Objective
We sought to examine rates of birth trauma in 2 groupings (all International Classification of Diseases, Ninth Revision codes for birth trauma, and as defined by the Agency for Healthcare Research and Quality Patient Safety Indicator [PSI]) among infants born by vaginal and cesarean delivery.
Study Design
Data on singleton infants were obtained from the 2004-2005 Healthcare Cost and Utilization Project Nationwide Inpatient Sample.
Results
The rates of Agency for Healthcare Research and Quality PSI and all birth trauma were 2.45 and 25.85 per 1000 births, respectively. Compared with vaginal, cesarean delivery was associated with increased odds of PSI birth trauma (odds ratio [OR], 1.71), primarily due to an increased risk for “other specified birth trauma” (OR, 2.61). Conversely, cesarean delivery was associated with decreased odds of all birth trauma (OR, 0.55), due to decreased odds of clavicle fractures (OR, 0.07), brachial plexus (OR, 0.10), and scalp injuries (OR, 0.55).
Conclusion
Infants delivered by cesarean are at risk for different types of birth trauma from infants delivered vaginally.
Cesarean delivery is the most common major surgical procedure in the United States and rates have increased from 22.8% in 1989 to 30.3% in 2005. There are known risks to mother and fetus during both vaginal and cesarean deliveries. Although a number of studies examine maternal safety associated with cesarean delivery, there are few studies that address neonatal safety. In the era of rapidly increasing cesarean delivery rates, neonatal safety data are urgently needed to monitor the quality of care and better counsel obstetric patients. In 2000, the Institute of Medicine published “To Err Is Human: Building a Safer Health System,” which focused attention on the importance of patient safety in all fields of medicine. In response, the Agency for Healthcare Research and Quality (AHRQ) has developed a group of Patient Safety Indicators (PSIs). A PSI is a set of International Classification of Diseases, Ninth Revision ( ICD-9 ) codes that represent outcomes considered avoidable through practice modification. These indicators were chosen by a group of experts through literature review, consensus development, and public comment, but have not been validated. A PSI was developed for birth trauma and included 7 types of neonatal birth trauma. Our study uses this AHRQ PSI for examining national rates of birth trauma. We also examine another clinical grouping of birth trauma codes (all birth trauma) made up of all codes found in the birth trauma section of the ICD-9 coding manual, to capture the total amount of neonatal birth trauma in the United States. In addition, we examine individual types of birth trauma. Furthermore, potential associations of birth trauma with clinical and demographic factors such as route of delivery, birthweight, and presence of fetal distress are examined. We hypothesized that the rates of individual types of neonatal birth trauma, rather than any 1 grouping, would vary by route of delivery.
Materials and Methods
Hospital discharge data from the Nationwide Inpatient Sample (NIS), 2004-2005, were obtained from the Healthcare Cost and Utilization Project. The Healthcare Cost and Utilization Project is a group of health care databases and related software tools that were developed through a partnership with private and public state level data collection organizations and sponsored by the AHRQ. The NIS is the largest all-payer inpatient care database publicly available in the United States.
The sampling universe for NIS includes US community hospitals that are open during any part of the calendar year and are designated as community hospitals by the American Hospital Association Annual Survey of Hospitals. Community hospitals are defined as all nonfederal general and specialty hospitals, with average length of stays <30 days, and whose facilities are open to the public. This definition includes specialty hospitals such as orthopedic, pediatric, obstetrics-gynecology, and ear-nose-throat institutions, as well as public hospitals and academic medical centers. A hospital is considered to be a teaching hospital if it has an American Medical Association-approved residency program, is a member of the Council of Teaching Hospitals, or has a ratio of full-time equivalent interns and residents to beds of ≥ .25. Veterans hospitals and other federal hospitals, rehabilitation hospitals, psychiatric hospitals, and alcohol/chemical dependency treatment facilities are not included in the sample. Data are gathered from all community hospitals within each participating state. Hospitals may vary from year to year based on state participation, but this change is accounted for by the below sampling strategy. In 2004 and 2005, 37 states contributed data.
For each year, the NIS is designed to approximate a 20% stratified sample of community hospitals in the United States and contains discharge data for approximately 8 million hospital stays from >1000 hospitals. The sampling frame for NIS uses 5 strata: type of ownership, number of hospital beds, teaching status, urban or rural location, and region of the country. For each sampled hospital, 100% of the discharges are retained. These sampling probabilities are used to create a “weight” for each hospital so when appropriate statistical tools are used, estimates reflect a national sample of community hospitals. That is why we present weighted data, not unweighted data. For each change to the sampling frame, AHQR compares individual years of the NIS with the corresponding National Hospital Discharge Survey and with the Medicare Provider Analysis and Review file to check for consistency and validate the dataset.
