'Only God can promise healing.': help-seeking intentions and lay beliefs about cures for post-traumatic stress disorder among Sub-Saharan African asylum seekers in Germany.

Background: Epidemiological studies have reported high rates of post-traumatic stress disorder (PTSD) among asylum seekers from Sub-Saharan Africa. In order to provide appropriate and culturally sensitive mental health care for this group, further knowledge about treatment preferences might be necessary. Objective: We aimed to provide insights into help-seeking intentions and lay beliefs about cures for PTSD held by asylum seekers from Sub-Saharan Africa living in Germany.
Methods: To address this objective, we used a quantitative and qualitative methodological triangulation strategy based on a vignette describing symptoms of PTSD. In the quantitative part of the study, asylum seekers (n = 119), predominantly from Eritrea (n = 41), Somalia (n = 36), and Cameroon (n = 25), and a German comparison sample without a migration background (n = 120) completed the General Help-Seeking Questionnaire (GHSQ). In the qualitative part, asylum seekers (n = 26) reviewed the results of the questionnaire survey within eight focus group discussions sampled from groups of the three main countries of origin.
Results: Asylum seekers showed a high intention to seek religious, medical, and psychological treatment for symptoms of PTSD. However, asylum seekers indicated a higher preference to seek help from religious authorities and general practitioners, as well as a lower preference to enlist psychological and traditional help sources than Germans without a migration background. Furthermore, asylum seekers addressed structural and cultural barriers to seeking medical and psychological treatment.
Conclusion: To facilitate access to local health care systems for asylum seekers and refugees, it might be crucial to develop public health campaigns in collaboration with religious communities. When treating asylum seekers and refugees from Sub-Saharan Africa, practitioners should explore different religious and cultural frameworks for healing and recovery in order to signal understanding and acceptance of varying cultural contexts.

Background

By 2018, the number of forcibly displaced people around the world had risen to an unprecedented 68.5 million, among them 25.4 million refugees and 3.1 million asylum seekers (United Nations High Commissioner for Refugees, 2018). Despite the progressing isolationist policy of the European Union (Fröhlich, 2018), European countries are receiving persistently high numbers of asylum seekers of Sub-Saharan African origins (Eurostat, 2019), who have been exposed to a high number of traumatizing events, such as war, armed conflict, and torture (Cavallera et al., 2016; Neuner et al., 2010). These circumstances, as well as post-migration stressors in the receiving countries, seem to have a strong detrimental impact on their mental health (Carta, Bernal, Hardoy, & Haro-Abad, 2005; Mewes, Reich, Skoluda, Seele, & Nater, 2017). Although high rates of post-traumatic stress disorder (PTSD) and other mental disorders have been found among asylum seekers from Sub-Saharan Africa (Kolassa et al., 2010; Neuner et al., 2004; Onyut et al., 2009; Slobodin & de Jong, 2015; Ssenyonga, Owens, & Olema, 2013; Tempany, 2009), too few receive support and professional help (Munz & Melcop, 2018; Sijbrandij et al., 2017).

Disparities in help-seeking behaviour have been documented between refugee populations and populations without a migration background, with asylum seekers and refugees being less likely than native-born populations to seek or be referred to mental health services, even when they experience comparable levels of distress (Kirmayer et al., 2011). In part, these disparities have been attributed to structural barriers, as the provision of appropriate mental health care for refugees and asylum seekers has been described as a challenge for the health care systems of host countries (Munz & Melcop, 2018; Sijbrandij et al., 2017; Sturm, Baubet, & Moro, 2010). In addition, the refugees themselves might suffer from a lack of knowledge regarding available mental health care services (Donnelly et al., 2011), unfamiliarity with such services (Ellis et al., 2010; Palmer, 2006), concerns regarding the confidentiality of professional interpreters (Bhatia & Wallace, 2007), and fears that their problems will not be understood by practitioners due to a lack of cultural competence (De Anstiss & Ziaian, 2010; Sandhu et al., 2013). Moreover, asylum seekers’ culturally shaped perceptions of mental disorders might influence their help-seeking intentions and impede contact with health care providers (Este, Simich, Hamilton, & Sato, 2017; Knettel, 2016; Mölsä, Hjelde, & Tiilikainen, 2010; Priebe, Giacco, & El-Nagib, 2016).

Past research has demonstrated that refugees and asylum seekers of Sub-Saharan African origin tend to rely on alternative sources of help rather than seeking mental health care services (Ellis et al., 2010; Fenta, Hyman, & Noh, 2006; Palmer, 2006). For instance, Ethiopian immigrants and refugees in Canada were found to be more likely to consult traditional healers than health care professionals for mental health problems (Fenta et al., 2006). Moreover, Ethiopian refugees and asylum seekers in the UK reported relying more on interpersonal solutions or religious treatment when experiencing mental health problems (Papadopoulos, Lees, Lay, & Gebrehiwot, 2004). Eritrean refugees in Switzerland considered mental health to be related to faith, and described spiritual or church support as a first-line treatment for mental health issues and trips to holy waters as a cure for mental health problems (Melamed, Chernet, Labhardt, Probst-Hensch, & Pfeiffer, 2019). Studies on the help-seeking preferences of Somali refugees living in Western resettlement countries found a reliance on family, friends, and the ethnic and religious community, and a preference for religious sources of help (Ellis et al., 2010; Markova & Sandal, 2016). A large body of research has emphasized the importance of the belief in God’s will and religious prohibitions for health care. Traditional and religious healing through readings from the Quran, eating special foods, and burning incense has been described as a treatment for mental health problems (Carroll, 2004; Clarkson Freeman, Penney, Bettmann, & Lecy, 2013; Palmer, 2006; Pavlish, Noor, & Brandt, 2010).

