Albanian Migrants in Cyclades: Contact with Mental Health Services and Implications for Practice

Abstract

The purpose of this study was to examine the potential differences between Albanian migrants and Greeks in the islands of Paros and Antiparos with regard to seeking help from the local public mental health service, namely the Mobile Mental Health Unit of NE Cyclades Islands (EPAPSY-NGO). The study’s instruments include the Global Assessment of Functioning (GAF) and a questionnaire for recording psychosocial profile data and information concerning contact with the services. The results showed that significantly less Albanian migrants contacted the mental health services in the past in comparison to Greeks. When using the service, the Albanians were more likely to see a psychiatrist in the Unit, rather than a psychologist-psychotherapist, and they more often received prescriptions for medication. There was a significantly higher dropout rate among Albanian migrants. This study highlights the different aspects of access and use of mental health services among Albanian migrants compared with native residents. Further research should focus on the factors related to early dropouts and difficulties accessing mental health services in rural areas, in order to develop more focused and effective interventions and improve the quality of care provided.

1. Introduction

Since 1990, Greece has been receiving a mass inflow of migrants and refugees, coming from Europe, other Mediterranean countries, Asia, and Africa. Serving as an intermediate destination towards European Union countries, from the Eastern Mediterranean, North Africa and Asia, Greece also borders Balkan countries that have faced various forms of socio-political unrest, which has led to even larger waves of migration.

A significant percentage of migrants in Greece is of Albanian origin [1,2,3], living not only in urban but in rural areas too, such as the Cyclades Islands, where they are employed in construction and tourism sector [1].

Τhe use of health care services by migrants has been the subject of various studies, while special emphasis has been given to non-urban areas due to the greater difficulties of access (limited number of services available in rural areas and/or difficulty of travelling to services in urban areas due to geographical distance) [4,5]. Most studies conducted in Europe and America regarding the access and use of health services by migrants highlight the inequality and complexity of the difficulties faced by foreigners in obtaining appropriate treatment and care [6,7,8,9,10]. Recent evidence also shows unmet health care needs for migrants, especially in the area of mental health, emphasizing the role of legal, communication and language barriers, as well as the role of discrimination [11,12].

Several studies in Canada have also shown that migrants and refugees are less likely to seek help and get a referral to mental health services than natives, even when experiencing high levels of discomfort [13,14,15,16]. This may be associated with factors related to new living conditions, the health care system, as well as cultural issues [17,18,19]. Other studies on the use of mental health services have highlighted the stigma associated with mental health problems as a significant barrier to seeking help [20].

A study conducted in 16 European countries, including Greece, examined the difficulties reported by health professionals (in primary health care hospitals and community mental health centers), regarding the provision of services to migrants [21]. Among the key issues identified were:

• language-related difficulties,
• migrants’lack of familiarity with the national health system,
• the provision of services to migrants who do not have the necessary documents,
• social deprivation and traumatic experiences for certain categories of migrants,
• the different ways of understanding the disease and treatment in different culture settings,
• negative attitudes of staff towards migrants,
• difficulties of specialists regarding diagnostic issues in migrants,
• the increased risk of marginalization in migrants [22].

In general, the difficulties concerning the access of health services by the migrants (the majority of whom are of Albanian origin) have been highlighted in other Greek studies as well. A survey among immigrants (N = 191) in Athens, 27% of whom were of Albanian origin, showed that only 20% reported having a good knowledge of services, 60% having moderate knowledge and 19% having poor knowledge [23]. The degree of knowledge of the Greek language seemed to be a key factor 50% had used public services during the last year, while more than half (53%) reported having great difficulty accessing services. Moreover, more than half of the participants (62%) stated that their needs were not met by health services, mainly due to long waiting lists in hospitals, difficulties in communicating with health professionals, high costs and complexity of the health system.

With respect to the use of mental health services by immigrants in Greece, there is extremely limited published data. Specifically, psychopathology and use of mental health services were studied in immigrant families who visited a community mental health service in Athens, with a request concerning their children [24]. No significant differences were found regarding psychopathology (frequency of specific diagnoses), as well as the use of psychiatric services. However, it seems that, in comparison to the native population, immigrant families did not seek help οn time.

The aforementioned findings highlight the need for further research regarding difficulties related to the contact of migrants with mental health services (e.g., timely seeking help, engagement in treatment, etc.), in order to develop specific interventions to tackle the issue effectively.

