Since the first umbilical cord blood transplant was performed in 1988 for a boy with Fanconi’s anemia,1 umbilical cord blood has been transformed from a waste product to be discarded to a source of stem cells with the potential to treat a variety of conditions ranging from leukemia to metabolic disorders to cerebral palsy. Plausible medical advantages of umbilical cord blood stem cells include their immunologic properties and the presence of mesenchymal in addition to hematopoietic stem cells. More important from a public-health perspective are the logistic advantages, including the ease of their collection, their ready availability as a stored resource for clinical and research indications,2 and the possibility of improving the likelihood of finding stem cell matches for underrepresented minorities.3 Currently these advantages remain primarily theoretical; the medical indications for umbilical cord stem cell transplant are limited, with bone marrow transplant offering equivalent survival and remaining the preferred method of treatment in all but a few rare clinical situations.4,5 Moreover, the abundance of private cord blood banking options coupled with the lack of a public cord blood bank alternative in most areas of the United States prevents the public health benefits, such as improved access to stem cell transplant for underrepresented minorities, from being realized.
The private umbilical cord blood banking companies in the United States market directly to consumers, reporting that banking cord blood provides a “biologic insurance” for their unborn child.6 However, a survey of private cord blood banks by the American Society for Blood and Marrow Transplantation found that, of the approximately 460,000 privately banked cord blood units, only 99 had been shipped for transplantation.4 Similarly, a survey of 93 pediatric hematopoietic stem cell transplantation physicians in the United States and Canada reported that nine autologous and 41 allogenic transplants had been performed using privately banked cord blood; of the 41 allogenic transplants, 36 were in families where the cord blood had been collected preemptively owing to a known disease in a family member. Moreover, the logistics of cord blood banking are not always straightforward; characteristics such as the gestational age at delivery, obstetric history of the mother, and the circumstances of the delivery can affect the adequacy of the cord blood collection, which has implications for the success rate of a transplant.7 In addition, many cord blood units have been found to have quality-control issues related to bacterial contamination, documentation, or labeling.8 Given the low likelihood of any individual needing a stem cell transplant as well as the limitations of umbilical cord blood as a source of stem cells, the American College of Obstetrics and Gynecology as well as the American Academy of Pediatrics have issued statements recommending against private cord blood banking unless there is a family member with a known diagnosis that could be treated by umbilical cord blood.9,10 However, private cord blood bank companies continue to emphasize the future potential of umbilical cord transplant.6,11,12
In the setting of contradictory information from professional medical organizations and private umbilical cord blood banking companies regarding this health intervention, we sought to investigate the cost-effectiveness of private umbilical cord blood.
MATERIALS AND METHODS
A decision-analytic model from the parental perspective was developed with TreeAgePro 2009 software (TreeAge Software Inc, Williamstown, MA) (Fig. 1). At the time of delivery, an individual could choose to bank umbilical cord stem cells or not. Stem cells were presumed to be collected and stored for 20 years because that is the most common “package” offered by cord blood banking companies. Over the course of this time, we assumed at baseline that there was a 0.04% or 1 in 2,500 probability for an individual to develop an indication for a stem cell transplant.13 This is a composite estimate including a wide range of indications such as acquired hematologic conditions (aplastic anemia), hematologic malignancy (leukemia), and metabolic disorders. Although many of these conditions have a variety of treatment options, not all of which include stem cell transplant, for this model, it was assumed that if an indication developed, a stem cell transplant would be performed. The literature indicates that between 25% and 56% of cord blood specimens are not able to be used, either because the cell count contained is too small or because of problems with the storage process, and this possibility was included in the model 8. Although it is not the current standard of care, the assumption for this model was that, if it was available, the umbilical cord blood sample was used (as opposed to using autologous bone marrow, for example). Otherwise, the child could undergo an allogenic bone marrow transplant or receive medical therapy, with the potential for mortality or graft-versus-host disease as downstream effects. The likelihood of finding a matched, unrelated bone marrow donor ranges from 75% in white patients to less than 10% in patients from underrepresented minority groups14,15; a baseline estimate of 25% was used, and this was varied widely in sensitivity analysis.