Regarding psychotherapeutic and medical health care, Somali refugees in Norway expected psychologists and general practitioners to provide concrete solutions that would effectively cure mental health problems (Markova & Sandal, 2016) and to immediately prescribe treatment (Pavlish et al., 2010). Furthermore, Pavlish et al. (2010) reported that Somali women in the USA expected to develop a personal relationship with their health care providers, which seems rather contradictory to Western health care systems. These divergent expectations are bound to result in multiple frustrations, with the potential to diminish the perceived quality of health care (Pavlish et al., 2010). In another study by Ellis et al. (2010), Somali refugee adolescents living in the United Sates did not perceive psychotherapy as a culturally accepted treatment. Somali and Ethiopian refugees in Australia proposed community-based solutions to problems of mental distress and preferred members of their ethnic communities to be trained to provide professional support (Kokanovic, Dowrick, Butler, Herrman, & Gunn, 2008). Furthermore, Sub-Saharan African migrant youth in Australia perceived strong local community support systems, trustworthiness of help-sources, high expertise of formal help-sources, and increasing mental health literacy as facilitators for seeking out mental health professionals (McCann, Mugavin, Renzaho, & Lubman, 2016).

However, to our knowledge, no study has included native comparison groups. Hence, the quantitative difference in help-seeking intentions between refugees and asylum seekers from Sub-Saharan Africa and host populations is not known. Moreover, most research analysed mental health service utilization and barriers in general, while we are unaware of any study addressing beliefs regarding cures for the symptoms of PTSD. In order to increase access to mental health care and to provide culturally sensitive care for refugees and asylum seekers, further knowledge about differences to the host population in terms of help-seeking intentions and beliefs about cures for PTSD might be necessary. Therefore, this paper aims to give further insight into help-seeking intentions and beliefs about cures for PTSD held by Sub-Saharan African asylum seekers residing in Germany. To address this research issue, we applied a qualitative and quantitative methodological triangulation strategy. The application of these two methodologies enabled us to generate comparable and generalizable data and to complement it with in-depth and culturally sensitive descriptions of constructs from an emic perspective. We believe that this reduces the problem of transferability of Western constructs to other cultural contexts (Bekhet & Zauszniewski, 2012; Karasz & Singelis, 2009).

In the quantitative part of the study, we hypothesized that Sub-Saharan African asylum seekers would express a higher intention to seek religious and medical treatment and a lower intention to seek psychological treatment than German participants without a migration background. To account for possible influences of countries of origin on help-seeking intentions, group differences between asylum seekers from the main countries of origin were explored. In the subsequent qualitative part of the study, the quantitative results were reviewed within focus group discussions.

Materials and methods

Procedure

The study consisted of two parts: a) a questionnaire survey in which participants responded to the General Help Seeking Questionnaire (GHSQ; Wilson, Deane, Ciarrochi, & Rickwood, 2005) and b) focus group discussions in which participants reviewed the results of the questionnaire survey (see also Grupp, Moro, Nater, Skandrani, & Mewes, 2018). Ethical approval for each part of the study was obtained from the local review board of the Department of Psychology, University of Marburg, Germany, and all participants provided informed consent prior to participation. Data were collected in 2016 throughout different cities in Germany.

For all participants, an inclusion criterion for study participation was age over 18 years, while training or working in a health profession was an exclusion criterion. As an exclusion criterion for the German participants, we employed the definition of the German Federal Statistical Office (2017) and defined participants as having no migration background when both of their parents were born in Germany. Asylum seekers had to have flight experience and an origin in a Sub-Saharan African country.

In the first part of the study, a convenience sample of n = 120 German participants without a migration background, as well as n = 119 asylum seekers from seven Sub-Saharan African countries, mainly Eritrea (n = 41), Somalia (n = 36), and Cameroon (n = 25), took part in a questionnaire survey using paper-and-pencil and online assessments (survey software UNIPARK & QuestBack©). In the group of asylum seekers the paper-and-pencil form was filled in by 112 participants and seven participants completed the online version. A combination of convenience and snowball sampling was applied. Asylum seekers were approached in their accommodation facilities, through collaboration with civic refugee initiatives, through language courses for adult immigrants, and by networking at cultural gatherings. The survey was provided in German, English, French, Tigrinya, and Arabic. Asylum seekers preferred the paper-and-pencil version, as they favoured the personal contact with researchers and expressed difficulties with their internet connection. However, past research supports the idea that web-based data collection provides, in general, equal responses as paper-and-pencil modes (Braekman et al., 2018). For the German participants without a migration background the online survey (n = 99) was promoted by the researchers through mailing lists, social media, and through personal contacts via word-of-mouth recommendation. The paper-and-pencil version (n = 21) was distributed in vocational schools and church congregations.

In the second part of the study, a smaller number of participants (n = 26) reviewed the findings of the quantitative part in focus group discussions. Given that the quantitative sample consisted mainly of individuals from Eritrea, Somalia, and Cameroon, focus group discussions were sampled according to these three countries of origin. In the quantitative part of the study, participants had the opportunity to express their interest in taking part in a subsequent interview study and interested persons were contacted by the first author after the questionnaire survey was terminated. Thus, a total of eight focus groups were conducted with participants from Eritrea (three focus groups; n = 10), Somalia (three focus groups; n = 8), and Cameroon (two focus groups; n = 8) who took part in the questionnaire survey beforehand. Focus groups were organized separately for different languages and countries of origin. Participants of each focus group mostly came from the same regions of their home country, mainly from the capital or urban regions. The focus groups took place in asylum seekers’ accommodation facilities throughout different cities in Germany and at a German university. The average duration of focus group discussions was 1 h 30 min. Each discussion was audio-taped and complemented by handwritten notes. The first author, a trained clinical psychologist, conducted focus group discussions in English and French with participants from Somalia and Cameroon. Focus group discussions with participants from Eritrea were carried out with the assistance of bilingual Tigrinya to German interpreters who received detailed instructions beforehand.

