Click for next page ( 213


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 212
CHAPTER 10 Prenatal Care and Low Birthweight: Effects on Health Care Expenditures Public policy decisions about health promotion strategies should incorporate an understanding of their costs as well as their benefits. Ideally, for each strategy summarized in Chapter 4, policymakers would be able to determine the net public and private costs per case of low birthweight averted. Equipped with data on the most cost-effective means to reduce low birthweight, policymakers would then weigh the additional costs of intervention against the public and private benefits from reducing the low birthweight rate. Unfortunately, lack of adequate data on the costs of these strategies, alone or in combination, placed such a goal beyond the comm~ttee's reach. In general, estimates of the costs of measures that should be implemented in the period before pregnancy to reduce the risks associated with low birthweight are not available. Information does exist on the costs of family planning, but it does not include calculations of the economic impact of projected changes in the low b~rthweight rate resulting from family planning practices, and the committee did not undertake such an analysis. Lack of adequate data also prevented the committee from estimating the additional public expenditures that would be required to finance the recommended public information program and research efforts. With regard to extending the availability of prenatal care, however, the committee found that a straightforward, common sense analysis could be performed regarding some of the financial implications involved in the provision of prenatal services to pregnant women. Formal cost-benefit and cost-effectiveness analyses, including estimates of the present value of low birthweight infants, were not feasible, however, because of problems in the quality and uniformity of available cost data, uncertainties about the life expectancy of low birthweight infants with significant morbidity, difficulties in delineating the services received, and uncertainties about target populations. Within the domain of prenatal care costs, the committee elected to undertake a narrow set of tasks. First, a target population of high-risk pregnant women was identified. Second, within the target population, the additional public fiscal outlays required to provide these women with more complete prenatal care services than they now obtain were estimated. Third, the direct medical care expenditures 212

OCR for page 212
213 resulting from a low-weight birth were assessed for a single year. Finally, the additional public fiscal outlays for adequate prenatal care of the target group mothers were compared with the potential savings in single-year medical care costs that might result from reduction in the low birthweight rate, resulting from increased receipt of prenatal care. The estimated costs of extending more complete prenatal services to the target group would represent potential expenditures of new dollars; the reductions in the costs of caring for low birthweight infants, by virtue of reducing the low birthweight rate in the target group, would be a return on these new dollars. Even the narrow agenda of this chapter is fraught with uncer- tainties. For example, the committee focused on high-risk women receiving public assistance; it di d not analyze outcomes for other women at high risk. To assess the costs of prenatal care, the committee relied on data from current obstetrical practice; it did not speculate on the costs of changes i n the content of care, such as those outlined in Chapter 8, that might be appropriate for higher isk mothers. Although prescription of a particular set of prenatal services to produce a defined reduction in low birthweight is not possible at this tome, the estimates of a reduced rate of low birthweight used in this analysis appear achievable. This judgment is based in part on studies reviewed in Chapter 6 that, despite qualifications, demonstrate reductions in low birthweight associated with adequate prenatal care, and in part on the multivariate analysis presented in Chapter 3, which indicates that a significant reduction in low birthweight rates could result from changes in the timing and frequency of prenatal visits. The analysis presented in this chapter supports the widely shared view that, with adherence to current standards for timing and frequency of prenatal visits, the Surgeon General's national goal of a 5 percent rate of low birthweight overall and a maximum rate of 9 percent in high-risk subpopulations,* can be met through provision of such prenatal services. The co~ittee's focus on mothers receiving public assistance permits a calculation of the direct budgetary consequences of subsidizing prenatal care. In most instances, both the predelivery services consumed by these high-risk mothers and the postdelivery care of their low-weight infants would be paid by governmental programs. If the postdelivery cost savings outweigh the additional predel ~ very expenditures, then a net reduction in public expenditures will be obtained. It must be emphasized, however, that net savings in governmental budgets is a 1 imited or iter ion. A soc iety concerned with the health and productivity of all its citizens might choose to reduce low-weight births even if the budgetary outlays were to exceed savings. *The Surgeon General's objective states, By 1990, low bir th weight babies . . . should constitute no Are than 5 percent of all live births . . . (and) no county and no racial or ethnic group of the population {e.g . black, Hispanic, Amer ican Indian) should have a rate of low birth weight infants . . . that exceeds 9 percent of all live births.

OCR for page 212
214 In the following analysis, the cost of prenatal care compr ises the expenditures required for routine prenatal services as currently provided by most qualified physicians and nurse-midwives. The costs of a more comprehensive package of care, inc.' uding psychological and social counseling, behavioral modif ication, nutr itional planning, and special diagnostic studies, are excluded. The cost of low birthweight reflects initial expenditures on intensive care; rehospitalization costs of survivors dur ing the f irst year and a single-year estimate of the annual long-term medical expenses of those who survive the first year of life and do not require ~nstitutional~zation. The cumulative direct long-term medical expenses of both institutionalized and noninstitutionalized survivors are excluded, as are the indirect costs of lost productivity from infants who survive with handicaps and from the family members who care for them. Overview of the Analysis In the sections below, the major components of the committee's computation are defined and discussed. Later in the chapter, the specif ic f igures used for the cost elements are presented. Target Population The target population was identified as the total national cohort of women aged 15 to 39 years who receive public assistance and who have less than 12 years' education. To assess the number of low-weight births in this group, the committee proceeded in three stages. First, it estimated the total size of the target population for 1980 from a 1/1,000 sample of the 1980 U.S. Census .2 In that year, an estimated 2,981,000 black and white women aged 15 to 39 years received public assistance. Among such women, an estimated total of 1,399,000 had completed less than a high school education (Population Po in figure 10.1~. Second, to estimate the total number of live births among the target group, the population was further d~saggregated by age (15 to 17, 18 to 19, 20 to 24, 25 to 29, 30 to 34, and 35 to 39 years) and race (black and white only). Age- and race-specific national fertility rates for 1979 were then applied to each subgroup. 3 Among the target population, a total of 110, 600 live births were estimated for 1980 (Population PI in Figure 10 .1 ) . Third, for each age/race subgroup of the target population, the committee estimated the total number of low-weight births from age- and rac~specific low birthweight rates in 1980.4 (The 1980 data are used here instead of the 1981 data presented in Table 3.6 of Chapter 3 to be consistent with the census data.) Among the 110,600 live births ~ n the target cohort, 12,719 (11.5 percent) were estimated to be low-weight babies (Population P2 in Figure 10.1~. The population of less educated women on public assistance was, in the committee's view, the most easily identifiable population for

