NDSU Department of Child Development and Family Science
North Dakota Department of Human Services
December 1999
James R. Beal, Ph.D.
Larry Burd, Ph.D.
Stephanie Dahl, M.D.Marilyn G. Klug, Ph.D.
Kimberly D. McCulloch, M.D.
From the University of North Dakota School of Medicine and Health Sciences
Dr. Burd is assistant professor in the Departments of Pediatrics and Neuroscience, the Director of the North Dakota Fetal Alcohol Syndrome Center, and Associate Director of the Child Evaluation and Treatment Program. Dr. Klug is in the Department of Pediatrics and Dr. Beal is in the Family Medicine. Dr. Dahl and Dr. McCulloch were medical students.
Corresponding Author and Reprint requests to:
Larry Burd, Ph.D. • Child Evaluation and Treatment Program
1300 S. Columbia Rd • Grand Forks, ND 58202
(701) 780 2477 • FAX: (701) 780-2599 • e-mail:laburd@mail.med.und.nodak.edu
Background: In rural North Dakota family physicians provide obstetrical care for a large portion of the population. This study evaluates cesarean section rates and indications for family physicians and obstetricians and examines fetal outcomes.
Methods: A retrospective review of computerized delivery data was performed. All cesarean deliveries (n = 317) at a North Dakota hospital were analyzed.
Results: Obstetricians, with a cesarean section rate of 21.3%, were above the Healthy People 2000 goal of a 15% cesarean section rate and above the 13.7% rate of their family practice colleagues. Patient demographics for the two groups were similar. Infant outcomes, as determined by birth weight and 1- and 5-minute apgar scores were similar, although patients receiving their prenatal care from family physicians had infants with significantly greater gestational age. Indications for cesarean section were very similar between the two groups. The category "multiple births" was the only significantly different indication.
Conclusion: Patients receiving prenatal care from family physicians can expect a lower cesarean section rate and satisfactory infant outcomes. Recommendations for lowering the cesarean section rate at rural hospitals can be made based on the outcomes of this study.
Keywords: Cesarean section; family physicians; obstetricians; neonatal outcomes
Cesarean sections are one of the most common surgical procedures performed in the United States. Approximately one million cesarean sections are performed annually and comprise greater than 45% of the total births in some areas (Deutchman, 1997; North Dakota Health Care Review, 1997). The overall rate of cesarean delivery in the United States was 21% in 1995 with a 15% primary cesarean section rate (Sachs, Kobelin, Castro, & Frigoletto, 1999). The American College of Obstetricians and Gynecologists reports that the most common indications for cesarean section are repeat cesarean section, dystocia or cephalopelvic disproportion, breech presentation, non-reassuring fetal heart tracing, failure to progress, failed attempted vaginal birth after previous cesarean (VBAC), and placenta previa (Quality Assessment and Improvement in Obstetrics and Gynecology, 1994).
Surgical delivery increases the risk of maternal morbidity and mortality compared to a normal vaginal delivery. The goal of Healthy People 2000, a project of the Department of Health and Human Services, is to decrease the US cesarean section rate to a 15% overall rate and 12% primary cesarean birth rate by the year 2000 (Sachs et al., 1999). This proposed reduction would decrease the maternal risks involved with a surgical procedure and is anticipated to save one $1 billion in health care costs per year (Deutchman et al., 1997). The proposed reduction in the number of Cesarean births seems to be a realistic goal. Efforts at some centers have lowered rates from 17.5% to 11.1 % with no observed change in neonatal or maternal outcome (Myers & Gleicher, 1993).
Many patients in rural communities seek obstetrical care from their local family physician. Multiple studies have reported that women who receive their prenatal care from family physicians have a significantly lower cesarean section rate than women cared for by obstetricians. It is often argued that obstetricians have a greater number of high-risk patients than family practice physicians, accounting for their increased cesarean section rates. In a recent study, in which women with a previous cesarean section or factors making them high risk were excluded, it was found that women managed by family practice physicians were less likely to undergo cesarean section (9.3 % vs. 16%). No differences in neonatal outcome between the physician groups were identified (Hueston, Applegat, Mansfield, King, & McClaflin, 1995). Another study compared the rates of cesarean sections between family physicians and obstetricians among patients with almost identical demographics and risk factors in a New York inner city neighborhood. The family practice clinic had a 7.6% cesarean section rate compared to a 14.4 % rate at the obstetrics and gynecology clinic. Outcomes were identical ("FPs Outperform," 1997).
