National Academies Press: OpenBook

Research Issues in the Assessment of Birth Settings: Report of a Study (1982)

Chapter: Appendix A: Review of the Safety of Maternity Care in Different Birth Locations

« Previous: 4: Variables
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 67
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 68
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 69
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 70
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 71
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 72
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 73
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 74
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 75
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 76
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 77
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 78
Suggested Citation:"Appendix A: Review of the Safety of Maternity Care in Different Birth Locations ." Institute of Medicine and National Research Council. 1982. Research Issues in the Assessment of Birth Settings: Report of a Study. Washington, DC: The National Academies Press. doi: 10.17226/18297.
×
Page 79

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

APPENDIX A Review of the Safety of Maternity Care in Different Birth Locations Gigliola Baruffi Tbe delivery of high quality maternity care is the desire of health professionals and the consumer. Although all the parties concerned agree that such care should be safe, the understanding of other compo- nents of quality and the assessment of their relative importance vary according to philosophy, values, way of life, and immediate interests. B9en the definition of what constitutes a •safe• birth varies. Thus, contrasting viewpoints and strong beliefs have developed concerning the advantages and disadvantages of in-hospital births versus those that take place out of the hospital. Although agreement on the importance of safety exists, few publications have evaluated the safety of mater- nity care. This paper reviews the literature addressing the physical safety of maternity care in different locations. Emphasis has been placed on the relationship of research methodologies and statistical analyses to the study objectives and the conclusions drawn. This review does not in- clude reports on specific obstetrical techniques, individual obstet- rical practices, selection of birth settings' satisfaction with ser- vices, comparison of different personnel' or emotional, psychological, and social factors. CONVEN'l'IONAL IN-HOSPITAL MATERNITY CARE Changes in obstetric practice over time and different approaches to maternity care in the hospital were studied by Chalmers, et al., (1976a, 1976b, 1976c). Tbe data were derived from the cardiff (Wales) Birth survey, in which information was collected on 39,864 births occurring in cardiff from 1965 to 1973. During this period hospital deliveries became routine practice (home deliveries decreased from 1 in 5 to 1 in 100), while the incidence of the following practices increaseda induction of labor (from 7.5 percent to 26.5 percent), episiotomies (from 24.4 per- cent to 46.7 percent), elective cesarean section (from 2.9 percent to 3.4 percent), and forceps delivery (from 6.4 percent to 16.6 percent). Perinatal mortality did not change during this period. Advantages and disadvantages of different approaches to the manage- ment of labor were studied by comparing infant outcomes resulting from 67

68 care provided to 9,907 women by two obstetrical teams at the Cardiff Maternity Hospital between 1968 and 1972. One team's approach was characterized by relatively greater use of induction and stimulation of labor, biochemical and electronic fetal monitoring, analgesia and anesthesia, and operative interventions during delivery. The study failed to demonstrate any advantage or disadvantage of a more •active management of labor.• No significant differences were found when fetal distress, Apgar score, perinatal mortality or low birth weight were used as outcome measures. These reports are based on retrospective data and have nonrandom- ized designs. The data were not originally intended to be used in eval- uation studies. Despite these limitations, which the authors recog- nized, the results of the studies are similar, thus strengthening the case for the validity of the findings. Yanover et al. (1976) evaluated early postpartum hospital discharge by studying the results of the Family Centered Perinatal Care program instituted by the Kaiser Permanente Medical Center in San Francisco. This program offered the opportunity for early postpartum discharge and home care by a perinatal nurse practitioner with the support of obste- tricians and pediatrician3. Of 362 low-risk women initially screened, 271 agreed to participate in the study and were interviewed. Of these, 143 women did not participate because of ineligibility based on failure to meet additional criteria, lack of interest, or other reasons. The remaining 128 were randomly assigned to the alternative care (experi- mental) group or the conventional care (control) group. Forty women did not complete the study because their medical status changed before delivery or during labor, or because of lack of interest, change of residence, or other reasons. Forty-four women remained in the experi- mental group and 44 in the control group. There were no group differ- ences in reasons for discontinuing participation in the study. The median postpartum hospital stay was 26 hours for the study group and 68 hours for the control group. There were no significant differences between the two groups in the number or type of maternal and neonatal morbidity during hospitaliza- tion or the six-week postpartum period. In addition to determining the safety of early home hospital discharge with follow-up care, the program proved economically feasible and was well accepted by the patients. This study is an example of a well-designed prospective study with random assignment to experimental and control groups. UNCONVENTIONAL HOSPITAL MATERNITY CARE Most of the literature on unconventional hospital maternity care (e.g., alternative birth centers or birth rooms) is descriptive in nature. Representative articles by Barton et al. (1980), Gillet (1979), Kerner et al. (1978), Klass et al. (1980), Rising (1976), Schmidt (1980), and Sumner (1976) provide varying degrees of specificity and emphasize dif- ferent aspects of care. The article by Barton et al. (1980) is one of the most specific and thorough treatments of this subject. The authors describe physical

