Abstract
Objective
To determine the incidence and risk factors for maternal morbidity during childbirth hospitalization.
Methods
Maternal morbidities were determined using ICD9-CM and vital records codes from linked hospital discharge and vital records data for 1,572,909 singleton births in California, 2005-2007. Sociodemographic, obstetric, and hospital volume risk factors were estimated using mixed effects logistic regression models.
Results
The maternal morbidity rate was 241/1000 births. The most common morbidities were episiotomy, pelvic trauma, maternal infection, postpartum hemorrhage, and severe laceration. Preeclampsia (Adjusted Odds Ratio [AOR] 2.96; 95% CI 2.8,3.13), maternal age over 35 years, (AOR 1.92; 1.79,2.06), vaginal birth after cesarean, (AOR 1.81; 1.47,2.23), and repeat cesarean birth (AOR 1.99; 1.87,2.12) conferred the highest odds of severe morbidity. Non-white women were more likely to suffer morbidity.
Conclusions
Nearly one in four California women experienced complications during childbirth hospitalization. Significant health disparities in maternal childbirth outcomes persist in the United States.
Keywords: Childbirth hospitalization, maternal morbidity, risk factors
INTRODUCTION
Maternal health during childbirth is an international priority presently under increased focus in the United States (US) in light of escalating overall maternal mortality and racial disparities in maternal mortality.[1-4] Recent death certificate coding changes in the US only partially explain the observed increase in maternal mortality.[2,5] Furthermore, many more women and families are affected by maternal morbidity during childbirth; severe maternal morbidity often may represent a “near miss”[4] with significant utility as an indicator of maternal health and clinical quality.
Contemporary reports of maternal morbidity during childbirth hospitalization in the US are either not population based, or use a sampling strategy of administrative survey data to estimate rates.[6-8] Administrative data often do not accurately reflect patient race/ethnicity, particularly for non-Whites, and may underestimate racial disparities.[9-11] Previous studies have not assessed hospital volume, a factor implicated in morbidity and mortality for certain surgeries and conditions such as neonatal mortality for premature infants.[12,13]
The purpose of this study was to evaluate the incidence of maternal morbidity during childbirth hospitalization in a large US population-based cohort of singleton births and assess risk factors for morbidity, including race/ethnicity and hospital volume. By combining birth certificate and administrative data, we incorporated variables that may have not been captured using hospital administrative data alone.
MATERIALS & METHODS
We evaluated maternal morbidity during childbirth hospitalization using data linked from California Vital Statistics records and state-wide hospital discharge data from the Office of State Wide Planning and Development (OSHPD). This linked data set contains maternal education level, race, ethnicity, insurance status, estimate of prenatal care, mode of birth, and diagnosis and procedure codes from the birth hospitalization for virtually all inpatient live births in California. The study period included all births in 2005, 2006, and 2007, which were the most recent years with available linked records at the time of the data request. We limited analysis to singleton births, due to the very different risk profile for the small fraction of births from multi-fetal pregnancy.
Data were linked using a probabilistic matching algorithm for large public health data sets as described by Jaro.[14,15] Linkage between Vital Statistics and hospital discharge data was established for greater than 96% of records. Linkage variables included baby’s birth date, birth hospital, delivery mode, mother’s birth date, and other maternal information. We excluded the records of 31,884 births (2.0% of total records) that occurred in military hospitals and freestanding birth centers that do not report hospital discharge data. Hospital identifiers were masked to protect privacy. The California Perinatal Quality Care Collaborative developed and executed the linkage strategy under grant support from the March of Dimes. Institutional Review Boards at Stanford University and University of California San Francisco approved the study.
We identified maternal morbidity using International Classification of Diseases, 9th Revision (ICD-9-CM) diagnosis and procedure codes[7,8,16-20] supplemented with birth certificate data.[21] (Supplementary Digital Content S1) After determining crude rates of morbidities for each year we classified women as having “pelvic floor morbidity,” “non-pelvic floor morbidity,” and “severe morbidity.” Mothers were considered positive for “pelvic floor morbidity” if they had ICD-9 codes for episiotomy, 3rd or 4th degree laceration, vulvar or perineal hematoma or other trauma; or indication of 3rd or 4th degree laceration on the birth certificate. Mothers were considered positive for “non-pelvic floor morbidity” if they had ICD-9 diagnosis codes for postpartum hemorrhage, maternal infection, pelvic trauma, deep vein thrombosis, wound complication, pulmonary embolus, uterine rupture, anesthetic complication, severe anesthetic complication, respiratory failure, maternal shock, maternal sepsis, heart failure, renal failure, puerperal cerebral vascular accident (CVA), or disseminated intravascular coagulation (DIC), or ICD-9 procedure codes for blood transfusion, dilatation and curettage, hysterectomy, or ventilation. Indications of these maternal morbidities from birth certificate fields were also included where applicable. Discharge dispositions of death or birth certificate indication of maternal death during labor and delivery were also included (n = 118).
We evaluated pelvic floor morbidity because severe lacerations can have significant negative consequences for women, and episiotomy increases the risk of 3rd and 4th degree laceration.[22] There is some controversy about whether cesarean birth should be considered a morbidity in contemporary obstetrics.[6,17] We calculated morbidity rates including and excluding cesarean birth. When included, cesarean accounted for 53% of overall morbidity. We determined that mode of birth is best considered a risk factor for morbidity, and excluded cesarean from our final composite indicators.
