Distinguishing Between Detective and Predictive Monitoring

May 23, 2016 | Sotera Wireless

“If it is accepted that any patient in hospital could deteriorate without prior warning or appropriate treatment, logic dictates that, if technically possible and affordable, all patients who are for active treatment should be continuously monitored.”[1]


According to the report from the 2008 Consensus Conference on the Afferent Limb of Rapid Response Systems, patient monitoring systems “hold the promise of both detecting and predicting patient deterioration.”[2]  Detection can be defined as the recognition of critical illness at a particular moment in time, while Prediction is the estimation of the risk of deterioration over various time frames.

Detective monitoring is a tool which enables improved patient surveillance, a primary goal of which is to rapidly identify a change in patient condition (independent of cause), and to communicate that information to the nurse.  This differs from predictive algorithms which attempt to extrapolate from the current state and assign a probability of occurrence to one or more specified future outcomes such as organ failure, ICU transfer, or death.  Thus one of the fundamental differences between detective and predictive monitoring is that predictions are often narrowly focused on very specific outcomes whereas detection will identify “all-cause” physiological changes.  For this reason, detective alarms have immediate implications for the ongoing care of the patient whereas predictive alarms have implications for future resource needs.

Detective and predictive monitoring are not mutually exclusive nor can they replace each other’s role; an optimal system will use both simultaneously: detective so as to improve the immediate delivery of care, and predictive to ensure resources are present in the event of a future downturn in clinical status, uncertain though it may be.


Benjamin Kanter, M.D., FCCP
Chief Medical Officer


[1] DeVita MA, Smith GB, Adam SK, et al. Identifying the hospitalized patient in crisis. A consensus conference on the affer­ent limb of Rapid Response Systems. Resuscitation. 2010;81:375-82.

[2] ibid