Other Immigrant Studies: Cancer and Cancer Registration

Transcription

1 Other Immigrant Studies: Cancer and Cancer Registration Prof. Oliver Razum Melina Arnold EUNAM Meeting Ferrara, 9 Sep

2 Outline Examples of typical projects MigHealthNet MEHO MEHO: WP Cancer in Migrants Migrant health from a lifecourse perspective Migration-sensitive Cancer Registration Survey Building migration-sensitive health indicators in cancer research Country Reports 2

3 MigHealthNet (1) Supported by the EC's Directorate-General Health and Consumer Protection (DG SANCO) and the Stavros Niarchos Foundation (completed in April 2009) Objectives to stimulate the exchange of knowledge and the formation of networks on migrant and minority health through the development of interactive databases in each of the participating countries to give professionals, policy makers including health authorities, researchers, educators and representatives of migrant and minority groups easy access to a virtual network of expertise See also: 3

4 MigHealthNet (2) Implementation of 20 country-specific wikis and one general wiki Participating countries: Austria Belgium Bulgaria Czech Republic Denmark Finland France Germany Greece Hungary Lithuania Netherlands Norway Poland Portugal Romania Sweden Switzerland Turkey United Kingdom See also: 4

5 MEHO (1) Migrant and Ethnic Health Observatory Funded under the framework of the Public Health Programme of the European Commission Objectives to develop indicators to monitor the health status of immigrants and ethnic minorities to approach the conceptual, methodological, ethical and practical issue of identifying immigrants and ethnic minorities in health databases to conduct valid comparisons between these groups within and between European countries and migrant groups 5 See also:

6 MEHO (2) Five critical health areas/ workpackages Mortality Cardiovascular Disease and Diabetes Cancer Infectious Diseases Self-perceived health and health care use Partners Erasmus MC Rotterdam, The Netherlands Amsterdam Medical Center/ University of Amsterdam, The Netherlands University of Edinburgh, Scotland Bielefeld University, Germany University of Copenhagen, Denmark University of P.J. Safarik Kosice, Slovakia See also: 6

7 Cancer in Migrants (I) Cancer risks vary geographically within and between populations Assessing the relevance of environmental ( nurture ) components genetic ( nature ) components Lifecourse perspective Allowing progress in etiology research Allowing migrant-specific prevention means 7

8 Cancer in Migrants (II) Greater likelihood to develop cancers with a relation to infectious diseases Oral cavity, nasopharynx, stomach, liver, gallbladder, cervix uteri, lymphoma Decreased risk for cancers associated with a Western lifestyle Colorectum, pancreas, lung, breast, ovary, kidney, bladder Arnold, M., O. Razum, and J.W. Coebergh, Cancer risk diversity in non-western migrants to Europe: An overview of the literature. European journal of cancer (Oxford, England: 1990), (14): p

9 Cancer in Migrants (III) 1. Lower all-cancer risk coincide in most studies irrespective of exact migrant origin health/epidemiologic transition 2. Susceptibility to certain cancers life-course perspective can help to identify important exposures during life 3. Cancer risks in migrants converge over time to be investigated Arnold, M., O. Razum, and J.W. Coebergh, Cancer risk diversity in non-western migrants to Europe: An overview of the literature. European journal of cancer (Oxford, England: 1990), (14): p

10 Migrant health from a lifecourse perspective Spallek J., Zeeb H., Razum O. What do we have to know from migrants past exposures to understand their health status? A lifecourse approach. Emerg. Themes Epidemiol Aug 15;8(1):6. 10

11 Migration-sensitive Cancer Registration 11

12 Migration-sensitive Cancer Registration Questionnaire-based survey among all European Cancer Registries (n=191) Categorization into exemplary and less than exemplary registries Exemplary = Used indicator of ethnicity: country of birth Less than exemplary = Used indicators of ethnicity: nationality or race or country of birth of father or country of birth of mother 12

13 Survey Results (I) 191 cancer registries contacted 182 registries got an online questionnaire 9 registries got a paper questionnaire 25 s returned 39 registries answered 4 registries with wrong address 21 registries got a paper version 118 registries reminded 3 answered 34 answered 2 answered 78 cancer registries answered response: 40.8 % 13 Reeske, A., J. Spallek, and O. Razum, Migrant and Ethnic Health Observatory (MEHO): Migrant-sensitive cancer registration in Europe Results of a survey conducted among European cancer registries., in Network Eurolifestyle, Neumann G and Kirch W, Editors. 2008, Thieme Verlag: Stuttgart. p