OCR for page 212
215 TARGET POPULATION OF HIGH-RISK WIN ( Po ) (1,39 ,000) TOTAL Lit BIRTHS TARGET GROUP (Pi) ~ (~10,600) ~ - ~ ~ LOWWEIGHT BIRTH FI=T SECOND THIN AMONG TARGET GROUP (P2) TRIMESTER TRIMESTER TRIMESTER ( 12, 719 ) ONSET OF ONSET OF ONSET OF 1 CARE (Pll) CARE (P12) CARE (P13) 02lE-MONTH SURVIVORS AMONG LOW-WEIGHT BIRTHS (P 3) j (11,701) 1 1 VERY LOWWEIGHT MODE==LY LOWMIGHT ON~ SURVIVOR SURVIVORS SURVIVORS AMONG LOWWE IGHT (1,170) (10,531) BIRTHS ~ 1 (11,167) REHOSPITALIZED REMOSPIT=IZ~ VERY LOWWEIGHT (P31) MODERATELY LOW-WEIGHT (P32 ) LONG TERM (448) (2,001) MORBIDITY (P4) (2,167) FIGURE 10 .1 Population calculations for cost analysis . analysis. Though many women with at least a high school education also could be classif fed as high r isk on the basis of other factors, the important relationship between education and low birthweight risk (chapters 2 and 3 ~ motivated the committee ' s choice. Moreover, because women receiving public assistance al so are likely to receive public subsidy for medical care, both the outlays for additional prenatal care and the savings from a reduced incidence of low birthweight involve governmental funds. In computing the number of births among the target group, the committee used age- and race-specif ice, but not education-specif ic, natal ity rates . Accordingly, the total namer of bir ths in the target population (and thus the total number of low-weight births) may be underestimated. Low-weight birth rates, however, were specific for infants whose mothers had less than a high school education. unit Cost of Prenatal Care The committee assumed that the additional cost of providing prenatal care to a mother with no prior care was equal to the current average charge for routine care in the United States. The average

OCR for page 212
216 charge was ascertained by reviewing a wide range of sources, including a small survey commissioned by the committee, described below. The current cost of prenatal care is denoted by C1 in Table 10.1. The use of an average pr ice for currently available, routine care entails two critical assumptions. First, the content of care (and hence the resource costs) for the target population would differ little from that currently available. Second, the increased demand for care resulting from public subsidy would not cause the price of care to rise. That is, the supply of resources for provision of care would be sufficiently elastic to meet the increased demand. Total Cost of Prenatal Care The committee examined national vital statistics data on the timing of prenatal care in relation to race and educational statuses For a TABLE 10.1 Total Cost Computations = unit cost of prenatal care = unit cost of initial hospitalization of low-weight infant C3 = unit rehospitalization cost of surviving low-weight infant C4 = unit long-term, single year morbidity cost of surviving low-weight infant TC1 = total additional cost of prenatal care = C1 x (P13 + .5P12) TC2 = total cost of initial hospitalization = C2 x P2 TC3 = total cost of rehospitalization = C3 x (P31 + P32) TC4 = total long-term morbidity cost = C4 x P4 TC5 = total cost of low-weight births = C2 ~ C3 ~ C4 in target group low birthweight rate per 100 in target population after introduction of prenatal care TC6 = total reduction in cost of care = t1 - r/11.5) x TCs of low-weight births

OCR for page 212
217 population of pregnant women with the same racial and educational distribution as the target group, the committee estimated the following distribution of timing of care: 57 percent initiate care in the first trimester, 32 percent in the second trimester, and 11 percent initiate care in the third trimester or receive no care (see groups Pll, P12' and P13 in Figure 10.1~. Given the total size of the target population, the estimated cost of care, and the estimated current distribution of care, the committee computed the estimated total additional cost of providing first-trimester care to the entire high-risk cohort (see Table 10.1~. Costs of Care of Low-We~ght Births The committee's analysis includes a single-year estimate of the costs of care per low-weight birth. The figure used has three components: initial hospitalization costs; the costs of repeat hospitalizations during the year following birth among those infants who were initially discharged alive; and an estimate of annual noninstitutional, nonhospital morbidity costs incurred by those infants surviving the first year of life. Each component of the cost of care for low-weight infants was taken from the literature reviewed below. Total costs for initial hospital- ~zation equaled the average cost per infant {C2) multiplied by the total population of 12,719 low-weight births in the target groups (P2 in Figure 10.11. Total costs for repeat hospitalization were computed separately for moderately low-weight and very low-weight births.5 Thus, among the 12,719 low-weight births in the target group, 11,701 were elected to survive the neonatal per iod (P3 I, compr Using about 1,170 very low-weight and 10,531 moderately low-we~ght infants. These separate population totals were in turn multiplied by the estimated probabilities of rehospitalization. This procedure yielded 448 very law-we~ght and 2,001 moderately low-weight rehospitalized infants (populations P31 and Ply. The latter population figures were multiplied by the estimated unit cost of rehospitalization (C3) to calculate the total costs. Finally, among the 11,719 infants who survived the neonatal period (P3), a total of 11,467 were estimated to survive the first year, of whom 2,167 were expected to suffer long- term morbidity (Pig. The annual morbidity costs for non~nst~tution- alized, nonhospitalized survivors were derived by multiplying the - estimated unit cost of morbidity (C4) by the latter population total. The population subtotals used In each stage of the analysis are diagrammed in Figure 10.1. The total cost computations are summarized in Table 10.1. In the estimate of the initial hospitalization cost=, the committee included an unknown number of infants (probably of moderately low birthweight) who would not have required or received intensive care in Level II or III neonatal intensive care units. This may appear to result In an overstatement of these costs. However, when the large cohort of moderately low b~rthweight infants is excluded from the initial hospitalization figures, the committee's conclusions remain

