Mode of Delivery Trends by Hospital Type: An 18-Year Comparison of a Military Hospital With University and Community Hospitals

ABSTRACT Objective: To determine the rate of mode of delivery over 18 years in a military teaching hospital as compared to university and community hospitals. Methods: From January 1992 to December 2009, we retrospectively calculated yearly rates for mode of delivery from a military hospital, two university hospitals, a community hospital in South Carolina and all community hospitals in Arkansas. Results: Over the 18-year period, 803,249 deliveries occurred from all hospitals. Overall the cesarean delivery rates have significantly increased across all practice models (22.7% + 0.9 versus 33.0% + 0.9, p = <0.001). The rate of increase has been greatest in university hospitals (21.8-37%) followed by community hospitals (26.7-32.9%) and the military hospital (19.6-29.2%). The rate of forceps-assisted deliveries has decreased dramatically across all practice models (11.6% + 1.3 versus 1.1% + 0.1, p = <0.001). The decline in forceps use was 6.4 to 1.1% in community hospital, 12.6 to1.4% in university hospitals, and 15.7 to 0.9% in military hospitals. Conclusions: The overall cesarean delivery rate has increased in all practice models but less so in the military. Forceps deliveries have dramatically decreased overall especially in the military hospital.

INTRODUCTION

Over the past 20 years, the percentage of cesarean deliveries (CDs) has increased for all women, of all ages, of all ethnic groups, and of all gestational ages.1,2 In the United States, the cesarean section rate dropped from 22.6% in 1991 to 20.7% in 1996 primarily because of an increased vaginal birth after cesarean (VBAC) rate of 33% between 1991 and 1996.3 By 2007, however, the overall CD rate increased to 31.1%. This rise in abdominal births was because of a dramatic fall in VBAC rates since 1996, with a subsequent increase in repeat CDs, as well as decreased operative vaginal deliveries and increased primary CDs.1 In addition, the current trend of more inductions of labor, fewer operative vaginal deliveries, and a rising primary CD rate has been noted since the mid- 1980s.4 A number of independent risk factors have been identified that have accompanied the rising CD rate including advancing maternal age, nulliparity, maternal obesity, cesar- ean birth on demand (without maternal or obstetric indica- tion), and fear of malpractice litigation.3,4,5,6 Despite the increase in CDs, the overall proportions, by indications for an abdominal birth have remained unchanged over time.7

Significant variations in CD rates have also been thought to reflect differences in practitioners' medical decision- making and variances in the available medical resources.8 Differences in CD rates have also been observed between patients with private insurance delivering in private hospital versus Medicaid patients delivering in public hospitals.9 In an evaluation for a managed health care plan evaluating deliveries in a U.S. military hospital, the primary and repeat cesarean section rates were found to be lower in military facilities compared with civilian health care facilities.10 There are a number of differences between military hospitals and civilian health care facilities that might account for this finding. In military teaching hospitals, all of the health care providers are salaried and not dependent on number of deliveries for financial remuneration, the physicians typically work shifts (no solo practice), work under the federal torts system for their malpractice coverage, and would be theoret- ically less influenced in managing and deciding on the appropriate mode of delivery, perhaps than those in other systems of care.1,4,6,8,9

There were two purposes for this investigation. First, we wished to determine the rate and mode of delivery over 18 years in a high-volume military teaching hospital. Second, we wanted to compare the modes of delivery between a military academic hospital and representative university- based academic practice and to that in community hospitals.

STUDY DESIGN

This study is a retrospective analysis of the deliveries occur- ring in January 1992 to December 2009 from the following institutions: (1) Naval Medical Center, Portsmouth Virginia (NMCP) (2) University of Arkansas for the Medical Sciences, Little Rock Arkansas (UAMS) (3) University of Mississippi Medical Center, Jackson Mississippi (UMC) (4) Spartanburg Regional Medical Center, Spartanburg South Carolina (SRMC), and (5) all community hospitals in the State of Arkansas. Our study was approved by the institutional review board as a ret- rospective study using the de-identifier data. The de-identified data from each source was the aggregate monthly total number of deliveries, as well as the total number of deliveries catego- rized into four mutually exclusive and exhaustive delivery types: spontaneous vaginal deliveries (SVDs), CDs, vacuum- assisted vaginal deliveries, and forceps-assisted vaginal deliv- eries. As aggregated monthly totals, the data did not contain any patient demographics, clinical information, or personal identifiers. All of the de-identified data was then combined into hospital categories as follows: community hospitals (Arkansas community hospitals + SRMC), university hospitals (UAMS + UMC), and military hospitals (NMCP).

