Search strategy and evidence selection

Search strategy and evidence selection

Search strategy

The following search strategy was used to search Ovid MEDLINE (R) 1946 to May week 2 2015:

"" or ""

Similar search strategies were adapted for Medline in Process, Embase, PsycInfo, Cochrane Library (all components), Pubmed, HEED, NHS Evidence and Web of Science. The searches returned a total of 15 references after duplicate removal.

Evidence selection

Retrieved results were independently sifted by two researchers using the selection criteria below, and disagreements discussed and resolved.

  • Population: Non‑critical care patients who may deteriorate and are not constantly monitored

  • Intervention: EarlySense monitoring system

  • Comparator: Standard ICU monitoring systems for heart rate (e.g. ECG) and respiratory rate (e.g. plethysmorgraphy/end tidal CO2)


  • Heart and respiratory rate diagnostic accuracy – sensitivity, specificity, PPV, NPV

  • Prediction of patient deterioration/medical events using heart and respiratory rate, and rate of false alarms

  • Rate of code blue events and ICU transfers

Following the first sift, 9 records were removed based on the following criteria:

  • Not relevant to selection criteria (out‑of‑scope setting)

  • Review articles, protocols or editorials

  • Duplicate results with other publications found. In these cases, peer‑reviewed journal publications, more recent publications, and larger cohort sizes (in ongoing studies with updates) were selected for inclusion.

Full articles were retrieved for the remaining 7 studies. Due to the paucity of data, all studies meeting the selection criteria were considered for inclusion. Ultimately, 6 references met the criteria and were included in this briefing.