle. However, the conclusions reached in this paper cannot be reached with the currently presented analyses. Authors need to address the analytical questions about extent of vulnerability to violence per subgroup, what increases their vulnerability in most precise terms than the general picture painted already which does not in itself make much sense. Then tell the role substances play in this story.

- Paragraph 2: These comparisons are incorrect without a clear understanding of the prevalence of DV among women from your analysis, so this is incorrect comparison. The same point applies to comparisons on data about children or the elderly. The analyses failed to demonstrate prevalence for each subgroup so there cannot be comparisons with subgroups.

- Paragraph 3: This is suggesting that Buddist society is more gender equitable, where is the data to support this? The argument does not make sense as patriarchy is a common factor across all societies. Authors need to provide a clear explanation of the religious differences from the background and link their interpretation to that. It may be that there is a vast difference based on religions but the current argument to explain these results is incorrect without studies to support it. Authors need to explain what they mean by ‘complicated relationships’. How is Buddhism organized and are relationships there complicated too or not?

Conclusions:

The conclusions need to be strengthened and be drawing out what the results have demonstrated about the study sample as sub-groups and as a whole, and indicating more specifically what recommendations suite which aspect of the results.

Abbreviations:

Authors must integrate in the texts and ensure each is explained in full at first mention and then abbreviated afterwards.

Ethical considerations

- must be integrated with the details on how consent was obtained and data collected. Add a little bit on what DV specific considerations were made? What guidelines were followed for conducting these interviews on such a sensitive subject? Explain data storage.

References:

Many of the references do not follow the journal’s preferred referencing style. Visit: https://journals.plos.org/plospathogens/s/submission-guidelines#loc-references for information on Vancouver style. Here is some simpler information about what you can do and examples provided.

PLOS uses the reference style outlined by the International Committee of Medical Journal Editors (ICMJE), also referred to as the “Vancouver” style. Example formats are listed below. Additional examples are in the ICMJE sample references. A reference management tool, EndNote, offers a current style file that can assist you with the formatting of your references. If you have problems with any reference management program, please contact the source company's technical support.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Mar 15;16(3):e0248587. doi: 10.1371/journal.pone.0248587.r002

Author response to Decision Letter 0

27 Dec 2020

Response go reviewers’ comments

Dear Editor,

We have used long time in re-anaysis and re-forming the presentation, checking in whole text and response to all comments of reviewers. Some comments, we could not response because there is no information available. However, we very hope that you and reviewers would be happy in this version.

Finally, we sorry that using a longer time to be able to response to your comments.

Best regards,

TK

Editor comment

We have received comments from three reviewers. One reviewer recommended rejection, one recommended minor revisions, and the other one recommended major revisions. The primary concerns are around the sampling and data analysis methods. I agree with two reviewers that stratified analyses should have been performed to explore the differences in domestic violence prevalence and its risk factors in different sub-populations. The sub-populations included girls and boys under 15 years, women and men, and female and male elderly over 60 years. The authors expressed an interest in studying the subgroups and should have reflected this in the analyses. The lack of sub-group results introduced a challenge in discussing and comparing this study's findings against other studies. We will consider this manuscript for further review if the authors could address these concerns.

1.We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

: Thank you very much for the points raised, we have a lot concern about that. The manuscript is checked and re-edited by the American Journal Experts.

2. Please include a copy of the questionnaire, in the original language, as Supporting Information.

: Thank you, it’s attached (Thai version).

3. In statistical methods, please clarify whether you corrected for multiple comparisons.

: We revised the whole section of data analysis. There are divided into three groups; children (5-15 years), women (aged 16-59 years), and elderly (aged 60 years and over).

All steps of the analysis, we sued the multiple logistic regression to compare between who had violence and did not to fine the final model before interpreting the factors associated with domestic violence in each group.

4. Cross-sectional study design cannot be used to determine causation. In this light, please revise your conclusions to indicate that your study presents associations between exposure variables and domestic violence. For example, your statement that "smoking and alcohol and amphetamine use, leads to domestic violence" implies causation which is cannot be determined using your study design.

: Thank you very much for such great comment. All points have been revised including the mathematic forms used in abstract, results, and discussion (page 2; lines 14-24; page 11, lines 7-15.

5. Please amend your list of authors on the manuscript to ensure that each author is linked to an affiliation. Authors’ affiliations should reflect the institution where the work was done (if authors moved subsequently, you can also list the new affiliation stating “current affiliation:….” as necessary).

: Thank you, we have done in double checked.

6. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please ensure that your ethics statement is included in your manuscript, as the ethics statement entered into the online submission form will not be published alongside your manuscript.

: Thank you so much, it has been moved into the proper place (page 6, line 10-16). All information in the ethical statement is the same put into the system during submission the manuscript.

8. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

: Thank you very much for the suggestion. All relevant supporting information have been placed in a proper place in the manuscript with corresponding to the journal guideline.

Reviewer #1: Thank you for the opportunity to review “Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study.” The authors have tackled an important public health issue ie violence against the 3 vulnerable groups. However, I have several pertinent issues of concern with the study/paper, which I will outline below.

i) One of the main concern the study methodology and analysis procedure. The author indicated in their methods section of the paper that the target populations of interest were children (<=15 yrs), women and the elderly. They have also outlined the questionnaire items used to measure violence in each of the 3 groups. The authors then went on to derived a measure for domestic violence from these items. Considering that the drivers of violence in the 3 groups are likely to be different, I would have preferred if the authors conducted a stratified analysis instead so they can look at these groups separately. The type of violence experienced by the 3 groups are different and most likely perpetrated by different members of the HH. Possibly the factors associated with violence experienced by the 3 groups are also different.

: Thank you so much for this valuable comment. We totally agree with you and then making different procedure for the analysis (three different group, three stratifies).

: The results show very interestingly in table 4 (violence to women), table 5 (violence to children), and table 6 (violence to the elderly)

ii) Another concern is the study methodology. The authors have not clearly outlined how participants from households were recruited. Did they include everyone in the household (children, mother & father and the elderly)? If they did, did they account for clustering by household in their analysis? If they did not, then they need to clearly explain in their methodology how HH members were selected.

: All eligible populations (Children aged 5-15 years, women aged 16-59 years, and elderly 60 years and above) who live in 20 selected Lahu villages were invited to the study.

iii) The authors have defined children as <=15yrs. They need to clearly indicate the minimum age for this group in the methods section (as part of inclusion criteria definition) and how data was collected from them (eg was it interviewer administered?). I understand in Table 1 they have outlined that the minimum age in the sample was 5yrs. But it is not clear whether 5yrs was a cut off or just minimum age for those interviewed. I am also rather concerned about the integrity of information on violence corrected from children (10yrs or less)

: We intended to collect data from children 5-15 years, which come from our 13 years experience in doing research among the hill tribe. Below 5 years is very difficult to obtain the validate information.

iv) The authors also indicated in the methods section that the target population of interest were children (<=15yrs), women(16-59yrs) and the elderly (>=60yrs). However, their data shows that they also have men (n=382) aged between 16-59yrs in the sample. I do not understand the rationale for including men (16-59yrs) in the study.