We defined “severe morbidity” following the methods of Kuklina[7] and Callaghan.[8] Cases with a length of stay ≥90th percentile for mode of birth and an ICD-9 or birth certificate code for severe postpartum hemorrhage, maternal sepsis, deep vein thrombosis, pulmonary embolism, uterine rupture, respiratory failure, heart failure, puerperal CVA, severe anesthetic complication, maternal shock, DIC, or renal failure, were classified as severe. When ICD-9 or birth certificate codes indicated the woman had hysterectomy, ventilation, unplanned return to operating room, transfer to intensive care, or maternal death, these events were classified as severe morbidity regardless of length of stay.
We conducted unadjusted and multivariable analyses to examine associations between predictor variables and maternal morbidity. Predictor variables included year of birth, maternal socio-demographic factors, selected maternal co-morbidities, mode of birth, and hospital level factors. For unadjusted analyses, we used chi-square test for group comparisons with significance level p<0.05. Socio-demographic factors were included based on reports of substantial socio-demographic disparities in maternal health outcomes.[1,22,23] Maternal age, education, race, ethnicity, insurance status, and adequacy of prenatal care[24] were determined from birth certificate data. Maternal co-morbidities evaluated were diabetes, preeclampsia/eclampsia, and chronic hypertension, determined by ICD-9 codes. These co-morbidities, especially hypertensive disorders, reasonably could be classified as either risk factors or maternal morbidities. We chose to evaluate them as risk factors for birth-related morbidity.
We categorized hospital volume by number of annual births in each facility during each year of the study period: <1000, 1000-3000, and >3000.[2] To account for clustering of births within hospitals, we used mixed effects logistic regression models for the multivariable analyses, with individual hospitals as a random effect, and predictor variables as fixed effects. Adjusted odds ratios (AOR) with 95% confidence intervals (CI) were estimated using PROC GLIMMIX in SAS 9.2 (SAS Institute, Cary, NC).
RESULTS
During 2005 to 2007, there were 1,572,909 live singleton births in California with linked records. Socio-demographic and pregnancy characteristics are reported in Table 1. The overall rate of maternal morbidity excluding cesarean was 241/1000 births. The morbidity rate declined 11% over the study period from 256/1000 in 2005 to 241/1000 in 2006 and 227/1000 in 2007 (p<.001). The most common morbidities were episiotomy, pelvic trauma, maternal infection, postpartum hemorrhage, and 3rd or 4th degree laceration, all of which were slightly lower in 2007 than in 2005. (Table 2) While postpartum hemorrhage overall was 9% lower in 2007 than in 2005, blood transfusion increased by 21% and severe postpartum hemorrhage increased by 10% (p<0.001). Although rare, hysterectomy increased by 27% over the study period (p=0.004). While pelvic floor morbidity and non-pelvic morbidity were 14% and 8% lower in 2007, respectively (p<0.001), severe morbidity was 9% higher in 2007 than in 2005 (p=0.02). Unadjusted relationships between potential risk factors and morbidity are shown in supplementary digital content (S2). Adjusted relationships are shown in Table 3.
Table 1. Maternal Characteristics, by Year.
| 2005 % |
2006 % |
2007 % |
Total | |
|---|---|---|---|---|
| Education a | ||||
| Less than High School | 28.4 | 38 | 28.2 | 481498 |
| High School | 28.3 | 18.1 | 27.3 | 374847 |
| Some college | 31.9 | 36.4 | 37 | 536326 |
| Some post-graduate | 11.4 | 7.5 | 7.6 | 134259 |
| Maternal age | ||||
| <18 | 3.3 | 3.3 | 3.3 | 51653 |
| 18-25 | 34.4 | 34.7 | 34.4 | 542198 |
| 26-35 | 49.4 | 48.9 | 49.1 | 772772 |
| >35 | 13 | 13.1 | 13.2 | 206285 |
| Race | ||||
| White | 78.9 | 78.4 | 77.3 | 1229313 |
| Black | 5.2 | 5.3 | 5.2 | 82760 |
| American Indian/Alaskan Native | 0.5 | 0.5 | 0.5 | 7586 |
| Asian | 11.4 | 11.4 | 11.9 | 181534 |
| Hawaiian/Pacific Islander | 0.5 | 0.5 | 0.5 | 7298 |
| Other/Multirace | 3.5 | 4 | 4.7 | 64418 |
| Ethnicity | ||||
| Non-Hispanic | 46.5 | 45.7 | 45.5 | 722099 |
| Hispanic | 52.2 | 52.9 | 53.2 | 830122 |
| Unknown | 1.2 | 1.4 | 1.3 | 20688 |
| Payor | ||||
| None/self | 2.9 | 2.8 | 2.8 | 44361 |
| Medi-cal/government | 49.5 | 50 | 50.5 | 784487 |
| Private | 47.7 | 47.2 | 46.7 | 739983 |
| Parity | ||||
| Primiparous | 39.2 | 39.5 | 39.8 | 620782 |
| Multiparous | 60.8 | 60.5 | 60.2 | 951130 |
| Prenatal care b | ||||
| Inadequate | 19.5 | 19.4 | 21.6 | 317364 |
| Adequate | 80.5 | 80.6 | 78.4 | 1255545 |
| Co-Morbidities | ||||
| Diabetes | ||||
| No | 93.6 | 93.2 | 92.6 | 1464614 |
| Yes | 6.4 | 6.8 | 7.4 | 108295 |
| Preeclampsia | ||||
| No | 94.2 | 94 | 94 | 1479681 |
| Yes | 5.8 | 6.0 | 6.0 | 93228 |
| Chronic Hypertension | ||||
| No | 99.1 | 99 | 99 | 1557890 |
| Yes | 0.9 | 1 | 1 | 15019 |
| Mode of Birth | ||||
| Vaginal | 70 | 69.4 | 68.7 | 1091172 |
| VBAC | 0.8 | 0.7 | 0.7 | 11477 |
| Primary Cesarean | 16.3 | 16.6 | 16.8 | 260363 |
| Repeat Cesarean | 12.9 | 13.3 | 13.8 | 209897 |
Abbreviation: VBAC, Vaginal Birth after Cesarean
Number of years of education were reported prior to 2006, rather than degree information. In 2006-2007 there was a transition of reporting education levels were according to new 2003 US standard birth certificate specifications. Categories reported here were developed to account for education level across the three years of data.
