Abstract
Background:
Many studies indicate that migrants in western countries have limited access to and low utilization of community mental health centres (CMHCs) despite the high prevalence of mental disorders.
Aims:
We aimed to compare migrant pathways to care across four CMHCs located in different Italian provinces and to identify pathway to care predictors.
Methods:
Migrants attending the four CMHCs between 1 July 1999 and 31 December 2007 were included in the study. Data were gathered retrospectively from clinical data sets and chart review.
Results:
Five hundred and eleven (511) migrants attended the four CMHCs, 61% were referred by GPs or other health services and 39% followed non-medical pathways to care (self-referral or through social and voluntary organizations), with important site variations. Younger age and being married were predictors of medical pathways to care; lacking a residence permit and having a diagnosis of substance abuse were related to non-medical pathways.
Conclusions:
Pathways to CMHCs are complex and influenced by many factors. Non-medical pathways to care seem to be frequent among migrants in Italy. More attention should be paid to developing psychiatric consultation liaison models that also encompass the social services and voluntary organizations.
Introduction
A large body of research indicates that migrants in western countries have limited access to and low utilization of community mental health centres (CMHCs) despite the high prevalence of mental disorders (Bhui, 1997; Bhui & Bhugra, 2002; Lloyd & Moodley, 1992). As Morgan, Mallett, Hutchinson, and Leff (2004) reported, early research considering patterns of service use among ethnic minorities took place against a background of growing evidence that there might be abnormally high rates of schizophrenia and other psychotic illness among African-Caribbeans. Rwegellera (1980), for example, as part of research primarily geared towards identifying rates of mental illness among different ethnic groups, found that African-Caribbeans were less likely to be referred by a GP and more likely to brought to mental health services by the police. Later, national data in the UK showed that women born in the Indian subcontinent and East Africa have a 40% higher suicide rate than women born in England and Wales (Raleigh & Balarajan, 1992); however, they did not have proportionally higher access to mental health care. Commander, Dharan, Odell, and Surtees (1997) found that in Birmingham, South Asians had the highest community rates of mental disorder, were the most frequent consulters of primary care and were less likely than white people to be recognized and referred to specialist care. As Bhui et al. (2003) reported in their systematic review on ethnic variations in pathways to care, the large majority of studies have been conducted in London and the UK.
In Italy, the CMHCs are a primary level structure, freely accessible to citizens and not necessarily after GP referral. Although several studies have already investigated pathways to psychiatric care for the general population (Amaddeo, Zambello, Tansella, & Thornicroft, 2001; Lora, 2009), only a few have investigated sources of migrant referrals to a CMHC in our country. One study (Tarricone et al., 2009), carried out at a CMHC in Bologna, highlighted the lower proportion of migrants referred by GPs or self-referred, and the higher rate of migrant referrals from the social and voluntary services. Other studies carried out at Varese did not find any differences in the pathway to psychiatric care between migrants (the large majority from South America) and natives (Costantini, 2009; Vender, 2009).
This prompted us to set up a multicentre study aiming to:
evaluate the socio-demographic and clinical characteristics of migrants attending four different CMHCs in Italy;
compare the pathways to care across these sites;
identify the possible socio-demographic predictors of pathways to care.
Methods
Setting
This study was conducted in four CMHCs in northern Italy. CMHCs are the core of the community-based mental health system in Italy and cover all activities pertaining to adult psychiatry in the outpatient setting (Lora, 2009). Patients are generally referred to the CMHC by primary care physicians but may also, in accordance with Italian psychiatric law, arrive directly at the CHMC on a drop-in basis. The main characteristics of the four centres’ catchment area and of the CMHCs themselves are described in Table 1.
Description of four community mental health centres: Resident population and mental health services.
