SEARCH

SEARCH BY CITATION

Keywords:

  • failure to rescue;
  • nursing interventions;
  • nursing outcomes;
  • perinatal measurement;
  • standard terminology

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

Objective

To reach consensus for words used by nurses to document elements of a perinatal failure to rescue process measurement tool.

Design

Exploratory study with mixed methods.

Setting

Virtual. Participants were recruited through an online perinatal nursing discussion list and completed Internet-based electronic surveys.

Participants

Twenty-nine (29) labor and delivery nurses with at least 5 years of bedside nursing experience and additional expertise in fetal heart monitoring.

Methods

Modified Delphi study with three rounds. Qualitative methods were used to analyze study results for round one. Rounds 2 and 3 were analyzed quantitatively with a desired level of consensus of 75%.

Results

Twenty-seven of 29 participants completed all three study rounds. Seventy-six distinct data elements related to careful monitoring, timely identification of problems, appropriate intervention, and activation of a team response were defined by consensus. Because classification of maternal and fetal risk determines assessment frequency in labor, specific criteria for classifying a woman or fetus as high risk or low risk were included in the definitions for which participants reached consensus.

Conclusion

Achieving consensus about the actual words used to document perinatal nursing elements provides the foundation for incorporating paper-based process measurement tools, such as perinatal failure to rescue (P-FTR) into electronic documentation systems. Standardizing the words perinatal nurses use in documentation facilitates data retrieval and analysis and increases the usefulness of process measurement tools such as perinatal failure to rescue. Further, building process measurement tools into electronic systems may facilitate real-time rather than retrospective recognition of process deficiencies and improve perinatal outcomes.

Perinatal nurses interact with data every day, and with more health care organizations and perinatal care providers implementing electronic health record systems, the amount of data entering these systems is staggering. The national focus on health care quality and patient safety, the ability to correlate perinatal nursing care to maternal and infant outcomes, and reporting requirements related to perinatal specific core quality measures suggested by the Joint Commission (2013) all require data. Across nursing specialties, specific quality indicators such as patient falls and pressure ulcers are categorized as “nurse-sensitive” meaning they may be influenced by nursing care. In labor, the perinatal nurse must properly assess maternal and fetal status, identify potential problems, and appropriately intervene meaning patient outcomes for mother and infant are sensitive to perinatal nursing care. Analyzing the data captured and recorded by perinatal nurses and the rest of the care team during labor and delivery may help to demonstrate appropriateness of care or identify process deficiencies that need correction. However, the task of retrieving perinatal data, translating those data into meaningful information, and generating knowledge from that information is cumbersome and resource intensive.

Because more hospitals and health systems have implemented electronic documentation systems, there has been an increase in the capture of discreet data elements with the use of options such as predefined checklists and drop-down boxes. These discreet data fields may be used to collect data in a particular way for one unit, to capture the same data another way across health care systems, or even to capture data differently among separate units in the same system. An example of discreet data variability would be a data selection option for the perinatal nurse to document heavy bleeding. In another system, a perinatal nurse might have the choice to document bleeding with a second data field for heavy. Even with increased discreet data capture, much of perinatal nursing care is still captured on paper in narrative notes or into free text fields in electronic systems, a practice that makes data difficult to retrieve, contributes to data variability, and inhibits the ability of electronic systems to exchange information. All these factors support the need for retrievable, standardized perinatal data along with useful tools that assess clinical processes to quantify perinatal safety and quality outcomes.

Perinatal nurses assess maternal and fetal status, identify problems, and appropriately intervene, meaning outcomes for mother and infant are sensitive to perinatal nursing.

An example of a process assessment tool for perinatal nursing is Simpson's (2005) paper-based perinatal failure-to-rescue (P-FTR) process measurement instrument. This validated perinatal nursing process measurement tool is endorsed by the Agency for Healthcare Research and Quality (AHRQ) (Simpson, 2006). The tool was created to retrospectively assess perinatal medical records to determine (a) if mother and fetus were appropriately monitored during labor, (b) if the correct nursing interventions occurred based on monitoring findings, and (c) in the case of nonreassuring fetal or maternal status, if the appropriate team was mobilized to accomplish either a timely cesarean birth or the timely transfer of mother and/or infant to a higher level of care (the tool is available online as Figure S1). In a validation study for P-FTR, Beaulieu (2009) reviewed the records of more than 50 women who underwent cesarean births because of nonreassuring fetal status. Beaulieu's findings supported using P-FTR as a process measurement instrument, but she noted significant difficulty locating the data elements in the medical record. Either important documentation was missing all together, or data elements were documented in a nonstandard way, making it difficult to correlate the elements of the medical record with P-FTR processes (Beaulieu, 2009). Incorporating data elements from tools such as P-FTR into an electronic system may make it possible for the tool's data elements to be reliably and consistently retrieved and allow P-FTR to be used at the point of care to help guide perinatal nursing care decisions in real time. However, the first step in moving P-FTR or any paper-based tool into an electronic system is to decrease the data variability by standardizing the definitions (values) for P-FTR data elements.

