PHEBP Benefits and risks of postmenopausal HRT (+tutorial) Flashcards
Suggest some specific elements of the review you might wish to focus on when
thinking about whether we can apply the findings to Mrs D?
https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD004143.pub5/epdf/full (REVIEW)
The table summarises findings related to healthy postmenopausal women who take Combined HT regularly
For which health event is the risk most elevated with
use of combined HT? Interpret the finding
Look at far right column- VTE has the highest number here. This means that 4.8
The confidence interval dosent contain 1
The risk in the population is as low as 2.49 or as high as 7.34 times this means the interval is not very precise, you want a tighter CI
Assumed risk- those not taking the HRT, the risk would be 2 per 1000. For those taking HRT, the risk would be 7/1000. This has a risk difference of 5- so an extra 5 VTE cases per 1000 per year
Compare the absolute risk
difference between VTE and
breast cancer
There is a protective effect
RR- the risk of OP is 22% lower in women taking HRT compared to women not taking HRT.
95% confident
What are the results for clinical fractures telling us? Hence what type of prevention could HT be used for here?
Hence could be used for primary prevention of postmenopausal osteoporosis
This table now summarises findings related to healthy postmenopausal women who take Oestrogen-only HT regularly
Comment on the main differences between the risks here and those seen with combined HT?
The RR is lower for combined HRT for Coronary events, breast cancer and clinical fractures, but they are not quite statistically significant- eg the one for breast cancer only just contains 1
Given all these health outcomes, how might you best relay this information to Mrs D? What other factors might you want to consider?
● With so many relative/absolute risks it would be impossible for a patient to digest all this info. Some kind of visual info graphic would be helpful.
● Convo w pt should inc her priorities along w medical evidence
● Family history of breast cancer means the cancer risks associated with combined HRT need discussion
● Although breast cancer risk appears lower with oestrogen-only HRT, this is only rec for women post-hysterectomy, she would need combined HRT to avoid risk of endometrial cancer from unopposed oestrogen
● The family history of osteoporosis (inc fractures) and her inactivity puts her at higher risk of osteoporosis in, so potential benefits of HRT (+ exercise) could be raised
Describe this study design and the advantage of it
Prospective studies were included using a nested case-control design,
Up to four randomly selected controls per case of invasive breast cancer matched on age, year of birth, and region.
All analyses included only postmenopausal women, defined as known age at natural menopause or unknown age at menopause but age of at least 55 years (since 90% with a natural menopause were postmenopausal by 55 years) and excluded younger women with a hysterectomy.
define who the cases and controls are here, and what the exposure is?
● Cases = those with breast cancer
● Controls = women without breast cancer
● Exposure = menopausal hormone therapy
(Note: the authors use the abbreviation MHT in the paper, but as it’s only postmenopausal women we have altered the following slides to say HT)
Prospective studies were included using a nested case-control design,
Up to four randomly selected controls per case of invasive breast cancer matched on age, year of birth, and region.
All analyses included only postmenopausal women, defined as known age at natural menopause or unknown age at menopause but age of at least 55 years (since 90% with a natural menopause were postmenopausal by 55 years) and excluded younger women with a hysterectomy.
Why were cases matched to up to 4 controls and why were they matched for age, year of birth and region?
● 4 controls per case are used to increase sample size and make estimates/ 95%CI more precise (>4:1 doesn’t impact 95%CI as much)
● Matching is generally used to take account of strong confounders
● As age increases, risk of breast cancer rises and potential exposure to HT is likely to increase too
● Year of birth accounts for cohort/calendar effects across extended period
● Region – accounts for regional differences in HT prescribing & breast cancer rates
.
The size of the box relates to the number of women included in each comparison- eg the larger the sample size, the bigger the number of cases
● The lines represent the 95% confidence intervals around the relative risks.
● They appear as white when they are completely inside the black box (and are black when they stretch out beyond the symbol).
● It is not the clearest type of visualisation! (RR’s don’t line up well either)
.
Past users have a lower risk rate for breast cancer compared to current users, however, they still have increased risk
Basically stopping HRT does not completely eradicate the risk of BC
The difference is more subtle
.
If they weren’t matched on age you would need to take age into account as a confounding factor
.
If you started young- greater increase in risk
If you started later- smaller increase in risk
Younger patients would be on HRT for longer- duration taking HRT will be relevant- this is slightly adjusted for by the 5-14 years
