A study from researchers at the Mayo Clinic in Rochester, analyzed the detection of when a patient is deteriorating which is critical to hospital care. For example, two of the most common causes of acute inpatient deterioration are sepsis and acute respiratory failure, which have an in-hospital mortality of 20–30% and are involved in 34–52% of in-hospital deaths. These conditions show increased mortality if interventions are delayed. Our study aimed to evaluate whether the accuracy of nursing judgement, based on the “worried criterion” in detecting impending physiological deterioration merits its inclusion in the electronic medical record. The study is available on JAMIA Open.
To improve the detection of acute inpatient deterioration, one solution is to use automated scores based on data from the electronic health record (EHR). Several scores using a combination of vital signs and other inputs, called early warning scores (EWS), have been developed to improve the recognition of inpatient physiological deterioration. However, EWS do not demonstrate accurate predictive capabilities when applied strictly, and to date they have failed to provide strong evidence of their ability to improve outcomes.
EWS are being incorporated into the electronic medical record and are used to inform the decision to activate rapid response teams (RRTs; a team of clinicians with specific expertise in responding to acutely deteriorating patients). Some of the most sophisticated EWS incorporate certain nursing assessments when using a data science approach, but they are generally limited to more objective assessments currently available in the EHR, such as neurological, skin, or nutritional status. While these approaches benefit from including a wider range of information not limited to vital signs or laboratory results, they are still missing an important piece of information that is currently not captured in the EHR: nursing assessment of patient risk of deterioration.
Despite this information not being routinely captured in the EHR, the most common criterion used to activate the RRT is the “worried criterion,” which is based on nurses’ pattern recognition. However, the predictive accuracy of nurses’ judgement of risk, whether based on analytical or intuitive pattern-recognition processes has not been evaluated.
The study aimed to evaluate whether the accuracy of nursing judgement, based on the “worried criterion” in detecting impending physiological deterioration merits its inclusion in the electronic medical record.
Identification of hospitalized patients with suddenly unfavorable clinical course remains challenging. Models using objective data elements from the electronic health record may miss important sources of information available to nurses.
The research team recorded nurses’ perception of patient potential for deterioration in 2 medical and 2 surgical adult hospital units using a 5-point score at the start of the shift (the Worry Factor [WF]), and any time a change or an increase was noted by the nurse. Cases were evaluated by three reviewers. Intensive care unit (ICU) transfers were also tracked.
Nurses’ pattern recognition and sense of worry can provide important information for the detection of acute physiological deterioration and should be included in the electronic medical record.