OCR for page 212
218 unchanged. Further, the low base charge used to determine the initial hospitalization costs probably does not adequately reflect all of the current hospital and professional charges for the care of low birth- weight infants. It is not possible to determine the degree to which these adjustments offset each other. The dollar figure used for the per diem charge for rehospitalization of infants would tend to produce an underestimation of costs, because it does not include any professional charges or reflect the higher than average hospital resource utilization by infants admitted to general pediatric wards or to intensive care units. The annual morbidity costs also are underestimated because they do not include charges for institutionalization or the discounted present value of expenditures over the lifetime of the low birthweight infant. This estimate of the ambulatory medical care costs of low birthweight infants must be interpreted cautiously, however, because it assumes that their chronic care is similar in its pattern of resource utiliza- tion to that of other children with chronic disabilities. This has not been establ ished . Computation of single-year expenditures attributable to the care of low-weight infants does not by itself gauge the incremental costs incurred by delivery of a low-weight neonate. For example, if all live births were shifted to the normal weight category, there would still be some expected costs due to hospitalization, rehospitalization, and long-term morbidity. Alternatively, no adjustment has been made for cost savings in infant care associated with the improved health of normal birthweight infants born to pregnant women who have received improved prenatal management. These factors have not been incorporated, because the committee believes they would not affect the outcome of the analysis significantly. More importantly, confining the cost estimates for the care of low birthweight infants used in this analysis to a single year underesti- mates the actual costs over the long term. Thus the committee's estimates of this variable must be viewed as conservative. Estimated Cost Savings If the current low birthweight rate of 11.5 percent among the target population were reduced only to 10 percent, then the total number of low-weight neonates would decline from 12,719 (population P2 in figure 10.1) to 11,060. If the subsequent survival probabilities, rehospitalization probabilities, and long-term morbidity probabilities remained unchanged, then the total costs associated with 1-weight births would be reduced by a factor of 10/11.5 = 0.87. Similarly, if the low birthweight rate were reduced to 9 percent, with all other probabilities unchanged, the population P2 would be only 9,954, and total costs would be reduced by a factor of 9/11.5 = 0.78. To assess the potential cost savings from reduced low-weight births in the target group, the committee computed the total costs for low birthweight rates of 11.5 percent, 10 percent, and 9 percent, respec- tively. Such a procedure assumes that the reduction in low-weight

OCR for page 212
219 rates does not affect subsequent probabilities _ tion, rehospitalization, or long-term morbidity. Tn Ah. r - ~1 the h - 1 law . al 1 nms:t magnitudes are of initial hospitaliza- ,~ ~ 7~ _ stated in constant 1984 dollars. Adjustments for changes in the general pr ice level were based on the medical care component of the consumer pr ice index. Additional Assumptions and Comments Many assumptions had to be made to perform this analysis, and they must be taken into consideration when interpreting the committee's conclusions. First, the dollar figures used as a basis for the cost calculations represent what the committee considered to be reasonable estimates of the current charges for the particular medical services discussed. No attempt was made to analyze the current sources of reimbursement by patients, public agencies, insurance companies, or other third parties for prenatal services or for the subsequent care of low birthweight infants. Similarly, no attempt was made to analyze the effect of reimbursement of less than charges on potential cost shifting by providers or on access to care. Second, the estimates used for health care charges (both for prenatal services and for services used by infants) do not generally include all of the medical services provided and thus underestimate the costs per patient, especially in the cost of care of the hospitalized low birthweight child. For example, charges for the medically indicated use of ultrasound to evaluate the fetus generally would not be included in routine prenatal care costs. Similarly, charges for the appropriate use of CT and ultrasound head scanning of low birthweight neonates for detection of intracranial hemorrhages and any related surgical procedures might not be included in estimates of initial hospitalization costs. Third, women at high risk of bearing a low birthweight infant who were not on public assistance because they did not apply or qualify, or who were not represented in the census data used in this analysis, are not included in the target group of pregnant women at high risk. The total costs both of prenatal care and the care of low bir thweight children would be increased by including these additional groups, but the increases might have a different pattern than that projected for the cohort presented. Fourth, the estimate of prenatal care costs may be understated to a degree that cannot be adequately quantified because of a greater use of these services by women at greater risk, and because changes in the content of prenatal care services may be implemented and add more to the expense. However, even if additional prenatal care costs were substantially greater, the committee's conclusions would still be valid. Fifth, this analysis does not take into consideration the possible costs of appropriate medical interventions during the prenatal period that might not have occurred without the increased provision of prenatal services to these high-risk pregnant women. For example, early identif ication of preterm labor might lead to the use of tocolytic agents and hospitalization before admission for delivery,

OCR for page 212
220 which would increase medical costs. Alternatively, early identification of toxemia might lead to diet counseling and bed rest, which could decrease the need for hospitalization or shorten the length of stay and thus reduce medical costs. Finally, no adjustment has been made for fetal wastage, which might reduce prenatal care costs and low birthweight child care costs. Alternatively, higher fertility rates than those utilized might be appropriate for this h~gh-risk cohort of pregnant women, and this would increase both the costs of prenatal services and of low bir thwe ight infant care. Description of Costs This section summarizes the costs reported in the literature for the provision of prenatal and delivery services to pregnant women, of neonatal intensive care to low birthweight infants, of subsequent postneonatal hospitalization to surviving infants, and of long-term medical care to survivors. It also presents and justifies the estimates used by the committee for each cost component in the analysis. All costs and charges have been adjusted for inflation to March 1984 dollars. Prenatal and Delivery Costs The range of costs and charges for prenatal care reported in the literature is substantial.7 This variability is due in large part to differences in the content of the prenatal services, to differences in the reimbursable costs or charges allowed by third party payers, and to variations in the methods of partitioning the total costs of maternity care into prenatal, delivery, and postpartum components. For example, the cost of the prenatal care services may be defined as the "reimbursable cost. provided by Medicaid, or as the professional "charges" that are paid fully or in part by private insurance companies or individuals. Further, it is customary for payment to physicians for prenatal services to be included as part of the comprehensive fee for delivery. Published data on the separate cost of prenatal services is limited. Therefore, an independent survey of prenatal care charges In the United States was under taken.8 One hundred perinatal health care professionals from 38 of the 50 states were queried and responses were obtained from 34 of these individuals. The average physician fee for prenatal care was $365; obviously this figure could be affected significantly by the experiences of the nonresponders. This survey also indicated the following averages for physician comprehensive fees for a cluster of services including prenatal care, delivery, and postpartum in-hospital care: $804 for a vaginal delivery and $1,179 for a cesarean delivery. The Health Insurance Association of America (HIAA), in a publication entitled "The Cost of Having a Baby, n reports that the