Data standardization is a necessary step for meaningful data retrieval and data exchange. When perinatal electronic documentation systems were first introduced, standardized data were often overlooked in favor of customizable systems that could be adapted to an individual agency. Today, for any information system vendor to meet certification criteria as part of the meaningful use of health information technology (HIT), interoperability and the use of data standards must be considered (Department of Health and Human Services, 2012).

In terms of nursing visibility, when a nursing term is standardized, it can be assigned a code. Once coded, the term can be measured. Physicians and other care providers have been documenting medical diagnoses and corresponding treatment in a coded format since the late 1800s, when the International Statistical Classification of Diseases and Related Health Problems (ICD) began (Clark & Phil, 1999). The latest version of ICD (ICD-10) is being prepared for widespread implementation and consists of codes that are internationally recognized. The data elements for perinatal core measures, for example, are currently retrieved using ICD coded records. Similarly, coded nursing specific standard languages exist, such as the North American Nursing Diagnosis (NANDA) nomenclature, Nursing Intervention Classification (NIC), and Nursing Outcomes Classification (NOC).

Unlike ICD, none of the standard nursing terminologies has been widely implemented, perhaps because ICD is tied directly to billing and reimbursement, and nursing care has never been billed for or reimbursed in the same manner. Standard nursing languages are developed by consultation with expert sources, such as professional nursing organizations, and they incorporate the latest evidence, so using them increases the likelihood of adherence to practice standards (Rutherford, 2008). In all, the American Nurses Association (ANA; 2006) has approved 12 standard languages through its Nursing Practice Information Infrastructure Committee (CNPII). Now that standards and interoperability are required as part of the certification process for electronic health record systems and patient outcomes must be measured and reported, the use of standard terminologies is expected to increase.

The purpose of the study was to achieve consensus among experienced labor and delivery nurses for standard definitions for P-FTR elements using an internet-facilitated modified Delphi approach. Two research questions were addressed: What are the consensus definitions for elements of P-FTR? Can P-FTR elements be mapped to four ANA-recognized standard languages? In this article, I address only the first research question.

Methods

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

Study Design

This was an exploratory study with mixed methods. A modified-Delphi technique was used to obtain consensus definitions for the data elements in P-FTR. Delphi studies originated in the 1950s at the RAND Corporation (Miller & Salkind, 2002). Participants in Delphi studies progress through a series of consecutive iterations (rounds) about a subject. For each round, participants respond with their levels of agreement about the subject and are aware of the relationships of their responses to the responses of other participants. Study rounds continue until a defined degree of consensus is achieved. Strengths of the Delphi approach include (a) the ability to use content experts on the given subject, (b) the ability for participants to be distant from each other, (c) allowance for panel size variation, and (d) flexibility in study design (Okoli & Pawlawski, 2004). Weaknesses include the potential for response fatigue with multiple study rounds. Prior to recruitment, I assumed that participants would be busy professionals with limited time to commit to research rounds. Because McKenna, Hasson, and Smith (2002) noted that response fatigue is more likely in Delphi studies with more than two study rounds, I attempted to limit study rounds to as few as necessary. See Figure 1 for a visual description of the Delphi study process. The study was approved by the Vanderbilt University Institutional Review Board and was conducted during the first quarter of 2011.

image

Figure 1. Delphi study model. P-FTR = perinatal failure to rescue.

Download figure to PowerPoint

Setting

The research setting was virtual with online recruitment and informed consent of participants conducted over a 2-week period in January, 2011. Participants were recruited through a popular perinatal nursing discussion list with more than 800 members. A general call for participation went out to the entire discussion list describing the purpose of the study and guidelines for participation. To mitigate selection bias, discussion list members were encouraged to forward the request for participation to colleagues meeting inclusion criteria who were not listserv members. To maintain participant confidentiality, potential participants meeting inclusion criteria were instructed to contact me by e-mail privately, rather than to respond with their interest to the entire discussion list. Prospective participants were individually sent informed consent statements. To verify consent, participants were asked to respond by e-mail to me with a specifically worded statement that the participant understood the consent document, their questions had been answered, and they agreed to participate. Participants completed three rounds of Internet-based surveys (see Figure 1).

