BACKGROUND: The publication in 2002 of the women's health initiative (WHI) trial revealed long-term risks and benefits of hormone replacement therapy (HRT). Increased cardiovascular disease, venous thromboembolism and breast cancer risks outweighed benefits on hip fracture, colorectal and endometrial cancer. We investigated whether the subsequent 50% fall in HRT use in England impacted on population rates of these outcomes. METHODS: Time-series analysis of hospital admissions, incidence and mortality amongst women aged 50-69, England 1997-2006. RESULTS: There was no relationship between reduced HRT prescribing after 2002 and trends in breast cancer, colorectal cancer or hip fracture. Amongst 50-59 year olds, the annual percentage change in venous thromboembolism hospitalizations fell from 0.0% [95% confidence interval (CI): -2.3 to 2.3%] between 1997 and 2000 to -5.7% (-7.7 to -3.6%) between 2000 and 2006 (P-value = 0.001); the annual change in endometrial cancer mortality increased from 0.7% (-3.2 to 4.8%) between 1997 and 2003 to 11.0% (0.2 to 22.9%) after 2003 (P-value = 0.07); and previously falling acute myocardial infarction hospitalizations (annual change: -6.8%) and stroke (-3.0%) stabilized (-0.4%) or increased (+0.8%), respectively, around 2001 (P < 0.0001). CONCLUSION: Although rates of venous thromboembolism (decline) and endometrial cancer mortality (increase) changed in line with WHI findings, the decline in venous thromboembolism may have started before 2002 and increased fatal endometrial cancers could be a chance finding.
BACKGROUND: The publication in 2002 of the women's health initiative (WHI) trial revealed long-term risks and benefits of hormone replacement therapy (HRT). Increased cardiovascular disease, venous thromboembolism and breast cancer risks outweighed benefits on hip fracture, colorectal and endometrial cancer. We investigated whether the subsequent 50% fall in HRT use in England impacted on population rates of these outcomes. METHODS: Time-series analysis of hospital admissions, incidence and mortality amongst women aged 50-69, England 1997-2006. RESULTS: There was no relationship between reduced HRT prescribing after 2002 and trends in breast cancer, colorectal cancer or hip fracture. Amongst 50-59 year olds, the annual percentage change in venous thromboembolism hospitalizations fell from 0.0% [95% confidence interval (CI): -2.3 to 2.3%] between 1997 and 2000 to -5.7% (-7.7 to -3.6%) between 2000 and 2006 (P-value = 0.001); the annual change in endometrial cancer mortality increased from 0.7% (-3.2 to 4.8%) between 1997 and 2003 to 11.0% (0.2 to 22.9%) after 2003 (P-value = 0.07); and previously falling acute myocardial infarction hospitalizations (annual change: -6.8%) and stroke (-3.0%) stabilized (-0.4%) or increased (+0.8%), respectively, around 2001 (P < 0.0001). CONCLUSION: Although rates of venous thromboembolism (decline) and endometrial cancer mortality (increase) changed in line with WHI findings, the decline in venous thromboembolism may have started before 2002 and increased fatal endometrial cancers could be a chance finding.
Authors: Richard J Santen; D Craig Allred; Stacy P Ardoin; David F Archer; Norman Boyd; Glenn D Braunstein; Henry G Burger; Graham A Colditz; Susan R Davis; Marco Gambacciani; Barbara A Gower; Victor W Henderson; Wael N Jarjour; Richard H Karas; Michael Kleerekoper; Roger A Lobo; JoAnn E Manson; Jo Marsden; Kathryn A Martin; Lisa Martin; JoAnn V Pinkerton; David R Rubinow; Helena Teede; Diane M Thiboutot; Wulf H Utian Journal: J Clin Endocrinol Metab Date: 2010-06-21 Impact factor: 5.958