Methodological Flaws in “The Clinical and Economic Impact of Nurse to Patient Staffing Ratios in Women Receiving Intrapartum Oxytocin”

On behalf of the 24,000 members of the Association of Women's Health, Obstetric and Neonatal Nurses (AWHONN), we are writing in response to an article published in the February 2014 issue of the American Journal of Perinatology, called “The clinical and economic impact of nurse to patient staffing ratios in women receiving intrapartum oxytocin.”

AWHONN is concerned with several fundamental flaws in the methodology of the research and thus the conclusions that result from those flaws. In the earlier mentioned study, AWHONN is concerned with the following:

1. Flaws in the data collection methods used: The data collected about how often women in labor receiving oxytocin had one-to-one nursing care was collected based on the opinion of nurse leaders. Hospitals were categorized based on the opinion of one nurse leader at each site who reported the percentage of time one-to-one nursing care was provided for women receiving oxytocin during labor. These nurse leaders self-reported data were never confirmed using chart reviews or financial administrative data and therefore cannot be considered as a valid or reliable method of grouping hospitals for this analysis. The self-reported data from the nurse leaders were submitted to researchers who are not independent from the system in which the nurse leaders work. Therefore, this study design introduces the potential for recall bias and undue influence on the individuals reporting their staffing data.

2. Flaws in the study inclusion and exclusion criteria: The authors provide no evidence that the outcomes selected are nurse sensitive. Patient acuity, level of service, type of hospital services provided, and birth volume at each hospital were not reported in the study or included in the analysis, yet each of these characteristics are known to affect nurse staffing ratios. Furthermore, more than one-half of the sample did not receive oxytocin and thus did not meet criteria for one-to-one care; they were not relevant to the research question, yet all patients were included in the study.

3. Flaws in analysis: Although the authors discuss large numbers in their series, there is no power analysis. This is particularly important for the birth asphyxia analysis as this is a rare outcome and it is unclear if the study was powered to detect a difference.

In addition, there are factual errors in the discussion of the implications of AWHONN's “Guidelines for professional registered nurse staffing for perinatal units.” For example, the recommended ratio for unstable antepartum patients is one-to-one, and the recommended ratio for a woman on oxytocin is one-to-one. The rationale for these recommendations is that both women are high risk. The authors assert that if a woman receiving oxytocin has a nurse patient ratio of one-to-one this could require the hospital leaders to reduce the nurse patient ratio for a woman at 28 weeks with a bleeding previa and thus put this woman in jeopardy. However, this does not recognize the fact that if the unit staffing allows for both patients to have one-to-one nursing care, as is recommended, there is much more capacity for resilience in the face of emergent developments and unplanned patient care needs. Furthermore, AWHONN's staffing guidelines state that they are recommendations for planning purposes, not mandates.

AWHONN recognizes there are many significant challenges in studying the effect of nurse staffing on patient outcomes in the perinatal setting. Thus, caution is needed when designing and interpreting staffing studies. Other potential confounding variables include the following: not all nurses have an equal amount of expertise; not all women and fetuses have the same nurse care for them throughout their entire labor; and not all units have the same amount of support personnel or the same geographic layout. Examples of geographic layout differences include the location and availability of a central station or satellite stations or the location of birthing centers on separate floors of the hospital. Both examples affect how often and how easily the nurses working on a given perinatal unit are able to interact and garner support from other nurses and other members of the health care team during an emergency. We also recognize the volume and nursing care needs of other patients do not remain constant.

Despite these research challenges, high-quality research in the area of perinatal nurse staffing is critically important. Unfortunately, the study by Clark et al is lacking the methodological rigor required to make a meaningful contribution to what is known about the impact of the nursing care provided to women who are receiving oxytocin for labor induction and/or augmentation. AWHONN urges the American Journal of Perinatology to apply the same standards of rigor in nurse staffing research as are applied to other areas of research.

• Reference

• 1 Clark SL, Saade GA, Meyers JA, Frye DR, Perlin JB. The clinical and economic impact of nurse to patient staffing ratios in women receiving intrapartum oxytocin. Am J Perinatol 2014; 31 (2) 119-124