Friday, September 01, 2006

Evaluation Points You Should Know for Health Promotion

Point 1: Methodology; establishing causation.

There are two main problems -

1) establishing causation

2) correlation

Many studies in this area use natural experiments as their method, meaning the IV as a health promotion activity is naturally occurring; e.g. whether people are exposed to the health promotion activity or not. Given that in such studies not all variables can be controlled and there are so many confounding variables that the one which contributes to the alteration cannot be singled out/identified.

We need to establish the cause in order to find out if something is working or not. How do you measure what you are doing? It is important to attempt to answer this as these experiments are often very expensive. If the causation cannot be established, is the experiment useful? People might improve or alter behaviour due to a change in habits, lifestyle, changing personal circumstances or even another health promotion activity running at the same time. If we cannot establish causation, we should at least evaluate the usefulness of such research.

Also, we need baseline data of a control population to measure any success against. This population needs to be matched on various important factors (ethnicity, class, age, etc) for a valid comparison to be made. This in itself is no small task. If not done well then the validity of any conclusions can be questioned.

There are significant problems in measuring the success of health promotion activities that are targeted at ‘invisible’ problems that are by their nature hard to measure. For example, contraceptive use, or improvement of mental health. Thus not all health promotion activities are open to measurement through an experimental approach.

Often, researchers will correlate their health promotion activity with improvement in health behaviour. For example, finding a correlation with x activity and y improvement in behaviour, e.g. health and safety adverts on telly resulted in less chip pan fires in the home. This is particularly true in national campaigns which will look at big trends; a certain behaviour will be investigated over a long period of time. However the problem here is that correlation does not mean causation.

There are other problems with the measurement of the effectiveness of health promotion activities due to not being able to measure the health behaviour in a reliable or valid fashion. For example in relation to smoking cessation campaigns, smokers might hide and smoke somewhere else, or smoke in private. This does not mean they have given up the behaviour but rather it is now hidden. This means we have an invalid measure, we can’t see them smoking but it doesn’t mean they are not doing it. Thus people may adjust their lifestyle to fit with a health promotion campaign rather stop or reduce a negative health behaviour.

Self Report Methods may be used as alternative ways of measuring the effectiveness of a health promotion activity. Unfortunately people embellish the truth. They may forget things, deny things to themselves and others, they may want to present a positive image of themselves. People are more likely to lie when they feel threatened and feel as though they will be persecuted. (e.g. heroin addiction).

A further problem is that the effect of a health promotion activity may take some time to show itself. Thus a longitudinal design is required. Without this the effects may go unnoticed due to the short length of the follow up study. Again this would reduce the validity of such investigation.

Point 2: Determinism.

Assuming that it is the health promotion activity that alters behaviour rather than any other possible factor(s). This could in some way be accounted for be an experimental design, however this is very hard when dealing with large populations. Correlation designs are often used to investigate the effectiveness of a health promotion activity however finding a correlation does not tell us anything about cause.

Further, some people argue that taking a deterministic approach to health promotion i.e. this health promotion activity causes X outcome, is a fundamentally flawed approach to adopt for use in this area. Given that there are numerous factors that affect one’s health, it would be somewhat naïve to think that a health promotion would be the determining factor in an improvement in a health behaviour. We must therefore assume that a health promotion activity is merely ONE of the many contributing variables. The measurement tools we use should reflect this, we should look at how the health promotion activity has affected or interacted with the other variables that affect people’s health. Then we may be able to see it’s usefulness.

Point 3: Validity.

The construct here is health. What is health? How can we define it? Seeing as we have problems defining health, we may have problems defining health promotion as a whole. For example – what do we mean by a reduction in smoking? 40 a day to 10 a day? Heavy to light use? Cutting out completely? Which of these is ‘healthy’. This needs to be clear in order to be able to measure it. Given the many problems we have defining health then health promotion activities will always face the difficulty of defining what it is they are measuring. However the more specific a researcher can be in regards to what they define as health and what is involved with this, the more likely they are to measure what they set out to measure.

If health is more than just the absence of illness then health promotion can legitimately focus on many areas of our lives and attempt to promote many types of behaviours and activities. But does this stretch health promotion so far that it becomes meaningless? Are teachers, social workers, police, etc, involved in health promotion?

Point 4: Ethics.

If we can’t adequately measure what is meant by health and healthy behaviours, is it right to tell people that they are unhealthy or that they have unhealthy behaviours? Is this health defined by drug companies, teachers, parents? In other words, whose agenda is this?

For example, thalidomide was given to pregnant women to reduce symptoms of morning sickness. This was very effective, thus it was an effective health promotion activity, however the children were born with stunted growth of one or more limbs. Another example is jogging, which has now been shown to be very dangerous for people’s joints. Therefore, many health behaviours may be promoted which are later (or even at the time) shown to cause serious problems in the future for individuals who undertake them.

Further, health promotion can contribute to making people’s attitudes more negative if a person sees themselves as not be doing any of the things promoted in health promotions. Is health promotion becoming a form of social control? For example, are teenage girls anorexic because of a massive complex about being overweight caused by numerous health promotions in, for example, teenage magazines? Health promotions may cause anxiety on a national level and they are very hard to avoid – on the television, bill boards, magazines, films etc.

Are people merely led to believe that we need health promotion when what we need is less of it due to the negative effects it has on us collectively by generating a culture of fear surrounding health behaviours.

Does informing people about drug use, etc, promote this activity simply through discussing it – does it normalise such behaviours to an extent that some people may no longer attach concern or stigma to such health behaviours and therefore more likely to take an active interest? In other words, is the absence of a moral message in some current health promotion activities by default advocating such activities?

Point 5: Problems of worksite health promotion activities.

Nathan (1984) discussed the advantages and disadvantages of health promotion in the workplace. He pointed out that prevention efforts benefit from a captive audience. In addition, programs in the workplace are typically convenient for employees, with many programs scheduled before and after work and even during lunchtime. Some companies allow employees to participate in health programs during work time. One problem with workplace programs is directly related to the advantage of having a captive audience: Employees may not feel free to decline participation, even if the program is ostensibly voluntary. Workers sometimes feel uneasy about attending programs that include treatment for drug or alcohol abuse. Employees may also feel pressured to participate in their company's programs to lose weight, stop smoking, and exercise more. Nathan recommended that participation records be kept separate from their employment records to ensure privacy, and Kizer (1987) argued that all employee participation should be voluntary.


Post a Comment

<< Home