Bologna West CMHC is the only centre with a programme specifically dedicated to migrants: the Bologna West Transcultural Psychiatric Team (BoTPT). This team was established as a pilot project 10 years ago. The BoTPT is composed of Bologna West CMHC mental health operators who dedicate part of their time to migrant psychiatric consultation. Core BoTPT personnel includes psychiatrists as well as social workers, psychiatric nurses, residents in psychiatry and medical anthropologists. If needed, a cultural mediator joins the multi-professional team. In the most difficult cases, the BoTPT team directly delivers psychiatric and psychosocial treatment to migrants; in other cases, the BoTPT team provides a consultation designed to identify the mental and psychosocial needs of the migrants and then redirects patients within the CMHC and other services. This team’s work is different from activities routinely delivered by the whole CMHC team since on request it may involve a cultural mediator and it implies higher attention to migrants’ cultural and social mental health needs. During the last 10 years, Bologna West CMHC psychiatrists have provided training programmes and consultation liaison activities for social workers with a view to ameliorating pathways to care for migrants with mental disorders. The BoTPT activities have been described in detail in previous papers (Tarricone et al. 2009; Tarricone et al., 2010).
In the other CMHCs involved in the present study, no dedicated or specific procedures are offered to migrant patients, who have free access to the usual services. After the first contact with the CMHC, which is usually a short triage visit with a nurse collecting personal data from the patient and the reason for seeking advice, an appointment for a first psychiatric consultation is set. The treatment needs are then evaluated, subsequent care is planned and it is decided whether to involve other professional figures (psychologist, social worker, dedicated psychiatric nurse or educator). If needed, municipal social workers and cultural mediators provided by an external agency may be involved in dealing with cases.
No main differences were found among social services organization in the four areas: social services are provided by the local city authority and training initiatives; integrated projects and intervention for migrants are developed jointly between the social services and the respective CMHC. The area of Bologna West and Sassuolo CMHC run dedicated social services for foreign citizens. In all four centres there are also local partnerships between psychiatric services and local non-profit organizations, which take care of migrants, exiles and refugees.
Study design and instruments
Subjects
All adult migrants consecutively attending the Varese, Bologna and Sassuolo CMHCs between 1 July 1999 and 31 December 2007 and the Codroipo CMHC from 2003 to 31 December 2007 were included in this study. No formal exclusion criteria were applied and migrant patients from every country were eligible for the study. Socio-demographic, clinical and health service utilization data of migrants attending the four CMHCs were gathered retrospectively by researchers (IT, MB, SF, CG, EB, CC) using clinical data sets and chart review. The treating psychiatrist made the primary diagnosis according to ICD-10 criteria (World Health Organization, 1992). Patients’ areas of origin were grouped as Maghreb, Sub-Saharan Africa, Asia, Eastern Europe, and Central or South America. The length of stay in Italy was classified in three categories (less than one year, one to four years, five years or more). The pathway to care was defined as the main source of referral of migrants to the CMHC, using the definition provided by Gater et al. (1991).
Statistical analysis
Comparisons among groups were performed using the χ2 test for categorical variables and ANOVA for continuous variables. Following significant χ2 or F tests, post-hoc comparisons were carried out with a Bonferroni correction to the probability level.
To estimate the joint effect of patient variables and site on the pathway to care, we adopted a multilevel statistical approach (Rice & Jones, 1997). This approach is suitable for data that are clustered in higher level units, e.g. patients clustered into sites, and allows one to estimate the effects of the variability of each of these components on the dependent variable. Specifically, a multilevel logistic regression was fitted by regressing the pathway to care (coded as medical/non-medical) against patient’s socio-demographic and clinical variables and site. The medical pathways included primary care, hospital and psychiatric ward, while non-medical pathways to care included self or family referral, and social or voluntary service referral.
Data analysis was carried out using SPSS for Windows version 17.0 and STATA version 10.1.
Results
Socio-demographic characteristics
During the study period, 511 migrant patients attended the four CMHCs: 190 in Bologna, 32 in Codroipo, 63 in Sassuolo and 226 in Varese. Table 2 summarizes the socio-demographic characteristics of the sample. The mean age was 35 years (SD = 11) and 220 subjects (43%) were male. Age and gender distributions varied across sites, with Bologna patients being significantly younger (M = 33, SD = 10) and more frequently men (n = 108, 57%). Ethnic composition also varied in a statistically significant way according to site: in Bologna and Sassuolo most subjects were from Maghreb, in Varese and Codroipo from East Europe. Fifty per cent were gainfully employed, with the highest proportion of regular employment in Varese (n = 128, 57%). Eighty-four per cent of the study population had a regular residence permit; migrants in Bologna CMHC were more frequently undocumented (n = 78, 28%). Fifty-one per cent of migrants had been in Italy for five years or more, with the highest proportion in the Sassuolo sample.