Kotelchuck Adequacy of Prenatal Care Utilization Index[24]
Table 2. Births and Morbidity Rates, by Year.
| 2005 | 2006 | 2007 | Total Period | |||
|---|---|---|---|---|---|---|
| Number of singleton births | 512869 | 527632 | 532408 | 1572909 | ||
|
| ||||||
| Morbidities per 1000 births | Events | |||||
| Composite Morbidities | 2005 | 2006 | 2007 | Total Period | n | |
|
| ||||||
| Pelvic floor morbiditya | 169 | 153 | 146 | 156 | 245095 | |
| Non-pelvic floor morbidityb | 99 | 100 | 91 | 97 | 152089 | |
| Any pelvic or non-pelvic floor morbidity |
256 | 241 | 227 | 241 | 379345 | |
| Cesarean birth counted as an additional morbidity |
516 | 508 | 504 | 516 | 800836 | |
| Severe morbidityc | 5.6 | 5.8 | 6.1 | 5.8 | 9190 | |
| Individual Morbidities | 2005 | 2006 | 2007 | Total Period | n |
% of women
with non-pelvic floor morbidity b |
|
| ||||||
| Episiotomy | 148 | 133 | 125 | 136 | 213229 | - |
| Pelvic trauma | 33.1 | 34.1 | 29.3 | 32.2 | 50593 | 33.3 |
| Maternal infection | 28.4 | 28.0 | 26.3 | 27.5 | 43312 | 28.5 |
| Postpartum hemorrhage | 28.3 | 28.3 | 25.7 | 27.4 | 43124 | 28.4 |
| 3rd or 4th degree laceration | 26.2 | 23.8 | 22.0 | 24.0 | 37759 | - |
| Blood transfusion | 5.7 | 6.4 | 6.9 | 6.3 | 9961 | 6.6 |
| Severe postpartum hemorrhage | 4.8 | 4.9 | 5.3 | 5.0 | 7854 | 5.2 |
| Anesthetic complication | 3.6 | 3.6 | 3.3 | 3.5 | 5548 | 3.6 |
| Wound complication | 3.4 | 3.7 | 3.3 | 3.5 | 5493 | 3.6 |
| Heart failure | 3.8 | 2.9 | 2.7 | 3.1 | 4897 | 3.2 |
| Other pelvic floor | 3.3 | 3.5 | 6.1 | 4.3 | 6766 | - |
| Dilatation & curretage | 2.7 | 2.5 | 2.7 | 2.6 | 4154 | 2.7 |
| Disseminated intravascular coagulation |
1.2 | 1.1 | 1.3 | 1.2 | 1890 | 1.2 |
| Uterine rupture | 1.0 | 1.0 | 1.1 | 1.0 | 1628 | 1.1 |
| Maternal Sepsis | 0.9 | 0.9 | 1 | 0.9 | 1486 | 1 |
| Hysterectomy | 0.7 | 0.8 | 0.8 | 0.8 | 1195 | 0.8 |
| Respiratory failure | 0.6 | 0.6 | 0.7 | 0.6 | 1006 | 0.7 |
| Ventilation | 0.5 | 0.5 | 0.5 | 0.5 | 761 | 0.5 |
| Maternal shock | 0.4 | 0.4 | 0.5 | 0.4 | 702 | 0.5 |
| Deep vein thrombosis | 0.3 | 0.3 | 0.3 | 0.3 | 502 | 0.3 |
| Severe anesthetic complication | 0.3 | 0.3 | 0.3 | 0.3 | 476 | 0.3 |
| Renal failure | 0.3 | 0.2 | 0.3 | 0.3 | 407 | 0.3 |
| Puerperal cardiovascular accident |
0.2 | 0.2 | 0.2 | 0.2 | 313 | 0.2 |
| Pulmonary embolus | 0.2 | 0.1 | 0.2 | 0.2 | 264 | 0.2 |
| Maternal death during childbirth hospitalization |
0.08 | 0.09 | 0.05 | 0.07 | 118 | 0.08 |
Pelvic floor morbidity includes episiotomy, 3-4th degree laceration, and vulvar or perineal hematoma or other trauma to vulva or perineum.
Cesarean birth without other morbidity not counted in “any morbidity” composite. Total of individual morbidities greater than 100% as some women had more than one complication.
