Primary cesarean delivery rates have increased in Illinois since 1997,1–6 although the reasons for this increase remain poorly understood.1,7 One contributing factor for this rise may be the change in the characteristics of the gravid population. For example, women 35 years or older account for a greater proportion of deliveries and are more likely to have a cesarean delivery than younger women.8 Yet, there also is evidence that factors unrelated to the patient population, such as changing physician practice patterns,9 influence the cesarean delivery rate. One factor that has been theorized to affect physician practice patterns and decision-making is the medical–legal environment.7,10 Physicians themselves acknowledge that their decision making is influenced by this environment11 and have stated that medical–legal pressures influence them to perform tests and procedures that are not clearly medically necessary.12 Indeed, obstetricians have claimed to be particularly affected by the medical–legal climate, a finding that is not surprising given the relatively large monetary settlements that have been awarded in birth injury malpractice cases.13
These findings, together with evidence that cesarean delivery rates may be affected by physician practice style, suggest that the perceived medical–legal pressures experienced by obstetricians may be contributing to the rising cesarean delivery rate. However, evidence of this association has not been consistently demonstrated. Baldwin and colleagues14 were unable to show an association between malpractice pressures and cesarean delivery. Although some investigators have found a positive relationship between primary cesarean delivery and medical malpractice pressures, these investigators did not control for changes in the characteristics of the population.10 Consequently, it remains unknown whether this association was due to confounding bias. Other investigators have discerned an association and controlled for potentially confounding factors but performed a cross-sectional analysis only, raising the possibility that higher professional liability insurance rates may coexist with, but not precede, higher cesarean delivery rates.15,16 Therefore, the aim of this study was to estimate whether the increase in the rate of primary cesarean delivery is independently associated with the increase in medical professional liability premiums.
MATERIALS AND METHODS
Using birth certificate data from the National Center for Health Statistics, we identified all deliveries occurring in Illinois from 1998 through 2003. Women who delivered a singleton between 37 0/7 and 44 6/7 weeks of gestation and had no prior cesarean delivery were eligible for the study. Women who were not candidates for a vaginal delivery due to a placenta previa or who had an unknown route of delivery were excluded from the analysis.
All maternal and infant data were stratified by maternal parity. The county-level primary cesarean delivery rate was calculated as the number of primary cesarean deliveries per 1,000 eligible women at 37 0/7 weeks of gestation. The cesarean delivery rate was determined for each year of the study and differences in the mean county-level cesarean delivery rates were evaluated using analysis of variance (ANOVA). Annual professional liability insurance premiums for obstetrician–gynecologists in Illinois were determined for each county in Illinois. These charges were derived from the ISMIE Mutual Insurance Company, because this organization accounts for the majority of medical liability policies for Illinois obstetricians (Alan Allphin, VP Underwriting, ISMIE Mutual Insurance Company Mutual Insurance Company, personal communication). Premiums were then adjusted to 2004 dollars using the medical care component of the Consumer Price Index. Annual differences in the mean county-level insurance premiums were evaluated using ANOVA.
The county-level characteristics of the obstetric population were then determined. To ensure confidentiality, the Centers for Disease Control and Prevention identified county of residence only for women residing in counties with a population greater than 100,000 residents. Women from counties with fewer than 100,000 residents (accounting for 10.3% of the eligible births in Illinois in 2000) were placed into a single group and had their “county-level” characteristics aggregated. The professional liability insurance premium for this group was determined as an average of these counties' premiums weighted by their mean population of residents in 2000.17 Univariable analyses were performed using the Student t test and χ2 analysis to determine the patient characteristics that were significantly associated with cesarean delivery.
The independent association between county-level primary cesarean delivery rates and obstetric professional liability premiums was evaluated using linear regression models. We hypothesized that physician behavior would be most influenced by premiums to which they had already been exposed. Thus, in these models, premiums for the year before the birth were used as an independent variable. Given the possibility that both cesarean delivery rates and premiums may be related to changes in the maternal characteristics of the population, analyses were further adjusted for the presence of at least one obstetric risk factor for cesarean delivery. Other demographic factors were evaluated as independent variables in the regressions as well, and any variable that changed the estimated effect of the insurance premiums on the cesarean delivery rate by at least 20% was included in the final regression model. All tests were two-tailed, and P<.05 was used to define statistical significance. Statistical analyses were performed using Stata 9.2 (Stata, Inc., College Station, TX). The Children's Memorial Research Center's Institutional Review Board approved this study.