The present study was carried out in two islands of Cyclades, Paros and Antiparos. It was part of a large-scale research also examining the possible links between psychopathology, family functioning and quality of life in Albanian and Greek beneficiaries of the Mental Health Mobile Units [25]. The population of Paros and Antiparos islands comprise of 13,750 and 1211 residents respectively [3]. In 2015, approximately 1/5 of the local population were migrants from Albania, who have permanently migrated in the area [26].

The study was conducted in these two islands due to the relatively great number of Albanian immigrants living there and due to the significant number of Albanians being treated for psychosocial difficulties by the only mental health service operating in the area the last 17 years, the Mental Health Mobile Unit of NE Cyclades Islands (EPAPSY-NGO) [27,28]. The Mental Health Mobile Units were established during the psychiatric reform in Greece, in order to offer services in rural areas, where other mental health services are limited, according to the main principles of community psychiatry [27,29,30]. Services provided by the Mobile Units include diagnostic assessment and therapy for psychiatric disorders and psychosocial difficulties, in collaboration with local health and social services. Moreover, mental health promotion and assessment of needs are included in the basic aims of the Units.

The data provided by the records of the Mobile Unit showed that Albanian immigrants consisted the 10% of the total number of service beneficiaries. Cyclades Islands experience a lack of specialised mental health services focused on migrants [27]. The present study is the first to examine migrants’ use of mental health services in the islands, with an aim to acknowledge the emerging difficulties, in order to develop appropriate strategies and interventions, tailored to the needs of beneficiaries of different cultural backgrounds.

The aim of the present study was to examine possible differences between Albanians and Greeks regarding their contact with mental health services (previous contact with a mental health professional, referral source, treatment plan, course of treatment) in the Cyclades.

2. Methods

2.1. Sample

The sample comprised 204 adult beneficiaries, 105 (51.5%) Greeks and 99 (48.5%) Albanians, who had an appointment for the first time with a mental health professional in the Mental Health Mobile Unit in Paros and Antiparos islands, during the years 2012–2016. The mean age in Greeks was 38.5 years (SD = 9.5) and in Albanians 37.7 (SD = 10.2). 66.7% were women in both ethnicities. Service users with a diagnosis of psychosis, developmental disorder and organic mental disorder were excluded. Those presenting affective disorders, neurotic, stress-related and somatoform disorders, as well as service users who belonged to the category Z03–Z99 ‘Factors influencing health status and contact with health services’ according to ICD-10 Classification of Diseases [31], were included in the study. Most beneficiaries with a diagnosis belonging in the category Z03–Z99 sought help from the service due to psychosocial problems (family conflict, social and financial problems, members of family with severe mental or physical disorders etc) [27].

According to the data recorded by the Mobile Unit, the last 5 years before the present study the analogy of Albanian immigrants to Greeks contacting the Unit for the first time was 1/10. There was an attempt to include in the study all Albanian beneficiaries seeking help from the Unit for the first time during the period 2012–2016, fulfilling the inclusive criteria. A total of 116 service users of Albanian origin received information concerning the present research project and were asked to take part. 99 took part in the study (the response rate was 85%). In Greeks one service user in every 10 seeking help from the Unit for the first time during 2012–2016 received information about the study and was invited to take part. A total of 119 Greeks were informed and 105 finally participated (88% response rate).

2.2. Measures

The following measures were used:

• Global Assessment of Functioning (GAF, [32]): Τhe measure has been translated into Greek, as a modified form of the Global Assessment Scale, with satisfactory psychometric properties [33]. Higher score indicates better level of functioning. In the present study, functioning was assessed by the mental health professional assigned with the case, at the beginning and at the end of treatment or the end of the study (if treatment was ongoing).
• Questionnaire for recording psychosocial profile data and contact with mental health services: This questionnaire is a special adaptation of the related questionnaire used by the Mobile Units of EPAPSY for the recording of data concerning service users. It was completed by the psychiatrist or clinical psychologist who made the initial assessment for each participant. Specifically, it consists of 24 items, recording the following data:
  • Socio-demographic variables.
  • Psychiatric diagnosis according to ICD-10 [31].
  • Data on the use of mental health services: source of referral (the service or professional making the initial referral to the Mobile Unit), source of information for the Mobile Unit, if they had visited a mental health professional in the past, the age of first visit to a mental health professional, psychiatric hospitalizations.
  • Treatment plan and course of treatment: the kind of therapist with whom they had the first contact in the Unit, treatment plan, medication before starting treatment in the Unit, medication as part of the treatment plan in the Unit, treatment status at the end of this research (ongoing treatment, dropout, completion of treatment), total number of sessions with a mental health specialist at the Mobile Unit.

