Research Intervention Fidelity: Tips to Improve Internal Validity of Your Intervention Studies
The clinical nurse specialist is often involved in intervention studies to determine whether a specific treatment or program improves some important outcome. However, a common problem with all intervention studies is the possibility of a type I or II error that is attributable to how the intervention is administered rather than how well the intervention worked.1 A type I error occurs when the researcher incorrectly concludes that there is a difference between groups (those who receive the intervention and those who do not receive the intervention), when the difference may be caused by the introduction of unintentional bias into the protocol.1 In contrast, a type II error occurs when the researcher concludes that there is no difference between groups, although there actually is a difference.1 Any variation in how the intervention is administered will decrease power and make it difficult to find a difference, although a difference may actually exist.1,2
Understanding how to prevent or minimize intervention variation is a key factor when designing a sound and rigorous research study. Ultimately, researchers want to accurately assess the impact of an intervention on an outcome of interest. The purpose of this article was to identify issues associated with intervention fidelity and to provide tips that researchers can use to enhance the validity of intervention study findings.