From 1998 through 2003, 1,046,317 women delivered singletons in Illinois. Of these women, 3,321 (0.3%) had a placental previa, 113,557 (10.9%) delivered preterm, 10,390 (1%) were reported to have delivered after the 44th week of gestation, 4,300 (0.4%) did not have their route of delivery recorded, and 97,228 (9.3%) had a previous cesarean delivery. The 817,521 remaining women (78.1%) were included in the analysis.
The mean (±standard deviation) county-level rate of primary cesarean delivery was 141±83 per 1,000 eligible women, and this rate rose significantly from 126±78 per 1,000 in 1998 to 163±93 per 1,000 eligible women in 2003 (P<.001). Figure 1 depicts the cesarean delivery rate stratified by maternal parity. For nulliparous women, the mean primary cesarean delivery rate was 218 per 1,000 eligible women, increasing from 199 per 1,000 in 1998 to 249 per 1,000 in 2003 (ANOVA, P<.001). Multiparous women, whose mean primary cesarean delivery rate was 63 per 1,000 women, also exhibited a significant increase in the rate of primary cesarean delivery (53 per 1,000 eligible women in 1998 to 72 per 1,000 in 2003 (ANOVA, P=.001)).
As shown in Figure 2, a significant increase was also found when examining the change in professional liability insurance premiums beginning in 1997, the year before the study period. The mean annual county premium, which was $60,766±$14,854 in 1997, increased to $83,167±$18,408 in 2002 (P<.001).
Compared with women delivering vaginally, women delivering by primary cesarean were slightly older and more likely to be nulliparous, to have completed high school, and to have infants with higher birth weights (Table 1). Multiple antepartum medical risk factors were also seen more commonly among those women who underwent cesarean delivery. Specific factors significantly associated with cesarean delivery were maternal pulmonary disease, diabetes mellitus, genital herpes, amniotic fluid abnormalities, hypertensive disorders of pregnancy, and uterine bleeding (Table 1).
Univariable analysis of nulliparous women showed that for each $10,000 increase in the previous year's county-level professional liability premium, the primary cesarean delivery rate increased by 16.6 per 1,000 women (95% confidence interval 12.1–21.2, P<.001). In multivariable analysis, this association continued to be significant: a $10,000 annual increase in professional liability premium was independently associated with an increase in the cesarean delivery rate of 15.7 per 1,000 women (Table 2).
A similar association between premiums and cesarean delivery was seen in multiparous women. In univariable analysis, for each annual $10,000 increase in the previous year's premium, the primary cesarean delivery rate increased by 4.7 per 1,000 women (95% confidence interval 2.7–6.6, P<.001). After adjusting for potential confounding factors in multivariable analysis, the association continued to be significant (Table 2).
In this analysis, we have demonstrated that an increase in professional liability premium is associated with a subsequent increase in the rate of primary cesarean delivery among both nulliparous and multiparous women. This finding confirms the existence of an association that has not been well established. Although some investigators have previously noted an association between cesarean delivery and malpractice pressures, they had not accounted for differences in patient characteristics that may have been responsible for the reported association. Other investigators have employed a cross-sectional design, with the result that the temporal association between malpractice pressures and cesarean delivery rates was unclear.10 In the present study, we have controlled for patient demographic factors as well as obstetric factors to limit the potential for confounding. Of note, we have not included intrapartum factors, such as prolonged labor, in our regression equation because these are factors on the causal pathway to cesarean delivery and not “risk factors” per se. With regard to the measure of medical–legal pressure, different variables may be considered to represent this pressure, and other investigators have used such variables as the number of claims or plaintiff attorneys in a given county.14,15 We chose insurance premiums to represent the medical–legal pressure on physicians in Illinois because these premiums, not claims or attorneys, are directly experienced and paid by all practicing obstetric clinicians in Illinois. An additional strength of the study is the longitudinal design and the use of a prior year's professional liability premiums to predict cesarean delivery rates. This methodology provides results that demonstrate not just an association, but evidence that a rise in professional liability insurance premiums precedes and is independently associated with a subsequent rise in cesarean delivery rates.
