It is standard practice for students in clinical education programs such as nursing and midwifery to collect information on the patients whom they serve. Such information typically includes logs, case studies, care plans, and other methods of documenting the type and quality of the clinical experience. The faculty of a nurse-midwifery education program in the United States used an intrapartum minimum data set to examine the care provided by nurse-midwifery students. Project goals were to use student-collected and analyzed data to help students appreciate the value of examining their own developing practice, to explore factors that affect their practice, and to begin to use collected data to ask clinical questions.
The processes of care common to midwifery practice in the intrapartum period include nonpharmaceutical methods of induction and pain relief, alternative positions for delivery, food and oral fluids in labor, and a variety of methods of perinea! management (Oakley, Murtlaud, Mayes, Hayashi, Peterson, Rorie, & Anderson, 1995). Although institutions keep a variety of data, their ability to inform clinicians about midwifery care is limited. Birth certificates, hospital discharge abstracts, and labor and delivery records all provide basic demographic information as well as outcome data about the hospital stay: type of delivery, complications, neonatal Apgars and weight, dates of services, and length of stay. However, these tools do not effectively capture the practice style of midwives.
Midwives in the United States work with a population that is, in general, quite healthy. Outcome data such as maternal and neonatal morbidity and mortality and length of stay are not likely to be any different from those of physicians serving the same population. Rates of operative delivery are not significantly different from those of family practice physicians, another provider group that sees essentially healthy women and uses obstetrician consultants for cesarean deliveries. This does not indicate that midwifery practice is indistinguishable from that of physicians, but rather attests to the need to measure practices that are unique to midwifery.
With these concerns in mind, the faculty chose the Nurse-Midwifery Clinical Data Set (NMCDS) for its focus on nursemidwifery care in the intrapartum period. The NMCDS was developed, piloted, and refined by Albers and her colleagues at the University of New Mexico (Albers, Anderson, Cragin, Daniels, Hunter, Sealer, & Teaf, 1996). The group reduced the size of an earlier tool (Greener, 1991) by eliminating items that occurred only rarely in a large nurse-midwifery practice. After multiple iterations and multicenter clinical use trials, the result was a 1-page, double-sided data sheet. The current version of the form takes approximately 5 minutes to complete and three minutes to enter into a software program. The tool is available to members of the American College of Nurse-Midwives at no charge and is easily used with Epilnfo shareware, making it inexpensive and accessible to midwifery practices (ACNM web site).
The NMCDS is designed for use in midwifery practices and includes demographic data, risk factors, intrapartum care measures, interventions, and outcome information. Specific items that reflect the process of midwifery care include nonpharmacologic methods of pain relief, methods of permeai management, and maternal position for delivery. The ease and brevity of the tool allows the clinician to collect data even in a busy clinical situation immediately after the event. Ib facilitate comparison and linkages with national data-gathering efforts, standard demographic information is included.
At the time of the project (1996-1998), the faculty taught in the Midwifery Education Program at Education Programs Associates/San Jose State University. The program was a certificate nurse-midwifery education program located in Northern California. Students were recruited from across the country, and were diverse in background, experience, and educational preparation in nursing. Of the 25 students in this study, 10 had master's degrees, 8 had associate degrees, 5 had bachelor's degrees, and 2 had diplomas. Prior clinical preparation included women's health practitioner (14), family nurse practitioner (3), licensed midwife (3), adult nurse practitioner and physician assistant (2 each), and foreign educated midwife (1).
The program was a modified-distance learning program that minimized the length of time that students spent on campus. Students completed prerequisite coursework in their home communities and attended additional clinical courses on campus for short, intensive one-to-four month periods. They then completed their clinical practica in distance sites, typically in their own communities.
The program, one of six certifícate nurse-midwifery education programs in the U.S., was self-paced and course requirements were flexible to account for prior education in advanced practice nursing. The strong clinical emphasis made the program attractive to students who saw themselves primarily as cliniciane. Typical students had minimal experience in research and chose the program for its clinical orientation. In addition, many students expressed significant anxiety and concern about "doing research." Interestingly, students who had taken research coursework as part of their master's degrees in nursing stated similar concerns as those who did not have a research background.
The faculty wanted to encourage students to develop expertise in critically analyzing and using research as a lifelong professional habit. Recognizing the need to first reduce students' expressed anxiety, the faculty chose to approach research as an adjunct to becoming an expert midwife. Looking at research through a clinical lens as a way of strengthening practice made it possible to engage students in research topics. This approach fit with both the students' interest and with the clinical strengths of the program.
The faculty chose to use the NMCDS as a data collection tool for this project because of its ability to provide detailed information about intrapartum practice. When clinicians are surveyed about practice patterns or statistics, the information is typically "last remembered" and anecdotal data, both of which tend to be biased toward the unusual, high risk, problem centered, or adverse outcome. Recalled information is more subjective and likely to be additionally biased by the self-perception of what is appropriate (Basso, Olsen, Bisanti & Karmaus, 1997; DiFranceisco, McAuliffe, & Sikkema, 1998; Mannisto, Pietinen, Vìrtanen, Kataja, & Uusitupa, 1999; Rookus & van Leeuwen, 1996).