69 facilities, staffing, admission procedures, and the selection and screening of patients for the Alternative Birthing Center (ABC) at the Illinois Masonic Medical Center in Chicago. Admission and transfer criteria at this facility are strict. Only women with a normal obste- trical history, no medical-obstetrical complications or detectable emotional problems, and those who are expected to have a spontaneous vaginal delivery are admitted. Any abnormal prenatal factor or any intrapartum or postpartumVneonatal indication for intervention warrants transfer. The authors conducted their study between March 1978 and March 1979. They reported the number of ABC applicants accepted, the number of women transferred and the reasons for transfer, maternal age and parity, type of delivery, and the incidence and type of maternal and neonatal complications. Of 309 women registered for the ABC, 218 (70 percent) were admittedJ 166 of those admitted (76 percent) delivered there, and 52 (24 percent) were transferred to the conventional unit. Twenty of the transferred women received cesarean sections for a •sub- stantial• rate of 9 percent. Eleven women (5 percent) transferred to the traditional postpartum unit after the ABC delivery. Although this is a valuable descriptive study, no conclusions can be reached about comparative safety, rate of complications, or medical intervention rate because of a lack of matched comparison groups or statistical analyses of the different outcomes of the ABC and transfer women. Goodlin (1980) studied 500 women giving birth at the ABC located at the University of california at Davis Medical Center and compared them to a control group of 500 women who gave birth in the conventional delivery room at the same institution. Control women were of the same low-risk status and socioeconomic class and were offered similar ante- natal childbirth education and care. It is not clear whether the two groups were matched for age, parity, and race. The author described in detail the obstetrical procedures at the ABC and those in the delivery room. The two settings differed greatly in the use of intravenous fluids, electronic fetal monitoring, anesthesia and analgesia, and attendants at birth. Twenty-three percent of the ABC women needed transfer to the delivery room. The investigator studied 42 variables pertaining to labor, delivery, and the neonatal and postnatal periods. Babies in the study were followed for a minimum of 4.5 monthSJ the aver- age follow-up time was 15.2 months. ~ere was a statistically signifi- cant difference between the two settings in 14 of the 42 possible com- plications. All but 2 of the 14--meconium aspiration and readmission-- were in favor of the ABC. The twelve other factors were: failure to progress in labor, oxytocin augmentation of labor, primary cesarean section, fetal distress, meconium-stained amniotic fluid, child abuse, both mild and severe congenital anomalies, central nervous system abnor- malities, jaundice, polycythemia, and scalp infection. Among the mater- nal factors, infections such as endometritis, mastitis, and infected episiotomy were higher among the ABC patients than for the delivery room group. In the ABC group there was one case of severe toxemia accompa- nied by antenatal fetal death, one of abruptio placentae during labor, and one postpartum uterine inversion. ~ere were no similar complica- tions among the delivery room women. The author attributed the •unex- pected• better neonatal outcomes at the ABC to different attitudes