Nevertheless, potential limitations should be acknowledged. This study is based on an analysis of two secondary data sets, and there is the potential for misclassification and ascertainment bias associated with such data. This analysis includes women delivering in a single state over a 6-year period. Consequently, the results of this analysis are not necessarily generalizable throughout the United States. Also, there remains the possibility that other unmeasured factors may actually be responsible for the observed association. The premium data were obtained from a single insurance provider. However, the insurance company, ISMIE Mutual Insurance Company, represents the majority of obstetrician–gynecologists in Illinois, and there is no evidence that the few who are not covered by this company are exposed to a significantly different rate environment. Lastly, these findings are based on an observational analysis and cannot confirm causality.
Even if this study cannot in and of itself prove causality, it is plausible that causality exists. Although there are indications for cesarean delivery, these indications, such as “nonreassuring fetal status,” are known to have a subjective component. Other investigators have demonstrated the marked geographic variation that exists in cesarean delivery rates, even after taking patient factors into account.7 Also, factors external to the patient, such as physician practice patterns, have been demonstrated to affect cesarean delivery decisions.9 Indeed, physicians themselves have noted that their decision to proceed with obstetric interventions is influenced by the medical–legal climate.11,15 The present analysis seems to support that self-assessment.
What remains uncertain is the effect this relationship may have upon patient outcomes. Conceivably, an increase in cesarean deliveries could enhance maternal and neonatal outcomes if these cesarean deliveries were increasingly used in indicated situations. Alternatively, an increase in cesarean deliveries in response to external pressures, and not patient factors, could result in greater maternal and neonatal morbidity, not just in the present pregnancy but in future pregnancies as well. Further research should elucidate the relationship between malpractice pressures and these short-term and long-term perinatal outcomes.
1. Martin JA, Kochanek KD, Strobino DM, Guyer B, MacDorman MF. Annual summary of vital statistics—2003. Pediatrics 2005;115:619–34.
2. MacDorman MF, Minino AM, Strobino DM, Guyer B. Annual summary of vital statistics–2001. Pediatrics 2002;110:1037–52.
3. Hoyert DL, Freedman MA, Strobino DM, Guyer B. Annual summary of vital statistics: 2000. Pediatrics 2001;108:1241–55.
4. Guyer B, Hoyert DL, Martin JA, Ventura SJ, MacDorman MF, Strobino DM. Annual summary of vital statistics—1998. Pediatrics 1999;104:1229–46.
5. Guyer B, Freedman MA, Strobino DM, Sondik EJ. Annual summary of vital statistics: trends in the health of Americans during the 20th century. Pediatrics 2000;106:1307–17.
6. Arias E, MacDorman MF, Strobino DM, Guyer B. Annual summary of vital statistics—2002. Pediatrics 2003;112:1215–30.
7. Baicker K, Buckles KS, Chandra A. Geographic variation in the appropriate use of cesarean delivery. Health Aff (Millwood) 2006;25:w355–67.
8. Kozinszky Z, Orvos H, Zoboki T, Katona M, Wayda K, Pal A, et al. Risk factors for cesarean section of primiparous women aged over 35 years. Acta Obstet Gynecol Scand 2002;81:313–6.
9. Luthy DA, Malmgren JA, Zingheim RW, Leininger CJ. Physician contribution to a cesarean delivery risk model. Am J Obstet Gynecol 2003;188:1579–85.
10. Rock SM. Malpractice premiums and primary cesarean section rates in New York and Illinois. Public Health Rep 1988;103:459–63.
11. American College of Obstetricians and Gynecologists, Department of Professional Liability/Risk Management. District VI Liability Lowdown: Overview of the 2006 ACOG Survey on Professional Liability. Washington (DC): ACOG; 2006. p. 1–3.
12. Studdert DM, Mello MM, Sage WM, DesRoches CM, Peugh J, Zapert K, et al. Defensive medicine among high-risk specialist physicians in a volatile malpractice environment. JAMA 2005;293:2609–17.
13. Studdert DM, Mello MM, Gawande AA, Gandhi TK, Kachalia A, Yoon C, et al. Claims, errors, and compensation payments in medical malpractice litigation. N Engl J Med 2006;354:2024–33.
14. Baldwin LM, Hart LG, Lloyd M, Fordyce M, Rosenblatt RA. Defensive medicine and obstetrics. JAMA 1995;274:1606–10.
15. Localio AR, Lawthers AG, Bengtson JM, Hebert LE, Weaver SL, Brennan TA, et al. Relationship between malpractice claims and cesarean delivery. JAMA 1993;269:366–73.
16. Tussing AD, Wojtowycz MA. Malpractice, defensive medicine, and obstetric behavior. Med Care 1997;35:172–91.