Although knowledge about clinician values and perception of practice style is important, it does not provide a clear and accurate picture of what actually occurs in a clinical setting. Collection of intrapartum data immediately after the event offers a "real time" snapshot of style of practice, demographics of the clients, risk factors, and use of interventions.
The faculty anticipated that as students collected data on their own practices, they would begin to recognize differences between what they subjectively assumed they were doing and what began to actually appear on the data sheets. It was hoped that over time students would become aware of heretofore unseen patterns as well as recognize their own transition from preceptor-directed practice at the beginning of the clinical experience to more self-directed practice toward the end. In addition, an examination of practice patterns by the class as a group would invariably bring about discussion of different ways to do things, successes and failures in management, and ideas about how to institute change in the workplace.
With these goals in mind, the investigators initiated the use of the NMCDS for intrapartum care. Following discussions about evidence-based practice, the faculty introduced the use of the data set. The following questions were asked to encourage a closer examination of practice:
1. What do you do?
2. Why do you do it this way?
3. Who/what influences your practice?
-institutional guidelines or rituals
-the particular woman's needs in labor
-personal preference of the student
4. How does your practice compare with that of your colleagues? Why is this so?
5. How can this data help you to answer clinical questions?
The students were oriented to the NMCDS during the intrapartum course of the education program. During the intrapartum experience each student was expected to complete a data sheet for every other delivery throughout the preceptorship. Every other delivery was selected for data collection to obtain sufficient information for each student to look at her own practice but avoid overwhelming quantities of paperwork. All data sheets were to be sent to the program office for the faculty to review. Because each student had at least 40 births, this offered an opportunity to observe changes in practice as they occurred with increasing clinical experience and comfort level. Although accuracy and thoroughness were emphasized, no efforts were made to crosscheck student data sheets with patient records. The project was designed to use the data sheets as a teaching tool rather than an investigation of the care itself.
Patient privacy was protected by strict adherence to institutional confidentiality agreements. Confidentiality issues were addressed by having each student use her name instead of the patient name. Identifying information was not collected on the form, and faculty kept the original data sheets and lists linking students with specific clinical sites in a secure file.
As each student completed the intrapartum clinical experience, the faculty collated the data sheets and reviewed data for frequencies and distribution of interventions. The brief analysis gave the faculty an idea of the experience of each class so they could facilitate group discussions about the project.
The last week of the education program was an on-campus integration week where the students gathered as a class for the last time. As one of the assignments students provided basic descriptive analysis of their collected NMCDS data: simple percentages and means for each of the items. Even such basic analysis was sufficient for the students to begin to see patterns of patient characteristics and care in their work.
This information was collected from each student and copies were made for the class members. In a class facilitated by the investigatore, the same initial questions were asked as students began to examine and compare their practices using the data they had collected. This proved to be an opportunity for lively discussion of clinical practice and its basis in evidence.
Limitations and Leseone
As with many clinical experiences, the results of the project differed somewhat from the original plan the faculty had developed. Problems noted included process, data entry error, and data omission.
Errors in Process. About half of the students did not turn in their data sheets in a timely way, eliminating the opportunity for faculty to correct errors in data entry. Distance from campus, the heavy time commitment of the intrapartum practicum, and the necessity of juggling school, work, and family obligations seem to have contributed to this problem. Approximately half of the students completed data sheets for every delivery instead of every other delivery, which weighted the information available to analyze. Although this could have been corrected by eliminating every other form to maintain congruence with the other students, all forms were retained. Ae previously stated, the purpose of the project was to involve students in examination of their own practices; thus errors in the collection process became part of the discussion and learning opportunity.
Errors in Data Entry. The faculty recognized that it would not be possible to crosscheck data entry for this project. This was not thought to be a significant concern because the purpose of the project was to introduce students to the process of data collection and analysis. Although virtually every student completed the data sheets, many were incorrectly completed. The most common problem was simply checking all answers that applied instead of entering a 1-3 ranking when euch was requested.
Errors in Omission. Students also tended to omit items, most commonly date of birth, years of education completed, height, weight, weight gain in pregnancy, and length of labor. Several students noted that these items were not on the intrapartum chart, but had to be found on the prenatal chart and in the case of weight gain, needed to be calculated. In large and busy practices the prenatal charte were sometimes not available at the time of delivery, or were filed in a different place than the intrapartum record.
Over the three years that the faculty conducted the project (1996-1998), it became obvious that it was necessary to provide clear and detailed information to the students about data collection and use of the tool. The information collected during the final year of the project was of a higher quality than that of the initial year.
As with the problems, rewards were both expected and unexpected. As the faculty had hoped, the analysis of the data and the sharing of information provided rich discussions about differences in practice. Although the students shared a common theoretical background from their coursework, they quickly noticed how diverse their actual practices were. Differences were particularly interesting when studente had completed preceptorships in different institutions within the same health care system.