This study was undertaken to determine the overall cesarean section rate at North Dakota's largest hospital (Meritcare). The second study objective was to determine if any difference existed in the rate of cesarean sections of women receiving their prenatal and labor care from family physicians versus obstetricians. Indications for performing cesarean section also were studied to determine if any differences existed between family physicians and obstetricians. Finally, infant outcomes based on 1- and 5- minute apgar scores, weight, and gestational age were evaluated and compared based on physician type.
A retrospective review of birth data for the year 1997 at a North Dakota hospital (Meritcare) was performed. All the birth data for the year were entered into a computer data bank by the obstetrics nurse administrator. The nurse administrator was able to obtain information on every delivery at the facility for the entire year. All patients who underwent cesarean section during 1997 were analyzed. Data selected for study included number of previous deliveries, indication for cesarean section, maternal age, marital status, gestational age of infant, birth weight, prenatal care physician, delivering physician, 1- and 5-minute apgar scores, and mother's smoking history. Maternal age, marital status, smoking history, and previous number of deliveries were variables used in t-test analysis to detect differences in patient demographics.
The total number of women who received care, number of infants delivered, and number of cesarean sections performed was determined. This information was then divided into two groups based on care provider status (i.e. family physician or obstetrician). Indications for performing cesarean section were studied to determine if any differences existed between family physicians and obstetricians. The hospital's existing computer program utilized seven categories of indications including: fetal presentation/breech, cephalopelvic disproportion (CPD), abnormal fetal heart rate pattern, placental complications, maternal condition, multiple gestation, and repeat/elective cesarean section. Fisher's Exact tests of significance were performed on this categorical data.
Finally, infant outcomes were evaluated. Infant birth weight (in grams), gestational age, and 1- and 5-minute apgar scores were compared in t-tests to detect differences in neonatal outcome in each physician group.
North Dakota has a total population of 666,000 people and approximately 8,600 births per year (North Dakota Department of Health, 1999). In 1997 1,565 women delivered 1,591 infants at the hospital study site (this was 18.5% of all North Dakota births for 1997). The obstetrician group had the majority of infants delivered (n = 1,372) (86.2%). Family physicians (n = 219) (13.8% ) delivered the remainder of the infants.
The patient groups were nearly identical and were not significantly different for any measured variable. Patient demographics are shown in table 1.
Table 1. Demographics of women who delivered by cesarean section in 1997 in a North Dakota
hospital.
-----------------------------------------------
Family
Practice Obstetricians p-value
-----------------------------------------------
Maternal age 29.2 28.7 0.628
Married 82% 77% 0.687
Smokers 12% 17% 0.305
Multiparous 51% 50% 0.661
-----------------------------------------------
Of 1,565 women who delivered infants, there were 318 cesarean sections. One patient was
dropped from this study as the prenatal care physician was not indicated in the records.
Thus, the cesarean section rate was 20.3% for the year for the hospital.
The cesarean section rate for obstetricians was 21.3 % and for family practice physicians 13.7%. These rates were significantly different with a Z-score of 5.86 (p < .001). The indications for cesarean sections for both family physicians and obstetricians are shown in table 2. The most common indication for both family physicians and obstetricians was repeat cesarean section. The only indication that was significantly different between the two groups was multiple gestations, all delivered by obstetricians.
Table 2. Indications for cesarean section among 1,565 deliveries in 1997 in a North Dakota
Hospital.
---------------------------------------------------
Family
Indication Practice Obstetrics p-value
---------------------------------------------------
Repeat 10 (36%) 90 (33%) 0.965
Fetal presentation 7 (25%) 54 (20%) 0.705
Fetal heart rate 6 (21%) 37 (14%) 0.199
Cephalopelvic- 5 (18%) 41 (15%) 0.441
disproportion
Multiple gestation 0 39 (14%) 0.016
Maternal condition 0 9 (3%) 0.965
No indication 2 (6.7%) 17 (5.9%) given
---------------------------------------------------
Infant outcomes in this study were addressed by examining apgar scores at 1 and 5 minutes,
birth weight, and number of weeks gestation and are listed in table 3. Infants of family
physician patients had a longer gestation than their counterparts (38.7 vs. 37.3 weeks).
Since gestational age and birth weight were significantly correlated (r = .819,
p < .001), birth weight was adjusted for gestational age. The adjusted mean weight
for family physician infants was 3048 grams, and the adjusted mean weight for obstetrician
infants was 3127 grams. These adjusted means were not significantly different (F = .595, P
= .441). Apgar scores were not significantly different at either 1 or 5 minutes.
Table 3. Infant outcomes in the 317 infants delivered by cesarean section.
