- Email: [email protected]
Willingness or unwillingness to perform cesarean section for impending preterm delivery at 24 weeks' gestation: A cost-effectiveness analysis
American Journal of Obstetrics and Gynecology (2005) 193, 1187–92
www.ajog.org
Willingness or unwillingness to perform cesarean section for impending preterm delivery at 24 weeks’ gestation: A cost-effectiveness analysis Gianina Cazan-London, MD, Ellen L. Mozurkewich, MD, Xiao Xu, PhD, Scott B. Ransom, DO, MBA, MPH University of Michigan Health Systems, Department of Obstetrics and Gynecology, Ann Arbor, MI Received for publication March 4, 2005; revised June 2, 2005; accepted June 29, 2005
KEY WORDS Extremely low birth weight 24 weeks’ gestation Cost-effectiveness Willingness Cesarean section
Objective: This study was undertaken to compare the costs and health outcomes of 2 management options when encountering a 24-week gestation in labor. Study design: We constructed a decision model for willingness versus unwillingness to perform cesarean section for fetal indication (aggressive vs nonaggressive management). We modeled chance nodes for stillbirth, neonatal death, and long-term survival, with and without major morbidity. Main outcome measures were intact (healthy) infant and live infant. Cost-effectiveness analysis was conducted from a societal perspective to determine the cost-effectiveness of the 2 strategies. Results: The probabilities of both intact survival (16.8% vs 12.9%) and survival with major morbidity (39.2% vs 19.4%) were higher with willingness to perform cesarean section. Nonaggressive management was less costly for delivery at 24 weeks’ gestation. Aggressive management strategy would cost $4,680,387 more than nonaggressive management for each additional intact infant, and $766,241 more per additional live infant. Conclusion: Although the probability of survival is increased by physician willingness to perform cesarean section, the more cost-effective strategy is unwillingness because of a strong relationship to the increased probability of survival with major morbidity when physicians are willing to perform cesarean section for fetal indications. Ó 2005 Mosby, Inc. All rights reserved.
The survival rate for extremely low birth weight (ELBW) infants born at the threshold of viability (25 or fewer weeks’ gestation) has markedly improved since the early 1990s, largely as a result of greater use of assisted ventilation in the delivery room, antenatal corticoste-
Presented at the Twenty-Fifth Annual Meeting of the Society for Maternal Fetal Medicine, Reno, Nev, February 7-12, 2005. Reprints not available from the author. 0002-9378/$ - see front matter Ó 2005 Mosby, Inc. All rights reserved. doi:10.1016/j.ajog.2005.06.084
roids, and postnatal surfactant therapy.1-6 However, this improvement in survival has not been associated with an equal improvement in postnatal morbidity.2-3,6 Extremely preterm births are associated with profound personal and societal burdens.7-10 Although the prevalence of such births is less than 1%, they account for nearly half of all cases of perinatal mortality.11 Many of those who survive incur serious morbidities such as cerebral palsy (CP), varying degrees of mental retardation (MR), blindness, or neurosensory hearing loss.
1188 Two studies, 12 years apart, have reported similar findings, in which a significant fraction of ELBW infants who had a ‘‘normal’’ neurologic and developmental evaluation in early life, had moderate-to-severe disability by school age.12,13 The authors hypothesized that as children age, neurologic and developmental requirements increase, as does the complexity of tests that are performed to evaluate overall growth and maturation. There are many implications when parents and physicians encounter the unanticipated imminent birth of an extremely preterm fetus. Obstetricians have traditionally focused on absolute survival when considering surgical intervention; however, neurodevelopmental handicaps must be discussed for legitimate informed consent. A thoughtful and balanced discussion regarding the risks requires a reliable means for prediction of neonatal and long-term outcomes for parents and health care providers who face this difficult decision. In the early 1990s, the Obstetric Determinants of Neonatal Survival Study evaluated information available before birth to predict the likelihood of survival with and without major morbidity for infants with birth weights of 1000 g or less.11,15,16 One of the most striking findings was that the physician’s belief of fetal viability and willingness to perform cesarean section (intervene in labor for fetal indications) at 24 weeks’ gestation was associated with an increased likelihood of survival. This study also found that willingness to perform a cesarean section doubled the probability of producing a survivor with serious morbidity. There is currently much societal disagreement regarding the merits of near-universal implementation of technology that can greatly increase the number of infants surviving near the threshold of viability but may also produce many neurologically and cognitively handicapped survivors. The purpose of our study is to use a cost-effectiveness analysis methodology to compare infant outcome and costs surrounding aggressive obstetric management at the time of imminent preterm delivery at 24 weeks of gestation relative to those of expectant management.