Measures and materials

After answering sociodemographic questions, participants of both parts of the study were presented with an unlabelled standardized vignette about a hypothetical friend describing symptoms of PTSD according to criteria outlined in the 10th International Classification of Diseases (see Maercker, 2013; World Health Organization [WHO], 1993). In the first part of the study, the vignette was integrated into the questionnaire, while in the second part, it was read aloud and disseminated in a printed version during the focus group discussions. A stressful event of an exceptionally threatening nature was indicated as a triggering event for the symptoms, following criterion A of ICD-10 (WHO, 1993). The traumatic event differed for asylum seekers and Germans without a migration background in order to increase the possibility that participants had experienced such an event and to improve the fit to the participants’ respective living conditions: Whereas ‘flight’ is a representative traumatic event experienced by asylum seekers, physical violence is among the most commonly experienced traumatic events in the German general population (Hauffa et al., 2011). Therefore, we operationalized physical violence with the event ‘armed robbery’ in the vignette for Germans without a migration background.

The vignette read as follows (Maercker, 2013):

“Since the [flight/armed robbery], I have become a totally different person. In the evenings, I lie in bed and then these thoughts and images come and I lie awake forever. Now I have reached a point where I realize I can’t go on like this anymore … Sometimes I scream at night and I wake up drenched in sweat because of the nightmares. If I have arrived somewhere and there is a noise, I wince. There it is again. I can’t turn it off, it’s like an electric shock that immediately goes straight up and triggers intense sweating. My wife/My husband accuses me of often being aggressive, easily irritable and she/he is afraid of my outbursts of rage. That’s why I prefer to withdraw myself because I always have a feeling that no one can be trusted anymore. Many things just don’t interest me anymore. Sometimes my environment appears distant and unreal and I have a feeling of “standing next to myself”, then I become totally numb. Afterwards I sometimes can’t remember what has happened. I have no hope left anymore … ”

Participants were asked to imagine the scenario and indicate their personal assumptions concerning the described condition. In the first (quantitative) part of the study, participants responded to the General Help Seeking Questionnaire (GHSQ; Wilson et al., 2005) and an extended version of the Posttraumatic Diagnostic Scale (PDS; Foa, Cashman, Jaycox, & Perry, 1997). In the second (qualitative) part, focus group discussions were structured using an adapted version of the Short Explanatory Model Interview (SEMI; Lloyd et al., 1998). Translations of both instruments were conducted using the forward- and backward-translation method (Flaherty et al., 1988).

General Help Seeking Questionnaire GHSQ

The GHSQ (Wilson et al., 2005) aims to assess intentions to seek help from different sources by rating the intention on a 7-point Likert scale ranging from 1 (‘extremely unlikely’) to 7 (‘extremely likely’) for each source (see Table 2 for the sources). Higher scores indicate higher intentions. For the present study, the instructions were altered to ‘Where should the person described above seek help for his/her problems?’. In addition, the item ‘traditional treatment’ was added to take into account traditional sources of help such as healers, shamans, herbal treatments, or animal sacrifices. The GHSQ had to be translated into French, Arabic, and Tigrinya; a German version was already available. The translations were conducted as described above.

Table 2.

Preferences for informal and formal help-seeking in the two groups.

	Asylum seekers				German participants
	likely (%)	unsure (%)	unlikely (%)		likely (%)	unsure (%)	unlikely (%)
GHSQ Informal help-sources*				GHSQ Informal help-sources*
Parent	75.4	10.5	14.0	Intimate partner	65.0	19.2	15.8
Intimate partner	70.8	13.3	15.9	Parent	51.7	27.5	20.8
Other relative/family member	60.8	16.1	23.2	Friend	64.1	21.7	14.2
Friend	56.1	25.4	18.4	Other relative/family member	45.1	34.2	20.8
GHSQ Formal help-sources*				GHSQ Formal help-sources*
Psychiatrist	68.4	17.5	14.0	Psychologist	93.3	3.3	3.3
Religious authority	67.5	15.8	16.7	Psychiatrist	84.1	9.2	6.7
Psychologist	64.3	22.3	13.4	General practitioner	45.0	20.0	35.0
General practitioner	60.1	23.9	15.9	Traditional treatment	43.3	28.3	28.3
Traditional treatment	22.1	32.7	45.1	Religious authority	17.5	33.3	49.2
None	12.8	33.6	54.2	None	0.8	2.5	96.7

*GHSQ = General Help Seeking Questionnaire.

Post-traumatic stress diagnostic scale PDS

Experienced traumatic events and post-traumatic symptoms were assessed using an extended version of the PDS (Foa et al., 1997). The PDS list of 11 specified potentially traumatizing events was extended by 12 traumatic events frequently experienced by asylum seekers from the Harvard Trauma Questionnaire (e.g. ‘brainwash’, ‘kidnapped or taken as a hostage’, and ‘lacked shelter’; Mollica et al., 1992) to determine experienced traumatic events. In addition, participants rated 17 items representing the cardinal symptoms of PTSD experienced in the past 30 days on a 4-point scale. These ratings sum up to a symptom severity score ranging from 0 to 51. The cut-offs for symptom severity rating are 1–10 mild, 11–20 moderate, 21–35 moderate to severe, and > 36 severe (McCarthy, 2008).

Short Explanatory Model Interview SEMI

The focus group discussions were moderated using key questions from the interview manual SEMI (Lloyd et al., 1998), developed to elicit explanatory models, exploring respondent’s cultural background, nature of presenting problem, help-seeking behaviour, interaction with a healer, and beliefs related to mental illness. We adhered to the originally provided questions by the SEMI (Lloyd et al., 1998), such as ‘Where would you seek help from?’ and ‘What should a practitioner do about it?’. Moreover, we extended the SEMI by adding probes that were derived from the results of the first, quantitative part, of the survey (‘What could a treatment look like?’; ‘Will a treatment cure this discomfort effectively?’; ‘What would a religious [traditional/biomedical] treatment look like?’; ”Would you want to utilize such a treatment?”). As we were interested in help-seeking intentions and beliefs about cures for PTSD, we only focused on responses to this part of the discussions.

Analyses

Statistical analysis

Statistical analyses were carried out using IBM SPSS Statistics version 17.0. Statistical significance was set at p < .05 (two-tailed).

Group differences in sociodemographic variables, traumatic events, and posttraumatic symptom load were investigated using Chi-square tests and t-tests (see Table 1).

Table 1.

Study sample and characteristics.