--------------------------------------------------------
Family
Practice Obstetrics p-value
--------------------------------------------------------
Gestational age (weeks) 38.7 37.3 0.044
APGAR - 1 minute 7.07 7.54 0.212
APGAR - 5 minute 8.53 8.64 0.606
Birthweight (in grams) 3,313 3,099 0.224
--------------------------------------------------------
At this North Dakota hospital 1,565 women delivered 1591 infants in 1997. Of those deliveries 318 (20.3%) were Cesarean sections. Interestingly, the statewide cesarean section rate was 19% (North Dakota Health Care Review, 1997). The rate of cesarean sections in North Dakota, a largely rural state, is slightly below the national average of 21% (Sachs et al., 1999).
Among patients with similar demographics, obstetricians in this study had a cesarean section rate of 21.3%. This rate is well above the national recommendation of 15% encouraged by the Department of Health and Human Services in their Healthy People 2000 agenda. Family physicians, with a cesarean section rate of 13.7%, are slightly lower than the national recommendation and are significantly below the rates of their obstetrician partners. These lower rates of cesarean delivery among family physicians are associated with no significant difference in neonatal outcome as determined by 1- and 5-minute apgar scores or birth weight. In addition, the gestational age of the infants delivered by cesarean section is greater in family physician patients. Thus, in rural North Dakota, patients of family physicians are less likely to have a cesarean delivery, and if they do, the infant is likely to have experienced a longer gestation than infants delivered by an obstetrician.
The indications for cesarean delivery among the two physician groups are quite similar, with the only significant difference occurring in the multiple births category. This result may be expected, as patients self select or are referred to obstetricians due to the higher risk of multiple pregnancies. The category that encompasses the largest percentage of both obstetrician and family physician cesarean sections is the repeat/elective category. This finding is concurrent with national trends (Sachs et al., 1999).
The topic of repeat/ elective cesarean sections is under intense scrutiny. There has been, in recent years, a growing trend toward vaginal birth after cesarean (VBAC) deliveries. With a VBAC there is the increased risk of uterine rupture (1%), with potentially devastating maternal and neonatal outcomes (Sachs et al., 1999). However, approximately 70% of VBAC attempts are successful (Gregory, Henry, Gellens, Hobel, & Platt, 1994). It is important to realize that not all patients are candidates for a VBAC because absolute and relative contraindications exist (Gregory et al., 1994). Many patients refuse to undergo a trial of labor, thus elevating the number of repeat cesareans. The medical records in this study do not delineate between elective repeat cesarean sections, failed trial of labor or contraindications to trial of labor, so it is impossible to determine how these numbers could be improved at this hospital. Instituting distinct codes on medical records to differentiate these categories of "repeat" cesarean section would improve future studies (Gregory et al., 1994).
This study had a number of limitations. High risk obstetrical patients may choose obstetricians rather than family physicians. Patients who expect a repeat cesarean may also be more likely to chose an obstetrician who will follow them throughout the pregnancy and delivery. These two factors alone may account for a significant increase in the number of cesarean deliveries for obstetricians. As mentioned above, it is not known how many patients failed a trial of labor, had contraindications to VBAC or chose elective repeat cesarean section.
The "indications" categorization is also a limitation of the study. The hospital data base did not have an "other" category, so all indications were placed in seven categories. The maternal condition category, for example, was the category for patients with pre-eclampsia, herpes simplex etc., but it is not possible to differentiate exactly what conditions existed in the patients who are included in this category.
The small number of family physician patients in this study is also a possible limitation. In addition, we have no maternal outcome data but are aware that there were no maternal deaths during the study period. Finally, regional differences in cesarean section practices and rates in the state exist and this study may not be entirely representative of the state, especially considering the rural nature of most North Dakota hospitals.
Studies have consistently shown lower cesarean birth rates among family physicians compared to obstetricians even after high risk patients are eliminated or among patients with identical higher risk backgrounds (Hueston et al., 1995; FPs Outperform," 1997). It has been suggested that the cesarean section rate for family physicians would be higher if all family physicians were trained to do cesarean sections, implying that obstetricians perform more cesarean sections because they can do the procedure themselves. However, there is data to demonstrate that family practice physicians who were trained to perform cesarean sections had a significantly lower rate of cesarean sections than obstetricians (Deutchman, Conner, Gobbo, & FitzSimmons, 1995). Differences in reimbursement of a cesarean section versus a vaginal delivery is not likely to be an explanatory factor since insurance companies now reimburse physicians the same amount for both procedures (Sachs et al., 1999).