Participants

Participants were determined to be labor and delivery experts if they met inclusion criteria: at least 5 years of recent labor and delivery experience in a Level I, Level II, or Level III inpatient perinatal unit (American Academy of Pediatrics & American College of Obstetrics and Gynecology, 2007) and expertise in electronic fetal heart monitoring (FHM). Criteria to demonstrate FHM expertise included having completed an Association of Women's Health, Obstetric and Neonatal Nurses (AWHONN) Intermediate or Advanced FHM course within the previous 2 years, being a currently approved AWHONN FHM instructor, and/or current national certification in electronic fetal monitoring (C-EFM) through the Nurses Credentialing Center (NCC). Self-reported experience with current perinatal quality improvement efforts, such as those coordinated through the Institute for Healthcare Improvement (IHI), having responsibility for quality improvement focused medical record review, and/or having previous experience using P-FTR were all desired skills, but participants were not excluded if they did not have this additional experience.

Unlike other quantitative research in which a power analysis provides the basis for sample size, there are no defined panel size criteria in Delphi studies (Akins, Tolson, & Cole, 2005). Delphi panels have ranged from as few as four participants to more than 100 (Skulmoski, Hartman, & Krahn, 2007). Further, discerning the number of active members in an online discussion list is difficult. Therefore, for this study, an initial recruitment target was 20 participants. Twenty-nine labor and delivery nurses met inclusion criteria and consented to participate in all Delphi study rounds. Total recruitment time from listserv invitation to exceeding the initial targeted sample size was approximately 2 weeks. Demographic information for each participant was collected during the first study round only and was not associated with survey responses, allowing individual participant responses to remain anonymous. Twenty-seven of 29 participants completed demographic information.

Participants represented labor and delivery nurses with years of experience ranging from five to greater than 20 years. More than 90% (n = 25) of participants had greater than 10 years of experience, and more than 60% (n = 17) had more than 20. Electronic fetal heart monitoring expertise was demonstrated by completion of an AWHONN FHM intermediate (85%, n = 23) or advanced (63%, n = 17) course within 2 years of study participation; 70% (n = 19) were AWHONN fetal heart monitoring course instructors, and more than 50% (n = 14) maintained national C-EFM in electronic fetal monitoring. Participant work settings represented all three levels of perinatal care, as defined by the Guidelines for Perinatal Care (American Academy of Pediatrics & American College of Obstetrics and Gynecology, 2007). More than 60% (n = 17) represented Level III facilities, 15% (n = 4) represented Level II, and 22% (n = 6) represented Level I facilities. Annual birth volumes for participant facilities ranged from more than 4,000 per year (22%, n = 6), 2,501 to 4,000/year (19%, n = 5), 1,501 to 2500/year (22%, n = 6), 501 to 1,500/year (22%, n = 6), and 500 births per year or less (15%, n = 4).

Delphi Study Procedure

Study rounds followed the framework outlined in Figure 1 and were facilitated by an Internet-based survey tool that allowed anonymous distribution, anonymous response, as well as the sending of prescheduled reminders during study rounds. At the end of each round, the survey tool permitted aggregation of responses, basic statistical analysis, and the ability to export results into statistical software. There is little agreement in the literature as to consensus percentages, with published ranges from 51% to over 90% (Hasson, Keeney, & McKenna, 2000; Jones & Hunter, 1995; Kennedy, 2004). Consensus for this study was defined as 75% agreement or greater. To mitigate response fatigue, after at least 75% consensus was achieved for any one question during a study round, that question was not included in subsequent study rounds.

In the first iteration (round) of a standard Delphi study, participants are asked an open-ended question or to respond to a simple statement. The standard Delphi approach was adapted for this study in that participants received a copy of the P-FTR instrument as a separate attachment along with the link to the first Internet-based survey. For the first round, participants were asked what words they actually used for each P-FTR element in their own nursing documentation. Unlimited free-text fields were provided for Round 1 response, and participants were encouraged to include comments for further clarification. The Round 1 survey was open for 2 weeks. First round results required significant qualitative analysis, undertaken by using a word color coding approach suggested by Creswell and Zhang (2009). Once Round 1 responses were collated, summarized Round 1 responses became the Round 2 survey (see Figure 1). For Rounds 2 and 3, participants were permitted to choose only from the collated Round 1 responses; the free-text option was no longer available. Consensus building was facilitated by limiting the ability for free-text responses after Round 1. However, in case participants did not agree with my qualitative analysis from Round 1, and to increase the credibility of the qualitative analysis, participants were encouraged during Round 2 to add comments and/or to seek clarification. No requests for changes or clarification were received.