: We so sorry for the mistake of process of case selection during the analysis. We have mad revised in whole process of data analysis.

v) The general norm for violence studies is to analyze data for male and females separately as they tend to have different risk factors. I would prefer that they present their analysis that way.

: Thank you very much for the comment. We have now making three subcategories for the analysis; children aged 5-15 years, women 16-59 years, and the elderly aged 60 years and over. The whole characteristics and other information have been separated.

Other general comments are:

• They can make their tables more succinct by reporting one category for all binary variables (‘yes”) rather both ‘yes’ and ‘no’

: Thank you for the comment, we have tried to make the best option, and present in Yes and No to confirm the readers on the validation of the presentation. We very hope you understand us.

• The manuscript can do with some proof reading and editing.

: Thank you so much, it was re-edited by the American Journal Experts.

• The author should consider having prevalence estimates for each type of violence (eg physical violence).

: Yes, the data are presented in the results section.

• With the study having 3 groups, it is not clear if participants were reporting their own behavior ( eg alcohol use) and wondering what could be the impact of these on the estimates.

: We have classified all items or factors according to three groups, in table 2. Very hope it helps.

• Sample size formula (if it is necessary to show it) should be explained in the context of the study, e.g what is p in this study?

: Thank you, we have carefully explain detail, please see page 4-5.

• They could have a separate table for the multivariate analysis.

: Yes, we did. There are now separated into different table, table 4 (children), 5 (women), and 6(elderly)

Reviewer #2: Background

1.Please clarify the scope of this study because there is different definition between domestic violence and intimate partner violence so the reader might more clear and understand (Domestic violence is violence in the family, the perpetrator is family member, whereas intimate partner violence is violence between partner)

: Yes, we intended to understand the domestic violence which is defined as a pattern of behavior in any relationship that is used go gain or maintain power and control over a family members. This definition is followed the UN definition of domestic violence. Page 6.

2. The authors should mention research gaps and why have to conduct this study in Lahu tribe.

: We did, page 4, lines 7-10.

Method

1. How the researchers randomly selected the village?

: We did selection the village by which were selected by a computer generate random number method, page 4, lines 17-18.

2. The sample size calculation is n=826 but the researcher collected the data from n=1,028. Please calrify and give the explanation

: We so sorry, some procedure in sample size calculation was incorrect. We have made correction, page 4 line 27-31, and page 5, line 1-5.

3. What is the reliability (pre-test) value from this study?

: The overall Cronbach alpha was 0.71, page 6, line 4.

Result

1. Heading of Table 3 show n= 647 but in detail in the table showed n= 1,028. However, some item "n" is lower than 1,028. Please check the consistency of the data

: So sorry for the mistake, all data in all table have been double checked and revised.

2. Recoomendation pls consider to give recommendation to 3 level; individual, community, and policy level

: We have added in conclusion section, page 28, lines 12-25.

Reviewer #3: General comments: Authors have made a good attempt to analyse the extent of domestic violence and factors associated with domestic among children, women and the elderly in a poor community in rural Thailand. This paper will contribute to understanding of domestic violence in low and middle income countries, and provide data for understanding subgroups. However, the authors have introduced interest in studying subgroups but have not done so in the analyses, and this presents a challenge in the analysis and discussion of the generalize results as they often seek to make comparisons or justifications of their results against subgroups elsewhere without actual analyses of their own to back up the comparisons. My recommendations are that the authors must revise their analyses and present the overall study population analysis and present children: girls and boys under 15 years, women and men (by whichever is legal age limit for marriage perhaps), elderly women and men over 60 years. The lack of data on perpetrators is a major limitation but inferences may be safely drawn with supporting study as to who the common perpetrators of DV usually are in such contexts in Thailand. The subgroup analyses will help authors make a case for the understanding of how substance abuse increases risk of violence for specific subgroups to some extent even without understanding the perpetrators data.

: Thank you so much, we have completely done on classifying to analysis the factors associated with three different groups; children (table 4), women (table 5), and elderly (table 6)

Abstract:

Revise according to the recommendations from the content of the manuscript

: Thank you, it’s revised and improved.

Background:

“Lead to”: Authors may not reach conclusion that certain variable ‘lead to’ certain risks with cross-sectional data including their own study as such studies only allows for determination of risk factors. We do speak of associations unless experimental studies were done.

: Thank you, we have changed as comment.

“Substance use” versus abuse: authors have described substance ‘use’ and not abuse this makes it difficult to determine risk outright without measuring the extent of misuse.

: Thank you, it’s changed properly in whole text.

The background starts well with the global picture but needs to locate the domestic violence problem and risk factors in the context of Thailand, and do so indicating the extent to which other studies have addressed this issue.

: Thank you,

: We have revised and mentioned many risk factors related to domestic violence in Thailand and its references, page 3, line 20-29

- Go into some detail about the Thai social life, income, earnings, religion, culture and describing who Thai people are, what other tribes or social groupings exist in Thailand and the extent of violence there as well as any risk factors or protective factors against DV.

: We have greatly investigated on the information between Thailand and the hill tribe, unfortunately there is little information found, and no information presented in protective factor. We would say that no-study had been done previously about domestic violence in the hill tribe. So sorry.

- Also contrast with other neighbouring countries where possible, at least in terms of the prevalence.

: We have explored the information, unfortunately there is no scientific information regard to domestic violence among tribe people in Myanmar or Laos as well.

: Information is presented to non-scientific writer which is mostly not relevant to domestic violence.

- Any cross-border issues that make the hill tribes different from other Thai society?

: The different of the hill tribe and Thai people is not from cross-border but rather from the difference of their culture and other life styles.

- Describe the context of substance use and substance abuse in Thailand. How are amphetamines accessed? Natural or other synthetic stimulants? History of substance abuse. What kind of alcohol is used or abused? How is it sourced in communities which have been described as poor? How do they access amphetamines?

: We have added in reference 16-17. Even Thailand has been reported on having amphetamine use, but surprising there is very few scientific papers available.

- I also propose that some effort into explaining violence against children younger than 15 years be made, and explaining how such studies were done in Thailand or neighbouring countries. What are the risk factors?

: We have added some information in the discussion part. However, there is very few paper published.

Page 3:

Paragraph 1:

- the mention of the SDGs: This is a goal and not a fact, so the authors need to talk about this SDG in the correct context. What are authors trying to say about this goal?

- at the end of paragraph: What do the authors mean ‘domestically’?

: Thank you very much for the great comment.

Paragraph 2:

- at the end of the paragraph, authors mention contributors to violence: Do authors mean ‘perpetrators’? how do they contribute? If as the perpetrators then authors should state that explicitly. And mention other contributors if they exist. In some context, we have instigators and some Asian studies refer to instigators.

