Obstetrics & Gynecology:
Paradoxical Trends and Racial Differences in Obstetric Quality and Neonatal and Maternal Mortality
Howell, Elizabeth A. MD, MPP; Zeitlin, Jennifer DSc, MA; Hebert, Paul PhD; Balbierz, Amy MPH; Egorova, Natalia PhD, MPH
Departments of Health Evidence & Policy, Obstetrics, Gynecology, and Reproductive Science, and Psychiatry, Mount Sinai School of Medicine, New York, New York; the Epidemiological Research Unit on Perinatal Health and Women's and Children's Health, UMRS 953, INSERM Paris, France; and the Department of Health Services, University of Washington School of Public Health, Seattle, Washington.
Corresponding author: Elizabeth A. Howell, MD, MPP, Department of Health Evidence & Policy, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1077, New York, NY 10029-6574; e-mail: firstname.lastname@example.org.
Supported by grant number R21HD068765 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health and Human Development or the National Institutes of Health. Dr. Zeitlin received funding from the European Commission, Research Directorate, Marie Curie, IOF Fellowship, grant number 254171.
Financial Disclosure The authors did not report any potential conflicts of interest.
Presented at Academy Health Annual Research Meeting, June 27–28, 2010, Boston, Massachusetts.
OBJECTIVE: To evaluate trends by race in Agency for Healthcare Research and Quality obstetric-related quality and safety indicators and their relationships to trends in inpatient maternal and neonatal mortality.
METHODS: We used the Nationwide Inpatient Sample from 2000 through 2009 and calculated obstetric hospital quality and patient safety indicators and inpatient maternal and neonatal mortality stratified by race. We examined differences in age and comorbidity-adjusted trends in black compared with white women over time in the United States and by geographic region. Proportions were analyzed by χ2 and trends by regression analysis.
RESULTS: Obstetric quality indicators varied by geographic region, but changes over time were consistent for both races. Cesarean deliveries increased similarly for black and white women, and vaginal births after cesarean delivery declined for both races but more rapidly for white women than for black women. Obstetric safety indicators improved over the study period for black and white women, with obstetric trauma decreasing significantly for both groups (28% compared with 35%, respectively) and birth trauma–injury to neonates declining for both, but changes were not significant. In striking contrast, inpatient maternal and neonatal mortality remained relatively constant during the study period, with persistently higher rates of both seen among black compared with white women (12.0 compared with 4.6 per 100,000 deliveries, P<.001 and 6.6 compared with 2.5 per 1,000 births, P<.001, respectively, in 2009).
CONCLUSION: Improvements in Agency for Healthcare Research and Quality quality indicators for obstetrics are not reflected in improvements in maternal and neonatal morbidity and mortality and do not explain continued racial disparities for outcomes in pregnancies in black and white women. Quality measures that are related to pregnancy outcomes are needed and these should elucidate obstetric health disparities.
LEVEL OF EVIDENCE: II
More than four million births occur annually in the United States, and childbirth is the largest category for hospital admissions for commercial payers and Medicaid programs.1,2 Complications associated with delivery are not rare and are sensitive to quality of care at delivery.3–6 Furthermore, persistent racial disparities in perinatal outcomes exist, with neonatal mortality rates in black women twice those in white women and maternal mortality rates in black women three to four times those in white women.7,8
Hospital quality is associated with neonatal and obstetric outcomes and quality among hospitals varies.9–11 Preventable causes account for one third to one half of maternal deaths including hospital practices such as the management of hypertension, deep venous thrombosis, chronic diseases, and hemorrhage during delivery.12–14 Given these facts, there is a rising focus on the quality of hospital care as a means to reduce both overall maternal and neonatal mortality and to narrow racial disparities in perinatal outcomes.
The Agency for Healthcare Research and Quality developed seven indicators of obstetric health care quality and safety that make use of readily available hospital inpatient administrative data.15 The objectives of this study were to examine differences between black and white women in these indicators, explore geographic variation in these indicators by race, and examine differences between black and white women in maternal and neonatal mortality over time. We hypothesized that data over the past decade would demonstrate improving trends in quality and decreasing neonatal and maternal mortality for both black and white women.
MATERIALS AND METHODS
We used data from the 2000 through 2009 Nationwide Inpatient Sample of the Healthcare Cost and Utilization Project, a federal–state–industry partnership sponsored by the Agency for Healthcare Research and Quality. The Nationwide Inpatient Sample is a stratified sample representing 20% of U.S. community hospitals.16 We examined all deliveries that occurred in hospitals with at least 10 deliveries annually, because we wanted to analyze hospitals with an obstetric volume high enough to preclude accidental births. The number of states participating in the Nationwide Inpatient Sample ranged from 28 in 2000 to 44 in 2009.16 Validity and reliability of the Nationwide Inpatient Sample have been studied extensively.17 Nationwide Inpatient Sample data are publicly available and do not include personal identifiers. The Mount Sinai Program for Protection of Human Subjects (institutional review board) deemed this research exempt.