Severe morbidity = ICD9-CM or birth certificate codes present for hysterectomy, ventilation, unplanned return to operating room, transfer to intensive care, or maternal death, OR an ICD9-CM or birth certificate code for severe postpartum hemorrhage, maternal sepsis, deep vein thrombosis, pulmonary embolism, uterine rupture, respiratory failure, heart failure, hysterectomy, puerperal CVA, severe anesthetic complication, maternal shock, DIC, or renal failure and length of stay ≥90th percentile for mode of birth.
Table 3. Adjusted Maternal Morbidity, by Risk Factors.
| Adjusted Odds Ratioa (95% Confidence Interval) | |||
|---|---|---|---|
| Predictor Variable | Pelvic floor morbidity |
Non-pelvic floor morbidity |
Severe morbidity |
| Education | |||
| Less than HS | 0.97 (0.95, 0.98) | 0.99 (0.98, 1.01) | 1.05 (0.99, 1.12) |
| High school | 1 reference | 1 reference | 1 Reference |
| Some college | 1.1 (1.08, 1.12) | 0.98 (0.96, 0.99) | 0.95 (0.89,1.01) |
| Some post-graduate | 1.24 (1.21, 1.27) | 0.92 (0.9, 0.94) | 0.81 (0.74, 0.89) |
| Maternal Age | |||
| <18 | 1.12 (1.09, 1.15) | 1 (0.97, 1.03) | 1.2 (1.07, 1.34) |
| 18-25 | 1 reference | 1 reference | 1 Reference |
| 26-35 | 1.04 (1.03, 1.06) | 0.94 (0.93, 0.95) | 1.22 (1.15, 1.29) |
| >35 | 1.16 (1.13, 1.18) | 1 (0.98, 1.02) | 1.92 (1.79, 2.06) |
| Race | |||
| White | 1 reference | 1 reference | 1 reference |
| Black | 0.59 (0.57, 0.60) | 1.11 (1.08, 1.14) | 1.47 (1.34, 1.60) |
| American Indian/Alaskan Native | 0.76 (0.70, 0.82) | 1.12 (1.04, 1.21) | 1.30 (0.99, 1.71) |
| Asian | 1.60 (1.57, 1.62) | 1.18 (1.16, 1.2) | 1.16 (1.08, 1.26) |
| Hawaiian/Pacific Islander | 0.81 (0.74, 0.88) | 1.37 (1.27, 1.47) | 1.19 (0.90, 1.59) |
| Other/Multi-race | 0.83 (0.80, 0.85) | 1.04 (1.01, 1.08) | 1.06 (0.94, 1.21) |
| Ethnicity | |||
| Hispanic | 0.83 (0.81, 0.84) | 1.15 (1.14, 1.17) | 1.11 (1.04, 1.18) |
| Non-Hispanic | 1 reference | 1 reference | 1 Reference |
| Unknown | 0.96 (0.85, 1.09) | 1.08 (0.95, 1.22) | 1.24 (0.80, 1.93) |
| Payor | |||
| Medi-Cal/Govt/Other | 0.97 (0.96, 0.99) | 1.06 (1.04, 1.08) | 1.24 (1.16, 1.32) |
| None/Self | 1.03 (0.997, 1.06) | 1.13 (1.09, 1.17) | 1.29 (1.12, 1.48) |
| Private | 1 Reference | 1 reference | 1 Reference |
| Parity | |||
| Primiparous | 1 Reference | 1 reference | 1 reference |
| Multiparous | 0.21 (0.21, 0.22) | 0.57 (0.56, 0.58) | 0.64 (0.61, 0.68) |
| Prenatal care b | |||
| Inadequate | 1.0 (0.99, 1.02) | 1.03 (1.01, 1.04) | 1.03 (0.98, 1.09) |
| Adequate | 1 Reference | 1 reference | 1 Reference |
| Co-Morbidities | |||
| Diabetes | |||
| Yes | 0.98 (0.96, 1.00) | 0.94 (0.92, 0.96) | 1.08 (1.01, 1.16) |
| No | 1 reference | 1 reference | 1 Reference |
| Preeclampsia | |||
| Yes | 0.93 (0.91, 0.96) | 1.27 (1.24, 1.3) | 2.96 (2.8, 3.13) |
| No | 1 reference | 1 reference | 1 Reference |
| Chronic Hypertension | |||
| Yes | 0.77 (0.72, 0.84) | 1.1 (1.05, 1.16) | 1.45 (1.29, 1.63) |
| No | 1 reference | 1 reference | 1 Reference |
| Hospital volume | |||
| < 1000 | 0.78 (0.7, 0.86) | 0.97 (0.89, 1.05) | 1.27 (1.06, 1.52) |
| 1000-3000 | 0.93 (0.91, 0.96) | 1.05 (1.01, 1.1) | 1.15 (1.02, 1.31) |
| > 3000 | 1 reference | 1 reference | 1 Reference |
| Mode of Birth | |||
| Vaginal | 1 reference | 1 reference | 1 Reference |
| VBAC | 2.16 (2.04, 2.27) | 1.51 (1.42, 1.59) | 1.81 (1.47, 2.23) |
| Primary Cesarean | 0.004 (0.003, 0.004) | 1.36 (1.34, 1.38) | 1.59 (1.51, 1.67) |
| Repeat Cesarean | 0.004 (0.003, 0.005) | 0.88 (0.86, 0.9) | 1.99 (1.87, 2.12) |
Abbreviation: VBAC, Vaginal Birth after Cesarean.
