• My File Drawer 
• Journals 
• Featured Journal 
• Journal Supplements 
• Insiders 
• Recommended Reading 

• My CE Planner 
• Browse or Search CE 
• Recommended CE 
• CESaver Collection 
• CESaver Accounts 
• CEConnection Employer Accounts 
• Oncology Times CE Tests 
• Nursing Assistant CE Articles 

• Clinical Updates 
• Drug Updates 
• AHA Connection 
• Focus On: Nursing Ethics 
• Focus On: Diabetes 
• Focus On: Trauma 
• Focus On: Asthma 
• Patient Education
• White Papers

• Free eNewsletters 
• NursingCenter eNews 
• eNews Archives 
• Conferences 
• Career Article Archive 
• Your Guide to Certification 
• How to Get a License 
• Future of Nursing Initiative

• Surveys 
• Puzzlers 

Purpose: To evaluate a perinatal team's clinical responses to the key components of the failure to rescue process.

Methods: This retrospective descriptive study involved a review of medical records and their accompanying fetal monitoring strips for 53 women who had a cesarean birth at term for a nonreassuring fetal heart rate pattern. The instrument was the Fetal Safety Failure to Rescue Process Tool based on the Agency for Healthcare Research and Quality's Failure to Rescue Patient Safety Indicator adapted for perinatal care processes. This tool measured four process measures: careful surveillance, timely identification of complications, appropriate interventions, and activating a team response. A mean total score was computed to determine quality of care based on all four of the process measures.

Results: The perinatal team's mean total score was 6.6 (SD = 1.0, range 3–8); the highest possible score was 8. The lowest score was found in the team's response with appropriate interventions (lateral positioning, intravenous fluid bolus, discontinuation of oxytocin, oxygen administration, amnioinfusion, administration of Terbutaline) based on fetal heart rate pattern. Interrater reliability of the tool was 90%.

Clinical Implications: This study provided information that may be useful in evaluating processes of care to ensure quality care for mothers and babies during labor. The findings formed the basis for implementing unit-specific educational programs, including (a) certification, continuing education, and documented competence in electronic fetal monitoring education to promote consistency in language and understanding of abnormal fetal heart rate patterns; (b) review of appropriate documentation of nonreassuring fetal heart rate patterns; (c) review and revision, if necessary, of established institutional standards and guidelines for appropriate interventions for nonreassuring fetal heart rate patterns; (d) changes in standardized forms to include times for notification of team members and improved communication, and (e) mechanisms for identifying system failures.

Failure to accurately assess maternal-fetal status, appropriately treat a nonreassuring fetal heart rate (FHR), correctly communicate maternal-fetal status to the physician/midwife, and respond to or initiate a chain of command are common sources of patient harm and obstetric malpractice claims (Dunn, Gies, & Peters, 2005; Joint Commission, 2004; Simpson & Knox, 2000). In efforts to identify and mitigate these sources of harm, patient safety structure, process, and outcome measures have been developed. Structure measures address the organizational context of care, such as existence of key clinical protocols for care known to be associated with patient risk (e.g., fetal monitoring). Process measures evaluate how care is provided (e.g., timely identification and treatment of nonreassuring FHR patterns) (Simpson, 2006). Outcome measures for complication rates, morbidity, and mortality are the most widely used evaluation method (Agency for Healthcare Research and Quality [AHRQ], 2004); however, they may not be the most sensitive indicators of quality when applied to perinatal care because general physiologic resilience of mothers and babies usually leads to good outcomes, even when the care provided is less than optimal (Simpson, 2006).

The concept of failure to rescue (failure to recognize, or act upon, early signs of distress) was first introduced in the 1990s as a measure of hospital performance in the care of the adult postsurgical patient population (Clarke & Aiken, 2003; Silber, Williams, Krakauer, & Schwartz, 1992). This concept has only recently been applied to obstetrics. Careful surveillance, timely identification of complications, appropriate interventions, and activating a team response, which are the four essential components of the rescue process, serve as the basis for application of failure to rescue in the perinatal setting (Simpson, 2006). Evaluating the rescue process during labor when the FHR pattern is nonreassuring can be used as an indicator of fetal safety (Simpson, 2006).

