--> Europe's Journal of Psychology ejop.psychopen.eu | 1841-0413 Research Reports Changes in Disease Perception, Coping Strategies and Diagnoses in the Case of First and Fourth Generations of Turkish Migrants in Germany

This study examines to what extent subjective illness concepts and symptoms of Turkish patients in Germany have changed over 40 years. Two groups of patients of Turkish origin from psychosomatic clinics in Germany took part in the study, the first group made up of 690 patients from the "first generation" of migrants and the second group comprising 350 patients from the "fourth generation". The study was conducted using standardised interviews ( The sociodemographic questionnaire, IPQ, COPE, BDI, SCL-90-R ). Differences were found in essential aspects of subjective illness perception and illness coping strategies as well as in the diagnosis of psychological illness. The fourth generation of Turkish patients that were born and have grown up in Germany showed more problem-oriented behaviour regarding their coping strategies and less use of dysfunctional strategies for treatment than patients of the first generation. The findings of the present study are important for treating Turkish patients and patients of Turkish origin suffering from psychological disorders since they indicate that their disease perception, coping strategies and diagnoses undergo changes in the course of generations. Hitherto these differences were not taken into account in treatment; in the future we should be moving towards customised treatment in order to achieve better therapeutic effects.

The medical-psychological perspective on migration is dominated by our conception of modern societies and recent, post-WWII migration causing individual and collective strains (Sluzki, 2010).Migration results in significant shifts in the psycho-social structure of any society and influences its socio-economic stratification, politics, culture, and public health system (Kizilhan & Bermejo, 2009).This change in a person's life has an influence on individual and collective identity as well as on the way the past is "digested" and on whether or not integration into one's host society is successful.The shaping of this new phase in life is dependent on individual and collective coping strategies and on the ability to make use of social networks (Sluzki, 2010).Establishing a new network of relationships in a culturally, ethnically and socially different context demands new social resources, a new orientation and new competences to take action (Kizilhan, Haag, & Bengel, 2011).
For example, in Turkey, traditional notions on family and responsibilities of the individual within this network were clearly defined and compliance with the given rules and norms controlled by the collective.On the other hand, attitudes towards these standards and values have changed in various ways with subsequent generations through migration; this also applies to the perception and treatment of illnesses (Özbek, 2006).However, in psychiatric and psychotherapeutic treatment, this aspect is not being taken into consideration sufficiently (Franz et al., 2007), --> Europe's Journal of Psychology ejop.psychopen.eu| 1841-0413 although it has been shown that notions on health differ significantly across cultures and generations (Barnes et al., 2004).The connection between attitudes on health and illness, on the one hand, and cultural norms, on the other, are particularly strong, which also means that any statement about an illness is culture-specific to a certain degree (Thomas et al., 2003).
The association of religious and magic notions (such as the existence of spirits, jinn, symbols or rituals) on the one hand and illness on the other is still of great relevance to many first generation migrants of Turkish origin (Heine & Assion, 2005).Leventhal, Meyer, & Nerenz (1980) assume a general process leading to the development, change and medical consequences of subjective illness concepts.According to a perceptual-cognitive model (Leventhal, Leventhal, & Cameron, 2001) there are cognitive and emotional illness representations that are significant for compliance, the course of an illness, and the effectiveness of its treatment.
In past studies exploring the subjective illness concepts of patients of Turkish origin, the cultural aspect was considered to be a relatively stable factor.Thus, the pilot study by Franz et al. (2007) among Turkish and German patients concluded that patients of Turkish origin tend to have more pessimistic notions of illnesses and, in contrast to German patients, would rather attribute them to external causes.Significant differences in coping behaviour between German and Turkish patients were also uncovered by Özelsel (1994): German patients preferred "emotional retreat" or "further mental digestion", while patients of Turkish origin tended to "play down" or "minimize" an illness.Patients who managed to "further mentally digest", were prepared to acknowledge information about their illness; patients who played it down were unable to cope with it.
Another comparative study between German patients on the one hand and Turkish patients of the first and second generations, on the other, also showed that Turkish patients more often attribute their illnesses to external causes such as "fate" or "God's will", they feel that they have less personal control over their treatment and have a significantly poorer understanding of their illness (Bircheneder, 2010).Unlike German patients, in order to cope with their illness, they do not apply a systematic long-term strategy, but try to act in a short-term, problem-oriented manner (Bircheneder, 2010, Kizilhan & Haag, 2011).
The study by Bircheneder (2010), which was conducted at the same psychosomatic clinics as the present research, gave us to the idea of comparing a German sample with first and fourth generations of Turkish patients in Germany.
The "first generation" of Turkish patients was defined as persons who had been socialised in their home country and had migrated to Germany as adults.The "second generation" was born in their home country and migrated as children to Germany.The "third" and "fourth generation" are made up of people born and raised in Germany.
For a better difference and comparability we choose to compare the first and fourth generation in this study; for the latter, parents had migrated to Germany as children (up to the age of 12).
Due to our experience in psychosomatic clinics with German and Turkish patients, we expected a difference in illness conceptions, coping strategies and diagnoses between the first and the fourth generation of Turkish patients.
Further we expected that the fourth generation of Turkish patients would show similar illness conceptions, coping strategies and diagnoses to a German group of equivalent age.