Mixed effects logistic regression model with individual hospital as random effect and predictor variables as fixed effects.
Kotelchuck Adequacy of Prenatal Care Utilization Index[24]
Pelvic Floor Morbidity
The overall rate of pelvic floor morbidity was 156/1000 births. In unadjusted analyses, women had higher rates of pelvic floor morbidity if they were under age 18 (279/1000), age 18-25 (181/1000), Asian (238/1000), Non-Hispanic (165/1000), uninsured (191/1000), more educated (some college 164/1000; some post-graduate 183/1000) or primiparous (255/1000). Women had higher rates of pelvic floor morbidity with vaginal birth (222/1000) than with vaginal birth after cesarean (VBAC) (172/1000), or if they gave birth in a higher volume hospital (>3000 births 162/1000). Inadequate prenatal care was associated with lower pelvic floor morbidity (141 vs 160/1000). Women who reported being Black, American Indian/Alaskan Native, and Hawaiian/Pacific Islander or who had hypertensive disorders had lower rates of pelvic floor morbidity (S2).
In the multivariable analysis VBAC remained strongly associated with pelvic floor morbidity (Table 3). Asian race, higher education, and age (<18 and >36 years of age) were also associated with increased odds of pelvic floor morbidity. Odds for pelvic floor morbidity were lower for women who gave birth at low volume hospitals or had chronic hypertension, were multiparous, or were Black, American Indian/Alaskan Native, Hawaiian/Pacific Islander, or Hispanic. Lack of insurance was not significantly associated with pelvic floor morbidity.
Non-Pelvic Floor Morbidity
The overall rate of non-pelvic floor morbidity was 97/1000 births. In unadjusted analysis, women under age 26 had higher rates of non-pelvic floor morbidity (<18, 128/1000; 18-25 106/1000), as did women who were primiparous (133/1000), had preeclampsia (144/1000) or chronic hypertension (137/1000), or gave birth by VBAC or primary cesarean (130/1000 and 135/1000, respectively). Hawaiian/Pacific Islander (126/1000) or other non-White race, and inadequate prenatal care were also risk factors for non-pelvic floor morbidity (S2). Mothers with Medi-Cal/government insurance had lower rates of non-pelvic floor morbidity (93/1000) compared to mothers with private insurance (101/1000), no insurance or self-pay (98/1000). Mothers who gave birth by repeat cesarean also had lower non-pelvic floor morbidity rates. In multivariable analysis Hawaiian/Pacific Islander and other non-white races, Hispanic ethnicity, non-private or no insurance, hypertensive disorders, VBAC, and primary cesarean birth were associated with higher non-pelvic floor morbidity (Table 3). Repeat cesarean section births, mulitparity, and postgraduate education were associated with lower rates of non-pelvic floor morbidity.
Severe Morbidity
The overall rate of severe morbidity during the study period was 5.8/1000 births. Severe morbidity was higher in women who had the co-morbidities of chronic hypertension (24/1000) and preeclampsia (20/1000) than those who did not (5.7/1000 and 4.9/1000, respectively), and was elevated in women with diabetes (9.3/1000 vs. 5.6/1000) (S2). Compared to vaginal birth, the rate of severe morbidity was higher in repeat cesarean (8.1/1000) and VBAC (8.5/1000), and more than doubled in primary cesarean birth (9.4/1000).
In crude analysis for severe morbidity, Black, American Indian/Alaskan Native, and Hawaiian/Pacific Islander women had higher rates of severe morbidity (9.4, 7.4 and 6.9/1000, respectively) compared to White women (5.5/1000), with Black women bearing a high burden of serious morbidities (Supplementary digital content S3). Women under 18 and over 35 had higher rates of severe morbidity (7.3 and 9.3) than women 18-25 (5.0/1000). Women who gave birth in higher volume hospitals had lower rates of severe morbidity.
In the multivariable analysis, post-graduate education and multiparity were associated with lower severe morbidity (Table 3). Odds for severe morbidity were increased in women over 35. Odds for severe morbidity were significantly higher in Black, Asian, and Hispanic women and women without private insurance, giving birth in hospitals with <3000 annual births, and giving birth by VBAC or cesarean section. Preeclampsia, maternal age over 35 years, VBAC and cesarean birth conferred the highest odds of severe morbidity.
DISCUSSION
Nearly one in four California women giving birth during the study period experienced morbidity during their childbirth hospitalization. As would be expected, hypertensive disorders were associated with increased risk of non-pelvic and severe morbidity, and preeclampsia was the strongest risk factor for severe morbidity. Black women had increased risk for severe morbidity. Women of Asian race had increased risk for all types of morbidity, though the risk was highest for pelvic floor morbidity. Women of all other non-White races had lower risk of pelvic floor morbidity but increased risk of non-pelvic floor morbidity. While cesarean birth had virtually no risk for pelvic floor involvement, primary cesarean birth conferred increased odds of both non-pelvic floor and severe morbidity; repeat cesarean birth was also significantly associated with severe morbidity. Women who gave birth by VBAC accounted for only 1% of births and had a morbidity pattern similar to Asian women, with increased risk for all types of morbidity and highest risk for pelvic floor morbidity.
Declining overall morbidity during the study period stands in contrast to reports of increasing maternal mortality;[4] however the rate of severe morbidity increased. In our study 379,345 women (241/1000 births) had morbidity. A substantial portion of morbidity was due to episiotomy, 3rd or 4th degree laceration, pelvic trauma, and postpartum hemorrhage, suggesting that much of this morbidity may be preventable.