	Asylum seekersn=119		German participantsn=120				Sample of the qualitative partn=26
Variable	M (SD)	range	M (SD)	range	t	p	Eritrea (n=10)	Somalia (n=8)		Cameroon (n=8)
Age (years)	27.97 (7.8)	18-54	37.24 (16.5)	18-79	-5.56	<.001	31.9	25.6		23.4
Education (years)	9.7 (3.6)	2-18	12.2 (1.7)	6-16	-6.64	<.001	10.8	5.6		9.5
					χ²	p
Males	70.7 %		35.0 %		30.1	<.001	80 %	87.5 %		100 %
Education					36.51	<.001
university degree	13.4 %		33.3 %
higher education entrance-level qualification	18.5 %		30 %
secondary school certificate	39.5 %		31.7 %
primary school education	21.0 %		1.7 %
no school-leaving qualification	5.0 %		1.7 %
Religion					63.7	<.001
Christianity	56.3 %		65.8 %
Islam	33.6 %		0.8 %
Judaism	1.7 %		0.8 %
No Religion	4.2 %		31.7 %
Other	0.8 %		0.8 %
					t	p
Importance of faith	2.90 (0.50)	0-3	1.52 (1.32)	0-3	10.65	<.001
Assistance from faith	3.67 (0.94)	0-4	1.84 (1.60)	0-4	10.74	<.001
Country of birth
Eritrea	n = 41 (34.5 %)
Somalia	n = 36 (30.3 %)
Ethiopia	n = 7 (5.9 %)
Sudan	n = 2 (1.7 %)
Cameroon	n = 25 (21.0 %)
Nigeria	n = 4 (3.4 %)
Togo	n = 4 (3.4 %)
Posttraumatic symptom severity score (PTSD)	15.25 (10.42)	0-41	7.09 (9.76)	0-36	5.88	<.001
					χ²	p
Number of experienced traumatic events	5.02 (4.91)	0-19	1.33 (2.03)	0-11	62.34	<.001
Prevalence of at least one traumatic event	73.9 %		54.2 %

To display the preference for help-seeking in each group separately, the values 1 (‘extremely unlikely’); 2 (‘very unlikely’), and 3 (‘unlikely’) of the GHSQ were combined in the category ‘unlikely’. Value 4 (‘unsure’) represented the category ‘unsure’ and values 5 (‘likely’), 6 (‘very likely’), and 7 (extremely likely”) were combined in the category ‘likely’ (see Table 2).

Before analysing differences between the two samples, initial correlational analyses and t-tests were conducted for each group separately in order to determine significant associations between different sociodemographic variables on the one hand and help-seeking intentions on the other hand (see the correlational matrix in the supplementary material, Table 4). Since sociodemographic variables differed significantly between the two groups, we decided to include the variables gender, age, religion, importance of faith, formal years of education, highest school leaving qualification, number of traumatic events, and symptom severity as covariates into the analytical model. Finally, sample differences in GHSQ items were investigated by conducting a one-way between-groups multivariate analysis of covariance (MANCOVA). Items of the GHSQ were used as dependent variables. The independent variable was sample – asylum seekers or no migration background (see Table 3).

Table 3.

Inter- and intragroup differences in help-seeking intentions.

	German participants	Asylum seekers	Intergroup differences			Intragroup differences
	M (SD)	M (SD)	F	p	Partial η2		M (SD)	F	p
GHSQ sources of help*
Intimate partner	4.85 (1.34)	5.08 (1.70)	1.48	.157	.069	Eritrea	4.76 (2.12)	2.31	.105
						Somalia	5.59 (1.52)
						Cameroon	4.70 (1.80)
Friend	4.80 (1.20)	4.54 (1.56)	0.93	.502	.044	Eritrea	4.32 (1.45)	1.18	.313
						Somalia	4.85 (1.60)
						Cameroon	4.39 (1.50)
Parenta,c	4.47 (1.42)	5.45 (1.81)	3.48	.001	.148	Eritrea	4.94 (2.01)	6.47	.002
						Somalia	6.21 (1.37)
						Cameroon	4.70 (1.89)
Other relative/family member	4.30 (1.17)	4.50 (1.70)	0.60	.791	.029	Eritrea	4.09 (1.71)	0.85	.431
						Somalia	4.62 (1.67)
						Cameroon	4.48 (1.81)
Religious authority	3.24 (1.44)	4.83 (1.84)	7.02	<.001	.259	Eritrea	4.68 (2.01)	0.03	.971
						Somalia	4.77 (1.94)
						Cameroon	4.65 (1.67)
Traditional treatment	4.12 (1.52)	3.14 (1.73)	3.88	<.001	.162	Eritrea	3.12 (1.30)	0.12	.887
						Somalia	3.29 (1.96)
						Cameroon	3.30 (1.82)
General practitionera	4.23 (1.47)	4.75 (1.71)	2.46	.012	.109	Eritrea	4.09 (1.82)	5.24	.007
						Somalia	5.32 (1.41)
						Cameroon	4.30 (1.74)
Psychiatristc	5.74 (1.29)	5.46 (1.58)	1.41	.187	.066	Eritrea	4.59 (1.94)	4.49	.014
						Somalia	5.68 (1.59)
						Cameroon	5.70 (1.40)
Psychologist	6.29 (0.99)	5.24 (1.52)	4.04	<.001	.167	Eritrea	3.97 (1.48)	15.25	<.001
						Somalia	5.56 (1.74)
						Cameroon	6.00 (1.13)
Noneb	1.43 (0.92)	2.87 (1.56)	8.34	<.001	.293	Eritrea	3.47 (1.46)	5.14	.008
						Somalia	2.26 (1.73)
						Cameroon	2.87 (1.39)

*GHSQ = General Help Seeking Questionnaire.

asignificantly higher intentions in Muslim asylum seekers.

bsignificantly higher intentions in Christian asylum seekers.

csignificantly higher intentions in male asylum seekers.

Additional analyses were conducted to explore differences between the three main countries of origin of asylum seekers: Eritrea, Somalia, and Cameroon (Table 3).