70 toward general health, pregnancy, and labor/delivery among the women electing to deliver at the ABC. Goodlin suggested that the higher inci- dence of postpartum maternal infections among the ABC might result from the location of the ABC in a hospital environment. In this study no attempt was made to analyze the association between obstetrical procedures and outcomes, although settings, procedures, and personnel are described and the study includes a control group. With procedures in the ABC different from those in the delivery room, the possibility exists that procedural differences as well as women's attitudes might be associated with different outcomes. Statistical analysis should be used to investigate this hypothesis. NONHOSPITAL MATERNITY CARE: BIRTH CENTERS Few studies evaluate the safety of out-of-hospital maternity care. A representative selection of publications offering a description of the philosophy, settings, procedures, and personnel of freestanding birth centers are the reports prepared by Bennetts et al. (1982), Ernst et al. (1975), Faison et al. (1979), Lubic (1976), and McCallum (1979). The most comprehensive and detailed description is provided by Faison et al. (1979), who noted the physical facilities, staffing, admission procedures, and the selection and screening of patients at the Child- bearing Center in New York City. The authors reported the number of applicants, the number accepted, the number of women transferred and the reasons for transfer, maternal age and parity, type of delivery, and the incidence and type of maternal and neonatal complications. Such a study is useful because it offers a description of events taking place at the birth center. Although this research cannot be used to draw conclusions about the safety of care provided, it does provide a good description of economical and satisfying care in the nonconventional setting. Another freestanding birth center was evaluated by Halle (1980). In this study, 43 women who delivered at a Southern California center were pair-matched on medical-obstetrical risk, parity, age, and race with 43 women who delivered at a nearby community hospital. No differ- ences were found in the incidence of intrapartum and neonatal problems, but patients at the birth center had a significantly higher incidence of postpartum complications such as operative or difficult delivery (mid-forceps, primary cesarean section, and vaginal breech), perineal lacerations, abnormally long labor, and postpartum infections, as well as neonatal infections and hematologic abnormalities. Although this evaluative study is methodologically sound, its major limitations are the small sample size and the limited data analysis. The small sample size makes it difficult to interpret the differences in the incidence of perinatal problems between the two settings. Fur- thermore, it was not possible for the author to assign patients ran- domly to the two settings. A better understanding of the findings might have resulted if mention of individual perinatal problems had been pro- vided in addition to their quantitative measurement. The study would also have benefited from a comparison of the process of care at the two

71 institutions and from an analysis of the relationship between those processes and outcomes. Bennetts (1981) studied a stratified, systematic sample of 1,938 low-risk women who began labor between 1972 and 1979 in 1 of 11 selected out-of-hospital alternative birth centers with nurse-midwifery services and both physician and hospital backup. The sample was found to be much like those described in other sample studies of single centers. The mean age of the patients was 25 years. Sixty-three percent were white, 34 percent Hispanic, 88 percent married, 45 percent had completed at least 2 years of college, nearly one-third were professionals, and more than one-third were housewives. Ninety-five percent of the patients delivered infants at term, mostly without complication. Nearly 60 per- cent of the labors were unmedicated. Seventy-nine percent of the in- fants were breastfed. Fifteen percent of the patients required transfer to the hospital after the onset of labor due to a change in their risk status. The level of education of the transfers was considerably higher than that of the nontransfers, and the transfers often had no living children. The control group was selected from a follow-up study of hospital deliveries in the United States, which was conducted in 1972 by the National Center for Health Statistics (1972a, 1972b). A group of 4,790 women matched by race, age, gravidity, and obstetrical risk was compared to the group of women using the services of the centers. The ABC group had made significantly more antenatal visits and had better compliance with postnatal visits. Intrapartum use of anesthetics in the hospital sample significantly exceeded that in the ABC sample. There were no statistically significant differences in the numbers of neonatal deaths that occurred in the ABC and in the hospital groups, but the ABCs had proportionately fewer deaths. This is the first national study of nonhospital birth centers oper- ated by certified nurse-midwives with physician and hospital backup. The author provides a comprehensive description of the administration and services of the selected centers. In addition, the study demon- strates the ability of the certified nurse-midwife to select a low-risk population using obstetrical and sociodemographic criteria. The number of perinatal visits, patient compliance to appear for postpartum exam- inations, and neonatal mortality rates indicate that the centers pro- vided safe care. The research methodology used in this study could not be evaluated because only an abstract of the original work was available at the time of this review. (However, see Bennetts et al., 1982, for more description.) Two studies to evaluate the safety of alternative maternity care are in progress. Baruffi (1979) is studying a representative sample consisting of 802 women who delivered at the Booth Maternity Center and a control group of 817 women who delivered at the Thomas Jefferson University Hospital. Both institutions are located in Philadelphia, Pennsylvania, and all deliveries took place in 1977 or in 1978. The design is a prospective, nonconcurrent study in which these women are followed from the early phase of pregnancy to the immediate postpartum period. Race, age, parity, education, and previous pregnancy losses were used as matching variables to establish a control group. Medical- obstetrical risk was measured by the Hobel method (see Appendix E for a