Method Sample and Data Collection
We investigated patients of Turkish origin in three psychosomatic hospitals in Bavaria and Baden-Württemberg.
In all three hospitals they were treated using single and group therapy sessions in conjunction with psycho-educative seminars.
Data collection in three psychosomatic hospitals was conducted between June 2007 and July 2011.We selected this long period of time because despite the clinics' capacity, patients of Turkish origin accounted only for a small share of the total number of patients treated by these clinics.For the first generation group, patients of Turkish origin were chosen who had been born and raised in their country of origin and had migrated to Germany as adults (i.e.above 23 years of age).The fourth generation group comprised patients of Turkish origin who had been born and raised in Germany and whose parents had migrated to Germany as children (i.e. at up to 12 years of age).Patients with war experiences, traumatisation or other extreme burdens, unclear residence status and severely suicidal patients were excluded from the investigation.
Due to the expected differences between the two groups with respect to gender, age, level of education and diagnosis, the two samples were parallelised in order to ensure comparability and to enhance the inner validity of the study.Thus, patients under 21 and older than 60, who were more frequent in the first generation group, were excluded.Due to noteworthy generational differences, average age was significant (F=7,209, p <.008) in the first group.Furthermore, all patients who were illiterate, had a university degree or had been treated in a psychosomatic hospital in the year of the study were excluded as well.Finally, we excluded patients who had been diagnosed with PTSD due to torture or displacement, as the fourth generation group barely showed this syndrome.
Overall, 720 patients of the first generation group and 372 of the fourth generation group participated in the survey.
Due to missing data and incorrectly filled-in questionnaires, a final 690 first generation and 350 fourth generation participants were considered for analysis.As mentioned in the introduction, we also compared these two groups to a sample of German patients (N=34) concerning their illness concepts, coping strategies and diagnoses.

Demographics
Average age for the first generation group was 57.9 years, and for the fourth generation group it was 27.9 years.
Gender distribution was 40% men and 60% women for both groups.Patients of the first generation group showed a significantly lower level of education (χ 2 =36.00, p<.001), a lower percentage of gainful employment (χ 2 =3.933, p<.047), and they were on a lower income level than their fourth generation counterparts.Persons of the first generation group were also married more often (χ 2 =10.394, p<.015) and had more children (p<.003).While this group had an average of 3.5 children, the mean value for the fourth generation participants was only 1.7.

Measures
The following test instruments were used in the Turkish version: The sociodemographic questionnaire by Koch (1997): This questionnaire has been in use in various surveys (Koch, 2005).It generates important background information about a person (e.g.place of origin, religion, migration history, family, work, financial situation, duration of illness, treatment, etc.).Religious affiliation and practice were examined using specific questions on "Religion and Belief", an autonomous category integrated into this questionnaire.
Illness Perception Questionnaire (IPQ-R): For the operationalisation of the subjective illness concept, we used the Turkish translation (Armay, 2006) of the revised version of the "Illness Perception Questionnaire" (Moss-Morris et al., 2002) and expanded it according to the modification proposed by Lujic (2008) by adding the factors "Fate", "Environment" and "Personal".It consists of eight sub-scales showing the components of illness perception as postulated by Leventhal et al. (1980).Overall, the Turkish translation of the IPQ-R is of satisfactory psychometric quality with respect to the reliability of the sub-scales (Cronbach-α between .68 and .89).
COPE-Inventory: COPE was designed to cover a wide range of both functional and dysfunctional coping strategies (Carver et al., 1989).With the exception of two scales, the range of reliability values was between good and very good (Cronbach-α between .75 to .96).For the collection of coping strategies, we used the Turkish translation of COPE (Ağargün, Beşiroğlu, Kıran, Özer, & Kara, 2005), whose validation resulted in high internal consistency values of around 0.79.