Prior assumption testing was carried out to verify the criteria of normality, homogeneity of variance-covariance matrices, and multicollinearity. Shapiro-Wilk tests revealed significant deviations from normal distribution in some items. However, as analyses of variance have proven to be robust against such deviations, analyses were nevertheless conducted (Glass, Peckham, & Sanders, 1972; Harwell, Rubinstein, Hayes, & Olds, 1992; Schmider, Ziegler, Danay, Beyer, & Bühner, 2010). Effect sizes are given as partial eta-squared, where .01–.06 = small effect; .06–.14 = moderate effect and > .14 = large effect (Cohen, 1988).

Interpretative Phenomenological Analysis (IPA)

Focus group discussions were recorded, transcribed, and analysed applying Interpretative Phenomenological Analysis (IPA; Smith & Osborn, 2003). To organize and manage data analysis, the software MAXQDA© version 12 was used.

As described by Smith and Osborn (2003), we followed a four-stage process, beginning with a close interpretative reading of the first transcript. Responses to the material were captured with initial notes and translated into emergent themes at one higher level of abstraction. Afterwards, these themes were connected, associated, and arranged in a table of superordinate themes. This procedure was repeated for each transcript. Subsequently, we established patterns cross-case and recorded them in a master table of themes for all of the transcripts. The audited themes were reviewed with other researchers and experts of the respective cultures to ensure that conclusions were drawn in a culturally sensitive manner and were well derived from the transcripts (Smith & Osborn, 2007, 2003).

Results

Sample

The group of asylum seekers taking part in the quantitative part of the study (n = 119; see Table 1) comprised 71% males and the mean age was 28 years, which broadly corresponds to German asylum statistics (Federal Office for Migration and Refugees, 2017). Their mean duration of stay in Germany lay at approximately two years (s.d. = 1.2). The mean age of participants from Eritrea was 28 years, from Somalia 25 years, and from Cameroon 30 years; with around 50% female participants per group. Germans without a migration background (n = 120) comprised 35% males and the mean age was 37 years (see Table 1). Over half of the investigated asylum seekers were Christians and one third were Muslims. Regarding German participants, two thirds were Christians and almost one third indicated that they were not religious. Asylum seekers stated stronger faith and higher perceived assistance from their faith. They had a lower educational level than German participants. Asylum seekers had experienced significantly higher numbers of different traumata than Germans without a migration background and showed a significantly higher symptom load.

The sample of the second part of the study consisted of n = 26, predominantly male, participants (see Table 1). Their mean age was 27 years, they had a mean of 8.5 years of formal education, and their mean duration of residence in Germany was 1.3 years. Participants from Eritrea and Cameroon were Christians, whereas participants from Somalia were Muslims.

Help-seeking intentions – results of the questionnaire survey

We hypothesized that asylum seekers from Sub-Saharan African would express a higher intention to seek religious and medical treatment and a lower intention to seek psychological treatment than Germans without a migration background.

In general, asylum seekers most often intended to seek help from informal sources, such as a parent (75%) or partner (71%), who seemed less relevant for the German participants as a source of help for the symptoms of PTSD (52% and 65%, see Table 2). German participants most often intended to seek help from psychologists (93%) or psychiatrists (84%), which ranked fifth (64%) and third (68%) in the group of asylum seekers. Religious help ranked fourth in asylum seekers (67.5%), but was the next to last source (17.5%) in the German comparison group (see Table 2).

Statistical comparisons showed a significantly higher intention in asylum seekers to rely on their parents when experiencing the symptoms of PTSD compared to German participants (large effect size; Table 3), and particularly older German participants (regarding seeking help from partner, parents, or friends; Table 3). Further analyses showed that the intention to seek advice from their parents was significantly higher in participants from Somalia than in participants from Cameroon. German participants showed higher intentions to seek help from traditional treatments and psychologists (large effect sizes) compared to asylum seekers. In contrast, asylum seekers indicated significantly higher intentions to seek help from a religious authority (large effect size) and a general practitioner (moderate effect size). However, older participants and participants from Eritrea indicated a comparably lower intention to seek medical help (see Table 3). Asylum seekers’ intention to not seek help at all was significantly higher compared to the German participants (large effect size). Furthermore, this intention was significantly higher in asylum seekers with a higher posttraumatic symptom load (see Table 4, supplement), and in participants from Eritrea compared to participants from Somalia and Cameroon (Table 3).

Beliefs about cures for PTSD – themes emerging from the focus group discussions

Participants of all focus group discussions identified the described PTSD symptoms in themselves or in somebody they knew. They attributed the symptoms to various causes such as traumatic life experiences, psychological and social causes, and post-migration stressors. The role of religious causes was emphasized, but supernatural causes, such as witchcraft, cursing, and evil spirits, were also discussed as culturally acceptable causes in explaining PTSD symptoms (see Grupp et al., 2018).

Regarding help-seeking and cures for the symptoms, different forms of treatment were proposed, which can be grouped into three superordinate themes: (a) religious treatment, (b) traditional treatment, and (c) medical and psychological treatment.

In general, initiating help-seeking behaviour was placed in a social context and perceived to be the responsibility of the social environment of the affected person. The different forms of treatment were often described as being supported by and taking place in the presence of parents, relatives, or religious and village communities.

“You, as an affected person, cannot set yourself off, it’s your family that will guide you to (…) faith and to healing waters and also your wife who sees your problem and will bring you to a psychologist. As an affected person you don’t know immediately how to look out for yourself. You are clouded and that’s why I see the family and the parents as the most important assistance.“ (male, 35 years, Eritrea)

Religious treatment

Seeking religious treatment was perceived to be the most effective form of curing PTSD symptoms (n = 26). Receiving prayers from a priest, sheikh, or imam and reciting particular verses of the Bible for Christian participants or suras of the Quran for Muslim participants, were believed to be of crucial importance. Participants explained that parents, relatives, and the social environment should assist the affected person spiritually. This spiritual assistance was described as praying in order to bless and protect the affected person and to ask for his or her salvation. It was perceived to significantly support the religious treatment and to be a crucial part of the healing process.