OCR for page 212
221 average charge for a physicians obstetrical services in the United States, including prenatal and delivery services, is $730 for a vaginal delivery and $941 for a cesarean delivery. HIAA data also indicate that the average national hospital charge for a delivery, based on a hospital per diem, is $1,648 for a vaginal and $2,853 for a cesarean delivery. Total charges for professional services (obstetric, anesthesia, pediatric) and the hospital maternity stay for the United States average $2,620.9 The committee also reviewed other data on prenatal charges. For example, in a report submitted to the state legislature, the Michigan Department of Public Health proposed a reimbursement sling of $350 per patient for providing an estimated 12 prenatal visits and a postpartum visit, nutritional and psychosocial assessment, laboratory services, prescriptions, and parent education for Medicaid patients through local health departments.~ Malitz, analyzing the cost implications of extending Medicaid in Texas to low-income pregnant women, estimated that ~birth-related" services to a majority of Medicaid recipients for prenatal care averaged $266 per woman. There is no description of the services provided to these Medicaid recipients. The Colorado Department of Health reported that prenatal services cost an average of $461 per woman, and the Ohio Department of Health estimated that the cost of providing prenatal services in public health clinics was $365 per patient.) 3 McManus reports, in an analysis of Medicaid's Impact on prenatal care to low-income women, that in 1982, obstetricians were reimbursed an average of $398 and ~specialists. $417 for prenatal, delivery, and postpartum care. There is no description, however, of the services rendered and no indication of the proportion of the total reimbursement that could be allocated to prenatal care alone. ~ 4 The Oakland, Calif., "OB Access Project. reported $870 as the average fee-for-service payment financing a rich array of obstetrical services including prenatal and delivery care to a group of Medi-Cal patients.~5 This contrasted with an average physician fee reimburse- ment of $620 for routine prenatal, intrapartum, and postpartum care under the regular Medz-Cal program. The $250 differential financed a more comprehensive set of prenatal services that went beyond routine care to include psychosocial and nutritional assessments, 6 childbirth education, and other care components described in Chapter 6. Based on the available literature and the survey, the committee estimates the cost for professional-services associated with prenatal care, excluding delivery and postpartum charges, to be approximately $400. The cost comparison analysis in this chapter uses this figure for the additional cost of prenatal care in determining the expense of providing prenatal services (as currently practiced) beginning in the first trimester to women on public assistance who are at high risk of deliver ing a low birthweight infant. . Initial Hospitalization Costs Reported costs and charges for the initial hospitalization of low birthweight infants in intensive care units vary enormously, from only

OCR for page 212
222 several thousand dollars to as high as S134,173.~ 17-28 A few of the studies are summarized below. The costs attributable to the use of standard newborn nurseries for the care of low birthweight babies are not described in this report. In an analysis of the cost of providing health services to low birthweight infants and low-income pregnant women, the Michigan Depar tment of Health r epor ted that the average approximate cost of neonatal intensive care was $20,000, with an average length of stay of 17 to 27 days; it was not indicated whether scheme were reimbursed costs or charge=.~ Pomerance et al. reported an average cost of $62,730 for the hospital component of neonatal intensive care provided to low birthweight infants born at less than 35 weeks of gestation. ~ 7 Bragonier, Cushner, and Hobel estimated the average charges for initial hospitalization (including neonatal intensive care), intrapartum care, and all professional services for low birthweight infants at $90,880 . ~ ~ Their analysis used the findings of Phibbs et al. that 16 percent of the total medical care expense for a low birthweight infant's neonatal intensive care stay can be attributed to professional fee charges, and the remainder to hospital and ancillary service charges. ~ 9 In a paper describing the hospital costs of caring for inf ants weighing 1,000 grams or less, Pomerance et al. analyzed the hospital charges, excluding physician fees and bad debt, for caring for 75 infants.2 ~ The average charge per infant admitted (regardless of survival) was $61,196. The survival rate for the 75 infants in the sample was 40 percent, with an average total charge of $99,993 per survivor. The average length of stay for survivors was 89 days, with a range of 51 to 194 days. Rajagopalan et al. analyzed the cost of neonatal intensive care for 492 infants admitted to Babies Hospital, Columbia Presbyterian Medical Center, New York.26 Seventeen percent of these infants had one or more operations and incurred an average cost of $35,834. These infants were responsible for 37 percent of the aggregate cost of care for the study population. The infants weighing less than 1,500 grams incurred an average cost of $37,392. Fifty percent of the sample had total expenses of less than $9,000. The average total neonatal intensive care e ~ ense for hospitalization and professional fees was $16,826. The lower the newborn infant ' s birthweight, the greater the resource consumption and cost of services.27 Smaller infants require ore extensive medical support for longer periods of time. Survivors consume more services than infants who die. Diagnosis also has a major impact on resource consumption and medical costs. Budetti estimated that the average sum of physician and hospital charges for a newborn infant ' s initial hospitalization in a Level II or III neonatal intensive care unit is $13, 616 {adjusted to 1984 dollars).28 ThiS figure is used in the committee's cost comparison analysis. Although it probably represents a low estimate, the quality of the data on which it is based makes it the most appropriate information availabi e for the committee's purposes. In Budetti's study, the average length of stay in neonatal intensive care units for low birthweight infants was 8 to 18 days, with a mean of 13 days .