We computed indicators of patient safety, hospital quality of care, and neonatal and maternal mortality. We used the three Agency for Healthcare Research and Quality patient safety indicators related to obstetrics: birth trauma–injury to the neonate, obstetric trauma with instrument, and obstetric trauma without instrument. Obstetric trauma refers to third- and fourth-degree perineal lacerations. We also examined the four Agency for Healthcare Research and Quality inpatient quality indicators related to obstetrics: cesarean delivery, primary cesarean, uncomplicated vaginal birth after cesarean delivery, and all vaginal births after cesarean delivery. The Agency for Healthcare Research and Quality specifies the numerator and denominator for each indicator. For example, to measure birth trauma–injury to the neonate, we included all deliveries with these International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes: 7670, 7671, 7673, 7674, 7675, 7677, 7678. As specified by the Agency for Healthcare Research and Quality, we excluded neonates with any diagnosis code of neonates with a birth weight of less than 2,000 g. We also excluded neonates with any injury to the brachial plexus or any diagnosis code of osteogenesis imperfecta. The denominator was all live-born births (newborns) with these ICD-9-CM codes: V3000, V3001, V3100, V3101, V3200, V3201, V3300, V3301, V3400, V3401, V3500, V3501, V3600, V3601, V3700, V3701, V3900, V3901 or with specified newborn admission codes and age at admission equal to zero days. Each patient safety indicator and inpatient quality indicator has specific inclusion and exclusion criteria detailed by the Agency for Healthcare Research and Quality.15
We computed inpatient maternal mortality using diagnosis-related group 370–375 and disposition equal to died. We computed inpatient neonatal mortality by identifying newborn admissions with disposition equal to died for infants up to 1 year of life, which has been recommended to avoid bias from not including deaths who are hospitalized for long periods and for whom death is related to perinatal causes.18 Ninety-nine percent of deaths occurred within 28 days of birth in this cohort. We identified newborns using the same algorithm as recommended by the Agency for Healthcare Research and Quality for identification of live newborns.
For each hospital, we computed risk-adjusted rates of patient safety and inpatient quality indicators. Risk adjustment consisted of controlling for age, sex, modified diagnosis-related group, and comorbidities as specified by the Agency for Healthcare Research and Quality. We examined risk-adjusted patient safety and inpatient quality indicators by each of four geographic regions: Northeast, Midwest, South, and West. Proportions were analyzed by χ2 and trends by linear regression of the adjusted annual rate on time. Because the ICD-9-CM codes used by the Agency for Healthcare Research and Quality for birth trauma–injury to the neonate changed in 2003, we used data from 2004 to 2009 to calculate trends in birth trauma rates.
We conducted race-specific analyses and limited our analyses to white and black women. We then compared specific risk-adjusted rates for quality and safety indicators and for neonatal and maternal mortality overall and by geographic region in black and white women. To examine whether trends in quality indicators differed by race, we included interaction terms between race and a linear time term. To account for nonlinear trend, we included both time and the square of time year as main effects and interactions with race. Race was missing for approximately 28% of the hospitals. Race data in Nationwide Inpatient Sample are incomplete because of differences in state and hospital procedures for collecting race data.19 We therefore conducted sensitivity analyses comparing rates of inpatient quality indicators and patient safety indicators for all hospitals compared with those hospitals with race for at least 80% of deliveries. We conducted these analyses to verify that rates and trends based on the restricted sample were similar to those in the overall sample of hospitals. We also conducted sensitivity analyses for delivery volume and compared rates of inpatient quality indicators and patient safety indicators for all hospitals and those with at least 10 deliveries.
Patient safety indicators improved from 2000 to 2009 overall. Obstetric trauma decreased by 28% for white women (201.9 to 145.7 per 1,000 vaginal deliveries with instrument) and by 35% for black women (140.4 to 90.9 per 1,000 vaginal deliveries with instrument) (Table 1; Fig. 1). Similar trends were evident for vaginal deliveries without instrument, which decreased by 44% for white women (43.7 to 24.3 per 1,000 vaginal deliveries without instrument) and by 43% for black women (23.5 to 12.3 per 1,000 vaginal deliveries without instrument). The proportion of instrumental deliveries also declined from 11% to 8% over the study period. Birth trauma–injury to the neonate declined 6% for black women and 21% for white women from 2004 through 2009. Rates of change for patient safety indicators were similar for black and white women except for a small but statistically significant difference in the rate of decline in obstetric trauma among vaginal deliveries without instrument. This decline was more rapid for white women than black women.
