JMIR Publications

ABSTRACT

Background:

Patients admitted to hospitals’ internal-medicine departments, are inherently at risk of deterioration. Early detection of deteriorating patients may be lifesaving. Frequent remote patient monitoring has the potential to allow such early.

Objective:

Analyze physiological parameters in such patients using a novel, wearable wireless monitor (WM).

Methods:

A prospective, observational study in two internal medicine departments in a large, tertiary medical center. Patients were considered at risk upon hospital admission and were monitored during hospital stay by standard monitoring device and a WM simultaneously. Data from the WM was blinded during hospitalization and analyzed retrospectively. We compared deterioration detection by either alert generated by medical staff, summation of parameters to the national early warning score as transmitted by the WN, ABCNO scores, and an early warning prediction algorithm, based on the WM output.

Results:

410 patients were recruited and 217 were included in the final analysis. The median age was 71 [IQR 62-78] and 130 (60%) were males. In-hospital deterioration events were diagnosed by the medical staff in 24 patients. A retrospective analysis of the WM output detected a NEWS score-based deterioration in 16 (67%) patients, preceding the staff by mean of 29 hours and an ABCNO score-based deterioration signal that preceded the staff warnings by mean of 38 hours in 18 (64%) patients. The WN-based prediction algorithm detected a deterioration signal in 22 (92%) patients, preceding the staff detection by an average of 36 hours.

Conclusions:

Frequent remote patient monitoring has the potential for earlier detection of hospitalized patients’ deterioration using both grouped-signal based scores and algorithm-based prediction. Clinical Trial: : Israeli ministry of health registration: MOH_2020-07-12_009133; NIH registration: NCT04220359.