In addition to autologous uses, the possibility of a sibling in the family requiring allogenic transplant also was considered, with a baseline probability of 0.07% or approximately 1 in 1,425. This probability is higher than the likelihood of an individual using his own stem cells because, in the case of leukemia or other malignancy, autologous stem cell transplant may not be preferred because cord blood stem cells are assumed to have an increased propensity for malignant transformation. In contrast, a child who developed these conditions would be able to use the allogenic stem cells from a sibling because these would not have the same underlying malignancy risk. The sibling faced the same potential complications. Information regarding the likelihood of requiring a stem cell transplant as well as survival and complication probabilities were obtained from the literature. Because the complication rate and survival likelihood after stem cell transplant depends on the indication for transplant and because the probability of requiring a stem cell transplant was a composite measure including many potential diagnoses, the likelihood of complications was varied widely (Table 1).
The cost applied was the cost of umbilical cord blood banking and storage for 20 years. Cost estimates were obtained from the private umbilical cord banking companies’ Web sites and ranged from $3,620 to $4,170. To provide the most conservative estimate of the cost-effectiveness of cord blood banking, the base case cost was estimated as the lowest price quoted from the major cord blood banking companies ($3,620); this estimate was varied widely in sensitivity analysis. Because the umbilical cord blood banking companies offer a large discount for paying upfront (more than the discount rates of 3–5% commonly used in the cost-effectiveness model), no additional discounting was included. Life years were based on neonatal life expectancy estimates from the national birth/death statistics. Based on the lack of data available regarding the utility of graft-versus-host disease and other potential health outcomes, life years was used as the measure of effectiveness.
Analysis consisted of projecting the total costs and life years for each strategy. The incremental cost-effectiveness ratio of each strategy was estimated and was considered cost-effective at a threshold of $100,000 per life year. Although the optimal threshold for cost-effectiveness in the United States is a matter of debate,16 the range of cost-effectiveness thresholds commonly is set at $50,000 to $100,000 per quality-adjusted life year.17 To be as liberal as possible and assess whether such an enterprise could even be marginally cost-effective, we used the upper range of $100,000 per quality-adjusted life year as our threshold. One-way and multiway sensitivity analysis was performed to examine the robustness of the findings. A Monte Carlo simulation was performed to test the robustness of the model to simultaneous multivariable changes in a theoretical cohort of women.
This study was exempt from institutional review board approval because no human participants or participant information was involved.
In the base case analysis, private umbilical cord blood banking results in a gain of 0.0026 life years (29.9848 for private cord blood banking compared with 29.9822 for no private cord blood banking) at an additional cost of $3,620 ($0 compared with $3,620); therefore, this was not a cost-effective intervention because it cost an additional $1,374,246 per life year saved.
In a one-way sensitivity analysis, private cord blood banking became cost-effective if the likelihood of a child needing an autologous stem cell transplant exceeded 1 in 110 or if the likelihood of a sibling needing an allogenic stem cell transplant rose to 1 in 43. If the cost of umbilical cord blood banking was reduced to $262, or approximately 7% of the baseline, cord blood banking became cost-effective. Given that the difficulty in finding a stem cell donor for children of underrepresented minorities has been one justification for public cord blood banking, the likelihood of finding an unrelated bone marrow donor was varied widely. Even if the likelihood of finding a related or unrelated bone marrow donor was reduced to zero, umbilical cord blood banking was still not cost-effective because it cost an additional $975,655 per life year saved. Because the reduction in graft-versus-host disease is seen as another advantage of cord blood stem cell transplant, this effect was also varied. Even if the likelihood of graft-versus-host disease was reduced to zero when umbilical cord stem cell transplant was used, this intervention still was not cost-effective because it cost an additional $1,342,732 per life year saved. Finally, the effect of family size was evaluated because, depending on the birth order of the pregnancy being considered and the number of siblings, the likelihood of using a stem cell transplant would increase. Contemplating a family size of five siblings, which incorporates 94% of United States households, private umbilical cord blood banking still was not cost-effective because it cost an additional $304,806 per life year saved.
Two-way sensitivity analysis was performed to examine the effect of variation in multiple parameters. Figure 2 depicts the effect of varying the cost of umbilical cord blood banking and the likelihood of a child needing an autologous transplant simultaneously. The white area of the graph is the portion in which the cost and probability would result in an incremental cost-effectiveness ratio of less than $100,000. If the cost of umbilical cord blood banking was reduced by 50%, the probability of needing an umbilical cord stem cell transplant would need to be more than 1 in 300, more than an eightfold increase from the baseline risk, for this to be a cost-effective intervention (Fig. 2).