An overarching goal of Healthy People 2020 is to reduce health disparities.[23] Substantial differences in pelvic floor morbidity in Asian women are consistent with a recent study of perineal trauma that suggests risk factors for perineal trauma may be mostly non-modifiable.[22] However, others have reported successful interventions for lowering severe laceration rates,[26] suggesting that like central line infections, perineal trauma may be more preventable than previously believed. Higher risk for severe morbidity in Black women is consistent with their much higher rates of maternal mortality in the US population.[1,2] This pressing public health problem has persisted over the past five decades.[1,2,27] We found that among women with childbirth morbidities, serious morbidities such as pulmonary embolus, respiratory failure, shock, and death were more common in Black mothers than in other racial groups. Focused exploration of the reason for conditions disproportionally affecting Black mothers could aid efforts to reduce maternal mortality in Black women.
Hospital volume was a risk factor for morbidity. Our finding of greater risk for severe morbidity in smaller volume hospitals is consistent with literature on decreased morbidity with higher volumes for certain conditions.[12,13] This may be due to having fewer resources to address serious complications when they occur without time for transfer to a higher level of care, and is a concerning finding considering that in 2008, 58% of US hospitals providing obstetrical care had a volume of less than 1000 annual births.[28] Circumstances leading to severe morbidity and requiring complex coordinated perinatal care may be similar to some surgical conditions in which lower volume is associated with increased mortality.[13] Unfortunately interventions to improve care coordination are challenging to replicate in disparate contexts, and measuring their efficacy is difficult.
The significant association between primary cesarean and morbidity after controlling for demographic and obstetric risk factors is similarly quite concerning. Some factors that may have led to a cesarean birth could also lead to morbidity, and our study design cannot determine causality. However, contrary to public health goals,[3] the primary cesarean rate is increasing, without demonstrated benefit to mother or infant.[29,30] For our primary analysis, we classified cesarean birth as a predictor rather than an outcome given its prevalence[6] and concerns that cesarean birth may best be considered a process rather than an outcome indicator.[17,25] If cesarean births were included they would account for more than half of childbirth morbidity and the overall morbidity rate would be 52%, substantially higher than the estimate of 48.5% for a national sample of birth hospitalizations in 2001-2005.[6] The association between primary cesarean and severe morbidity, particularly, lends urgency to efforts to reevaluate practices regarding primary cesarean. This urgency is reinforced by the association of severe morbidity with both VBAC and repeat cesarean, modes of birth that can only be avoided through prevention of primary cesarean.
We acknowledge the inherent limitations of hospital discharge data and vital statistics data including the time lag between the provision of services and data analysis. If there is a systematic bias in the coding of administrative data, it is likely to be toward under-reporting of some morbidities.[16,19,20] There is also likely to be some misclassification of co-morbidities when assessed by ICD-9 codes. We note that variation in reported education levels across the three years in our study is greater than expected, and there may have been misclassification of education levels in 2006 due to changes in California reporting requirements [21]. Furthermore, the social determinants of maternal health are complex. Education level and insurance status represent a fairly crude approximation of socioeconomic status. However, compared to hospital discharge data alone, the linkage of vital statistics data with administrative discharge data provides an improved level of analysis of socio-demographic risk, and is a strength of this study. While there are limitations to analyzing the data from a single state, particularly in relation to racial and ethnic population representativeness, approximately one in seven US births occur in California, and our three years of data provide a cohort size that allows for analysis of rare outcomes. We were not able to evaluate the contribution of body mass index, behavioral risk factors, or labor interventions to maternal morbidity, and these are important areas of investigation. California began reporting mother’s pre-pregnancy weight, delivery weight, height, and smoking status in electronically collected vital records data in 2007.
Our mixed effects model and volume analysis account for hospital characteristics to some degree, but do not illuminate specifically what is most important at the hospital level. Continued research is needed to better understand the range of factors contributing to observed racial and ethnic differences in maternal morbidity and mortality. While some disparities may be attributable to biologic factors that are difficult to control at the time of birth, some could potentially be addressed through targeted interventions to reduce specific morbidities.
In summary, maternal morbidity at the time of birth is common, and while affecting women from all socioeconomic strata, it is more common and more consequential in specific high-risk groups. Our findings regarding the association between severe morbidity and low birth volume hospitals and between primary cesarean births and non-pelvic and severe morbidity suggest opportunities for improving care. Establishing and simulating the implementation of emergency protocols could help facilities with more limited resources respond quickly and effectively to serious complications when they arise. Reducing maternal morbidity and reducing cesarean births among low-risk women are public health priorities.[3] Renewed attention to reducing use of cesarean birth could have a synergistic effect on maternal morbidity.
Supplementary Material
This table lists the ICD-9-CM and Vital Statistics Codes used to calculate individual morbidities.
This table shows the unadjusted relationships between potential risk factors and morbidity.
Figure shows percent of serious morbidity cases in relation to the proportion of the population studied, for White, Black, Asian, and Hawaiian/Pacific Islander women.
Abbreviations: pph, postpartum hemorrhage; DVT, deep vein thrombosis; Severe anesthetic, severe anesthetic complication; DIC, disseminated intravascular coagulation; CVA, cerebrovascular accident
Acknowledgments
We thank Ms. Chengshi Jin and Dr. Lance Pollack for programming support.
