Every year in the United States 4 million women give birth, making childbirth the leading cause of hospital admission and a major source of hospital revenues.1 Over the past decade, there has been an increase in the use of epidural and spinal analgesic techniques for vaginal delivery, as well as an increase in the neuraxial technique for both planned and unplanned cesarean deliveries.1,2 Obstetric anesthesia care represents a significant portion of clinical services provided by anesthesiologists in the hospital setting. Recent studies suggest that 3.2 million women (80% of women giving birth) receive analgesia and anesthesia care during labor and delivery in the United States.2
Extensive efforts have been made in obstetrics and obstetric anesthesia to improve maternal safety, including evidence-based practice guidelines and reviews of mortality and morbidity, monitoring maternal morbidity and diffusing information to health care providers, and identifying specific interventions.3–6 These efforts have led to a marked reduction in the rate of anesthesia-related adverse events (ARAEs).7,8 The rate of ARAEs deemed clinically serious (e.g., failed intubation, high neuraxial block) during the period 2004 to 2009 reported in the multicenter SCORE project was 0.3 per 1000 deliveries.2 However, ARAEs still occur in 5 per 1000 deliveries, suggesting that most ARAEs in obstetrics are minor.9,10 Nonetheless, these minor events may result in mother dissatisfaction, prolonged hospital stay, and increased costs.
Although the rate of ARAEs is a valuable indicator of obstetric anesthesia safety, it provides little information about the impact of ARAEs on the mother’s health and well-being, the payer, or the hospital. Consequences for the payer can be assessed with the cost associated with ARAEs but have not been adequately examined. As efforts nationwide focus on improving both the quality and value of health care delivery, it is important to quantify such costs through a cost identification analysis.
In recent years, there has been a small increase in studies analyzing costs in obstetric anesthesia; however, these studies have been limited to narrow fields of focus, such as stillbirths and neuraxial anesthesia to facilitate external cephalic version.11,12 Analyses based on insurance claims such as the Closed Claims Analysis of Liability Associated with Obstetric Anesthesia may offer a broader perspective,13 but they still fail to fully capture the true hospital costs. The Closed Claims Analysis reports that the cost associated with ARAEs in obstetric anesthesia is nearly twice that associated with complications in other anesthesia specialities14 and has not decreased over the past 2 to 3 decades. Meanwhile, investigators in other specialties, such as orthopedics and gastroenterology, have consulted hospital discharge records collected in administrative databases to both describe hospital costs and perform cost analyses.15–17 Given the challenge to identify specific cost drivers within these databases, many studies within health services research have used a charge-to-cost ratio method to roughly estimate costs from hospital charges.18,19 Recently, Oofuvong et al.20 applied a similar methodology in pediatric anesthesia to examine excess costs associated with perioperative respiratory events. The aim of this study was to estimate the cost associated with ARAEs during labor and delivery in New York State for the year 2010.
Human Subjects Protections
The study protocol was reviewed by the IRB of Columbia University Medical Center and was granted exemption under 45 CFR 46 (not human subjects research). The study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for reporting of observational studies.21
The study sample consisted of all women admitted for labor and delivery in New York State between January 1, 2010, and December 31, 2010. Hospital discharge record data for these women collected in the deidentified State Inpatient Database (SID) for New York were analyzed. The SIDs are part of the Healthcare Cost and Utilization Project sponsored by the Agency for Healthcare Research and Quality (AHRQ). The SID captures all inpatient discharges from community hospitals in participating states since 1988. Community hospitals are nonfederal, short-term, general, and other specialty hospitals, including university hospitals. For each discharge, the SID includes patient characteristics, total hospital charges, one hospital identifier, and up to 15 procedural and diagnostic codes defined in the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Detailed information on SID is publicly available.a
Discharges with female sex were first selected. Then, discharges indicating labor and delivery were identified with a combination of ICD-9-CM diagnosis and procedure codes developed by Kuklina et al.22 (Appendix 1.1). Excluded from the analyses were 8742 discharges with missing values for total hospital charges.