: We are trying to say that under the context of Thai society which men dominate in family, including the referred paper said that the men was defined as the domestic makers.

: We sorry if the response is not going to what your questions. If you could make more easier would be great appreciated.

Last paragraph describing hill tribes needs to go into the Setting/study site information in the methods section and explain the following: What is the population size relative to the largest and smallest tribe? Describe the lifestyle that distinguishes Lahu people from others?

: Thank you for the comment. We studied in the Lahu because one of our small previous study presented that their have a large proportion who used alcohol. Moreover, some information were found on the domestic violence among the Lahu people. This become our current study. Unfortunately, we do not have information about very some sensitive behaviors such as substances use including domestic violence in other tribes. Then, we decide not to mention the information of other tribes. We so sorry, we will do more study before could be able to come up the scientific information these points.

Last paragraph of the Background section: Provide a justification for the paper explaining the conceptual framework used for this study as mentioned in the methods section. Thus explaining why a paper on prevalence and associated factors. Why was the study conducted? And why its important for the paper to be published?

: Thank you so much for such great comments here.

: We have modified, improved, and revised as your comments. Page 4, lines 7-10.

Methods:

Paragraph 1 - Explain size of the Lahu hill area. Sources of living such as access to water, income, who is working or not, what is the family structure?

: Thank you, it’s revised in whole paragraph. Page 4-5.

Paragraph 2 - Study sample recruitment needs to be explained. The study aim suggests data was collected for children, women and older persons. It is not clear though what the sampling unit was for the study. Was it families or just individuals in the community?

: Thank you, it’s revised and explained in page 6, line 14-28.

Paragraph 4 regarding the questionnaire:

- Which questions were inappropriate for the Lahu people due to culture and beliefs?

: Do you kick your wife? In this question, we can ask directly. Many questions, they did not answer if they did not trust interviewer.

: Do you think you have a problem with your husband in amphetamine use?

- Last sentence about pilot is incomplete. Explaining the pilot is good. So finish the sentence please.

: Thank you and sorry for the mistake. It’s improved.

Paragraph 5 about the questionnaire:

- Authors need to present a table of each set of questions measured, describing what the variable is made up of, and explaining the source of the variables especially if used in other studies or new including those verified and changed with inputs from the Lahu people.

: We used our own questionnaire developed. The detail is explained in page 5 and 6

: The detail of questionnaire both Thai and English have been attached.

- Please name that dependent variable and clarify what it comprised of it is a combination of variables. What is the outcome variable made up of?

: DV is domestic violence which is very special forms occurred in Lahu family, see page 6, lines 4-12.

: We have known the forms from our pilot step. All forms have been developed to questions which present the result in Table 3.

- Domestic violence questions need to be explained in more detail as an outcome variable. The explanation given here is different compared to what is presented in the results. Clearly explain all variables here first, what they entail including any variation that may be presented in the results. Also provide the time frame that is reported for each form of DV, lifetime experience or in the past 12 months?

: It’s improved and explained in page 6, line 5-12.

Paragraph 6:

- Informed consent forms are meant to remain with the authors particularly to demonstrate that consent was sought and obtained from participants. Informed consent is also a process, it is not clear how this was done. Rather explain how information about the study was provided to participants, what was were these essential aspects that were explained, and how consent was given by particpants.

: Thank you, it’s explained in page 6, lines 11-16.

- Sentence: “People who were not able to provide essential information due to personal health problems were excluded from the study” does not explain the eligibility criteria for sample to participate in the study. Please explain the criteria used, was it one factor or multiple factors that were considered?

: Thank you, it’s explained. Page 4, lines 24-25.

- Issue of language used by participants: Move this to the questionnaire section. Was this a paper-based questionnaire? Explain how data was collected in more precisely, like language, who collected the data, how they were trained, their level of qualification. What the role of the village health volunteers including what they usually do.

: Thank you, we were helped from the village health volunteers in filling general information. However, those who could not read Thai, the forms were filled by psychiatric nurse with the help of community heath volunteers,

- How was a confidential room determined? Maybe explain this in a different way? Merge the note about ethical clearance at the end of this paragraph.

Data analysis:

: A confident room is provided by village headmen at the community hall. Page 6, line 22-23.

Logistic regression modelling is a process, explain that process. What was the outcome variable, candidate variable, how they were determined first, were bivariate analyses performed first, how were candidate variables selected and fitted into the model, any process of elimination engaged? This is crucial information to determine what was done and whether it can be replicated and reach the same conclusions the authors did.

: The DV is the having domestic violence (Y/N). Then, it forces us to us the logistic regression for the analyses. The stepe of the analysis, we used “Enter” mode which is allowed researcher to consider a IV put into the model because we have to concern on statistic significant and public health significant.

: Detail have been explained in page 6, line 30-31, and page 7 lines, 1-4.

Results section:

- Table 1 and Table 2 and Table 3 are difficult to decipher because the study is looking at 3 sets of participants: children under 15, women, older persons older than 60.

: Thank you, we have revised in analyses and interpretations in whole set of data. Please see table 1-6 which are classified into three main groups.

- Authors need to demonstrate the same characteristics for the study population alongside these characteristics for the study population sets. Including data from men and boys and classifying elders by difference between women, men, girls, boys, older women, older men. The socio-demographics do not make sense without a clear understanding of what the univariate data looks like for all sets of individuals. This is important as data also needs to account for any child marriage which is possible under 15 years, number of family members living in the same household, etc. Vulnerability to violence is also determined by gender. Cross tabulation of the differences by gender is also key in this analysis and table.

: We agree with you, and did reanalysis in the whole set of data, please see table 1, page 7.

- In Table 2 there are other drugs that have been measured which are not explained in the methods section. This justifies why a table including types of substances is needed.

: From our pilot stage, most of substances use among the Lahu had been clarify, then we intended to measure only this 6 items.

- Table 3: Authors can provide an overall estimation of DV for the population of interest, but still need to extent the analysis to demonstrate the differences between children, women and elderly persons, so that there is clarity on the extent of vulnerability in the population including their subgroups, this is most useful to provide guidance to which groups health interventions should prioritise and address and what actual interventions could be considered for which subgroups or to address which factors associated with vulnerability of which subgroup. Lumping all children, women and elderly persons is not justified as an end in itself, but would be more valuable if authors are able to single out the most vulnerable, even at a comparative level. The analysis is insufficient at this level and needs to demonstrate the complexities that are entailed in domestic violence as a multi-layered construct that can be differentiated for women into intimate partner violence or domestic violence by family members other than husband if women are married, or child abuse by parents or other family members, or even strangers if the latter was measured, etc.

: Defiantly, we have revised the analysis in table 3, please see pages 11-12.

- In Table 3, for each type of domestic violence need to provide a composite estimate as done with domestic violence. This will explain the extent of physical or emotional abuse overall.