Beck Depression Inventory (BDI):
The degree of depressive symptoms was registered with the Beck Depression Inventory (Beck et al., 1961) in this case its Turkish version (Hisli, 1988;Hisli, 1989).The validation of the questionnaire resulted in a high internal consistency of the Turkish version, with values of around .85.There is a range of validation studies for this version, assuming a very valid and reliable record of the construct "depression" with patients of Turkish origin (Aktürk et al., 2005).The cut-off values of 11 and 18 also apply to the Turkish version.

Symptom Checklist (SCL-90-R):
The Symptom Checklist (Derogatis, 1977) measures the subjectively perceived impairment of a person through physical and psychological symptoms within a timeframe of seven days.The Turkish version of SCL-90-R, previously tested for its reliability and validity by Dag (1991), was used for this research.

Statistical Analyses
The data was analysed with SPSS 15.0.1 for Windows (2006).Patients of the first and fourth generation were compared with regard to their illness perception (IPQ-R) and coping strategies (COPE) by calculating χ 2 tests, t tests and with the help of a univariate, factorial analysis of variance (ANOVA).The correlations between IPQ-R and COPE, on the one hand, and between IPQ and strain (BDI and SCL-90-R), on the other, were calculated separately for each group.The modified cause scale of the IPQ-R (Lujic, 2008) was checked using reliability analysis.

Psychological Syndrome
As presented in Figure 1, significant differences could be found for somatoform disorders, with the first generation group being more highly affected (28.6%) compared to 9.2% for fourth generation migrants (χ 2 =10.28, p<.036).
The relationship was reversed for the diagnosis of personality disorders (χ

Stress Symptom
Table 1 shows average differences in depressiveness (BDI) and in psychological stress symptom (SCL-90-R).
All nine sub-scales of SCL-90-R, plus the overall stress value (GSI) show significantly higher values for the first generation group (p<.001).First generation participants also reported much higher values for depression (BDI value at 36.4 points) than their fourth generation counterparts (BDI value at 20.6 points).With a t value of -3.746, this difference can be considered highly significant (p<.001).
Concerning the relation between depression and gender, women of the first generation group showed the highest value (56.6%) (Figure 2).Among men of the same group, 52.3% also featured high values.In the younger group, 29.7% of women and 24.8% of men reached higher values.

Causes of Subjective Illness
Figure 3 show that first generation patients of Turkish origin attribute their illnesses to external causes much more frequently than fourth generation patients.
The Turkish patients of the first generation believe to have less control above their illness ("destiny factor"; F=16.044, p<.001) and less influence on the treatment ("environmental factor"; F=8.408, p<.005).
Both groups often perceive stress and exhaustion, but also their personality and family affairs, as causes of their illness.The results of the fourth generation group are consistent with the ones of the German groups in the 2010 study discussed above (Bircheneder, 2010).
Compared to fourth generation patients, first generation participants feel they have no control over their illness (F=17.183,p<.001; d=1.2) and little control over its treatment (F=17.214,p<.001; d=1.05).
Using the sub-scale "coherence" of the IPQ-R, we examined to what extent the patients have an understanding of their illness.There is a noteworthy difference between the two groups in this aspect (F=4.874,p<.015; d=0.54): first generation patients were significantly less capable of understanding their illness (F=4.874,p<.015; d=0.54).

Illness Coping Strategies
Table 2 shows that the coping strategies "planning" (F=5.652,p<.022) and "active coping" (F=10.316,p<.002) were applied significantly more often by the members of the fourth generation group.These individuals tended to think more about their illness and develop strategies and plans of action in order to overcome the problem.The strategies "patience" and "instrumental support" showed no significant differences between first and fourth generation respondents.In their efforts to cope with their illness, patients of the first generation group were more likely to turn to religious belief (F=10.120,p<.008; d=0.12) than their fourth generation counterparts.The factor "education" also generates a strong effect (d=0.18) and it is exclusively significant for the coping strategy "humour" (F=5.088,p<.015; d=0.12).Patients in both groups with a high level of education try to tackle their illness with humour.The strategy "denial" is applied more often by the fourth generation group members (F=5.641, p<.010; d=0.21).The interaction between the two factors religion and education is significant (F=4.913,p<.022): highly educated member of the fourth generation as well as patients of the first generation with strong religious bonds resort to this coping strategy more frequently.
The coping mechanisms "behaviour-oriented escape", "mental give-up", "use of alcohol and other drugs" as well as "focus on and abreaction of emotions" are seen by fourth generation as particularly dysfunctional for successful coping with illnesses."Behaviour-oriented escape" as a strategy is applied more often by the fourth generation group (F=8.763,p<.005; d=0.52).Also noteworthy is the interaction of the factors religion and education (F=5.530,p<.021).In particular, fourth generation patients with strong religious bonds apply this coping strategy, while first generation patients who make use of it tend to have a lower level of education.

