Registry data for disease mapping & spatial epidemiology in Finland

Transcription

1 Registry data for disease mapping & spatial epidemiology in Finland Mika Gissler THL National Institute for Health and Welfare, Helsinki, Finland Nordic School of Public Health, Gothenburg, Sweden

2 Topics 1. Existing health information sources 2. Data on residence in registers 3. Study example on a health environment hazard in Helsinki, Finland 4. Conclusions

3 Why good possibilities to register-based studies? Traditions: population statistics have been collected more than 250 years and health statistics more than 150 years. First real health registers were started in the s, when improved computers were available: health care personnel, cancer register. Personal identification number for all in Several data quality studies have shown the high quality of routinely collected registers. Data protection allows research use of register data.

4 Finnish health registers Cancers 1953 Health care personnel 1955 Tuberculosis and STIs 1958 Congenital anomalies 1963 Occupational diseases 1964 Special medication 1964 Adverse drug reactions 1966 Hospital discharges 1967 Mass Screenings 1968 Causes-of-death 1969 Abortions and sterilisations 1977 Exposure to cancerhazardous material 1979 Endoprostheses 1980 Drugs (surveillance) 1982 Visual impairments 1983 Births 1987 Infectious diseases 1989 Dental implants 1994 Prescribed drugs 1994 Outpatient visits in public hospitals 1998 Outpatient visits in health care centres 2011 All these registers include personal identification number

5 Other important registers Social welfare registers Background data updated continuously by Statistics Finland Pensions 1962 Social Benefits (Social Insurance Institution) 1964 Social assistance 1985 Children taken into custody 1991 Institutionalised care at social institutions 1994 Education Income Socioeconomic status Country of birth / language Citizenship Marriages and divorces Emigrations and immigration Link between parents and children/siblings All these registers include personal identification number

6 Information on residence All registers include information on persons residence. Most often this is the actual residence, but not necessary (e.g. students). Foreigners have an own code: 200. Hospital Discharge Register even includes information on the country of residence.

7 The number of municipals decreases

8 Exact information on residence at CPR Basic information related to the identification of people and buildings is registered in the Population Information System. Personal data recorded in the system includes name, personal identity code, address, citizenship and native language, family relations and date of birth and death (if applicable) Building data registered includes the building code, location, owner, area, facilities and network connections, intended use and year of construction Real estate data registered includes the real estate unit identifier, owner s name and address, and buildings located on the property More information:

9 Exact information on residence at CPR Information can be used in scientific research, and it is relatively easy to get the permission. The process to get the permission and the data is usually fast. The data is free, but the researcher have to may for the extra work to form the data: / case Minimum 600 Discount with large datasets (at least cases)

10 Costs for CPR data Cases Costs Costs /case , , , , , , , ,02

11 Study example: Registers and environmental health risks - Follow-up of urban population living above a former waste dump area in Helsinki, Finland Gissler M: Journal of Official Statistics 2010, in press.

12 Gathering of unbiased information on health risks is difficult Large-scale health examination studies are seldom feasible. Health questionnaire or interview surveys problematic: A high risk for selection, recall or reporting bias for the exposed population, especially if the potential health hazards have been discussed in public. The best way to gather research data is to use existing registers. No selection bias, no loss in follow-up.

13 Myllypuro blockhouse residential area was built in the 1970s. The area had earlier been mostly forest, but 4.5 hectares had been used as a waste dump in

14 In , twelve blockhouses and a day care centre were built on the former dumpsite. All these houses were owned by the City of Helsinki, with the exception of one private house. Over inhabitants have been living in these houses.

15 City of Helsinki decided to destroy the blockhouses and to give the residents a financial compensation for their loss.

16 Methods Residence information the exact GIS-coordinates were identified by the Environment Centre of Helsinki City the Central Population Register (CPR) ID numbers, marriages, migration data, spouses, children Follow-up was based on register information: Cause-of-Death Register (Statistics Finland) Medical Birth Register and Finnish Register of Congenital Malformations (STAKES)

17 Exposure and indicators Four different exposure measures were formed. Fertility and sub-fertility indicators: Live born children per 1000 women IVF deliveries and adoptions Perinatal health indicators: Proportion of boys Low-birth weight, preterm birth and Apgar scores Infant mortality Congenital anomalies

18 The mean number of live born children, by year cohort 2,5 2 1,5 1 Waste dump Other parts 0,

19 The proportion of families with adopted child, % 1,0 % 0,8 % 0,6 % 0,4 % Waste dump Other Myllypuro 0,2 % 0,0 % Total Educationadjusted Primary infertility only

20 The proportion of IVF deliveries in ,8 % 1,6 % 1,4 % 1,2 % 1,0 % 0,8 % 0,6 % 0,4 % 0,2 % 0,0 % IVF deliveries Age-adjusted Waste dump Other Myllypuro Other Helsinki

21 Perinatal and infant health Years Variation Finland Boys, % NS Infant mortality, % * Preterm birth, % NS Low-birth weight, % NS Low Apgar score, % NS Congenital anomalies, % NS * P=0.026, if mother lived in the area during pregnancy (N=4 infant deaths).

22 Conclusions of methods The identification of exposed group and follow-up was unbiased and unselected. The completeness and reliability of the Finnish population and health registers is good. It was possible to enumerate the exposed population without selection bias, to track the exposed population without loss to follow-up, and to form different exposure measures. The data collection and analyses were more difficult than expected.

23 Conclusions of methods The type, occurrence, and severity of health outcomes have to be evaluated by using similar health information and data collection methods for cases and their controls. The small size of study data and the rarity of several outcome measures impeded the analysis and the evaluation of clinical and public health aspects of our findings.

24 Finally Register-based studies are feasible, e.g. crosssectional, longitudinal and trend studies. Combination of data from other registers and from other sources - such as medical records and biobank material - is feasible. Detailed GIS-information is available. Administrative borders causes no problems, in some cases delays. Data protection questions have not been an issue. Informed consent is not required. The data must not be used in individual decision making.

25 Future New health data sources: The collection of primary health care data in public sector starts in The nation-wide electronic patient journal is to be available in Finland in The Finnish Information society strategies 2020 & The EU INSPIRE directive All data should be available for low costs or no costs. The INSPIRE Directive (May 2007) is to establish an infrastructure for spatial information in Europe to support Community environmental policies, and policies or activities which may have an impact on the environment.

26 Acknowledgements City of Helsinki funded the Myllypuro study in Dr Annukka Ritvanen (THL) and Dr Antti Pönkä (City of Helsinki) participated in the Myllypuro study. Kimmo Nummela gave the permission to use his Myllypuro photos in this presentation.