: Thank you, it’s revised and improved.

- Authors need to explain the risk of smoking and how it links to perpetration of violence, as this variable doesn’t make sense in terms of how it increases vulnerability to DV. Do this in the background section.

: We have modified the discussion in this part and added some information. Page 26, lines 1-12.

- Table 4: The analysis appears to have skipped a stage as not all variables need to go into the multivariate analysis, so authors need to demonstrate candidate variables through bivariate analysis and look at some degree of association of each variable with the outcome variable. So bivariate analysis is required. Authors much explain Why are some analyses adjusted for and others are not?

: Sorry, please forget the old style of the analysis and presentations. We did re-analysis in classifying into three groups, please kindly see table 4-6 in new results.

Discussion:

- The authors are evidently struggling to explain the study prevalence in it entirety without a clear indication per children, women, elderly women or men subsets analysed. The data does not provide data to indicate the extent of violence against children. The story of substance use in general is also not telling a clear story without understanding the extent to which increased risk and perhaps authors should also report on who is using, and whether combinations of substances are used and determine risk from that angle. However, the conclusions reached in this paper cannot be reached with the currently presented analyses. Authors need to address the analytical questions about extent of vulnerability to violence per subgroup, what increases their vulnerability in most precise terms than the general picture painted already which does not in itself make much sense. Then tell the role substances play in this story.

: Thank you, you are right. We have made revision in whole paragraph after getting information from the analysis in three groups.

- Paragraph 2: These comparisons are incorrect without a clear understanding of the prevalence of DV among women from your analysis, so this is incorrect comparison. The same point applies to comparisons on data about children or the elderly. The analyses failed to demonstrate prevalence for each subgroup so there cannot be comparisons with subgroups.

: Yes, we agree with you. We have revised in this whole paragraph after getting the three groups analyses.

- Paragraph 3: This is suggesting that Buddist society is more gender equitable, where is the data to support this? The argument does not make sense as patriarchy is a common factor across all societies. Authors need to provide a clear explanation of the religious differences from the background and link their interpretation to that. It may be that there is a vast difference based on religions but the current argument to explain these results is incorrect without studies to support it. Authors need to explain what they mean by ‘complicated relationships’. How is Buddhism organized and are relationships there complicated too or not?

: Unfortunately, after carefully re-analysis in three different groups, religion was not found the association with domestic violent. Then, it’s deleted from the sdicussion.

Conclusions:

The conclusions need to be strengthened and be drawing out what the results have demonstrated about the study sample as sub-groups and as a whole, and indicating more specifically what recommendations suite which aspect of the results.

: Thank you, It’s revised and improved.

Abbreviations:

Authors must integrate in the texts and ensure each is explained in full at first mention and then abbreviated afterwards.

: Thank you, we have checked all relevant text.

Ethical considerations

- must be integrated with the details on how consent was obtained and data collected. Add a little bit on what DV specific considerations were made? What guidelines were followed for conducting these interviews on such a sensitive subject? Explain data storage.

: Thank you for the comment, page 6, lines 4-12.

: the ethical has been moved to methods section. The content is also extended.

References:

Many of the references do not follow the journal’s preferred referencing style. Visit: https://journals.plos.org/plospathogens/s/submission-guidelines#loc-references for information on Vancouver style. Here is some simpler information about what you can do and examples provided.

PLOS uses the reference style outlined by the International Committee of Medical Journal Editors (ICMJE), also referred to as the “Vancouver” style. Example formats are listed below. Additional examples are in the ICMJE sample references. A reference management tool, EndNote, offers a current style file that can assist you with the formatting of your references. If you have problems with any reference management program, please contact the source company's technical support.

: Thank you so much, all references have been re-checked and re-formatting.

Attachment

Submitted filename: Response go reviewers.docx

Click here for additional data file. (29.2KB, docx)

Decision Letter 1

Siyan Yi

22 Jan 2021

PONE-D-20-27497R1

Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study

PLOS ONE

Dear Dr. Apidechkul,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Mar 08 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Siyan Yi, MD, MHSc, PhD

Academic Editor

PLOS ONE

Additional Editor Comments:

Editor’s comments

We thank the authors for addressing outstanding comments from the reviewers. The revised manuscript has been improved. However, the English writing quality remains well below an acceptable level, particularly. I have read the entire manuscript; however, my comments are not exhaustive. The manuscript requires substantial support from an experienced English writer with sufficient understanding of the research context.

In your response to reviewers, please provide the details on how you addressed the comments (what changes were made, where in the text are revised), not just saying those particular points were revised and improved (but what were revised or improved and where). Where data were reanalyzed, explain briefly the new findings and ensure that revisions have been made to reflect the new findings (what were added or removed and where). To most of the comments, particularly those from Reviewer #3, the responses are not sufficient without clear explanations. In your revised manuscript preparation, please have a closer look at the Instructions for Authors and follow them strictly, including word count limit, styles, spacing adjustments, etc.

Abstract

  1. The abstract must be condensed as it contains far more than the recommended 350 word count limit.

  2. Line 26, page 1: Please remove ‘sexual problems’ as it is outside of the study’s scope and is not applicable to all target populations under the study (e.g., children).

  3. Please use consistent terminologies; e.g., ‘the elderly’ in place of ‘elderly,’ ‘elderly adults,’ ‘elderly individuals,’ ‘older people,’ ‘elderly persons,’ etc. throughout the text.

  4. Line 26, page 1: ‘types of’ is unnecessary.

  5. Line 30, page 1: The word ‘adults’ is unnecessary.

  6. Line 1, page 2: A cross-sectional study was conducted among participants who belonged to the Lahu hill tribe and lived in 20 selected villages in Chiang Rai Province, Thailand.

  7. Lines 2-5: A validated questionnaire was used to collect personal information and experiences related to domestic violence in the past year from children aged 5-15 years, women (aged 16-59 years), and elderly aged ³60 years.

  8. Line 5, page 2: Please specify whether the ‘logistic regression analyses’ were bivariate, multiple, or both.

  9. Line 6, page 2: Please remove ‘at the significance level µ=0.05.’

  10. Results section must be shortened, removing most of the descriptive socio-demographic results.

  11. The word ‘analysis’ is a countable noun, and its plural form should be ‘analyses.’

  12. Line 10, page 2: The use of ‘while’ is incorrect.

  13. Line 12, page 2: …children who smoked…

  14. Line 15, page 2: …a family head…

  15. Lines 14-20, page 2: The large sentence must be restructured as it is difficult to read. It may be broken down into small sentences. Also, when you present the associations between variables, please make the significance level clear, not just say ‘more likely.’

Introduction

  1. Line 7, page 3: In scientific writing, numbers smaller than 10 should be spelt out.

  2. Line 19, page 3: …the elderly living in developing countries…

  3. Lines 21-23, page 3: Please revise the sentence to minimize wording redundancies (children, women, the elderly) and improve the referencing. The references 11, 12, and 13 should be moved to join 14 and 15 at the end of the sentence.