Rounds 2 and 3 surveys were each available to participants for 2 weeks. At the end of each round, I informed participants if another study round was necessary and if so when to expect the next round's survey. Survey content for Rounds 2 and 3 was identical. However, because consensus of at least 75% was reached for many P-FTR elements after Round 2, only the questions for which 75% consensus had not been achieved were included in the Round 3 survey. Thus, each survey was shorter, to reduce the potential for participant attrition due to response fatigue. Due to minimal variation in responses between Rounds 2 and 3, and to further minimize response fatigue, the study concluded after three rounds. Twenty-seven of 29 participants completed all three study rounds.

Results

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

A total of 78 distinct P-FTR elements was identified by study participants (n = 27). When broken out by category, careful monitoring/timely identification contained 58 elements; 25 of the 58 elements related to maternal or fetal risk (16 high risk and 9 low risk), 19 were FHM elements, and 14 related to uterine monitoring or labor assessment. Appropriate intervention contained 13 elements related to perinatal nursing intervention, and activation of a team had seven elements. In the few cases where elements did not have at least 75% consensus, all definitions are reported.

Careful Monitoring/Timely Identification

Process measurement during labor includes determining whether mother and fetus were monitored according to established parameters and determining whether care providers accurately identified normal (reassuring) and abnormal (nonreassuring) maternal and/or fetal characteristics. P-FTR measures the process of maternal and fetal monitoring and assessment. During Round 1, all 27 study participants (100%) agreed that the standard fetal monitoring terminology approved by the National Institute of Child Health and Human Development (NICHD) (Macones, Hankins, Spong, Hauth, & Moore, 2008) should be used to document fetal heart rate (FHR) characteristics. These characteristics include 19 distinct elements relating to (a) FHR baseline (beats per minute [BPM], bradycardia, tachycardia), variability (absent, minimal, moderate, marked), (b) accelerations (present, prolonged), (c) decelerations (early, variable, late, prolonged), and (d) impression (Category I, II, or III, normal, equivocal, abnormal). Interestingly, although 24 of 27 participants (88.9%) agreed that FHR categories should be documented, participants could not reach consensus as to whether it is appropriate for the nurse to document his or her interpretation of the FHR tracing using categories. Participants were asked how they would document nonreassuring (Category II or Category III) findings; only 51.9% (n = 14) indicated they would document the actual FHR Categories II or III, whereas 48.1% (n = 13) noted they would document the specific FHR characteristics (FHR, variability, type of deceleration) observed, leaving the category interpretation to another clinician. This percentage was essentially unchanged between Rounds 2 and 3.

There is institutional variation related to the frequency of assessment and documentation in labor.

Recommendations for the frequency of maternal and fetal assessment during labor vary based on maternal and fetal risk. Therefore, implicit within the concept of careful monitoring is the identification of mother and fetus as either high risk or low risk. Some situations, such as the administration of oxytocin for labor, place a woman and fetus at high risk. Other situations and conditions do not have clearly defined recommendations, so during Round 1, study participants listed diagnoses, characteristics, or conditions that they perceived (or their organization perceived) classified a woman or fetus as high risk. Similarly, participants listed characteristics for classification of mother or fetus as low risk. Tables 1 and 2 include participants’ perception of high risk characteristics for mother and fetus. Table 3 includes low risk maternal and fetal characteristics as perceived by study participants.

Table 1. Maternal high-risk characteristics (every 15 minute assessment in active labor)
Risk FactorPercent (n)
Note
  1. aCharacteristics that received less than 75% consensus. VBAC = vaginal birth after cesarean.

Previous cesarean (VBAC)96.3 (26)
Multiple gestation88.9 (24)
Abnormal vital signs81.5 (22)
Induction/augmentation of labor with oxytocin77.8 (21)
Less than 37 weeks gestation77.8 (21)
Maternal substance abusea63 (17)
Epidural anesthesia/analgesiaa51.9 (14)
Age greater than 35 yearsa18.5 (5)
Complication 
Hypertensive disorders (includes preeclampsia)100 (27)
Diabetes (controlled or uncontrolled, includes insulin drip)96.3 (26)
Cardiac disorders85.2 (23)
Vaginal bleeding (suspicious of or known previa or abruption)81.5 (22)
Renal disordersa70.4 (19)
Pulmonary disordersa66.7 (18)
Auto-immune disordersa63 (17)
Morbid obesitya55.6 (15)
Table 2. Fetal high-risk characteristics (every 15 minute assessment in active labor)
Fetal Risk FactorPercent (n)
Note
  1. aCharacteristics that received less than 75% consensus.