Participants of both Christian and Muslim faith strongly emphasized the use of sanctified or holy water in the context of a religious treatment (n = 26). Participants from Cameroon applied the French expression L’eau bénie (blessed water) and participants from Somalia used the expression Taleeth or Tahliil (holy water) for the water that was sanctified by a religious authority through the recitation of special prayers and verses of the Bible or suras of the Quran.

“One time I go to Sheikh and he gives me this small water and he reads Quran and it’s called Taleeth and I get a help inside.” (male, 19 years, Somalia)

Holy water was described as effective for curing every kind of condition, physical as well as mental distress, and also discomfort caused by supernatural forces such as possession by evil spirits. Participants from Eritrea (n = 10) in particular described the process of healing through different forms of holy water in a very detailed manner. They emphasized the importance of Mai Digam (sacred water) and Tsebel (smaller volumes of sanctified earth or water) as a universal remedy. According to the participants, depending on the specific discomfort and the instructions of a priest, Tsebel should normally be taken orally or should be used to wash the body. Furthermore, the curative effect of taking a bath in Mai Xolot (divine sources and waterfalls) was described in detail: Under the guidance of a priest, persons in need would come to ‘get their sins washed away’ while the priest cites verses of the Bible or says prayers.

“Sick people, the old, and other people who are disturbed by spirits (…) they are coming baptizing in that place. (…) in most of the places they take off their clothes and they start. (…) The priest speaks with the spirit from what he has got his sickness (…).” (male, 35 years, Eritrea)

Participants explained that a holy snake was believed to live in some special divine sources (Mai Xolot). A person who had incurred heavy guilt, regretting the sinful behaviour, and bathing in Mai Xolot, would be entangled by this snake and be freed from the guilt.

“If you are of heavy guilt, the snake will come. (…) The snake entangles this person who is full of sin. (…) when a priest questions him and he admits his sins, he will be released from this snake.” (male, 31 years, Eritrea)

Traditional treatment

Participants across the focus groups (n = 15) discussed herbal treatments and natural remedies as cures for the described symptoms. These could be made from special trees, plants, roots, or barks that have a curative effect on physical suffering, mental distress in general, or spiritual problems, such as the possession by evil spirits. Depending on the treatment, the herbal remedies can be produced and applied in diverse ways (external or internal) and forms (e.g. such as liquid or oil).

“(…) natural resources that have ever existed. (…) Some sort of trees which are considered to be blessed. They are pasteurized. They go through a process where they become in a liquid form, whereby this person drinks. (…) Sometimes they put through the nose. They put through the ears.” (male, 30 years, Somalia)

Some participants from Eritrea (n = 8) discussed inhaling frankincense or the smoke of a burned root (Kaberitcho) for persons possessed by a spirit, in order to identify the kind of spirit and its reasons for possessing the body.

“(…) when he smells Kaberitcho that spirit starts speaking (…).” (male, 35 years, Eritrea)

“He cannot cure him. You can led know what kind of spirit [it is].” (male, 29 years, Eritrea)

Another discussed form of treating the symptoms of PTSD, or problems of any kind, was a sacrificial offering performed by the affected person’s family in the country of origin. Participants from Cameroon described this as appeler au pays.

“You can call someone in your home country. They will try to look what is going wrong.” (male, 23 years, Cameroon)

“You can ask them to send money. With this money you can buy a goat or a chicken. And one will kill the chicken. Or to buy candies and do a Sadaqa [charity], share it in the neighborhood or in the village so that he can receive the blessing.” (male, 25 years, Cameroon)

The most frequently discussed form of sacrifices across the focus groups was the sacrifice of an animal by a traditional or spiritual healer. This was debated very contentiously, often interpreted as the devil’s work and witchcraft, and was rejected by most of the participants (n = 20) as a cure for mental or spiritual problems.

“Because I have seen something about in country. Maybe in bush. This, a mad person, they can’t look for a doctor. But they salute [sacrifice] a cock or a hen and then they give the sick person to that blood. Then that person returns. Become ok. (…) That is a traditional medicine. But not good. I don’t like traditional medicine.” (male, 24 years, Somalia)

The traditional or spiritual healers performing these sacrifices were described in greater detail by participants from Cameroon and Eritrea. Even though they distinguished between traditional practitioners and sorcerers, the division between the two were fluid. Participants from Cameroon referred to traditional or spiritual healers as Marabouts.

“A Marabout is like a doctor for the mind. A doctor of spirits.” (male, 26 years, Cameroon)

“He [Marabout] performs spiritual practices. He utilizes tree barks. (…). He sacrifices animals. He takes the animal’s blood.” (male, 25 years, Cameroon)

Within this context, participants used the French term faire des pratiques (to perform practices) and linked it to a traditional practice within the supernatural realm: A person could ask the Marabout to perform Grimba (magic, fetish) in order to achieve an objective, to attack someone spiritually, or protect oneself from spiritual attacks.

“They perform special practices (faire des pratiques). In order to make it work. (…) You go to the village and one will protect you with these practices, with Grimba.” (male, 25 years, Cameroon)

Participants from Eritrea discussed different types of traditional or spiritual healers. Debtera were described as spiritual or religious magical healers trained in exorcism, traditional medicine, and supernatural practices in a more religious realm. Another form of traditional healers or sorcerers were described as Tonkolti.

“These are people who possess miraculous powers. (…) They can help but they can earn a lot of money at the same time. (…) Sorcerers, Tonkolti, are people with knowledge who show you (…) possibilities to solve problems. (…) A Tonkolti can offer assistance that releases you from this curse. Perhaps even with rituals like chickens that one should circle above the head. (…) he gives you hints which kind of chicken and what color it should be. That can combat the cause.” (male, 31 years, Eritrea)

Tonkolti (sorcerers) and Debtera (spiritual healers) were described as being aware of ‘both sides’: how to heal but also how to harm and curse a person. Participants therefore emphasized the need to view these healers with caution.