2.3. Procedure

The mental health professional who made the initial assessment completed the questionnaire for the psychosocial profile and contact with mental health services. In this questionnaire, diagnosis given by the psychiatrist, based on the ICD-10 diagnostic system [31] was also included. In addition, the mental health professional recorded the GAF score [32] for each participant. This scale was completed twice: once after the initial clinical evaluation and once at the completion or discontinuation of treatment or at the time of completion of the study (while treatment was ongoing). The total number of sessions, the treatment status and the treatment plan implemented, were completed by the same professional at the end of the study, based on the records kept in the service.

All beneficiaries taking part in the study gave informed consent for their participation in the current research project. Ethical approval was obtained by the Institutional Board of EPAPSY. The present research project was conducted according to the ethical standards delineated in the 1964 Declaration of Helsinki.

Data was collected during the years 2012–2016.

2.4. Statistical Analysis

Cross sectional analysis calculating mean scores, standard deviation (SD), median and interquartile range were used to describe quantitative variables. Frequencies were used to describe qualitative variables. Pearson’s χ² test or Fisher’s exact test were used to compare analogies. Student’s t-test or non-parametric Mann-Whitney were used to compare quantitative variables between the two groups. Pearson or Spearman r were used to test the relationship between two quantitative variables. Kruskal-Wallis non-parametric test was used to compare quantitative variables between more than two groups. IBM SPSS (IBM Corp., Armonk, NY, USA) ver. 19.0 was used to analyse the data.

3. Results

Most service users participating in the study were of female gender in Greeks and Albanians (66.7%). The mean age for Greeks was 38.5 years (SD = 9.5) and for Albanians 37.7 years (SD = 10.2). The level of education was significantly higher in Greek than Albanian participants. No statistically significant difference was found in family status between the two ethnic groups. Greeks were employed full time in a greater percentage than Albanians (56.3% versus 35.4%). There was a higher level of unemployment in Albanians, compared to Greeks (23.2% versus 17.5%). Albanians more often than Greeks were employed part-time (11.7% versus 26.3%), to a statistically significant level (p = 0.001). Moreover, significantly more service users of Albanian origin than those of Greek origin had no social insurance (24.2% versus 2.9%, p < 0.001). Basic sociodemographic characteristics are included in

Table 1. Basic sociodemographic characteristics.

		Ethnicity				P Pearson’s χ2 Test
		Greeks		Albanians
		N	%	N	%
Gender	Male	35	33.3	33	33.3	1000
	Female	70	66.7	66	66.7
Age (mean, SD)		38.5 (9.5)		37.7 (10.2)		0.563 +
Educational Level	Primary School	1	1.0	18	18.2	<0.001 *
	High School	6	5.7	38	38.4
	Lyceum	49	46.7	34	34.3
	Technological Studies (after Lyceum)	30	28.6	4	4.0
	University	19	18.1	5	5.0
Family Status	Single	22	21.0	15	15.2	0.228 *
	Married	73	69.5	72	72.7
	Separated	7	6.7	12	12.1
	Widowed	2	1.9	0	0.0
	Cohabiting	1	1.0	0	0.0
						P Fisher’s Exact Test
Employment Status	Full-time job	58	56.3	35	35.4	0.001
	Part-time job	12	11.7	26	26.3
	Student	1	1.0	3	3.0
	Housewives	8	7.8	12	12.1
	Retired	6	5.8	0	0.0
	Unemployed	18	17.5	23	23.2
Social Insurance Status	Greek National Health Service (EOPYY)	97	93.4	75	75.7	<0.001
	Other	3	2.9	0	0.0
	No insurance	3	2.9	24	24.2

⁺ Student’s t-test. * Fisher’s exact test.

, for Greeks and Albanians (N = 204).

No statistically significant difference was found between Albanians and Greeks, concerning psychiatric diagnosis. Almost 1/3 of Greek beneficiaries and 45.5% of Albanians has taken a diagnosis belonging in Affective Disorders. 17.1% of Greeks and 14.1% of Albanians had been diagnosed with Neurotic, stress-related and somatoform disorders. Finally, 46% of Greeks and 35.4% of Albanians were assigned with a diagnosis in the category Z00–Z99 of ICD-10.

Percentages of Greeks and Albanians belonging in the three diagnostic categories are presented in

Table 2. Psychiatric Diagnosis.

		Nationality				P Fisher’s Exact Test
		Greeks		Albanians
		N	%	N	%
Psychiatric Diagnosis (ICD-10)	[F30–F39] Affective Disorders	35	33.3	45	45.5	0.288
	[F40–F48] Neurotic, stress-related and somatoform disorders	18	17.1	14	14.1
	[Z03–Z99] Factors influencing health status and contact with health services	52	49.5	40	40.5

.