72 review of obstetrical risk assessment methods). Pregnancy outcomes being studied include neonatal morbidity as measured by a neonatal risk score of ~ 10, neonatal mortality, length of stay in the nursery, and intrapartum and postpartum maternal fever (temperature > 38°C for ~ 48 hours). Selected process-of-care variables include induction and stimu- lation of labor, electronic fetal monitoring, analgesia and anesthesia, episiotomies, use of outlet forceps, cesarean sections, breastfeeding, childbirth education, and length of postpartum stay. Bivariate and multivariate analyses are being used to examine the relationship of pregnancy outcomes to medical-obstetrical risk and process of care within and between the two institutions. Preliminary results suggest no difference in pregnancy outcomes between the two settings (Baruffi et al., 1981). Ziskin (1980) is comparing care provided by three birth settings: a nonhospital birth center, a hospital birth room, and a hospital deliv- ery suite in Englewood, New Jersey, during a five-year period (1976- 1981). The sample consists of 500 women from the birth center, 300 from the birth room, and 5,000 from the hospital, all of whom are patients at low medical-obstetrical risk. Only Caucasian women not receiving Medicaid are included. Maternal age, education, gravidity, and parity will be controlled by statistical analysis. Variables to be studied include several measures of maternal and neonatal morbidity, fetal and neonatal mortality, and process of care. The data will be subjected to bivariate and multivariate statistical analyses. NONHOSPITAL MATERNITY CARE: HOME BIRTHS Home birth is the aspect of nonconventional maternity care that gener- ates the most concern among professionals and the most controversy between providers of care and consumers. The following discussion is limited to studies addressing the physical safety of home births. Thus, papers not considered here include those on the philosophy of home births, on the emotional, social, and economic advantages, and on the reasons for selecting home birth settings. Cameron et al. (1979) compared differences in planned home deliv- eries in Salt Lake County, Utah, in 1972 (62 deliveries) and 1975 (105 deliveries). Birth certificate data were studied and delivery status was determined to be planned or not planned for the place of birth listed on the certificate. The planning status was determined by study- ing the listed birth attendant and place of delivery on the birth cer- tificate. Eighty-four women agreed to interviews, which took place two to 15 months after the birth (average, 8 months). (Fifty-five of these women were from the 1975 sample1 29 were from the 1973 group.) The age, race, marital status, and socioeconomic status of the 1975 study group were similar to the 1973 Utah population of women who bore children (also, see Cameron, 1979). However, 19 percent of the women in the 1975 study group had received inadequate prenatal care compared to 5 percent of the 1973 home birth group. Inadequate prenatal care was defined as no care, less than five prenatal visits, or care begun in the third tri- mester. Neonatal outcomes in the 1975 group included four low birth

73 weight infants (defined as weight at birth less than 2,500 grams), one infant with birth injury, and one infant with congenital malformation. The birth weight distribution and the incidence of birth injuries and congenital malformations for the entire state were not provided for comparison by the authors. Cameron et al. ranked their interview data to determine the most important reasons for deciding to plan a home birth. T.he five elements individuals reported desiring werea 1) control over their own delivery, 2) a family-centered experience, 3) no interference with normal pro- cesses, 4) personalized care, and 5) low cost. T.he 55 women interviewed in 1975 were questioned about their infants• health. Although most of the infants were reported to have good health, two had been hospitalized (one for hernia repair, one for jaundice). Most infants had notre- ceived preventive health care (immunizations) at the time of the inter- view. It is unfortunate that the data were not subjected to any statis- tical analysis, and that the comparison made with total state births included only a few of the variables studied. However, the small sample size makes interpretation of results difficult (see Appendix F). Dingley (1977, 1979) studied planned out-of-hospital births in Oregon in 1976 and 1977 by linking birth certificates with infant death certificates and full-term fetal death certificates. In 1976, 959 births (2.7 percent of all state births) occurred outside the hospital. In 1977 the figures increased to 1,492 infants (3.9 percent of all state births). Out-of-hospital births were examined by identifying the type of attendants (e.g., whether or not they were licensed and the size of their obstetrical practice), place of birth (e.g., home, clinic, or other residence), parental characteristics (e.g., education, maternal age, parity), trimester when prenatal care was initiated, number of prenatal visits, birth weight, and neonatal complications. TWenty-two percent of the attendants were licensed and delivered 61 percent of the infants. Sixty percent of the births took place at home, 32 percent in clinics, 7 percent at other residences, and less than 1 percent in •other• unspecified locations. Compared to statistics for the total state population, women giving birth outside the hospital were more educated, younger, and had more children. For both total state and out-of-hospital births, less than one percent of the women received no prenatal care. Of the nonhospital group, however, women attended by licensed personnel had received prenatal care similar to women across the state, but those attended by unlicensed personnel had fewer prenatal visits. Infants born outside the hospital were heavier and the neonatal mortality rate for this group was lower, but the fetal death rate was higher. This study provides detailed comparisons of out-of-hospital births with all state births. But it is difficult to interpret the observed differences because no statistical analysis was performed on the data. State data of the type used in this study are primarily useful for des- criptive purposes and for generating research hypotheses. FOr further examples, see Appendix F. Shy et al. (1980) studied nonhospital deliveries in washington State between 1975 and 1977. Of 3,203 infants in this category, 1,247 were born in birth centers, 1,614 in home residences, and the rest in clin-