Socio-demographic characteristics and psychiatric diagnosis in the overall sample and by site.
Missing data: a4 patients, b7 patients, c4 patients, d17 patients, e77 patients, f2 patients
BO = Bologna, CO = Codroipo, SS = Sassuolo, VA = Varese
Psychiatric diagnoses
Common mental health disorders were diagnosed in 70% of the sample (n = 354); psychotic disorders accounted for 19% of cases; substance-related disorders were found in 14 patients (3%). Twenty-four subjects suffered from other disorders, while 20 had no psychiatric diagnosis. Statistically significant differences were found in the distribution of psychiatric diagnoses across sites, with the highest prevalence of psychotic disorders in Sassuolo (n = 20, 32%).
Pathways to care
Sixty-one per cent of patients were referred to CMHCs by the health services: specifically, 39% were referred by GPs, 11% by the general hospital, and 11% by a psychiatric ward. Non-medical pathways to care accounted for the remaining 39% of cases, mostly in the form of informal (self or family) referral (22% of cases), whereas social and voluntary organizations were responsible for 7% and 5% of referrals, respectively. Table 3 shows the differences in pathways to psychiatric care across the four sites: health service referrals, particularly GP referrals, were most frequent in Varese, accounting for 77% of the total referrals, while non-medical routes of care, including self-referral, social services and voluntary organizations, were much more common among migrants in Bologna (59%).
Pathway to care.
Missing data: a3 patients
BO = Bologna, CO = Codroipo, SS = Sassuolo, VA = Varese
In the multilevel logistic regression analysis, including site and patients’ variables, the site effect on the pathway to care proved to be significant (Wald test = 6.728, p < .01).
Patients’ factors related to pathways to care are summarized in Table 4. Being married, of a younger age and having a residence permit proved to be related to an increased likelihood of medical referral, after adjustment for the site effect. On the contrary, having a diagnosis of substance abuse was associated with the non-medical pathway to care.
Correlates of pathway to care.
Note: Odds ratios (OR) exceeding unity with a confidence interval (CI) not including unity denote a positive association with the medical pathway to care. Odds ratios below unity with a confidence interval not including unity denote a negative association with the medical pathway to care.
Discussion
To our knowledge, this is the first study in Italy that has attempted to analyse the pathways to public psychiatric services taken by migrants attending CMHCs. The pathway to mental health care for migrants is complex and is influenced by many social and demographic factors, in addition to clinical factors, as has already been shown by other studies predominantly conducted in the UK. Our results indicate that a sizeable number of migrants followed a non-medical route to care. This is in line with other studies reporting a higher proportion of non-medical referrals among ethnic minorities (Borowsky et al., 2000; Commander et al., 1997; Husain, Waheed, Tomenson, & Creed, 2007). One study conducted in the Italian South Verona CMHC reported that non- medical referrals among patients listed in the Psychiatric Care Register account for about 30% of all new episodes (Amaddeo et al., 2001). Thus, there is some evidence that also in Italy the non-medical route to care is more common among migrants, as already demonstrated in the UK.
In our study, the pathway to care was associated with some socio-demographic characteristics, such as age, marital status, having a residence permit, as well as by the clinical variable of substance abuse (after controlling for the site effect). Specifically, being older, unmarried, lacking a resident permit and having a diagnosis of substance abuse predicted the non-medical pathway to care, with self-referral being the most frequent way of access to psychiatric care. The association found between non-medical referral and lack of a residence permit is not surprising: in Italy, migrants without a residence permit are not allowed to attend GP facilities and may only receive urgent care, which is usually provided by special doctors. We could hypothesize that being unmarried may entail low social support and a dearth of relevant others, as already suggested by Morgan et al. (2005). In fact, significant others within an individual’s social network have been shown to play a major role in shaping how, when and what type of help is sought (Morgan et al., 2005). Finally, as already shown by other research, older people and people with drug addiction problems might need a special therapeutic approach to facilitate access to care and use of services (Gilmer, Ojeda, Fuentes, Criado, & Garcia, 2009; Kay-Lambkin, 2008; Mowbray, Perron, Bohnert, Krentzman, & Vaughn, 2010).
