Reliable Blood Pressure Self-measurement in the Obstetric Waiting Room

Abstract

Patients often fail to adhere to clinical recommendations when using current blood pressure self-measurement (BPSM) methods and equipment. As existing BPSM equipment is not able to detect non-adherent behavior, this could result in misdiagnosis and treatment error. To overcome this problem, we suggest introducing an alternative method for achieving reliable BPSM by measuring additional context meta-data for validating patient adherence. To facilitate this, we have developed ValidAid, a context-aware system for determining patient adherence levels during BPSM. The aim of this study was to validate this new reliable BPSM method based on ValidAid in the clinical setting. Specifically, we wanted to evaluate ValidAid's ability to accurately detect and model patient adherence levels during BPSM in the clinic. The validation was done by asking 41 pregnant diabetic patients scheduled for self-measuring their blood pressure (BP) in the waiting room at an obstetrics department's outpatient clinic to perform an additional BPSM using ValidAid. We then compared the automatically measured and classified values from ValidAid with our manual observations. We found that a) the pregnant diabetics did not adhere to given instructions when performing BPSM in the waiting room, and that b) the ValidAid system was able to accurately classify patient adherence to the modeled recommendations. A new method for ensuring reliable BPSM based on the ValidAid system was validated. Results indicate that context-aware technology is useful for accurately modeling important aspects of non-adherent patient behavior. This may be used to identify patients in need of additional training, or to design better aids to actively assist the patients during measurements. ValidAid is also applicable to other self-measurement environments including the home setting and outpatient clinics in remote or underserved areas as it is built using telemedicine technology and thus well-suited for remote monitoring and diagnosis.