  4. Line 24, page 3: Please remove ‘who were’ as it is unnecessary.

  5. Lines 26-27, page 3: “Given people’s economic constraints and substance use, alcohol use has become an integrated factor contributing to domestic violence [18].” This sentence is difficult to understand – how is the first part linked to the latter?

  6. Line 8, page 4: Add ‘,’ before ‘including.’

Methods

  1. Lines 16-17, page 4: The first and second paragraph should be combined as the first one does not make much sense. Consider: “A cross-sectional study was conducted in 20 villages randomly selected out of 216 Lahu villages in Chiang Rai Province, Thailand [21], using a random number generation method.”

  2. Lines 18-19, page 4: Please remove “However, in the study process, only 20 Lahu villages were randomly selected from the list” as does not add any meaning to the text and sounds redundant.

  3. Lines 19-20, page 4: The sentence, “All the Lahu people who were living in the selected 20 villages were invited to participate in the study” is not true as you included only children aged 5-15, women, and the elderly.

  4. Line 22, page 4: Again, the same sentence was repeated, which should also be removed.

  5. Lines 24-25, page 4: The sentence “People who were able to provide essential information regarding the study protocol, living selected villages at the date of data collection were eligible for the study” is not readable. Why should the target populations be able to provide essential information regarding the study protocol? Living in the selected villages? The inclusion and exclusion criteria must be clearly stated.

  6. Line 27, page 4: The study sample size…

  7. Line 1, page 5: “Therefore, the minimum required sample size for the study was 196 participants for each sub-population (children aged 5-15, women aged 16-59, and the elderly aged ³60).” Then you can remove the whole following sentence (However, in this study, three vulnerable groups (children aged 15 years and below, women aged 16-59, and elderly aged 60 year and over) were selected for the study, then, a total of 588 participants were required.”

  8. Line 7, page 5: Did this study have a conceptual framework? If so, it must be described in the text. If not, this expression about it should be removed.

  9. Lines 13-15, page 5: The complex sentence needs to be re-structured by breaking it down into a few sentences or using correct punctuations to link the sub-sentences.

  10. Lines 17-21, page 5: The long sentence should be rewrite as it is difficult to read and with grammatical errors.

  11. Lines 25-26, page 5: Why was substance use limited to only cigarette, alcohol, and amphetamine use? Also, how is cigarette smoking related to domestic violence?

  12. Lines 27-28, page 5: Are sexual harassment and being forced to have sex considered harm related to sex, not sexual violence?

  13. Line 30, page 5: The word ‘ignorant’ may have been incorrectly used.

  14. Lines 1, page 6: Is ‘domestic abuse’ different from ‘domestic violence?’ It is the first mention of domestic abuse in the paper. Also, please say ‘women aged 16-59’ consistently.

  15. Line 4, page 6: ‘Cronbach’s alpha.’ Also, state clearly on what is the Cronbach’s alpha for? Domestic violence scale? Across the three participant groups? References to support the scales used in this study must be cited. The use of the scales (coding, recoding, cut-offs used to define the variables…), should be clearly described.

  16. Several grammatical mistakes are easily found in the questionnaire development and data collection section. Please revise them carefully. For example, the following three sentences are all incorrect: did you have experienced of kicking out of the house from family member in the past year? Did you have experienced of being forced to drink alcohol, smoke, or use substances in the past year from family member? Did you have experienced of being forced to ask for money or borrow items from others?

  17. Please make ‘statistical/data analysis’ plural (data analyses).

  18. Line 1, page 7: Please specify whether the ‘logistic regression analyses’ were bivariate, multiple, or both. The description of the variable selection for the multiple regression analysis models is hard to understand. The sentences “Some variables were controlled the effect in the model which were determined as the confounder factors for the prediction. In the final model, all significant variables and controlled variables were fitted before making interpretations” are unreadable.

  19. The paragraph on the ethics also needs improvement in writing quality. Did this study need more than one protocol? If so, why? Please correct the wording – e.g., what you need from the participants was their consent, not the form. Please also correct several grammatical errors in the paragraph.

Results

All tables and narratives of the tables need to be revised.

  1. Lines 19-20, page 7: Please clarify if the 646 was the number of participants recruited or included in the analyses. If the number of the participants recruited was different from that in the analyses, please provide the details of the number of participants excluded with reasons.

  2. Line 22-23, page 7: Please remove ‘While,’ which was incorrectly used, and change ‘elderly persons’ to ‘the elderly.’ Also, please use ‘;’ to separate full sentences, not ‘,’.

  3. Lines 21-24, page 7: You don’t need to repeat the number of participants in each group as it was clearly stated in the first sentence.

  4. Table 1:

  • Please do not use ‘.’ in titles.

  • The use of ‘;’ is also incorrect.

  • The table’s head is not understandable – Why was the head repeated? Please have a look at other published papers and revise them accordingly

  1. Line 2-5, page 9: Please correct all the sentences in this paragraph, which are unreadable and grammatically incorrect.

  2. Table 2:

  • The title is incomplete.

  • The contents of the table are not matched with the table’s head – while the head notes n (%) for Yes and No, one column presents number, the other presents %. Also, it’s unnecessary to present both ‘Yes’ and ‘No’ results. Let readers do the math.

  • While only smoking, alcohol use, and amphetamine use was mentioned in the methods, other types of substances are shown in the table. Please make them consistent. Please also include the last item, ‘Family member who used alcohol’ in the measurements in the methods.

  1. The narrative of Table 3 also needs improvement, correcting the misuse of punctuations typos, and inconsistencies. Also, only one category should be presented in tables for binary variables.

  2. Line 1, page 14: Please use ‘bivariate logistic regression’ in place of ‘univariate analysis.’ Univariate analyses involve only one variable without a comparison or association and are used to describe participants’ characteristics (e.g., mean, median, %, etc.). Also, ‘smoking behavior’ is not a correct term in this context – you just asked whether they smoked (yes or no), not exploring the way they smoked.

  3. In description of relationships, please specify the direction and significance level of the associations, not just saying the variables are associated.

  4. Since the description of the regression models was not at all clear, the reader cannot understand why only religion and education were controlled for in Table 4. Same for Tables 5 and 6. Also, please use correct terminologies (e.g., multiple logistic regression analysis).

  5. Lines 2-5, page 14: The two sentences tell the same thing and should be combined. This is also applied to Table 5 and 6.

  6. Data in Table 4 tell clearly that the sample sizes were not calculated to represent each study population. As a result, the it was too small for children.

  7. As commented earlier, please keep only one head for all tables.

  8. The footnotes under Tables 4-6 are not understandable – why were µ values needed to tell a significance level, while p-values were already presented in the tables?