The majority of the participants contacted the Unit without referral from another service or professional, with a higher percentage of Greeks (61.9%), compared to Albanians (45.5%). The percentage of Albanians who were referred by Primary Health Care was higher, compared to Greeks. However, the differences between Greeks and Albanians were not statistically significant in terms of source of referral and information regarding the Unit (

Table 3. Source of initial referral and information about the Unit.

		Nationality				P Fisher’s Exact Test
		Greeks		Albanians
		N	%	N	%
Referral	Without a referral	65	61.9	45	45.5	0.057
	Primary Health Care	15	14.3	29	29.3
	Private Doctor	6	5.7	9	9.1
	Social Services	8	7.6	7	7.1
	Public Authorities	1	1.0	0	0.0
	Church	1	1,0	0	0.0
	Other	6	5.7	3	3.0
Source of Information	Beneficiary of the Mobile Unit	49	46.7	34	34.3	0.063
	Primary Health Care	16	15.2	26	26.3
	Social Service	7	6.7	7	7.1
	Media	1	1.0	0	0.0
	Private Doctor	6	5.7	12	12.1
	School	9	8.6	13	13.1
	Public Authorities	1	1.0	2	2.0
	Family	11	10.5	4	4.0
	Public Event	1	1.0	0	0.0
	Other	4	3.8	1	1.0

).

Table 4. Previous contact with mental health services and treatment plan.

		Nationality				P Pearson’s χ2 Test
		Greeks		Albanians
		N	%	N	%
Previous Contact with a Mental Health Professional	Yes	42	40.0	21	21.2	0.004
	No	63	60.0	78	78.8
Age of First Contact with a Mental Health Professional Mean		36.3 (10.1)		36.8 (10.2)		0.724 +
Psychiatric Hospitalizations	Yes	3	2.9	1	1.1	0.622 *
	No	100	97.1	94	98.9
Kind of First Therapist	Psychologist	96	91.4	59	59.5	<0.001 *
	Psychiatrist	9	8.6	40	40.8
Treatment Plan	Individual Counselling/psychotherapy	57	54.8	56	56.6	<0.001*
	Medication	5	4.8	27	27.3
	Psychotherapy/counselling and medication	17	16.2	13	13.1
	Family therapy	25	24.0	1	1.0
	Referral to another service	0	0.0	1	1.0
	Other	0	0.0	1	1.0
Medication Before Contacting the Mobile Unit	Yes	15	14.2	15	15.2	0.579
Medication as Part of the Treatment Plan in the Unit	Yes	22	20.9	40	40.4	0.008

⁺ Student’s t-test. * Fisher’s exact test.

provides information on previous contact with a mental health professional, psychiatric hospitalizations, kind of therapist at the Mobile Mental Health Unit, medication, and treatment plan for each ethnic group.

The percentage of Greeks who had contact with a mental health professional in the past was of higher statistical significance than the percentage of Albanians (40.0% versus 21.2% respectively, p < 0.05). The average age of Greeks who had first contact with a mental health professional was 36.3 years (SD = 10.1 years) and that of Albanians was 36.8 years (SD = 10.2 years). The percentage of participants who had psychiatric hospitalizations in the past was at similarly low levels in both nationalities (3 Greeks and 1 Albanian had a history of psychiatric hospitalizations). There were statistically significant differences between the two nationalities, in terms of the therapist to whom they were initially addressed within the Mobile Unit. 91.4% of Greeks had a first appointment with a psychologist in the Unit, compared to 59.5% of Albanians (p < 0.001). The percentage of Albanians who first saw a psychiatrist was significantly higher, 40.8%, compared to 8.6% among Greeks (p < 0.001).

Regarding the treatment plan, it seems that there were statistically significant differences between Greeks and Albanians (p < 0.001). Thus, while the greatest percentage of Greeks and Albanians followed individual counselling/psychotherapy—54.8% versus 56.6% respectively—taking medication (without psychotherapy/with parallel counselling) was the second choice in the case of Albanians (27.3%) and with a statistically significant difference in comparison to Greeks (4.8%, p < 0.001).

Although there was no statistically significant difference between Greeks and Albanians concerning receiving medication before seeking help from the Unit, it was found that the percentage of Albanians who ultimately received medication based on the treatment plan in the Unit was significantly higher than that of Greeks (40.4% versus 20.9%). Furthermore, the percentage of Greeks who had family therapy was significantly higher than that of Albanians (24% versus 1%). Family therapy was the third choice of treatment for Greeks.