Monthly aggregate data was combined into annual totals stratified by hospital type. Rates of delivery types were cal- culated as the proportion of total deliveries for a given year, and then converted into a percentage of all annual deliveries. After visual inspection of delivery trends, it was decided that the analysis of change in rates over time for each delivery type would be conducted by comparing the average rate over the first 6 years, 1992-1997, to the average rate over the last 6 years, 2004 -2009. Comparison of trends in delivery type across hospital type would then be accomplished through a difference-in-differences approach, where the difference in delivery rate, early to late, in one hospital type would be compared to the difference in delivery rate, early to late, in each of the other two hospital types. For all rate comparisons t-tests were used. All statistical analysis was conducted using Stata MP v12, (StataCorp., College Station, Texas).

RESULTS

Between January of 1992 and December of 2009 there were 58,733 deliveries at the military hospital, 100,842 from the two university institutions, and 643,674 from the community sites for 803,249 deliveries from all of the hospitals. Figures 1 to 4 describe the changes in mode of delivery over time within each individual hospital setting. In the military hospi- tal during the 1990s, the percentage of forceps deliveries declined, offset by an increase in the percentage of SVDs. During the 2000s an increase in the rate of CDs was concur- rent with a decrease in SVDs at the military hospital. The rate of vacuum delivery essentially remained unchanged (1-3%) over the entire time frame (Fig. 4). In the university medical centers during the 1990s the percentage of SVDs declined, because of a concomitant increase in the rate of CDs through the late 1990s. Since 2000 the rates of CDs have risen from approximately 30% to an average of 38% with an increase in SVDs from 48% to almost 60%. These figures in the past 10 years have been offset by a steep decline in operative delivery rates, both forceps and vacuum, at university institu- tions (Figs. 3 and 4). The community sites during the first epoch (1992-1999) saw a decrease in CDs and forceps deliv- eries (Figs. 1 and 3) although there was an increase during this first time period in SVDs and vacuum, and assisted deliv- eries (Figs. 2 and 4) in these community hospitals. However, during the period 2004-2009 the rate of CDs increased significantly, which as is expected, was concurrent with a decrease in the rate of SVDs. Since 2004, the rate of opera- tive vaginal deliveries (forceps and vacuum), have remained stable at 2 to 3% and 4%, respectively (Figs. 3 and 4).

Table I compares the delivery rate from the early period, 1992-1997, to the delivery rate for the late period, 2004-2009, for each delivery type across each practice model. There were no statistically significant changes in SVD rate, either overall, or across practice models. CD rates increased significantly overall (22.7% + 0.9 versus 33.0% + 0.9, p = <0.001). The rate of CD also increased significantly across all practice models. The rate of forceps-assisted deliveries decreased dra- matically overall (11.6% + 1.3 versus 1.1% + 0.1, p = <0.001), and it also decreased significantly across all practice models. The rate of vacuum-assisted delivery decreased significantly overall (3.5% + 0.4 versus 2.3% + 0.3, p = 0.020). The major- ity of this decline was the result of declines at University Hospitals (4.7% + 0.1 versus 1.6% + 0.2). Rates did not change significantly at either community hospitals or military hospitals between the periods studied.

Table II compares the differences in rate change for each delivery type, early to late, across practice models. There were no statistically significant differences in change in rate of SVD across practice models. CD rates increased signifi- cantly more at university hospitals compared to both commu- nity hospitals (9.1%, SE 1.8, p < 0.001) and Military Hospitals (5.6%, SE 1.8, p = 0.006). Forceps-assisted delivery rates decreased more at university hospitals than community hospitals (-6.0%, SE 0.9, p < 0.001), although the greatest difference was seen between military hospitals and community hospitals (-9.6%, SE 2.7, p = 0.002). The rate of vacuum- assisted delivery decreased more at university hospitals than either military hospitals (-3.2%, SE 0.8, p < 0.001) or commu- nity hospitals (-2.5%, SE 0.7, p = 0.003).