Less than 37 weeks gestation88.9 (24)
Growth restriction (IUGR) or small for gestational age (SGA)88.9 (24)
Multiple gestation81.5 (22)
Known fetal anomaliesa74.1 (20)
Oligo/polyhydramniosa74.1 (20)
Meconium stained fluida66.7 (18)
Prolonged rupture of membranes (>18 hrs)a66.7 (18)
Greater than 40 weeks gestationa33.3 (9)
Suspected macrosomiaa25.9 (7)
Table 3. Maternal and fetal low risk characteristics (every 30 minute assessment in active labor)
Maternal characteristicPercent (n)
Note
  1. aCharacteristics that received less than 75% consensus.

Without complications96.3 (26)
>37–40 weeks gestation (term)88.9 (24)
Vital signs within normal limits88.9 (24)
Negative medical history81.5 (22)
Received prenatal care77.8 (21)
Not receiving oxytocina63.0 (17)
Spontaneous labor and/or spontaneous rupture of membranesa48.1 (13)
Without an epidurala40.7 (11)
Fetal characteristic 
> 37–40 weeks (term) gestation85.2 (23)
Singleton85.2 (23)
Fetal growth within normal limits81.5 (22)
Vertex presentationa74.1 (20)
Mother without risk factorsa70.4 (19)
No known fetal anomaliesa70.4 (19)
Amniotic fluid levels within normal limitsa66.7 (18)
Clear fluid (if membranes ruptured)a59.3 (16)

An important assumption when assessing careful monitoring processes is that the woman is in active labor, because recommendations for monitoring assessment frequency apply to active labor and P-FTR is meant to assess monitoring during active labor. Various textbook definitions of active labor exist; all include the presence of uterine contractions, cervical dilatation, and cervical change. However, there is institutional variation related to the frequency of assessment and documentation, regardless of labor status. Therefore, consensus definitions for the elements of active labor were included as questions to study participants. Participants were asked how they document contractions in active labor. Twelve participants (44.4%) indicated they would document regular contractions, the same number documented regular contractions, moderate to strong, and three (11.1%) documented regular, painful contractions. No consensus was achieved for this element. By consensus of 85.2% (n = 23), participants agreed that cervical dilatation to classify a woman in active labor should be 4 centimeters or greater. Cervical change should be documented as present or progressive, according to 85.2% (n = 23) of participants.

Uterine monitoring characteristics included as part of careful monitoring in labor include contraction frequency in minutes and uterine resting tone (noted as documented by all participants) as well as contraction strength (mild, moderate, or strong), which 92.6% (n = 25) of participants routinely documented. Although 74.1% (n = 20) of participants indicated they consistently documented average contraction length (in seconds), this was slightly less than the 75% consensus desired for the study.

Appropriate Intervention

When a potential or real problem with mother or fetus during labor is identified by the nurse, it is very important that the nurse intervenes appropriately. Study participants identified 13 distinct elements associated with appropriate intervention. A common intervention for women receiving oxytocin for active labor is for the oxytocin to be discontinued if non-reassuring maternal or fetal status occurs. Twenty-one participants (77.8%) indicated they documented Pitocin discontinued (DC), compared to 22.2% (n = 6) of participants who would merely document Pitocin off. Interestingly, all study participants documented using the trade name Pitocin rather than the generic name oxytocin.

When maternal or fetal status indicates the need for supplemental oxygen administration, 88.9% (n = 24) of participants documented that the FHR pattern (which they may describe) remains unchanged, oxygen at 10 liters per minute (L/min). Only two participants indicated they would document only the administration of oxygen, without noting that there had been no improvement in FHR characteristics.

