We found that 11% (16/142, 95% CI 7% to 18%) of participants reported a history of being personally tortured in a manner that met the UN definition of torture. Two associated factors for higher rates of torture were region of origin (Africa) and having been in the United States for a shorter period of time. However, the observation that subjects in the United States for a shorter period of time were more likely to be tortured is likely due to the fact that the duration for which subjects had been in the United States was associated with the region of origin (P < 0.01). For example, while half of all subjects who had been in the United States for less than 3.3 years (first quintile) were of African origin, only 5% of those who had been in the United States for greater than 31 years (fifth quintile) were of African origin.
Our finding that 39% of subjects reporting torture had disclosed this history to a health care provider is extraordinarily high when compared with the existing literature. Eisenman et al.16
reported only 3% ever telling a clinician about political violence, and 0% reported that their current clinician asked about political violence. There are several hypotheses for this unexpected finding. The greater Boston area is a resettlement site for many immigrant communities, and there are multiple organizations that serve the needs of these populations, including specific attention to the needs of survivors of torture and human rights abuses. Some of these organizations are Physicians for Human Rights, The International Institute of Boston, and The Boston Center for Refugee Health and Human Rights. The presence of, and awareness about these organizations in communities may increase awareness of torture among both patients and providers. We have also provided training about caring for survivors of torture to health care providers in multiple clinical sites in the greater Boston area, directly raising awareness of this topic among primary care providers.
Our findings on the prevalence of torture in African primary care patients (41%) is similar to that reported by Jaranson et al.14
in a community-based population of East African refugees (25% to 69%). The prevalence of torture in subjects from Central and South America and the Caribbean was 6%, compared with 8% reported by Eisenman16
in Latino primary care patients. Our prevalence of torture was higher than Eisenman15
reported in a primary care sample of 121 patients in NYC (6.6% vs 11%). It is possible that differences in subjects' countries of origin may account for this difference.
Several important limitations should be considered when interpreting these data. Few studies have examined the validity of self-reported history of torture. The reference that we cite (Montgomery and Foldspang18
) reports good validity of a personal report of torture to a clinical determination of torture as defined by the Tokyo Declaration. The validity of our questions in determining torture as established by the UN Convention has not been directly established. We did not evaluate socioeconomic status and it is possible that the prevalence of torture may vary with socioeconomic status. We excluded subjects who were known to the Boston Center for Refugee Health and Human Rights, which is based at Boston Medical Center. This was appropriate, as two-thirds of patients seen at Boston Center for Refugee Health and Human Rights have been referred for care at the Boston Center for Refugee Health and Human Rights by outside sources (attorneys and resettlement agencies) and including such patients would inappropriately enrich our sample. Conversely, as one-third of the patients at the Boston Center for Refugee Health and Human Rights are referred from within Boston Medical Center excluding such patients will deplete the sample and yield an underestimation of the true prevalence of torture in the clinic population. The magnitude of this effect, however, is quite small due to the relative sizes of the Primary Care Clinics at Boston Medical Center (>20,000 unique patients/year, >33% foreign-born) and The Boston Center for Refugee Health and Human Rights (359 patients last year). Inclusion of subjects removed from the sampling pool due to internal referral would have increased our prevalence estimate from 11% (95% CI 7% to 18%) to 13%. It is also possible that the prevalence we report is an overestimate due to the presence of a specialized center for survivors of torture within the institution. Although patients of the Boston Center for Refugee Health and Human Rights were not included in this survey, relatives and acquaintances of such patients, who themselves likely would have a high rate of exposure to violence, may have been drawn to the institution for this reason. It is possible that overestimation of prevalence rates could be due to high utilization of health care services, as seen with domestic violence populations.20
Weighting the subject selection process by health care utilization was not possible because the surveys were administered anonymously.
We collected no information on nonparticipants. While it is possible that some survivors of torture would choose to avoid this study because of fear and stigma, we are unable to confirm this conjecture. While the interview instrument used in this study has been validated previously, we were not able to confirm that the instrument operates effectively across the many cultures represented by the participants in this study. We did not ask whether participants were refugees or asylum seekers. This might have provided useful data for primary care providers, as immigration status may be an important easily identifiable associated factor. It is unclear whether the point prevalence we report can be generalized to the foreign-born patients in other primary care practice settings. It is important to realize that the actual point prevalence of torture will vary among clinical practices depending on the proportion of foreign-born patients from different countries and various parts of the world. In addition, prevalence will change over time with country-specific situations, such as wars, oppressive leaders, and politics.
The high prevalence of torture in foreign-born primary care patients highlights the importance of clinical interview and exam skills for primary care providers to identify patients who have experienced torture or potential vicarious trauma. Lack of recognition and treatment may result in significant psychological and physical sequelae.
The clinical presentation of survivors of torture has been shown to be highly varied.21,22
For example, patients may present to their primary care providers with chronic headache or organic brain syndromes due to head trauma, nerve palsies due to suspension, genital pain due to genital torture, foot pain due to falanga, chest pain, abdominal pain, hearing loss, or dental trauma.10
Often, there are no telltale marks, and physicians are not generally trained to detect the specific sequelae of torture.23
In addition, mental illness, including posttraumatic stress disorder, depression, anxiety, adjustment disorder, and psychosomatic illness, are all prevalent in torture survivors, but may not be easily diagnosed in the absence of an appropriate history.24,25
This lack of recognition may result in unnecessary investigations, or labeling patients as “hypochondriacs.” At worse, the lack of a history will result in failure to get treatment and prolongation of suffering.
Our results showed that this survey was their first disclosure to anyone in the United States of being personally tortured or having a family member tortured for one-third of the subjects. Survivors of torture may try to avoid medical care due to fear of further persecution, deportation, and humiliation. They may not identify themselves to physicians, even when seeking services. Such patients may harbor a basic mistrust of physicians and may be reluctant to tell their caregivers about their history. For communities without dedicated immigrant and refugee services, providers may need more diligence to elicit a torture history in foreign-born patients.
In our population, variables associated with a higher risk of torture were recent arrival to the United States, and immigration from the African and Asian continents. We believe that clinicians should routinely ask patients from the African and Asian continents who are recent arrivals to the United States about a history of torture. Further studies of large numbers of foreign-born patients across a broad spectrum of primary care practices are needed to stratify risk factors for torture in clinical settings, and to provide further guidance to clinicians for torture history screening in primary care settings.
Screening programs, educational initiatives, and interventions for treatment should be further studied. Physicians seeing immigrant patients in their practices should be familiar with the general backgrounds of their patients' countries of origin, common medical and psychological sequelae of torture, and should be knowledgeable about specialized referral centers for survivors of torture. The Boston Center for Refugee Health and Human Rights has a web course available for providers on caring for survivors of torture (http://www.bcrhhr.org
). Information about specialized treatment centers for survivors of torture can be found at The National Consortium of Torture Treatment Programs Web site.26