OCR for page 212
228 costs used here, there would still be a net cost savings in attaining the Surgeon General ' s goal. ~ Savings in the Cost of Caring for Low Birthweight Children A decrease in the number of low bar thweight infants would lead directly to decreases in neonatal mortality and morbidity and in the volume of medical services used to care for these children, especially in terms of days of hospitalization in neonatal intensive care units and rehospitalization during the first year of life. To determine the reduction in costs that might result from the provision of prenatal care to the target population, the committee examined incremental reductions in the low b' rthwe~ght rate from 11. 5 percent to 10 percent to 9 percent. These decrements would represent a 13 percent and a 22 percent reduction in the low b~rthweight rate, which is within the range of changes that may be anticipated (chapters 3 and 6 ~ . The committee estimated only the resulting annual expendi- tures for direct medical care dur ing the initial hospitalization, for rehospitalization of neonatal survivors dur ing the f irst year, and for medical services provided for noninstitutionalized, long-term survivors dur ing 1 year . Initial Hospitalization Cost Savings The 12,719 low birthweight infants (P2 in Figure 10.1) born to the target population are assumed to require Level II or III neonatal intensive care dur ing the per iod of initial hospitalization. Using Budetti ' s charges of $13, 616 (C2 unit costs in Table 10 .1) for Level II and III neonatal intensive care, the total initial hospitalization cost (TC2 in Table 10.1) for these low birthweight infants would be $173,181,904. A projected reduction in the low birthweight rate to 9 percent would reduce the number of low birthweight infants to 9,954 . Initial hospitalization costs at this 9 percent rate would be $135,533,664--a cost savings of $37,648,240. If only a 10 percent low birthweight rate were achi eyed by the provision of prenatal care to these higher isk pregnant women, the total initial hospitalization costs would be $150 ,592,960--a cost savings of $22, 588, 944 (Table 10 . 5) .3 9 The application to the initial hospitalization of all low birth- weight infants of Budetti's estimated charges for hospitalization in Level II and III intensive care units tends to overestimate these expenses. For example, some infants, generally a subpopulation within the 2,250- to 2,500-gram weight group, may be cared for in Level I nurseries and some will have a pattern of care similar to normal birthweight infants . To evaluate the f iscal impact of these ad just- ments on the cost compar ison, the committee estimated the effect of excluding the initial hospitalization and subsequent care costs attributable to all infants weighing 2,250 to 2,500 grams, which should eliminate the costs related to almost all of the moderately low b~rthweight infants not receiving intensive care, as well as the costs of many other infants who would have received intensive care (and,

OCR for page 212
229 TABLE 10.5 The Cost of Initial Hospitalization of Low Birthweight (LBW) Infants and Savings at Different LBW Rates LBW Rate Total Costa Cost Saving sb (percent) (dollars) (dollars) 11.5 173,181,904 10 150,592,960 22,588,944 9 135,533,664 37,648,240 aBased on an average neonatal intensive care charge (professional and hospitals of $13,616 (1984 dollars). bThe cost savings are the reduced expenditures for initial hospital~- zation of fewer low birthweight infants at the 10 percent and 9 percent low birthweight rates, respectively. thus, whose costs should be included) . In 1981, this cohor t represented 43 percent of all low birthweight infants. The effect of this percen- tage reduction on the total low birthweight infant costs at each low birthweight rate (Table 10.9) does not invalidate the committees conclusions, although it reduces the magnitude of the net cost saVingS.ho 41 Budetti's original estimate did not include all charges for Level lI and III care of infants.28 In addition, the adjustment to 1984 dollars does not take into account the growing use of increasingly expensive resources in caring for low birthweight infants, resulting from changes in medical diagnosis, treatment, and technology since Budetti gathered his data. These considerations may explain, in part, why many of the cost estimates for neonatal intensive care found in more recent studies are higher than those of Budetti. Finally, the trend toward a reduction in the number of term low birthweight infants (Chapter 3) will raise the proportion of very low birthweight and preterm moderately low birthweight infants, which in turn could result in increasing expenditures for the initial hospitalization of low birthweight infants. On balance, in the judgment of the committee, the costs of the initial hospitalization of low birthweight infants used in this analysis have not been overestimated. Rehospitalization Cost Savings During the First Year The costs of rehospitalization of low birthweight infants who survive the neonatal period were calculated utilizing data reported by McCormick and the Virginia Per~natal Assoczation.29 The rehospitalization rate was 38.3 percent for very low birthweight infants (1,500 grams or less) and 19 percent for moderately low birthweight infants (~1,501 to 2,500 grams). The average lengths of stay were 16.2 days and 12.5 days, respectively. An average hospital charge of $372 per day is used as

OCR for page 212
230 TABLE 10.6 Rehospitalization Costs During Initial Year for Very Low Birthweight (VLBW) and Moderately Low Birthweight (MLBW) Infants at Different Low Birthweight (LBW) Rates Total Number of Hospital LBW Rate Number of Infants Total Days Costs (percent) Infants Rehospitalized of Care (dollars VLBW 11.5 1, 170 448 7, 258 2, 699, 976 10 1,017 390 6,318 2,350 ,296 9 915 351 5, 687 2, 115, 564 MLBW 11~5 10,531 2 ,001 25,013 9,304 ,836 10 9J158 1,740 21,750 8,091,000 9 8,243 1,567 19,588 7~286r736 the unit cost (C3 in Table lOel) .3 1 man ~ ~ ~ ~= ~~~ _' ~~ - estimate are descr ibed above . An 8 percent mortality rate was used to estimate the number of low Birthweight infants who would not survive the neonatal period.33 42 Estimated total costs (TC3 in Table 10.1) for rehospitalization of the 2 ,449 low Birthweight infants (at a low Birthweight rate of 11.5 percent) who would survive the neonatal period and require rehospital- ization are $12,004,812 (Table 10.6). At a 10 percent low birthwe~oht -able ~ Elite u"t ~ Inn or C11 IS rate, rehospitalization cost savings are projected to be $1, 563, 516 (Table 10.7~; at a 9 percent rate, these savings are projected to be $2,602 ,512. Long-Term, Single-Year Morbidity Cost Savings Breslau estimates that 18.9 percent of low birthweight infants surviving the first Year of life will have long-term morbidity.36 Salkever estimates n~l;~1 1lni ~ m^~;~:~1 ~~At in :~ m_t~~ ~ ~ ~ ~ ~ =~ the an- .~_~_ In.- ~~4 ~~~ -1-~: He . 1~ rear nonlnstltutlonalized children with limitations of activities due to chronic disease to be more than $1,405 per child.37 The committee projected that there would be an additional 2 percent mortality among the low Birthweight infants during the first year of life, leaving 11,467 surviving infants, of whom 2,167 (P4 in Figure 10.1) would suffer long-term morbidity.32 The long-term morbidity costs for these infants would be $3, 044, 635 per year . This f igure does not include the costs of care