DISCUSSION

Similar to practices across the globe, our study has shown an increase in CD versus vaginal deliveries. All the hospitals analyzed in our study showed similar trends of increased CDs, decreased operative vaginal deliveries over the 18-year epoch. Over the entire period evaluated, the cesarean rate has increased in the three service categories from 23 to 33%. Over the 18 years in the military community and university hospital settings, the increase in cesarean births has averaged 50%, 100%, and 125% increases respectively. Although the overall increase in primarily and repeat CDs has been the primary reason for this change, it is partially explained by the decrease in operative vaginal deliveries, particularly in the military institutions and university practice settings in the last half of our study period. The reasons for the decline in operative vaginal deliveries among these two practice models are uncertain; however, the trend toward fewer operative vaginal deliveries has been noted over the last decade.11 In the university hospital settings, the decreased number of forceps deliveries may be because of fewer faculty who are comfortable in teaching forceps deliveries and fewer new facility wanting to resort to operative vaginal deliveries. These factors may also explain why the operative vaginal delivery rate has fallen within military installation as well as university institutions. Although vacuum-assisted deliveries have remained consistent and have slightly increased over the years in community hospitals, this may simply reflect the long-term stability of private practice versus academic and military medicine institutions (James A. Scardo, Personnel communication). In any case, it appears that the incidence of operative vaginal delivery may continue to decline in the future as we have seen with VBAC.6

Despite rising cesarean section rates in all areas, military hospitals still have less total cesarean section rate than com- munity hospitals and university institutions. The reason for a significant lower rate of cesarean section in the military insti- tutions compared to community hospitals and university practices is uncertain. Possible explanations are that the patient population has early access to prenatal care and since such care is free they are unlikely to miss prenatal visits because of financial constraint or transportation issues, which are common in community hospitals and university practices, respectively. In addition, among military practices the patients tend to be younger, generally healthier (particularly active duty subjects), and thus may have less comorbid conditions, which may account for an increase in abdominal deliveries. In com- munity hospitals where the cesarean section rate is highest, the physicians may proceed to an abdominal delivery more rapidly than in a military sitting. In university practice settings, the intensity of illness and presence of many comorbid factors also may increase the malpractice experience, whereas in military hospitals there is a system of adjudication for alleged malprac- tice it does not carry the same weight as in nonmilitary prac- tices for both doctors and hospitals. There are certainly unknown factors that may account for the increase in cesarean section rate, particularly in university hospitals compared to military institutions. However, the military facility in this study represents one of the busiest Obstetric Departments in the Department of Defense Hospital system, and consists of a large residency program (6 per year) with a full complement of subspecialties. As such, it is a regional referral center and the acuity of patients is more comparable to a university hospital than that at a community setting. Therefore, acuity alone can- not account for the military facility having the lowest cesarean section rate compared to the university hospital setting. Further research in these areas is recommended.

The primary strength of our study is the large dataset over 18 years representing three different practice models. Although, the sites were selected by investigator interest, rather than randomly designated, we can find no study that has previously compared delivery trends in the three settings of military versus university versus community hospitals. Another strength of the current investigation is that the same set of de-identified data was selected over all years from all sites. Thus, the numbers are large, and the consistency of the data is also present regardless of the site of practice. Finally, this is one of the few studies that include operative vaginal deliveries and further subdivide these births by forceps or vacuum in the three service categories over the entire 18-year epoch.

We acknowledge there are several limitations to our study. The findings may be biased because of being a retrospective analysis and the absence of clinical outcome data in each case prevented us from controlling for potential confounders. However, with such a large patient population (>800,000 deliveries) individual chart review was not possible. Simi- larly, the analysis of birth certificate data, which was consid- ered, was not undertaken because of frequent errors in the clinical portion of both certificate data, which is filled out frequently by nonmedical personnel. We also, recognize that we are comparing a single military medical treatment facility with isolated community and university facilities and that our findings may not be generalizable. However, it has been shown that at least in the southern region of the United States the data has been shown to be fairly consistent.12 Therefore, since all the hospitals in the three categories of practice settings are within the southern United States based on the regional partitions as defined by the National Center for Health Statistics,13 we feel that the conclusion of the study are as unbiased as possible as they are limited to one region.

In conclusion, our study has shown that, consistent with United States and worldwide data, the cesarean section rate is rising, although the rate of SVD and operative delivery are falling. For many of these reasons listed earlier, military institutions may have an advantage in having lower cesarean section rates and higher vaginal birth rates when compared to community hospitals and university teaching institutions. Evaluating modes of delivery over time in various hospital settings offers insight into practice patterns of these facilities. Future studies regarding patient demographics, indications for delivery, and maternal and fetal outcomes may further add to our understanding of this complex issue. If a region or hospital model shows a statistically significant improvement in these desired outcomes, an inquiry into how and why this is accomplished is warranted. The goals of each individual practitioner, department, hospital, or region remain finding the safest balance of spontaneous vaginal, operative vaginal, and CD for mother and infant.

