Postmenopausal hormone drugs and breast and colon cancer: Nordic countries 1995-2005

Objectives: The purpose of this study was to utilize the variation in the level and timing of the change in postmenopausal hormone therapy (HT) use between the Nordic countries to assess the population-level impact of decline in HT use on the breast and colon cancer incidences. Methods: Nationwide HT-sales data in defined daily doses (DDDs) per 1000 inhabitant in 1995-2005 in Finland, Iceland, Norway and Sweden were obtained from drug control authorities. Breast and colon cancer incidence data by 5-year age-groups were obtained from the National Cancer Registers. By time series analysis we estimated in each age-group and country how much a change in HT-sales changes cancer incidence. Results: The decline of HT-sales varied; the decline from the highest sales was 61% in Sweden, 51% in Norway, 43% in Iceland and 25% in Finland. With the exception of Finland, the breast cancer incidence increased from 1995 to the year following the year with maximum HT-sales and decreased after that year. In the model combining countries and years, changes in HT-sales predicted the change in breast cancer incidence, an average 7% for each 10 DDD units of HT-sales. No clear association between HT-sales and colon cancer incidence was found. Conclusions: The time and country specific data suggest, that on the population level, a notable drop from high level of HT use somewhat decreases breast cancer incidence or breaks its increasing trends. The suggested protective effect of HT for colon cancer was not seen. (C) 2008 Elsevier Ireland Ltd. All rights reserved.