News and views

Arguments persist over NICE's HRT recommendations

Small clouds of dust still hover over last year's NICE guidelines on the menopause and refuse to settle. As ever, the air is stirred by the relative safety and benefits of hormone therapy (HRT), and how NICE interpreted the evidence. In May three supporters of the NICE guideline development committee, including its chairman, wrote to the British Medical Journal in defence of how evidence underpinning the guideline was presented. ¹ This defence was in response to a less than complimentary editorial in the BMJ back in January which claimed that ‘methodological deficiencies' undermine the NICE conclusions. ² Specifically, the editorial complained that ‘the guideline lacks appropriate and complete quantitative summary estimates of the risks and benefits from taking MHT (hormone therapy)'.

The argument of the editorial, one of whose authors was a Million Women Study collaborator, was that the estimates of relative risk for HRT should be calculated from all relevant randomised controlled trials for cardiovascular disease and from trial and observational data for cancer – thus, the ‘complete quantitative summary estimates of the risks'. Only these, said the BMJ editorial, would ensure that practice is guided by the quantitative sum total of the appropriate evidence, and not overly influenced by the results of individual studies or subgroup analysis.

There was also implied criticism of NICE's evident emphasis on HRT in the BMJ's conclusion that menopause is ‘a normal transition' and not ‘an oestrogen deficiency syndrome', and that ‘women and their healthcare providers should have confidence that most will manage their symptoms without pharmacotherapy'.

Not surprisingly, the ‘NICE response' was to refute the claims and defend its methodology, arguing that its approach – a complex meta-analysis known as network meta-analysis – ‘allows the simultaneous comparison of different treatments to achieve an outcome for the population of interest'. As for the lack of quantitative summary estimates, the response argued that evidence in each of the risk/benefit domains ‘was synthesised separately' for two main reasons – that estimation of risk is different for different disorders, and to communicate risk clearly with respect to each of these disorders.

Of course, the guidelines are now the guidelines, and clinical practice will be guided more by them and (no doubt) less by the MHRA's safety reports, whose assessments apparently are based on full quantitative evidence. The MHRA's latest pronouncement (in 2015) was still that the lowest effective dose of HRT should be used for the shortest possible time, and that the decision to start, continue or stop HRT should be made jointly by a woman and her doctor, based on the best advice available and her own personal circumstances, including her age, her need for treatment and her medical risk factors.

There are certainly overlaps between most guidelines in the menopause, but a factional edge remains clear, sharply illustrated in several reactions to the NICE guidelines and which now in the BMJ bear a familiar stamp.

Clear and better information still needed

A report published in this journal last year, whose authors included its two editors, suggested that health professionals since the turn of the millennium ‘continue to let our patients down with poor provision of information, inaccurate or wrong information, or access to the right care'. ¹

These downbeat conclusions were largely based on the results of an online survey run for 20 weeks in 2014 via the website www.menopausematters.co.uk and analysed alongside results from a similar survey of 2007. The bottom line of the comparison was not as gloomy as the conclusions suggested – although only around one quarter of respondents felt their views on the menopause and hormone therapy had changed for the better, and 50% felt their family doctor did not recognise the importance of the menopause (with one-third even sensing resistance to HRT). However, compared to 2007, the authors did find significantly more women aware of the different risks associated with different types of HRT, and more able to make informed choices.

The report has now formed the basis of a recent Menopause Live update produced for its members by the International Menopause Society, and this too noted that ‘concerns remain that the information available to women is still insufficient to allow them to make informed decisions regarding treatment'. ² Some improvements found in the menopausematters survey – in making informed choice, for example – were described by Menopause Live as ‘reassuring’. However, the IMS update continued that if most women replying to the survey were well educated and well informed (as seems likely) it seems more than probable that there is still insufficient information for the majority of women, or that that information presented is unclear. Burger and colleagues in a 2012 report on the fall-out of the Women's Health Initiative argued that untreated women from 2002 onwards have ‘lost the best years of their lives’. ³

In their report last year Cumming et al. saw an opportunity for more and better information supplied via the internet, and a need for greater ‘web literacy’ among users to assess the quality of information. This too was echoed by the IMS, which noted that today more than one-third of women access their menopause information online.