Clinically this is an important study because it highlights the fact that family physicians are managing their obstetrical patients well and are seeking assistance from their obstetrician colleagues when appropriate. This study brings into focus the need to decrease the cesarean section rates in rural North Dakota, as well as in urban areas where cesarean section rates are above the goal of 15%. The repeat/ elective category appears to be the easiest target in the effort to keep cesarean section rates low. As mentioned above, however, this category should ideally be broken into three categories: elective cesarean section (refuses VBAC), contraindications to VBAC, and failed VBAC (Gregory et al., 1994). The subgroup that could most likely be targeted is the elective group. Patient education is the key to persuading more patients to attempt a trial of VBAC. On a national level, a collective effort to decrease truly elective cesarean sections could reduce the cesarean section rate by 2.6% (Gregory et al., 1994).
Decreasing the number of repeat cesarean sections is an admirable goal. However, the adage "Once a cesarean, always a scar" is a reminder that preventing the "scar" and decreasing primary cesarean sections rates may be a better focus and ultimately the key to decreased cesarean section rates (Paul & Miller, 1995). In the United States, the indication dystocia seems to be proportionately higher than in other countries (Paul & Miller, 1995; Notzen, Cnattingius, Bergsjo, Cole, Taffel, Lorentz, & Dalveit, 1994). It is not unreasonable to aim for a decrease in the rates of cesarean section performed for the dystocia indication by 3%-4% (Paul & Miller, 1995). Improved labor induction methods and patient selection may significantly impact this indication (Paul & Miller, 1995).
Other studies have addressed strategies to decrease cesarean section rates. Mount Sinai Medical Center in Chicago reported a successful program to decrease cesarean section rates. Objective criteria for common indications of cesarean section were implemented. The program also included a stringent requirement for a second opinion on cesarean sections, a detailed review of all cesarean sections performed, and a review of practice parameters of individual physicians who performed the cesarean sections. The cesarean section rate fell from 17.5% to 11.5%. The outcomes for mother and infant were not adversely affected (Myers & Gleicher, 1988). We would recommend consideration of this strategy. This approach, as well as the others mentioned, will likely be useful in rural settings as well as urban settings.
Because surgical delivery carries increased risk of maternal morbidity, vaginal birth should be the goal, if feasible, in each delivery. However, physicians must not let hospital published statistics or economic concerns govern their decision making process. It should be obvious that maternal and fetal health should never be jeopardized secondary to concern about cesarean section statistics. When appropriate clinical indications are present, cesarean section should be rapidly performed to ensure a positive maternal and infant outcome.
References
ACOG. Quality Assessment and Improvement in Obstetrics and Gynecology. The American College of Obstetrics and Gynecology. Washington, DC. 1994.
Deutchman, M.E., Conner, P., Gobbo, R., FitzSimmons, R. (1995). Outcomes of Cesarean Sections Performed by Family Physicians and the Training They Received: A 15-Year Retrospective Study. Journal of American Board Family Practice, 8(2), 81-90.
Deutchman, M.E. (1997). Cesarean Section in Family Medicine. Position paper prepared for AAFP Commission on Quality and Scope of Practice.
"FPs Outperform Ob. Gyns. In Urban Maternity Care." (1997, October 15). Family Practice News, 27(20), 1-2.
Gregory, K.D., Henry, O.A., Gellens, A.J., Hobel, C.J., Platt, L.D. (1994). Repeat Cesareans: How Many Are Elective? Obstetrics and Gynecology, 84, 574-578.
Hueston, W.J., Applegat, J.A., Mansfield, C.J., King, D., McClaflin, R.R. (1995). Practice variations between family physicians and obstetricians in the management of low-risk pregnancies. Journal of Family Practice, 40(4), 345-351.
Myers, S. A, Gleicher, N. (1988). A successful program to lower cesarean-section rates. New England Journal of Medicine, 319, 1511-1516.
North Dakota Health Care Review. Cesarean Section Cooperative Project. Quality Quest. Summer 1997.
Notzon, F.C., Cnattingius, S., Bergsjo, P., Cole, S., Taffel, S., Lorentz, I. Daltveit, A.K. (1994). Cesarean section delivery in the 1980s: International comparison by indication. American Journal of Obstetrics and Gynecology, 170, 495-503.
Paul, R.H., Miller, D.A. (1995). Cesarean birth: How to reduce the rate. American Journal of Obstetrics and Gynecology, 172, 1903-1907.
Sachs, B.P., Kobelin, C,. Castro, M.A., Frigoletto, F. (1999). The Risks of Lowering the Cesarean-Delivery Rate. New England Journal of Medicine, 340, 54-57.
| Home |
NDSU Department of Child Development and Family Science
North Dakota Department of Human Services