In the case of suspected decreased umbilical cord circulation, an amnioinfusion may be ordered. All 27 study participants indicated they would document this intervention as amnioinfusion at __milliliters (ml) per hour. If a tocolytic medication such as terbutaline, is necessary for the resolution of tachysystole or as an intervention for nonreassuring fetal status, 26 of 27 participants (96.3%) would document that terbutaline 0.25 milligrams was given subcutaneously (SQ) and include the location (of the injection). If nonreassuring fetal status occurs during the second stage of labor, modified pushing, such as instructing the mother to push with every second or third contraction or suggesting she not push at all to allow passive fetal descent, may be suggested as an intervention to improve FHR characteristics. In this case, 81.5% (n = 22) of participants would document pushing with every second (third) contraction due to __. Study participants would include pertinent FHR characteristics as part of this documentation. Only 11.1% of participants (n = 3) would document only that the patient was pushing every second or third contraction, without documenting FHR characteristics, and only two participants (7.4%) would document that they had encouraged(instructed) to push with every second (third) contraction.

Notably, there was one intervention for which study participants could not reach at least 75% consensus. With respect to patient laterality, when patient repositioning during labor is necessary, 63% (n = 17) of participants responded they would document that they turned the patient to her left (L) or right (R) side, whereas 37% (n = 10) indicated they would document that the patient was turned to the left or right lateral position.

Activation of a Team Response

Sometimes, appropriate interventions by the labor nurse and/or the health care provider do not improve maternal or fetal status. In this case, other members of the health care team need to be notified to accomplish a cesarean birth or the transfer of mother or infant to a higher level of care. P-FTR measures processes related to team activation and study participants identified seven distinct documentation elements related to team activation.

Several P-FTR elements related to processes regarding notification of the healthcare provider and requests for the provider to come to the bedside and subsequent responses. There was no definite consensus among study participants as to how notification of provider should be documented. More than 40% (n = 11) of participants indicated they documented MD/CNM notified of __ with the specifics of the situation included; more than 33% (n = 9) documented provider (name) notified of ___, and nearly 26% (n = 7) documented MD/CNM notified of situation__, background__, assessment__, recommendation/request__ (SBAR) format. Next the nurse must document the provider's response and arrival at the bedside. The majority of participants (55.6%, n = 15) documented this as the time the provider was notified and the time he or she arrived at the bedside, respectively. Although more than 50% represents a majority, this was less than the desired 75% consensus. Over one third (33.3%, n = 9) would document MD/CNM at bedside at__(time) and 11.1% (n = 3) would document only MD/CNM at bedside without a corresponding time.

It is important for the bedside perinatal nurse and perinatal nursing administrators to appreciate the need for data standardization.

In the event of a clinical disagreement as to the urgency of the request to come to the bedside, the nurse must document attempts to resolve any dispute; this is often referred to as initiating the chain of command, and 92.6% of participants (n = 25) indicated they would document ___(charge nurse, house supervisor, attending, etc) notified. Only two participants (7.4%) would actually document the words chain of command initiated. P-FTR also assesses for whether there is provider support for the nurse's appropriate interventions and communication among other members of the healthcare team. To capture these events in documentation, 81.5% (n = 22) of study participants indicated they would document MD/CNM at bedside, no new orders (new orders) received; only 18.5% (n = 5) would document MD/CNM agrees with assessment and interventions.

If a cesarean birth is indicated, the notification and mobilization of the surgical and neonatal resuscitation teams must be recorded. Although not quite the 75% consensus desired, 70.4% (n = 19) of study participants would document C/S called at ___(time). Nearly 30% (n = 8) would document OR team present.

Two P-FTR elements relate to team response relative to the newborn. If neonatal resuscitation (NRP) is required, 92.6% (n = 25) of participants would document the individual interventions (tactile stimulation, ventilation, chest compressions, etc.) performed. Only two participants (7.4%) would document neonatal resuscitation per NRP guidelines. If a neonatal transport is indicated, 70.4% (n = 19) would document transport (or transfer) team notified and approximately 30% (n = 8) would document transport (or transfer) team en route (or arrived) at ____ (time).

Discussion

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

Implications for Practice

For any specialty or discipline, standardizing data definitions is the first step toward incorporation of data elements into electronic systems. The results of this study suggest standard definitions for a significant number of perinatal nursing data elements. Appropriate monitoring of a woman and fetus during active labor as measured by the elements of P-FTR includes the documentation of FHM and uterine contraction characteristics. For FHM characteristics and contraction frequency, the NICHD-recommended, standard terminology already existed (Macones et al., 2008). Participants in this study affirmed by consensus that the NICHD terminology should be used in perinatal nursing documentation. Because participants represented a wide range of inpatient settings, study findings demonstrate that NICHD terminology is broadly used across a variety of inpatient units. However, participants could not agree on whether the FHR summary Categories I, II, and III should be documented by the nurse. Several comments by participants indicated that the summarization into categories represented making a “diagnosis” and therefore was beyond the nurse's scope of practice. Others disagreed. Such disagreement has implications for FHM research. In revising the FHM documentation guidelines in 2008, the NICHD suggested that grouping various FHM characteristics into categories may facilitate research by allowing clinicians and scientists to concentrate research efforts on those FHM characteristics that were, by definition, equivocal (Category II) (Macones et al. 2008). If Category II FHM strips cannot be reliably queried, such research may be delayed.

