Evaluation of studies of treatment harm
- 1McMaster University Hamilton, Ontario, Canada
- 2University of Auckland Auckland, New Zealand
Previous Users’ guides have described the evaluation of studies of the effectiveness of a treatment or intervention and studies of causation or aetiology (ie, studies assessing the relation between certain exposures and the development of a specific disease or condition). This Users’ guide will focus on related questions of the harmful effects of interventions or treatments—ie, the undesirable outcomes of treatments prescribed by healthcare providers. The criteria are those identified in the original JAMA users’ guide by Levine et al.1
Before we get started, a few preliminary notes. The concepts and criteria used to evaluate observational studies of treatment harm are the same as those used to assess studies of causation or aetiology. For example, an observational study may be conducted to determine the effects of second hand smoke in the home on the development of asthma in children. In such a study, children exposed to second hand smoke would be compared with those who were not exposed to see if the exposed and unexposed groups differed in terms of developing asthma. In this Users’ guide, we will be focusing on the harms of a treatment or intervention. For example, a study by Madsen et al assessed whether children who received mumps, measles, and rubella (MMR) vaccinations (intervention group) were more likely to develop autism than those who did not receive vaccinations (control group).2 In this Users’ guide, we will primarily refer to the intervention group (the equivalent of the exposed group) and the control group (the unexposed group).
When considering studies of treatment harm, readers may encounter various terms used to refer to “harm”—studies may refer to risks, adverse events, side effects (often in relation to drugs), or the safety of an intervention.
You are a nurse practitioner working in a paediatric primary care clinic. …