Materials and methods We constructed a decision tree using PrecisionTree (Palisade Corporation, Ithaca, NY) to compare 2 different clinical strategies for management of a 24-week gestation with imminent delivery. The first strategy is aggressive obstetric management, in which a physician is willing to intervene for fetal indications (nonreassuring fetal status or malpresentation). Within the scope of this study, we used willingness to perform cesarean section as a surrogate, as once the decision is made to proceed with aggressive management, intensive fetal heart rate monitoring and emergent cesarean for nonreassuring
Cazan-London et al fetal status would be implemented in most clinical settings. In the alternative strategy, the physician follows a nonaggressive scheme, ie, expectant management with no fetal heart rate monitoring and unwillingness to perform a cesarean section. The decision tree depicts the paths and probabilities of all relevant clinical outcomes, including intrauterine fetal death, neonatal death after admission to neonatal intensive care, and survival with various health conditions (Figure 1). For the purpose of this study, survival was based on 120 days of life or discharge from the hospital, as described by Bottoms et al.15 We defined major morbidity as survival with cerebral palsy, mental retardation, blindness/retinopathy of prematurity (ROP) grades 3 and 4, hearing loss, and ‘‘other’’ morbidity, which included necrotizing enterocolitis requiring surgery, oxygen dependence at discharge or 120 days, and seizures. Survival without major morbidity was considered ‘‘intact’’ (healthy) survival. This decision analysis was conducted from a societal perspective, and the time horizon of the model is the newborn infant’s entire lifetime. The health outcomes examined are intact survival and overall survival (intact plus nonintact). We calculated the total cost of each management strategy, as well as the incremental cost-effectiveness ratios for both intact survival and overall survival. Sensitivity analyses were also conducted to assess the uncertainties and reliability associated with the model parameters.
Data sources We conducted a thorough review of the obstetric and pediatric literature on preterm birth at the threshold of viability. The probability estimates at the chance nodes of the decision tree were derived primarily from the most recent available data concerning outcomes in these clinical situations.2-4,6-8,13-15,18-22 The probability (p) estimates with plausible ranges are shown in Table I. At chance nodes with binary outcomes (eg, death or survival) the converse probability was defined as 1-(p). Cost estimates for each outcome were derived from St. John et al7 and the Morbidity and Mortality Weekly Report (MMWR).9 Cost data are presented in Table II and all values are reported in 2004 US dollars. Costs were adjusted to 2004 dollars using the Consumer Price Index for all urban consumers.24 Initial hospitalization is defined as all inpatient care before the first discharge to home (survivors) or death (nonsurvivors). The cost per infant for initial hospitalization included hospital costs and physician fees, and was calculated using the corresponding institutional cost-charge ratio. The per-person cost estimates for each developmental disability (after initial hospitalization) were obtained from a publication in MMWR that estimated the
Cazan-London et al
1189
Figure 1 Decision tree describing the path for aggressive (willing to perform cesarean on behalf of the fetus) and non-aggressive (unwilling to perform cesarean) obstetrical management at the time of imminent preterm delivery at 24 weeks of gestation.
lifetime costs associated with MR, CP, hearing loss, and vision impairment in the United States.9 They included direct medical and nonmedical costs (eg, physician visits, inpatient hospital stays, assistive devices, and home and automobile modifications), as well as indirect
costs associated with productivity loss and premature mortality. Because the literature reviewed in this study did not provide information on maternal delivery and hospitalization costs, we did not incorporate any maternal hospitalization cost.
1190 Table I
Cazan-London et al Probability variables
Variable Live birth Aggressive management Nonaggressive management Survival Aggressive management Nonaggressive management Intact survival Aggressive management Nonaggressive management Major morbidities CP MR Blindness/ROP Hearing loss Other
Table II
Value
Range
Reference
1.00
d
15
0.95
d
15
0.56
0.17-0.68
17,7,13,14,18
0.34
0.17-0.68
17,7,13,14,18
0.30
0.03-0.58
15,2,3,4,8,13,14
0.40
0.03-0.58
15,2,3,4,8,13,14
0.23 0.27 0.21 0.10 0.19
d d d d d
14,2,4,13,18,19 20,4,8,13,14 4,6,14,20,21,22 4,3,6,13,14 2,3,4,6,8,21,22
Cost variables
Variable Initial hospitalization Survivor Neonatal death Stillborn burial
Cost in 2004 dollars ($) 197,402 27,869 100
Lifetime costs per developmental disability CP 951,080 MR 1,047,117 Hearing loss 430,619 Blindness/ROP 584,486 Other 206,532
Reference 7 7 Local funeral home 9 9 9 9 Author’s estimate
Results Our analysis presented a 16.8% (range: 1.1%-35.4%) versus 12.9% (range: 0.5%-26.7%) probability of an intact survivor with aggressive management (willingness to perform a cesarean section) and nonaggressive management (unwillingness to perform cesarean section), respectively; whereas the probabilities of nonintact survival were 39.2% (range: 7.1%-66.0%) versus 19.4% (range: 6.74%-62.7%), respectively. Taking into account the high probability of not surviving during these preterm deliveries, with aggressive management, the overall cost per intact infant is $2,379,531 and per surviving infant is $713,859. In
Figure 2 survival.