Anesthesia-Related Adverse Events
ARAEs were identified with a combination of ICD-9-CM diagnosis and procedure codes as previously described by Cheesman et al.9 The crude rate of ARAEs was calculated as the ratio of the number of discharges recording at least one adverse event to the total number of discharges. Adverse events were divided into 6 groups based on ICD-9-CM codes (Appendix 1.2): (1) pulmonary adverse events; (2) cardiac adverse events; (3) central nervous system adverse events; (4) other or unspecified adverse events; (5) adverse events related to neuraxial anesthesia, including headache following lumbar puncture, blood patch, adverse effects of spinal anesthetics, and epidural, extradural, or subdural abscess; and (6) adverse events related to anesthetic drugs. In addition, an ARAE was defined major if associated with 1 of the 3 following maternal conditions: (1) death; (2) cardiac arrest; (3) or severe cardiac, respiratory, or neurologic morbidity. In-hospital death was directly recorded from the SID. Cardiac arrest and severe morbidity were identified with ICD-9-CM codes (Appendix 1.3).
Patient, Delivery, and Hospital Characteristics
The following characteristics were recorded directly from the SID: age, race/ethnicity, insurance type, median household income for patient’s ZIP code, type of admission (week/weekend, elective/nonelective), length of hospital stay, and total hospital charges. Charlson and Bateman comorbidity indexes were calculated with algorithms based on ICD-9-CM codes.23,24 Unlike the Charlson index, the Bateman index for use in obstetric patients includes pregnancy-related risk factors for adverse maternal outcome, such as severe preeclampsia and eclampsia or multiple gestation.
Delivery was categorized into 4 groups: planned cesarean delivery, unplanned cesarean delivery, induced vaginal delivery, and noninduced vaginal delivery. A planned cesarean delivery was defined as a discharge not indicating labor and not indicating one of the following maternal and fetal conditions requiring urgent delivery: abnormal fetal heart rate tracing, placental abruption, uterine rupture, or umbilical cord prolapse.25 An induced vaginal delivery was defined as a discharge indicating an induction of labor (Appendix 1.4).
For each hospital, the number of deliveries and the cesarean delivery rate were calculated from the SID. The number of deliveries was categorized as <250, between 250 and 499, between 500 and 999, and ≥1000. The cesarean delivery rate was categorized as <25.0%, between 25.0% and 29.9%, between 30.0% and 34.9%, and ≥35.0%. Other hospital characteristics were obtained from the American Hospital Association (AHA) file for the year 2010. From the AHA file, the following characteristics were obtained: accreditation by the Joint Commission, residency training approval by the Accreditation Council for Graduate Medical Education, medical school affiliation reported to American Medical Association, hospital ownership, and hospital location. Hospital ownership was categorized as government nonfederal control or nongovernment not-for-profit control. A teaching hospital was defined as a hospital with an affiliation to a medical school or a residency training approval. In the AHA file, hospital location is based on the Core Based Statistical Areas type and includes 4 categories: metropolitan statistical areas, metropolitan divisions, micropolitan statistical areas, and rural. A micropolitan statistical area corresponds to at least one urban cluster that has a population of at least 10,000 but <50,000. For the purpose of the study, rural hospital location included micropolitan or rural.
The primary outcome was the adjusted excess cost associated with ARAEs and expressed in 2010 US dollars. The secondary outcomes were the percentage increase, cost ratio, cost per admission, and total cost associated with ARAEs. The primary and secondary outcomes were also calculated for (1) major and minor ARAEs and (2) for the 6 types of ARAES previously defined.
Unadjusted costs were calculated as the product of total hospital charges and the group average all-payer inpatient cost-to-charge ratio provided by the Healthcare Cost and Utilization Project. For the year 2010, the mean group average all-payer inpatient cost-to-charge ratio (1 SD) was 0.47 (0.16). Costs were adjusted with a linear regression model described in the section “Statistical Analysis.” This model takes into account the number of ARAEs reported in each discharge.
The excess cost (or cost ratio) was calculated as the difference (or ratio) between the mean cost of discharges with and without ARAEs. The percentage increase was calculated as the ratio of the difference between the mean cost of discharges with and without ARAEs (i.e., excess cost) to the mean cost in discharges without ARAEs. The excess cost (or cost ratio) and the percentage increase can be viewed as a proxy for the severity of adverse events.
The cost per admission (or total cost) was calculated as the product of the excess cost and the incidence (or number) of ARAEs. The cost per admission and total cost can be viewed as the financial burden associated with ARAEs at the patient level and the state level, respectively.
