3.2.4 Intervention studies
Whereas the types of study described so far, including case-control studies, are purely observational in nature (i.e. they only observe what is happening), intervention studies involve intervening with a group of people, and include an equivalent group which acts as a ‘control’. The most popular study of this kind is the randomised controlled trial (RCT). RCTs divide the population to be studied into groups on a random basis; one group is then subjected to a treatment, procedure or intervention, the other not. If the two groups are matched in terms of their characteristics, then any measurable differences between them should be due to the intervention. Ideally, the RCT should be carried out using a double-blind method: that is, neither the researcher nor the subject knows who is in the intervention or control arm of the study.
Intervention studies are conducted in the following way. Initially, observations (or measurements) are made on a population which is then divided (by random sampling) into two equivalent groups. Of these, one is subjected to the intervention. After some predetermined period following the intervention, the observations are repeated and the results for the two groups compared to establish whether or not the intervention had any effect.
Drug trials are by far the most common type of intervention studies. Their purpose is to discover the effects and effectiveness of new drugs developed by the pharmaceutical industry. Objections to trials such as these are often on ethical grounds. If a study is investigating the effects of a particular treatment on cancer sufferers, for instance, then it could be seen as unethical to withhold the potentially beneficial treatment from the control group who would also need to be cancer sufferers. However, conventionally, the control group would be receiving the current best treatment. Other ethical problems relate to whether the participants are fully informed about the potential effects of participating. The 1964 Declaration of Helsinki on medical ethics (see World Medical Organisation, 1996) makes it clear that RCT participants must receive adequate information about the aims, methods, expected benefits and possible hazards involved in the intervention. Participants must feel free to withdraw without putting at risk their previous treatment.
Thinking point: Can you think of any groups in the population who may be vulnerable to exploitation in RCTs?
Older people have traditionally been highly deferential to the medical profession and may feel that they cannot refuse to take part for fear of losing the goodwill of their doctor. Also, great care needs to be taken with people for whom English is not their first language to ensure that they fully understand what is involved. To safeguard the rights of vulnerable groups, all trials must gain ethical approval from local ethical committees before they are allowed to recruit people to their study
Randomised controlled trials use an experimental approach which is largely restricted to single-factor interventions and interventions in closed systems such as schools or health clinics, because these are the contexts in which it is possible to exercise more control and to have more knowledge about the various factors that may influence the outcomes. The experimental approach assumes a greater degree of control over extraneous factors than is usually possible or desirable in most social settings.
Thinking point: What might be the main problems with trying to use an experimental approach in public health settings?
The random assignment of people in a community to intervention and control groups can be problematic given that people are social and cultural beings whose lives cannot become separated from one another. Many interventions are meant to influence whole communities and in these situations the problem of trying to create experimental conditions is great.