One-way sensitivity analysis: probability of intact
Figure 3 vival.
One-way sensitivity analysis: probability of sur-
contrast, for nonaggressive management, the costs are $1,688,562 and $675,425, respectively. Total costs per infant delivered under each management strategy and the incremental cost-effectiveness ratios are summarized in Table III. Sensitivity analyses were performed over a plausible range of probabilities as shown in Table I. To account for the uncertainty of surviving at 24 weeks’ gestation, we varied the probability of overall survival and intact survival, respectively, using 1-way sensitivity analyses. Results are presented in Figures 2 and 3. As the probability of overall survival varies from 0.17 to 0.68, the total cost per birth under aggressive management (willingness) was much higher than nonaggressive management (unwillingness) at all probability points (Figure 2). Similarly, as the likelihood of intact survival changes from 0.03 to 0.58, the optimal choice of obstetric management is always nonaggressive management if costs are considered (Figure 3). We also conducted a 2-way sensitivity analysis allowing both the probability of overall survival and the probability of intact survival to vary. At all combinations of the 2 probabilities, nonaggressive management cost less when compared with aggressive management.
Comment Very preterm deliveries present exceptionally high risk of neonatal deaths or major morbidities among survivors. The overall costs to parents, health care providers and society at large are substantial with either management decision. Although aggressive management improves the probability of producing an intact survivor, the more cost-effective management option, from a societal perspective, is nonaggressive management in
Cazan-London et al Table III
1191
Cost-effectiveness of obstetric management strategies for impending preterm delivery at 24 weeks’ gestation
Strategy
Cost per birth ($)
Intact survivors per 100 births
Cost per additional intact survivor ($)
Survivors per 100 births
Cost per additional survivor ($)
Nonaggressive Aggressive
218,162 399,761
13 17
4,680,387
32 56
766,241
the setting of a 24 weeks’ gestation with imminent spontaneous or indicated delivery. Compared with expectant management, aggressive management would cost an extra $766,241 for every additional surviving infant and $4,680,387 for each additional intact infant. Such results are largely attributable to the 2-fold increased probability of surviving with major and costly morbidity when physicians are willing versus unwilling to perform cesarean section for impending delivery (39.2% vs 19.4%).11,15 On the other hand, the likelihood of healthy survivors of all neonates delivered at 24 weeks of gestation is also higher in the aggressive management scheme as opposed to nonaggressive management (17% vs 13%).15 It is therefore a very difficult decision that families and physicians must make when faced with having to choose the management strategy for a pregnancy at the cusp of viability. Limitations of this study should also be acknowledged. Data on long-term outcomes of infants born at 24 weeks necessarily reflect neonatal practices of 5 to10 years ago. It is possible that improvement in neonatal care might result in a decreased likelihood of major morbidity among survivors. However, recent data showing decreased brain volumes and abnormal brain morphology among ELBW survivors suggests that some morbidities associated with extreme prematurity may not be reduced by improvement in neonatal care.25 The available data in the MMWR report did not account for costs related to family caregiver and residential care for persons not living in households, as they were not available. The prevalence of disabilities reported was based on the Metropolitan Atlanta Developmental Disabilities Study and may not be representative of national prevalences.23 Finally, because of data nonavailability, our analyses did not incorporate cost information regarding maternal hospitalization. Further research is required to fully understand the long-term consequences of delivery choices for extremely preterm pregnancies. A prospective trial comparing the obstetric management of these pregnancies would be optimal to fully understand the consequences; however, ethical considerations may prevent such an evaluation. In conclusion, both absolute survival and long-term handicaps must be considered when proceeding with aggressive intervention in extremely preterm pregnancies. Because physicians’ beliefs and management strategies (fetal viability and willingness to perform cesarean) appear to significantly affect fetal outcomes in the
presented scenario, physicians must strive to present an objective and balanced discussion of the risks and benefits. The long-term implications of decisions made in often urgent circumstances must still be clearly explained to and understood by the patient and her family. As our profession develops new and innovative strategies to care for patients, we must go further than perfect the interventions. We must also understand the long-term implications and be willing to avoid some available services to have a net positive impact.