“It’s like the training of a doctor, who knows the cure for an illness but also the opposite side. (…) These people who possess the knowledge to mislead you might be the same who possess the knowledge of religious faith and move deliberately against the faith in the direction of the devil worship. They might bewitch you and find the cure at the same time. And call themselves healers, saviors, and enrich themselves.” (male, 35 years, Eritrea)

In general, the reputation of traditional healers was rather divided: Some participants (n = 6) confirmed their strong belief in the healing abilities and the effectiveness of the treatment practices performed by traditional healers. Others (n = 20) strongly rejected these practices. Participants across the focus groups (n = 15) emphasized that they perceived these healers to be opposed to religion and a faith in God. They were rejected because participants suspected them either of performing the devil’s work and witchcraft or of being salesmen who only want to earn money.

“Within certain tribes these practices don’t happen anymore. And there are other tribes in our country, where these practices take place. And it happens that there are suddenly crooks. (…) And they use these rites to enrich themselves. Personally I will never solicit a Marabout because the Marabout in any case is a salesman.” (male, 28 years, Cameroon)

Medical and psychological treatment

Participants across the focus groups (n = 22) emphasized the importance of medical and psychological help for the described symptoms of PTSD. Only a small number of participants (n = 4) did not regard the described symptoms as a mental disorder with a medical or psychological indication and doubted the helpfulness of these treatments.

Seeking medical or psychological help was often described in two steps. The first step was to seek help from a religious authority or community. In a second step, and depending on the advice of the religious leader, the social environment of the affected person could consult a doctor for medical advice and treatment.

“There are the two ways: That you go to the doctor, even the doctor is an assistant of God. And I personally would seek help from the church before going to a doctor.” (male, 25 years, Eritrea)

Attitudes towards pharmacological treatment were rather divided. Some participants endorsed the curative effect of medicaments, while others rejected the idea of taking medication when facing symptoms of PTSD. This was often linked to a need to be understood and listened to.

“(…) But there are sicknesses where you are not supposed (…) to give out tablets. You see it’s like you have to feel. You have to digest the person. You have to give this person a time to listen to you.” (male, 29 years, Somalia)

In some cases, a rejection of medication was placed in a supernatural framework, e.g. when the supposed causes were spiritual possession or being attacked by a curse and participants feared that such culture-specific problems might not be understood by Western practitioners.

“The spirit who possesses your body will talk to you. (…) It will tell you why it has possessed your body. (…) The possessed will not go to a doctor. If he gets a syringe he will die immediately. That is why these people don’t go to doctors, hospitals. They go directly to the sacred waters of Mai Xolot.” (male, 35 years Eritrea)

“My friend, (…) his wife had a problem. And he tells the social [social worker]. But they didn’t believe him. (…) But they don’t understand what is this problem. This is Djinn [spirit]. You need another prescription. Such like Quran.” (male, 25 years, Somalia)

Participants (n = 13) expressed further obstacles to seeking medical or psychotherapeutic treatment. Some participants worried that their traumatic experiences could be too burdensome for German practitioners. Furthermore, they described that practitioners might not be able to comprehend their account because of the foreign and divergent experiences and realities of life.

“Is like what our people are going through is insensitive to the people who own this land. (…) When someone is talking to a psychologist (…). And he tells everything. It’s very hard for this doctor to digest. But we as people, we feel the pain. We know what this person is talking about. (…) But if I try to (…) explain this to the psychologist. It’s very hard for him to digest. What I am telling him.” (male, 30 years, Somalia)

Furthermore, participants described difficulties regarding the functioning of an unfamiliar health care system and orienting themselves within this system.

“There are so many obstacles. This society works completely differently. (…) Who is my contact person regarding this inner anxiety? Do you tell everybody? Or do you tell just your gynecologist? Or the general practitioner?” (female, 34 years, Eritrea)

“If I wish to go to a doctor I have to pay. This is a real problem. I will prefer to stay at home with my problems. Because (…) the doctor will say (…) your treatment will take maybe three to six months, which will cost you 3000 Euros.” (male, 28 years, Cameroon)

Discussion

The present study aimed to provide insights into beliefs about cures for PTSD held by Sub-Saharan African asylum seekers residing in Germany. Moreover, the help-seeking intentions of this group were compared to those of Germans without a migration background. To address this research issue, we used a quantitative and qualitative methodological triangulation strategy.

We found that asylum seekers are likely to seek religious, medical, and psychological treatment for PTSD as well as help from their parent or partner. The initial hypothesis was supported: Whilst asylum seekers indicated higher intentions to seek religious help and treatment by general practitioners, Germans without a migration background showed greater intentions to seek psychological and traditional treatments. In the qualitative part of the study, different forms of treatments were proposed by the asylum seekers, which were grouped into three superordinate themes: (a) religious treatment, (b) traditional treatment, and (c) medical and psychological treatment.

The support of family and friends was equally as important to both groups. Only the reliance on parents was more important for asylum seekers than for Germans without a migration background. However, as asylum seekers in the present study were on average ten years younger than the German participants, this may have influenced this finding despite the statistical control for influences of age. The results of the quantitative part are in line with the findings of the qualitative part of the study, as asylum seekers emphasized that healing must be understood as embedded within a social context. Seeking treatment can therefore be interpreted as a social act that involves the affected person’s social environment and needs to be initiated by parents, relatives, or community elders. This corresponds to previous research reporting that Eritrean and Somali refugees perceive trusted friends and family to be responsible for recognizing and attempting to find treatment for mental health problems (Bettmann, Penney, Clarkson Freeman, & Lecy, 2015; Melamed et al., 2019). Community and family cohesion should therefore be considered as crucial elements of recovery (Schnyder et al., 2016) and should be taken into account when treating immigrant and refugee populations from Sub-Saharan Africa (Baubet & Moro, 2013; Ehntholt & Yule, 2006; Murray, Davidson, & Schweitzer, 2010). However, as Melamed et al. (2019) point out, the reliance on friends and family for initiating treatment might also pose a barrier to treatment seeking for asylum seekers, who may be living far away from their families and may not yet have trusting relationships around them.