ACKNOWLEDGMENT

We thank Donna Eastham, BA, University of Arkansas for Medical Sciences, for her editing skills and her help in submitting this manuscript.

REFERENCES

1. MacDorman MF, Menacker F, Declercq E: Cesarean birth in the United States: epidemiology, trends, and outcomes. Clin Perinatol 2008; 35: 293-307.

2. Getahun D, Strickland D, Lawrence JM, Fassett MJ, Koebnick C, Jacobsen SJ: Racial and ethnic disparities in the trends in primary cesarean delivery based on indications. Am J Obstet Gynecol 2009; 201: 422.e1-7.

3. Menacker F, Curtin SC: Trends in cesarean birth and vaginal birth after previous cesarean 1991 -99. Natl Vital Stat Rep 2001; 49: 1- 16.

4. O'Leary CM, de Klerk N, Keogh J, et al: Trends in mode of delivery during 1984-2003: can they be explained by pregnancy and delivery complications? BJOG 2007; 114: 855-64.

5. Kominiarek MA, Vanveldhuisen P, Hibbard J, et al: The maternal body mass index: a strong association with delivery route. Am J Obstet Gynecol 2010: 203; 264. e1-7.

6. Yang YT, Mello MM, Subramanian SV, Studdert DM: Relationship between malpractice litigation pressure and rates of cesarean section and vaginal birth after cesarean section. Med Care 2009; 47: 234 -42.

7. Choudhury AP, Dawson A: Trends in indications for cesarean sections over 7 years in a Welsh district general hospital. J Obstet Gynaecol 2009; 29: 714-7.

8. Hanley GE, Janssen PA, Greyson D: Regional variation in the cesarean delivery and assisted vaginal delivery rate. Obstet Gynecol 2010; 115: 1201-8.

9. Lipkind HS, Duzyj C, Rosenberg TJ, Funai EF, Chavkin W, Chiasson MA: Disparities in cesarean delivery rates and associated adverse neo- natal outcomes in New York City Hospitals. Obstet Gynecol 2009; 113: 1239-47.

10. Linton A, Peterson MR: Effect of managed care enrollment on primary and repeat cesarean delivery rates among U.S. Department of Defense health care beneficiaries in military and civilian hospitals worldwide, 1999-2002. Birth 2004; 31: 254-64.

11. Bofill JA, Rust OA, Perry KG, Roberts WE, Martin RW, Morrison JC: Operative vaginal delivery: a survey of fellows of ACOG. Obstet Gynecol 1996; 88(6): 1007-10.

12. Learman LA: Regional differences in operative obstetrics: a look to the south. Obstet Gynecol 1998; 92: 514 - 9.

13. National Center for Health Statistics. Regional Partitioning. Available at http://www.cdc.gov/nchs/data/series/sr_02/sr02_130.pdf; assessed November 6, 2013.

CDR Joshua D. Dahlke, MC USN*; CAPT Everett F. Magann, MC USN (Ret.)[dagger]; Tommy M. Bird, PhD[dagger];

LCDR Jesse Rohloff, MC USN[double dagger]; James A. Scardo, MD§; John C. Morrison, MD?

*Department of Obstetrics and Gynecology, Women and Infants Hospital of Rhode Island, 101 Dudley Street, Providence, RI 029051.

[dagger]Department of Obstetrics and Gynecology, University of Arkansas for the Medical Sciences, 4301 West Markham Street, Little Rock, AR 72205.

[double dagger]&lt;org>Department of Obstetrics and Gynecology, Naval Medical Center Portsmouth, 620 John Paul Jones Circle, Portsmouth, VA 23708.

§Department of Obstetrics and Gynecology, Spartanburg Regional Medical Center, 101 East Wood Street, Spartanburg, SC 29303.

||Department of Obstetrics and Gynecology, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216.

This article was selected for Poster Presentation at the Central Associa- tion of Obstetricians and Gynecologists 78th Annual Meeting, Nassau, Bahamas, October 26 -29, 2011.

The views expressed in this article are those of the author(s) and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, or the U.S. Government. This work was prepared as part of our official duties as a military service member. Title 17 U.S.C. 105 provides that "Copyright protection under this title is not available for any work of the United States Government." Title 17 U.S.C. 101 defines a U.S. Government work as a work prepared by a military service member or employee of the U.S. Government as part of that person's official duties.

doi: 10.7205/MILMED-D-13-00431