During change-of-shift handoffs, nurses share information about patients' status and needs that influences clinical judgment. Early warning scoring systems have been developed to help the health care team identify deterioration in patients and avoid potential errors; however, some nurses doubt the value of these scores and are reluctant to use the scoring systems. Researchers evaluated the degree of agreement between acute care nurses in their judgments of patient stability and risk compared with numerical ratings reflected in three widely used early warning scores (MEWS, NEWS, and ViEWS).

Data were collected from acute care nurses on one surgical and two medical units in a tertiary acute care hospital. Over the course of the study, 62 nurses carried out 444 handoffs for 158 patients. After the handoff, incoming and outgoing nurses rated a patient's risk of deterioration on the Patient Acuity Rating (PAR) scale.

Agreement between nurses for the same patient was fair to moderate. Correlations between nurses' judgments using the PAR and patients' early warning scores were low (0.20 to 0.22; P < 0.001). Levels of agreement between nurses didn't differ depending on their years of experience and educational level. Focus group interviews revealed that nurses had doubts about the concept of probability when predicting risk of patient deterioration. Instead, they were more comfortable thinking about patients' relative risk, and compared their assigned patients with their other assigned patients to guide their ratings of risk.

The authors point out that the study was conducted in only one hospital and didn't account for potentially confounding variables. Also, the number of incidents was low, which made it difficult to evaluate the predictive power of nurses' clinical judgments.