This study was supported by NIH/NCRR/OD UCSF-CTSI Grant Number KL2 RR024130. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. The linkage of OSHPD and California Vital Statistics datasets was made possible through a grant from the March of Dimes. Sponsor role: none.
Footnotes
Declaration of Interests:
The authors have no conflicts of interest to report.
Contributor Information
Dr. Audrey Lyndon, Department of Family Health Care Nursing, University of California, San Francisco; California Maternal Quality Care Collaborative, Stanford, California.
Dr. Henry C. Lee, Department of Pediatrics, University of California, San Francisco; California Perinatal Quality Care Collaborative, Stanford, California.
Dr. William M. Gilbert, Sutter Health Sacramento-Sierra Region, Sacramento, California; California Maternal Quality Care Collaborative, Stanford, California.
Dr. Jeffrey B. Gould, Department of Pediatrics, Stanford University, Palo Alto, California; California Perinatal Quality Care Collaborative, Stanford, California.
Dr. Kathryn A. Lee, Department of Family Health Care Nursing, University of California, San Francisco.
References
- 1.Xu JQ, Kochanek KD, Murphy SL, Tejada-Vera B. National vital statistics reports. 19. Vol. 58. National Center for Health Statistics; Hyattsville, MD: 2010. Deaths: Final data for 2007. http://www.cdc.gov/nchs/data/nvsr/nvsr58/nvsr58_19.pdf. [PubMed] [Google Scholar]
- 2.The California Pregnancy-Associated Mortality Review: Report from 2002 and 2003 Maternal death reviews. California Department of Public Health, Maternal Child and Adolescent Division; Sacramento, CA: 2011. [Google Scholar]
- 3.US Department of Health and Human Services . Healthy People 2020 Maternal, Infant, and Child Health Objectives. [Accessed July 12, 2011]. http://www.healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?topicId=26. [Google Scholar]
- 4.Berg CJ, Callaghan WM, Syverson C, Henderson Z. Pregnancy-related mortality in the United States, 1998 to 2005. Obstet Gynecol. 2010;116(6):1302–1309. doi: 10.1097/AOG.0b013e3181fdfb11. doi: 10.1097/AOG.0b013e3181fdfb11. [DOI] [PubMed] [Google Scholar]
- 5.Hoyert DL. Maternal mortality and related concepts. National Center for Health Statistics. Vital Health Stat. 3(33):2007. [PubMed] [Google Scholar]
- 6.Berg CJ, Mackay AP, Qin C, Callaghan WM. Overview of maternal morbidity during hospitalization for labor and delivery in the United States: 1993-1997 and 2001-2005. Obstet Gynecol. 2009 May;113(5):1075–1081. doi: 10.1097/AOG.0b013e3181a09fc0. [DOI] [PubMed] [Google Scholar]
- 7.Kuklina EV, Meikle SF, Jamieson DJ, Whiteman MK, Barfield WD, Hillis SD, Posner SF. Severe obstetric morbidity in the United States: 1998-2005. Obstet Gynecol. 2009 Feb;113(2 Pt 1):293–299. doi: 10.1097/AOG.0b013e3181954e5b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Callaghan WM, Mackay AP, Berg CJ. Identification of severe maternal morbidity during delivery hospitalizations, United States, 1991-2003. Am J Obstet Gynecol. 2008 Aug;199(2):133 e131–138. doi: 10.1016/j.ajog.2007.12.020. [DOI] [PubMed] [Google Scholar]
- 9.Burrows LJ, Meyn LA, Weber AM. Maternal morbidity associated with vaginal versus cesarean delivery. Obstet Gynecol. 2004 May;103(5 Pt 1):907–912. doi: 10.1097/01.AOG.0000124568.71597.ce. [DOI] [PubMed] [Google Scholar]
- 10.Boehmer U, Kressin NR, Berlowitz DR, Christiansen CL, Kazis LE, Jones JA. Self-reported vs administrative race/ethnicity data and study results. Am J Public Health. 2002 Sep;92(9):1471–1472. doi: 10.2105/ajph.92.9.1471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kressin NR, Chang BH, Hendricks A, Kazis LE. Agreement between administrative data and patients’ self-reports of race/ethnicity. Am J Public Health. 2003 Oct;93(10):1734–1739. doi: 10.2105/ajph.93.10.1734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chung JH, Phibbs CS, Boscardin WJ, Kominski GF, Ortega AN, Needleman J. The effect of neonatal intensive care level and hospital volume on mortality of very low birth weight infants. Med Care. 2010 Jul;48(7):635–644. doi: 10.1097/MLR.0b013e3181dbe887. [DOI] [PubMed] [Google Scholar]
- 13.Birkmeyer JD, Siewers AE, Finlayson EV, Stukel TA, Lucas FL, Batista I, Welch HG, Wennberg DE. Hospital volume and surgical mortality in the United States. N Engl J Med. 2002 Apr 11;346(15):1128–1137. doi: 10.1056/NEJMsa012337. [DOI] [PubMed] [Google Scholar]
- 14.Jaro MA. Probabilistic linkage of large public health data files. Stat Med. 1995;14(5-7):491–498. doi: 10.1002/sim.4780140510. [DOI] [PubMed] [Google Scholar]
