Background
An alarmingly high proportion of mental disorder patients have a higher risk of morbidity and mortality, particularly patients with severe mental disorders such as schizophrenia, bipolar disorder, or depression [
1‐
6]. The high burden of somatic diseases is the main contributor for excess morbidity and mortality, i.e., about 60% of excess mortality is due to somatic problems [
7,
8]. In particular, cardiovascular diseases (CVD) and infectious diseases (e.g., pneumonia, tuberculosis and chronic viral infections) are identified as major comorbid somatic diseases in patients with mental disorders [
9‐
11]. For example, a review of studies reported that people with depression have a 50% greater risk of CVD [
12]. The prevalence of hepatitis among people with severe mental illness is approximately 5 to 11 times higher than the estimated population rates of this infection [
12,
13]. Studies have suggested that such somatic comorbidities may even be greater among immigrant psychiatric patients, particularly among refugees with post-traumatic stress syndrome and depression [
14,
15].
Despite such a high burden of somatic comorbidities, rates of undiagnosed and untreated somatic illness are greater in mental disorder patients than in the general population [
16]. Mental disorder patients are also subject to unacceptably high levels of disparity in healthcare access and utilization, which contribute to poor somatic health outcomes [
16]. In particular, such disparities could be much worse among immigrant patients where underutilization of healthcare services is highly prevalent compared with non-immigrant patients [
17,
18]. For instance, our recent research work found that the majority of immigrant groups in Norway have generally lower utilization rates of specialist mental healthcare compared to ethnic Norwegians [
19]. However, there is little research investigating utilization of somatic medical services among psychiatric immigrant patients. This study therefore aims to examine the use of specialist somatic healthcare services among psychiatric patients with and without immigrant backgrounds.
Mental ill individuals in general may face a number of barriers to use and access to healthcare services, which are associated with patient- and illness-related factors as well as factors related to healthcare providers and the healthcare system. Such factors may influence the recognition and management of somatic illness. Particularly, patients with immigrant backgrounds may experience a number of such barriers. For instance, they might lack information regarding the healthcare system, little knowledge about health behaviours, or inconsistencies between their expectations and what healthcare providers are able to offer [
20]. Moreover, access to care might not be straightforward, as a result of waiting lists or a tendency for lower medical referral rates amongst immigrants. Such factors can impact upon an already under-utilised system leading to significant unmet needs amongst psychiatric immigrant patients.
In this study, we therefore aimed to examine differences in the utilization of specialist somatic healthcare between immigrant and ethnic Norwegian psychiatric patients. Specifically, the study describes differences in rates of outpatient visits and hospital admission for circulatory system diseases and infectious diseases since psychiatric patients seem to be particularly affected by these disease categories. The rationale of the study is to identify utilization patterns of somatic healthcare which could suggest specific needs among immigrant psychiatric patients compared to ethnic Norwegian psychiatric patients. The empirical knowledge provided by this study will ultimately inform about inequalities in use and access to somatic healthcare, which will be vital in efforts to improve the coordination of care across the somatic and mental healthcare delivery system for vulnerable groups.
Methods
Study design and population
The base data file, comprising all individuals listed in the Norwegian population register per 1 January 2008 (approximately 4.7 millions), was constructed by linking socio-demographic information from Statistics Norway with data from the Norwegian Patient Registry. The analysed sample was restricted to those who had at least one contact, either outpatient or hospital admission or both, with specialist mental healthcare during 2008–2011 and were living in Norway at the start of 2008 (N = 329,363).
Immigrants were defined in this study as “1st generation”, i.e., born abroad by non-Norwegian parents. Because of data protection requirements, specific country background information was only available for the larger immigrant groups, while other immigrants had been pooled into broader background categories (e.g., Latin America, West Europe) by the data provider. This background information was used to construct a variable indicating five world region origins among the immigrants, in addition to Norwegians used as the reference category. Western migrants consist of those from Nordic countries, West and Central Europe, and overseas Western countries. Those from Bosnia-Herzegovina, Poland, Serbia, Russia, and other countries in these parts of Europe were classified as East European migrants. Somalia and ‘other Africa’ were combined into one category. Turkish immigrants and all migrants with an Asian background were pooled into the Asian migrant category, while the Latin America category was kept unchanged.
Variables
The outcome variables were dichotomized and indicated whether the psychiatric patients in the study sample had attended one or more outpatient appointments or had been admitted to a somatic hospital during the four years 2008–2011, either due to infectious diseases or to diseases of the circulatory system, i.e., Chapters I and IX in the International Classification of Diseases (ICD), 10th Revision [
21].
