| Learning Objective #1: Describe a non-specific treatment control group | |||
| Learning Objective #2: Discuss construct validity issues related to non-specific treatment control groups |
Objective: The selection of a control group in an intervention study is usually based on addressing threats to internal validity but may also include consideration of threats to construct validity. The control condition should be different from the treatment condition, i.e. lacking the critical therapeutic components, yet avoid attrition by involving the participants in some type of credible treatment experience. A nonspecific-treatment control group or "minimal treatment" condition can address both issues. Within a focused yet neutral control treatment, the investigator can include common therapeutic factors that may account for the study results. This paper discusses issues related to the design of a nonspecific-treatment control group in an insomnia intervention study for breast cancer survivors. Study Design: Breast cancer survivors (BCS) are randomly assigned to behavioral treatment (BT) or sleep education and hygiene control (SEHC). Participants complete pre- and post-treatment instruments to measure sleep, fatigue, mood, and quality of life. Upon completion of the non-specific treatment, the SEHC participants are offered BT, if desired. Intervention: The BT group is a multi-component intervention consisting of stimulus control instructions, sleep restriction therapy, and sleep education and hygiene. The SEHC condition is the nonspecific-treatment control group. Non-specific effects associated with therapist directed intervention such as therapist contact, treatment contact hours, treatment rationale, etc. will be discussed. Construct validity issues will be examined by comparing and contrasting the two conditions and their components. Implications: In order to demonstrate why BT reduces insomnia and its importance in treating BCS, the use of a non-specific treatment control group or another treatment condition is necessary. Previous research reveals that sleep-focused control treatments are highly credible yet have limited efficacy. Although these control treatments may not identify which treatment construct produces the behavioral change, they can reduce construct validity threats.
Funding: National Cancer Institute CA91869
Back to Research Methods
Back to 14th International Nursing Research Congress
Sigma Theta Tau International
10-12 July 2003