SMC, Swedish Mammography Cohort

Summary

Strengths: The longitudinal design provides good opportunities to study public health, gender and life styles. The broad variety of variables contributes to analyses of diet, physical activity, medical history, age at menarche, history of oral contraceptive use, age at menopause, postmenopausal hormone use and lifestyle factors such as cigarette smoking history and use of dietary supplements. The SMC database will be extended with genetic information. This will facilitate studies of genetic susceptibility and of interplay/interactions between lifestyle factors and genes in the development of chronic diseases and aging. Weaknesses: We do not have biological samples from the whole cohort.

Type of data

Data Source
Registry + Survey

Type of Study
Survey same
Cohort study

Data gathering method
Telephone
Registries
Self administered questionnaire
Other: Cognitive tests Saliva, blood, urine, fat biopsies

Access to data

Conditions of access
On site access, as well as downloadable files are available after agreement. Participants were informed that their data can be analysed abroad.

Type of available data (e.g. anonymised microdata, aggregated tables, etc.)
Aggregated tables or anonymised microdata (depending on type of collaboration)

Formats available
SAS, STATA, Excel

Coverage

Coverage Years of collection, reference years, and sample sizes
The SMC is a population-based cohort study established in central Sweden (Västmanland and Uppsala counties) and includes about 90,000 women born 1914-1948. Baseline data in the SMC study were collected in 1987. The cohort includes questionnaire-based information about demographic factors and modifiable lifestyle factors such as diet, physical activity, smoking, weight, use of dietary supplements, some drugs “over counter”, and alcohol. There is also information on some sociologic and work-related factors. Other self-reported information regarding wellbeing, different symptoms related to aging (not available in the Swedish health registries), social aspects, housing, longevity of parents etc., is also available in the SMC. The baseline questionnaire (wave 1) was carried out between 1987 and 1990 and had a sample size of 61,433 women. A second questionnaire (wave 2) was carried out in 1997 with a sample size of 39,227 women. The third wave was carried out in 2008 with a sample of 25,332 women. The fourth wave was in 2009 with a sample of 30,621 women. Biomaterial was collected between 2003 and 2008, and included a sample size of 5,330 women.

First year of collection
1987

Stratification if applicable
Born 1914-1948, women, central Sweden (Västmanland and Uppsala counties)

Base used for sampling

Geographical coverage and breakdowns
Central Sweden (Västmanland and Uppsala counties)

Age range
Born 1914-1948, 40-74 years old at baseline

Statistical representativeness
Population representative

Coverage of main and cross-cutting topics
The data is based on self-reported answers regarding present occupational status (full-time work, part-time work, house-wife, retired, disability, pension and unemployed). Information about work status is also obtained by linkage to Statistics Sweden (LISA register). With the information from the cohort, we are able to investigate aging and specific major chronic diseases such as cancer (breast, colon, bladder, kidney, etc.), cardiovascular diseases, osteoporosis, diabetes, age-related cataracts, lower urinary tract symptoms (LUTS), obesity, etc. Follow-up of the cohort is accomplished through annual matching with national and regional registers with high completeness of diagnoses and population registers at the Statistic Sweden. In a random subsample of the SMC we have DEXA measurements of bone status from 5,022 women (worlds-unique size of such material with other prospective data) and biological samples (blood, urine, fat biopsis) from 5330 women. Saliva samples (DNA source) are available from additional 9,000 women.

Linkage

Standardisation
The dataset is linked to Statistics Sweden where work occupation classification (SSYK) is used, which is based on ISCO-88.

Possibility of linkage among databases
There is ID for these over 60,000 participating women, which gives us a possibility to link to Swedish registers; we are updating their health status annually.

Data quality

Entry errors if applicable
Among the 60,000 participants there is some missing information (partial lack of answers to some questions).

Breaks
ICD codes for specific diseases are all translated into the newest version ICD 10.

Consistency of terminology or coding used during collection
Yes

Governance

Contact information
Professor Alicja Wolk
Karolinska Institutet, Institute of Environmental Medicine, IMM
Nobels väg 13
171 77 Stockholm Sweden Phone: +46 8 524 861 70
Email: Alicja.Wolk(at)ki.se
Url: http://ki.se/ki/jsp/polopoly.jsp?d=40981

Timeliness, transparency
First publication regarding women from the Swedish Mammography Cohort, SMC was published year 1995. Time between collection of data and publication of results varies due to kind of research design, but takes approximately one year.