Acknowledging that recommendations regarding the frequency of maternal and fetal assessment in active labor vary based on risk, study participants achieved consensus on which diagnoses and conditions place a woman or her fetus at high risk, necessitating frequent assessment and which mothers and fetuses might be considered low risk, needing less frequent assessment. The frequency of assessment has implications for perinatal staffing. Assessment frequency for women in active labor considered to be high risk should be assessed every 15 minutes in the first stage of labor and every 5 minutes during the second stage (AWHONN, 2009). Low-risk women may be assessed less frequently. Consistently considering risk factors and monitoring mother and fetus accordingly would allow ongoing analysis of how much time the perinatal nurse spends on documentation and also perhaps free the nurse from documenting quite as frequently for those who do not need such frequent assessment. If process measurement tools such as P-FTR could be incorporated into electronic systems and capture FHM documentation in real-time rather than retrospectively, it might be possible to assess how much documentation happens at the point of care and how much may actually be entered later. Such information could also support staffing decisions.

Although it may seem simplistic to some clinicians, P-FTR is used to assess careful monitoring for active labor. Therefore, it was necessary to define active labor for purposes of process measurement using P-FTR. Study participants agreed that the definition of active labor includes cervical dilatation of 4 centimeters or greater, present or progressive cervical change, and the presence of regular uterine contractions. However, some nurse participants would also document contraction strength (moderate or strong) or include the word painful. Conversely, though study participants would not all document contraction strength based on whether a woman was in active labor, by consensus participants did agree that contraction strength was an important aspect of uterine activity assessment in careful monitoring, although not all would document contraction duration. Study findings suggest that despite the availability of evidence-based guidelines for what constitutes a complete maternal assessment for the laboring woman, variation exists in practice.

The fact that 29 experienced perinatal nurses quickly consented to participate in this modified Delphi study and 27 of them completed all three study rounds suggests that there is interest in the ability to effectively and efficiently use perinatal nursing data. Although not representative of everything the perinatal nurse does, the concepts included in P-FTR represent key elements of nursing care for the laboring woman and her unborn child, including careful monitoring of mother and fetus, timely identification of problems, appropriate interventions, and activation of the necessary team members to handle an emergent situation or to transfer a mother or infant in trouble. Consensus methods used in this study could be replicated for other perinatal nursing elements, including nurse staffing and perinatal acuity (Simpson, 2013). For example, participants identified characteristics that contribute to high risk, or low risk, maternal and fetal status. Some of these characteristics, such as maternal comorbid medical diagnoses, may not be routinely captured by perinatal documentation systems for purposes of assessing risk or contributing to the patient's overall acuity. Perinatal system vendors, clinicians, and researchers should work together to make sure all the data elements needed to make care, staffing, or process decisions are present in a standard format.

Although this study is foundational to perinatal nursing data, it contributes to work by researchers concerned with standardized data elements for nursing assessment processes in other areas, such as pressure ulcers (Bavaresco & Lucena-Ade, 2012), clinical pathways (Dykes, Currie, & Cimino, 2003), nursing clinical assessment (Dykes et al., 2009), evidenced-based guidelines (Dontje & Coenen, 2011), and nursing competencies (Meretoja & Koponen, 2012).

In perinatal care, there are several national initiatives underway concerned with the standardization of perinatal care related data definitions. AWHONN (2013) recently released a draft set of Perinatal Nursing Care Quality Measures for public comment; the final version of these measures may be available within the next year. The modified-Delphi approach used in this study could be used to contribute to the validation of these measures. Other initiatives include the standardization of data elements related to maternal, infant and perinatal death (Barfield, 2011), healthcare associated infections in the neonate (Bizarro, 2012), and standard obstetric data definitions for gestational age, induction of labor, and others (American College of Obstetricians and Gynecologists, 2012). This study contributes additional standard perinatal nursing definitions and supports the use of consensus study methods, such as the Delphi technique, in data standardization work. Other examples of the use of Delphi studies in perinatal nursing include lactation practices (Becker & Roberts, 2009), perinatal care outcome measures (Devane, Begley, Clarke, Horey, & O'Boyle, 2007), and midwifery duties (McKenna et al., 2002).