To simulate the situation in which private cord blood banking would be optimal, the viability of cord blood was assumed to be 100%, graft-versus-host disease with cord blood transplant was reduced to zero, the likelihood of finding an unrelated bone marrow donor was reduced to zero, and mortality without stem cell transplant was assumed to be 100%. Under these conditions, private cord blood banking still was not cost-effective because it cost an additional $352,931 per life year saved. The greatest likelihood of needing a stem cell transplant that has been quoted in the literature, which includes the use of stem cells for a variety of indications where this is not currently first-line treatment, is 0.13%.18 Using this probability for the likelihood of autologous transplant or allogenic transplant in a sibling along with the other optimal probabilities for stem cell transplant described above, private umbilical cord blood banking still is not cost-effective because it costs an additional $134,217 per life year saved. In sensitivity analysis, even under these optimal conditions for private cord blood banking, the likelihood of needing a stem cell transplant would have to be at least 1 in 515 in order for this intervention to be cost-effective.
Finally, Monte Carlo simulation was performed to test the robustness of the model to simultaneous changes in input probabilities. Figure 3 depicts a scatterplot with each point representing a separate trial in which the input probabilities and costs are sampled from a distribution of possible values. Points that lie below the dotted line represent trials in which private cord blood banking was cost-effective at a threshold of $100,000. In 99.2% of trials, private cord blood banking was not cost-effective because it cost more than $100,000 per life year saved. In 52.8% of trials, private cord blood banking was dominated, meaning that it was less effective and more expensive than the alternative.
In this analysis, private cord blood banking is not a cost-effective intervention for an individual family because the cost required greatly outweighs the benefit gained. Any cost-effectiveness analysis incorporates some degree of assumption and uncertainty; however, even envisioning the best possible scenario in support of private cord blood banking based on currently available information, this is not a cost-effective intervention. Although the American College of Obstetricians and Gynecologists and the American Association of Pediatrics already have made statements recommending against private cord blood banking, practitioners’ day-to-day experience shows that many patients still choose to bank umbilical cord blood with private companies. One rational explanation for why patients would choose to participate in private umbilical cord blood banking is that they simply overestimate the probability of the need for its use. Thus, the discrepancy between the benefit perceived by families and the lack of benefit seen in this analysis and in the opinions of our professional societies has important implications for the way that we counsel and share information with patients.
With the exception of physicians treating families with rare blood disorders likely to require stem cell transplant, the desire for cord blood banking primarily comes from the parent in response to marketing or peer information that they have received. Direct-to-consumer marketing of pharmaceutical products has increased radically since 1997, when the U.S. Food and Drug Administration relaxed their rules regarding this type of advertising. Proponents of this type of marketing argue that patients benefit from the clinical information that is provided, whereas opponents counter that no benefit in health outcomes has resulted from the expansion of direct-to-consumer advertising.19,20 As for the issue of cord blood banking in particular, previous research has shown that pregnant women have a very poor understanding of the current indications for stem cell transplant as well as of the probability that their child will need this intervention.21–23 Our analysis shows that, unless the applications for cord blood transplant are increased exponentially or the cost of banking drops dramatically, private umbilical cord blood banking is not a cost-effective intervention. The highest likelihood of needing a stem cell transplant before age 20 that is quoted in the literature is 0.13% or approximately 1 in 800, far below the threshold identified in our analysis.18 The remote likelihood of any individual child requiring her own or her sibling’s cord blood stem cells is a concept that prospective parents must be made aware of as they weigh the resources required to bank umbilical cord blood.
Our analysis has inherent limitations. No decision-analytic model is able to capture fully the complexity of a clinical situation or to include all of the factors that a clinician or parent integrates when making a decision. Stem cell research is an evolving field. The hematopoietic stem cells in cord blood are uniquely suited for transplantation both because of their immune properties and because of the relative ease of collecting them. Even more exciting is the idea that the mesenchymal stem cells contained within cord blood someday could serve as a reservoir to repair and replace cell lines that may have become damaged as a result of diseases of adulthood such as Alzheimer and diabetes. Our analysis is limited in that it cannot take into account potential future applications. Moreover, we were not able to quantify the peace of mind that parents may feel knowing that this resource is available to a child or her siblings. However, the argument can be made that parents’ concerns are a result of the information they have available regarding the risks and benefits of cord blood banking; so perhaps providing additional information explaining the current applications of cord blood would provide the same reassurance. Finally, this analysis considers only the costs and benefits of private cord blood banking. Previous analysis regarding the optimal size of a U.S. public cord blood bank indicates that significant societal benefit could be gained by making cord blood an available source of stem cells for unrelated patients and particularly for underrepresented minorities, significantly enhancing the current bone marrow registry system.24 Unfortunately, donation to a public cord blood bank is not available in most hospitals at this time.