Hospital discharge diagnoses were classified using the ICD-9, Clinical Modification ( ICD-9-CM ) codes. Singleton live born infants were identified using ICD-9-CM diagnosis codes and classified as either vaginal birth (V30.00, V39.00) or cesarean birth (V30.01, V39.01). Birthweight was defined as low (764.01-764.08, 764.11-764.18, 765.01-765.08, 765.11-765.18), high (766.0, 766.1), or average. Since there are no specific codes for average birthweight, newborns without codes specifying low birthweight or high birthweight were considered average. Presence of fetal distress was defined by ICD-9-CM codes 763.81, 763.82, 763.83, 768.2, 768.3, and 768.4.
Neonatal birth traumas were defined as subdural and cerebral hemorrhage (767.0), epicranial subaponeurotic hemorrhage (767.11), other injuries to scalp (767.19), fracture of clavicle (767.2), other injures to skeleton (767.3), injury to spine and spinal cord (767.4), facial nerve injury (767.5), injury to brachial plexus (767.6), other cranial and peripheral nerve injuries (767.7), other specified birth trauma (767.8), and birth trauma unspecified (767.9). Other specified birth trauma (767.8) includes hematoma or injury to sternocleidomastoid; hematoma or rupture of spleen, liver; teste; vulva; viscera; kidney or stomach; injury or damage to eye or traumatic glaucoma; fetal laceration by scalpel. Other injuries to scalp (767.19) included caput succedaneum, cephalohematoma, and chignon (from vacuum extraction). If an infant had a diagnosis of subdural and cerebral hemorrhage and was defined as preterm (765.01-765.09, 765.11-765.19, 765.21-765.28), then the diagnosis was not considered a birth trauma. If an infant had a diagnosis of other injuries to skeleton or injury to spine and spinal cord and had a diagnosis of osteogenesis imperfecta (756.51), then the diagnosis was not considered a birth trauma.
Birth traumas were examined in 2 groups as well as individually. The AHRQ PSI birth trauma includes the 7 types of birth trauma: (1) subdural and cerebral hemorrhage, (2) epicranial subaponeurotic hemorrhage, (3) other injuries to skeleton, (4) injury to spine and spinal cord, (5) other cranial and peripheral nerve injuries, (6) other specified birth trauma, and (7) birth trauma unspecified. In addition, we created and examined a new all-inclusive group of neonatal birth trauma, called “all birth trauma,” which includes the 7 PSI birth traumas as well as 4 additional birth traumas: (1) other injuries to scalp, (2) fracture to clavicle, (3) facial nerve injury, and (4) injury to brachial plexus.
Primary payer was defined as public (Medicare/Medicaid), private (private insurance), or other (self-pay, no insurance). For each hospital, the number of liveborn singleton deliveries per year were categorized as ≤400, 401-1300, or >1300. Hospital teaching type and location were combined and defined as rural, urban nonteaching, and urban teaching.
To account for the complex sampling design, we used software (SAS-callable SUDAAN, v. 9.1; Research Triangle Institute, Research Triangle Park, NC) to analyze the data. Rates, along with the 95% confidence interval (CI), were calculated per 1000 singleton live hospital births. Logistic regression was used to estimate both unadjusted and adjusted odds ratio (aOR) as well as corresponding 95% CI. Interaction between route of delivery and neonatal birthweight, and route of delivery and presence of fetal distress, was evaluated and considered to be present if P < .01. Programming and data results were confirmed by 2 independent researchers. Since the NIS data do not contain personal identifiers and are publicly available administrative data, the Centers for Disease Control and Prevention determined this research to be exempt research not requiring review by an institutional review board.