74 ics, nonresidence homes, or en route. Home deliveries were found to be at higher risk than those in birth centers. The higher risks were more frequently associated with grand multiparity, advanced maternal age, multiple gestation, and low birth weight. Women who delivered at home were less frequently attended by trained personnel, had received later prenatal care, and had made fewer prenatal visits. All of these differ- ences were statistically significant. After controlling for birth weight, infant mortality was found to be higher among home births than among birth center deliveries, but the difference did not reach statis- tical significance. Although the authors differentiate out-of-hospital births by place of birth, they are aware of their inability to categorize them by plan- ning status (i.e., planned or unplanned out-of-hospital births). Major maternal and infant variables affecting outcome were considered, and an appropriate statistical analysis was used. But it was not possible to study outcomes by controlling for obstetrical risk status. A second limitation of the study is the fact that infant mortality was standard- ized only by birth weight. Although maternal characteristics such as race, age, and parity were identified for the various groups, they were not controlled in the final analysis. This could have been accomplished by multivariate analysis. Burnett et al. (1980) studied home deliveries in North carolina from 1974 to 1976. The investigators determined whether the deliveries were planned or unplanned, whether there was a trained birth attendant (a lay midwife), and whether prenatal care and screening were performed. They found that the women attended by lay midwives had been classified by prenatal screening as medically low-risk pregnancies. Planned home de- liveries not attended by lay midwives had not been screened prenatally. Prenatally screened women, in spite of their high-risk demographic profile (e.g., poor, little education) had the lowest neonatal mortality (3 per 1,000 births). But women who were not prenatally screened had a higher neonatal mortality rate (30 per 1,000 live births) in spite of their low-risk demographic profile (e.g., not poor, more education). The neonatal mortality rate among unplanned home delivery was the high- est (e.g., 120 per 1,000 live births). ~is study of nonhospital births categorizes place of birth by plan- ning status, that is, whether the birth was scheduled to take place where it did. A detailed description of the assumptions and criteria used in assigning planning status is presented. The authors compare sociodemographic characteristics of nonhospital births with those of total state births and control for birth weight when comparing neonatal mortality rates between the various groups. The authors fully discuss the limitations imposed by the use of birth certificate data and by the selection of neonatal mortality as an outcome measure. Statistical analysis is limited to the calculation of relative risk of neonatal mortality and its 95 percent confidence limits. In this study, as in that conducted by Shy et al. (1980), the authors did not control for maternal sociodemographic characteristics in addition to birth weight. Nevertheless, the Burnett et al. study adds emphasis to the importance of differentiating between planned and unplanned home births.

