Introduction

The Rohingya population of Myanmar is one of the world’s most oppressed minority groups. The Citizenship Act of 1982 removed the Rohingya from the list of officially recognized ethnic minority groups. It denied them many fundamental rights, including citizenship, freedom of movement, access to health care and education, marital registration, and the ability to vote, making them the world’s largest stateless group. In August 2017, in continuation of past persecution events, the Myanmar army began a massive clearance operation in Rakhine, which was then home to approxmately 1.2 million Rohingya individuals. The operation saw an estimated 7800 Rohingya killed. The movement drove the Rohingya out of Myanmar, leading to a massive exodus of about 750 000 Rohingya refugees to Bangladesh’s Cox’s Bazar district.

Several studies have recorded the experience of abuses faced by Rohingya refugees before displacement. Rohingya refugees were refused medical care for physical or sexual assault in Myanmar. A major component of these violent incidences was also emotional and verbal abuse. Rohingya group leaders echoed and corroborated these descriptions in different reports. These occurrences of physical, sexual, and emotional abuse were reported to have a lasting effect on the survivors’ mental health and to be associated with the potential diagnosis of psychological conditions, such as posttraumatic stress disorder (PTSD).

Posttraumatic stress is a syndrome of PTSD characterized by distracting ideas, nightmares, memories of past traumatic events, avoidance of trauma reminders, hypervigilance, and sleep disturbance. Such aspects are associated with extreme psychological, occupational, and interpersonal dysfunction. However, patients with PTSD experience pronounced cognitive, affective, and behavioral reactions to stimuli, resulting in hallucinations, extreme anxiety, and fleeing or combative behavior. These symptoms may result in emotional numbness and reduced involvement in daily activities and, in the extreme, may result in alienation from others. Among patients with PTSD, depressive disorders, anxiety disorders, and drug misuse are 2 to 4 times more prevalent than among patients without PTSD. Moreover, PTSD may increase the risk of suicide attempts. We used the Impact of Event Scale–Revised (IES-R) for the evaluation of posttraumatic stress symptoms (PTSSs). This self-report instrument was designed to include all 3 groups of symptoms of PTSD (ie, interference, avoidance, and hyperarousal) associated with a particular life-threatening incident.

During the predisplacement and postdisplacement periods, refugees faced multiple stressors. Individuals with an experience of abuse were at risk of increasing mental health problems.

The challenges facing Rohingya refugees while living in Bangladesh have been reported in many studies. They live in small, overcrowded temporary shelters in refugee camps without adequate food, clean water, or toilets. Moreover, their lives are on hold, and they are unsure about their future. A 2017 study analyzed the daily environmental stressors among 148 Rohingya adults and found worse mental health outcomes for refugees. Compared with 2 years ago, the Rohingya refugees’ basic needs and health care have mostly improved. However, postdisplacement factors are still important for improving the mental health of Rohingya refugees.

Screening is effective only when combined with high-quality services for mental well-being. One of the challenges to ensuring appropriate services for Rohingya refugees in Bangladesh is the lack of statistical data on the group’s mental health status. In this report, we plan to identify the prevalence of PTSSs and associated predisplacement and postdisplacement factors of PTSSs among Rohingya adults living in Bangladesh after the massive clearance operation.

Methods

Study Design and Participants

From September 17, 2019, to January 11, 2020, we conducted a cross-sectional survey among Rohingya refugees residing in Cox’s Bazar, Bangladesh. More than 1 million Rohingya refugees are now staying in Bangladesh. Most are clustered in Ukhia and Teknaf, the district’s 2 upazilas (administrative regions). The largest refugee settlement in the world, Kutupalong, centered in Ukhia, is home to more than 600 000 refugees alone. The prospective participants were recruited from Kutupalong via the process of a 3-stage sampling technique. First, with the assumption of equal population size in each camp, we chose 8 camps randomly from the Kutupalong refugee camp and expansion areas that consist of 23 camps. Second, we selected at least 160 households from each selected camp, with a target of 1280 households in the study. We targeted more households to be included in our study than the required sample size at 80% power, 95% CI of 0.05 to 1.96, with an assumption that 40% of the population had a mental illness, and a design effect of 2. We applied a systematic sampling technique, and the first household was randomly chosen from the approximate geographical center of the camp. Data collectors proceeded to the next closest household until 160 households were sampled. Third, we had a single respondent per household interviewed, preferably the head of the household. The female head of the household or other available adult member of the household was surveyed when the male head of the household was not available. Household members have been described as those who lived for at least 1 month under the same roof and shared cooking and eating facilities from the same source. We also ensured that the participant lived in the camp for at least 2 years after the displacement. The details of the sampling allocation are given in eTable 1 in the Supplement. Participants provided verbal consent for the study because they were reluctant to sign their names or provide their fingerprints on any piece of paper; in addition, most participants were analphabetic. They were reassured that all the information collected would be kept strictly confidential and would not be used for anything other than research purposes. However, they were provided with a consent paper with detailed contact information of the research investigators for any future query. The institutional review board at North South University, Bangladesh, approved the study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Recruitment and Training

The data were obtained and cleaned by a team of 4 enumerators consisting of 2 men and 2 women from the Health Management BD (HMBD) Foundation. The HMBD Foundation, a local nongovernmental organization (NGO) that has been working in Rohingya refugee camps since the influx, chose the local community leaders (called a “maji”) from each of the selected 8 camps. The local leader from each of the camps was then informed about the study’s research and ethics. They assisted the HMBD Foundation in recruiting 8 local male residents from each camp who could communicate in both the Bengali (Bangla) and Rohingya languages. A team of data collectors was then formed and included 2 persons (ie, one person from the camp and another person from the HMBD Foundation). The interview was held in both the Rohingya and Bangla languages: the camp data collector asked questions in the Rohingya language, and then the HMBD Foundation member checked the answer by asking the same question in Bangla. Our 2 research investigators from North South University (A.H. and T.A.K.) arranged a 1-day practical training session about ethics and data collection. The enumerators and data collectors were also briefed about the study objectives, methods, and questionnaire. The researchers also taught the data collectors about the techniques of report building and preserving neutrality as well as information on ethical problems, privacy concerns, cultural awareness, and risk management for mental health. After the training, a pilot study was arranged for the 8 study teams and was evaluated as a single unit. The aim was to observe the capacity to comprehend the relevant techniques and troublesome situations while interviewing. We made necessary corrections after the pilot study. Afterward, each trained team visited their designated camp together to collect the data using a semistructured questionnaire.

Data Collection

It was made clear to the respondents that participation in the study was entirely voluntary. The face-to-face interview took place 1 person at a time, to ensure privacy. The respondents were given no monetary or food-item incentives. The questions were read aloud to the respondents 1 question at a time during the interview, and the respondents were asked which of the scale choices was acceptable. The coinvestigators reviewed the data collection sheets for completeness, accuracy, and internal consistency, which were confirmed by the principal investigator.

Sociodemographic Variables

The first section of the questionnaire assessed sociodemographic characteristics, including age (in years), sex (male or female), height (in inches), weight (in kilograms), marital status (never married, widowed or divorced, and married), and the number of family members in a household. Later we categorized family members as 2 or less, 3 to 4, and more than 4 members. Educational level was subdivided into 2 groups: respondents who attended school and those who cannot read or write.

Predisplacement Abuse

We used the word “abuse” here, which involves the violence recognized by the intention of imposing control or dominance over Rohingya people to cause rage, harm, resentment, humiliation, coercion, and helplessness. We collected data on the types of abuse that refugees were exposed to before the forced displacement to Bangladesh. The categories were indirect abuse or not exposed to any abuse, verbal or emotional abuse, physical abuse, and both physical and sexual abuse. Verbal or emotional abuse was defined as a nonphysical act that impairs an individual’s psychological integrity and may take the form of coercion, defamation, verbal abuse, or harassment. It might also include being forced into labor and separation from or witnessing abuse against family or community members. Physical abuse was regarded as any force applied to any part of the body, such as shaking, burning or scalding, choking, hair pulling, hitting, slapping, kicking, or threatening and/or attacking with a knife, gun, or other type of weapon. It also also might have included restraining, tying, or locking up against the individual’s will. Sexual abuse was defined as sexual humiliation (forced masturbation or nudity), sexual slavery, rape (vaginal, oral, anal, or attempted), genital abuse (beatings, electric shock, or mutilation), castration, penis amputation, sterilization, or forced marriage, cohabitation, or sexual activity (with a stranger, family member, or corpse). However, to differentiate between physical and sexual abuse, the respondents were intentionally asked with caution whether the specific event included the sexual organs or not. Because Rohingya women were reluctant to respond to the question of sexual abuse as an individual option, we added the option of both physical and sexual abuse together. Another response option was that the refugee did not experience any direct abuse but was exposed to grief for losing family members or property or experiencing separation, anxiety, or other trauma.

Postdisplacement Factors

Many of the Rohingya refugees worked in the camp, especially as day laborers who performed critical infrastructure work. In addition to assisting with outreach and coordination in the sprawling camps, the paid volunteer work was essential to initiatives to build roads, prevent landslides, and clear sewage. We included employment status as either employed (paid work) or as unemployed. Based on an individual’s perception, refugees were also assessed on whether they were receiving sufficient humanitarian aid for the family or not. In addition, we obtained data on the presence of any existing physical disability (eg, deafness, blindness, or amputation) of a stateless refugee.

Outcome Measurement: PTSSs

The study applied the psychometric properties of the IES-R criteria to assess the severity of PTSSs within the study population. This scale is a short, 22-item self-report questionnaire that is not a diagnostic tool for PTSD but an appropriate instrument to measure the subjective response to a specific traumatic event experienced by an adult. Three clusters of symptoms are included in the IES-R: the hyperarousal, interference, and avoidance subscales. Respondents were asked to describe a particular event and then indicate how much each event identified has upset or disturbed them during the past 7 days. We calculated the total subjective stress IES-R score (range, 0-88) from the 3 subscales. The total scale of Cronbach α was 0.87 (95% CI, 0.86-0.88), indicating a high degree of reliability. Posttraumatic stress symptoms were categorized into 4 categories: no symptoms (total IES-R score, ≤23), mild symptoms (score, 24-32), moderate symptoms (score, 33-38), and severe symptoms (score, ≥39) of PTSD. The categorization and interpretation of the IES-R score were described in a report from the Hartford Institute for Geriatric Nursing. In addition, we categorized the total IES-R score into 2 groups following the recommendation of Creamer et al: less than 33, which identified refugees with no or minimum symptoms of PTSD, and 33 or above, which identified refugees with moderate or severe symptoms of PTSD.

Statistical Analysis

Data were analyzed using R, version 3.6.2 (R Project for Statistical Computing). The questionnaire, R scripts, and data are available online. All categorical variables (presented as frequencies and percentages) were assessed using descriptive statistics. We estimated prevalence ratios (PRs) and their 95% CIs using a multivariable logistic model after adjusting for potential confounders. The PR is the ratio between the likelihood of an outcome in the exposed group and the likelihood of an outcome in the unexposed group. By using the delta method, we obtained the SEs for PRs. In the adjusted model, we controlled for demographic factors, such as age, sex, educational level, marital status, history of predisplacement abuse, current paid employment status in refugee camps, number of family members, and self-reported sufficient humanitarian aid for the family. We also obtained variance inflation factors in the logistic regression model to evaluate potential multicollinearity in the model (eTable 2 in the Supplement).

The pattern of missing data in the study sample is presented in eTable 3 in the Supplement. We found that the proportion of missing data ranged from 0.01% (1 of 1183) for family size to 12.5% (148 of 1036) for body mass index (BMI; calculated as weight in kilograms divided by height in meters squared). Also, we did not include missing data for covariates for 32 participants in multivariable analysis.

Results

Response Rate

Of the 1280 households sampled, 17 were excluded because the head of the household did not consent to participate. An additional 57 households were excluded because no persons were eligible to be included in the study during the study period. Finally, in our analysis, we had 1184 households, for a 92.5% response rate. A thorough calculation of the response rate is given in the eAppendix in the Supplement.

Characteristics of the Participants

Table 1 shows the sociodemographic, predisplacement, and postdisplacement factors of the participants according to the categories of severity of PTSSs. Of the 1184 Rohingya refugees who participated, 625 (52.8%) were men, and 559 (47.2%) were women. The mean (SD) age of respondents was 35.1 (13.4) years. We found that 509 of 1036 respondents (49.1%) had a normal BMI between 18.5 and 24.9, and 431 (41.6%) were overweight or obese. A total of 994 of 1182 respondents (84.1%) were married, and 766 of 1182 respondents (64.8%) had a family of more than 4 members. In addition, 751 of 1156 respondents (65.0%) did not go to school and could not read or write. In Bangladesh, refugees are not legally permitted to work, but 276 of 1165 respondents (23.7%) reported having paid employment opportunities in the refugee camps. We found that 136 of 1177 respondents (11.6%) experienced predisplacement physical and sexual abuse in Myanmar. Sixty-four of 1183 respondents (5.4%) had physical disabilities. Also, 496 of 1182 respondents (42.0%) reported that they did not receive adequate humanitarian aid for their family during the last 7 days.

Table 1. Characteristics of Participants by Severity of Posttraumatic Stress Symptoms.

Moreover, we investigated the association between postdisplacement factors and demographic variables. It appears that Rohingya women were not receiving opportunities for paid employment equal to the opportunities that men received. People younger than 45 years were more engaged in paid employment. Also, adults younger than 35 years reported experiencing more physical and sexual abuse relative to older age groups. The Rohingya adults who had more than 4 family members reported a lack of relief from humanitarian agencies.

Prevalence of PTSSs and Unadjusted PRs

Of the 1184 adult Rohingya refugees, 552 (46.6%) had severe PTSSs, and 274 had moderate PTSSs (23.1%). Severe mental health symptoms were more prevalent in male refugees (303 of 625 [48.5%]) than in female refugees (249 of 559 [44.5%]) (Table 1). The prevalence of severe PTSSs was 57.4% (296 of 516) for refugees aged 35 years or older and 38.3% (256 of 668) for those younger than 35 years. The pattern of severe PTSSs increased with increasing age of the respondents: 18 to 24 years, 36.9% (103 of 279); 25 to 34 years, 39.3% (153 of 389); 35 to 44 years, 58.9% (123 of 209); 45 to 54 years, 59.4% (107 to 180); and 55 years or older, 52.0% (66 of 127). Among married respondents, 480 of 994 (48.3%) had severe PTSSs, while among the never-married respondents, 52 of 153 (34.0%) had severe PTSSs. Of the 766 households with more than 4 family members, 382 (49.9%) had severe PTSSs; of the 62 households with less than 3 members, 22 (35.5%) had severe PTSSs. We found that 363 of the 751 respondents who did not have formal education (48.3%) had severe PTSSs. Employed refugees had a lower prevalence of severe PTSSs (113 of 276 [40.9%]) than nonworking participants (430 of 889 [48.4%]). Of the 64 respondents with physical disabilities, 41 (64.1%) had severe PTSSs, while 511 of the 1119 respondents without physical disabilities (45.7%) had severe PTSSs. Also, 281 of 496 respondents who did not receive adequate humanitarian aid during the last 7 days (56.7%) reported severe PTSSs. Moreover, 87 of the 1177 respondents who experienced physical and sexual abuse before displacement (64.0%) reported severe PTSSs.

After categorizing traumatic distress into 2 categories, we calculated the unadjusted PR (Table 2). The PR of mental health symptoms increased with age; with refugees aged 18 to 24 years as a reference, the PR for those aged 25 to 34 years was 1.22 (95% CI, 1.08-1.39), and the PR for those aged 55 years or older was 1.48 (95% CI, 1.28-1.70). Male sex may be a factor associated with severe PTSSs (PR, 1.06; 95% CI, 0.98-1.14). A significant association between receipt of adequate humanitarian aid and PTSSs was observed (PR, 0.75; 95% CI, 0.70-0.81). The results indicate that refugees in Myanmar who experienced both physical and sexual abuse had an increased risk of experiencing severe mental health symptoms (PR, 1.59; 95% CI, 1.43-1.76). Also, results indicated that paid employment in the refugee camps was associated with fewer severe mental health symptoms (PR, 0.91; 95% CI, 0.83-1.01).

Table 2. Unadjusted Prevalence Ratios by Sociodemographic, Predisplacement, and Postdisplacement Factors on Traumatic Distress.

Multivariable Analysis: Adjusted PRs

We estimated adjusted prevalence ratios (aPRs) using a multivariable logistic regression model adjusted for potential confounders (Table 3). It appears that sufficient humanitarian aid is associated with reduced risk of symptoms of traumatic distress (aPR, 0.50; 95% CI, 0.38-0.65). The results also suggest that both physical and sexual abuse before displacement were associated with a significant increase in mental health symptoms (aPR, 2.09; 95% CI, 1.41-3.07). After adjustment, results also indicate that paid employment opportunities in refugee camps were associated with a reduced risk of developing mental health symptoms (aPR, 0.69; 95% CI, 0.52-0.91).

Table 3. Adjusted Prevalence Ratios of Posttraumatic Stress Symptoms After Adjusting for Potential Confounders.

Discussion

Nearly 1 million Rohingya people live in the refugee camps of Bangladesh, more than 750 000 of whom have been living there since August 2017. However, compared with 2 years ago, the living conditions in the camps have generally improved, but refugees still live in small, overcrowded temporary shelters in the camps, without sufficient food, clean water, or toilets. Their lives are on hold, and their futures are uncertain. Alongside the traumatic experiences that many Rohingya refugees have experienced, these postmigration factors may contribute to a growing desperation. This study provides a detailed view of the symptoms of traumatic distress encountered by Rohingya refugees in Bangladesh.

Our study indicates that 46.6% of respondents had severe PTSSs and that 23.1% of respondents had moderate PTSSs in the third year after the forced evacuation. In various refugee communities globally, a varying proportion of diagnosed mental health problems has been observed. The prevalence of PTSD in Syrian refugees living in Turkey and Iraq was 83.4% after remaining in the camps for approximately 1 year and 60% after remaining in the camps for approximately 3 years. The prevalence of mental illness was 54% among the war-affected Ugandan population, where two-thirds of that population was displaced for more than 5 years. Also, 48.8% of Afghan refugees who fled their country nearly 20 years ago and have now lived in Australia for 1 to 5 years have met the criteria for PTSD. One study of Syrian refugees living in Germany found that only 13% had mental health disorders. However, a small study conducted in 2017 among camp-based Rohingya refugees who fled to Bangladesh found that 36% of adults had PTSSs and that 89% had symptoms of depression.

There are currently insufficient resources for mental health services in Rohingya refugee settings, and the number of mental health professionals is too low to cover the entire Rohingya population in need. The high prevalence of PTSSs suggests that a scale-up of mental health care is needed, which could be met by increasing medical workers’ capacity in refugee health facilities to diagnose and treat patients with mental disorders.

In Bangladesh, refugees are not legally entitled to work. The inability to survive without jobs has led many refugees, especially men, to illegally seek paid employment. Many of the Rohingya refugees work in the refugee camps, especially as paid day laborers who help construct roads, prevent landslides, and clear sewage. A few of the refugees have small grocery stores. The paid work allows some Rohingya refugees to supplement their own family’s food rations and provides opportunities to obtain a variety of food. Paid employment offers not just the freedom to buy needed items but also a chance to work with others. Work offers hope that is crucial to healing from mental illness. Our results indicate that those with paid employment are less likely to have symptoms of traumatic distress than unemployed refugees. One meta-analysis found poorer mental health outcomes for refugees with fewer economic opportunities.

Humanitarian aid from the government or NGOs plays an essential role in Rohingya refugees’ lives. The refugees are solely dependent on humanitarian assistance. Our findings show that sufficient humanitarian aid to a family is associated with a lower risk of developing mental health symptoms compared with those who received insufficient assistance. Similar results found that a lack of social support was associated with poorer mental health.

Before the mass exodus to Bangladesh, sex-based abuse toward refugees was well documented in many reports. We found that physical abuse was more frequent among male refugees than female refugees and that both physical and sexual abuse were equally prevalent among male and female refugees. Our findings indicate that, among Rohingya refugees subjected to physical and sexual abuse, the prevalence of severe PTSSs was high. Many studies show that women were found to have a higher incidence of mental health disorders after rape or sexual assault.

Approximately half the respondents in our sample were male. The study findings show that symptoms of traumatic distress were more prevalent among male refugees than female refugees. Several studies have shown that male refugees had a higher prevalence of poor mental health. A qualitative study among refugees from Afghanistan speculates that the lack of job opportunities plays a crucial role in mental distress among men after years of forced migration. Rohingya women do not get enough paid jobs in conservative Rohingya society, making it difficult for them to earn a living. Lack of paid employment may be associated with the increase in depression among women. Our research also indicates that refugees with physical disabilities were at a much greater risk of elevated PTSD symptoms than those without disabilities. This finding is consistent with previous studies that found that disability was associated with a high prevalence of mental illness.

In our sample, the risk of traumatic distress increased with increasing age regardless of sex. The findings are consistent with studies conducted among Syrian and Sudanese refugees. Our research also found that refugees who were unable to read or write had a higher prevalence of severe mental health symptoms than those who had schooling. Currently, adult refugees do not have educational opportunities in the refugee camps, and job placement is not based on education. Therefore, among adults in the refugee camps, education has no significance. This finding is comparable to previous research that showed that poor education was correlated with higher rates of PTSD.

Also, married respondents were more likely to have PTSSs than those who were never married. This finding is comparable to findings in previous research. Our study further found a high risk of PTSSs for refugees with family sizes of more than 4 members. The Rohingya population believed that family planning methods clash with their faith. Thus, married people are more likely to have a large family and be unable to afford enough food or fulfill other needs; this explains why the variables of marital status and family sizes have a positive association with traumatic distress.

Strengths and Limitations

There are some strengths to our research. First, a local NGO working with Rohingya refugees for more than 3 years helped us collect data from the 8 camps. Our data reduced the information bias by involving interviewers from Rohingya refugee camps. Second, we recruited households through a random sampling technique. Third, to assess symptoms of mental well-being, we collected data on a variety of variables that were not taken into account in current registries or population-scale monitoring efforts.

Several limitations should be taken into account when interpreting our results. First, our research used a scale to measure PTSSs that was not validated for use in the Rohingya community. Second, the method of this short duration of the cross-sectional survey could fail to address the transient nature of the population. Third, we did not assess mental health treatment availability and use in the last 2 years of the postmigration period. This limitation could be significant because accessing or using mental health services may mitigate the prevalence of PTSSs among respondents. Fourth, we did not gather details on the number of assaults experienced by the respondents or the number of many family members who were involved in those events. We had a question concerning the loss of family members during the attack. Unfortunately, with this question, some of the respondents got too emotional to respond, and several respondents misunderstood the question about the concept of immediate family members. We stopped asking these questions later in the study. Similarly, several respondents were worried about their health conditions, but they were not certain about chronic diseases because they did not visit any health facilities. Therefore, many respondents were unable to answer this question. We did not include several postmigration stressors, such as the refugee camp climate, access to health care or education, varieties of food, and limited access to certain services. These variables may have confounding effects on mental health.

Conclusions

The high prevalence of self-reported mental health symptoms among Rohingya refugees in this study suggests that the trauma of displacement and the violent consequences of military clearance operations are still present. Different support services, such as access to education or training on stress management for their violent memories, may reduce the burden of severe mental health symptoms. Our findings indicate the importance of ensuring that every household receives sufficient humanitarian assistance. Employment opportunities at Rohingya refugee camp settings hold promise as a potential intervention to reduce the burden of mental health symptoms.

Summary

The Rohingya people in Myanmar are among the most oppressed groups globally. Myanmar's 1982 Citizenship Act removed their official recognition as an ethnic minority. This denied them basic rights like citizenship, freedom to move, healthcare, education, marriage registration, and voting. This made them the world's largest stateless population. In August 2017, the Myanmar army started a large operation in Rakhine. About 1.2 million Rohingya lived there at the time. This operation led to an estimated 7,800 Rohingya deaths. It forced around 750,000 Rohingya to flee to Bangladesh's Cox's Bazar district.

Studies have shown the abuse Rohingya refugees faced before being displaced. They were denied medical care for physical or sexual assault in Myanmar. Emotional and verbal abuse was also a big part of these violent events. Rohingya leaders have confirmed these reports. These physical, sexual, and emotional abuses had lasting effects on survivors' mental health. They were linked to conditions like post-traumatic stress disorder (PTSD).

Post-traumatic stress, a feature of PTSD, includes troubling thoughts, nightmares, memories of past trauma, avoiding trauma reminders, being overly alert, and sleep problems. These issues cause severe problems in a person's mind, work, and relationships. People with PTSD can have strong reactions to triggers, such as hallucinations, extreme anxiety, or wanting to flee or fight. These symptoms can cause emotional numbness and reduce participation in daily life. In severe cases, it can lead to feeling cut off from others. Depressive disorders, anxiety disorders, and drug misuse are 2 to 4 times more common in people with PTSD. PTSD can also increase the risk of suicide attempts. This study used the Impact of Event Scale–Revised (IES-R) to measure post-traumatic stress symptoms (PTSSs). This tool helps people report symptoms of PTSD related to a life-threatening event. It covers symptoms like re-experiencing the trauma, avoiding reminders, and being overly aware.

Refugees faced many stressful situations before and after being displaced. Those who experienced abuse were at higher risk for mental health problems.

Many studies have described the challenges Rohingya refugees face in Bangladesh. They live in small, crowded temporary shelters in camps without enough food, clean water, or toilets. Their lives are on hold, and their future is unclear. A 2017 study looked at daily stressors among 148 adult Rohingya. It found worse mental health outcomes for these refugees. Compared to two years ago, basic needs and healthcare for Rohingya refugees have mostly improved. However, factors after displacement are still important for improving their mental health.

Screening for mental health only works when combined with good mental well-being services. A challenge in providing proper services for Rohingya refugees in Bangladesh is the lack of information on their mental health status. This report aims to find out how common PTSSs are among adult Rohingya in Bangladesh after the large clearance operation. It also looks at the factors before and after displacement that are linked to these symptoms.

Methods

Study Design and Participants

From September 2019 to January 2020, a survey was conducted among Rohingya refugees in Cox’s Bazar, Bangladesh. Over 1 million Rohingya refugees live in Bangladesh, mostly in Ukhia and Teknaf. Kutupalong, the world's largest refugee camp in Ukhia, houses over 600,000 refugees. Participants were chosen from Kutupalong using a three-step sampling method. First, 8 camps were randomly selected from the 23 camps in Kutupalong, assuming each had the same population size. Second, at least 160 households were chosen from each selected camp, aiming for 1280 households in total. More households were targeted than needed to achieve 80% power and a 95% confidence interval of 0.05 to 1.96, assuming 40% of the population had a mental illness, with a design effect of 2. A systematic sampling technique was used, starting with a randomly chosen household from the center of each camp. Data collectors then moved to the nearest household until 160 households were sampled. Third, one adult per household was interviewed, preferably the head of the household. If the male head was unavailable, the female head or another available adult was interviewed. Household members were defined as those living together for at least one month, sharing cooking and eating facilities. Participants also had to have lived in the camp for at least two years after displacement.

Participants gave verbal consent because many were unable to read or write and were hesitant to sign or use fingerprints. They were assured that all information would be confidential and used only for research. They received contact information for researchers for any future questions. The study was approved by the institutional review board at North South University, Bangladesh, and followed the STROBE reporting guidelines.

Recruitment and Training

Data was collected and cleaned by a team of four, two men and two women, from the Health Management BD (HMBD) Foundation. This local NGO works in Rohingya refugee camps. The HMBD Foundation selected local community leaders ("maji") from each of the eight chosen camps. These leaders were informed about the study's research and ethics. They helped the HMBD Foundation recruit eight local men from each camp who could speak both Bengali (Bangla) and Rohingya. A team of data collectors was then formed, including one person from the camp and one from the HMBD Foundation. Interviews were conducted in both Rohingya and Bangla. The camp data collector asked questions in Rohingya, and the HMBD Foundation member checked the answers by asking the same questions in Bangla. Two research investigators from North South University (A.H. and T.A.K.) provided a one-day practical training session on ethics and data collection. Enumerators and data collectors were informed about the study's goals, methods, and questionnaire. Researchers also taught data collectors how to build rapport, stay neutral, and handle ethical issues, privacy, cultural awareness, and mental health risks. After training, a pilot study was conducted with all eight study teams as a single unit. This aimed to observe their ability to understand techniques and difficult situations during interviews. Necessary corrections were made after the pilot study. Afterward, each trained team visited their assigned camp together to collect data using a semi-structured questionnaire.

Data Collection

Respondents were informed that participation was voluntary. Face-to-face interviews were conducted privately, one person at a time. No money or food was given as incentives. Questions were read aloud, one at a time, and respondents chose from given options. Co-investigators reviewed data sheets for completeness, accuracy, and consistency, which the principal investigator then confirmed.

Sociodemographic Variables

The questionnaire's first part gathered information on age, sex, height, weight, marital status (never married, widowed or divorced, married), and the number of family members in a household. Family members were later grouped as two or less, three to four, and more than four. Educational level was divided into two groups: those who attended school and those who could not read or write.

Predisplacement Abuse

The term "abuse" in this study refers to violence intended to control or dominate Rohingya people, causing anger, harm, resentment, humiliation, coercion, and helplessness. Data was collected on the types of abuse refugees experienced before their forced displacement to Bangladesh. Categories included indirect abuse (or no abuse), verbal or emotional abuse, physical abuse, and both physical and sexual abuse. Verbal or emotional abuse was defined as non-physical acts that harm a person's psychological well-being, such as coercion, defamation, verbal insults, or harassment. This could also include forced labor, separation from family, or witnessing abuse against family or community members. Physical abuse involved any force applied to the body, such as shaking, burning, scalding, choking, hair pulling, hitting, slapping, kicking, or threats or attacks with weapons. It also included restraining, tying, or locking someone against their will. Sexual abuse encompassed sexual humiliation (forced masturbation or nudity), sexual slavery, rape (vaginal, oral, anal, or attempted), genital abuse (beatings, electric shock, or mutilation), castration, penis amputation, sterilization, or forced marriage, cohabitation, or sexual activity (with a stranger, family member, or corpse). To distinguish between physical and sexual abuse, respondents were carefully asked if the event involved sexual organs. Because Rohingya women were reluctant to respond to sexual abuse as a standalone option, it was combined with physical abuse. Another option for response was that the refugee did not experience direct abuse but suffered grief from losing family members or property, or experienced separation, anxiety, or other trauma.

Postdisplacement Factors

Many Rohingya refugees worked in the camps, often as day laborers, performing important infrastructure tasks. This paid volunteer work was crucial for building roads, preventing landslides, and clearing sewage. It also helped with outreach and coordination in the large camps. Employment status was recorded as either employed (paid work) or unemployed. Refugees also reported if they felt they were receiving enough humanitarian aid for their family. Data was also collected on whether a stateless refugee had any physical disabilities (e.g., deafness, blindness, or amputation).

Outcome Measurement: PTSSs

The study used the IES-R to measure the severity of PTSSs in the population. This short, 22-item questionnaire helps adults describe their subjective response to a specific traumatic event. It is not a diagnostic tool for PTSD. The IES-R has three symptom clusters: hyperarousal, intrusion, and avoidance. Respondents described a specific event and then rated how much each event had upset or disturbed them over the past seven days. The total IES-R score (0-88) was calculated from the three subscales. The total scale had a Cronbach α of 0.87 (95% CI, 0.86-0.88), indicating high reliability. PTSSs were grouped into four categories: no symptoms (total IES-R score, ≤23), mild symptoms (score, 24-32), moderate symptoms (score, 33-38), and severe symptoms (score, ≥39). The Hartford Institute for Geriatric Nursing provided these categories and interpretations. Additionally, the total IES-R score was divided into two groups, following Creamer et al.'s recommendation: less than 33 for no or minimal PTSD symptoms, and 33 or above for moderate or severe PTSD symptoms.

Statistical Analysis

Data analysis was performed using R, version 3.6.2. The questionnaire, R scripts, and data are publicly available. Descriptive statistics were used for all categorical variables, presented as frequencies and percentages. Prevalence ratios (PRs) and their 95% CIs were estimated using a multivariable logistic model, adjusted for potential confounding factors. The PR shows how much more likely an outcome is in an exposed group compared to an unexposed group. Standard errors for PRs were calculated using the delta method. The adjusted model accounted for demographic factors such as age, sex, education level, marital status, history of predisplacement abuse, current paid employment in refugee camps, number of family members, and self-reported sufficient humanitarian aid for the family. Variance inflation factors were also obtained in the logistic regression model to check for potential multicollinearity.

The pattern of missing data in the study sample was analyzed. The proportion of missing data ranged from 0.01% (1 out of 1183) for family size to 12.5% (148 out of 1036) for body mass index (BMI). Missing data for covariates were not included for 32 participants in the multivariable analysis.

Results

Response Rate

Out of 1280 sampled households, 17 were excluded because the head of the household did not agree to participate. An additional 57 households were excluded because no eligible individuals were available during the study period. Ultimately, 1184 households were included in the analysis, resulting in a 92.5% response rate. A detailed calculation of the response rate is provided in the eAppendix in the Supplement.

Characteristics of the Participants

Table 1 shows the sociodemographic, predisplacement, and postdisplacement factors of participants, categorized by the severity of PTSSs. Of the 1184 Rohingya refugees who participated, 625 (52.8%) were men, and 559 (47.2%) were women. The average age was 35.1 years (standard deviation of 13.4). Among 1036 respondents, 509 (49.1%) had a normal BMI (between 18.5 and 24.9), and 431 (41.6%) were overweight or obese. A total of 994 out of 1182 respondents (84.1%) were married, and 766 out of 1182 respondents (64.8%) had more than four family members. Additionally, 751 out of 1156 respondents (65.0%) had not attended school and could not read or write. While refugees in Bangladesh are not legally allowed to work, 276 out of 1165 respondents (23.7%) reported having paid employment opportunities in the refugee camps. Among the respondents, 136 out of 1177 (11.6%) had experienced predisplacement physical and sexual abuse in Myanmar. Sixty-four out of 1183 respondents (5.4%) had physical disabilities. Also, 496 out of 1182 respondents (42.0%) reported not receiving adequate humanitarian aid for their family during the last seven days.

Furthermore, the study examined the connection between factors after displacement and demographic variables. It appeared that Rohingya women had fewer opportunities for paid employment compared to men. Individuals under 45 years of age were more likely to be in paid employment. Also, adults under 35 years reported more experiences of physical and sexual abuse than older age groups. Rohingya adults with more than four family members reported a lack of adequate relief from humanitarian agencies.

Prevalence of PTSSs and Unadjusted PRs

Of the 1184 adult Rohingya refugees, 552 (46.6%) had severe PTSSs, and 274 (23.1%) had moderate PTSSs. Severe mental health symptoms were more common in male refugees (303 out of 625, or 48.5%) than in female refugees (249 out of 559, or 44.5%) (Table 1). The prevalence of severe PTSSs was 57.4% (296 out of 516) for refugees aged 35 years or older and 38.3% (256 out of 668) for those younger than 35 years. The pattern of severe PTSSs increased with age: 18 to 24 years, 36.9% (103 out of 279); 25 to 34 years, 39.3% (153 out of 389); 35 to 44 years, 58.9% (123 out of 209); 45 to 54 years, 59.4% (107 out of 180); and 55 years or older, 52.0% (66 out of 127). Among married respondents, 480 out of 994 (48.3%) had severe PTSSs, while among never-married respondents, 52 out of 153 (34.0%) had severe PTSSs. Of the 766 households with more than four family members, 382 (49.9%) had severe PTSSs; of the 62 households with fewer than three members, 22 (35.5%) had severe PTSSs. Among the 751 respondents without formal education, 363 (48.3%) had severe PTSSs. Employed refugees had a lower prevalence of severe PTSSs (113 out of 276, or 40.9%) compared to unemployed participants (430 out of 889, or 48.4%). Among the 64 respondents with physical disabilities, 41 (64.1%) had severe PTSSs, while 511 out of 1119 respondents without physical disabilities (45.7%) had severe PTSSs. Also, 281 out of 496 respondents who did not receive adequate humanitarian aid during the last seven days (56.7%) reported severe PTSSs. Moreover, 87 out of 1177 respondents who experienced physical and sexual abuse before displacement (64.0%) reported severe PTSSs.

After dividing traumatic distress into two groups, the unadjusted PR was calculated (Table 2). The PR of mental health symptoms increased with age; using refugees aged 18 to 24 years as a reference, the PR for those aged 25 to 34 years was 1.22 (95% CI, 1.08-1.39), and for those aged 55 years or older, it was 1.48 (95% CI, 1.28-1.70). Being male might be a factor associated with severe PTSSs (PR, 1.06; 95% CI, 0.98-1.14). A significant link between receiving adequate humanitarian aid and PTSSs was observed (PR, 0.75; 95% CI, 0.70-0.81). The results show that refugees in Myanmar who experienced both physical and sexual abuse had a higher risk of severe mental health symptoms (PR, 1.59; 95% CI, 1.43-1.76). Additionally, paid employment in refugee camps was linked to fewer severe mental health symptoms (PR, 0.91; 95% CI, 0.83-1.01).

Multivariable Analysis: Adjusted PRs

Adjusted prevalence ratios (aPRs) were estimated using a multivariable logistic regression model that accounted for potential confounding factors (Table 3). It appears that sufficient humanitarian aid is linked to a reduced risk of traumatic distress symptoms (aPR, 0.50; 95% CI, 0.38-0.65). The results also suggest that both physical and sexual abuse before displacement were significantly linked to an increase in mental health symptoms (aPR, 2.09; 95% CI, 1.41-3.07). After adjustment, the results also show that paid employment opportunities in refugee camps were linked to a reduced risk of developing mental health symptoms (aPR, 0.69; 95% CI, 0.52-0.91).

Discussion

Nearly 1 million Rohingya people live in Bangladeshi refugee camps, with over 750,000 residing there since August 2017. While living conditions in the camps have generally improved over the past two years, refugees still inhabit small, overcrowded temporary shelters lacking sufficient food, clean water, or toilets. Their lives remain in limbo, and their futures are uncertain. These post-migration factors, combined with the traumatic experiences many Rohingya refugees have endured, may contribute to increasing despair. This study provides a detailed look at the symptoms of traumatic distress experienced by Rohingya refugees in Bangladesh.

The study indicates that 46.6% of respondents had severe PTSSs and 23.1% had moderate PTSSs during the third year after forced displacement. Various proportions of diagnosed mental health problems have been observed in refugee communities worldwide. For example, PTSD prevalence in Syrian refugees was 83.4% after about one year in camps in Turkey and Iraq, and 60% after about three years. Among war-affected Ugandans, two-thirds of whom were displaced for over five years, mental illness prevalence was 54%. Similarly, 48.8% of Afghan refugees who fled nearly 20 years ago and have lived in Australia for 1 to 5 years met PTSD criteria. However, a small 2017 study of camp-based Rohingya refugees in Bangladesh found that 36% of adults had PTSSs and 89% had depression symptoms.

Currently, there are not enough mental health services in Rohingya refugee settings. The number of mental health professionals is too low to serve all Rohingya in need. The high prevalence of PTSSs indicates a need to increase mental health care. This could be achieved by training more medical workers in refugee health facilities to diagnose and treat mental disorders.

In Bangladesh, refugees are not legally allowed to work. The inability to survive without jobs has led many, especially men, to seek illegal paid employment. Many Rohingya refugees work in the camps, often as paid day laborers, helping to build roads, prevent landslides, and clear sewage. A few refugees also own small grocery stores. Paid work allows some Rohingya refugees to buy extra food for their families and get a wider variety of food. This work offers not just the freedom to buy needed items but also a chance to work with others. Work provides hope, which is essential for healing from mental illness. The study's results show that those with paid employment are less likely to experience traumatic distress symptoms than unemployed refugees. One large review of studies found that refugees with fewer economic opportunities had worse mental health outcomes.

Humanitarian aid from governments or NGOs is crucial for Rohingya refugees. They depend entirely on this assistance. The study's findings show that sufficient humanitarian aid to a family is linked to a lower risk of developing mental health symptoms compared to those who received insufficient help. Similar research found that a lack of social support was linked to poorer mental health.

Before the mass exodus to Bangladesh, reports widely documented sex-based abuse against refugees. The study found that physical abuse was more frequent among male refugees than female refugees. Both physical and sexual abuse were equally common among male and female refugees. Findings indicate a high prevalence of severe PTSSs among Rohingya refugees who experienced physical and sexual abuse. Many studies show that women have a higher rate of mental health disorders after rape or sexual assault.

Approximately half of the study participants were male. The findings indicate that symptoms of traumatic distress were more common among male refugees than female refugees. Several studies have shown that male refugees often experience poorer mental health. A qualitative study among refugees from Afghanistan suggests that a lack of job opportunities plays a significant role in mental distress among men after years of forced migration. In conservative Rohingya society, women often lack paid employment opportunities, making it difficult for them to earn a living. A lack of paid employment may be linked to an increase in depression among women. The research also indicates that refugees with physical disabilities faced a much higher risk of elevated PTSD symptoms compared to those without disabilities. This finding is consistent with previous studies that have shown a link between disability and a high prevalence of mental illness.

In the study sample, the risk of traumatic distress increased with age, regardless of sex. These findings are consistent with studies conducted among Syrian and Sudanese refugees. The research also found that refugees who could not read or write had a higher prevalence of severe mental health symptoms than those who had schooling. Currently, adult refugees do not have educational opportunities in the camps, and job placement is not based on education. Therefore, among adults in the refugee camps, education does not have much significance. This finding is similar to previous research showing a correlation between poor education and higher rates of PTSD.

Also, married respondents were more likely to have PTSSs than those who had never married. This finding is consistent with previous research. The study also found a high risk of PTSSs for refugees with family sizes of more than four members. The Rohingya population believes that family planning methods conflict with their faith. Thus, married people are more likely to have large families and may struggle to afford enough food or meet other needs. This explains why marital status and family size are positively associated with traumatic distress.

Strengths and Limitations

This research has several strengths. First, a local NGO that has worked with Rohingya refugees for over three years assisted in collecting data from the eight camps. Using interviewers from Rohingya refugee camps helped reduce bias in the information. Second, households were recruited using a random sampling method. Third, to evaluate mental well-being, the study collected data on various factors not included in existing records or large-scale monitoring efforts.

Several limitations should be considered when interpreting the results. First, the study used a scale to measure PTSSs that had not been officially tested for use within the Rohingya community. Second, the cross-sectional survey's short duration might not fully capture the temporary nature of the population. Third, the study did not assess the availability or use of mental health treatment during the two years after displacement. This is an important limitation because accessing mental health services could reduce the prevalence of PTSSs among respondents. Fourth, detailed information on the number of assaults experienced by respondents or how many family members were involved in these events was not collected. There was a question about the loss of family members during the attack. Unfortunately, some respondents became too emotional to answer, and several misunderstood the concept of immediate family members. These questions were eventually removed from the study. Similarly, many respondents were concerned about their health but were unsure about chronic diseases because they had not visited health facilities. Therefore, many could not answer this question. Several post-migration stressors, such as the camp climate, access to healthcare or education, food variety, and limited access to certain services, were not included. These variables might have confusing effects on mental health.

Conclusions

The high prevalence of self-reported mental health symptoms among Rohingya refugees in this study indicates that the trauma of displacement and the violent impact of military operations are still present. Various support services, such as access to education or training in stress management for their traumatic memories, could help reduce the burden of severe mental health symptoms. The findings highlight the importance of ensuring that every household receives enough humanitarian assistance. Employment opportunities in Rohingya refugee camp settings show promise as a possible way to reduce the burden of mental health symptoms.

Introduction

The Rohingya people in Myanmar are one of the world’s most oppressed minority groups. In 1982, the Citizenship Act removed the Rohingya from the list of recognized ethnic groups. This law took away many basic rights, including citizenship, freedom to move, access to healthcare and education, the ability to marry legally, and the right to vote. This made the Rohingya the largest group of stateless people in the world. In August 2017, the Myanmar army started a large operation in Rakhine, where about 1.2 million Rohingya lived. This operation led to the deaths of an estimated 7,800 Rohingya. It also forced about 750,000 Rohingya to flee Myanmar, seeking safety in Bangladesh’s Cox’s Bazar district.

Studies have documented the abuse Rohingya refugees faced before they were displaced. They were denied medical care for physical or sexual assault in Myanmar. Emotional and verbal abuse was also a significant part of these violent incidents. Rohingya community leaders have confirmed these descriptions in various reports. These experiences of physical, sexual, and emotional abuse were reported to have lasting effects on the mental health of survivors. They were also linked to the potential development of mental health conditions like posttraumatic stress disorder (PTSD).

Posttraumatic stress is a set of symptoms related to PTSD. These include unwanted thoughts, nightmares, memories of past traumatic events, avoiding things that remind one of the trauma, being overly alert, and trouble sleeping. These symptoms can severely impact a person's psychological well-being, work life, and relationships. People with PTSD may have strong reactions to certain triggers, such as hallucinations, extreme anxiety, or a strong urge to flee or fight. These symptoms can lead to emotional numbness, less involvement in daily activities, and, in severe cases, feeling distant from others. Compared to people without PTSD, those with PTSD are 2 to 4 times more likely to experience depression, anxiety disorders, and substance misuse. PTSD can also increase the risk of suicide attempts. This study used the Impact of Event Scale–Revised (IES-R) to measure posttraumatic stress symptoms (PTSSs). This survey tool helps assess all three groups of PTSD symptoms—intrusion, avoidance, and hyperarousal—related to a specific life-threatening event.

Refugees faced many stressful situations both before and after being displaced. Individuals who experienced abuse were at a higher risk of developing mental health problems.

Numerous studies have reported the challenges faced by Rohingya refugees in Bangladesh. They live in small, crowded temporary shelters in refugee camps, often lacking enough food, clean water, or toilets. Their lives are in limbo, and they are unsure about their future. A 2017 study that looked at daily environmental stressors among 148 Rohingya adults found that refugees had poorer mental health outcomes. While basic needs and healthcare for Rohingya refugees have largely improved compared to two years ago, factors after displacement still play a crucial role in improving their mental health.

Screening for mental health issues is effective only when paired with high-quality mental well-being services. One obstacle to providing proper services for Rohingya refugees in Bangladesh is the lack of statistical information about their mental health status. This report aims to identify how common PTSSs are and what factors before and after displacement are linked to these symptoms among Rohingya adults living in Bangladesh after the large-scale military operation.

Methods

Study Design and Participants

From September 17, 2019, to January 11, 2020, a survey was conducted among Rohingya refugees living in Cox’s Bazar, Bangladesh. More than 1 million Rohingya refugees currently reside in Bangladesh, with most clustered in Ukhia and Teknaf, two administrative regions of the district. Kutupalong, in Ukhia, is the world's largest refugee settlement and houses over 600,000 refugees. Participants were recruited from Kutupalong using a three-stage sampling method. First, eight camps were randomly chosen from the 23 camps and expansion areas within Kutupalong, assuming each camp had an equal population size. Second, at least 160 households were selected from each chosen camp, aiming for a total of 1280 households for the study. More households were targeted than the required sample size to achieve 80% power and a 95% confidence interval of 0.05 to 1.96, based on an assumption that 40% of the population had a mental illness, and a design effect of 2. A systematic sampling technique was used, with the first household randomly picked from the approximate center of each camp. Data collectors then moved to the next closest household until 160 households were sampled. Third, one adult respondent per household was interviewed, preferably the head of the household. If the male head was unavailable, the female head or another available adult member was surveyed. Household members were defined as those who had lived under the same roof for at least one month and shared cooking and eating facilities from the same source. Participants were also required to have lived in the camp for at least two years after displacement. Detailed sampling information is available in eTable 1 of the Supplement. Participants gave verbal consent for the study, as they were hesitant to sign their names or provide fingerprints on any paper, and most were unable to read or write. They were assured that all collected information would be kept strictly confidential and used only for research. However, they were given a consent paper with contact information for the research investigators for future questions. The institutional review board at North South University, Bangladesh, approved the study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Recruitment and Training

A team of four enumerators, two men and two women from the Health Management BD (HMBD) Foundation, collected and cleaned the data. The HMBD Foundation, a local non-governmental organization working in Rohingya refugee camps since the influx, selected local community leaders (known as “maji”) from each of the eight chosen camps. These leaders were informed about the study’s research and ethics. They helped the HMBD Foundation recruit eight local male residents from each camp who could communicate in both Bengali (Bangla) and Rohingya languages. A data collection team was then formed, consisting of two people: one from the camp and one from the HMBD Foundation. Interviews were conducted in both Rohingya and Bangla. The camp data collector asked questions in Rohingya, and the HMBD Foundation member checked the answer by asking the same question in Bangla. Two research investigators from North South University organized a one-day practical training session on ethics and data collection. The enumerators and data collectors were also briefed on the study objectives, methods, and questionnaire. The researchers also taught data collectors techniques for building rapport, maintaining neutrality, and addressing ethical issues, privacy concerns, cultural awareness, and mental health risk management. After the training, a pilot study was conducted for the eight study teams, evaluated as a single unit. The purpose was to observe their ability to understand relevant techniques and handle difficult situations during interviews. Necessary corrections were made after the pilot study. Afterward, each trained team visited their assigned camp together to collect data using a semi-structured questionnaire.

Data Collection

Respondents were clearly informed that participation in the study was entirely voluntary. Face-to-face interviews were conducted one person at a time to ensure privacy. No monetary or food incentives were given to respondents. During the interview, questions were read aloud one at a time, and respondents were asked to choose from the provided scale options. Coinvestigators reviewed the data collection sheets for completeness, accuracy, and internal consistency, which were confirmed by the principal investigator.

Sociodemographic Variables

The first part of the questionnaire collected information about people's social and personal backgrounds. This included their age in years, sex (male or female), height in inches, weight in kilograms, marital status (never married, widowed or divorced, or married), and the number of family members in their household. Later, the number of family members was grouped into three categories: two or fewer, three to four, and more than four members. Education level was divided into two groups: those who attended school and those who could not read or write.

Predisplacement Abuse

The term "abuse" in this context refers to violence intended to control or dominate Rohingya people, causing anger, harm, resentment, humiliation, pressure, and helplessness. Data were collected on the types of abuse refugees experienced before they were forced to move to Bangladesh. Categories included indirect abuse or no exposure to abuse, verbal or emotional abuse, physical abuse, and both physical and sexual abuse. Verbal or emotional abuse was defined as any non-physical act that harms a person's psychological well-being. This could involve coercion, defamation, verbal insults, or harassment. It might also include forced labor, or witnessing or being separated from family or community members who were abused. Physical abuse was considered any force applied to the body, such as shaking, burning, scalding, choking, hair pulling, hitting, slapping, kicking, or threatening or attacking with a knife, gun, or other weapon. It could also include restraining, tying, or locking someone up against their will. Sexual abuse was defined as sexual humiliation (forced masturbation or nudity), sexual slavery, rape (vaginal, oral, anal, or attempted), genital abuse (beatings, electric shock, or mutilation), castration, penis amputation, sterilization, or forced marriage, cohabitation, or sexual activity (with a stranger, family member, or dead body). To distinguish between physical and sexual abuse, respondents were carefully asked whether the specific event involved sexual organs. Because Rohingya women were reluctant to answer questions about sexual abuse as a separate option, the option of both physical and sexual abuse was combined. Another response option was for refugees who did not experience direct abuse but experienced grief from losing family members or property, or from separation, anxiety, or other trauma.

Postdisplacement Factors

Many Rohingya refugees worked in the camp, particularly as day laborers performing crucial infrastructure work. This paid volunteer work was vital for road construction, landslide prevention, and sewage clearance, in addition to supporting outreach and coordination across the expansive camps. Employment status was categorized as either employed (paid work) or unemployed. Based on individual perceptions, refugees were also assessed on whether they received sufficient humanitarian aid for their family. Furthermore, data were collected on whether a stateless refugee had any existing physical disability, such as deafness, blindness, or amputation.

Outcome Measurement: PTSSs

The study used the psychometric properties of the IES-R criteria to measure how severe posttraumatic stress symptoms (PTSSs) were in the study population. This is a brief, 22-item self-report survey. It is not a tool for diagnosing PTSD but is suitable for measuring an adult's personal response to a specific traumatic event. The IES-R includes three symptom groups: hyperarousal, intrusion, and avoidance subscales. Respondents were asked to describe a particular event and then indicate how much that event had bothered or disturbed them over the past seven days. The total subjective stress IES-R score (ranging from 0 to 88) was calculated from the three subscales. The overall Cronbach’s alpha for the scale was 0.87 (95% CI, 0.86-0.88), indicating high reliability. Posttraumatic stress symptoms were divided into four categories: no symptoms (total IES-R score, ≤23), mild symptoms (score, 24-32), moderate symptoms (score, 33-38), and severe symptoms (score, ≥39). The categorization and interpretation of the IES-R score were based on a report from the Hartford Institute for Geriatric Nursing. Additionally, the total IES-R score was grouped into two categories, following Creamer et al.'s recommendation: less than 33, which indicated refugees with no or minimal PTSD symptoms, and 33 or above, which indicated refugees with moderate or severe PTSD symptoms.

Statistical Analysis

Data analysis was carried out using R, version 3.6.2 (R Project for Statistical Computing). The questionnaire, R scripts, and data are available online. Descriptive statistics were used to assess all categorical variables, presented as frequencies and percentages. Prevalence ratios (PRs) and their 95% confidence intervals (CIs) were estimated using a multivariable logistic model, adjusted for potential confounding factors. The PR represents the ratio of the likelihood of an outcome in the exposed group compared to the unexposed group. Standard errors for PRs were obtained using the delta method. In the adjusted model, demographic factors such as age, sex, education level, marital status, history of predisplacement abuse, current paid employment status in refugee camps, number of family members, and self-reported sufficient humanitarian aid for the family were controlled for. Variance inflation factors were also obtained from the logistic regression model to assess potential multicollinearity (eTable 2 in the Supplement).

The pattern of missing data in the study sample is presented in eTable 3 in the Supplement. The proportion of missing data ranged from 0.01% (1 of 1183) for family size to 12.5% (148 of 1036) for body mass index (BMI; calculated as weight in kilograms divided by height in meters squared). Missing data for covariates for 32 participants were not included in the multivariable analysis.

Results

Response Rate

Out of the 1280 households sampled, 17 were excluded because the head of the household did not agree to participate. An additional 57 households were excluded because no eligible individuals were available for the study during the study period. Ultimately, 1184 households were included in the analysis, resulting in a 92.5% response rate. A detailed calculation of the response rate is provided in the eAppendix of the Supplement.

Characteristics of the Participants

Table 1 shows the social, personal, predisplacement, and postdisplacement factors of participants, grouped by the severity of posttraumatic stress symptoms (PTSSs). Among the 1184 Rohingya refugees who took part, 625 (52.8%) were men, and 559 (47.2%) were women. The average age of respondents was 35.1 years, with a standard deviation of 13.4 years. It was found that 509 out of 1036 respondents (49.1%) had a normal body mass index (BMI) between 18.5 and 24.9, while 431 (41.6%) were overweight or obese. A total of 994 out of 1182 respondents (84.1%) were married, and 766 out of 1182 respondents (64.8%) had more than four family members. Additionally, 751 out of 1156 respondents (65.0%) had not attended school and could not read or write. Although refugees are not legally allowed to work in Bangladesh, 276 out of 1165 respondents (23.7%) reported having paid employment opportunities in the refugee camps. It was found that 136 out of 1177 respondents (11.6%) experienced physical and sexual abuse before displacement in Myanmar. Sixty-four out of 1183 respondents (5.4%) had physical disabilities. Furthermore, 496 out of 1182 respondents (42.0%) reported not receiving enough humanitarian aid for their family during the previous seven days.

Rohingya women were less likely to receive paid employment opportunities compared to men. Individuals younger than 45 years were more often engaged in paid employment. Adults younger than 35 years reported experiencing more physical and sexual abuse compared to older age groups. Rohingya adults with more than four family members reported a lack of assistance from humanitarian agencies.

Prevalence of PTSSs and Unadjusted PRs

Among the 1184 adult Rohingya refugees, 552 (46.6%) reported severe PTSSs, and 274 (23.1%) reported moderate PTSSs. Severe mental health symptoms were more common among male refugees (303 of 625 [48.5%]) than female refugees (249 of 559 [44.5%]) (Table 1). The prevalence of severe PTSSs was 57.4% (296 of 516) for refugees aged 35 years or older and 38.3% (256 of 668) for those younger than 35 years. The rate of severe PTSSs increased with age: 36.9% (103 of 279) for those aged 18 to 24 years; 39.3% (153 of 389) for those aged 25 to 34 years; 58.9% (123 of 209) for those aged 35 to 44 years; 59.4% (107 of 180) for those aged 45 to 54 years; and 52.0% (66 of 127) for those aged 55 years or older. Among married respondents, 480 of 994 (48.3%) had severe PTSSs, while among never-married respondents, 52 of 153 (34.0%) had severe PTSSs. Of the 766 households with more than four family members, 382 (49.9%) had severe PTSSs; of the 62 households with fewer than three members, 22 (35.5%) had severe PTSSs. It was found that 363 of the 751 respondents who had no formal education (48.3%) had severe PTSSs. Employed refugees had a lower prevalence of severe PTSSs (113 of 276 [40.9%]) compared to unemployed participants (430 of 889 [48.4%]). Among the 64 respondents with physical disabilities, 41 (64.1%) had severe PTSSs, while 511 of the 1119 respondents without physical disabilities (45.7%) had severe PTSSs. Additionally, 281 of 496 respondents who reported not receiving adequate humanitarian aid during the last seven days (56.7%) had severe PTSSs. Furthermore, 87 of the 1177 respondents who experienced physical and sexual abuse before displacement (64.0%) reported severe PTSSs.

After dividing traumatic distress into two categories, the unadjusted prevalence ratios (PRs) were calculated (Table 2). The PR of mental health symptoms increased with age; using refugees aged 18 to 24 years as a reference, the PR for those aged 25 to 34 years was 1.22 (95% CI, 1.08-1.39), and the PR for those aged 55 years or older was 1.48 (95% CI, 1.28-1.70). Being male might be a factor associated with severe PTSSs (PR, 1.06; 95% CI, 0.98-1.14). A notable link was observed between receiving adequate humanitarian aid and PTSSs (PR, 0.75; 95% CI, 0.70-0.81). The results show that refugees in Myanmar who experienced both physical and sexual abuse had a higher risk of severe mental health symptoms (PR, 1.59; 95% CI, 1.43-1.76). Additionally, the results indicated that paid employment in the refugee camps was associated with fewer severe mental health symptoms (PR, 0.91; 95% CI, 0.83-1.01).

Multivariable Analysis: Adjusted PRs

Adjusted prevalence ratios (aPRs) were estimated using a multivariable logistic regression model that accounted for potential confounding factors (Table 3). The findings suggest that sufficient humanitarian aid is linked to a reduced risk of traumatic distress symptoms (aPR, 0.50; 95% CI, 0.38-0.65). The results also indicate that experiencing both physical and sexual abuse before displacement was associated with a notable increase in mental health symptoms (aPR, 2.09; 95% CI, 1.41-3.07). After adjustment, the results further suggest that paid employment opportunities in refugee camps were linked to a reduced risk of developing mental health symptoms (aPR, 0.69; 95% CI, 0.52-0.91).

Discussion

Nearly one million Rohingya people live in refugee camps in Bangladesh, with over 750,000 having been there since August 2017. While living conditions in the camps have generally improved compared to two years ago, refugees still reside in small, crowded temporary shelters, often lacking adequate food, clean water, or toilets. Their lives remain uncertain, with an unclear future. These post-migration factors, combined with the traumatic experiences many Rohingya refugees have endured, may contribute to a growing sense of despair. This study provides a detailed look at the symptoms of traumatic distress experienced by Rohingya refugees in Bangladesh.

This study shows that 46.6% of respondents had severe posttraumatic stress symptoms (PTSSs), and 23.1% had moderate PTSSs three years after their forced evacuation. The proportion of diagnosed mental health problems varies across different refugee communities worldwide. For example, among Syrian refugees in Turkey and Iraq, the prevalence of PTSD was 83.4% after about one year in camps and 60% after approximately three years. Among war-affected Ugandans, where two-thirds had been displaced for over five years, 54% experienced mental illness. Similarly, 48.8% of Afghan refugees who fled their country nearly 20 years ago and lived in Australia for one to five years met the criteria for PTSD. In contrast, a small study in 2017 found that 36% of camp-based Rohingya refugees in Bangladesh had PTSSs, and 89% had depression symptoms, while another study of Syrian refugees in Germany reported only 13% with mental health disorders.

Currently, mental health services in Rohingya refugee settings are insufficient, and there are not enough mental health professionals to serve all Rohingya people in need. The high prevalence of PTSSs indicates a need to increase mental health care. This could be achieved by improving the capacity of medical workers in refugee health facilities to diagnose and treat patients with mental disorders.

In Bangladesh, refugees are not legally allowed to work. The inability to survive without jobs has led many refugees, especially men, to seek illegal paid employment. Many Rohingya refugees work in the camps as paid day laborers, helping to build roads, prevent landslides, and clear sewage. A few refugees own small grocery stores. Paid work allows some Rohingya refugees to supplement their families' food rations and access a wider variety of food. Paid employment offers not only the freedom to buy necessary items but also opportunities to work with others. Work provides hope, which is crucial for healing from mental illness. This study's results indicate that those with paid employment are less likely to experience symptoms of traumatic distress compared to unemployed refugees. A meta-analysis found that refugees with fewer economic opportunities experienced poorer mental health outcomes.

Humanitarian aid from governments and non-governmental organizations plays a vital role in the lives of Rohingya refugees, who depend solely on this assistance. This study’s findings indicate that sufficient humanitarian aid to a family is associated with a lower risk of developing mental health symptoms compared to those who received insufficient assistance. Similar studies have also shown that a lack of social support is linked to poorer mental health.

Before the large-scale displacement to Bangladesh, reports extensively documented gender-based abuse against refugees. This study found that physical abuse was more common among male refugees than female refugees. Both physical and sexual abuse were equally prevalent among male and female refugees. The findings indicate that Rohingya refugees who experienced physical and sexual abuse had a high prevalence of severe posttraumatic stress symptoms (PTSSs). Many studies show that women often have a higher rate of mental health disorders after rape or sexual assault.

Approximately half of the study participants were male. The findings indicate that symptoms of traumatic distress were more common among male refugees than female refugees. Several studies have shown that male refugees often experience poorer mental health. A qualitative study involving refugees from Afghanistan suggested that a lack of job opportunities significantly contributes to mental distress among men after years of forced migration. In conservative Rohingya society, women often do not have enough paid jobs, making it difficult for them to earn a living. This lack of paid employment may be linked to an increase in depression among women. The research also indicates that refugees with physical disabilities faced a much higher risk of elevated PTSD symptoms compared to those without disabilities. This finding aligns with previous studies showing a link between disability and a high prevalence of mental illness.

In the study sample, the risk of traumatic distress increased with age, regardless of sex. These findings are consistent with studies conducted among Syrian and Sudanese refugees. The research also found that refugees who could not read or write had a higher prevalence of severe mental health symptoms than those who had received schooling. Currently, adult refugees in the camps lack educational opportunities, and job placement is not based on education. Therefore, education has little importance for adults in the refugee camps. This finding is similar to previous research showing a correlation between poor education and higher rates of PTSD.

Married respondents were also more likely to experience posttraumatic stress symptoms (PTSSs) than those who had never married. This finding is consistent with previous research. The study also found a high risk of PTSSs for refugees with more than four family members. The Rohingya population believes that family planning methods conflict with their faith. As a result, married individuals are more likely to have large families and may struggle to afford enough food or meet other needs. This explains why marital status and family size are positively linked to traumatic distress.

Strengths and Limitations

This research has several strengths. First, a local non-governmental organization that has worked with Rohingya refugees for over three years assisted in collecting data from eight camps. The involvement of interviewers from Rohingya refugee camps helped reduce information bias. Second, households were recruited using a random sampling technique. Third, the study collected data on various factors related to mental well-being that are not typically included in current registries or large-scale population monitoring efforts.

Several limitations should be considered when interpreting the results. First, the research used a scale to measure posttraumatic stress symptoms (PTSSs) that has not been validated for use within the Rohingya community. Second, the cross-sectional survey method, due to its short duration, may not capture the changing nature of the population. Third, the study did not assess the availability and use of mental health treatment during the past two years of the post-migration period. This limitation could be significant because access to or use of mental health services might reduce the prevalence of PTSSs among respondents. Fourth, detailed information was not gathered on the number of assaults experienced by respondents or how many family members were involved in these events. A question about the loss of family members during the attack was initially included, but some respondents became too emotional to answer, and several misunderstood the question about the concept of immediate family members. This question was later removed from the study. Similarly, many respondents were concerned about their health but were uncertain about chronic diseases because they had not visited health facilities, preventing them from answering this question. Several post-migration stressors were not included, such as the refugee camp climate, access to healthcare or education, variety of food, and limited access to certain services. These variables might have confounding effects on mental health.

Conclusions

The high number of self-reported mental health symptoms among Rohingya refugees in this study shows that the trauma of displacement and the violent impact of military operations are still present. Various support services, such as access to education or training in stress management for their traumatic memories, could help reduce the burden of severe mental health symptoms. This study's findings highlight the importance of ensuring that every household receives enough humanitarian assistance. Employment opportunities in Rohingya refugee camp settings show promise as a possible way to reduce the burden of mental health symptoms.

Summary

The Rohingya people in Myanmar are among the world's most persecuted groups. In 1982, the Citizenship Act took away their official recognition as an ethnic group, denying them basic rights like citizenship, free movement, healthcare, education, marriage registration, and the right to vote. This made them the world's largest stateless population. In August 2017, the Myanmar army started a large operation in Rakhine, where about 1.2 million Rohingya lived. Around 7,800 Rohingya were killed, forcing about 750,000 to flee to Bangladesh's Cox’s Bazar district.

Studies have documented the abuse Rohingya refugees faced before being displaced. They were denied medical care for physical or sexual assault in Myanmar. Emotional and verbal abuse were also common. Rohingya community leaders confirmed these accounts. These experiences of physical, sexual, and emotional abuse had lasting effects on mental health and were linked to conditions like posttraumatic stress disorder (PTSD).

PTSD involves symptoms such as intrusive thoughts, nightmares, traumatic memories, avoiding things that remind one of the trauma, being overly watchful, and trouble sleeping. These issues significantly affect a person's psychological well-being, work, and relationships. Individuals with PTSD can have strong reactions to triggers, leading to hallucinations, severe anxiety, and a desire to flee or fight. Symptoms may also include emotional numbness, reduced participation in daily life, and, in severe cases, feeling isolated from others. Compared to those without PTSD, individuals with PTSD are 2 to 4 times more likely to experience depression, anxiety, and substance use. PTSD can also increase the risk of suicide attempts. The Impact of Event Scale–Revised (IES-R) was used to assess posttraumatic stress symptoms (PTSSs). This self-report tool covers all three PTSD symptom groups: interference, avoidance, and hyperarousal, related to a specific life-threatening event.

Refugees faced many stressful situations before and after displacement. Those who experienced abuse were at a higher risk for mental health problems.

Research has detailed the difficulties faced by Rohingya refugees in Bangladesh. They live in cramped, temporary shelters in refugee camps without enough food, clean water, or toilets. Their lives are on hold, and their future is uncertain. A 2017 study found that daily environmental stressors contributed to worse mental health among 148 Rohingya adults. While basic needs and healthcare have generally improved compared to two years ago, post-displacement factors remain crucial for improving the mental health of Rohingya refugees.

Mental health screening is effective only when combined with high-quality services. A challenge in providing adequate services for Rohingya refugees in Bangladesh is the lack of statistical information on their mental health status. This report aims to determine how common PTSSs are and identify the factors before and after displacement that contribute to these symptoms among adult Rohingya living in Bangladesh after the large-scale military operation.

Methods

Study Design and Participants

From September 17, 2019, to January 11, 2020, a survey was conducted among Rohingya refugees living in Cox’s Bazar, Bangladesh. Over a million Rohingya refugees currently reside in Bangladesh, with most concentrated in Ukhia and Teknaf, two administrative regions. Kutupalong, in Ukhia, is the world's largest refugee settlement, housing over 600,000 refugees. Participants were chosen from Kutupalong using a three-stage sampling method. First, eight camps were randomly selected from the 23 camps and expansion areas of Kutupalong, assuming each camp had a similar population size. Second, at least 160 households were chosen from each selected camp, aiming for a total of 1280 households. More households were targeted than the required sample size to ensure 80% power, a 95% confidence interval of 0.05 to 1.96, assuming 40% of the population had a mental illness, and a design effect of 2. A systematic sampling technique was used, starting with a randomly chosen household near the camp's geographical center, and then proceeding to the nearest household until 160 households were sampled. Third, one adult respondent was interviewed per household, preferably the head of the household. If the male head was unavailable, the female head or another available adult family member was interviewed. Household members were defined as those who had lived under the same roof for at least one month and shared cooking and eating facilities from the same source. Participants were also required to have lived in the camp for at least two years after displacement. Details of the sampling allocation are available in eTable 1 of the Supplement. Participants gave verbal consent because they were reluctant to sign or use fingerprints on paper, and most were unable to read or write. They were assured that all collected information would be kept confidential and used only for research purposes. They also received a consent paper with contact information for the research investigators for future inquiries. The institutional review board at North South University, Bangladesh, approved the study, which followed the STROBE reporting guideline.

Recruitment and Training

A team of four enumerators, two men and two women from the Health Management BD (HMBD) Foundation, collected and cleaned the data. HMBD Foundation, a local non-governmental organization working in Rohingya refugee camps since the initial influx, selected local community leaders, called "maji," from each of the eight chosen camps. These local leaders were informed about the study's research and ethical considerations. They helped HMBD Foundation recruit eight local male residents from each camp who could speak both Bengali (Bangla) and Rohingya. A team of data collectors was then formed, consisting of one person from the camp and one from the HMBD Foundation. Interviews were conducted in both Rohingya and Bangla: the camp data collector asked questions in Rohingya, and the HMBD Foundation member checked the answers by asking the same questions in Bangla. Two research investigators from North South University organized a one-day practical training session on ethics and data collection. The enumerators and data collectors were also briefed on the study's objectives, methods, and questionnaire. Additionally, researchers taught data collectors how to build rapport, maintain neutrality, and handle ethical concerns, privacy issues, cultural awareness, and mental health risk management. After the training, a pilot study was conducted with the eight study teams, evaluated as a single unit. The purpose was to observe their ability to understand techniques and manage difficult situations during interviews. Necessary corrections were made after the pilot study. Afterward, each trained team visited their assigned camp together to collect data using a semi-structured questionnaire.

Data Collection

Respondents were clearly informed that participating in the study was entirely voluntary. Face-to-face interviews were conducted individually to ensure privacy. Respondents did not receive any monetary or food incentives. During the interview, questions were read aloud one by one, and respondents were asked to choose the most appropriate answer from a scale. Co-investigators reviewed the data collection sheets for completeness, accuracy, and internal consistency, which the principal investigator confirmed.

Sociodemographic Variables

The first part of the questionnaire collected information on age (in years), sex (male or female), height (in inches), weight (in kilograms), marital status (never married, widowed or divorced, and married), and the number of family members in a household. Later, family members were grouped into three categories: two or fewer, three to four, and more than four members. Educational level was divided into two groups: those who attended school and those who could not read or write.

Predisplacement Abuse

The term "abuse" here refers to violence intended to control or dominate Rohingya people, causing anger, harm, resentment, humiliation, coercion, and helplessness. Data was gathered on the types of abuse refugees experienced before being forced to flee to Bangladesh. Categories included indirect abuse or no exposure to abuse, verbal or emotional abuse, physical abuse, and both physical and sexual abuse. Verbal or emotional abuse was defined as any non-physical act that harms an individual's psychological well-being, such as coercion, defamation, verbal insults, or harassment. This could also include forced labor, or witnessing or being separated from family or community members who were abused. Physical abuse involved any force applied to the body, like shaking, burning, scalding, choking, hair pulling, hitting, slapping, kicking, or threatening or attacking with weapons. It might also include being restrained, tied up, or locked against one's will. Sexual abuse encompassed sexual humiliation (forced masturbation or nudity), sexual slavery, rape (vaginal, oral, anal, or attempted), genital abuse (beatings, electric shock, or mutilation), castration, penis amputation, sterilization, or forced marriage, cohabitation, or sexual activity (with a stranger, family member, or corpse). To distinguish between physical and sexual abuse, respondents were carefully asked if the specific event involved sexual organs. Because Rohingya women were reluctant to answer questions about sexual abuse as a separate option, it was combined with physical abuse. Another response option was for refugees who did not experience direct abuse but suffered grief from losing family members or property, or experienced separation, anxiety, or other trauma.

Postdisplacement Factors

Many Rohingya refugees worked in the camp, often as day laborers performing essential infrastructure tasks. This paid volunteer work was crucial for building roads, preventing landslides, and clearing sewage, in addition to supporting outreach and coordination in the sprawling camps. Employment status was categorized as either employed (paid work) or unemployed. Based on their individual perception, refugees were also asked if they received enough humanitarian aid for their family. Additionally, data was collected on whether a stateless refugee had any existing physical disability (e.g., deafness, blindness, or amputation).

Outcome Measurement: PTSSs

The study used the IES-R, a scale with established psychometric properties, to measure the severity of PTSSs in the study population. This is a brief, 22-item self-report questionnaire. It is not a diagnostic tool for PTSD but effectively measures a person's subjective response to a specific traumatic event. The IES-R includes three symptom clusters: hyperarousal, interference, and avoidance. Respondents were asked to describe a particular event and then indicate how much that event had bothered them in the past seven days. The total subjective stress IES-R score (ranging from 0 to 88) was calculated from the three subscales. The total scale's Cronbach's alpha was 0.87 (95% CI, 0.86-0.88), showing high reliability. Posttraumatic stress symptoms were classified into four levels: no symptoms (total IES-R score, ≤23), mild symptoms (score, 24-32), moderate symptoms (score, 33-38), and severe symptoms (score, ≥39). The Hartford Institute for Geriatric Nursing provided the scoring and interpretation guidelines for the IES-R. Furthermore, the total IES-R score was divided into two groups, following Creamer et al.'s recommendation: scores below 33 indicated no or minimal PTSD symptoms, and scores of 33 or above indicated moderate or severe PTSD symptoms.

Statistical Analysis

Data analysis was carried out using R, version 3.6.2 (R Project for Statistical Computing). The questionnaire, R scripts, and data are available online. Descriptive statistics were used to assess all categorical variables, which are presented as frequencies and percentages. Prevalence ratios (PRs) and their 95% CIs were estimated using a multivariable logistic model, after adjusting for potential confounding factors. The PR represents the ratio of the likelihood of an outcome in the exposed group compared to the likelihood in the unexposed group. Standard errors for PRs were obtained using the delta method. In the adjusted model, demographic factors such as age, sex, educational level, marital status, history of predisplacement abuse, current paid employment status in refugee camps, number of family members, and self-reported sufficient humanitarian aid for the family were controlled for. Variance inflation factors were also obtained in the logistic regression model to check for potential multicollinearity (eTable 2 in the Supplement).

The pattern of missing data in the study sample is presented in eTable 3 in the Supplement. The proportion of missing data ranged from 0.01% (1 of 1183) for family size to 12.5% (148 of 1036) for body mass index (BMI; calculated as weight in kilograms divided by height in meters squared). Missing data for covariates for 32 participants were not included in the multivariable analysis.

Results

Response Rate

Out of 1280 sampled households, 17 were excluded because the head of the household did not agree to participate. An additional 57 households were excluded because no eligible individuals were available during the study period. Ultimately, 1184 households were included in the analysis, resulting in a 92.5% response rate. A detailed calculation of the response rate is provided in the eAppendix of the Supplement.

Characteristics of the Participants

Table 1 shows the sociodemographic, predisplacement, and postdisplacement factors of participants, categorized by the severity of PTSSs. Of the 1184 Rohingya refugees who participated, 625 (52.8%) were men, and 559 (47.2%) were women. The average age of respondents was 35.1 years, with a standard deviation of 13.4 years. Among 1036 respondents, 509 (49.1%) had a normal BMI between 18.5 and 24.9, and 431 (41.6%) were overweight or obese. A total of 994 of 1182 respondents (84.1%) were married, and 766 of 1182 respondents (64.8%) lived in households with more than four family members. Additionally, 751 of 1156 respondents (65.0%) had not attended school and could not read or write. While refugees in Bangladesh are not legally permitted to work, 276 of 1165 respondents (23.7%) reported having paid employment opportunities in the refugee camps. It was found that 136 of 1177 respondents (11.6%) experienced predisplacement physical and sexual abuse in Myanmar. Sixty-four of 1183 respondents (5.4%) had physical disabilities. Furthermore, 496 of 1182 respondents (42.0%) reported not receiving adequate humanitarian aid for their family during the previous seven days.

Moreover, the study examined the relationship between postdisplacement factors and demographic variables. It appeared that Rohingya women had fewer opportunities for paid employment compared to men. Individuals younger than 45 years were more likely to be in paid employment. Also, adults younger than 35 years reported experiencing more physical and sexual abuse than older age groups. Rohingya adults with more than four family members reported a lack of sufficient aid from humanitarian agencies.

Prevalence of PTSSs and Unadjusted PRs

Among the 1184 adult Rohingya refugees, 552 (46.6%) experienced severe PTSSs, and 274 (23.1%) had moderate PTSSs. Severe mental health symptoms were more common in male refugees (303 of 625 [48.5%]) than in female refugees (249 of 559 [44.5%]) (Table 1). The prevalence of severe PTSSs was 57.4% (296 of 516) for refugees aged 35 years or older and 38.3% (256 of 668) for those younger than 35 years. The occurrence of severe PTSSs increased with age: 36.9% (103 of 279) for those aged 18 to 24 years; 39.3% (153 of 389) for those aged 25 to 34 years; 58.9% (123 of 209) for those aged 35 to 44 years; 59.4% (107 to 180) for those aged 45 to 54 years; and 52.0% (66 of 127) for those aged 55 years or older. Among married respondents, 480 of 994 (48.3%) had severe PTSSs, while among never-married respondents, 52 of 153 (34.0%) had severe PTSSs. In households with more than four family members, 382 of 766 (49.9%) had severe PTSSs; in households with fewer than three members, 22 of 62 (35.5%) had severe PTSSs. Among the 751 respondents without formal education, 363 (48.3%) had severe PTSSs. Employed refugees had a lower prevalence of severe PTSSs (113 of 276 [40.9%]) compared to unemployed participants (430 of 889 [48.4%]). Of the 64 respondents with physical disabilities, 41 (64.1%) had severe PTSSs, while 511 of 1119 respondents without physical disabilities (45.7%) had severe PTSSs. Additionally, 281 of 496 respondents who did not receive adequate humanitarian aid in the past seven days (56.7%) reported severe PTSSs. Furthermore, 87 of 1177 respondents who experienced physical and sexual abuse before displacement (64.0%) reported severe PTSSs.

After categorizing traumatic distress into two groups, the unadjusted prevalence ratios (PRs) were calculated (Table 2). The PR for mental health symptoms increased with age; using refugees aged 18 to 24 years as a reference, the PR for those aged 25 to 34 years was 1.22 (95% CI, 1.08-1.39), and for those aged 55 years or older, it was 1.48 (95% CI, 1.28-1.70). Being male might be a factor associated with severe PTSSs (PR, 1.06; 95% CI, 0.98-1.14). A significant link was found between receiving adequate humanitarian aid and PTSSs (PR, 0.75; 95% CI, 0.70-0.81). The results indicated that refugees in Myanmar who experienced both physical and sexual abuse had a higher risk of severe mental health symptoms (PR, 1.59; 95% CI, 1.43-1.76). Also, the findings suggested that paid employment in refugee camps was linked to fewer severe mental health symptoms (PR, 0.91; 95% CI, 0.83-1.01).

Multivariable Analysis: Adjusted PRs

Adjusted prevalence ratios (aPRs) were calculated using a multivariable logistic regression model, with adjustments for potential confounding factors (Table 3). The findings suggest that sufficient humanitarian aid is linked to a reduced risk of traumatic distress symptoms (aPR, 0.50; 95% CI, 0.38-0.65). The results also indicate that experiencing both physical and sexual abuse before displacement was significantly associated with an increase in mental health symptoms (aPR, 2.09; 95% CI, 1.41-3.07). After adjusting for other factors, the results also showed that paid employment opportunities in refugee camps were associated with a reduced risk of developing mental health symptoms (aPR, 0.69; 95% CI, 0.52-0.91).

Discussion

Nearly a million Rohingya people live in Bangladeshi refugee camps, with over 750,000 having been there since August 2017. Although living conditions in the camps have generally improved compared to two years ago, refugees still reside in small, overcrowded temporary shelters, lacking sufficient food, clean water, or toilets. Their lives remain stagnant, and their future is uncertain. These post-migration factors, combined with the traumatic experiences many Rohingya refugees have endured, may contribute to growing despair. This study provides a detailed look at the symptoms of traumatic distress experienced by Rohingya refugees in Bangladesh.

The study found that 46.6% of respondents had severe PTSSs, and 23.1% had moderate PTSSs in the third year after their forced evacuation. The proportion of diagnosed mental health problems varies across refugee communities globally. For example, the prevalence of PTSD among Syrian refugees living in Turkey and Iraq was 83.4% after about one year in camps and 60% after about three years. Among the war-affected Ugandan population, two-thirds of whom were displaced for over five years, the prevalence of mental illness was 54%. Similarly, 48.8% of Afghan refugees who fled their country nearly 20 years ago and have lived in Australia for one to five years met the criteria for PTSD. However, a small study in 2017 among camp-based Rohingya refugees in Bangladesh found that 36% of adults had PTSSs and 89% had symptoms of depression.

Current mental health services in Rohingya refugee settings are insufficient, with too few mental health professionals to serve the entire population in need. The high prevalence of PTSSs indicates a need to increase mental healthcare, which could be achieved by expanding the capacity of medical workers in refugee health facilities to diagnose and treat mental disorders.

In Bangladesh, refugees are not legally allowed to work. The inability to survive without jobs has led many refugees, especially men, to seek illegal paid employment. Many Rohingya refugees work in the camps, primarily as paid day laborers who help build roads, prevent landslides, and clear sewage. A few also operate small grocery stores. Paid work allows some Rohingya refugees to supplement their families' food rations and access a variety of food. Employment offers not only the freedom to buy necessities but also an opportunity to work with others. Work provides hope, which is vital for recovering from mental illness. The findings indicate that those with paid employment are less likely to experience symptoms of traumatic distress than unemployed refugees. A meta-analysis similarly found worse mental health outcomes for refugees with fewer economic opportunities.

Humanitarian aid from governments and NGOs is crucial for Rohingya refugees. They depend entirely on this assistance. The study findings show that receiving enough humanitarian aid is linked to a lower risk of developing mental health symptoms, compared to those who received insufficient help. Similar research has found that a lack of social support is associated with poorer mental health.

Before the mass migration to Bangladesh, sex-based abuse against refugees was widely documented. The study found that physical abuse was more common among male refugees than female refugees, and both physical and sexual abuse were equally prevalent among both sexes. Findings indicate that Rohingya refugees who experienced physical and sexual abuse had a high prevalence of severe PTSSs. Many studies show that women are more likely to develop mental health disorders after rape or sexual assault.

Approximately half of the study participants were male. The findings show that symptoms of traumatic distress were more common among male refugees than female refugees. Several studies have also shown that male refugees experienced a higher prevalence of poor mental health. A qualitative study among refugees from Afghanistan suggested that a lack of job opportunities significantly contributes to mental distress among men after years of forced migration. Rohingya women have limited access to paid jobs in their conservative society, making it difficult for them to earn a living. A lack of paid employment may be linked to an increase in depression among women. The research also indicates that refugees with physical disabilities faced a much higher risk of elevated PTSD symptoms compared to those without disabilities. This finding aligns with previous studies showing that disability is associated with a high prevalence of mental illness.

In the study sample, the risk of traumatic distress increased with age for both sexes. These findings are consistent with studies conducted among Syrian and Sudanese refugees. The research also found that refugees who could not read or write had a higher prevalence of severe mental health symptoms than those who had attended school. Currently, adult refugees in the camps do not have educational opportunities, and job placement is not based on education. Therefore, education holds little significance for adults in the refugee camps. This finding aligns with previous research indicating a correlation between poor education and higher rates of PTSD.

Married respondents were more likely to experience PTSSs than those who were never married. This finding is similar to results from previous research. The study also found a high risk of PTSSs for refugees in families with more than four members. The Rohingya population believes that family planning methods conflict with their faith. Thus, married individuals are more likely to have large families and may struggle to afford enough food or meet other needs; this explains why marital status and family size are positively associated with traumatic distress.

Strengths and Limitations

The research has several strengths. First, a local NGO that has worked with Rohingya refugees for over three years helped collect data from the eight camps. The involvement of interviewers from Rohingya refugee camps reduced information bias. Second, households were recruited using a random sampling technique. Third, to assess mental well-being, data was collected on various factors not included in current registries or large-scale population monitoring efforts.

Several limitations should be considered when interpreting the results. First, the research used a scale to measure PTSSs that had not been validated for use within the Rohingya community. Second, the cross-sectional survey's short duration might not fully capture the population's transient nature. Third, the study did not assess the availability and use of mental health treatment during the past two years of the post-migration period. This limitation is significant because accessing or using mental health services could potentially reduce the prevalence of PTSSs among respondents. Fourth, detailed information was not gathered on the number of assaults experienced by respondents or the number of family members involved in those events. A question was included about the loss of family members during the attack, but some respondents became too emotional to answer, and several misunderstood the question about the concept of immediate family members. These questions were eventually removed from the study. Similarly, many respondents were concerned about their health but were unsure about chronic diseases because they had not visited any health facilities, making them unable to answer this question. Several post-migration stressors, such as the refugee camp climate, access to healthcare or education, food variety, and limited access to certain services, were not included. These variables might have confounding effects on mental health.

Conclusions

The high prevalence of self-reported mental health symptoms among Rohingya refugees in this study indicates that the trauma of displacement and the severe consequences of military operations are still present. Various support services, such as access to education or training in stress management for their traumatic memories, could help reduce the burden of severe mental health symptoms. The findings highlight the importance of ensuring that every household receives adequate humanitarian assistance. Employment opportunities in Rohingya refugee camp settings show promise as a potential intervention to reduce the burden of mental health symptoms.

Summary

The Rohingya people in Myanmar are treated very poorly. In 1982, a law took away their rights, like being citizens, moving freely, getting healthcare and schooling, getting married, and voting. This made them the largest group of people in the world without a country. In 2017, Myanmar's army attacked, killing about 7,800 Rohingya. This forced about 750,000 Rohingya people to leave their homes and go to Bangladesh.

Before they left Myanmar, many Rohingya people faced abuse. They could not get medical help after being hurt or sexually attacked. Many also faced hurtful words and emotional abuse. These bad experiences deeply affected their minds and could lead to problems like post-traumatic stress disorder, or PTSD.

PTSD is a mental health problem where people have bad thoughts, nightmares, and memories of terrible past events. They might avoid things that remind them of the trauma, be easily scared, and have trouble sleeping. These issues can make it hard to live a normal life, work, and get along with others. People with PTSD might feel numb, pull away from others, and are more likely to have other mental health problems or try to harm themselves.

Refugees faced many stressful events before and after leaving their homes. Those who were abused were more likely to have mental health problems.

Life in the refugee camps in Bangladesh is hard. People live in small, crowded shelters without enough food, clean water, or toilets. They do not know what will happen in the future. While some basic needs have gotten better, the ongoing problems in the camps still affect the refugees' mental health.

It is important to check for mental health problems and offer good care. But there is not enough information about the mental health of Rohingya refugees in Bangladesh. This report aimed to find out how many Rohingya adults in Bangladesh have PTSD-like symptoms and what factors before and after leaving their homes are linked to these symptoms.

How the Study Was Done

Study Plan and People

Researchers did a study from September 2019 to January 2020 with Rohingya refugees in Bangladesh. Over a million Rohingya refugees live there, mostly in certain areas. The study included people from Kutupalong, the biggest refugee camp. Researchers picked 8 camps out of 23. Then, they chose about 160 homes from each camp, aiming for 1,280 homes. They picked homes by starting in the middle of each camp and moving to the next closest one. Only one adult from each home was interviewed, usually the head of the household. The person had to have lived in the camp for at least two years.

People gave their spoken permission to be part of the study because many did not want to sign papers. They were told that their information would be kept private and only used for the study. The study was approved by a special board in Bangladesh.

Getting and Training People

A team of four people, two men and two women, helped collect and clean the study information. They worked with a local group that helps Rohingya refugees. This group chose local leaders from each camp, who then helped find 8 local men from each camp who could speak both Rohingya and Bengali. A team was made with one person from the camp and one from the local group. The person from the camp asked questions in Rohingya, and the other person checked the answers in Bengali.

Researchers trained these teams for one day on how to collect information and be fair. They also taught them about being kind, keeping information private, and helping people who might be upset. After the training, they did a test run to make sure everyone understood. Then, the trained teams went to their camps to collect the information using a set of questions.

Collecting Information

People were told they did not have to be part of the study if they did not want to. Each person was interviewed alone to keep things private. They did not get money or food for taking part. The questions were read out loud, and people chose the answer that fit them best. The researchers checked the collected information to make sure it was complete and correct.

Personal Details

The first part of the questions asked about things like age, gender, height, weight, if they were married, and how many people lived in their home. The number of family members was later put into groups: 2 or less, 3 to 4, and more than 4. Education was split into two groups: those who went to school and those who could not read or write.

Abuse Before Leaving Home

"Abuse" here means violence used to control or harm Rohingya people, making them angry, hurt, ashamed, or helpless. The study asked about different types of abuse people faced before they had to leave Myanmar. These types included no abuse, hurtful words or emotional abuse, physical abuse, and both physical and sexual abuse.

Hurtful words or emotional abuse meant things that hurt someone's mind, like threats, insults, or forcing them to work, or seeing family members get hurt. Physical abuse included hitting, kicking, burning, choking, or attacking with weapons. Sexual abuse included forced sexual acts or harm to sexual body parts. Researchers were careful when asking about sexual abuse, especially for women, and combined physical and sexual abuse as one option to make it easier for people to answer. Another option was for those who did not experience direct abuse but were sad from losing family or property.

Factors After Leaving Home

Many Rohingya refugees worked in the camps, often as day laborers building roads or clearing drains. This study looked at whether people had paid jobs or not. It also asked if people felt they were getting enough help (like food and supplies) for their families. Researchers also collected information on whether anyone had a physical disability, like being deaf, blind, or missing a limb.

Measuring PTSD-like Symptoms

The study used a special survey called the IES-R to measure how strong PTSD-like symptoms were. This survey has 22 questions and helps understand how much a person is bothered by a past bad event. It asks about three main types of symptoms: being easily upset, avoiding things, and being overly watchful. People were asked to think about a specific bad event and how much it bothered them in the past week. A total score was given, ranging from 0 to 88. Higher scores mean more symptoms.

Symptoms were put into four groups: no symptoms (score 23 or less), mild (24-32), moderate (33-38), and severe (39 or more). For some parts of the study, scores were also split into two groups: less than 33 (no or very few symptoms) and 33 or more (moderate or severe symptoms).

Analyzing the Numbers

The study used computer programs to look at the collected information. They looked at how often different things happened (like how many people were male or female). They also looked at the chances of having PTSD-like symptoms based on different factors, like age or past abuse. They used special math to make sure the results were fair, considering other factors like age, education, and job status.

Some information was missing, but researchers made sure this did not affect the main findings too much.

Results

How Many People Answered

Out of 1,280 homes chosen, 17 did not want to join, and 57 had no eligible people. In the end, 1,184 homes were included in the study. This means 92.5% of the chosen homes took part.

Details About the People

Of the 1,184 Rohingya refugees in the study, 625 (52.8%) were men and 559 (47.2%) were women. The average age was about 35 years old. Almost half (49.1%) had a normal weight, and many (41.6%) were overweight. Most (84.1%) were married, and most (64.8%) had more than 4 family members. Many (65.0%) could not read or write. While refugees are not allowed to work legally in Bangladesh, 23.7% reported having paid jobs in the camps.

About 11.6% of people said they experienced both physical and sexual abuse before leaving Myanmar. A small number (5.4%) had physical disabilities. Also, 42.0% felt they did not get enough help (like food) for their families.

The study also found that women did not get as many paid jobs as men. Younger adults were more likely to have paid jobs and to have experienced physical and sexual abuse. Families with more than 4 members often said they did not get enough help from aid groups.

How Many Had PTSD-like Symptoms and Related Factors

Of the 1,184 adult Rohingya refugees, 552 (46.6%) had severe PTSD-like symptoms, and 274 (23.1%) had moderate symptoms. More men (48.5%) than women (44.5%) had severe symptoms. Severe symptoms were more common in older refugees (57.4% for those 35 or older) than younger ones (38.3% for those younger than 35). The number of severe symptoms generally went up with age.

Among married people, 48.3% had severe symptoms, compared to 34.0% of those who were never married. Homes with more than 4 family members had more severe symptoms (49.9%) than homes with fewer members (35.5%). Of those who could not read or write, 48.3% had severe symptoms. People with paid jobs had fewer severe symptoms (40.9%) than those without jobs (48.4%). Among those with physical disabilities, 64.1% had severe symptoms, compared to 45.7% of those without disabilities. Also, 56.7% of those who felt they did not get enough aid had severe symptoms. A high number (64.0%) of those who faced physical and sexual abuse before leaving Myanmar had severe symptoms.

The chance of having mental health symptoms went up with age. Getting enough aid was linked to fewer symptoms. People who experienced both physical and sexual abuse had a much higher chance of having severe mental health symptoms. Having paid work in the camps was linked to fewer severe mental health symptoms.

Factors That Were Important (After Adjusting for Other Things)

When researchers looked at everything together, they found that getting enough humanitarian aid was linked to a lower chance of having symptoms of trauma. Also, physical and sexual abuse before leaving home was strongly linked to more mental health symptoms. Paid jobs in refugee camps were linked to a lower chance of developing mental health problems.

Discussion

Almost one million Rohingya people live in refugee camps in Bangladesh, most having arrived since 2017. While living conditions have gotten a bit better, refugees still live in crowded, temporary shelters without enough food, clean water, or toilets. They feel stuck and unsure about their future. These ongoing problems, along with the past trauma, make their situation desperate. This study shows the deep mental health impact on Rohingya refugees in Bangladesh.

The study found that 46.6% of people had severe PTSD-like symptoms, and 23.1% had moderate symptoms, even three years after being forced to leave their homes. Other studies in different refugee groups have found varying numbers of mental health problems, sometimes very high. This suggests that many Rohingya refugees need mental health care, and there are not enough trained helpers to go around. More medical workers need to be trained to help diagnose and treat these mental health issues.

In Bangladesh, refugees are not allowed to work legally. This makes many, especially men, look for illegal jobs. Some Rohingya work as paid day laborers in the camps, building roads or cleaning. Some run small shops. These jobs help families get more food and feel a sense of purpose. Work can offer hope, which is important for healing from mental illness. This study found that those with paid jobs were less likely to have symptoms of trauma. Other research also shows that refugees with fewer job chances have worse mental health.

Help from governments or aid groups is very important for Rohingya refugees, as they depend on it. This study found that families who received enough aid had a lower risk of mental health symptoms. This is similar to findings that show a lack of social support can lead to poorer mental health.

Abuse, especially sexual abuse, towards refugees before they left Bangladesh has been widely reported. This study found that physical abuse was more common for men, and both physical and sexual abuse were equally common for men and women. People who faced physical and sexual abuse had many severe PTSD-like symptoms. Other studies have shown that women often have more mental health problems after sexual assault.

About half of the people in this study were men. The study showed that men had more symptoms of trauma than women. Other studies have also found that men refugees often have worse mental health. One idea is that not having jobs plays a big role in mental distress for men after being forced to move. Rohingya women in their culture may not get as many paid jobs, which also makes it hard for them to earn a living and can lead to more sadness. The study also found that refugees with physical disabilities were much more likely to have PTSD symptoms, which matches other research.

In this study, the risk of trauma symptoms went up with age for both men and women, similar to studies on other refugee groups. The study also found that refugees who could not read or write had more severe mental health symptoms. There are no learning opportunities for adults in the camps, and education does not help them get jobs, so it may feel useless. This also matches past research that links low education to higher rates of PTSD.

Married people in the study were more likely to have PTSD-like symptoms than those who were never married, which is like other research. The study also found a high risk of PTSD symptoms for refugees in large families (more than 4 members). Rohingya people often have many children due to their beliefs, which can make it hard to afford enough food and other needs. This explains why being married and having a large family are linked to trauma symptoms.

Good Points and Things to Consider

This study had some good points. First, a local aid group that had worked with Rohingya refugees for over three years helped collect the information from 8 camps. Having interviewers from the Rohingya camps helped get more accurate information. Second, homes were chosen randomly. Third, the study collected information on many different things that were not usually tracked, which helped understand mental health.

However, there were some things to consider when looking at the results. First, the survey used to measure PTSD-like symptoms had not been fully checked to make sure it worked perfectly for the Rohingya community. Second, this type of short study might not fully capture how people's mental health changes over time. Third, the study did not look at whether people got mental health treatment in the last two years, which could have changed their symptoms. Fourth, researchers did not collect details on how many times people were attacked or how many family members were involved. Some people got too upset to answer these questions, and some misunderstood questions about losing family. So, these questions were stopped later in the study. Similarly, many people were not sure about long-term illnesses because they had not seen a doctor. The study also did not include other stressful things after leaving home, like the camp weather, access to doctors or schools, food choices, or limited services. These things could also affect mental health.

Conclusion

The many reports of mental health problems among Rohingya refugees in this study show that the pain of being forced from their homes and the violence they faced are still very real. Different types of help, like chances to learn or training to deal with bad memories, could lessen the burden of severe mental health symptoms. The study highlights how important it is for every family to get enough aid. Giving Rohingya refugees chances to work in the camps is a good way to help reduce mental health problems.