The argument has been made that the decision to bank umbilical cord blood is a personal one, best made without the influence of analyses such as these.18 However, this perspective does not take into account the inherent difficulty that humans have in conceptualizing risk. Studies of human decision making show that, when considering very rare events, decision-makers tend to overweight their likelihood.25 In addition, the most recently presented information has a stronger influence and may cause underweighting or neglecting of more likely events.26 In this case, marketing from cord blood banks greatly increases the exposure of parents to the possibility of a future need for stem cell transplant. In many cases, the emphasis is placed on potential future applications of stem cells for the treatment of neurologic disorders or diabetes despite the fact that these treatments are only experimental at this point. Although the probabilities quoted are low and statements regarding the uncertain benefit are included, both the inherent difficulty in conceptualizing low-probability events and the increased availability of this information results in additional weighting, affecting decision making. Without explanation of the quoted probabilities, it is difficult for prospective parents to put this expenditure in perspective with the many other competing needs of their newborn.
In addition to the individual monetary tradeoffs for an individual family, there is significant societal cost resulting from continued emphasis on commercial umbilical cord blood banking. Despite an analysis commissioned by the Institute of Medicine regarding the optimal size and distribution of an umbilical cord blood bank24 and the plan to establish such a resource, there currently is not a public cord blood banking option in most areas of the United States.27 The option of private cord blood banking will remain as long as this continues to be a lucrative industry; patient education is the key to shifting the focus to a public cord blood banking system, which has been successful in other countries.28,29 Improving patient understanding of the true indications and probabilities associated with cord blood stem cell transplant is an integral step in supporting the development of such a resource. Once regional cord blood banks are established, physicians will be able to offer patients an option with real public health and societal benefit.
Given the current cost of banking and storage as well as the applications for umbilical cord blood stem cells, private umbilical cord blood banking is cost-effective for only families in which a child has a very high risk of requiring a stem cell transplant. Parents contemplating this intervention should be made aware of the remote likelihood that cord blood will be used for a child in their family. Because private cord blood banking companies have a significant conflict of interest in providing this information, physicians and other obstetric care providers must help to provide evidence-based information to patients. In the future, as it becomes available in the United States, public cord blood banking will offer a better alternative from a personal and a societal perspective.
1. Gluckman E, Broxmeyer HA, Auerbach AD, Friedman HS, Douglas GW, Devergie A, et al. Hematopoietic reconstitution in a patient with Fanconi’s anemia by means of umbilical-cord blood from an HLA-identical sibling. N Engl J Med 1989;321:1174–8.
2. Barker JN, Krepski TP, DeFor TE, Davies SM, Wagner JE, Weisdorf DJ. Searching for unrelated donor hematopoietic stem cells: availability and speed of umbilical cord blood versus bone marrow. Biol Blood Marrow Transplant 2002;8:257–60.
3. Sullivan MJ. Banking on cord blood stem cells. Nat Rev Cancer 2008;8:555–63.
4. Ballen KK, Barker JN, Stewart SK, Greene MF, Lane TA, American Society of Blood and Marrow Transplantation.. Collection and preservation of cord blood for personal use. Biol Blood Marrow Transplant 2008;14:356–63.
5. Thornley I, Eapen M, Sung L, Lee SJ, Davies SM, Joffe S. Private cord blood banking: experiences and views of pediatric hematopoietic cell transplantation physicians. Pediatrics 2009;123:1011–7.
7. McGuckin CP, Basford C, Hanger K, Habibollah S, Forraz N. Cord blood revelations- The importance of being a first born girl, big, on time and to a young mother! Early Hum Dev 2007;83:733–41.
8. Moise KJ Jr. Umbilical cord stem cells. Obstet Gynecol 2005;106:1393–407.
9. Umbilical cord blood banking. ACOG Committee Opinion No. 399. American College of Obstetricians and Gynecologists. Obstet Gynecol 2008;111:475–7.