- 15.Herrchen B, Gould JB, Nesbitt TS. Vital statistics linked birth/infant death and hospital discharge record linkage for epidemiological studies. Comput Biomed Res. 1997 Aug;30(4):290–305. doi: 10.1006/cbmr.1997.1448. [DOI] [PubMed] [Google Scholar]
- 16.Lydon-Rochelle MT, Holt VL, Nelson JC, Cardenas V, Gardella C, Easterling TR, Callaghan WM. Accuracy of reporting maternal in-hospital diagnoses and intrapartum procedures in Washington State linked birth records. Paediatr Perinat Epidemiol. 2005 Nov;19(6):460–471. doi: 10.1111/j.1365-3016.2005.00682.x. [DOI] [PubMed] [Google Scholar]
- 17.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 Jun;200(6):681 e681–687. doi: 10.1016/j.ajog.2009.02.033. [DOI] [PubMed] [Google Scholar]
- 18.Lu MC, Fridman M, Korst LM, Gregory KD, Reyes C, Hobel CJ, Chavez GF. Variations in the incidence of postpartum hemorrhage across hospitals in California. Matern Child Health J. 2005 Sep;9(3):297–306. doi: 10.1007/s10995-005-0009-3. [DOI] [PubMed] [Google Scholar]
- 19.Yasmeen S, Romano PS, Schembri ME, Keyzer JM, Gilbert WM. Accuracy of obstetric diagnoses and procedures in hospital discharge data. Am J Obstet Gynecol. 2006 Apr;194(4):992–1001. doi: 10.1016/j.ajog.2005.08.058. [DOI] [PubMed] [Google Scholar]
- 20.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 Oct;109(4):1174–1181. doi: 10.1213/ane.0b013e3181b2ef75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Department of Health Services . Certificates of Live Birth and Fetal Death Medical Data Supplemental Worksheets. Author; State of California Health and Human Serivces, Sacramento, CA: 2005-2007. [Google Scholar]
- 22.Landy HJ, Laughon SK, Bailit JL, Kominiarek MA, Gonzalez-Quintero VH, Ramirez M, Haberman S, Hibbard J, Wilkins I, Branch DW, et al. Characteristics associated with severe perineal and cervical lacerations during vaginal delivery. Obstet Gynecol. 2011 Mar;117(3):627–635. doi: 10.1097/AOG.0b013e31820afaf2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.U.S. Department of Health and Human Services . Disparities. [Accessed July 11, 2011]. Healthy People 2020.http://www.healthypeople.gov/2020/about/disparitiesAbout.aspx [Google Scholar]
- 24.Kotelchuck M. An evaluation of the Kessner Adequacy of Prenatal Care Index and a proposed Adequacy of Prenatal Care Utilization Index. Am J Public Health. 1994 Sep;84(9):1414–1420. doi: 10.2105/ajph.84.9.1414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Simpson KR. Failure to rescue: Implications for evaluating quality of care during labor and birth. J Perinat and Neonat Nurs. 2005;19(1):24–34. doi: 10.1097/00005237-200501000-00008. [DOI] [PubMed] [Google Scholar]
- 26.Hals E, Oian P, Pirhonen T, Gissler M, Hjelle S, Severinsen AM, Solsletten C, Hartgill T, Pirhonen J. A multicenter interventional program to reduce the incidence of anal sphincter tears. Obstet Gynocol. 2010;116(4):901–908. doi: 10.1097/AOG.0b013e3181eda77a. [DOI] [PubMed] [Google Scholar]
- 27.Tucker MJ, Berg CJ, Callaghan WM, Hsia J. The Black-White disparity in pregnancy-related mortality from 5 conditions: differences in prevalence and case-fatality rates. Am J Public Health. 2007 Feb;97(2):247–251. doi: 10.2105/AJPH.2005.072975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Simpson KR. An Overview of Distribution of Births in United States Hospitals in 2008 with Implications for Small Volume Perinatal Units in Rural Hospitals. J Obstet Neonat Nurs. 2011;40:432–439. doi: 10.1111/j.1552-6909.2011.01262.x. DOI: 10.1111/j.1552-6909.2011.01262.x. [DOI] [PubMed] [Google Scholar]
- 29.Bailit JL, Gregory KD, Reddy UM, Gonzalez-Quintero VH, Hibbard JU, Ramirez MM, Branch DW, Burkman R, Haberman S, Hatjis CG, et al. Maternal and neonatal outcomes by labor onset type and gestational age. Am J Obstet Gynecol. 2010 Mar;202(3):245 e241–245 e212. doi: 10.1016/j.ajog.2010.01.051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zhang J, Troendle J, Reddy UM, Laughon SK, Branch DW, Burkman R, Landy HJ, Hibbard JU, Haberman S, Ramirez MM, et al. Contemporary cesarean delivery practice in the United States. Am J Obstet Gynecol. 2010 Oct;203(4):326 e321–326 e310. doi: 10.1016/j.ajog.2010.06.058. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
This table lists the ICD-9-CM and Vital Statistics Codes used to calculate individual morbidities.
This table shows the unadjusted relationships between potential risk factors and morbidity.
Figure shows percent of serious morbidity cases in relation to the proportion of the population studied, for White, Black, Asian, and Hawaiian/Pacific Islander women.
Abbreviations: pph, postpartum hemorrhage; DVT, deep vein thrombosis; Severe anesthetic, severe anesthetic complication; DIC, disseminated intravascular coagulation; CVA, cerebrovascular accident