OCR for page 212
231 TABLE 10.7 Rehospitalization Cost Savings Total Rehospital- Cost LBW Rate VT~BW Costs MLBW Costs ization Costs Savingsa (percent) (dollars) (dollars) (dollars) (dollars) 10 2,350,296 8,091,000 10,441,296 1,563,516 9 2, 115, 564 7 ,286, 736 9 ,402 ,300 2, 602 ,512 aThe cost savings are the reduced expenditures for rehospitalization of fewer low birthweight infants at the 10 percent and 9 percent low birthweight rates, respectively. for infants with disabilities severe enough to require institutionaliza- tion, because adequate data to make such an estimate were not available. Annual long-term morbidity cost savings from a reduction of the low birthweight rate to 10 percent would be $396,210; at 9 percent, the savings would be $661,755 (Table 10.81. Obviously, in order to estimate the cumulative savings in long-term morbidity costs, these annual calculations would need to be multiplied by the average life expectancy of low birthweight infants and take into consideration the discounted present value of such long-term expeditures. Adequate data on the life expectancy of these infants are not available, but there is little reason to suspect that the longevity of infants surviving the first year would be markedly reduced. Thus, in using an estimated cost saving for only 1 year, the committee has substantially understated the cost savings for noninstitutional ambulatory direct medical care of low birthweight children in the calculation of net cost savings.39 40 These estimates assume that low birthweight infants with long-term morbidity are comparable in their resource utilization to other groups of chronically ill children. Although there are some data suggesting that the proportion of very low birthweight infants with disabilities decreases with age, and expenditures for institutionalization may be changing because of deinstitutionalization policies, adequate information is not available to make a more discriminating analysis at this time. Long-term follow-up studies of low birthweight infants are needed to explore these and many other issues. Overall Cost Savings Table 10.9 summarizes the costs of caring for low birthweight infants in the target population (Figure 10 . 1) . Table 10.10 presents the reductions in the direct medical costs of care for low birthweight infants that might occur if the rates of such births were reduced to 10 percent and 9 percent, respectively. It also includes a calculation of the net reduction in fiscal outlays for medical care that might occur -

OCR for page 212
232 TABLE 10.8 Long-Term, Single-Year Morbidity Costs at Different Low Birthweight (LBW) Rates Nuder of Annual Direct Infants with Medical Cos t LBW Rate Long-Term Expenses Savingsa (percent) Morb' city (dollars) (dollars) 11.5 2,167 3 ,044 ,635 10 1,885 2 ,648,425 396 ,210 9 1,696 2,382,880 661,755 NOTE: This assumes a 2 percent mortality rate after rehospitalization during the first year of life, resulting in 11,466 low birthweight infants at risk of long-term morbidity. aThe cost savings are the annual reduced expenditures for long-term morbidity of fewer low birthwe~ght infants at the low birthweight rates of 10 percent and 9 percent, respectively. If these infants have a life expectancy of 60 years, for example, the cost savings might be 60-fold greater discounted by the present value of such long-term expenditures. as a consequence of increasing expenditures for prenatal care services to pregnant women in the target population. Within the limits of the assumptions of the co~it~ee's analysis, the provision of more adequate prenatal care services to a cohort of women who are at high risk of delivering a low birthweight infant could reduce total expenditures for direct medical care of their low b~rthweight infants by $3.38 for each additional $1 .00 spent on their prenatal care. ~ ~~ This would occur if increasing the amount of prenatal care obtained by these women decreased their rate of low birthweight from its present level of 11.S percent to the 9 percent level, which is the Surgeon General's goal for a maximum low birthwe~ght rate in high-risk groups in the United States. At a low birthweight rate of 10.76 percent, the savings In low birthweight infant care costs would equal the additional costs of prenatal care services. Finally, although the committee ' s calculations and examples focus only on governmental fiscal outlays for a selected Group of women receiving public assistance, there are, of course, many more women at high risk of bearing a low birthweight infant. Low lairthweight infants born to these women~incur costs that are met by private insurance or out-of-pocket expenditures, or are unmet and result in hospital bad debts. The provision of more adequate prenatal care to these women, too, is likely to reduce such -fiscal outlays .

