Why Wear Duration Mattered in an Early SCRAM Validation Study
A validation study can report one average detection rate while the device performs differently at different points in a deployment. Marques and McKnight analyzed wear duration in an early study of SCRAM and a prototype wrist monitor.
How the investigators examined wear
Twenty-two participants wore both devices during laboratory and field observations. The SCRAM analysis divided 271 drinking episodes into shorter- and longer-wear groups. The groups averaged approximately 3.3 and 12.5 days of wear, respectively. The researchers found evidence of poorer SCRAM performance with longer wear and identified water accumulation in the sensor housing as a suspected explanation.
What to verify before using the finding in court
Identify the model and generation of the actual device and compare its installation, service and replacement dates with the event date. The paper discussed a later version that warranted its own evaluation. The historical association supports checking how the particular device was maintained and validated. It does not establish that a current device had the same condition.
Marques and McKnight, Alcoholism: Clinical and Experimental Research (2009)