Discussion

  1. In general, the discussion section requires extensive support from an experience writer. Many statements are not understandable because of the poor quality, many others are too broad and not specifically based on the study findings. The discussions are mostly the comparisons with other studies (A supports B, A is confirmed by B), without context-specific interpretation and linkage to social policy implications.

  2. Lines 12-14, page 26: The sentence “The WHO [28] reported that 30.0% of domestic violence against women had different prevalence estimates in different communities that were higher than the estimates in the Lahu community in Thailand” is not at all clear.

  3. Line 16, page 26: Please clarify what ‘clinical populations’ means?

  4. Lines 1-12, page 27: The discussion on the relationship between smoking and domestic violence relied so much on personal assumptions that are not well supported by the findings or literature.

  5. Lines 13-18, page 27: The paragraph almost unreadable at all due to a poor structure and grammatical mistakes.

  6. Lines 1-2, page 28: What does ‘…domestic violence was dominated by alcohol use…’ mean? Alcohol use may be a cause of domestic violence, but how it dominates the violence is not understandable.

  7. Limitations are not well presented.

  • The sentence does not make sense – the study’s limitations are not determined by the refusal rate.

  • It’s not clear how and to what extend data collection by community health volunteers affected the data quality? What measures were taken to address the concerns? Did the volunteers receive any training? Information on data collection training should be clearly provided in the Methods.

  • Same for the following sentence. Please discuss clearly what were the methodological issues around the collection of sensitive data?

  1. Conclusion also need to be entirely re-written. Please avoid repeating the results or include further discussion. Instead, please summarize key findings and provide relevant recommendations that are well supported by the findings.

References

The reference list does not meet the journal’s requirements. Please rework on it more carefully.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: Accept. The authors have adequately addressed the comments raised in a previous round of review. This manuscript is now acceptable for publication.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Mar 15;16(3):e0248587. doi: 10.1371/journal.pone.0248587.r004

Author response to Decision Letter 1

19 Feb 2021

Response to Editor’s comments

Dear Editor,

I do not know how to thank you for the help along the paper. I have learnt a lot from your comments in each item. Your advices comments make us improvement all skills and practices in how to write a good academic paper.

After improvement, the paper has been once again checked by the American Journal Experts.

I do very hope that you happy in this version.

Thank you once again,

TK

Additional Editor Comments:

Editor’s comments

We thank the authors for addressing outstanding comments from the reviewers. The revised manuscript has been improved. However, the English writing quality remains well below an acceptable level, particularly. I have read the entire manuscript; however, my comments are not exhaustive. The manuscript requires substantial support from an experienced English writer with sufficient understanding of the research context.

: Thank you very much for your comments. The manuscript has been re-edited by American Journal Experts, with reference. no 3987-18F0-6D5E-AB8E-72FF .

In your response to reviewers, please provide the details on how you addressed the comments (what changes were made, where in the text are revised), not just saying those particular points were revised and improved (but what were revised or improved and where). Where data were reanalyzed, explain briefly the new findings and ensure that revisions have been made to reflect the new findings (what were added or removed and where). To most of the comments, particularly those from Reviewer #3, the responses are not sufficient without clear explanations. In your revised manuscript preparation, please have a closer look at the Instructions for Authors and follow them strictly, including word count limit, styles, spacing adjustments, etc.

Abstract

1. The abstract must be condensed as it contains far more than the recommended 350 word count limit.

: The abstract has been revised to 328 words.

2. Line 26, page 1: Please remove ‘sexual problems’ as it is outside of the study’s scope and is not applicable to all target populations under the study (e.g., children).

: Thank you. This term has been deleted.

3. Please use consistent terminologies; e.g., ‘the elderly’ in place of ‘elderly,’ ‘elderly adults,’ ‘elderly individuals,’ ‘older people,’ ‘elderly persons,’ etc. throughout the text.

: Thank you. This terminology has been replaced in the entire text.

4. Line 26, page 1: ‘types of’ is unnecessary.

: Thank you. This phrasing has been improved.

5. Line 30, page 1: The word ‘adults’ is unnecessary.

: Thank you. This term has been removed.

6. Line 1, page 2: A cross-sectional study was conducted among participants who belonged to the Lahu hill tribe and lived in 20 selected villages in Chiang Rai Province, Thailand.

: Thank you. This has been changed.

7. Lines 2-5: A validated questionnaire was used to collect personal information and experiences related to domestic violence in the past year from children aged 5-15 years, women (aged 16-59 years), and elderly aged 60 years and over.

: Thank you. This has been changed.

8. Line 5, page 2: Please specify whether the ‘logistic regression analyses’ were bivariate, multiple, or both.

: A binary logistic regression was used. The wording has been added.

9. Line 6, page 2: Please remove ‘at the significance level µ=0.05.’

: Thank you. This has been deleted.

10. Results section must be shortened, removing most of the descriptive socio-demographic results.

: Thank you. This has been deleted. Moreover, the abstract has been modified to meet the 350-word limit.

11. The word ‘analysis’ is a countable noun, and its plural form should be ‘analyses.’

: Thank you. However, we changed the word to “in the study”.

12. Line 10, page 2: The use of ‘while’ is incorrect.

: Thank you. This wording has been improved.

13. Line 12, page 2: …children who smoked…

: Thank you. This has been deleted.

14. Line 15, page 2: …a family head…

: Thank you. This has been changed.

15. Lines 14-20, page 2: The large sentence must be restructured as it is difficult to read. It may be broken down into small sentences. Also, when you present the associations between variables, please make the significance level clear, not just say ‘more likely.’

: This has been revised.

Introduction

16. Line 7, page 3: In scientific writing, numbers smaller than 10 should be spelt out.

: Thank you. This has been corrected.

17. Line 19, page 3: …the elderly living in developing countries…

: Thank you. This has been improved.

18. Lines 21-23, page 3: Please revise the sentence to minimize wording redundancies (children, women, the elderly) and improve the referencing. The references 11, 12, and 13 should be moved to join 14 and 15 at the end of the sentence.

: Thank you. This has been improved.

19. Line 24, page 3: Please remove ‘who were’ as it is unnecessary.

: Thank you. This has been improved.

20. Lines 26-27, page 3: “Given people’s economic constraints and substance use, alcohol use has become an integrated factor contributing to domestic violence [18].” This sentence is difficult to understand – how is the first part linked to the latter?

: Thank you. This has been revised and improved.

21. Line 8, page 4: Add ‘,’ before ‘including.’

: Thank you. This has been improved.

Methods

22. Lines 16-17, page 4: The first and second paragraph should be combined as the first one does not make much sense. Consider: “A cross-sectional study was conducted in 20 villages randomly selected out of 216 Lahu villages in Chiang Rai Province, Thailand [21], using a random number generation method.”

: Thank you very much for the help. The paragraphs have been accordingly revised and improved.

23. Lines 18-19, page 4: Please remove “However, in the study process, only 20 Lahu villages were randomly selected from the list” as does not add any meaning to the text and sounds redundant.

: Thank you. This has been removed.