References 1. Meadow W, Lee G, Lin K, Lantos J. Changes in mortality for extremely low birth weight infants in the 1990’s: implications for treatment decisions and resource use. Pediatrics 2004;113:1223-9. 2. Fanaroff AA, Hack M, Walsh MC. The NICHD neonatal research network: changes in practice and outcomes during the first 15 years. Semin Perinatol 2003;27:281-7. 3. Battin M, Ling EWY, Whitfield MF, Mackinnon M, Effer SB. Has the outcome for ELGA infants improved following recent advances in neonatal intensive care? Am J Perinat 1998;15:469-77. 4. Vohr BR, Wright LL, Dusick AM, Mele L, Verter J, Steichen JJ, et al. Neurodevelopmental and functional outcomes of ELBW infants in the NICHD and Human Development Neonatal Network, 1993-1994. Pediatrics 2000;105:1216-26. 5. Depp R, Lemons J. Perinatal care at the threshold of viability. Washington (DC): The College; 2002. ACOG practice bulletin, number 38. 6. Hack M, Fanaroff A. Outcomes of children of extremely low birthweight and gestational age in the 1990’s. Early Hum Dev 1999;53:193-218. 7. St. John EB, Nelson KG, Cliver SP, Bishnoi R, Goldenberg RL. Cost of neonatal care according to gestational age at birth and survival status. Am J Obstet Gynecol 2000;182:170-5. 8. Piecuch RE, Leonard CH, Cooper BA, Kilpatrick SJ, Schlueter MA, Sola A. Outcome of infants born at 24-26 weeks gestation: I, survival and cost. Obstet Gynecol 1997;90:803-8. 9. Honeycutt A, Dunlap L, Chen H, Al Homsi G, Grosse S, Schendel D. Economic costs associated with mental retardation, cerebral palsy, hearing loss, and vision impairmentdUnited States, 2003. CDC MMWR Weekly 2004;53:57-9. 10. Gilbert WM, Nesbitt TS, Danielsen B. The cost of prematurity: quantification by gestational age and weight. Obstet Gynecol 2003;102:488-92. 11. Iams JD, Mercer BM. What we have learned about antenatal prediction of neonatal morbidity and mortality. Semin Perinatol 2003;27:247-52. 12. den Ouden L, Hille ETM, Bauer L, Verloove-Vanhorick SP. School performance in very preterm children. Lancet 1993;342:550-1. 13. Marlow N, Wolke D, Bracewell MA, Samara M. Neurologic and developmental disability at six years of age after extremely preterm birth. N Engl J Med 2005;352:71-2.
1192 14. Wood NS, Marlow N, Costeloe K, Chir B, Gibson AT, Wilkinson AR. Neurologic and developmental disability after extremely preterm birth. N Engl J Med 2000;343:378-84. 15. Bottoms SF, Paul RH, Iams JD, Mercer BM, Thom EA, Roberts JM, et al. Obstetric determinants of neonatal survival: influence of willingness to perform cesarean delivery on survival of extremely low-birth-weight infants. Am J Obstet Gynecol 1997;176:960-6. 16. Bottoms SF, Paul RH, Mercer BM, MacPherson CA, Caritis SN, Moawad AH, et al. Obstetric determinants of neonatal survival and morbidity in extremely low birth weight infants. National Institute of Child Health and Development Network of Maternal-Fetal Medicine Units. Am J Obstet Gynecol 1999;180: 665-9. 17. Lorenz JM. Management decisions in extremely premature infants. Semin Neonatol 2003;8:475-82. 18. Lemons JA, Bauer CR, Oh W, Korones SB, Papile LA, Stoll BJ, et al. Very low birth weight outcome of the National Institute of Child, Health and Human Development Neonatal Research Network, January 1995 through December 1996. NICHD Neonatal Research Network. Pediatrics 2001;107:E1. 19. Lorenz JM, Paneth N, Jetton JR, den Ouden L, Tyson JE. Comparison of management strategies for extreme prematurity
Cazan-London et al
20.
21.
22.
23.
24.
25.
in New Jersey and the Netherlands: outcomes and resources expenditure. Pediatrics 2001;108:1269-74. Emsley HCA, Wardle SP, Sims DG, Chiswick ML, D’Souza WD. Increased survival and deteriorating developmental outcome in 23 to 25 week old gestation infants; 1990-4 compared with 1984-9. Arch Dis Child Fetal Neonatal Ed 1998;78:F99-104. Killpatrick SJ, Schlueter MA, Piecuch R, Leonard CH, Rogido M, Sola A. Outcome of infants born at 24-26 weeks gestation: I, survival and cost. Obstet Gynecol 1997;90:803-8. El-Metwally D, Vohr B, Tucker R. Survival and neonatal morbidity at the limits of viability in the mid 1990’s: 22 to 25 weeks. J Pediatrics 2000;137:616-22. Yeargin-Allsopp M, Murphy CC, Oakley GP, Sikes RK. A multiple-source method for studying the prevalence of developmental disabilities in children: the Metropolitan Atlanta Developmental Disabilities Study. Pediatrics 1992;89:624-30. US Department of Labor Bureau of Labor Statistics. Consumer price indexdall urban consumers. Available at: http://www.bls.gov/ data. Accessed on March 3, 2005. Peterson BS, Vohr B, Stave LH. Regional brain volume abnormality and long-term cognitive outcomes in preterm infants. JAMA 2000;284:1939-47.