Changes in inpatient quality indicators were evident over the study period with primary cesarean delivery rates increasing and vaginal births after cesarean delivery decreasing. Cesarean delivery rates for white women increased from 197.5 per 1,000 deliveries in 2000 to 303.0 per 1,000 deliveries in 2009, an increase of 53%. Likewise, cesarean delivery rates for black women increased from 225.1 per 1,000 deliveries in 2000 to 330.8 per 1,000 deliveries in 2009, an increase of 47%. The rate of change was similar for both groups. Throughout the study period, risk-adjusted cesarean delivery rates were higher for black women than for white women (Fig. 1).
In contrast, the rate of decline for vaginal birth after cesarean delivery was more rapid for white women than black women. Rates of vaginal birth after cesarean delivery decreased by 74% for white women compared with 66% for black women (P<.001), and the decline occurred earlier for white women (Fig. 1). Throughout the study period, risk-adjusted vaginal birth after cesarean delivery rates remained higher among black women than white women.
Quality and patient safety indicators varied by geographic region for white and black women, but trends over time were similar for both groups (Table 2). For example, although primary cesarean delivery rates increased for black and white women over the 10-year period, there was wide variation in rates of cesarean delivery depending on geographic region. Rates in 2009 ranged from a low for white and black women (170.9 and 185.2 per 1,000 births, respectively) in the Midwest to a high (206.9 and 219.7 per 1,000 births, respectively) in the South. Similar to national trends, the decline in vaginal birth after cesarean delivery rates occurred more rapidly for white women than for black women, and this decline appeared to be most pronounced for the Northeast region.
Unlike dramatic changes in patient safety and inpatient quality indicators during the study period, inpatient maternal and neonatal mortality did not change substantially between 2000 and 2009 (Fig. 2) with persistently higher rates of both seen in black women compared with white women (12.0 compared with 4.6 per 100,000 deliveries, P<.001 and 6.6 compared with 2.5 per 1,000 births, P<.001, respectively, in 2009).
Trends in mortality ...Image Tools
Sensitivity analyses were conducted to compare rates of inpatient quality indicators and patient safety indicators for all hospitals compared with those hospitals with race for at least 80% of deliveries. Rates and trends in the overall sample were very similar to those in the group of hospitals with race data (Table 3). We also conducted sensitivity analyses for delivery volume and compared rates of inpatient quality indicators and patient safety indicators for all hospitals and those hospitals with at least 10 deliveries and found no significant difference in reporting between these two groups.
This article provides a nationally representative overview of obstetric quality for black and white women as measured using the Agency for Healthcare Research and Quality quality indicators. Our analyses demonstrate that some parameters of obstetric-related quality improved from 2000 to 2009. We found that patient safety indicators, in particular obstetric trauma during delivery, decreased markedly overall and for black and white women during this period. At the same time, inpatient quality indicators also changed for women. Cesarean delivery rates rose, whereas vaginal birth after cesarean delivery rates decreased during this period and these changes are consistent with published literature of this period and with changes in American College of Obstetricians and Gynecologists guidelines on safety of vaginal birth after cesarean delivery.20–22 However, our findings suggest a paradox: despite improvements in some parameters of hospital quality, inpatient maternal and neonatal mortality remained relatively stable.
During this period of increased attention to hospital quality, parameters of hospital quality and patient safety in obstetrics had similar patterns for black and white women. However, one exception was the rate of decline in vaginal birth after cesarean delivery rates, which was higher for white women than for black women and raises the hypothesis that the rate of uptake of practice-based guidelines may vary for patients by race. Whether this finding is the result of hospital site of care, patient decision-making, or health care provider decision-making cannot be determined with the use of this data set. Our findings that black women had lower rates of obstetric trauma and higher rates of cesarean deliveries than white women and that cesarean delivery rates vary by region are consistent with previous literature.23–26
Our findings suggest that the seven Agency for Healthcare Research and Quality obstetric quality and safety indicators are not associated with maternal and neonatal mortality. The Agency for Healthcare Research and Quality indicators were extensively tested for feasibility, validated according to strict criteria, and do assess domains within the Institute of Medicine's suggested six domains of quality (safety, effectiveness, patient-centered, timeliness, efficiency, and equity).15,27 However, they may not be associated with the underlying factors that are most important for explaining variation between hospitals in maternal and neonatal mortality.
