Consider whether the people being compared were cared for similarly.
For a comparison to be reliable, the people in the comparison groups should be cared for in the same way except for the health actions being compared. Otherwise, differences in outcomes may be due to differences in the care the participants received, rather than the effects of the health actions being compared.
Explanation
If people in one comparison group receive additional treatments or more care and attention than people in the other comparison group, this can affect the reliability of the results of a comparison of health actions (things people do for their health).
Healthcare providers who are aware of the health action to which people are allocated may act differently based on their attitudes for or against a particular health action. This can affect the outcome the researchers are interested in. One way of preventing this is to keep the healthcare providers unaware of (“blind” to) which people have been allocated to which health action.
In a comparison, the people who could be “blinded” include the people receiving the health actions being compared, the people delivering treatments, data collectors, people who assess the outcomes, data analysts, and the people who write about the comparison. People sometimes use the term ‘double-blinded’ but this can be a misleading term as it does not say exactly who was blinded. It is better to consider specifically who was blinded and how that could lead to overestimation or underestimation of treatment effects.
It is not always possible to blind care providers in randomized trials, and it is rarely possible in non-randomized studies of the effects of health actions, such as cohort studies or case-control studies. When blinding is not possible, it is important to consider whether people in the comparison group were treated differently (aside from receiving a different health action) and that this could influence the results of the study.
Example
In a randomized trial of a type of psychological treatment called cognitive behavioural therapy (CBT) for hypochondriasis (persistent fear or belief that one has a serious, undiagnosed illness) compared with no CBT, a detailed letter with advice was sent to the doctors whose patients were allocated to receive CBT. This led to the researchers not being sure whether differences in outcomes were caused by CBT alone or whether the letter to the doctors could have altered how they cared for the participants allocated to CBT. In addition, participants in the CBT group received more attention than those who did not receive CBT. So, it is uncertain how much of the observed difference in outcomes was due to attention, support, concern, and positive expectation and not specifically to CBT.
Remember: Be careful about believing the results of comparisons if people in the groups that were compared were not cared for similarly (apart from the health actions being compared). The results of such comparisons can be misleading.
- Video: Placebo effect, control groups, and the double blind experiment This is a Simple Learning Pro video lasting 3:35 minutes.
- Blog: Comparison groups should be treated equally. Students 4 Best Evidence.