www.ajog.org
Willingness or unwillingness to perform cesarean section for impending preterm delivery at 24 weeks’ gestation: A cost-effectiveness analysis Gianina Cazan-London, MD, Ellen L. Mozurkewich, MD, Xiao Xu, PhD, Scott B. Ransom, DO, MBA, MPH University of Michigan Health Systems, Department of Obstetrics and Gynecology, Ann Arbor, MI Received for publication March 4, 2005; revised June 2, 2005; accepted June 29, 2005
KEY WORDS Extremely low birth weight 24 weeks’ gestation Cost-effectiveness Willingness Cesarean section
Objective: This study was undertaken to compare the costs and health outcomes of 2 management options when encountering a 24-week gestation in labor. Study design: We constructed a decision model for willingness versus unwillingness to perform cesarean section for fetal indication (aggressive vs nonaggressive management). We modeled chance nodes for stillbirth, neonatal death, and long-term survival, with and without major morbidity. Main outcome measures were intact (healthy) infant and live infant. Cost-effectiveness analysis was conducted from a societal perspective to determine the cost-effectiveness of the 2 strategies. Results: The probabilities of both intact survival (16.8% vs 12.9%) and survival with major morbidity (39.2% vs 19.4%) were higher with willingness to perform cesarean section. Nonaggressive management was less costly for delivery at 24 weeks’ gestation. Aggressive management strategy would cost $4,680,387 more than nonaggressive management for each additional intact infant, and $766,241 more per additional live infant. Conclusion: Although the probability of survival is increased by physician willingness to perform cesarean section, the more cost-effective strategy is unwillingness because of a strong relationship to the increased probability of survival with major morbidity when physicians are willing to perform cesarean section for fetal indications. Ó 2005 Mosby, Inc. All rights reserved.
The survival rate for extremely low birth weight (ELBW) infants born at the threshold of viability (25 or fewer weeks’ gestation) has markedly improved since the early 1990s, largely as a result of greater use of assisted ventilation in the delivery room, antenatal corticoste-
Presented at the Twenty-Fifth Annual Meeting of the Society for Maternal Fetal Medicine, Reno, Nev, February 7-12, 2005. Reprints not available from the author. 0002-9378/$ - see front matter Ó 2005 Mosby, Inc. All rights reserved. doi:10.1016/j.ajog.2005.06.084
roids, and postnatal surfactant therapy.1-6 However, this improvement in survival has not been associated with an equal improvement in postnatal morbidity.2-3,6 Extremely preterm births are associated with profound personal and societal burdens.7-10 Although the prevalence of such births is less than 1%, they account for nearly half of all cases of perinatal mortality.11 Many of those who survive incur serious morbidities such as cerebral palsy (CP), varying degrees of mental retardation (MR), blindness, or neurosensory hearing loss.
1188 Two studies, 12 years apart, have reported similar findings, in which a significant fraction of ELBW infants who had a ‘‘normal’’ neurologic and developmental evaluation in early life, had moderate-to-severe disability by school age.12,13 The authors hypothesized that as children age, neurologic and developmental requirements increase, as does the complexity of tests that are performed to evaluate overall growth and maturation. There are many implications when parents and physicians encounter the unanticipated imminent birth of an extremely preterm fetus. Obstetricians have traditionally focused on absolute survival when considering surgical intervention; however, neurodevelopmental handicaps must be discussed for legitimate informed consent. A thoughtful and balanced discussion regarding the risks requires a reliable means for prediction of neonatal and long-term outcomes for parents and health care providers who face this difficult decision. In the early 1990s, the Obstetric Determinants of Neonatal Survival Study evaluated information available before birth to predict the likelihood of survival with and without major morbidity for infants with birth weights of 1000 g or less.11,15,16 One of the most striking findings was that the physician’s belief of fetal viability and willingness to perform cesarean section (intervene in labor for fetal indications) at 24 weeks’ gestation was associated with an increased likelihood of survival. This study also found that willingness to perform a cesarean section doubled the probability of producing a survivor with serious morbidity. There is currently much societal disagreement regarding the merits of near-universal implementation of technology that can greatly increase the number of infants surviving near the threshold of viability but may also produce many neurologically and cognitively handicapped survivors. The purpose of our study is to use a cost-effectiveness analysis methodology to compare infant outcome and costs surrounding aggressive obstetric management at the time of imminent preterm delivery at 24 weeks of gestation relative to those of expectant management.
Materials and methods We constructed a decision tree using PrecisionTree (Palisade Corporation, Ithaca, NY) to compare 2 different clinical strategies for management of a 24-week gestation with imminent delivery. The first strategy is aggressive obstetric management, in which a physician is willing to intervene for fetal indications (nonreassuring fetal status or malpresentation). Within the scope of this study, we used willingness to perform cesarean section as a surrogate, as once the decision is made to proceed with aggressive management, intensive fetal heart rate monitoring and emergent cesarean for nonreassuring
Cazan-London et al fetal status would be implemented in most clinical settings. In the alternative strategy, the physician follows a nonaggressive scheme, ie, expectant management with no fetal heart rate monitoring and unwillingness to perform a cesarean section. The decision tree depicts the paths and probabilities of all relevant clinical outcomes, including intrauterine fetal death, neonatal death after admission to neonatal intensive care, and survival with various health conditions (Figure 1). For the purpose of this study, survival was based on 120 days of life or discharge from the hospital, as described by Bottoms et al.15 We defined major morbidity as survival with cerebral palsy, mental retardation, blindness/retinopathy of prematurity (ROP) grades 3 and 4, hearing loss, and ‘‘other’’ morbidity, which included necrotizing enterocolitis requiring surgery, oxygen dependence at discharge or 120 days, and seizures. Survival without major morbidity was considered ‘‘intact’’ (healthy) survival. This decision analysis was conducted from a societal perspective, and the time horizon of the model is the newborn infant’s entire lifetime. The health outcomes examined are intact survival and overall survival (intact plus nonintact). We calculated the total cost of each management strategy, as well as the incremental cost-effectiveness ratios for both intact survival and overall survival. Sensitivity analyses were also conducted to assess the uncertainties and reliability associated with the model parameters.
Data sources We conducted a thorough review of the obstetric and pediatric literature on preterm birth at the threshold of viability. The probability estimates at the chance nodes of the decision tree were derived primarily from the most recent available data concerning outcomes in these clinical situations.2-4,6-8,13-15,18-22 The probability (p) estimates with plausible ranges are shown in Table I. At chance nodes with binary outcomes (eg, death or survival) the converse probability was defined as 1-(p). Cost estimates for each outcome were derived from St. John et al7 and the Morbidity and Mortality Weekly Report (MMWR).9 Cost data are presented in Table II and all values are reported in 2004 US dollars. Costs were adjusted to 2004 dollars using the Consumer Price Index for all urban consumers.24 Initial hospitalization is defined as all inpatient care before the first discharge to home (survivors) or death (nonsurvivors). The cost per infant for initial hospitalization included hospital costs and physician fees, and was calculated using the corresponding institutional cost-charge ratio. The per-person cost estimates for each developmental disability (after initial hospitalization) were obtained from a publication in MMWR that estimated the
Cazan-London et al
1189
Figure 1 Decision tree describing the path for aggressive (willing to perform cesarean on behalf of the fetus) and non-aggressive (unwilling to perform cesarean) obstetrical management at the time of imminent preterm delivery at 24 weeks of gestation.
lifetime costs associated with MR, CP, hearing loss, and vision impairment in the United States.9 They included direct medical and nonmedical costs (eg, physician visits, inpatient hospital stays, assistive devices, and home and automobile modifications), as well as indirect
costs associated with productivity loss and premature mortality. Because the literature reviewed in this study did not provide information on maternal delivery and hospitalization costs, we did not incorporate any maternal hospitalization cost.
1190 Table I
Cazan-London et al Probability variables
Variable Live birth Aggressive management Nonaggressive management Survival Aggressive management Nonaggressive management Intact survival Aggressive management Nonaggressive management Major morbidities CP MR Blindness/ROP Hearing loss Other
Table II
Value
Range
Reference
1.00
d
15
0.95
d
15
0.56
0.17-0.68
17,7,13,14,18
0.34
0.17-0.68
17,7,13,14,18
0.30
0.03-0.58
15,2,3,4,8,13,14
0.40
0.03-0.58
15,2,3,4,8,13,14
0.23 0.27 0.21 0.10 0.19
d d d d d
14,2,4,13,18,19 20,4,8,13,14 4,6,14,20,21,22 4,3,6,13,14 2,3,4,6,8,21,22
Cost variables
Variable Initial hospitalization Survivor Neonatal death Stillborn burial
Cost in 2004 dollars ($) 197,402 27,869 100
Lifetime costs per developmental disability CP 951,080 MR 1,047,117 Hearing loss 430,619 Blindness/ROP 584,486 Other 206,532
Reference 7 7 Local funeral home 9 9 9 9 Author’s estimate
Results Our analysis presented a 16.8% (range: 1.1%-35.4%) versus 12.9% (range: 0.5%-26.7%) probability of an intact survivor with aggressive management (willingness to perform a cesarean section) and nonaggressive management (unwillingness to perform cesarean section), respectively; whereas the probabilities of nonintact survival were 39.2% (range: 7.1%-66.0%) versus 19.4% (range: 6.74%-62.7%), respectively. Taking into account the high probability of not surviving during these preterm deliveries, with aggressive management, the overall cost per intact infant is $2,379,531 and per surviving infant is $713,859. In
Figure 2 survival.
One-way sensitivity analysis: probability of intact
Figure 3 vival.
One-way sensitivity analysis: probability of sur-
contrast, for nonaggressive management, the costs are $1,688,562 and $675,425, respectively. Total costs per infant delivered under each management strategy and the incremental cost-effectiveness ratios are summarized in Table III. Sensitivity analyses were performed over a plausible range of probabilities as shown in Table I. To account for the uncertainty of surviving at 24 weeks’ gestation, we varied the probability of overall survival and intact survival, respectively, using 1-way sensitivity analyses. Results are presented in Figures 2 and 3. As the probability of overall survival varies from 0.17 to 0.68, the total cost per birth under aggressive management (willingness) was much higher than nonaggressive management (unwillingness) at all probability points (Figure 2). Similarly, as the likelihood of intact survival changes from 0.03 to 0.58, the optimal choice of obstetric management is always nonaggressive management if costs are considered (Figure 3). We also conducted a 2-way sensitivity analysis allowing both the probability of overall survival and the probability of intact survival to vary. At all combinations of the 2 probabilities, nonaggressive management cost less when compared with aggressive management.
Comment Very preterm deliveries present exceptionally high risk of neonatal deaths or major morbidities among survivors. The overall costs to parents, health care providers and society at large are substantial with either management decision. Although aggressive management improves the probability of producing an intact survivor, the more cost-effective management option, from a societal perspective, is nonaggressive management in
Cazan-London et al Table III
1191
Cost-effectiveness of obstetric management strategies for impending preterm delivery at 24 weeks’ gestation
Strategy
Cost per birth ($)
Intact survivors per 100 births
Cost per additional intact survivor ($)
Survivors per 100 births
Cost per additional survivor ($)
Nonaggressive Aggressive
218,162 399,761
13 17
4,680,387
32 56
766,241
the setting of a 24 weeks’ gestation with imminent spontaneous or indicated delivery. Compared with expectant management, aggressive management would cost an extra $766,241 for every additional surviving infant and $4,680,387 for each additional intact infant. Such results are largely attributable to the 2-fold increased probability of surviving with major and costly morbidity when physicians are willing versus unwilling to perform cesarean section for impending delivery (39.2% vs 19.4%).11,15 On the other hand, the likelihood of healthy survivors of all neonates delivered at 24 weeks of gestation is also higher in the aggressive management scheme as opposed to nonaggressive management (17% vs 13%).15 It is therefore a very difficult decision that families and physicians must make when faced with having to choose the management strategy for a pregnancy at the cusp of viability. Limitations of this study should also be acknowledged. Data on long-term outcomes of infants born at 24 weeks necessarily reflect neonatal practices of 5 to10 years ago. It is possible that improvement in neonatal care might result in a decreased likelihood of major morbidity among survivors. However, recent data showing decreased brain volumes and abnormal brain morphology among ELBW survivors suggests that some morbidities associated with extreme prematurity may not be reduced by improvement in neonatal care.25 The available data in the MMWR report did not account for costs related to family caregiver and residential care for persons not living in households, as they were not available. The prevalence of disabilities reported was based on the Metropolitan Atlanta Developmental Disabilities Study and may not be representative of national prevalences.23 Finally, because of data nonavailability, our analyses did not incorporate cost information regarding maternal hospitalization. Further research is required to fully understand the long-term consequences of delivery choices for extremely preterm pregnancies. A prospective trial comparing the obstetric management of these pregnancies would be optimal to fully understand the consequences; however, ethical considerations may prevent such an evaluation. In conclusion, both absolute survival and long-term handicaps must be considered when proceeding with aggressive intervention in extremely preterm pregnancies. Because physicians’ beliefs and management strategies (fetal viability and willingness to perform cesarean) appear to significantly affect fetal outcomes in the
presented scenario, physicians must strive to present an objective and balanced discussion of the risks and benefits. The long-term implications of decisions made in often urgent circumstances must still be clearly explained to and understood by the patient and her family. As our profession develops new and innovative strategies to care for patients, we must go further than perfect the interventions. We must also understand the long-term implications and be willing to avoid some available services to have a net positive impact.