As an example, several students completed preceptorships in four different sites that belong to the same large HMO system. Although the practice guidelines for each eite within the system were identical, actual clinical practice varied widely in the use of ambulation in labor, monitoring practices, and the use of both pharmacologie and nonpharmacologic methods of pain relief. Studente realized that there are factors other than practice guidelines that influence day-to-day practice of midwifery, and the resulting differences can be very significant. These "ah-ha" experiences furthered discussion about the often dramatic differences in both practice and the perception of practice. Studente were encouraged by their peers to provide rationale for practice styles and to examine practices for their theoretical basis.
Unexpectedly, the abundance of data provided the campus-based faculty with a more complete picture of the clinical sites- a serendipitous educational benefit. As with any off-campus clinical site, practice information typically comes from observation of students' on-site visits, student logs, conversations with preceptors, and feedback from students. Each is valuable, but offers "snapshot" date. Loge provide minimal information about the process of care, site visits are subject to the vagaries of a particular day and time, and feedback tends to focus on problems and concerns and is by definition extremely subjective.
The NMCDS data, however, gave a detailed picture of nurse-midwifery practice. Client demographics alone provided important information about how best to focus teaching efforts to the specific population and culture. Because the focus of the NMCDS is the process of midwifery care, the faculty were able to understand more clearly the experience that students were likely to have. This enabled better site selection and matching for each student.
SUMMARY AND CONCLUSIONS
Research is an intimidating prospect for many studente. This brief and fairly simple foray into data collection and analysis did not convince all students to embrace research. It did, however, entice many to consider how to incorporate data collection into their own beginning practices. Students who had stated their aversion to research found themselves arguing for the necessity of date collection to provide evidence for practice changes. There were also students who found the experience yet another busywork paper trail invented by the faculty. They resented the time and effort required for completion of another form during their busy clinical experiences. Other students remained adamantly opposed to considering any changes in what they saw as "appropriate" practice despite evidence to the contrary.
An ongoing goal of education is to facilitate lifelong learning and investigation into clinical questions. The faculty plan to survey the graduates to ascertain what data collection efforts, if any, they are currently using. The educational program described in this project closed in December 1998 because of funding restrictions. One of the faculty in this project (Carr) and current colleagues are continuing to use the NMCDS to encourage students to examine and analyze their own practices. An additional use of the data set is for student-initiated research projects that focus on specific clinical questions. One faculty member was part of an effort to institute a similar project in a master'sbased nurse-midwifery education program. That database has currently provided data for two master's degree projects.
The uses of the NMCDS are not limited to graduate students or advanced practice clinicians. The tool is being used in an undergraduate maternity nursing clinical course to assist studente in seeing differences in practice and the possible reasons for those differences.
The Division of Research of the ACNM has sponsored the development of similar date collection tools for both ambulatory prenatal and well-woman care. The wellwoman tool is currently in the pilot stage (1999), and will be refined and further tested this year. The uee of these data collection tools by advanced practice nurses would provide a wealth of data for both policy setting and clinical research.
Educational programs frequently combine coursework for various advanced practice nursing pathways, and nursemidwifery and nurse practitioner students would benefit from jointly focusing on practice patterns and participating in the analysis of student-collected data. The standardized tools for intrapartum, wellwoman care, and antepartum care will be available to all of the 50 ACNM education programs as well as to practicing clinicians.
- ACNM. (Feb. 15, 2001). [On-lineJ Available at: http://www.acnm.org.
- Albere, L.L., Ander aon. D., Cragin, L., Daniele, S.M., Hunter, C., Sedler, K.D., & leaf, D. (1996). Factors related to perinea! trauma in childbirth. Journal of Nurse-Midwifery, 41, 269-276.
- American College of Nuree-Mid wives. (2000). Resources and bibliography: Data collection. The Nurse-Midwifery Clinical Data Set [Online]. Available: http://www.midwife.org/profi' data.htm
- Basso, O., Olsen, J., Bisanti, L., & Karmaus, W. (1997). The performance of several indicators in detecting recall bias. European study group on infertility and subfecundity. Epidemiology, 8(3), 269-274.
- DiFranceisco, W, McAuIiSe, T-L,, & Sikkema, K.J. ( 1998). Influences of survey instrument format and social desirability on the reliability of self-reported high risk sexual behavior. AIDS and Behavior, 2(4), 329-337.
- Greener, D. (1991). Development and validation of the nurse-midwifery clinical data set. Journal of Nurse-Midwifery, 36, 174-180.
- Mannieto, S., Pietinen, P., Vlrtanen, M., Kataja, V., & Uusitupa, M. (1999). Diet and the risk of breast cancer in a case-control study: Does the threat of disease have an influence on recall bias? Journal of Clinical Epidemiology, 52 (5), 429-439.
- Oakley, D-, Murtland, T., Mayes, F., Hayashi, R., Peterson, BA., Rorie, C., & Anderson, F. (1995). Processes of care: Comparisone of certified mirse-midwives and obstetricians. Journal of Nurse-Midwifery, 40, 399-409.
- Rookus, M.A, & van Leeuwen, F.E. (1996). Induced abortion and risk for breast cancer: Reporting (recall) bias in a Dutch case-control study. Journal of the National Cancer Institute, 88(23), 1759-1764.