OCR for page 212
233 TABLE 10 .9 Sugary of Medical Costs of Low Birthwezght (LBW) Child Care at Different Low Birthweight Rates LBW Rate Number of (percent) LBW Infants Initial Long-Term Hospital i- Rehospitali- Morbidity Total LBW Ration Costs zation Costs Costs Infant Costsa (dollars) (dollars) (dollars) (dollars) 11.5 12,719 173,181,904 12,004,812 3,044,635 188,231,351 10 11,060 150,592,960 10,441,296 2,648,425 163,682,681 9 9,954 135,533,664 9,402,300 2,382,880 147,318,844 aThese are calculations of the total cost of care (TC5 ) based on the formulas given in Table 10.1. Summary The bird of infants weighing less Ian 2, 500 gram, Id particularly those of 1,500 grams or less, imposes a large economic burden on our nation by contributing substantially to neonatal mortality, to disability among surviving infants, and to the cost of health care. The provision of adequate prenatal services, as currently practiced, ~ ~ ~ ~ 3 _ _~ _ ~ ~ to all pregnant women WhO receive public assistance ana woo nave Attained less than a hiah school education wou] d require increased ~ ~ , _ expenditures, but would decrease the overall rascal outlays or govern- mental agencies for the care or the low b~rthweisht infants born to these higher isk women. Savings in the cost of care of low birthweight children would probably more than offset the additional cost for the prenatal services. Similarly, further net savings in overall fiscal outlays for the care of low birthweight children would be likely to result from the provision of appropriate prenatal services to other groups of women who are at high risk of delivering low birthwe~ght infants. TABLE 10.10 Cost Savings at Different Low Birthweight (LBW) Rates Reduction in Outlays for Total LBW Cost of LBW Additional Net Cost Infant Costs Infant Care Prenatal Care Savings LaW Rate (TC5) tTC6) (TC1) (TC6TC1) (percent) (dollars) (dollars) (dollars) (dollars) 11.5 188,231,351 10 163, 682, 681 9 147,318,844 24 ,548, 670 12, 107 ,200 12 ,439 ,470 40 ,912,507 12,107 ,200 28 ,805,307 NOTE: See Table 10 .1 for def initions and formulas.

OCR for page 212
234 References and Notes 1. Public Health Service: Promoting Health/Prevent~ng Disease: Objectives for the Nation. Washington, D.C.: U.S. Government Pr inting Off ice, Fall 1980 . 2. Analysis of 1980 Census Public Use, Microdata Sample, provided by Northern Ohio Data and Information Service. Cleveland, Ohio, Spr ing 1984 . 3. National Center for Health Statistics: Advance Repor t of Final Natality Statistics, 1979, Vol. 30, No. 6 (supplement 2~. Table 3: Birth rates by age of mother, live birth order and race of child: United States, 1979. DHHS No. (PHS) 81-1120. Public Health Service. Washington, D.C.: U.S. Government Printing Off ice, September 29, 1981. 4 . National Center for Health Statistics: Advance Repor t of Final Natality Statistics, 1980, Vol. al, No. 8 (supplement) . Table 13: Number and percent low birthweight and live births by birthweight, by age of mother and race of child: United States, 1980 . DEHS No. (PHS) 83-1120. Public Health Service. Washington, D.C.: U.S. Government Printing Office, November 30, 1982. National Center for Health Statistics: Vital Statistics of the United States, 1981, Vol . 1, Natality. Table 1.45: Percent distribution of live births by month of pregnancy prenatal care began, by years of school completed by mother and race of child. Public Health Service. U.S. Government Pr inting Off ice. In press . 6 . Shapiro S. McCormick t4C, Star f ield BH, Rr ischer JP, and Bross D: Relevance of correlates of infant deaths for significant morbidity at one year of age. Am. J. Obstet. Gynecol . 136: 363-373, 1980 . See also McCormick MC, Shapiro S. and Star field BH: Injury and its correlates among 1-year-old children: Study of children with both normal and low birthweights. Am. J. DiS. Child. 135:159-163, 1981. See also Shapiro S. McCormick MO, Starfield OH, and Crawley B: Changes in morbidity associated with decreases in neonatal mortality. Pediatr ics 72: 408-415, 1983 . 7 . Committee to Study the Prevention of Low Bir thweight: Prenatal care and low birthweight: Effects on health care expenditures. Unpubl ished paper prepared by S Smookler and RE Berhman . Washington, D.C.: Institute of Medicine, 1984 . 8. Committee to Study the Prevention of Low Birthweight: A rapid survey of prenatal care charges in the United States. Unpublished paper prepared by C Rorenbrot. Washington, D.C.: Institute of Medicine, May 1984. 9. Health Insurance Association of America: Baby . Washington , D.C., 1983 . Michigan Department of Public Health: Prenatal Care: A Healthy Beginning for M~chigan's Children. Report of the Director 's Special Task Force . Lansing , Mich ., 1984 . 11. Malitz D: Cost benefit analysis of extending Texas Medicaid coverage to provide prenatal care to pregnant women. the Texas Depar tment of Human Resources, May 1983 . 10. The Cost of Having a Submitted to

OCR for page 212
235 12. Colorado Department of Health: Cost Benefit Analysis of Excess Prematurity Versus Prenatal Care. Denver, Colo., 1977. 13. Lazarus W: Right from the Start: Improving Health Care for Ohio's Pregnant Women and Their Children. Columbus, Ohio: Children's Defense Fund-Ohio, 1983 . 14. Committee to Study the Prevention of Low Birthweight: The role of Medicaid in delivering prenatal care to low income women. unpublished paper prepared by MA McManus. Washington, D.C.: Institute of Medicine, November 1983 . 15. Rorenbrot C: Risk reduction in pregnancies of low income women, Mobius 4:34-43, July 1984. 16. In an HMO setting, the costs of providing prenatal nutrition counseling and a smoking cessation program were estimated to be $118.00 per patient. Ershoff D, Aaronson N. Danaher B. and Wasserman FWs Behavioral, health and cost outcomes of an HMO-based prenatal health education program. Public Health Rep. 98:536-547, 1983. 17. Pomerance JJ, Schifrin BS, and Meredith JL: Womb rent. Am. J. Obstet. Gynecol. 137:486-490, 1980. 18. Bragonier JR, Cushner IM, and Hobel CJ: Social and personal factors in the etiology of preterm birth. In Preterm Birth: Causes, Prevention and Management, edited by F Fuchs and PG Stubblefield, pp. 64-85. New York: Macmillan, 1984. 19. Phibbs CS, Williams RL, and Phibbs RE: Newborn risk factors and costs of neonatal intensive care. Pediatr ics 68: 313-321, 1981. 20. Rorenbrot C, Aalto C, and Laros R: The cost-effectiveness of stopping preterm labor with beta-adrenergic treatment. N. Engl. J. Med. 310:691-696, 1984. 21. Pomerance JJ, Ukrainski CT, Ukra I, Henderson H. Nash AH, and Meridith JL: Cost of living for infants weighing 1,000 grams or less at birth. Pediatrics 61:908-910, 1978. 22. McCarthy JT, Roops BL, Boneyfield PR, and Butterfield LJ: Who pays the bill for neonatal intens~ve care? J. Pediatr. 95: 755-761, 1979. Kaufman SL and Shepard SS: Costs of neonatal intensive care by length of stay. Inquiry 19:167-178, 1982. 24. Weitz J: Improvements for Maternity and Infant Care. Washington, D.C.: Children's Defense Fund, July 11, 1983. Boyle ME, Torrance GW, Sinclair J. and Horwood SP: Economic evaluation of neonatal intensive care of very-low-birth-weight infants. N. Engl. J. Med. 308:1330-1337, 1983. 26. Rajagopalan R. St~ckle G. Rairam R. and Driscoll J: Some clinical determinants of the cost of neonatal intens~ve care. unpublished paper, 1984. Walker DB, Feldman A, Vohr BR, and Oh W: Cost-benefit analysis of neonatal intensive care for infants weighing less than 1,000 grams at birth. Pediatr ics 74: 20-25, 1984 . 28. Office of Technology Assessment, U.S. Congress: The Implications of Cost-Effectiveness Analysis of Medical Technology, Background Paper No. 2: Case studies of medical technologies. Case Study No. 10: The costs and effectiveness of neonatal intensive care. _ . .