Quality measures focused on more crucial processes of care on the pathway to maternal and neonatal mortality (such as the use of antenatal steroids in the setting of preterm delivery or the use of hemorrhage protocols in the delivery suite) are needed and may be targets for quality improvement activities.13,28,29 The Joint Commission recently endorsed the use of antenatal steroids as a perinatal quality measure.30 Others have endorsed quality measures such as elective delivery before 39 weeks of gestation, health care-associated bloodstream infections in newborns, deep venous thrombosis prophylaxis, and antibiotics in different settings including preterm premature rupture of membranes.31–34
Our findings also suggest that quality measures sensitive to disparities, or measures that are helpful in reducing disparities, are needed in obstetrics. The National Quality Forum has endorsed disparities-sensitive quality measures in other areas of medicine.35 One such potential obstetric measure is the use of antibiotics in the setting of preterm premature rupture of membranes. Given the high prevalence of preterm premature rupture of membranes among black mothers,36 its strong association with neonatal morbidity and mortality,37 and the evidence base for use of antibiotics in this setting,33 this measure may have the potential to help hospitals narrow disparities.
There are limitations with our study. Maternal mortality is underreported and rates of inpatient maternal mortality from the Nationwide Inpatient Sample are lower than reports enhanced by case ascertainment.38,39 Race data in the Nationwide Inpatient Sample are incomplete because of differences in state and hospital procedures for collecting race data and this could create bias if hospitals with missing race had different rates for these indicators or if differentials between white and black women did not follow the patterns observed in other hospitals.19 However, sensitivity analyses demonstrated no differences between hospitals that coded race and those that did not code race in terms of quality and safety indicators. We were also not able to sufficiently control for socioeconomic and environmental factors that likely contribute to disparities in outcomes with the use of administrative data. There are limitations inherent to use of ICD-9-CM codes to conduct medical research. Nevertheless, one of the strengths of Agency for Healthcare Research and Quality quality indicators is that they are ascertainable from routine administrative data.
Our study examined national trends in obstetric quality and safety and neonatal and maternal mortality stratified for black and white patients. We found that trends in Agency for Healthcare Research and Quality obstetric quality indicators and maternal and neonatal mortality were not aligned. Despite improvements in some parameters of obstetric quality, neonatal and maternal mortality rates stagnated and these indicators are unable to inform care focused on narrowing disparities. Given that minority women account for 50% of births and persistent racial disparities in maternal and neonatal mortality persist, quality measures that address disparities are very much needed. Future research should document what modifications to existing measures or which new measures might help elucidate disparities that exist as well as the improvement initiatives that could be used to reduce morbidity and mortality and the intractable racial disparities in perinatal outcomes.
1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Kirmeyer S. Births: final data for 2004. Natl Vital Stat Rep 2006;55:1–101.
2. Andrews RM. The national hospital bill: the most expensive conditions by payer, 2008. In: HCUP Statistical Brief 59. Rockville (MD): Agency for Healthcare, Research, and Quality; 2008.
3. Asch DA, Nicholson S, Srinivas S, Herrin J, Epstein AJ. Evaluating obstetrical residency programs using patient outcomes. JAMA 2009;302:1277–83.
4. Gregory KD, Fridman M, Shah S, Korst LM. Global measures of quality- and patient safety-related childbirth outcomes: should we monitor adverse or ideal rates? Am J Obstet Gynecol 2009;200:681.e1–7.
5. Srinivas SK, Epstein AJ, Nicholson S, Herrin J, Asch DA. Improvements in US maternal obstetrical outcomes from 1992 to 2006. Med Care 2010;48:487–93.
6. Guendelman S, Thornton D, Gould J, Hosang N. Obstetric complications during labor and delivery: assessing ethnic differences in California. Womens Health Issues 2006;16:189–97.
7. Berg CJ, Callaghan WM, Syverson C, Henderson Z. Pregnancy-related mortality in the United States, 1998 to 2005. Obstet Gynecol 2010;116:1302–9.
8. U.S. Department of Health and Human Services. Healthy People 2010. In: Understanding and improving health. 2nd ed. Washington, DC: U.S. Government Printing Office; 2000.
9. Howell EA, Hebert P, Chatterjee S, Kleinman LC, Chassin MR. Black/white differences in very low birth weight neonatal mortality rates among New York City hospitals. Pediatrics 2008;121:e407–15.
10. Clark SL, Frye DR, Meyers JA, Belfort MA, Dildy GA, Kofford S, et al.. Reduction in elective delivery at <39 weeks of gestation: comparative effectiveness of 3 approaches to change and the impact on neonatal intensive care admission and stillbirth. Am J Obstet Gynecol 2010;203:449.e1–6.
