Related Experiment Video
Updated: Sep 16, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Economic evaluation of a one-time screening for bipolar disorder using the EarlyDetect model in primary care in
Negar Razavilar1, Brian Chan1, Pratap Chokka2
1Institute of Health Economics, Alberta, Canada.
Background:
EarlyDetect is a digital self-administered screening tool that has been recently developed to help clinicians diagnose bipolar disorder (BD). Improved diagnosis of this mental health condition is expected to reduce health care utilization and cost burden. We present an early health technology assessment (eHTA) threshold analysis exploring the potential value of EarlyDetect.
Methods:
This study is not a traditional cost-effectiveness analysis but an exploratory eHTA using a threshold analysis approach. A decision analytic model was developed to evaluate the five-year impact of EarlyDetect compared to Mood Disorder Questionnaire (MDQ) or no screening on a cohort of individuals presenting to a primary care physician with a new depressive episode. Model parameters were derived from published literature or expert opinion. The percentage of individuals misdiagnosed was estimated. Total five-year health care costs were calculated from the Canadian public health care payer perspective and reported in Canadian dollars. Results were presented across the range of cut-off values.
Results:
Over a five-year period, EarlyDetect decreased health care costs by up to $6868 and decreased the number of misdiagnosed individuals by 3.5 % when compared to no-screening. For MDQ, health care costs were reduced by up to $4690 with a reduction of misdiagnosed individuals of 2.3 %.
Conclusions:
BD screening with EarlyDetect and MDQ may lower health care costs and decrease the number of individuals misdiagnosed compared to no screening. These benefits may be greater for EarlyDetect than with MDQ. This analysis has limitations and includes several assumptions. The results should be interpreted with caution.