Although a reassuring FHR pattern is generally an accurate predictor of fetal well-being, a nonreassuring FHR pattern has a high false-positive rate (Simpson & Knox, 2000). It is estimated that even the most ominous FHR patterns are associated with, at most, a 50% to 65% incidence of neonatal neurologic depression (Martin, 1998). Because there is no accurate method via electronic fetal monitoring (EFM) to distinguish between fetuses with a nonreassuring FHR pattern that are hypoxemic or acidemic and those that are not, timely interventions should be initiated in response to all nonreassuring FHR patterns to potentially prevent adverse fetal outcomes. When a nonreassuring FHR pattern is identified, cesarean birth may not necessarily be the most appropriate first course of action (Hendrix & Chauhan, 2005). Intrauterine resuscitation measures performed in a timely fashion may correct the precipitating event (see Table 1) (American College of Obstetricians and Gynecologists [ACOG], 2005; Feinstein, Torgersen, & Atterbury, 2003; Simpson, 2007).

Table 1. Potential Intrauterine Resuscitation Measures Based on the Individual Clinical Situation

In a review of published literature to date on the association between FHR patterns and fetal acidemia, Parer, King, Flanders, Fox, and Kilpatrick (2006) found that except for a sudden profound bradycardia, newborn acidemia with decreasing variability in combination with decelerations developed over a period of time approaching 1 hour. Barring any acute obstetric emergency, such as a prolapsed umbilical cord, uterine rupture, or placental abruption, in the context of careful assessment of maternal-fetal status every 15 to 30 min based on risk status, these findings suggest that there is ample time to rescue a fetus in jeopardy. The 60-min timeframe for fetal deterioration noted by Parer et al. (2006) lends credence to the failure to rescue tool as a useful determinant of fetal safety because clinical management can be evaluated during this critical period.

Figure. No caption available.

The research question was: What is the perinatal team's total score as measured by the Fetal Safety Failure to Rescue Process Tool for clinical responses to nonreassuring FHR patterns during labor? This study used a nonexperimental descriptive design to retrospectively review all medical records and their accompanying EFM strips of women at term who had a cesarean birth for a nonreassuring FHR pattern during a 6-month period. The setting was a large metropolitan teaching hospital in the Southern United States. The hospital is a regional high-risk perinatal center with more than 3,000 births per year. Institutional Review Board approval was obtained from the study setting and Case Western Reserve University.

The sample was determined by review of the birth log for a 6-month period. Inclusion criteria were gestational age of >37 weeks based on the prenatal record, maternal age of >18 years and a cesarean birth for a nonreassuring FHR pattern as the indication by the obstetrician who performed the surgery. All 53 cases that met inclusion criteria were included in the study.

The Fetal Safety Failure to Rescue Process Tool based on the AHRQ (2004) patient safety indicator and adapted for perinatal care processes developed by Simpson (2006) was used in this study with permission. The tool consists of the four components of the rescue process (see Table 2). Expectations for careful monitoring were categorized according to risk factors as outlined in Guidelines for Perinatal Care (ACOG & American Academy of Pediatrics [AAP], 2007). Women with identified risk factors require more frequent monitoring. Risk factors or high-risk conditions have been defined by ACOG as fetal growth restriction, type 1 diabetes, pre-eclampsia, and labor induction/augmentation (AAP & ACOG, 2007; ACOG, 2005). No specific guidelines were found that defined “low-risk” patients; rather, low risk was determined by the absence of any discernable risk factors.

Table 2. Failure to Rescue Process Tool

Timely identification is scored according to accurate interpretation and clinical data based on agreement between medical record documentation and EFM strip. Timely identification must occur within the timeframe outlined in the careful monitoring component (e.g., every 15 or 30 min according to risk status). Based on the FHR pattern, appropriate interventions (see Table 1) and activation of the perinatal team members and response of team members were noted and scored. Content validity for the Fetal Safety Failure to Rescue Process Tool was achieved using review and feedback from 10 expert perinatal nurses (Simpson, 2006). Interrater reliability (ranging from 85%–92%) was established by perinatal nurses using the tool on three different random samples of medical records coded for nonreassuring FHR and the corresponding fetal monitoring tracing (Simpson, 2006).