24. Lines 19-20, page 4: The sentence, “All the Lahu people who were living in the selected 20 villages were invited to participate in the study” is not true as you included only children aged 5-15, women, and the elderly.

: You are correct, and this sentence should not appear in this section. We have modified it and moved it to the next section.

25. Line 22, page 4: Again, the same sentence was repeated, which should also be removed.

: Thank you for the comment. As we removed the sentence in response to comment no. 24, we need to maintain the sentence in this section.

26. Lines 24-25, page 4: The sentence “People who were able to provide essential information regarding the study protocol, living selected villages at the date of data collection were eligible for the study” is not readable. Why should the target populations be able to provide essential information regarding the study protocol? Living in the selected villages? The inclusion and exclusion criteria must be clearly stated.

: Thank you. The sentences have been revised.

27. Line 27, page 4: The study sample size…

: Thank you. This has been improved.

28. Line 1, page 5: “Therefore, the minimum required sample size for the study was 196 participants for each sub-population (children aged 5-15, women aged 16-59, and the elderly aged ³60).” Then you can remove the whole following sentence (However, in this study, three vulnerable groups (children aged 15 years and below, women aged 16-59, and elderly aged 60 year and over) were selected for the study, then, a total of 588 participants were required.”

: Thank you. This has been revised and improved accordingly.

29. Line 7, page 5: Did this study have a conceptual framework? If so, it must be described in the text. If not, this expression about it should be removed.

: Thank you very much for the great comment. This has been removed.

30. Lines 13-15, page 5: The complex sentence needs to be re-structured by breaking it down into a few sentences or using correct punctuations to link the sub-sentences.

: Thank you very much. The sentence has been revised and improved.

31. Lines 17-21, page 5: The long sentence should be rewrite as it is difficult to read and with grammatical errors.

: Thank you very much. This has been improved and shortened.

32. Lines 25-26, page 5: Why was substance use limited to only cigarette, alcohol, and amphetamine use? Also, how is cigarette smoking related to domestic violence?

: This is because in our literature review, it was found that substances are frequently used by Lahu people. Cigarette smoking is related to amphetamine use, and it was thus included in the study.

33. Lines 27-28, page 5: Are sexual harassment and being forced to have sex considered harm related to sex, not sexual violence?

: Thank you for the comment. We have provided considerable discussion to describe these harms. Basically, sexual violence is a sexual act or attempt to engage in a sexual act by violence or coercion and includes acts to traffic a person or acts against a person directly. This is a clear definition. However, we need to cover only sexual harassment and being forced to have sex, which were specifically found in the Lahu women. Therefore, we decided to use this term. We hope very much that you understand us.

34. Line 30, page 5: The word ‘ignorant’ may have been incorrectly used.

: I am very sorry for the mistake. Actually, the English in this article was improved by American Journal Experts (AJE) before submission. However, I agree with you that there remain many errors. Before resubmission, the article has been re-edited by AJE.

35. Lines 1, page 6: Is ‘domestic abuse’ different from ‘domestic violence?’ It is the first mention of domestic abuse in the paper. Also, please say ‘women aged 16-59’ consistently.

: I am sorry for the confusion. During the literature review, I discovered that scholars use the two terms, domestic abuse and domestic violence, interchangeably. At the time, I wanted to use all words to cover every use and put both words in the paper. However, to be consistent, “domestic abuse” has been deleted.

36. Line 4, page 6: ‘Cronbach’s alpha.’ Also, state clearly on what is the Cronbach’s alpha for? Domestic violence scale? Across the three participant groups? References to support the scales used in this study must be cited. The use of the scales (coding, recoding, cut-offs used to define the variables…), should be clearly described.

: The Cronbach alpha is used to detect the reliability of questions related to domestic among children, women, and the elderly. I have double checked and out on the numbers and reference in.

37. Several grammatical mistakes are easily found in the questionnaire development and data collection section. Please revise them carefully. For example, the following three sentences are all incorrect: did you have experienced of kicking out of the house from family member in the past year? Did you have experienced of being forced to drink alcohol, smoke, or use substances in the past year from family member? Did you have experienced of being forced to ask for money or borrow items from others?

: Thank you for the comment. I requested that AJE re-edit the manuscript. I hope that the revised version is acceptable.

38. Please make ‘statistical/data analysis’ plural (data analyses).

: Thank you. This has been improved.

39. Line 1, page 7: Please specify whether the ‘logistic regression analyses’ were bivariate, multiple, or both. The description of the variable selection for the multiple regression analysis models is hard to understand. The sentences “Some variables were controlled the effect in the model which were determined as the confounder factors for the prediction. In the final model, all significant variables and controlled variables were fitted before making interpretations” are unreadable.

: We used a binary logistic regression with both univariate and multivariate steps.

: The mentioned sentence has been revised.

40. The paragraph on the ethics also needs improvement in writing quality. Did this study need more than one protocol? If so, why? Please correct the wording – e.g., what you need from the participants was their consent, not the form. Please also correct several grammatical errors in the paragraph.

: Thank you very much. I am sorry for the many grammatical errors (even though this version was checked by AJE). The language was improved and revised again by AJE, no. 3987-18F0-6D5E-AB8E-72FF .

Results

All tables and narratives of the tables need to be revised.

41. Lines 19-20, page 7: Please clarify if the 646 was the number of participants recruited or included in the analyses. If the number of the participants recruited was different from that in the analyses, please provide the details of the number of participants excluded with reasons.

: The 646 participants were included in the analyses.

: No participants were excluded from the analyses.

42. Line 22-23, page 7: Please remove ‘While,’ which was incorrectly used, and change ‘elderly persons’ to ‘the elderly.’ Also, please use ‘;’ to separate full sentences, not ‘,’.

: Thank you very much. This has been improved.

43. Lines 21-24, page 7: You don’t need to repeat the number of participants in each group as it was clearly stated in the first sentence.

: Thank you. I agree with you, and this has been improved.

44. Table 1:

• Please do not use ‘.’ in titles.

: This has been removed.

• The use of ‘;’ is also incorrect.

: This has been improved.

• The table’s head is not understandable – Why was the head repeated? Please have a look at other published papers and revise them accordingly

: Thank you for the great comment. I have learned a lot from this comment.

: The table headings have been revised.

45. Line 2-5, page 9: Please correct all the sentences in this paragraph, which are unreadable and grammatically incorrect.

: All sentences were improved.

46. Table 2:

• The title is incomplete.

: This has been revised and improved.

• The contents of the table are not matched with the table’s head – while the head notes n (%) for Yes and No, one column presents number, the other presents %. Also, it’s unnecessary to present both ‘Yes’ and ‘No’ results. Let readers do the math.

: Thank you very much for the comment. You are correct. This is an important mistake and has been corrected. The column is the n and % of each group.