References 1. Meadow W, Lee G, Lin K, Lantos J. Changes in mortality for extremely low birth weight infants in the 1990’s: implications for treatment decisions and resource use. Pediatrics 2004;113:1223-9. 2. Fanaroff AA, Hack M, Walsh MC. The NICHD neonatal research network: changes in practice and outcomes during the first 15 years. Semin Perinatol 2003;27:281-7. 3. Battin M, Ling EWY, Whitfield MF, Mackinnon M, Effer SB. Has the outcome for ELGA infants improved following recent advances in neonatal intensive care? Am J Perinat 1998;15:469-77. 4. Vohr BR, Wright LL, Dusick AM, Mele L, Verter J, Steichen JJ, et al. Neurodevelopmental and functional outcomes of ELBW infants in the NICHD and Human Development Neonatal Network, 1993-1994. Pediatrics 2000;105:1216-26. 5. Depp R, Lemons J. Perinatal care at the threshold of viability. Washington (DC): The College; 2002. ACOG practice bulletin, number 38. 6. Hack M, Fanaroff A. Outcomes of children of extremely low birthweight and gestational age in the 1990’s. Early Hum Dev 1999;53:193-218. 7. St. John EB, Nelson KG, Cliver SP, Bishnoi R, Goldenberg RL. Cost of neonatal care according to gestational age at birth and survival status. Am J Obstet Gynecol 2000;182:170-5. 8. Piecuch RE, Leonard CH, Cooper BA, Kilpatrick SJ, Schlueter MA, Sola A. Outcome of infants born at 24-26 weeks gestation: I, survival and cost. Obstet Gynecol 1997;90:803-8. 9. Honeycutt A, Dunlap L, Chen H, Al Homsi G, Grosse S, Schendel D. Economic costs associated with mental retardation, cerebral palsy, hearing loss, and vision impairmentdUnited States, 2003. CDC MMWR Weekly 2004;53:57-9. 10. Gilbert WM, Nesbitt TS, Danielsen B. The cost of prematurity: quantification by gestational age and weight. Obstet Gynecol 2003;102:488-92. 11. Iams JD, Mercer BM. What we have learned about antenatal prediction of neonatal morbidity and mortality. Semin Perinatol 2003;27:247-52. 12. den Ouden L, Hille ETM, Bauer L, Verloove-Vanhorick SP. School performance in very preterm children. Lancet 1993;342:550-1. 13. Marlow N, Wolke D, Bracewell MA, Samara M. Neurologic and developmental disability at six years of age after extremely preterm birth. N Engl J Med 2005;352:71-2.
1192 14. Wood NS, Marlow N, Costeloe K, Chir B, Gibson AT, Wilkinson AR. Neurologic and developmental disability after extremely preterm birth. N Engl J Med 2000;343:378-84. 15. Bottoms SF, Paul RH, Iams JD, Mercer BM, Thom EA, Roberts JM, et al. Obstetric determinants of neonatal survival: influence of willingness to perform cesarean delivery on survival of extremely low-birth-weight infants. Am J Obstet Gynecol 1997;176:960-6. 16. Bottoms SF, Paul RH, Mercer BM, MacPherson CA, Caritis SN, Moawad AH, et al. Obstetric determinants of neonatal survival and morbidity in extremely low birth weight infants. National Institute of Child Health and Development Network of Maternal-Fetal Medicine Units. Am J Obstet Gynecol 1999;180: 665-9. 17. Lorenz JM. Management decisions in extremely premature infants. Semin Neonatol 2003;8:475-82. 18. Lemons JA, Bauer CR, Oh W, Korones SB, Papile LA, Stoll BJ, et al. Very low birth weight outcome of the National Institute of Child, Health and Human Development Neonatal Research Network, January 1995 through December 1996. NICHD Neonatal Research Network. Pediatrics 2001;107:E1. 19. Lorenz JM, Paneth N, Jetton JR, den Ouden L, Tyson JE. Comparison of management strategies for extreme prematurity
Cazan-London et al
20.
21.
22.
23.
24.
25.
in New Jersey and the Netherlands: outcomes and resources expenditure. Pediatrics 2001;108:1269-74. Emsley HCA, Wardle SP, Sims DG, Chiswick ML, D’Souza WD. Increased survival and deteriorating developmental outcome in 23 to 25 week old gestation infants; 1990-4 compared with 1984-9. Arch Dis Child Fetal Neonatal Ed 1998;78:F99-104. Killpatrick SJ, Schlueter MA, Piecuch R, Leonard CH, Rogido M, Sola A. Outcome of infants born at 24-26 weeks gestation: I, survival and cost. Obstet Gynecol 1997;90:803-8. El-Metwally D, Vohr B, Tucker R. Survival and neonatal morbidity at the limits of viability in the mid 1990’s: 22 to 25 weeks. J Pediatrics 2000;137:616-22. Yeargin-Allsopp M, Murphy CC, Oakley GP, Sikes RK. A multiple-source method for studying the prevalence of developmental disabilities in children: the Metropolitan Atlanta Developmental Disabilities Study. Pediatrics 1992;89:624-30. US Department of Labor Bureau of Labor Statistics. Consumer price indexdall urban consumers. Available at: http://www.bls.gov/ data. Accessed on March 3, 2005. Peterson BS, Vohr B, Stave LH. Regional brain volume abnormality and long-term cognitive outcomes in preterm infants. JAMA 2000;284:1939-47.