75 Cox et al. (1976) studied 155 home deliveries among 1,937 total deliveries that took place between October 1970 and February 1972 in a study or •catchment• area in Middlesex, England. The socioeconomic characteristics of the home and hospital groups were similar. Of the home birth clients, three mothers in labor were transferred to the hospital as emergencies. Nineteen deliveries originally planned as home births were changed to hospital bookings during pregnancyJ five women who planned hospital deliveries opted for home births. A review of the home births showed that 61 (39 percent) of the women had one or more high-risk factors. Among them, 15 (10 percent) should have been booked for hospital delivery from the time of the first prenatal visit, and 46 (30 percent) should have been changed to hospital bookings dur- ing pregnancy. Various neonatal problems were either ignored or unrec- ognized. No perinatal deaths occurred among the home births, but the perinatal death rate in the catchment area from which the study popu- lation was drawn was 21.7 per 1,000 total births. The authors noted the lack of adherence to established criteria for both place of booking and transfer from home to the hospital. They also noted that adequate postpartum care did not often follow for those discharged early from the hospital. This is a descriptive study of prenatal and neonatal care provided in an epidemiologic catchment or study area within a community. The advantage of this prospective study design is counterbalanced by the lack of rigorous comparison between home and hospital deliveries. Cox and colleagues were more concerned with the weaknesses detected in the process of care and the referral system than with the comparison of process-of-care variables and outcomes in home versus hospital births. Thus, conclusions concerning the safety of nonhospital births cannot be drawn from this study. Fedrick et al. (1978) examined data from the 1958 British Perinatal Mortality Survey (see Butler and Alberman, 1969J Butler and Bonham, 1963). Women aged 20 to 34, who delivered at term and had normal preg- nancies (except for hypertension), were studied by place of booking and place of delivery (i.e., where the delivery was originally planned to occur and where it actually took place). Although perinatal death rates were lower for women delivering out of hospital, the findings were reversed when booking status was examined. Perinatal death rates were statistically significantly lower for women booked for hospital delivery than for women booked for domiciliary or general practitioner unit delivery. This occurred despite the higher incidence of adverse obstetrical history and low socioeconomic status among hospital-booked births. The authors discuss perinatal death rates by place of booking but not by place of delivery. Had they compared perinatal mortality rates by place of booking with those by place of birth and had they found a statistically significant difference, this could have indicated that the health system was functioning well. In other words, higher hospital rates would be explained by the transfer into the hospital of women originally booked for home delivery but whose risk status changed during pregnancy. However, the reported differences were not analyzed statis- tically, thus their significance levels cannot be ascertained. Also,

76 there were no controls for differences in sociodemographic character- istics and obstetrical history. Without such controls, it is not pos- sible to draw conclusions from differences found between the groups. Furthermore, obstetrical practice and awareness of risk factors have changed in the 30 years since the data in this study were gathered. Thus, the findings of Fedrick et al. may no longer be relevant to mater- nity care. Yet, the data ~ase provides information on a time when 35 percent of all births took place at home. Mehl and his colleagues (1975, 1977) reviewed the medical records of 1,146 home births attended by five home delivery services in north- ern California between 1970 and 1975. These investigators provided detailed descriptions of demography (e.g., urban or rural), attendants, population served, process of care, outcomes, and complications. The incidence of various events among home births was compared to the inci- dence of oimilar events in the birth population of the state of Califor- nia or as reported in the literature. No maternal deaths were noted, and the perinatal mortality rate of 9.5 per 1,000 births was lower than the California average. No control group was used in this self-selected study population1 thus, the descriptive information does not allow con- clusions to be drawn about the relative safety of home births. Mehl and Peterson (1976) compared medical records for 1,046 home births in northern California and in Madison, Wisconsin, to an equal number of births from two community hospitals in Madison, Wisconsin. The two groups were pair-matched on maternal age, education, parity, gestational age, major risk factors, and total risk score. Both popu- lations were from the upper middle class and were 98 percent Caucasian. No significant differences were found between the two groups on neonatal and fetal mortality, number of neurologically abnormal infants, and incidence of low birth weight infants. Hospital-birth women received significantly more intravenous oxytocin, anesthesia, and analgesia, and had more low- and mid-forceps deliveries, more cesarean sections, more episiotomies, and more lacerations. Among labor and delivery complica- tions, fetal distress, elevated blood pressure, meconium staining, shoulder dystocia, and postpartum hemorrhage occurred more often in the hospital births. Bleeding during labor and posterior delivery occurred more often among the home births. Birth injuries, total oxygen adminis- tered, and respiratory distress syndrome were observed more often among the hospital births. This study was carefully planned and executed, and the investigators paid attention to the variables known to be associated with pregnancy outcomes. They were also aware of the possibility that the results could be influenced by the limitations inherent in a study based on a review of medical records and on patients who had selected their own birth settings. However, the need for further analysis of the data is given only cursory attention. This is a major limitation because no attempt is made to link obstetrical procedures and outcomes in the two groups or to compare this relationship between the two groups. A multi- variate analysis of differences in outcomes between the home and hospi- tal births, controlling for differences in procedures and characteris- tics of women, would have provided additional information.