Table 5. Treatment status and total number of sessions in the Mobile Unit.

		Nationality				P Pearson’s χ2 Test
		Greeks		Albanians
		N	%	N	%
Treatment Status	Ongoing therapy	33	31.4	19	19.2	0.004
	Drop-out	17	16.2	35	35.4
	Completion of therapy	55	52.4	45	45.5
Total number of sessions, mean, median		20.6 (15.2)	17 (9–32)	6.8 (5.4)	6 (3–8)	<0.001 ++

⁺⁺ Mann-Whitney test.

provide information on the treatment status and total number of sessions in the Mobile Unit for Greeks and Albanians.

Albanians interrupted treatment (dropout) at a significantly higher rate compared to Greeks, 35.4% versus 16.2% (p < 0.05). The average total number of sessions within the Mobile Mental Health Unit was significantly higher among Greeks, 21 versus 7 among Albanians (p < 0.001).

Gender correlations with variables related to service contact, treatment plan and course of treatment are presented in

Table 6. Gender correlations with variables related to service contact, treatment plan and course of treatment.

		Greek			Albanians
		Gender		P Pearson’s χ2 Test	Gender		P Pearson’s χ2 Test
		Male	Female		Male	Female
		N (%)	N (%)		N (%)	N (%)
Previous Contact with a Mental Health Professional	Yes	9 (25.7)	33 (47.1)	0.035	8 (24.2)	13 (19.7)	0.602
	No	26 (74.3)	37 (52.9)		25 (75.8)	53 (80.3)
Treatment plan	Individual Counselling/Psychotherapy	15 (42.9)	42 (60.9)	0.141 *	14 (42.4)	42 (63.6)	<0.001 *
	Medication	3 (8.6)	2 (2.9)		17 (51.5)	10 (15.2)
	Psychotherapy/counselling and medication	5 (14.3)	12 (17.4)		1 (3)	12 (18.2)
	Referral to another service	0 (0)	0 (0)		0 (0)	1 (1.5)
	No therapy needed	0 (0)	0 (0)		1 (3)	0 (0)
	Family therapy	12 (34.3)	13 (18.8)		0 (0)	1 (1.5)
	Other	3 (8.6)	1 (1.4)		5 (15.2)	0 (0)
Therapy status	Ongoing therapy	7 (20)	26 (37.1)	0.191	7 (21.2)	12 (18.2)	0.755
	Drop out	6 (17.1)	11 (15.7)		10 (30.3)	25 (37.9)
	Completion of treatment	22 (62.9)	33 (47.1)		16 (48.5)	29 (43.9)
Total number of sessions, mean (SD)		20.3 (16.2)	20.8 (14.7)	0.678 **	5.1 (3.5)	7.6 (6)	0.050 **

* Fisher’s exact test. ** Mann-Whitney test.

.

In the group of Greek participants, women had visited a mental health professional in the past at a significantly higher rate compared to men (47.1% versus 25.7% respectively, p < 0.05). There were no statistically significant differences in other variables.

Albanian men were found to take medication in higher rates than women (51.5% versus 15.2% respectively, p < 0.001). Furthermore, they had counselling/psychotherapy at a lower rate than women (42.4% versus 63.6% respectively, p < 0.001). Overall, women had a significantly higher number of sessions within the Mobile Unit compared to men (7.6 versus 5.1, p ≤ 0.05).

Correlations between treatment status and medication, GAF score, diagnosis, and treatment plan were also studied. Greeks were found to have higher rates of ongoing treatment (p < 0.05) in patients with mood disorders (51.5%), compared with patients with neurotic, stress-related and somatoform disorders (12.1%) and those who belonged to the category Z00–Z99 (Factors affecting the state of health and contact with services, 33.3%). GAF score in the 2nd measurement was significantly higher (p < 0.001) in those who completed therapy (82.8) compared to those still in therapy (78) and those who stopped therapy (75.3). The change in GAF score was significantly lower (p < 0.001) for those who eventually discontinued treatment (1.4), compared to those who completed treatment (7.9) or were still in treatment (6).

Albanians were found to have a significantly higher rate of dropouts (p < 0.05) in patients suffering from mood disorders (68.6%) than patients suffering from neurotic, stress-related and somatoform disorders (11.4%) and those who belonged to the category Z00-Z99 (14.3%). The percentage of patients who took medication as part of the treatment plan was significantly lower (p < 0.05) in patients who completed treatment (24.4%) compared to those who stopped treatment (57.1%) and those still in treatment (63.2%). The GAF score in both the 1st and 2nd measurements was significantly higher (p < 0.001) in those who completed treatment (74.3 and 81.2 respectively), compared to the relevant score in those who stopped treatment (67.3 and 69.6 respectively), as well as those still in treatment (65 and 72.1 respectively). The change in GAF score was significantly lower (p ≤ 0.001) for those who stopped treatment.