11. Morales LS, Staiger D, Horbar JD, Carpenter J, Kenny M, Geppert J, et al.. Mortality among very low-birthweight infants in hospitals serving minority populations. Am J Public Health 2005;95:2206–12.
12. Geller SE, Cox SM, Callaghan WM, Berg CJ. Morbidity and mortality in pregnancy: laying the groundwork for safe motherhood. Womens Health Issues 2006;16:176–88.
13. Skupski DW, Lowenwirt IP, Weinbaum FI, Brodsky D, Danek M, Eglinton GS. Improving hospital systems for the care of women with major obstetric hemorrhage. Obstet Gynecol 2006;107:977–83.
14. Berg CJ, Harper MA, Atkinson SM, Bell EA, Brown HL, Hage ML, et al.. Preventability of pregnancy-related deaths: results of a state-wide review. Obstet Gynecol 2005;106:1228–34.
18. Phibbs CS, Baker LC, Caughey AB, Danielsen B, Schmitt SK, Phibbs RH. Level and volume of neonatal intensive care and mortality in very-low-birth-weight infants. N Engl J Med 2007;356:2165–75.
20. Rose VL. ACOG urges a cautious approach to vaginal birth after cesarean delivery. Am Fam Physician 1999;60:1245–6.
21. Coleman VH, Erickson K, Schulkin J, Zinberg S, Sachs BP. Vaginal birth after cesarean delivery: practice patterns of obstetrician-gynecologists. J Reprod Med 2005;50:261–6.
22. Declercq E, Young R, Cabral H, Ecker J. Is a rising cesarean delivery rate inevitable? Trends in industrialized countries, 1987 to 2007. Birth 2011;38:99–104.
23. Goldberg J, Hyslop T, Tolosa JE, Sultana C. Racial differences in severe perineal lacerations after vaginal delivery. Am J Obstet Gynecol 2003;188:1063–7.
24. 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.
25. Bryant AS, Washington S, Kuppermann M, Cheng YW, Caughey AB. Quality and equality in obstetric care: racial and ethnic differences in caesarean section delivery rates. Paediatr Perinat Epidemiol 2009;23:454–62.
26. Baicker K, Buckles KS, Chandra A. Geographic variation in the appropriate use of cesarean delivery. Health Aff (Millwood) 2006;25:w355–67.
27. Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, DC; Institute of Medicine; 2001.
28. Howell EA, Stone J, Kleinman LC, Inamdar S, Matseoane S, Chassin MR. Approaching NIH guideline recommended care for maternal-infant health: clinical failures to use recommended antenatal corticosteroids. Matern Child Health J 2010;14:430–6.
29. Lee HC, Lyndon A, Blumenfeld YJ, Dudley RA, Gould JB. Antenatal steroid administration for premature neonates in California. Obstet Gynecol 2011;117:603–9.
31. Main EK. New perinatal quality measures from the National Quality Forum, the Joint Commission and the Leapfrog Group. Curr Opin Obstet Gynecol 2009;21:532–40.
32. Korst LM, Gregory KD, Lu MC, Reyes C, Hobel CJ, Chavez GF. A Framework for the Development of maternal quality of care indicators. Matern Child Health J 2005;9:317–41.
33. Cousens S, Blencowe H, Gravett M, Lawn JE. Antibiotics for pre-term pre-labour rupture of membranes: prevention of neonatal deaths due to complications of pre-term birth and infection. Int J Epidemiol 2010;39(suppl 1):i134–43.
34. Mann S, Pratt S, Gluck P, Nielsen P, Risser D, Greenberg P, et al.. Assessing quality obstetrical care: development of standardized measures. Jt Comm J Qual Patient Saf 2006;32:497–505.
35. National Quality Forum. Healthcare disparities and cultural competency consensus standards: disparities-sensitive measure assessment. Washingto, DC: National Quality Forum; 2012.
36. Shen TT, DeFranco EA, Stamilio DM, Chang JJ, Muglia LJ. A population-based study of race-specific risk for preterm premature rupture of membranes. Am J Obstet Gynecol 2008;199:373.e1–7.
37. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008;371:75–84.
38. Lang CT, King JC. Maternal mortality in the United States. Best Pract Res Clin Obstet Gynaecol 2008;22:517–31.
39. Deneux-Tharaux C, Berg C, Bouvier-Colle MH, Gissler M, Harper M, Nannini A, et al.. Underreporting of pregnancy-related mortality in the United States and Europe. Obstet Gynecol 2005;106:684–92.
© 2013 The American College of Obstetricians and Gynecologists