To provide a comparison with the cost of ARAEs, the outcomes variables were also calculated for 2 pregnancy-related complications: postpartum hemorrhage and preeclampsia/eclampsia (Appendix 1.4).
Results are expressed as number (%) or mean ± 1 SD. The statistical analysis was performed with R version 3.0.2 (R Foundation for Statistical Computing, Vienna, Austria).
Univariate comparisons between discharges with and without ARAEs used unpaired Wilcoxon tests for quantitative variables and χ2 or Fisher exact tests for qualitative variables.
Unadjusted costs were compared according to patient, pregnancy, delivery, number of ARAES reported per discharge, and hospital characteristics with unpaired Wilcoxon tests or Kruskal-Wallis tests. All characteristics with differences in unadjusted costs at a P value <0.05 along with the hospital identifier were entered in a linear regression model using the backward selection approach. In the model, costs were expressed in a log form. For candidate variables with missing values, multivariate imputation was used. Performance of the final model was assessed with its R2 and the visual inspection of the relationship between fitted and residuals values.
The 95% confidence intervals (CIs) for the adjusted excess cost and the secondary outcomes were calculated with bootstrap resampling with replacement (B = 2000) and the percentiles method.
There were a total of 228,104 delivery-related discharges in New York State in 2010, including 79,035 (34.6%) cesarean deliveries. At least one ARAE was recorded in 1053 discharges (4.62 per 1000 discharges) and one major ARAE in 19 discharges (0.08 per 1000 discharges). The distribution of the 6 types of adverse events is presented in Table 1.
Compared with discharges without ARAEs, discharges with ARAEs were significantly more likely to involve older parturients, have a private insurance, have a Charlson comorbidity index ≥1 or a Bateman comorbidity index ≥2, and involved induced vaginal delivery or cesarean delivery (Table 2). Delivery with ARAEs was more likely to occur in hospitals with nongovernment control, nonteaching status, rural location, low volume of deliveries, or high cesarean delivery rate. On average, discharges with ARAEs had significantly longer hospital stay, higher total hospital charges, and higher unadjusted costs than discharges without ARAEs (Table 2).
Significant differences in unadjusted costs were observed according to patient, pregnancy, delivery, and hospital characteristics (Table 3). Higher unadjusted costs were observed in hospitals accredited by the Joint Commission, in nongovernment controlled, teaching, urban, high volume, or high cesarean delivery rate hospitals. The final cost-adjustment model is presented in Appendix 2, and its performance assessed with R2 was 0.7.
Cost Associated with ARAEs
The adjusted excess cost associated with ARAEs during labor and delivery was $1189 (95% CI, $1033–$1350), yielding total excess cost of $1.252 million (95% CI, $1.088–$1.421 million) for New York State in 2010. For minor ARAEs, the excess cost was $1173 (95% CI, $1006–$1340); for major ARAEs, it was $2105 (95% CI, $987–$3316). The total adjusted excess cost was $1.213 million for minor ARAEs and $0.40 million for major ARAEs.
The excess cost varied according to the 6 types of ARAEs (Table 4), with higher excess costs for pulmonary, cardiac, and central nervous system adverse events and for adverse events related to anesthetic drugs. The percentage increase and the cost ratio according to the type of adverse events displayed a similar pattern.
The overall cost per admission associated with ARAEs was $5.49 (95% CI, 4.77–6.23). For minor ARAEs, it was $5.32 (95% CI, 4.56–6.07), and for major ARAES, it was $0.18 (95% CI, 0.08–0.28). The cost per admission according to the type of adverse events is reported in Table 4; because of a higher incidence, it was higher for adverse events related to neuraxial anesthesia and other and unspecified systemic adverse events.
Cost Associated with Pregnancy-Related Complications
The excess cost associated with ARAEs ($1189; 95% CI, 1033–1350) was higher than the one associated with postpartum hemorrhage ($679; 95% CI, 608–748), and similar to the one associated with preeclampsia and eclampsia ($1328; 95% CI, 1272–1378). A similar pattern was observed for the cost ratio and the percentage increase. Because of a higher incidence of the 2 pregnancy-related complications, the cost per admission and the total cost for these 2 complications were higher than for ARAEs (Table 5).