Table 7. Correlation of treatment status with diagnosis, medication, treatment plan and GAF score.

		Greek				Albanians
		Treatment Status			P Fisher’s Exact Test	Treatment Status			P Fisher’s Exact Test
		Ongoing Therapy	Drop-Out	Completion of Therapy		Ongoing Therapy	Drop-Out	Completion of Therapy
		N (%)	N (%)	N (%)		N (%)	N (%)	N (%)
Psychiatric Diagnosis(ICD-10)	[F30–F39] Mood disorders	17 (51.5)	3 (17.6)	15 (27.3)	0.046	6 (31.6)	24 (68.6)	15 (33.3)	<0.001
	[F40–F48] Neurotic, stress related and somatoform disorders	4 (12.1)	5 (29.4)	9 (16.4)		7 (36.8)	4 (11.4)	3 (6.7)
	[Z00–Z99] Factors related to health status and contact with services	11 (33.3)	7 (41.2)	30 (54.5)		4 (21.1)	5 (14.3)	26 (57.8)
	Other	1 (3)	2 (11.8)	1 (1.8)		2 (10.5)	2 (5.7)	1 (2.2)
Medication	No	21 (63.6)	14 (82.4)	43 (78.2)	0.226 *	7 (36.8)	15 (42.9)	34 (75.6)	0.002 *
	Yes	12 (36.4)	3 (17.6)	12 (21.8)		12 (63.2)	20 (57.1)	11 (24.4)
Treatment Plan	Individual counselling/Psychotherapy	16 (48.5)	11 (64.7)	30 (55.6)	0.070	7 (36.8)	16 (45.7)	33 (73.3)	0.007
	Medication	4 (12.1)	1 (5.9)	0 (0)		9 (47.4)	13 (37.1)	5 (11.1)
	Psychotherapy/counselling, medication	8 (24.2)	1 (5.9)	8 (14.8)		2 (10.5)	6 (17.1)	5 (11.1)
	Referral to another service	0 (0)	0 (0)	0 (0)		0 (0)	0 (0)	1 (2.2)
	Family therapy	5 (15.2)	4 (23.5)	16 (29.6)		1 (5.3)	0 (0)	0 (0)
GAF Score (1st Measurement), Mean (SD)		72.1 (6.7)	73.9 (6.3)	74.9 (6.8)	0.072 **	65 (8.5)	67.3 (8)	74.3 (7.7)	<0.001 **
GAF Score (2nd Measurement, Mean (SD)		78 (4.7)	75.3 (6.2)	82.8 (3.4)	<0.001 **	72.1 (5.4)	69.6 (7.7)	81.2 (3.6)	<0.001 **
Change of GAF Score, Mean (SD)		6 (4.9)	1.4 (2.8)	7.9 (5.6)	<0.001 **	7.1 (6.1)	2.3 (3.3)	7 (6.3)	0.001 **

* Pearson’s χ² test. ** Kruskal-Wallis test.

provides information on correlation of treatment status with diagnosis, medication, treatment plan and GAF score.

4. Discussion

The present study aimed to examine the parameters related to the contact of Albanian migrants with mental health services provided by the Mobile Unit, as well as possible differences or similarities in comparison to Greek residents in the same area. The research findings demonstrated that Albanians were more often referred to the Mobile Mental Health Unit by Primary Health Care, however, this difference did not seem to be statistically significant.

High percentages of Greeks and Albanians contacted the Mobile Unit without referral from another service or professional, a fact that probably highlights the ways in which Primary Health Care services operate in the islands, as they do not seem to act as a gatekeeping for psychiatric cases. In addition, the aforementioned finding shows that people of both nationalities probably have a good level of knowledge of the existence of the Mobile Unit, as well as the fact that they can seek help directly.

In a relevant study on referrals to 4 community mental health centers in Italy [34], it was found that, from a total of 511 immigrants, 61% had been referred by the PHC (mainly younger and married people) and 39% contacted the service on their own or were referred by social or volunteer services. In the present research, the rate of referral from the PHC for Albanian immigrants was significantly lower (30%), possibly due to differences regarding the healthcare system, as well as to particular variables related to the island complex (i.e., limited services).