Discussion
The present study examined 650 patients of the first generation and 350 patients of the fourth generation of Turkish migrants -and occasionally compared them also to 34 German patients in the same age range as fourth generation members -hospitalised at psychosomatic clinics in Germany for possible differences in illness perception, in coping with illness and with regard to psychological disorders.
The results of this study indicated that patients of Turkish origin of the first generation have a lower level of school education and lesser proficiency in German than the fourth generation group.They practise religious rituals and behaviour much more often in everyday life than their younger counterparts.
Further, the traditional and family-oriented first generation of patients perceive the causes of their illness more often to be the result of external influences such as "destiny" or "God's will", i.e. they attribute their illness more often to external causes in general, feel that they have significantly less personal and treatment control, and have a much poorer understanding of their illness.In particular, the "fate factor", measured using IPQ-R, was found to be more significant (F=16.044,p<.001) than the "environment factor" (F=8.408,p<.005).
In order to cope with their illness, fourth generation patients are more likely to apply the planning strategy (F=5.652,p<.022) and try to respond actively (F=10.316,p<.002) and in a problem-oriented way.Fourth generation patients resemble German patients in the way they tackle illness (Bircheneder, 2010).While no notable eating or personality disorders could be identified among first generation illness syndromes, fourth generation patients reported illnesses similar to those of German patients.
In other studies in the field of cross-cultural psychology, including the study on schizophrenia carried out by al-Issa (1995) and Cantor-Graae and Selten (2005), it was assumed that psychological disturbances undergo changes in different generations of migrants.Backed by studies on schizophrenia and migration (al-Issa, 1995;Cantor-Graae & Selten, 2005), our findings show that the risk of schizophrenia is higher for the fourth than the first generation group.The possibility that the differences in illness perception, illness digestion and varying diagnoses are due to the age of respondents alone was already ruled out in the study by Nickel et al. (2006).They reported for instance that patients of Turkish origin of different ages with similar cultural backgrounds and length of stay in Germany showed no significant differences with regard to the frequency of depression, anxiety, somatoform disorders, eating disorders or personality disorders (Nickel et al., 2006).
The findings of this study have significance for the treatment of patients of Turkish origin of different generations in psychiatric and psychosomatic hospitals.We found that first, due to their specific cultural background, first generation patients of Turkish origin develop a different understanding of their illness and different ways of coping with it, and second, they undergo change over generations in conjunction with different cultural influences.While somatisation disorders and depression are predominant among first generation patients in psychosomatic hospitals, eating and personality disorders, among others, are more prevalent among patients of the fourth generation.As yet, these differences have not been considered yet and will probably be dealt with prospectively with regard to group-oriented treatments in order to get better therapeutic results.
As for the limitations of our study, it should be noted that despite distinct differences between the two generations in terms of age and socialisation, due to their common origin, it was not possible to exclude all cultural factors.As such, respondents came from different ethnic and religious groups in Turkey.Also, the study was conducted only at psychosomatic clinics and there are no data here about ambulatory patients.In future studies, it would be important to compare respondents of different generations from Turkey with those living in Germany.
162, p<001), which were present more frequently in the fourth generation group.Diagnoses of schizophrenia tended to be more frequent for the fourth rather than the first generation.

Figure 1 .
Figure 1.Diagnoses for the Groups "First and Fourth Generation"

Figure 2 .
Figure 2. Depressiveness, classified into "First and Fourth Generation" as well as gender, measured with BDI.

Table 1
Comparing Means (t Tests) of Both Groups with Regard to Depressiveness (BDI) and Stress Symptom (SCL-90-R)

Table 2 F
Values and Significance of the Two-factorial Analyses of Variance Conducted on the 15 Coping Scales of COPE.