The major finding of this study is that ARAEs during labor and delivery are associated with significant costs after adjustment for patient, pregnancy, delivery, and hospital characteristics. On average, ARAEs were attributable to $1189 in excess cost, representing a 25% difference over the average cost of a delivery without any ARAE. The financial burden resulting from ARAEs in New York State in 2010 was estimated as $1.2 million.
The excess cost (cost ratio) and percentage increase calculated in this study can be viewed as a proxy for the severity of the adverse events. As expected, the excess cost (cost ratio) and percentage increase were higher for major adverse events (i.e., adverse events associated with death, cardiac arrest, or severe morbidity) than for minor adverse events ($2105 vs $1173). Similarly, the excess cost according to the 6 types of events was lower for those with a low proportion of major events such as adverse events related to neuraxial anesthesia ($1177).
Interventions aimed at reducing adverse events and associated costs should target frequent adverse events and serious or critical adverse events. Criticality is defined as the product of the incidence of the event and its severity (i.e., excess costs).26,27 In this study, criticality corresponds to the cost per admission and had a mean value of $5.49. The cost per admission was higher for minor than for major adverse events ($5.32 vs $0.18) and for adverse events related to neuraxial anesthesia ($4.26). On a cost-benefit basis, the marked disparity in cost per admission suggests that interventions aimed at reducing adverse events related to neuraxial anesthesia merit higher priority over other types of adverse events. More than 90% of adverse events related to neuraxial anesthesia are postdural puncture headaches.25 In this setting, development of ultrasound imaging for epidural placement may reduce dural puncture and postdural puncture headache and associated costs.28 However, it should be noted that on a cost-benefit basis, any such intervention would need to achieve a proportional reduction in adverse event risk relative to the roughly $5 excess cost per admission, which may prove a difficult goal to achieve.
This study was consistent with recently published studies reporting a difference in adverse event rates based on insurance type (Table 2).29 Furthermore, the difference in unadjusted costs among insurance types also aligned with the differences in event rates (Tables 2 and 3). One possible explanation proposes that the difference in event rates—and we suggest potentially the unadjusted costs as well—may be tied to the different reimbursement rates among private insurance, Medicare, and Medicaid.29 While Medicare traditionally serves as public insurance for patients 65 years and older, it does cover some patients of all ages who receive Social Security disability benefits and therefore provides for obstetric health care.30 Given that these younger patients likely represent the obstetric Medicare population, their disability-related comorbidities may account for the difference in event rate compared with Medicaid, which offers public insurance to patients of all ages based on income level.
As with any discussion of health care costs, comparisons are crucial to fully comprehend the numbers. We therefore compared costs for ARAEs with costs for 2 common pregnancy-related complications responsible for significant maternal morbidity8: postpartum hemorrhage and preeclampsia/eclampsia (Table 5). The excess cost (cost ratio) and percentage increase associated with ARAEs were higher than those associated with postpartum hemorrhage, the main driver of the increase in maternal morbidity observed in the United States during the past 2 decades.6 They were similar to those associated with preeclampsia/eclampsia. As a result of their lower incidence, the financial burden associated with ARAEs was lower than those associated with these 2 pregnancy-related complications. For instance, the total cost of ARAEs at the New York State level was $1.2 million, but $3.9 million for postpartum hemorrhage and $13.3 million for preeclampsia/eclampsia.
As the obstetric anesthesia community expands its ongoing efforts to improve the quality and safety of anesthesia care, the relative excess cost per admission may be instructive in both setting expectations and prioritizing such endeavors. Compared with the $58.06 and $17.07 excess cost per admission for preeclampsia/eclampsia and postpartum hemorrhage, respectively, the $5.49 excess cost per admission for ARAEs suggests that any further, significant reduction in the risk of ARAEs during labor and delivery may be unreasonable to expect on a cost-benefit basis. This should be viewed as good news, as it suggests that recent efforts to improve the quality and value of obstetric anesthesia care have been largely successful. Furthermore, this study provides a relatively simple methodology to roughly estimate excess cost per admission, which may aid providers and policy makers in efficiently directing scarce resources toward quality and safety improvement efforts.