In addition, according to the psychiatric history of the participants, it appears that a significantly lower percentage of Albanians (21%) were referred to a mental health professional in the past, compared to the corresponding percentage in Greeks, which was twice as high (40%). However, the frequency of diagnoses (mood disorders, neurotic disorders, psychosocial issues) in the two categories of nationalities was the same and the percentages of those who were already receiving medication for psychiatric symptoms when they contacted the Mobile Unit were at similar levels (14% in Greeks, 15% in Albanians).

The lower percentage of Albanians who were previously referred to a mental health professional, compared to Greeks, is in the same direction with the findings of another study in Greece, conducted by a Community Mental Health Centre in Athens, showing a significant delay in seeking help by immigrants, compared to natives [24].

The smaller percentage of Albanians who had sought help from a mental health professional in the past indicates partly the difficulties that immigrants face in using mental health services [8,12,35]. In a relevant study migrants, especially those suffering from affective symptoms were significantly underrepresented in mental health services in Finland [36]. Possible factors related to the difficulties of access to services are difficulty in communicating, lack of information about health services, the charge to be paid by the immigrant if he/she is not insured and/or does not have the necessary legal documents, lack of trust in mental health professionals, and the different ways of understanding the disease and the treatment [11,21,22,37,38].

Stigma is also associated with the exclusion of immigrants from mental health services [17]. Prejudice towards mental health issues and difficulty in seeking help in Albanians may be related to the lack of familiarity with these issues, given the country’s communist regime’s restrictions regarding the exchange of scientific developments in psychiatry with other European countries [39]. It may also be associated with a lack of mental health structures and anti-stigma actions in Albania [40]. The notion that one can seek help from professionals only when they suffer from severe mental illness could serve as an additional obstacle, since in Albania it is considered as a taboo to see a psychologist and talk about one’s problems, perhaps due to prejudice and shame threatening the perceived honor and pride of the family (“you have to look good”) [39].

In addition, it should be taken into account that in the present study the place of residence is an island area, with limitations in terms of available services. Meanwhile, financial requirements regarding travelling to Athens (tickets, accommodation, absence from work) make it quite difficult to access similar mental health services there, especially for immigrants facing greater financial hardship [4,5].

Another important finding of the present study is that Albanians more often asked to see a psychiatrist in the Mobile Unit, at a rate almost five times higher than in the case of Greeks—Greeks initially asked to see a psychologist. A potential hypothesis is that Albanians themselves were more likely to seek help by a doctor, due to the idea that a psychiatrist is better suited to treat mental health problems, rather than a psychologist-psychotherapist. The aforementioned observation may also be related to a tendency to “medicalize” symptoms or the difficulties encountered by immigrants, which may be a determining factor in avoiding consulting a psychologist first [41]. Factors such as the lack of mental health services in the country of origin, unfamiliarity with issues of multifaceted treatment of mental illness and adherence to the biomedical model in psychiatry in Albania, seem to be related to the possible expectation of migrants who participated in the present study to receive medical treatment for the difficulties they encounter [40].

In the same direction, studies have shown that people from ethnic minorities in Britain and the US less often seek help from psychotherapists and more often discontinue treatment in comparison to natives [42].

Furthermore, regarding Greeks, while the percentage of participants consulting initially a psychiatrist was small (9%), about 21% received medication as part of the treatment plan. As a result, it seems that Greeks initially seek help from a psychiatrist less often compared to Albanians, which could possibly be related to perceptual components, i.e., the idea that psychiatrists mainly treat severe psychiatric disorders.

In terms of treatment plan, Albanians, especially men, more often received a prescription for medication compared to Greeks (40% versus 21%). This finding is consistent with a study showing that male immigrants were more likely to receive medication rather than psychotherapy in comparison to women, as part of a treatment plan [43]. Also, 13% of Albanians received medication in combination with psychotherapy (16% in Greeks). Thus, in the end, 40% of Albanians received medication, which is in line with the percentage of those who made their first appointment with a psychiatrist.

Furthermore, attitudes towards psychotherapy or medication may vary depending on the cultural context. In many societies that do not have a tradition of speech-based therapies, psychotherapeutic techniques are received with suspicion [44], while taking medication may be more acceptable. In addition, it is important to consider therapists’ prejudice or insufficient knowledge and understanding of cultural issues, which may also influence the treatment plan for immigrants [43].

Furthermore, in cultures where the individual is an integral part of society and family (as in countries with communist regimes), individual psychotherapy is not easily accepted [45]. The symbolic meaning given to medicine has an impact on how the migrant feels and behaves around receiving medication, especially in countries where the biomedical model in psychiatry is predominant, compared to other psychosocial interventions [40].