77 CONCLUSIONS The study and evaluation of the quality of maternity care need improve- ment in several areas. TO accomplish this, a number of study designs can be used to investigate different aspects of maternity care. Atten- tion must be paid to including, defining, and measuring psychosocial variables and to assessing their impact on pregnancy outcomes. Outcomes other than morbidity need to be identified, defined, and measured quan- titatively. Variables measuring the process of care must be explored in relation to population characteristics as well as to outcomes. Studies should be conducted to evaluate maternity care in different locations, with a variety of providers, and for different populations, i.e., mere reports of experiences in a single institution are not evaluative studies. Innovative investigations should provide the necessary infor- mation for the rational selection of high-quality maternity care. REFERENCES Barton, J. J., s. Rovner, K. Pula, and P. A. Read. 1980. Alternative birthing center: Experience in a teaching obstetric service. American Journal of Obstetrics and Gynecology 137:377-384. Baruffi, G. 1979. Evaluation Study of an Alternative Birthing Center. Grant Number 5 ROl BS03762, NCBSR, PBS, DBBS. Baruffi, G. E., and w. s. Dellinger, Jr. 1981. Alternative birthing: an evaluation of quality of care. Paper presented at the meeting of the American Public Health Association, Los Angeles, california, November 4, 1981. Bennetts, A. B. 1981. Out-of-hospital childbearing centers in the United States: A de,criptive study of the demographic and medical- obstetric characteristics of women beginning labor therein: 1972- 1979. Ph.D. thesis. University of Texas Health SCience Center at Houston. Bennetts, A. B., and R. w. Lubic. 1982. The freestanding birth centre. Lancet 1:378-380. Burnett, c. A., J. A. Jones, J. Rooks, c. B. Chen, c. w. Tyler, and c. A. Miller. 1980. Home delivery and neonatal mortality in North Carolina. Journal of American Medical Association 244:2741-2745. Butler, N. R., and E. D. Alberman. 1969. Perinatal Problems. Edinburgh: Livingstone. Butler, N. R., and D. G. Bonham. 1963. Perinatal Mortality. Edinburgh: Livingstone. Cameron, J. 1979. Professor camerons's response. American Journal of Public Health 69:1285. Cameron, J., E. s. Chase, and s. O'Neal. 1979. Home birth in Salt Lake County, Utah. American Journal of Public Health 69:716-717. Chalmers, I., J. G. Lawson, and A. c. Turnbull. 1976a. Evaluation of different approaches to obstetric carea Part I. British Journal of Obstetrics and Gynaecology 83:921-929. Chalmers, I., J. G. Lawson, and A. c. Turnbull. 1976b. Evaluation of different approaches to obstetric carea Part II. British Journal of Obstetrics and Gynaecology 83a930-933. 1