This study has several limitations. First, data came from a single state, New York. Variation in the practice of obstetrics and obstetric anesthesia across states in the United States suggests that the excess cost and other cost variables calculated in this study may not provide a reliable estimate of national figures.31 Second, costs are probably underestimated because they focus on medical treatment and were limited to direct hospital charges during the delivery hospitalization. We did not include indirect costs related to morbidity and mortality, such as follow-up care, lost productivity, or costs associated with a readmission (e.g., readmission to perform an epidural blood patch). Because of the low numbers of deaths and serious adverse events, this study lacked the statistical power to estimate morbidity and mortality costs. Any future cost-benefit analysis focused on potential interventions would require such an estimate to be complete. Third, costs are estimated from charges and a cost-to-charge conversion factor. Although this approach has been extensively used in health economics,18,19 it does not allow the identification of cost drivers through an itemized analysis.
In conclusion, ARAEs during labor and delivery are associated with significant costs. While further improvements in the safety and quality of obstetric anesthesia care are still necessary and important, the small excess cost per admission associated with ARAE suggests that a further reduction in the risk of ARAEs may not be feasible on a cost-benefit basis.
International Classification of Diseases, Ninth Revision, Clinical Modification Codes Used in This Study
2.2. Relationship Between the Costs Expressed in a Log Form Estimated with the Linear Regression Model and the Residuals
Name: Tristan D. Hunt, BS.
Contribution: This author helped design the study, conduct the study, collect data, analyze data, and prepare the manuscript.
Attestation: Tristan D. Hunt approved the final manuscript, attests to the integrity of the original data and the analysis reported in this manuscript, and is the archival author.
Name: Jean Guglielminotti, MD, PhD.
Contribution: This author helped design the study, conduct the study, collect data, analyze data, and prepare the manuscript.
Attestation: Jean Guglielminotti approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Guohua Li, MD, DrPH.
Contribution: This author conceived of the study, oversaw the development and implementation of the research plan, supervised statistical analysis and interpretation of the study results, and contributed to the critical revision of the manuscript.
Attestation: Guohua Li approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
This manuscript was handled by: Franklin Dexter, MD, PhD.
a http://www.hcup-us.ahrq.gov/sidoverview.jsp. Accessed December 23, 2015.
1. Martin JA, Hamilton BE, Ventura SJ, Osterman MJ, Mathews TJ. Births: final data for 2011. Natl Vital Stat Rep 2013;62:1–69.
2. D’Angelo R, Smiley RM, Riley ET, Segal S. Serious complications related to obstetric anesthesia: the serious complication repository project of the Society for Obstetric Anesthesia and Perinatology. Anesthesiology 2014;120:1505–12.
3. The California Pregnancy-Associated Mortality Review. Report from 2002 and 2003 Maternal Death Reviews. 2011Sacramento: California Department of Public Health, Maternal Child and Adolescent Health Division.
4. Creanga AA, Berg CJ, Ko JY, Farr SL, Tong VT, Bruce FC, Callaghan WM. Maternal mortality and morbidity in the United States: where are we now? J Womens Health (Larchmt) 2014;23:3–9.
5. American Society of Anesthesiologists Task Force on Obstetric A. Practice guidelines for obstetric anesthesia: an updated report by the American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Anesthesiology 2007;106:843–63.
6. Centers for Disease Control and Prevention. Severe Maternal Morbidity in the United States. Centers for Disease Control and Prevention, 2015Atlanta, GA:
7. Hawkins JL, Chang J, Palmer SK, Gibbs CP, Callaghan WM. Anesthesia-related maternal mortality in the United States: 1979-2002. Obstet Gynecol 2011;117:69–74.
8. Callaghan WM, Creanga AA, Kuklina EV. Severe maternal morbidity among delivery and postpartum hospitalizations in the United States. Obstet Gynecol 2012;120:1029–36.
9. Cheesman K, Brady JE, Flood P, Li G. Epidemiology of anesthesia-related complications in labor and delivery, New York State, 2002-2005. Anesth Analg 2009;109:1174–81.
10. Guglielminotti J, Li G. Monitoring obstetric anesthesia safety across hospitals through multilevel modeling. Anesthesiology 2015;122:1268–79.
11. Gold KJ, Sen A, Xu X. Hospital costs associated with stillbirth delivery. Matern Child Health J 2013;17:1835–41.
12. Carvalho B, Tan JM, Macario A, El-Sayed YY, Sultan P. Brief report: a cost analysis of neuraxial anesthesia to facilitate external cephalic version for breech fetal presentation. Anesth Analg 2013;117:155–9.