14 Tab : Main results of the survey conducted among European cancer registries in 2007/08 (n = 78) Survey Results (II) Indicator of migrant background Country of birth Nationality Race No migrantspecific data Total 35 (44.9 %) 6 (7.7 %) 3 (3.8 %) 34 (43.6 %) Possibility to conduct migrant-specific analyses Yes 9 (11.5 %) 0 (0.0 %) 1 (1.3 %) 4 (5.1 %) Migrant-specific analyses actually carried out Routine analyses 1 (1.3 %) 0 (0.0 %) 0 (0.0 %) 0 (0.0 %) Additional research 6 (7.7 %) 0 (0.0 %) 1 (1.3 %) 2 (2.6 %) Numerator and denominator population available Yes 2 (2.6 %) 12 (15.4 %) 0 (0.0 %) 14 Reeske, A., J. Spallek, and O. Razum, Migrant and Ethnic Health Observatory (MEHO): Migrant-sensitive cancer registration in Europe Results of a survey conducted among European cancer registries., in Network Eurolifestyle, Neumann G and Kirch W, Editors. 2008, Thieme Verlag: Stuttgart. p

15 Building migration-sensitive health indicators in cancer research 15

16 Building migration-sensitive health indicators in cancer research (I) For indicators we need Cases (numerator of an indicator, e.g. risk) Background population (denominator) Numerator: Cases are collected in cancer registries Survey about migrant specific data in European cancer registries Denominator: Background population Registries might have data or information Other data sources, e.g. national statistics Data linkage 16

17 Building migration-sensitive health indicators in cancer research (II) How to merge numerator and denominator: Direct methods e.g. record linkage Indirect methods data linkage (ideally using personal identifiers) name-based approaches Numerator-only approaches e.g. PCIR, relative survival, descriptive analyses 17

18 Country Reports 18

19 Country Reports: An overview Country Finland: Finnish Cancer Registry (populationbased, national level) Migrants in (1) General population (%), (2) Largest migrant groups (origin)* (1) 4.4% (foreign country of birth) (2) Russia, Estonia Indicators available in (1) Population data, (2) Cancer registry data (1) Country of birth, Citizenship, Language (2) None Routine analyses? Studies on cancer in migrants No. Using linkage procedures Main barriers Time consuming permission procedures Costs of additional data extraction Small number of cancer cases among migrants Current/Future Potentials Data linkage through personal identity code PID Inclusion of country of origin in CR data Germany: Regional registries (all 16 federal states covered in 2011, completeness fits international standards in 14 federal states) (1) ~20% (migration background) (2) Eastern Europe (re settlers), South and South Eastern Europe (Turkey) (1) None (Country of birth in Mikrocensus data) (2) None. No. Using a namebased approach to identify Turkish cancer cases Setting up a historical cohort of resettlers from the FSU Data protection not officially clarified German history (prosecution of ethnic minorities) Organisation of CRs on federal state level by federal state law lacking data comparability Lacking of migrant indicators in population data Indirect methods (e.g. name-based approaches) Numerator-only analyses Data linkage procedures 19

20 Country Reports: An overview (cont.) Country Netherlands: Netherlands Cancer Registry (populationbased, national level) Scotland: Scottish Cancer Registry (populationbased, national level) Sweden: Swedish Cancer Register (populationbased, national level) Migrants in (1) General population (%), (2) Largest migrant groups (origin)* (1) 19.5% (foreign background) (2) Turkey, Morocco, Western Europe, Suriname, Netherlands Antilles, Indonesia (1) 3.8% (foreign country of birth) (2) UK, US, Western Europe, Pakistan, India, Bangladesh, China (1) 13.8% (foreign country of birth) (2) Europe (mostly Scandinavia), Iraq, Iran, Former Yugoslavia, Poland Indicators available in (1) Population data, (2) Cancer registry data (1) Country of birth (2) Country of birth (1) Ethnic group (category classification, predefined in Scottish census) (2) Ethnic group (1) Country of birth, foreign background (2) None. Routine analyses? Studies on cancer in migrants No. Using linkage procedures No. Using name-based approaches for migrants from India, Pakistan and China First attempts using linkage with CHI Yes. Using the Health and Migration Cohort Using the Family- Cancer Database Main barriers No resources for routine analyses Lacking completeness of country of birth variable in incidence analyses Existing data often incomplete Inconsistent coding of ethnic groups between data sources Only small ethnic minority groups Lack of information on ethnicity (variation within the same country of origin) Technical solutions are still under development Current/Future Potentials Introduction of citizen service number (CSN) will facilitate linkage Linkage procedures using CHI are promising and can in future overcome most barriers PIN facilitates linkage of registers and enables all kinds of analyses (incidence, mortality, survival) and covariates (e.g. SES)

21 Conclusion Major barriers identified - Incompleteness of nationality data in the registries - Naturalized immigrants - Background population/denominator must be defined in the same way as numerator - Heterogeneity of migrant groups and origins - Heterogeneity of applied definitions and measures of association - 100% overlap of catchment area of registry and statistical region for background population is needed 21

22 Projects 22

23 Suggested readings 23