Gender was coded 0 for males and 1 for females. Information about age was only available in ten-year bands, due to data protection stipulations; the age variable (per 1 January 2008) was categorized into 0–19 years, 20–59 years, and 60 years and above.
Statistical analysis
After describing the study population, logistic regression models were used to analyse use of specialist somatic healthcare during 2008–2011 among those with a Norwegian background and the immigrants classified according to world regions of origin. Estimates were adjusted for age and gender. The results from these regression models are reported as marginal effects (predicted probabilities with robust standard errors – β (se)). Reporting marginal effects makes interpretation more easy since the marginal effects indicate average change in the probability of the outcome (P (y = 1)) when taking into account the distribution of the other independent variables across all observations. The reported results denote the predicted probability of having had at least one contact with specialist somatic healthcare (either outpatient, or hospital admission) because of conditions of the circulatory system or because of infectious diseases during the study years. Estimates were judged as statistically significant when p-values ≤0.05. The analyses were made by means of Stata SE/14.
Discussion
Our study found that the majority of non-Western immigrant patients have lower rates of utilization of somatic healthcare services for diseases of the circulatory system. In particular, the rate of hospitalization was significantly lower in most immigrant groups, except for psychiatric patients with African backgrounds. Prior research findings have revealed that immigrant psychiatric patients, particularly refuges and asylum seekers, have a higher burden of somatic comorbidities, e.g., hypertension, diabetes and cardiovascular diseases [
14,
15,
22,
23]. Based on this evidence, we suggest that underutilization is likely among non-Western immigrants, in the sense that comorbidities between mental disorders and circulatory system diseases that would normally trigger specialist healthcare for Norwegians, do not lead to specialist healthcare for many non-Western immigrants. Specifically, it seems that under-utilisation of somatic healthcare services is unlikely to be due to lack of availability or need. This may lead to lack of detection or treatment of circulatory system diseases in immigrant psychiatric patients, or late diagnosis resulting in poor prognosis, treatment responses and outcomes [
9,
24,
25].
In contrast, we found less disparities in the utilization rates of somatic specialist healthcare for infectious diseases, especially for the outpatient services. The findings rather reveal that African psychiatric patients had higher outpatient and hospital admission rates for infectious diseases. This could be due to the heightened attention given to screening on arrival and management of infectious diseases among immigrants, especially for those from high burden countries, e.g., African immigrants. Such increased attention from both health personnel and policy makers towards prevention and management of infectious diseases could contribute in reducing inequalities in access and utilization of somatic care for infectious diseases as well as improve coordination of care across the somatic and mental healthcare delivery system.
This study indicates that immigrants from non-Western countries are especially faced with a double barrier to health care. As shown in an earlier paper, immigrants have less utilization of mental health services in general [
19], and those who access mental health services have a higher barrier to get somatic services. There is an urgent need to inform policy makers and health care workers about these disparities and address them accordingly.
One strength of the present study is that it has investigated use of specialist somatic healthcare among immigrants with high-quality register data which cover the entire population, implying that selection bias is practically absent. Only a very small number of private-practising physicians, psychologists, psychiatrists and hospitals are not obliged to report on their activities to the Norwegian Patient Registry. Thus, the patients analysed here will cover practically all patients in specialist psychiatric care during the study years, and the estimated utilization rates in specialist somatic healthcare are likely to be quite precise. Another strength is that the data provided by Statistics Norway include all registered immigrants in Norway, meaning that each immigrant world origin was represented by a sample of usable size. Further, the registry offers data concerning the origin of patients making it possible to explore based on world regions of origin rather than heterogeneous groups, i.e., western vs. non-western immigrants or immigrants from high vs. low income countries. The study period 2008–2011 coincides with the ‘Great Recession’ in Europe which was followed by austerity measures in the health care of many countries. However, since the international financial crisis had few effects in Norway, the time period analysed here did hardly differ from previous or later years.
Various limitations may nevertheless have affected the reported results of this study. The outcome variables indicate whether the psychiatric patients had at least one outpatient visit or at least one hospital admission for circulatory conditions or infectious diseases during the observation period, but they do not distinguish between few and many visits and hospital stays. Only registered inhabitants are included on the register, thereby, migrants staying in Norway without being registered, and asylum seekers whose applications have not been decided on, are not part of the analyzed samples, implying that particularly vulnerable migrants are not analysed here. A weakness which affects the present study as well as most other investigations based on administrative registers, is that the information available is often very limited and possibilities for testing hypotheses and detailed explanations will be restricted. Research of the themes addressed in this study will benefit from data more suitable for exploring explanatory mechanisms.
Acknowledgements
The availability of cleaned and linked registered data was through the research project “Health care utilization among immigrants in Norway”, funded by the Research Council of Norway (project number 222100).