Total scores recorded in this study were compared to the accepted standard; scores should meet 100% (total score of 8) of applicable criteria for responses to nonreassuring FHR patterns. Scores of 0, 1, or 2 were given for each of the four categories. A total score and subscore were assigned for each of the 53 cases. A mean total score was computed to determine quality of care based on all four of the process measures. The mean score provided a unit of measurement to quantify the quality of care. Subscores were evaluated for each of the four categories to determine areas in need of improvement.

Each record and EFM strip was carefully reviewed by the principal investigator and a research assistant with more than 15 years of intrapartum experience for the presence of the four rescue process components and each of their attributes. The National Institute of Child Health and Human Development [NICHD] (1997) definitions were used to evaluate characteristics of nonreassuring FHR patterns (e.g., baseline rate, variability, and presence or absence of accelerations and specific decelerations). The NICHD definitions do not include interpretative statements such as reassuring or nonreassuring. Common expectations for interventions for nonreassuring FHR patterns as noted by ACOG (2005) and the Association of Women's Health, Obstetric, and Neonatal Nurses (Feinstein et al., 2003) (see Table 1) were used to evaluate whether intrauterine resuscitation techniques for specific FHR patterns were initiated in a timely manner. These expected interventions are listed on the tool to assist with the evaluation process (Simpson, 2005). Interrater reliability between the principal investigator and the research assistant was 90%. The principal investigator and research assistant completed a 14-hr national advanced fetal monitoring course before conducting this study.

Mean maternal age was 26 years. Fifty-five percent of the women were nulliparous. The two most common diagnoses on admission were spontaneous labor (40%) and labor induction (43%). Mean total score for clinical responses to nonreassuring FHR patterns was 6.6 (SD = 1.0). Total scores for individual cases ranged from 3 to 8. Fifty-five percent (n = 29) of cases scored at least a 7 for clinical responses in each of the four categories. Seventeen percent (n = 9) of cases met 100% of criteria for clinical responses in each of the four categories. In 11.3% of cases, total scores were less than 6. The lowest subscore was found in the team's response with appropriate interventions. Frequencies and mean scores for each of the four components are outlined in Table 3.

Table 3. Components of Rescue in This Sample

All 53 cases met 100% for careful monitoring. The most common characteristic of nonreassuring patterns identified was recurrent variable decelerations, followed by late decelerations. Although all cases of nonreassuring FHR patterns were identified within the timeframe outlined by expectations for careful monitoring, 28% (n = 15) of cases did not meet criteria for accurate interpretation and agreement between medical record documentation and the EFM tracing (e.g., medical record documentation noted moderate variability, whereas EFM strip displayed minimal variability; medical record documentation noted variable decelerations, whereas EFM strip demonstrated prolonged decelerations).

Although all 53 cases ultimately resulted in nonreassuring FHR patterns, there were key areas of inconsistency in documentation during aspects of the labor course, including lack of documentation of variable or late decelerations and lack of intervention. There were also inconsistencies in abbreviations/codes used to identify specific aspects of the tracing. Abbreviations not found in the documentation legends (and not standard in the institution's documentation) and terms such as “irregular” variable and “early” variable, which are not a part of the NICHD (1997) definitions, were found in 11% of cases (n = 6). In 70% (n = 37) of cases, team members responded appropriately to nonreassuring FHR patterns through notification of the primary care provider, timely response for evaluation, and appropriate and timely response by members of the perinatal team for surgical intervention. However, failure of 30% (n = 16) of cases to meet activation of team response criteria may be due (primarily) to lack of documentation of response. There was no documentation in the nurse's notes as to the time decision for a cesarean birth was discussed with the patient in (n = 5 cases). In 10 cases, time between decision for cesarean birth and incision time exceeded 30 min.

Several of the criteria for activation of team response were unable to be determined because standard documentation in this institution does not include specific times for notification of certain team members, although they attended the birth. For example, specific times for notification of the anesthesia provider, scrub technician, first assistant, and neonatal team were unable to be determined in all 53 cases. However, documentation of their presence was indicated elsewhere in the medical record.

Figure. No caption available.

Statistical analyses were performed comparing total scores and intervention subscores with selected demographic and clinical factors. Using Pearson's correlation, significant results were found between increasing maternal age and appropriate interventions (r = 0.009). No significant differences were found when comparing the diagnosis of spontaneous labor versus labor induction and intervention subscores (p = .12).