Asylum seekers showed a higher intention to seek help from religious authorities and strongly emphasized the importance of religion for recovery and the predominantly religious character of treatment. Moreover, medical and psychotherapeutic help-seeking seems to be mediated by religious gatekeepers and authorities. Therefore, it is advisable to offer to incorporate a religious dimension into psychotherapy, which might enhance patients’ therapy motivation, engagement in treatment, and their feeling of being understood and accepted (Markova & Sandal, 2016; Slewa-Younan et al., 2017). In this regard, Whitley (2012) equates cultural competence to religious competence. Like the cultural orientation, religious orientation influences patients’ beliefs, values, attitudes, and conventions. Koenig (2008) proposes that clinicians can respectfully address patient’s religious needs by assessing a spiritual history and engaging in appropriate consultation with clergy.

In contrast to previous research (Fenta et al., 2006; Palmer, 2006), asylum seekers in our study expressed a lower intention to seek help from traditional treatment practices than participants without a migration background. Our finding might be explained by divergent concepts of traditional treatments. Participants without a migration background might have interpreted traditional treatment as natural remedies or homoeopathic medicine. Indeed, previous studies have found that Western populations view the treatment of mental illness with vitamins and herbs more positively than treatment with psychotropic medication (Angermeyer & Matschinger, 1996; Jorm et al., 1997). The asylum seekers in our study showed a multifaceted and rather broad picture of culturally accepted traditional treatments: While natural treatments with herbal remedies were considered as appropriate, supernatural and magical practices performed by traditional healers or sorcerers were mainly rejected.

Asylum seekers’ intentions to seek help from general practitioners was higher as those of Germans without a migration background, a finding which corresponds to previous research (Maier, Schmidt, & Mueller, 2010; Papadopoulos et al., 2004). These findings are relevant for general practitioners, who often have the main responsibility for guiding asylum seekers within Western health care systems (Varvin & Aasland, 2009). Given that the investigated asylum seekers explained that culture-specific illness beliefs lead to a reluctance to take medicaments, transcultural training for general practitioners and health care staff might be helpful to ensure a culturally sensitive handling of refugee populations.

With regard to psychological treatment, high average mean scores in the group of asylum seekers, and particularly in those from Cameroon and Somalia, suggest a high likelihood of seeking help from this source. However, their intention was lower than that of Germans without a migration background. This is in accordance with previous studies, which reported low rates of mental health care utilization in immigrant and refugee populations (Ellis et al., 2010; Fenta et al., 2006; Kirmayer et al., 2011; Palmer, 2006). Furthermore, asylum seekers’ intention to not seek help at all was higher compared to the German participants. This was especially the case in participants from Eritrea and in asylum seekers with stronger posttraumatic symptoms. These results are in line with previous research reporting a low uptake of mental health care in Iraqi refugees with higher levels of PTSD symptomatology (Slewa-Younan et al., 2015). A potential explanation might lie in the finding that individuals who frequently use experiential avoidance and avoidant coping strategies may be at greatest risk of increasing their PTSD symptoms through strong avoidance behaviours (Orcutt, Pickett, & Pope, 2005; Pineles et al., 2011). As they might expect that seeking help will entail them being confronted with their traumatic experiences, the increased avoidance behaviour might also result in reduced help-seeking behaviour. In conclusion, asylum seekers from Eritrea with a higher symptom load might constitute a particularly vulnerable group among asylum seekers in Germany.

Additionally, and in line with previous research, asylum seekers in our study expressed a lack of knowledge and orientation regarding available mental health care services, and concerns that their problems might not be understood due to cultural distance (De Anstiss & Ziaian, 2010; Donnelly et al., 2011; Kirmayer et al., 2011; Sandhu et al., 2013). To reduce these barriers, it may be useful to inform newly arrived asylum seekers about the functioning and locations of different sources of help, or to support them with booking appointments (Bhatia & Wallace, 2007; Mewes & Reich, 2016).

Limitations

We investigated a diverse group of asylum seekers with regard to countries of origin, cultural groups of Sub-Saharan Africa, and religion. Although we accounted for differences regarding the main countries of origin and several sociodemographic factors, help-seeking intentions may still vary due to differences in other characteristics. Due to the limited number of participants from the respective countries of origin, caution is warranted in drawing conclusions about the impact of culture on participants’ help-seeking intentions and the generalizability thereof.

We did not control for prior treatment experience, and the two groups differed in terms of traumatic experiences and PTSD symptoms, which may have led to different answers regarding help-seeking intentions.

Furthermore, the two investigated groups differed in terms of age, gender, education, and religion. While we took these differences into account and included several variables in the analytical model, conclusions about the group differences in terms of culture still remain limited.

With regard to the qualitative part of the study, focus group discussions with predominantly male participants were organized and moderated by a White, female member of the majority society, which may have induced a response bias and social desirability. Due to these differences or a fear of stigmatization, some participants may have been reticent to express criticism of the local health care system or culture-specific opinions and practices.

In the present study, we focused on help-seeking intentions and beliefs about cures of PTSD of asylum seekers from Sub-Saharan Africa. However, it would have been interesting to incorporate the German participants’ perspective within focus group discussions as well, especially with regard to the relatively high endorsement of alternative and traditional treatment for PTSD. The study leaves this important topic open for future research.

The present study incorporates a rather male perspective on cures for PTSD, as participants were predominantly male, and an impact of gender on our results can therefore not be ruled out. For future research, it would be interesting to investigate gender differences in beliefs about cures for PTSD.

Conclusion

Our findings underline the need for practitioners to explore cultural and religious frameworks of healing and recovery in order to demonstrate understanding and acceptance of varying cultural contexts in which treatments can happen (Sturm et al., 2010). This can be achieved by addressing a patient’s cultural and religious needs. Clinicians may also encourage patients to engage in communal networks associated with their religious congregation and mobilize patients’ religious resources to promote resilience, recovery, and healing (Whitley, 2012). Furthermore, healthcare providers could target those in need by developing public health campaigns in collaboration with religious and cultural communities, which might help disentangling the potential for religiously reinforced mental health stigma (Peteet, 2019). In addition, this might help explaining the functioning of the local health care system, which might in turn facilitate the access for asylum seekers and refugees.