Limitations

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

Limitations of Delphi studies include the potential inability to recruit and retain a strong expert panel and the potential for response fatigue due to multiple study rounds. The majority of participants in this research had more than 20 years of labor and delivery experience, and all had significant FHR experience. Early Delphi study research noted that response fatigue is more likely in Delphi studies with more than two study rounds (Okiki & Pawlawski, 2004). Each study round survey was shortened wherever possible to mitigate response fatigue and, when variation in responses was unchanged after three rounds, data collection ceased.

The number of participants may also be a study limitation. Too few participants may not accurately reflect the subject area, whereas too many participants may make reaching consensus more difficult without numerous study rounds. Rather than conducting a power analysis to generate a suggested number of participants, I purposively sought participants who worked in varied inpatient perinatal settings. Study participants represented Level I, Level II, and Level III inpatient units.

In today's world, researchers must appreciate that participants recruited as subject experts are busy professionals with limited time to commit to research activities. Because participant involvement is necessary through all study rounds, clear expectations were included as part of the informed consent process, and rounds moved expediently to mitigate response fatigue and attrition. The potential for attrition was also mitigated by generating automated reminders through the online survey tool, modifying each survey so that it included only questions for which consensus had not already been achieved, and discontinuing additional study rounds when it was determined that further consensus was unlikely.

There are also limitations with internet facilitated research. The first and most obvious limitation is the requirement that study participants have Internet access. A potential complication to the fair treatment of participants in Internet research is the unintentional exclusion of subjects who cannot be identified through internet means. Discussion list participants were encouraged to forward the call for study participation to others outside the internet community. However, because consent was requested by e-mail, and surveys were electronic, Internet access was necessary. Although not every potential participant has Internet access, Internet participant recruitment allows for a broader recruitment reach across geographic boundaries as well as faster response time. Internet-facilitated survey research may also be less expensive than the cost of copying, mailing, and postage.

Study participants were recruited through an online perinatal nursing discussion list with more than 800 members; one might assume the large membership equates to a large pool of potential study participants. However, it was not known how many list members actively contribute to the list and how many are merely lurkers, subscribers to a discussion list who do not actively participate (Mendelson, 2007). Researchers should therefore use caution when setting recruitment goals through online lists, and this was another reason participants were recruited based on their experience and work setting. Also, authenticity in Internet-based research may be complicated (Im & Chee, 2002, 2006) and involves trust that the respondent is indeed the respondent. In this study, the threat to authenticity was mitigated by the specification of inclusion criteria for participants, along with the participant's reply and electronic signature, specifically requiring each participant to affirm that he or she met the inclusion criteria, understood the expectations relative to the study, and agreed to participate.

Conclusion

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

This study suggests standard definitions for a significant number of perinatal nursing care elements and provides a framework for eventual incorporation of perinatal process measurement tools, such as P-FTR, into electronic systems. The next step in this study was to address the second research question by mapping the standard definitions to selected standard nursing terminologies to determine which elements exist in which terminology and which are missing all together.

The findings may contribute to other current perinatal data standardization work underway by AWHONN, the American College of Obstetricians and Gynecologists, and others, and the consensus methods used may also be useful as standardization work intensifies. Wherever possible, correlating any standard perinatal data definition to a corresponding coded terminology is helpful and may be necessary for true data analysis and benchmarking.

Clinicians and others working in HIT appreciate the need for data standardization. However, it is important for the bedside perinatal nurse and perinatal nursing administration to also appreciate this need. Unless data standardization is widely deployed across systems and settings, the data collected by systems will continue to be just that, discreet data elements or free text data that must be queried individually, by hand. Such data are difficult to retrieve and may not contribute enough information from which to generate the knowledge to change practice and improve outcomes.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

Biography

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information
  • Catherine H. Ivory, PhD, RN-BC, is an assistant professor in the School of Nursing, Vanderbilt University, Nashville, TN.

Supporting Information

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusion
  8. REFERENCES
  9. Biography
  10. Supporting Information

Additional Supporting Information may be found in the online version of this article:

FilenameFormatSizeDescription
jogn12273-sup-0001-SuppMat.pdf151KTable 1: Evidence of Skin Breakdown of the Neonate during Use of Nasal Continuous Positive Airway Pressure (CPAP)

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.