Hormone replacement therapy and mammographic screening outcomes in Western Australia.

OBJECTIVES

The study objectives were three-fold: to estimate the effect of hormone replacement therapy (HRT) use on the sensitivity of mammographic screening; to examine the odds of having an interval cancer as a function of duration of HRT use and to compare the size, grade, lymph node and hormone receptor status of tumours for HRT users and non-users.

SETTING

Perth, Western Australia, where a free, population-based mammographic screening service targets women aged 50-69 years.

METHODS

The cohort consisted of 113,310 women who had 119,296 screening episodes between January 1998 and December 1999. Unconditional logistic regression was used to model the odds of having an interval cancer as a function of the effect of the duration of HRT use (measured in years). Using subsequent screening rounds, chi2-tests were used to examine whether HRT users differed from HRT non-users in terms of tumour characteristics.

RESULTS

Ninety-seven screen-detected and 31 interval cancers were diagnosed among women who were on initial screening rounds, and 393 screen-detected and 153 interval cancers were diagnosed among women on subsequent screens, two years post screening. Two-year sensitivity outcomes were significantly lower for HRT users on initial screening rounds. For women on subsequent screening rounds, the odds of having an interval cancer increased with the duration of HRT use (odds ratio 1.09, 95% confidence interval 1.04-1.14, P<0.04) after controlling for age, family history, tumour grade, size and pathology type. The histopathological characteristics of cancers did not differ for HRT users compared with HRT non-users.

CONCLUSION

The sensitivity of mammographic screening was reduced and the odds of having an interval cancer were increased for HRT users compared with HRT non-users. For women on subsequent screening episodes, the odds of having an interval cancer increased with duration of HRT use.