13. Davies JM, Posner KL, Lee LA, Cheney FW, Domino KB. Liability associated with obstetric anesthesia: a closed claims analysis. Anesthesiology 2009;110:131–9.
14. Chadwick HS, Posner K, Caplan RA, Ward RJ, Cheney FW. A comparison of obstetric and nonobstetric anesthesia malpractice claims. Anesthesiology 1991;74:242–9.
15. Bohensky MA, Ademi Z, deSteiger R, Liew D, Sundararajan V, Bucknill A, Kondogiannis C, Brand CA. Quantifying the excess cost and resource utilisation for patients with complications associated with elective knee arthroscopy: a retrospective cohort study. Knee 2014;21:491–6.
16. Stepanova M, Mishra A, Venkatesan C, Younossi ZM. In-hospital mortality and economic burden associated with hepatic encephalopathy in the United States from 2005 to 2009. Clin Gastroenterol Hepatol 2012;10:1034–41.e1.
17. Whitmore RG, Schwartz JS, Simmons S, Stein SC, Ghogawala Z. Performing a cost analysis in spine outcomes research: comparing ventral and dorsal approaches for cervical spondylotic myelopathy. Neurosurgery 2012;70:860–7.
18. Tellez MG, Mackersie RC, Morabito D, Shagoury C, Heye C. Risks, costs, and the expected complication of re-injury. Am J Surg 1995;170:660–3.
19. Kahler ZP, Beam DM, Kline JA. Cost of treating venous thromboembolism with heparin and warfarin versus home treatment with rivaroxaban. Acad Emerg Med 2015;22:796–802.
20. Oofuvong M, Geater AF, Chongsuvivatwong V, Chanchayanon T, Sriyanaluk B, Saefung B, Nuanjun K. Excess costs and length of hospital stay attributable to perioperative respiratory events in children. Anesth Analg 2015;120:411–9.
21. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JPSTROBE Initiative. STROBE InitiativeThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453–7.
22. Kuklina EV, Whiteman MK, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, Marchbanks PA. An enhanced method for identifying obstetric deliveries: implications for estimating maternal morbidity. Matern Child Health J 2008;12:469–77.
23. Quan H, Li B, Couris CM, Fushimi K, Graham P, Hider P, Januel JM, Sundararajan V. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol 2011;173:676–82.
24. Bateman BT, Mhyre JM, Hernandez-Diaz S, Huybrechts KF, Fischer MA, Creanga AA, Callaghan WM, Gagne JJ. Development of a comorbidity index for use in obstetric patients. Obstet Gynecol 2013;122:957–65.
25. Guglielminotti J, Wong CA, Landau R, Li G. Temporal trends in anesthesia-related adverse events in cesarean deliveries, New York State, 2003-2012. Anesthesiology 2015;123:1013–23.
26. Guida E, Rosati U, Pini Prato A, Avanzini S, Pio L, Ghezzi M, Jasonni V, Mattioli G. Use of failure modes, effects, and criticality analysis to compare the vulnerabilities of laparoscopic versus open appendectomy. J Patient Saf 2015;11:105–9.
27. Joint Commission. Failure Mode and Effects Analysis in Health Care: Proactive Risk Reduction. 20103rd ed. Oakbrook Terrace, IL: Joint Commission Resources.
28. Shaikh F, Brzezinski J, Alexander S, Arzola C, Carvalho JC, Beyene J, Sung L. Ultrasound imaging for lumbar punctures and epidural catheterisations: systematic review and meta-analysis. BMJ 2013;346:f1720.
29. Spencer CS, Roberts ET, Gaskin DJ. Differences in the rates of patient safety events by payer: implications for providers and policymakers. Med Care 2015;53:524–9.
30. Centers for Medicare and Medicaid Services. Inpatient hospital services covered under Part A. In: Medicare Benefit Policy Manual. 2014:Baltimore: Centers for Medicare and Medicaid Services, 22.
© 2016 International Anesthesia Research Society
31. Osterman MJ, Martin JA. Epidural and spinal anesthesia use during labor: 27-state reporting area, 2008. Natl Vital Stat Rep 2011;59:1–13.