This was a retrospective study using medical records and as such had the limitations associated with medical record review. As with any tool that uses medical records as the basis for evaluation of data, lack of documentation is a limitation and presented a challenge in determining whether appropriate care was provided but was not documented. With this specific tool, because some of the expected documentation was missing in the context of an urgent or emergent cesarean birth when clinicians were giving first priority to patient care rather than medical record documentation, this issue may have been more pronounced. The perinatal team consisted of an aggregate of caregivers, including attending physicians, resident physicians, and nurses, and individual analyses for these different providers was not possible and would otherwise be counterproductive. The purpose of the tool is to measure fetal safety based on the team's response. It also should be noted that these are the first reported findings using the Failure to Rescue Process tool and that this tool has not been tested extensively in a research context beyond the pilot study by Simpson (2006).

The practical significance of this study was to evaluate quality of perinatal care, specifically fetal safety during labor, and to direct mechanisms for improving care quality. The perinatal team's mean total score of 6.6 (highest possible = 8) suggested that clinical responses to nonreassuring FHR patterns met applicable criteria more than 50% but less than 100% of the time. Ideally, rescue process scores should be 8, indicating 100% adherence with rescue process measures; scores less than 8 are important indicators for areas of needed improvement. The lowest score was found in the team's response with appropriate interventions (lateral positioning, IV fluid bolus, discontinuation of oxytocin, oxygen administration, amnioinfusion, administration of Terbutaline) based on FHR pattern. Of all the interventions, lateral positioning was the most common intervention, occurring 89% of the time. Less commonly occurring interventions and contributors to low subscore and total scores were administration of IV fluids bolus and maternal oxygen therapy. It is possible that interventions were performed but not documented. It is also possible that other factors entered into the decision to choose a cesarean birth over intrauterine resuscitation and continuation of labor. Decision-making style and the experience of providers with patient outcomes may explain differences in operative intervention among obstetrical providers (Peipert, Hogan, Gifford, Chase, & Randall, 1999). Despite the lack of some documented interventions, it appears from the records that no adverse outcomes for the baby occurred (as per recorded Apgar scores and absence of admission to the neonatal intensive care unit).

Using the Failure to Rescue Process Tool presented some challenges, particularly in evaluating subscores for each of the four categories. Scores of 0 or 2 were easily determined based on available documentation. If none of the criteria was met within the category, a score of 0 was assigned; if all of the criteria were met, a score of 2 was assigned. Scores were assigned based on the preponderance of “yes” and “no” responses. No score was assigned for attributes in the category that were either “not applicable” or “unable to be determined” (UTBD). The challenge was in determining a score for criteria that were met (documented) some—but not all—of the time. With the exception of careful monitoring, most often scores were mixed. In the area of activation of team response, many of the components within that category were UTBD, which did not positively or negatively affect the score in that category. “Support by physician or nurse midwife for appropriate interventions by other members of the perinatal team” was subjective and/or not always able to be determined. Timely response by team members was often UTBD because specific times for their arrival were not documented, but their presence was documented elsewhere in the medical record. Criteria that were UTBD were not given a score. Had the criteria of UTBD been given a score of 0, the subscore for that category would have been affected more negatively.

Decision rules regarding scoring should be made when evaluating records with more than one nonreassuring event during the course of labor. Because the standard is to achieve 100% compliance with clinical responses, in this study more than one event of a nonreassuring FHR meant that the lowest score was assigned for that record. The results of this study suggested that ongoing review of clinical practices and policies and consistency in interpreting FHR patterns may close gaps in care and medical record documentation and promote fetal safety. The findings formed the basis for implementing unit-specific educational programs, which include (a) certification, continuing education, and documented competence in EFM education to promote consistency in language and understanding of abnormal FHR patterns; (b) review of appropriate documentation of nonreassuring FHR patterns; (c) review and revision, if necessary, of established institutional standards and guidelines for appropriate interventions for nonreassuring FHR patterns; (d) changes in standardized forms to include times for notification of team members and improved communication, and (e) mechanisms for identifying system failures.