Support for moderate female drinking

A prospective study of more than 20,000 women in the Diet, Cancer, and Health Study in Denmark suggests that those who increase their alcohol intake over a five-year period have a higher risk of breast cancer and lower risk of coronary heart disease than women whose alcohol intake remains stable.¹ The results echo similar findings from other studies of overall drinking, though little has been reported so far on the effect of change in alcohol intake.

This study followed up subjects from 1993 to 2012, cross-linking incidence and mortality data in coronary heart disease (CHD) and breast cancer from Danish population registries to a self-reported alcohol consumption questionnaire. The level of alcohol consumption recorded between 1993 and 1998 was then compared with that between 1999 and 2003, a mean average gap of 5.4 years.

Analysis showed that women who increased their alcohol consumption by seven or 14 drinks per week (one or two drinks a day), had hazard ratios for breast cancer of 1.13 and 1.29 when compared to women who kept their drinking stable. By contrast, a similar change in alcohol intake reduced hazard ratios for CHD. The calculations controlled for age, HRT use, dietary habits, education and BMI.

The investigators concluded that their results ‘support current recommendations of a light to moderate' alcohol intake.

Spud u can't like

Frequent helpings of boiled, baked, or mashed potatoes, as well as French fries, have been linked to an increased risk of hypertension in a huge prospective study of adult women and men.¹ The US researchers – from Brigham and Women's Hospital and Harvard Medical School – suggest that replacing one daily serving of potatoes with one serving of a non-starchy vegetable would help lower the blood pressure risk.

Although potatoes are one of the world's most commonly consumed foods and have recently been included as vegetables in US government healthy meals programmes (because of their high potassium content), they are not counted as vegetables by the WHO or by Britain's Department of Health – and are thus either in or out of the five-a-day, depending on where you live. ‘Potatoes are botanically classified as a vegetable, but they are classified nutritionally as a starchy food,' a DoH spokesperson told the Daily Mail back in 2011.

The study followed more than 187,000 men and women from three large US studies for more than 20 years – 62,175 women from the original Nurses’ Health Study, 88,475 from the Nurses’ Health Study II, and 36,803 men in Health Professionals Follow-up Study; all were non-hypertensive at baseline. Dietary intake, including frequency of potato consumption, was assessed using a questionnaire, while hypertension was reported by subjects based on professional diagnosis. Nine different response categories could be selected for potato consumption, ranging from ‘never or less than once a month' to ‘six or more a day'.

Taking account of potential confounders (dietary, BMI, smoking, exercise), results showed that four or more servings a week of baked, boiled or mashed potatoes was associated with an increased risk of hypertension compared with less than one serving a month. However, this increased risk was only evident in women, and not in men. However, a higher consumption of French fries was associated with an increased risk of hypertension in both women and men.

An editorial from Australia linked to the study was somewhat cool and proposed that overall dietary patterns and risk of disease are more useful than a focus on individual foods or nutrients.

We will continue to rely on prospective cohort studies, but those that examine associations between various dietary patterns and risk of disease provide more useful insights for both policy makers and practitioners than does a focus on individual foods or nutrients.

A functional designer estrogen – OK for mice

The first report of a truly synthetic designer estrogen which treated postmenopausal symptoms in mice ‘without raising cancer risk' has appeared in a journal of the American Academy for the Advancement of Science.¹ Researchers from three institutions in the USA claim design of a compound which mimics estrogen but only weakly binds to the estrogen receptor.

In ovariectomised mice (to mimic postmenopausal conditions) the drug reduced weight gain and body fat and protected blood vessels from injury without stimulating cell proliferation in the breast or uterus. They are thus described as structurally novel estrogens that preferentially activate a subset of estrogen receptor pathways and result in favourable target tissue–selective activity. Such ‘pathway-preferential' estrogens, say the investigators, may yet meet the huge global need for estrogens with favourable pharmacological profiles which support desirable activities for menopausal women, such as metabolic and vascular protection, but that lack any stimulatory activities on the breast and uterus.