78 Chalmers, I., J. E. Zlosnik, K. A. Johns, and H. Campbell. 1976. Obstetric practice and outcome of pregnancy in Cardiff residents, 1965-73. British Medical Journal 1:735-738. Cox, c. A., J. s. Fox, P. M. Zinkin, and A. E. B. Matthews. 1976. Critical appraisal of domiciliary obstetric and neonatal practice. British Medical Journal 1:84-86. Dingley, E. 1977. Birthplace alternatives. Oregon Health Bulletin 55(10):1-4. Dingley, E. 1979. Birthplace and attendants: Oregon's alternative experience. Women and Health 4:239-253. Ernst, E. K. M., and M. P. Forde. 1975. Maternity care: An attempt at an alternative. Nursing Clinics of North America 10:241-249. Faison, J. B., B. J. Pisani, R. G. Douglas, G. s. Cranch, and R. w. Lubic. 1979. The Childbearing Center: An alternative birth setting. Obstetrics and Gynecology 54:527-532. Fedrick, J., and N. R. Butler. 1978. Intended place of delivery and perinatal outcome. British Medical Journal 1:763-765. Gillett, J. 1979. Childbirth in Pithiviers, France. Lancet 2:894-896. Goodlin, R. 1980. Low risk obstetric care for low-risk mothers. Lancet, 1:1017-1019. Halle, J. N. 1980. Elective birth center delivery for low-risk pregnancy: Effect on perinatal outcome. M.s. thesis. california State University, Los Angeles. Hobel, c. J. 1976. Recognition of the high-risk pregnant woman. ~ Management of the High Risk Pregnancy, William N. Spellacy, ed. Baltimore: University Park Press. Kerner, J., and c. B. Ferris. 1978. An alternative birthing center in a community teaching hospital, Obstetrics and Gynecology 51:371-373. Klass, K., and K. Capps. 1980. Nine years experience with family- centered maternity care in a community hospital. Birth and the Family Journal 7:175-180. Lubic, R. w. 1976. Alternative patterns of nurse-midwifery care: I. The Childbearing Center. Journal of Nurse-Midwifery 21(3):24-27. McCallum, w. T. 1979. The El Paso Maternity Center. Birth and the Family Journal 6:259-266. Mehl, L., and G. H. Peterson. 1976. Home birth versus hospital birth. Paper presented at the meeting of the American Public Health Association, Miami, Florida, October 20, 1976. Mehl, L., G. H. Peterson, N. s. Shaw, and D. c. Creevy. 1975. Complications of home birth. Birth and the Family Journal 2:123-131. Mehl, L., G. H. Peterson, M. Whitt, and w. E. Hawes. 1977. Outcomes of elective home births: A series of 1,146 cases. Journal of Reproductive Medicine 19:281-290. National Center for Health Statistics. 1972a. 1972 United States National Natality Followback Survey. Rockville, Md.: National Center for Health Statistics. (Computer tape.) National Center for Health Statistics. 1972b. 1972 United States National Natality Followback Survey: Technical Appendix. Rockville, Md.: National Center for Health Statistics. Rising, s. s. 1976. Alternative patterns of nurse-midwifery care: The consumer-professional balance. Journal of Nurse-Midwifery 21:25-27. I I ~

79 Schmidt, J. 1980. The first year at Stanford University's Family Birth Room. Birth and the Family Journal 11169-174. Shy, K. K., F. Frost, and J. Ullom. 1980. Out-of-hospital delivery in Washington State, 1975 to 1977. American Journal of Obstetrics and Gynecology 137a547-552. Sumner, P. 1976. Six years experience of prepared childbirth in a home- like labor delivery room. Birth and the Family Journal 3a79-82. Yanover, M. J., D. Jones, and M. D. Miller. 1976. Perinatal care of low-risk mothers and infantsa Early discharge with home care. New England Journal of Medicine 2941702-705. Ziskin, L. 1980. Study of alternative birthing sites. Unpublished research grant proposal.

Next: Appendix B : Research Issues Concerning Reimbursement for Childbirth Services »
Research Issues in the Assessment of Birth Settings: Report of a Study Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The Institute of Medicine (IOM) of the National Academy of Sciences (NAS) collaborated in this study with the Board on Maternal, Child, and Family Health Research of the Commission on Life Sciences of the National Research Council (NRC) to determine methodologies needed to evaluate current childbirth settings in the United States. Although the proportion of non-hospital births runs as high as 4.4 percent annually in Oregon, insufficient data exist to permit complete evaluation of the various birth settings. The application of good research methods should lead to scientific findings that provide the basis for informed, rational decision making about alternative settings for childbirth.

A committee of 11 experts was appointed to review current knowledge, provide background knowledge, and identify the kinds of research designs useful for assessing such matters as the safety, quality of maternity care, costs, psychological factors, and family satisfaction of different birth settings. The committee was also charged with preparing a report that could be used to solicit, evaluate, and fund proposals for studies on childbirth settings. The committee did not design specific studies to be carried out, but rather attempted to point out issues that should be considered by researchers because it believed that the best proposals would arise from investigator-initiated research. Gaps in research could be filled by requests for proposals developed by agency staff and the agency peer review committee. In addition, IOM staff members and several consultants provided background papers for the committee's consideration. The research that results from this report will be useful to policymakers and to consumers searching for information to aid in making decisions about birth settings. Research Issues in the Assessment of Birth Settings summarizes the study.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!