Expert opinion: Understanding the evidence

Clinicians need to be able to offer patients an informed assessment of risks and benefits, writes Dr Sarah Gray

Women’s health represents a significant proportion of primary care workload, particularly for female clinicians. To carry out this work effectively, we need to be able to provide explanations and analyse risks and benefits in terms that are relevant for patients. To do this, we have to understand the difference between evidence of absence of therapeutic effect, and absence of evidence.

Negative research findings are often reported in the media with much greater prominence than positive findings or less alarming reassessment of data. This makes it very difficult for us to counsel our patients about objective decision-making.

The policy of placing an embargo on research results until after they have been published keeps the newspapers ahead of the professionals and can undermine our credibility. It is worthwhile reading the original papers and following the subsequent editorials and letters generated by them, to understand how the papers were received.

Interpreting research on HRT

It can be difficult to work out what the published research actually means for patients. The first principle would be to understand the characteristics of the population that has been studied. How similar were they to your practice population and, in particular, can the data from the study be extrapolated to the situation of the individual in front of you?

Let us examine the Women’s Health Initiative (WHI) as an example.1 This meets all of the accepted criteria for grade A research, being a large, multicentre, randomised, double-blind placebo-controlled trial. The study was designed to test the hypothesis that HRT would prevent heart disease.

Observational data had convincingly shown that women taking HRT experienced about half the acute cardiac events of non-treated women from the same population. Was this a difference between the women themselves, or was it the hormones that made the difference?

The women studied ranged in age from 50 to 79, with an average age of 63 at randomisation. The study population was largely asymptomatic. There were high rates of risk factors for cardiovascular disease, such as hypertension and obesity. Almost 8 per cent had already experienced a defined coronary event.

Regardless of age, those women with their uterus intact were randomised to a continuous combined hormone regimen or placebo, and followed for just over five years. A small excess of events was observed in the first year of use, although there was no overall benefit or harm by the end of the five-year period.

The press briefing led to alarmist headlines, because it was not made clear that these women were unlike the symptomatic healthier and younger population for whom we may consider initiating hormone treatment in the UK.

When the results of the WHI were analysed again in 10-year cohorts, the initial excess risk disappeared for women aged 50–59.2 Evidence is accumulating from this and other sources that starting HRT soon after menopause appears not to represent a cardiac risk. This is information that we can usefully convey to our typical patient.

The WHI has, however, confirmed that there are increased risks of stroke and venous thromboembolism (VTE). This increase is related to the size of the background risks. The regimen of combined hormones showed a doubling of VTE and a 30 per cent increase in ischaemic stroke. Age is very relevant and for younger women with low background risk, their excess attributable risk is small. Women need to understand the size of this risk to balance it against the perceived benefits for symptom relief, urogenital atrophy, or osteoporosis prophylaxis.

Conveying risks to patients

Talking to patients about benefits and risks is not easy to grasp if relative risks or percentage increases are used. How can women understand what it means for them? What is it relative to?

It is important to express risks in absolute terms that can be represented visually and easily appreciated. In the WHI, the excess of cardiac incidents was reported by the tabloids as a 25 per cent increase in heart attacks. This equated to eight events for every 10,000 person years of use. This was indeed about 25 per cent greater than the background risk of 30 in 10,000 but still affected fewer than 1 in 1,000 women.

We still do not know how much drug dosage (possibly too high for this age group), dose and type of progestogen and the route of administration affects these risks. Absolute risks may be further simplified by using the WHO classification: <1 in 10,000 is ‘very rare’, >1 in 10,000 but <1 in 1,000 is ‘rare’ and >1 in 1,000 is ‘appreciable’.

When we talk to women we need to discuss the strength of the available evidence so that they can share in the management of uncertainty and make as informed a choice as possible. The medical indemnity organisations advise that provided discussion is overt in covering issues of risk and benefit and is appropriately documented, informed choice is not a problem for them.

Women’s health issues can have great impact on quality of life. The right intervention for the right reason can make a difference. Choice cannot be informed if the information is lacking, of poor quality, or inadequately explained. We must try to get it right.

- Dr Sarah Gray is a GP specialist in women’s health in Truro, Cornwall


1. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002; 288: 321-33.

2. Hormone replacement therapy and coronary heart disease: the role of time since menopause and age at hormone initiation. J Womens Health (Larchmt) 2006 Jan-Feb: 15 (1): 51-3.

Want news like this straight to your inbox?
Sign up for our bulletins

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins

Already registered?
Sign in

More from MIMS

Antidiabetic Agents, Use in Renal Impairment

Antidiabetic Agents, Use in Renal Impairment

Dosing of diabetes drugs in patients with renal impairment.

Insulin Delivery Devices

Insulin Delivery Devices

Features and compatibility of insulin pens.

Drug shortages - live tracker

Drug shortages - live tracker

Check on the latest drug supply issues with MIMS' shortages...

Blood Glucose Testing Strips and Meters

Blood Glucose Testing Strips and Meters

Compatible test strips for blood glucose meters.