OCR for page 212
236 Prepared by P Budetti, MA McManus, N Bar r and, and LA Heinen. GPO Stock No. 052-003-00845-9. Washington, D.C.: U.S. Government Pr Tinting Office, 1981. 29. McCormick MC, Shapiro S. and Star field BE: Rehospitalization In the first year of life for higher isk survivors. Pedants Tics 66:991-999, 1980. Also see Virginia Perinatal Association, Inc.: Cost/Benefit Analysis of Virginia Senate Bill 200 (E~anding Medicaid Eligibility to Include First Time Pregnant Women). This cost-benef it analysis is based on a 1984 Medicaid cost-saving formula devised by the Children's Defense Fund, Washington, D.C. The 1983 Virginia Statewide Perinatal Services Plan is a major compendium of the perinatal health statistics used in this analys ~ s . 30. Hack M, DeMonterice D, Merkatz IR, Jones P. and Fanaroff A: Rehospitalization of the very low birthweight infant--a continuum of per inatal and environmental morbidity. Am. J. DiS. Child. 135: 263-266, 1981. Freeland MS and Schendler CE: Health spending in the 1980s: Integration of cl inical practice patterns with management. Health Care Financ. Rev. 5 :1-68, 1984. National Association of Children's Hospitals and Related Institutions, Inc.: Survey for Surgeon Generalts Conference on Handicapped Children and Their Families, 1982. The survey reports 33 . _ data from 11 children's hospitals. Goldenberg R. Koski J. Ferguson C, Wayne J. Hale C, and Nelson R: Infant mortality: The relationship between neonatal and post-neonatal mortality during a period of increasing perinatal center utilization. J. Pediatr., in press. Butler J. Budetti P. McManus P. Stenmark S. and Newacheck P: Health care e ~ enditures for children with chronic disabilities. Center for the Study of Families and Children, Institute for Public Policy Studies, Vanderbilt University. Paper prepared for Public Policies Affecting Chronically Ill Children and Their Families, September, 1982. Smyth-Staruch K, Breslau N. Weitzman M, and Gortmaker S: Use of health services by chronically ill and disabled children. Med. Care 22:310-328, 1984. 36. Committee to Study the Prevention of Low Birthweight: Medical care costs of prematurity. Unpublished paper prepared by N Breslau. Washington, D.C.: Institute of Medicine, 1984. 37. Salkever D: Parental opportunity costs and other economic costs of children's disabling conditions. In Chronically Ill Children, A Stacked Deck, edited by NJ Hobbs and J Perrin. San Francisco: Jossey-Bass Inc., 1984. Breslau N. Salkever D, and Smyth-Staruch K: Women's labor force activity and responsibilities for disabled dependents: A study of families with disabled children, J. Health Soc. Behav. 23:169-183, 1982. 3 9. The cost savings are the reduced expenditures for initial hospitalization of fewer low birthweight infants at the 10 percent and 9 percent low birthweight rates, respectively. 38.

OCR for page 212
237 40. The net cost savings are the differences in total low birthweight infant costs at low birthweight rates of 10 percent and 9 percent, respectively, compared to the total low birthweight costs at 11.5 percent, less the $12,107,200 expenditure for additional prenatal services. 41. In an alternative approach to this problem, it could be assumed that all infants weighing 1,500 grams or less would require intensive care during initial hospitalization at a unit cost of $40~000 per infant and that 40 percent of infants weighing 1,501 to 2,500 grams would require intensive care at a unit cost of $20,000 per infant. If 1,800 infants born to women in the target population were in the former group and 10,900 in the latter group, the total cost of intensive care during initial hospitalization would be $159,200,000 (~$40,000 x 1,800] ~ [$8,000 x 10,9001~. This would not affect the validity of the committee's basic conclusion that there would be a net cost saving from additional prenatal care services if the low birthweight rate were reduced to 9 percent. 42. This calculation assumes an 8 percent mortality rate during initial hospitalization (per Goldenberg et al.33) , resulting In 11,597 low birthweight survivors and rehospitalization rates of 38.3 percent for very low birthweight infants (1,500 Trams or _ _ ~ ~ , ~ _ _ ~ _, _ _ less) and 19 percent for moderately low birthweight infants (1,501 to 2,500 grassy. Average lengths of stay are assumed to be 16.2 days for very low birthweight infants and 12.5 days for moderately low birthweight infants. A hospital charge of $372 per day is used in this calculation. The ratio of very low birthweight to moderately low birthweight infants is assumed to be .10, based on data from the Robert Wood Johnson Foundation's regionalization of perinatal care project.

OCR for page 212