Summary Report on the MIPEX Health Strand and Country Reports

The Migrant Integration Policy Index (MIPEX) Health strand is a questionnaire designed to supplement the existing seven strands of MIPEX, which in its latest edition (2015) monitors policies affecting migrant integration in 38 different countries. The Health strand questionnaire is based on the Recommendations on Mobility, migration and access to health care adopted by the Council of Europe in 2011, which were based in turn on a consultation process that lasted two years and involved researchers, intergovernmental organizations, non-governmental organizations and a wide range of specialists in health care for migrants. The questionnaire measures the equitability of policies relating to four issues: (A) migrants’ entitlements to health services; (B) accessibility of health services for migrants; (C) responsiveness to migrants’ needs; and (D) measures to achieve change. The work described in this report formed part of the EQUI-HEALTH project carried out by the International Organization for Migration (IOM) from 2013 to 2016, in collaboration with the Migration Policy Group (MPG) and COST Action IS1103, Adapting European Health Services to Diversity (ADAPT).Part I of this report shows that many studies have already been carried out on migrant health policies, but because they tend to select different countries, concepts, categories and methods of measurement, it is difficult to integrate and synthesize all these findings. The MIPEX Health strand set out to surmount this obstacle by collecting information on carefully defined and standardized indicators in all 38 MIPEX countries, as well as Bosnia and Herzegovina and the former Yugoslav Republic of Macedonia. Indicators were scored on a three-point Likert scale and added up to form scales relating to the four issues listed above, as well as summary scales for “Access” (sections A and B), “Quality” (C and D) and the total score. Where separate policies apply to migrant workers, asylum seekers and UDMs, data are disaggregated for each group. Migration within the European Union/European Free Trade Association region is not studied because special measures exist to harmonize access to health care within this region. In keeping with the fact that policies in the health sector are influenced by multiple actors, a multilevel concept of policy is used. Development and piloting of the questionnaire were undertaken by the ADAPT network, while the data were collected by independent experts working in each country. The methodological issues involved in transforming qualitative data into qualitative scales in this way are also discussed in Part I. Computer simulations showed that varying the assumptions used to make these transformations had little effect on the results obtained. Part II describes the conceptual framework underlying the questionnaire and the way in which aspects of policy were operationalized and scored in the 38 indicators. This is followed in Part III by a detailed description of the pattern of results found in 34 European countries on each item in the questionnaire. • Section A covered legal entitlements to health-care coverage for migrants, also taking into account the administrative barriers that often make it difficult for migrants to actually obtain this coverage. A score of 100 on this section would represent complete parity with nationals. Migrant workers score 71 on this scale, asylum seekers 60, and undocumented migrants only 35. For migrant workers, requirements related to employment or length of stay often obliged them to take out private insurance or pay their own medical bills. Asylum seekers were seldom entitled to the complete basket of health-care services, while coverage for undocumented migrants ranged from practically non-existent to almost the same as for nationals (subject to a means test). Scores on section B (Accessibility) also showed that countries differed greatly in the efforts that were made to inform migrants about their rights to health care and how to exercise them, as well as other measures to help them find their way into care. Often, health workers appeared to be as badly informed about entitlements as migrants themselves. For undocumented migrants, the threat – real or perceived – of being reported to the authorities was a significant barrier to access in a number of countries. • Section C (Responsiveness) shows the widest variations between countries: eight countries take no measures whatsoever to meet the special needs of migrants, while six have scores above 70. • Section D showed that the data collection, research, planning, consultation and coordination that are needed to develop good policies existed in few countries. Part IV reports the results of statistical analyses of these data. First, the reliability, validity and structure of the scales are examined. Scores on all the individual questions are quite highly intercorrelated (Cronbach’s alpha = .86), but factor analysis shows that the four sections to some extent measure different dimensions of policy – as indeed they are supposed to. Overall scores on sections C and D (together measuring “Quality”) are strongly correlated (r = .67, p < .01), but sections A and B (“Access”) are only weakly related to them. In keeping with this, countries such as France and Iceland give very good access to migrants but make almost no adaptations, while the United Kingdom appears to have the opposite priorities. Despite these inconsistencies, the average score on all sections of the Health strand gives a reasonable indication of the overall “migrant-friendliness” of a country’s health system. Remarkably, section A on “Entitlements” shows no correlation with section D on “Achieving change”. Clearly, the “change” that the latter relates to has much more to do with what goes on inside health services than with migrants’ ability to access them. Second, the relations between Health strand scores and background variables are examined. Here, it is often difficult to disentangle the effects of different variables, because the latter tend to be strongly intercorrelated. Health strand scores are related to GDP, health expenditure, the percentage of migrants in a country, scores on the other strands of MIPEX, and the date of accession of countries to the European Union (EU) – that is, before or after 2000. Strikingly, the strongest predictor among these highly intercorrelated variables turns out to be the last one: the policy environment for migrant integration, especially regarding health, is much more negative in the 13 countries that joined the EU after 2000 than in the EU15. Further research is needed to shed light on this difference, which can be seen clearly in the scores on each section. Two other interesting findings are that the type of health financing used in each country (tax-based or insurance-based) makes a difference to quality but not to access. It is widely assumed that tax-based systems are more inclusive, but this does not appear to be the case for migrants. On the other hand, such systems do seem better at introducing measures to adapt services to the needs of migrants. Again, further research is called for. Finally, the results show that the “traditional countries of immigration” (Australia, Canada, New Zealand and the United States), which are often assumed to have better developed policies on migrant health than European countries, tend to have higher scores but also show the effects of recent political shifts. While the Affordable Care Act in the United States has improved access for migrants, governments in Canada and Australia have – as in some parts of Europe – rolled back earlier measures to make their health systems “migrant-friendly”. These results are only the first of many that are expected to result from the availability of the comprehensive, standardized data in the MIPEX Health strand.