Implementation and incorporation of suggested perinatal safety initiatives with a re-evaluation of rescue processes may be helpful in identifying effective measures for quality improvement during labor and birth. This study did not specifically explore education level of team members or time of day when cesarean births occurred, which may be useful in determining if there is a difference in identification and intervention of nonreassuring FHR patterns based on these variables. Future studies also may focus on other methods and markers for measuring patient safety in this population.

A systematic review of clinical processes is a key element in supporting a culture of safety in perinatal care (ACOG, 2003). This study found that evaluation of selected rescue process measures may be useful in identifying areas for quality improvement in perinatal care. With modifications to the rescue process tool, future research may also include a focus on measurement of other processes to evaluate safety in perinatal and neonatal care, which may be beneficial in promoting safer care for mothers and babies.

Agency for Healthcare Research and Quality. (2004). Patient safety indicators. Retrieved September 4, 2008, from www.qualityindicators.ahrq.gov [Context Link]

American Academy of Pediatrics and American College of Obstetricians and Gynecologists. (2007). Guidelines for perinatal care (6th ed.). Elk Grove Village, IL: Authors. [Context Link]

American College of Obstetricians and Gynecologists. (2005). Intrapartum fetal heart rate monitoring (Practice Bulletin No. 70). Washington, DC: Author. [Context Link]

American College of Obstetricians and Gynecologists Committee on Quality Improvement and Patient Safety. (2003). Patient safety in obstetrics and gynecology. Washington, DC: Author. [Context Link]

Clarke, S. P., & Aiken, L. H. (2003). Failure to rescue. American Journal of Nursing, 103, 42–47. [Context Link]

Dunn, P. A., Gies, M. L., & Peters, M. A. (2005). Perinatal litigation and related nursing issues. Clinics in Perinatology, 32, 277–290. [Context Link]

Feinstein, N., Torgersen, K., & Atterbury, J. (2003). Fetal heart rate monitoring: Principles and practices. Washington, DC: Association of Women's Health, Obstetric, and Neonatal Nurses. [Context Link]

Hendrix, N. W., & Chauhan, S. P. (2005). Cesarean delivery for nonreassuring fetal heart rate tracing. Obstetrics and Gynecology, 32, 273–286. [Context Link]

Joint Commission on Accreditation of Healthcare Organizations. (2004). Preventing infant death and injury during delivery. Sentinel Event Alert, (30), Oakbrook Terrace; Author. [Context Link]

Martin, C. B. (1998). Electronic fetal monitoring: A brief summary of its development, problems, and prospects. Journal of Obstetrics and Gynecology and Reproductive Biology, 78, 133–140. [Context Link]

National Institute of Child Health and Human Development. (1997). Electronic fetal heart rate monitoring: Research guideline for interpretation. Journal of Obstetric, Gynecological and Neonatal Nursing, 26, 635–640. [Context Link]

Parer, J. T., King, T., Flanders, S., Fox, M., & Kilpatrick, S. (2006). Fetal acidemia and electronic fetal heart rate patterns: Is there evidence of an association? Journal of Maternal-Fetal and Neonatal Medicine, 19, 289–294. [Context Link]

Peipert, J. F., Hogan, J. W., Gifford, D., Chase, E., & Randall, R. (1999). Strength of indication for cesarean delivery: Comparison of private physician versus resident service labor management. American Journal of Obstetrics and Gynecology, 181, 435–439. [Context Link]

Silber, J. H., Williams, S. V., Krakauer, H., & Schwartz, J. S. (1992). Hospital and patient characteristics associated with death after surgery. Medical Care, 30, 615–629. [Context Link]

Simpson, K. R. (2005). Failure to rescue implications for evaluating quality of care during labor and birth. Journal of Perinatal Neonatal Nursing, 19, 23–33. [Context Link]

Simpson, K.R. (2006). Measuring perinatal patient safety: Review of current methods. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 35, 432–442. [Context Link]

Simpson, K. R. (2007). Intrauterine resuscitation during labor: Review of current practice and supportive evidence. Journal of Midwifery and Women's Health, 52, 229–237. [Context Link]

Simpson, K. R., & Knox, G. E. (2000). Risk management and electronic fetal monitoring: Decreasing risk of adverse outcomes and liability exposure. Journal of Perinatal and Neonatal Nursing, 14, 40–52. [Context Link]

Key Words: Failure to rescue process; Nonreassuring fetal heart rate patterns; Perinatal patient safety; Process measurement