The CHMCs involved in our study are based in the three northern Italian regions where immigrant population has a higher prevalence and the highest growth rate stability in the last few years as compared, say, to central and southern Italy (Caritas/Migrantes, 2008). Analysis of the socio-demographic characteristics of our sample showed that on average migrants attending the four psychiatric services are more commonly female (57%), which is the same proportion found by the PROG-CSM survey among natives attending CMHCs in Italy (Lora, 2009). Moreover, migrants in our study seem to be younger than natives attending CMHCs (PROG-CSM survey, 2008). This result is not surprising, considering the lower mean age of migrants compared with natives in the Italian population at large. The first diagnostic group among migrants attending the four CMHCs is common mental disorders (CMD). CMD were likewise found to be the most frequent diagnosis among natives by the PROG-CSM study; however, such disorders have a higher prevalence among our migrant sample (70%) than with the natives in the PROG-CSM study (47.5%). The second diagnostic group is psychosis in our migrant sample, as well as in the national native sample of the PROG-CSM study.
Our findings also show a significant amount of heterogeneity across the national territory: migrants following a social route to care amounted to 59% in Bologna vs 23% in Varese. The site variation persisted even after adjusting for some socio-demographic variables, suggesting that other factors, possibly related to the characteristics of the migrant population or to the service organization, might be involved. Thus, one possible explanation for inter-centre variations in migrant pathways to care might be that the variables used were too crude to fully capture differences among migrants (e.g. social networks, cultural attitudes toward CMHCs, etc.) in the different centres, an issue to be addressed in future research. Another issue that comes into play is the degree of awareness of ‘mental health’ services and social services on the part of immigrants in each region, an issue that needs to be explored by further qualitative research. Even though no major organizational differences among CMHCs and social services are seen in northern Italy, we should still remember that, of the four sites, Bologna is the only one with metropolitan characteristics, as well as long-established services for welcoming refugees and asylum seekers. On the other hand, the more common social pathways to care of migrants at Bologna could represent an index of higher sensitivity by the Bologna CMHC in detecting psychiatric care needs among migrants attending social services, due to the existence of a dedicated clinical programme (BoTPT), which is also deeply interconnected with social services by means of consultation liaison activities. As suggested by Consoli and Lemogne (2009), the organizational characteristics of centres with regard to available health care heavily affect the expression of psychological complaints among services users. Thus, the Bologna consultation liaison of psychiatric activities with social organizations could increase the demand for psychiatric help among social service users.
Limitations
This study has an important limitation in that it lacks any direct native control group and has a relatively small sample size, which hampers our ability to draw conclusions about site differences related to the case mix; in addition, due to the retrospective design, only some of the many factors related to pathway to care could be considered. Moreover, data on psychiatric morbidity among migrants in Italy are still missing and thus our results did not allow us to see whether the rate of psychiatric referral of migrants actually corresponds with their needs for psychiatric care. The study nonetheless provides useful hints for future research in this area.
Conclusion
This preliminary pilot study is designed to generate further multi-centre research in Italy to better understand migrants’ pathways to psychiatric care. For the time being, our results enable us to form hypotheses concerning the factors predicting migrant pathways to care and the differences that we found between centres. These issues should be addressed more deeply in further studies.
Our findings, despite requiring further confirmation and specification, have important implications for clinical practice and service organization, showing that migrant pathways to psychiatric care in Italy are not only based on primary care agencies but also on social and voluntary services. Thus, more attention should be paid to fostering the development of a psychiatric consultation liaison model that encompasses not only primary care services, but also social services and voluntary organizations. Finally, further research is needed to evaluate the cost-effectiveness of specialized teams integrated in mental health services and in facilitating pathways to psychiatric care among migrants who do not follow a medical route.