Results
Our study population included a weighted sample of 8,176,523 live singleton newborns born in a hospital in 2004 and 2005. The proportion of singleton neonates born by cesarean delivery was 29.52%. The proportions of low and high birthweight infants were 4.44% and 5.77%, respectively. The proportion of infants with fetal distress prior to delivery was 0.60% ( Table 1 ).
| Variable | n a | n b | Percent | AHRQ PSI birth trauma rate c (95% CI) | Unadjusted OR for PSI birth trauma (95% CI) | All birth trauma rate c (95% CI) | Unadjusted OR for all birth trauma (95% CI) |
|---|---|---|---|---|---|---|---|
| Sex of infant d | |||||||
| Male | 861,461 | 4,176,147 | 51.19 | 2.63 (2.35–2.91) | 1.16 (1.09–1.24) | 28.74 (27.25–30.22) | 1.27 (1.23–1.31) |
| Female | 821,455 | 3,982,713 | 48.80 | 2.27 (2.00–2.54) | Referent | 22.80 (21.41–24.19) | Referent |
| Primary payer | |||||||
| Public | 699,742 | 3,385,847 | 41.41 | 2.23 (2.00–2.46) | 0.90 (0.75–1.07) | 25.38 (23.55–27.21) | 0.99 (0.90–1.09) |
| Private | 865,135 | 4,201,515 | 51.39 | 2.63 (2.25–3.02) | 1.06 (0.87–1.30) | 26.26 (24.68–27.85) | 1.03 (0.93–1.13) |
| Other | 121,657 | 589,161 | 7.20 | 2.48 (2.07–2.89) | Referent | 25.58 (23.28–27.88) | Referent |
| Delivery route | |||||||
| Cesarean | 498,451 | 2,413,979 | 29.52 | 3.46 (3.13–3.79) | 1.71 (1.56–1.87) | 17.07 (16.02–18.11) | 0.57 (0.55–0.59) |
| Vaginal | 1,188,083 | 5,762,544 | 70.48 | 2.03 (1.77–2.29) | Referent | 29.53 (27.92–31.13) | Referent |
| Birthweight | |||||||
| Low | 75,053 | 362,921 | 4.44 | 3.26 (2.78–3.75) | 1.42 (1.23–1.63) | 15.94 (14.55–17.34) | 0.63 (0.59–0.68) |
| Normal | 1,513,724 | 7,341,681 | 89.79 | 2.30 (2.04–2.57) | Referent | 25.04 (23.67–26.40) | Referent |
| High | 97,757 | 471,921 | 5.77 | 4.17 (3.60–4.74) | 1.81 (1.60–2.05) | 46.09 (43.33–48.84) | 1.88 (1.79–1.98) |
| Fetal distress | |||||||
| Present | 10,080 | 48,855 | 0.60 | 6.75 (5.02–8.48) | 2.79 (2.13–3.65) | 64.15 (52.61–75.69) | 2.61 (2.17–3.14) |
| Absent | 1,676,454 | 8,127,668 | 99.40 | 2.43 (2.16–2.69) | Referent | 25.62 (24.26–26.98) | Referent |
| Admission on weekend d | |||||||
| Yes | 342,303 | 1,659,934 | 20.30 | 2.64 (2.32–2.95) | 1.10 (1.01–1.18) | 28.57 (26.96–30.18) | 1.14 (1.11–1.17) |
| No | 1,344,230 | 6,516,584 | 79.70 | 2.41 (2.14–2.67) | Referent | 25.16 (23.80–26.51) | Referent |
| No. of singleton deliveries/y at hospital d | |||||||
| ≤400 | 83,916 | 427,354 | 5.23 | 1.98 (1.50–2.47) | 0.79 (0.60–1.04) | 19.93 (17.78–22.08) | 0.76 (0.67–0.86) |
| 401–1300 | 358,435 | 1,736,210 | 21.23 | 2.35 (1.93–2.76) | 0.93 (0.75–1.16) | 26.29 (23.87–28.71) | 1.01 (0.89–1.13) |
| ≥1301 | 1,244,183 | 6,012,958 | 73.54 | 2.52 (2.18–2.85) | Referent | 26.14 (24.40–27.89) | Referent |
| Hospital location and teaching type d | |||||||
| Rural | 185,118 | 937,600 | 11.47 | 2.32 (1.62–3.02) | 0.81 (0.57–1.14) | 22.28 (19.95–24.61) | 0.83 (0.72–0.96) |
| Urban nonteaching | 771,587 | 3,706,117 | 45.33 | 2.10 (1.79–2.42) | 0.73 (0.59–0.92) | 25.96 (23.94–27.98) | 0.97 (0.86–1.10) |
| Urban teaching | 729,829 | 3,532,807 | 43.20 | 2.86 (2.38–3.34) | Referent | 26.68 (24.37–28.99) | Referent |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
