Understanding risk

There is a growing body of evidence from across the world supporting a range of lifestyle changes as a means of reducing the risk of developing dementia. Scientists have investigated a number of health, lifestyle, and environmental factors suspected of having an influence. Results from these studies are often given considerable publicity in the popular press with dramatic headlines such as:

• Exercise 'cuts Alzheimer's risk‘
• Fish ‘fights Alzheimer’s’
• Curries are cure for Alzheimer’s
• Mental aerobics ‘halt dementia’

Although many studies have taken place, some carry more weight than others. Some have used animal models which have shown promising results in the laboratory but these do not necessarily transfer to the human population. Even where research with humans has taken place it may have involved small numbers of people or have covered a short time-scale.

There are other ‘health warnings’ for how you should interpret research findings. Much of the ‘best’ evidence comes from studies of large groups of people (population studies) but the evidence may not apply to a particular individual. There is no guarantee that acting on even the best evidence available will help a particular person avoid or delay developing dementia.

The strongest type of research evidence comes from studies where people are randomly chosen to be given or not given a treatment, or to carry out a certain strategy. The group of people who are not given the treatment or strategy is called the ‘control’ group. The researchers can then see if there is a difference in what happens to each group. However, randomised controlled trials like this are not practical for the kinds of lifestyle factors which may affect dementia risk over very long periods of time. Because there have been no controlled studies that have shown the effectiveness of any particular change in lifestyle, we cannot say for certain that a change in lifestyle will actually result in a reduction of risk.

Instead, we have to use epidemiological evidence. This is based on studies which follow groups of people for very long periods of time (‘longitudinal studies’). Large studies are more useful than small scale studies. There are a number of these studies across the world which have been running for many years, and which report new findings every so often. However, many of the conclusions remain tentative because it is hard to show direct links with any one risk factor – unless all other potential risk factors can be eliminated.

Another problem is that it is not possible to be sure of what is the cause and what is the effect. For example, if a study shows that people who do more crosswords have a lower risk of developing dementia years later, this may be due to the mental activity having a protective effect. But it is also possible that the reason that some people don’t do as many crosswords is that they have some very early and undetectable brain changes which would eventually be diagnosed as dementia.

Alzheimer’s disease in particular may take 30-40 years to develop(1) , emphasising the need to take action earlier in life which may help prevent it. But this also presents researchers with a huge challenge. How can they collect, track and analyse behavioural data going back 20, 30 or 40 years to investigate how that behaviour has influenced incidence of dementia today? Some data, such as blood pressure or weight may be confirmed by checking medical records but asking people to recall what they ate or drank during that time provides less reliable data.

Identifying risk factors

The causes of some diseases are easy to identify, which makes it easier to take steps to prevent people developing the disease in the first place. This is particularly true for infectious diseases like measles where vaccination programmes have dramatically reduced the incidence of the disease.

For other diseases, the situation is much more complicated, particularly where the condition develops over a long time, like many forms of dementia. A person’s chances of developing the disease may be influenced by many, interacting, factors and it may be impossible to identify one single risk factor that is enough to cause the disease.

Epidemiology, which studies the factors affecting the patterns of disease, has led to the discovery of many risk factors for conditions like heart disease and cancer. As people change their lifestyles by giving up smoking, taking regular exercise or eating a healthier diet, changes in the patterns of these diseases are emerging.
Risk factors for dementia are also now becoming clearer and there are several important leads which are explored later in this booklet.

What is risk?

Many of the newspaper headlines about medical research make startling claims:

• Obesity triples the risk of dementia, women warned
• Regular exercise in younger years ‘reduces Alzheimer’s risk by 60%’

What do these figures mean for a particular individual? How ‘big’, or small, is a particular risk?

In simple terms, a health related risk factor is something that increases our chances of something negative happening to us. So we may read that having no children is a risk factor for breast cancer. But this does not mean that having no children causes breast cancer, just that more women without children get breast cancer than women who have children. This also underlines another issue about risk – risks are measured by looking at large numbers of people, not individuals, so what is true for a large population may not be true for an individual.

Let us look at how one dementia-related study was reported:
‘HRT "doubles" Alzheimer's risk' was the headline but what did the research actually show? The American Women's Health Initiative Memory Study (2) followed 4,500 postmenopausal women aged 65 and older for an average of four years. Half the women received a daily tablet of combination hormone replacement therapy while the rest were given a placebo or dummy pill. The study found that taking combination HRT doubled the risk for probable dementia in women aged 65 and older. But what does ‘double the risk’ mean?

In this study, after an average of four years, 40 women given the HRT and 21 women given the placebo developed dementia, leading to the suggestion that taking combination HRT doubles the risk, since 40 is (nearly) two times 21.

‘Doubling’ the risk sounds very big and worrying. However, the overall individual risk of developing dementia in women in this age group is small to begin with. If the findings of this study are applied to a population of 10,000 women who took the combined HRT, there would be an extra 23 cases of dementia per year; so the risk to individuals remains small.

Presenting risk

There are two main ways that risk can be presented – as absolute risk and/or relative risk. The same research results can appear alarmingly different depending on which method is used, particularly in newspaper reports.

The Patient UK website(3) describes absolute and relative risk very clearly:

“Absolute risk is your risk of developing a disease over a time-period. We all have absolute risks of developing various diseases such as heart disease, cancer, stroke, etc. The same absolute risk can be expressed in different ways. For example, say you have a 1 in 10 risk of developing a certain disease in your life. This can also be said a 10% risk, or a 0.1 risk - depending if you use percentages or decimals.”

"Relative risk is used to compare the risk in two different groups of people. For example, the groups could be 'smokers' and 'non-smokers'. All sorts of groups are compared to others in medical research to see if belonging to that group increases or decreases your risk of developing certain diseases. For example, research has shown that smokers have a higher risk of developing certain diseases compared to (relative to) non-smokers.”

So, one study might suggest that risk is reduced by 50% which sounds impressive, while another study might say the risk is reduced by 1%, which has much less impact. Yet both are talking about the same level of reduction in risk – the first about relative risk and the second about absolute risk.

Unfortunately, it is not always clear in press reports whether they are writing about absolute or relative risk. But it is important to bear in mind the difference between the two when reading reports.

Having established that risk can be expressed in different ways, what does an increase (or reduction) in risk really mean?

If dementia affects one person in 50 between the ages of 65 and 70, that can be described as a 2% risk or a 0.02 risk or a 2 in 100 risk. If an imaginary research study reported that taking a magic pill every day reduced that risk by 50%, this would mean that the risk of developing dementia from taking that pill is 50% lower than the existing 2% risk for those who don’t take the pill. That is the relative risk decrease from taking one magic pill a day.

But how big a difference does that really make? If the existing risk is 2%, 50% of 2% is 1%, therefore 1% is the size of the decrease in absolute risk. So instead of two people in every hundred developing dementia, one person in every hundred people taking the magic pill will develop dementia.

Prevention or delay?

Because dementia is more common in older people, even delaying the onset by five years is predicted, in time, to halve the number of people with dementia. It is not possible to tell from the research whether a risk factor affects whether someone gets dementia at all or whether it would happen anyway, but later, by which time some people would have died from other causes.

The next chapter will look at the specific risk factors for dementia.

Next section: What are the risk factors for dementia?
References
Glossary
Return to risk reduction contents page

Useful links:

Information about dementia