On the other hand, the observation that Greeks and Albanians had individual counselling/psychotherapy as part of the treatment plan in similar rates is of significant importance. The main difference arose only in regard to family therapy as a psychotherapeutic method, which was applied at a rate of 1% for Albanians, compared to 25% for Greeks. The finding in question may be related to the difficulty of committing other members of the Albanian family to the treatment plan.

In addition, another important finding is that a much higher percentage of Albanian immigrants stopped treatment (dropouts) prematurely compared to Greeks (35% versus 16%). Higher rates of dropouts were observed in Albanians with mood disorders, as well as those with poor general functioning at the beginning of treatment (first GAF-rated assessment). Moreover, psychotherapy rather than medication as part of the treatment plan for Albanians was associated with a greater degree of adherence to treatment, as it was found that the rate of medication as part of the initial treatment plan was lower in Albanian patients who completed their treatment. In the same context, the total number of sessions was generally lower in immigrants than in Greeks, mainly for men. The lower number of sessions, especially for men, may also be associated with a lower degree of engagement to treatment (i.e., due to the perception that for men it is less socially acceptable than for women to seek help from mental health services).

The study’s findings concerning higher rates of dropouts in Albanians may be related not only to the difficulties of immigrants to complete treatment, but also to the possible ineffective response of the service itself towards their needs. Difficulties in the use of language have been associated with reduced engagement to treatment, as the communication between professionals and immigrants is directly affected [46].

The rates of dropouts for the given period of time and the sample of Greeks and Albanians in the present study, seem to be significantly lower than those found in another study conducted in Greece [47], although there are several differences between the samples (children were included, larger sample and time period, etc.).

Another study in Italy on the reasons for dropouts at an Intercultural Psychiatry Unit in Bologna showed that, out of 162 migrants, a percentage of 18% had stopped treatment 6 months later [34]. The percentage is much lower than the corresponding one in the present study, however the time period of 6 months is much shorter. Statistically stronger prognostic factors for early discontinuation of treatment were non-Asian nationality, recent immigration history, and lack of social interventions developed by the mental health service.

5. Limitations

A basic limitation of the present research project is generalisation of the findings, due to the fact that it is not population based, as the sample consisted of beneficiaries of a particular mental health service living in an area with specific characteristics. Another limitation is bias regarding subjects who do not accept to take part in the study [48]. Furthermore, there was an imbalance concerning gender, as significantly more women were represented in the sample. At last, the size of the sample is an extra limitation.

6. Implications for Practice

Taking into account the basic principles of a ‘culturally sensitive’ service, as well as the findings of the present study, suggestions could be made concerning the implementation of the following strategies within the Mobile Unit, with an aim to improve the quality of services provided to immigrants.

In order to reduce the rates of early termination of treatment and develop more effective therapeutic interventions for migrants, it would be important to implement systematic training programs for the professionals working in the service. Furthermore, it might be useful to appoint a therapist with sufficient training in transcultural issues, in order to form an initial assessment. Additionally, the use of psychometric scales in Albanian could offer a better assessment of migrants’ psychosocial needs.

The work of a cultural mediator, even via phone if physical presence is not possible, is necessary in order to conduct an effective assessment of immigrants’ difficulties and develop an effective treatment plan. The work of bilingual (Albanian and Greek language) mental health professionals would also be helpful.

Interventions that include psychological therapies (individual psychotherapy, counseling, family therapy) should also be offered to immigrants, as it was found that, with regards to the Albanian population, greater engagement in treatment was achieved with the implementation of such interventions as part of the treatment plan.

In addition, an innovative approach to the matter could pertain to the involvement of migrants themselves in the process of developing more sensitive approaches and treatments for people of different nationalities.

Furthermore, it would also be important to organize focused awareness and training activities for fellow professionals who work in the community and get in contact with immigrants (staff of social services, Health Centers, teachers, etc.) with an aim to increase cultural sensitivity.

The use of leaflets providing information on various mental disorders and treatments, translated to the native language of immigrants, could also increase awareness on these issues. Furthermore, it would be useful to have brochures with a description of the services provided by health and social organizations locally and nationwide.

In conclusion, further research on the factors associated with early dropouts and difficulties in the use and access of mental health services regarding migrant populations in rural areas could be useful, in order to develop more focused and effective interventions and improve the quality of care provided. In this direction, qualitative studies on migrants views and expectations regarding psychiatric disorders, seeking help by mental health professionals and available therapies could also shed more light on the issue.