To the Editor: Dr Gitlin and colleagues1 reported the results of the Care of Persons with Dementia in their Environments (COPE) randomized controlled trial, assessing a biobehavioral home-based intervention to support physical function and quality of life for patients with dementia and the well-being of their caregivers. The authors reported in their flowchart that 15 of 117 randomized patient/caregiver dyads in the intervention group and 13 of 120 in the control group were excluded because they did not complete the 4-month follow-up. The primary analysis as described by the authors is exposed to attrition bias because the analysis was per protocol, threatening the benefits of randomization and the initial comparability of the 2 groups. As stated in the CONSORT extension to nonpharmacologic treatment,2 the authors should report the type of statistical analysis (by intent to treat [ITT] or per protocol) and should prefer the ITT analysis.
Pitrou I. Well-being of Patients With Dementia and Their Caregivers After a Biobehavioral Home-Based Intervention. JAMA. 2010;304(22):2482-2484. doi:10.1001/jama.2010.1789