• While only smoking, alcohol use, and amphetamine use was mentioned in the methods, other types of substances are shown in the table. Please make them consistent. Please also include the last item, ‘Family member who used alcohol’ in the measurements in the methods.

: Thank you for the comment. All types of substance use are mentioned in questionnaire.

: ‘Family member who uses alcohol’ is now mentioned in the Method section.

47. The narrative of Table 3 also needs improvement, correcting the misuse of punctuations typos, and inconsistencies. Also, only one category should be presented in tables for binary variables.

: All the sentences have been revised.

: I am sorry if there is confusion in the presentation. Basically, we want to present the total number of participants and the number that experienced domestic violence. Domestic violence is divided into physical and mental violence. Sexual abuse is presented only for the women.

: We have also resummarized the information.

48. Line 1, page 14: Please use ‘bivariate logistic regression’ in place of ‘univariate analysis.’ Univariate analyses involve only one variable without a comparison or association and are used to describe participants’ characteristics (e.g., mean, median, %, etc.). Also, ‘smoking behavior’ is not a correct term in this context – you just asked whether they smoked (yes or no), not exploring the way they smoked.

: This section was revised again by AJE.

49. In description of relationships, please specify the direction and significance level of the associations, not just saying the variables are associated.

: Thank you. This has been improved.

50. Since the description of the regression models was not at all clear, the reader cannot understand why only religion and education were controlled for in Table 4. Same for Tables 5 and 6. Also, please use correct terminologies (e.g., multiple logistic regression analysis).

: Thank you. This has been revised and improved.

: After reviewing this point, I completely understand what you mean. The sentences have been revised and improved in all descriptions in Tables 4-6.

51. Lines 2-5, page 14: The two sentences tell the same thing and should be combined. This is also applied to Table 5 and 6.

: Thank you for the comment. All relevant sentences have now been properly combined.

52. Data in Table 4 tell clearly that the sample sizes were not calculated to represent each study population. As a result, the it was too small for children.

: You are correct. We all have learned a lot from this comment. When we work on the next project, we will carefully consider the sample size; some groups are large, while some groups are small.

: Thank you very much.

53. As commented earlier, please keep only one head for all tables.

: Thank you. All table headings have been revised.

54. The footnotes under Tables 4-6 are not understandable – why were µ values needed to tell a significance level, while p-values were already presented in the tables?

: Thank you. Based on my knowledge, the alpha value is the cut-off point, and we set the statistic to 0.05, while the p-value is obtained from the real data that can be any probability value. So, to determine the significance, we need to compare the alpha value and the p-value before deciding whether to accept or reject the hypothesis. Therefore, please consider this explanation.

Discussion

55. In general, the discussion section requires extensive support from an experience writer. Many statements are not understandable because of the poor quality, many others are too broad and not specifically based on the study findings. The discussions are mostly the comparisons with other studies (A supports B, A is confirmed by B), without context-specific interpretation and linkage to social policy implications.

: Thank you for such a wonderful comment. We have extensively expanded the paragraphs.

56. Lines 12-14, page 26: The sentence “The WHO [28] reported that 30.0% of domestic violence against women had different prevalence estimates in different communities that were higher than the estimates in the Lahu community in Thailand” is not at all clear.

: Thank you. The sentence has been revised.

57. Line 16, page 26: Please clarify what ‘clinical populations’ means?

: I am sorry for the unclear interpretation from the original cited paper, which mentioned people attended to in a hospital.

58. Lines 1-12, page 27: The discussion on the relationship between smoking and domestic violence relied so much on personal assumptions that are not well supported by the findings or literature.

: Thank you for the great comment. These references were replaced by the proper references.

59. Lines 13-18, page 27: The paragraph almost unreadable at all due to a poor structure and grammatical mistakes.

: Thank you. The paragraph has been revised.

60. Lines 1-2, page 28: What does ‘…domestic violence was dominated by alcohol use…’ mean? Alcohol use may be a cause of domestic violence, but how it dominates the violence is not understandable.

: You are correct that alcohol use contributes to domestic violence. The sentences have been revised.

61. Limitations are not well presented.

• The sentence does not make sense – the study’s limitations are not determined by the refusal rate.

: Thank you. I agree with you, and the sentences have been deleted.

• It’s not clear how and to what extend data collection by community health volunteers affected the data quality? What measures were taken to address the concerns? Did the volunteers receive any training? Information on data collection training should be clearly provided in the Methods.

: We did not use the community health volunteers to collect the information. We asked them to help in explaining some questions in case the participants did not clearly understand the meaning or the context of the question. This did not occur often because the participants who could not speak Thai were excluded from the study as part of the exclusion criteria. Therefore, we did not train the community participants to collect the data.

• Same for the following sentence. Please discuss clearly what were the methodological issues around the collection of sensitive data?

: Basically, as we know from our more than 10 years of experience in conducting research with Lahu people, domestic violence and other family conflicts are kept secret, and they do not tell these stories in public, particularly to individuals of a different gender. Therefore, all forms of violence are sensitive issues for them. Therefore, we used gender matching during the interview to improve the accuracy of the information.

: We have expanded the explanation in the paragraph.

62. Conclusion also need to be entirely re-written. Please avoid repeating the results or include further discussion. Instead, please summarize key findings and provide relevant recommendations that are well supported by the findings.

: Thank you very much. All of these paragraphs have been rewritten.

References

The reference list does not meet the journal’s requirements. Please rework on it more carefully.

: I am very sorry for the mistake and misunderstanding. I have revised all references to follow the journal guidelines from the following URL: https://journals.plos.org/plosone/s/submission-guidelines#loc-references.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: Accept. The authors have adequately addressed the comments raised in a previous round of review. This manuscript is now acceptable for publication.

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Attachment

Submitted filename: Response_to_editors_Harms_FINAL.docx

Click here for additional data file. (40.8KB, docx)

Decision Letter 2

Siyan Yi

2 Mar 2021

Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study

PONE-D-20-27497R2

Dear Dr. Apidechkul,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Siyan Yi, MD, MHSc, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Acceptance letter

Siyan Yi

4 Mar 2021

PONE-D-20-27497R2

Factors associated with domestic violence in the Lahu hill tribe of northern Thailand: A cross-sectional study

Dear Dr. Apidechkul:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Siyan Yi

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Appendix. Questionnaire used in the study (Thai version).

    (PDF)

    Click here for additional data file. (88.8KB, pdf)
    S2 Appendix. Questionnaire used in the study (English version).

    (PDF)

    Click here for additional data file. (91.5KB, pdf)
    S3 Appendix. Data file of the study.

    (SAV)

    Click here for additional data file. (493.1KB, sav)
    Attachment

    Submitted filename: Response go reviewers.docx

    Click here for additional data file. (29.2KB, docx)
    Attachment

    Submitted filename: Response_to_editors_Harms_FINAL.docx

    Click here for additional data file. (40.8KB, docx)

    Data Availability Statement

    All relevant data are within the manuscript and its Supporting Information files.

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES