Twenty seven years of treating survivors of torture and organized violence - associations between torture, gender and ethnic minority status among refugees referred for treatment of PTSD.

Background: Victims of torture and organized violence are among the most vulnerable refugees. The nature of torture promotes shame, secrecy and silencing of the survivors, and there is a need for studies to provide a testimony to the experiences of survivors. Objective: To contribute to the global knowledge base by exploring the nature and frequency of different types of torture as reported by survivors and to explore potential associations between basic sociodemographic variables and the likelihood of reporting different kinds of torture and of attending treatment for PTSD. Method: This study is based on archival data from 27 years of clinical practice treating survivors of torture and organized violence (N = 1249) in a specialist outpatient clinic. Data was coded from patient files by two research assistants. Statistical analysis included independent samples t-tests, Pearson Chi Square tests and regression.
Results: The study found strong associations between gender of survivors and the reporting of different kinds of torture, most strongly with respect to sexual torture. Furthermore, the study found significant differences between survivors belonging to an ethnic minority and survivors who belonged to the majority populations within their countries of origin with regards to the extent of the torture, they report being subjected to. Patients who received treatment reported more torture experiences than those who did not. Conclusions: Findings from the present study document the experiences of different torture methods as reported by survivors, and suggest that belonging to an ethnic minority population is associated with an increased risk of being subjected to more severe torture, which has implications for both research and interventions aimed at treating survivors of torture.

Introduction

According to the United Nations (UN), torture is defined as an event when the following three criteria are fulfilled: (i) strong pain or suffering is inflicted, either physically or mentally; (ii) forced confessions, information, or punishments are sought; and (iii) a public authority carries out, encourages or consents to the use of torture. Torture is often used as a political instrument to help a country’s rulers keep control by breaking down the personality of individuals. Torture methods can be both physical and mental, and these are equally detrimental. Often, methods that do not create permanent, visible damage are used in order to make it difficult to document torture (Sjölund, Kastrup, Montgomery, & Persson, 2009). Victims of torture and organized violence are among the most vulnerable refugees (Kalt, Hossain, Kiss, & Zimmerman, 2013; McColl et al., 2010; Quiroga & Jaranson, 2005; Sigvardsdotter, Vaez, Hedman, & Saboonchi, 2016). Studies consistently show that being a victim of torture increases the risk of long-term mental and physical health problems (McColl et al., 2010; Olsen, Montgomery, Bøjholm, & Foldspang, 2007). For example, refugees who report experiences of torture are four times more likely to suffer from post-traumatic stress disorder (PTSD) than other refugees, and 2.5 times more likely to suffer from depression (Steel et al., 2009). In addition to the individual physical and psychological pain and suffering, survivors who develop PTSD are at increased risk of experiencing substance abuse problems (Jacobsen, Southwick, & Kosten, 2001), problems in interpersonal relationships (Weine et al., 2004) and family violence (Timshel, Montgomery, & Dalgaard, 2017). Furthermore, a number of studies suggest that traumatic experiences of torture and organized violence may have a negative impact on the offspring of survivors, even when the children are born in the host country (Dalgaard, Todd, Daniel, & Montgomery, 2016; De Haene, Grietens, & Verschueren, 2010; Kellermann, 2001; Lambert, Holzer, & Hasbun, 2014).

Shame, secrecy and denial

One of the purposes of torture is to dehumanize and humiliate the victim by causing fear and suffering (Gorman, 2001). Survivors of torture often experience shame following the experience, and many victims are afraid to reveal their experiences even within the context of treatment for PTSD. Retelling the experiences may cause a feeling of re-traumatization and revictimization and this causes many survivors and their families to live in a semi-permanent state of secrecy and denial (Montgomery, 2004). Furthermore, the survivors may lack faith in doctor/patient confidentiality or may feel afraid of making the listener feel uncomfortable if the detailed nature of the experiences is revealed (Gorman, 2001). Therefore, the true prevalence and severity of torture is likely underestimated.

Many survivors are in urgent need of rehabilitation and treatment interventions, however, the complexity and magnitude of the problems with which survivors present constitute an ongoing challenge requiring clinicians to continuously develop their practice and to learn from their clinical experiences (Weiss et al., 2016).

For almost three decades survivors of torture and organized violence with subsequent symptoms of PTSD have been referred for treatment at Dignity Danish Institute Against Torture in Denmark. Patients are initially interviewed by medical doctors and psychologists, and this study is based on data obtained from coding of the patient files between 1982 and 2009. All patients referred to the clinic were experiencing symptoms of psychological distress and PTSD, however, the way in psychological complaints were measured and described in patient files, varies greatly as a reflection of changes in clinical practice over the years, and thus a reliable coding and classification of symptoms was not possible based on the available data. The present project is an attempt to study survivors of torture and organized violence in a historical context, to systematically uncover what we know about the patients based on 27 years of clinical practice. Thus, the present study may be seen as a testimony and a contribution to the knowledge base, which may inform clinicians worldwide when navigating the difficult path of balancing between disclosure and silencing in conversations with potentially traumatized refugees (De Haene, Rober, Adriaenssens, & Verschueren, 2012).

Associations between gender and ethnicity and the type and extent of torture

A review of gender differences in the epidemiology of posttraumatic stress disorder (PTSD) concluded that one of the most consistent findings across the literature is that women face a higher risk of developing this disorder, and that this increased risk may be due to the type of trauma they experience (Olff, Langeland, Draijer, & Gersons, 2007). In line with this, a number of studies show gender differences in the reported rates of being subjected to different types of torture (Masmas et al., 2008; Willard, Rabin, & Lawless, 2014) and that female survivors of torture report more sexual torture than male survivors (Kalt et al., 2013; Lunde & Ortmann, 1990). However, more recent research suggests that male victims of sexual torture are highly unlikely to report the experiences. In homophobic environments, homosexual survivors of rape may be seen as inviting rape by their very nature. In around 70 countries, same-sex relations are criminalized and the taboo on homosexuality probably discourages all men who experience sexual torture from reporting it. In line with this hypothesis, a study based on screening of patient records and interviews with healthcare professionals concludes that sexual torture of men was a “regular, unexceptional component of violence in wartime Croatia, not a rare occurrence“ (Oosterhoff, Zwanikken, & Ketting, 2004)

Globally, ethnic minority populations face a higher risk of experiencing oppression, conflict and persecution within their countries of origin when compared to members of the majority population (Lalani & International, 2010; Lim, Metzler, & Bar-Yam, 2007; Newland, 1993), which suggests that members of ethnic minority populations may also face higher risks of experiencing torture and organized violence.

Aim

The primary aim of this exploratory study was to systematically uncover the characteristics of survivors of torture and organized violence in a historical context, and to explore potential associations between sociodemographic variables and the extent and types of torture that survivors report having been subjected to. Secondly, based on the literature presented in the introduction, we proposed two hypotheses:

There would be substantial gender differences in the extent and types of torture that survivors have been subjected to

That belonging to an ethnic minority within one’s country of origin is associated with having been subjected to more torture

Finally, the study aimed to uncover potential differences between patients who received treatment and patients who dropped out after the initial intake sessions.

Methods

Procedure

This study is based on records of the treatment of patients at Dignity Danish Institute Against Torture between 1982 and 2009. For the data collection process, two research assistants went through patient records and recorded basic socio-demographic variables, along with measures for torture exposure. A sample consisting of 5% of the total were recorded and coded by both assistants, to ensure interrater reliability and uniformity in the process for data entry. The data collection focused on information disclosed in the visitation and preliminary-exam phase of the treatment. The data collection was approved by the Danish Data Protection Agency.

Records of all 2,141 patients who were referred to Dignity Danish Institute Against Torture for treatment in the time period mentioned above were included in the dataset, which means that a segment of those included did not receive treatment beyond the initial referral. This also includes relatives referred to the clinic in connection with a partner’s or parent’s treatment. For the purposes of this study, we include all those patients who underwent treatment at the clinic, as well as those who, through the visitation process, were confirmed as survivors of torture or organized violence, and therefore within the target group, but for various reasons did not receive treatment at the clinic. Of the respondents included in the following analyses, 904 received treatment at the clinic, while another 345 were referred to other treatment options following the visitation process, for a total of 1,249 patients in the final dataset.

Measures

For each patient, data were collected on 10 basic socio-demographic variables, including date of birth, gender, country of origin, education, and marital status. Additionally, the torture exposure for each patient was recorded, both as a dichotomous variable, and separated by the types of torture reported during the preliminary assessments. The variables recording various torture methods were developed based on previous work on the categorization of torture methods (Modvig, Jensen, Nielsen, & Vesterby, 2011), and adjusted to the recording methods used in the original patient records. An overview of the types of torture can be found in Table 3. Finally, as research has indicated that cumulative traumatic experiences are related to more severe PTSD symptoms, a summed variable of the listed torture methods was constructed, to serve as a proxy for the brutality and extent of patients’ torture exposure. While this measure is imperfect, the nature of the retrospective data collection process does not provide a better measure.

Table 3.

Overview, by gender, of the percentage of types of torture reported by torture survivors treated between 1982 and 2009

Types of torture experienced	Percent of total N = 1135	Percent of males N = 978	Percent of females N = 157	P-valuea
Systematic beating	45.7	48.8	26.8	p < .001
Unsystematic beating	91.8	93.1	83.4	p < .001
Positional torture	28.9	30.8	17.2	p < .001
Suspension	42.6	45.6	24.2	p < .001
Suffocation	10.7	10.7	10.2	NS
Thermal torture	21.9	23.3	13.4	p < .010
Cuts and amputation	4.8	4.8	5.1	NS
Electrical torture	37.6	39.8	24.2	p < .001
Deprivation	51.2	52.7	42.0	p < .050
Sexual torture	15.9	14.3	36.1	p < .001
Threats	55.2	55.1	56.1	NS
Humiliations	17.9	17.2	22.3	NS
Witnessed torture	34.4	34.4	35.0	NS
Sensory stress	4.3	4.2	5.1	NS

ap-Values from Pearson Chi Square test.

Statistical analyses

All analyses were conducted with IBM SPSS Statistics 25. Independent samples t-tests were used to examine gender and ethnic differences in age at intake, length of education (in years), and cumulative torture experience. Pearson Chi Square tests examined gender- and ethnic differences in torture exposure. Finally, regression analysis was used to examine the relationship between ethnicity and cumulative torture exposure, along with several socio-demographic controls and the differences between patients who subsequently received treatment at the clinic and those who did not.

Results

Regarding the gender distribution in the clinic, 80% of the patients treated between 1982 and 2009 were male, while only 20% were female. More than half of the patients were married, and the mean age at the beginning of treatment was 37.3 years (SD = 9.1). The patients have received an average of 11.1 years of education, which is within the range of the Danish average, which increased from 8.2 years in 1985 to 12.7 years in 2009.

Table 1 also specifies the socio-demographic break-down for the segments of the respondents that have and have not registered an ethnic minority association. The ethnic minority population makes up about 16% of the respondents in the target group, and they differ from the other patients on several socio-demographic measures. The ethnic minority group has a larger proportion of men, they are younger, and these patients have, on average, received 1.7 fewer years of education than the remainder of the target group. All these differences are statistically significant.

Table 1.

Overview of demographic differences for non-minority and ethnic minority patients treated between 1982 and 2009

Variables	All respondents N/%	Non-minority N/%	Ethnic minority N/%	P-value
Gender	N = 1249	N = 1038	N = 203	p < .001a
Male	1000/80.1%	814/78.4%	183/90.1%
Female	249/19.9%	224/21.6%	20/9.9%
Marital status	N = 1123	N = 935	N = 184
Single	162/13.0%	134/14.3%	27/14.7%
Living with partner	34/2.7%	28/3.0%
Married	727/58.2%	600/64.2%	124/67.4%
Divorced	117/9,4%	100/10.7%	17/9.2%
Widow(er)	28/2.2%	26/2.8%
Married but not living together	55/4.4%	47/5.0%
Age at treatment start, mean (SD)	N = 1249 37.3 (9.1)	N = 1038 37.7 (9.1)	N = 203 35.2 (8.4)	p < .001b
Top 5 countries of origin	N = 1241	N = 1038	N = 203
	Iraq 388/31.3%	Iraq 335/32.3%	Palestine 57/28.1%
	Iran 267/21.5%	Iran 255/24.6%	Iraq 53/26.1%
	Lebanon 103/8.3%	Lebanon 79/7.6%	Turkey 29/14.3%
	Chile 75/6.0%	Chile 75/7.2%	Lebanon 24/11.8%
	Afghanistan 60/4.8%	Afghanistan 60/5.8%	Kurdistan 22/10.8%
Years of education, mean (SD)	N = 933 11.1 (4.1)	N = 771 11.4 (4.0)	N = 161 9.7 (4.6)	p < .001b

ap-Values from Pearson Chi Square test.

bp-Values from independent samples t-test.

While a substantial segment of the ethnic minority population in the target group have reported their country origin simply as ‘Palestine’ or ‘Kurdistan’, Iraq, Turkey, and Lebanon also appear in the top five most reported countries of origin for the ethnic minority subgroup.

Table 2 shows the rate of exposure to torture among the target group. Nearly all the male patients (98.6%) are registered torture survivors, along with a majority of the female patients (70.5%). Additionally, of those that have been exposed to torture, the male patients have, on average, been exposed to more different types of torture.

Table 2.

Overview, by gender, of the torture experienced by torture survivors treated between 1982 and 2009

Variables	All respondents N/%	Male respondents N/%	Female respondents N/%	P-value
Exposed to torture	1202/93.3%	978/98.6%	224/70.5%	p < .001a
Number of registered torture types, mean (SD)	N = 1135 4.6 (2.2)	N = 978 4.8 (2.2)	N = 157 3.9 (2.1)	p < .001b

ap-Values from Pearson Chi Square test.

bp-Values from independent samples t-test.

Table 3 shows the exposure rates of different types of torture in the overall target group and for male and female patients separately. Overall, beatings, threats, deprivation, and suspension are the most commonly reported types of torture among the patients. Notably, the male patients have significantly higher rates of exposure to several of the types of torture, except for sexual torture, where the female patients are more than twice as likely to report exposure. These findings thus confirm the first hypothesis that there are significant gender differences in the frequency and types of torture that patients report have been subjected to.

In addition to gender difference in torture exposure, the data shows a trend towards higher rates of exposure among the ethnic minority populations. While the difference in torture exposure rate seen in Table 4 is not statistically significant, the difference in average number of types of torture experienced is significantly higher for the ethnic minority segment of the patients.

Table 4.

Overview, by minority status, of the torture reported by torture survivors treated between 1982 and 2009

Variables	All respondents N/%	Non-minority N/%	Ethnic minority N/%	P-value
Exposed to torture	1202/93.3%	996/93.3%	198/96.0%	NS
Number of registered torture types, mean (SD)	N = 1135 4.6 (2.2)	N = 941 4.5 (2.2)	N = 192 5.1 (2.4)	p < .005a

ap-Values from independent samples t-test.

Table 5 shows an overview of the exposure rates by types of torture, and for the specified types of torture, the ethnic minority patients are significantly more likely to have been exposed to systematic beating, suspension, and thermal torture. While the data overall indicates that patients belonging to an ethnic minority have been more exposed to torture, this segment of the patient population also contains a lower proportion of female patients, which, as outlined above, have lower overall exposure rates to torture than the male patients.

Table 5.

Overview, by minority status, of the percentage of types of torture reported by torture survivors treated between 1982 and 2009

Types of torture experienced	Percent of total N = 1135	Percent of non-minority N = 941	Percent of ethnic minority N = 192	P-valuea
Systematic beating	45.7	42.6	60.9	p < .001
Unsystematic beating	91.8	92.0	91.1	NS
Positional torture	28.9	27.9	33.9	NS
Suspension	42.6	41.0	50.5	p < .050
Suffocation	10.7	11.1	8.9	NS
Thermal torture	21.9	19.4	34.4	p < .001
Cuts and amputation	4.8	5.1	3.6	NS
Electrical torture	37.6	36.5	43.8	NS
Deprivation	51.2	50.6	54.7	NS
Sexual torture	15.9	15.4	18.8	NS
Threats	55.2	55.5	54.7	NS
Humiliations	17.9	18.0	17.7	NS
Witnessed torture	34.4	34.4	34.4	NS
Sensory stress	4.3	4.6	3.1	NS

ap-Values from Pearson Chi Square test.

To determine whether the difference in torture exposure rates between ethnic minority and majority patients is driven by the unequal gender distribution, Table 6 has the results of the regression models. Models 1 and 2 show that a patient’s ethnic minority status is positively related to the number of types of torture, to which a patient has been exposed. The relationship is statistically significant and robust when controls are added for socio-demographic factors. This confirms that the difference identified in Table 4 between ethnic minorities and the rest of the patient group, is not a result of the gender distribution.

Table 6.

Coefficients for OLS regression for models with sum of torture types as the dependent variable. Standard errors reported in parentheses

Variables	Model 1	Model 2	Model 3	Model 4
Ethnic minority	0.563** (.173)	0.456* (.195)	-	-
Palestinian	-	-	0.362 (.252)	0.237 (.281)
Kurdish	-	-	0.714** (.221)	0.617* (.245)
Female	-	−0.834** (.211)	-	−0.836** (.211)
Years of education	-	0.054** (.018)	-	0.053** (.018)
Age	-	−0.015(.008)	-	−0.015 (.008)
Constant	4.541** (.071)	5.671** (.437)	4.541** (.071)	5.694** (.437)

*p < 0.05; **p < 0.01.

Additionally, the models (3 and 4) show that the relationship is not universal for Palestinians and Kurds, the two types of ethnic minorities reported in the dataset. The positive relationship in models 1 and 2 appear to be driven by the Kurdish minority, which has been exposed to a significantly higher number of torture methods than the base population. To supplement these findings, additional analyses were completed, in which the patients were categorized based on their reported territories, so that those of Kurdish ethnicity, who reported their country of origin as Kurdistan, were grouped, and those of Kurdish ethnicity with a different country of origin were grouped together. The same procedure was followed for the Palestinian patients. This analysis indicates that these subgroups of Palestinians and Kurds differ in their levels of torture exposure. The Kurdish minority with other reported countries of origin and the Palestinian minority from Palestine are the only two groups that show torture exposure levels that are significantly higher than the base population. Results of the regression model can be seen in Table 7. In sum, the analyses confirm the hypothesis that belonging to an ethnic minority within one’s country of origin is associated with having been subjected to more types of torture.

Table 7.

Coefficients for OLS regression for models with sum of torture types as the dependent variable. Standard errors reported in parentheses

Variables	Model 1
Kurdish minority	0.524* (.265)
Kurdistan	1.045 (.562)
Palestinian minority	−0.725 (.470)
Palestine	0.699* (.334)
Female	−0.833** (.211)
Years of education	0.052** (.018)
Age	−0.016 (.008)
Constant	5.753** (.437)

*p < 0.05; **p < 0.01.

A total of 345 (30%) of the 1135 patients did not receive treatment. Linear regression analysis was used to control for differences between the groups. Age, gender, minority status, education level where a difference was detected between the groups were entered as covariates. Although both groups did report experiences of torture, those who received treatment were generally more likely to report these experiences, significantly so for suspension (p < .001) suffocation (p < .001), deprivation (p < .001), threats (p < .001) and witness of torture (p < .001).

Discussion

Underreported torture

As the data within the present study are based on journal notes, torture exposure not reported by patients during the initial assessment within the clinic will not appear in the data. It is well known that some types of torture are often unreported and might not be disclosed to medical professionals or psychologists. We therefore expect the numbers in this article to be conservative estimates of occurrence. More specifically, gender difference in exposure to sexual torture is a difference in reported sexual torture, which might not be an accurate estimate of the occurrence of sexual torture. This interpretation is supported by studies documenting the extent of sexual violence committed by men against men during armed conflicts (Manivannan, 2013; Sivakumaran, 2007; Solangon & Patel, 2012). However, the findings within the present study, in line with previous research, show that men are indeed also victims of sexual torture, which is important knowledge for practitioners working with refugees in order to ask the appropriate questions, even when the answer is difficult for the survivors to cope with.

Ethnic minority as a proxy for marginalization

The finding that survivors belonging to an ethnic minority population within their countries of origin report more types of torture than survivors belonging to the majority population sheds further light on the struggles of survivors belonging to this group. Unfortunately, it was not possible to construct a variable measuring the severity of reported torture in terms of duration and frequency of exposure within the different categories of torture, which might have further supported the hypothesis that belonging to an ethnic minority is associated with a greater risk of being subjected to more torture. The fact that no statistically significant correlation was found between the dichotomous torture exposure variable and ethnic minority status may be explained by the fact that the primary purpose of the clinic is to treat torture survivors, which leads to a non-representative sample with a very high proportion of survivors of torture. This should therefore not be seen as evidence that ethnic minorities in the general population are not more at risk. The nature of the data for the present study did not reliably allow for coding of other types of variables which may be associated with an increased risk of being subjected to torture. In this way, ethnic minority status may serve as a proxy for marginalization that can occur based on other parameters. The key parameter might be likelihood of police contact which in many regions of the world would also be increased for religious or sexual minorities or citizens living in extreme poverty.

Psychological torture

The present data did not allow for a systematic coding and classification of different types of psychological torture and other forms of ill treatment during captivity such as mock executions, sleep deprivation, the abuse of specific personal phobias, prolonged solitary confinement, etc. for the purpose of extracting information, as these were not systematically recorded in the patient files in the same way as physical forms of torture. This may be seen as problematic as there is evidence to suggest that psychological torture and other forms of ill treatment are not substantially different from physical torture in terms of the severity of mental suffering they cause, the underlying mechanism of traumatic stress, and their long-term psychological outcome. Future studies should thus aim to capture to full range of torture methods including psychological torture such as cruel, inhuman, and degrading treatment (Başoğlu, Livanou, & Crnobarić, 2007).

Patients who received treatment and those who did not

It is worth considering what factors may account for the difference between the patients who received treatment and those who did not. All patients were clinically assessed to be in need of treatment and were eligible for treatment in the clinic. The fact that 30% of torture survivors referred to the clinic have not received treatment, despite having been clinically assessed to be in need of treatment is unsatisfactory. It may in part be explained by the unstable living conditions of refugees in Denmark. In Denmark, individuals and families are required to agree to live for three years in an assigned community when accepted as refugees. The policy is based on the assumption that immersion in ethnically Danish local communities will facilitate integration (Larsen, 2011). However, this leads to a situation where many refugees move after the end of the three-year period. It is not surprising that those who received treatment reported more experiences of torture. It is a very vulnerable inaccessible group of patients, who may find it difficult to share torture experience, particularly in the initial phase of treatment. The group of patients who did not receive treatment were also considered to be in need of treatment, and many did report experiences of torture. It is unfortunate that they did not receive adequate help. Avoidance is a central PTSD symptom, and an essential part of treatment is building an alliance with the patient and motivating the patient for treatment. Overcoming the avoidance is particularly difficult for the most traumatized patients and it is thus not unlikely that attending the assessment sessions may have been retraumatizing for patients and brought about an avoidant reaction. Findings may also be seen as a testimony to the importance of a systematic clinical focus on addressing the patient’s avoidance and motivating the patient for treatment, from the very first contact. Although beyond the scope of the present study, future research should explore the mechanisms leading to drop-out in order to increase the treatment engagement for this very vulnerably group of torture survivors.

Conclusion

This study provides a testimony to the experiences of survivors of torture and organized violence based on 27 years of clinical practice. The humiliating and dehumanizing nature of torture promotes a culture of shame, secrecy and denial in survivors which may affect not just the survivors but also their families. In order to support survivors living in host countries, it is highly important for clinicians and frontline workers, who work with potentially traumatized refugee populations, to be aware of the nature of the traumatic experiences in order to ask the appropriate questions, even when the survivors struggle to come forward with their past traumatic experiences and the associated trauma sequelae. This is where the present study contributes. As hypothesized, the present study revealed gender differences in reported torture with men reporting more types of torture, while women are more than twice as likely to report sexual torture. The reported torture rates are, however, most likely only a conservative estimate of the occurrence, and it is highly possible that men are more likely to underreport sexual torture. In line with previous studies, the present study also found an association between being an ethnic minority within one’s country of origin and reporting more types of torture. This association was robust even when controlling for gender and other sociodemographic variables.

The retrospective nature of the data collection process did not allow us to extract data on other variables than the one’s reported, but it is highly possible that belonging to an ethnic minority in one’s country of origin can be seen as a proxy for marginalization and that the underlying risk factor is likelihood of police contact which in many countries of the world would also be higher for religious or sexual minorities. Furthermore, due to the available data, the present study was unable to systematically record and classify the full range of psychological torture methods, which should be addressed in future studies, as there is evidence to suggest that these methods are as detrimental as physical torture. Finally, the study found that 30% of patients, despite a need for help, did not receive treatment, which emphasizes the need for future research to explore the mechanisms leading to drop out of treatment further.

Limitations

The present study is based on a nonrepresentative sample and a retrospective data collection procedure, which may compromise the accuracy and generalizability of findings. As the data is based on clinicians’ written journal notes within patient files it is possible that there are some inconsistencies in the original recordings of data due to human error or shifting focus on different types of trauma over the past three decades. Furthermore, as discussed in the present article, experiences of torture are often underreported due to shame and fear of stigma or revictimization and thus many survivors may not have reported all of their experiences with torture. Many patients may not have revealed their experiences of torture in the initial assessment sessions, as they may not have had the confidence to do so, and thus it is possible that more experience were reported during the course of treatment, when a more trusting relationship between patients and their therapists had developed.

The present data did not allow for a reliable classification and coding of different types of psychological complaints and symptoms, which would likely also have provided further insights.

Future research should explore not just the reporting of different types of torture but also the intensity and brutality in terms of frequency and duration of each type of torture and the associations with victim characteristics such as gender and ethnic minority status and belonging to other marginalized populations such as religious and sexual minorities.