Authors

  1. Rosenberg, Karen

Abstract

According to this study:

 

* A higher nursing skill mix (that is, one that includes more RNs) is associated with improvements in 12 patient outcomes.

 

* These 12 outcomes are length of stay; ulcer, gastritis, and upper gastrointestinal bleeds; acute myocardial infarction; restraint use; failure to rescue; pneumonia; sepsis; urinary tract infection; mortality/30-day mortality; pressure injury; infections (other than urinary tract infection); and shock, cardiac arrest, or heart failure.

 

 

Article Content

Nurse staffing levels and their impact on patient outcomes have been studied, but the effect of nursing skill mix on outcomes has received less attention. A systematic review was conducted to expand on previous research examining the association between nursing skill mix-which is described as the proportion of total nurse hours provided by RNs-and patient outcomes. The researchers assessed patient outcomes sensitive to nursing care in acute care hospitals.

 

Of the 63 studies included, most were observational studies without a comparison group. Outcomes were considered sensitive to nursing skill mix if at least three-quarters of the studies reporting on that outcome found an inverse significant relationship between them (that is, a higher proportion of RNs in the skill mix was associated with fewer adverse patient outcomes). Based on these criteria, 12 patient outcomes were sensitive to nursing skill mix: length of stay; ulcer, gastritis, and upper gastrointestinal bleeds; acute myocardial infarction; restraint use; failure to rescue; pneumonia; sepsis; urinary tract infection; mortality/30-day mortality; pressure injury; infections (other than urinary tract infection); and shock, cardiac arrest, or heart failure. Any relationship between skill mix and other patient outcomes was inconclusive.

 

The results should be interpreted with caution, according to the authors, because of methodological flaws in the studies. For instance, few studies included experimental designs, and many collected self-reported data using single measurement methods. More experimental or longitudinal studies are needed to establish causal relationships, the authors conclude.

 
 

Twigg DE, et al J Adv Nurs 2019 Sept 4 [Epub ahead of print].