The Cognitive Health Outcomes Investigation Of The Comparative Effectiveness Of Dementia Screening (Choice) Study: Results From The Randomized Controlled Trial

Screening for dementia in primary care (PC) is not recommended by the US Preventive Services Task Force due to lack of empirical data on the benefits and harms for patients. The CHOICE trial, a large multi-site, trial, was conducted to measure the benefits and harms of routine screening for dementia in PC. CHOICE was a single-blinded, randomized controlled trial with 1:1 allocation. A total of 4,005 individuals aged ≥ 65 years without dementia, MCI, or serious mental illness receiving care within two cities in Indiana, USA. Subjects were randomized to screening (using the Memory Impairment Screen and Mini-Cog) or no screening. Subjects who screened positive (on either test) were referred to collaborative brain care program for diagnostic assessment and evidence-based care, if cognitive impairment was diagnosed. Outcomes included health-related quality of life (HRQOL) measured by the 15 item Health Utility Index and depressive symptoms and anxiety measured by the Patient Health Questionnaire-9, and the Generalized Anxiety Disorder Scale-7 at 1, 6 and 12 months. Mixed effects models including group and time interactions were used to compare study outcomes. Patient age was 74.5 (SD 7.0), 33.4% were male, 71.0% were white, 95.6% had at least a high school education, and had 2.9 (SD 3.0) co-morbidities. There were no significant differences in socio-demographics between the groups at baseline. Among the 2,008 patients screened 6.7% screened positive. No significant interactions between group and time were found for all outcomes indicating that the effect of the intervention did not differ across group over time. Patients who were screened had similar levels of depressive symptoms (p=0.10) or anxiety (p=0.71) at 1 month and similar levels of HRQOL of life at 12 months (p=0.80). Of those who screened positive and accepted collaborative care (38%) they were not more likely to have depression (p=0.35) and anxiety (p=0.32) or higher HRQOL (p=0.70), but, they were significantly more likely to have higher health care utilization at 12 months (p=0.04) compared to patients who refused follow-up care. Routine dementia screening in PC does not appear induce risk by increasing depression or anxiety or to provide benefit, as measured by HRQOL .