10. American Academy of Pediatrics Section on Hematology/Oncology; American Academy of Pediatrics Section on Allergy/Immunology, Lubin BH, Shearer WT. Cord blood banking for potential future transplantation. Pediatrics 2007;119:165–70.
13. Johnson FL. Placental blood transplantation and autologous banking–caveat emptor. J Pediatr Hematol Oncol 1997;19:183–6.
14. Beatty PG, Boucher KM, Mori M, Milford EL. Probability of finding HLA-mismatched related or unrelated marrow or cord blood donors. Hum Immunol 2000;61:834–40.
15. van Rood JJ, Oudshoorn M. Eleven million donors in Bone Marrow Donors Worldwide! Time for reassessment? Bone Marrow Transplant 2008;41:1–9.
16. Owens DK. Interpretation of cost-effectiveness analyses. J Gen Intern Med 1998;13:716–7.
17. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York (NY): Oxford University Press; 1996.
18. Nietfeld JJ. Opinions regarding cord blood use need an update. Nat Rev Cancer 2008;8:823.
19. Gilbody S, Wilson P, Watt I. Benefits and harms of direct to consumer advertising: a systematic review. Qual Saf Health Care 2005;14:246–50.
20. Donohue JM, Cevasco M, Rosenthal MB. A decade of direct-to-consumer advertising of prescription drugs. N Engl J Med 2007;357:673–81.
21. Fox NS, Stevens C, Ciubotariu R, Rubinstein P, McCullough LB, Chervenak FA. Umbilical cord blood collection: do patients really understand? J Perinat Med 2007;35:314–21.
22. Fernandez CV, Gordon K, Van den Hof M, Taweel S, Baylis F. Knowledge and attitudes of pregnant women with regard to collection, testing and banking of cord blood stem cells. CMAJ 2003;168:695–8.
23. Sugarman J, Kurtzberg J, Box TL, Horner RD. Optimization of informed consent for umbilical cord blood banking. Am J Obstet Gynecol 2002;187:1642–6.
24. Howard DH, Meltzer D, Kollman C, Maiers M, Logan B, Gragert L, et al. Use of cost-effectiveness analysis to determine inventory size for a national cord blood bank. Med Decis Making 2008;28:243–53.
25. Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica 1979:263–91.
26. Hertwig R, Barron G, Weber EU, Erev I. Decisions from experience and the effect of rare events in risky choice. Psychol Sci 2004;15:534–9.
27. Meyer EA, Gebbie KM, Hanna KE. Cord blood: establishing a national hematopoietic stem cell bank program. Washington (DC): National Academy Press, 2005.
28. Boo M. Public cord blood banking may play an important role in the emergence of unrelated transplant in developing countries. Transfusion 2008;48:207–8.
29. Gluckman E, Rocha V, Boyer-Chammard A, Locatelli F, Arcese W, Pasquini R, et al Outcome of cord-blood transplantation from related and unrelated donors. Eurocord Transplant Group and the European Blood and Marrow Transplantation Group. N Engl J Med 1997;337:373–81.
30. Brunstein CG, Wagner JE. Umbilical cord blood transplantation and banking. Annu Rev Med 2006;57:403–17.
31. Rocha V, Wagner JE, Sobocinski KA, Klein JP, Zhang MJ, Horowitz MM, et al. Graft-versus-host disease in children who have received a cord-blood or bone marrow transplant from an HLA-identical sibling. Eurocord and International Bone Marrow Transplant Registry Working Committee on Alternative Donor and Stem Cell Sources. N Engl J Med 2000;342:1846–54.
32. Davey S, Armitage S, Rocha V, Garnier F, Brown J, Brown CJ, et al. The London Cord Blood Bank: analysis of banking and transplantation outcome. Br J Haematol 2004;125:358–65.
33. Jacobsohn DA, Hewlett B, Ranalli M, Seshadri R, Duerst R, Kletzel M. Outcomes of unrelated cord blood transplants and allogeneic-related hematopoietic stem cell transplants in children with high-risk acute lymphocytic leukemia. Bone Marrow Transplant 2004;34:901–7.
34. Hahn T, McCarthy PL Zhang MJ, Wang D, Arora M, Frangoul H, et al. Risk factors for acute graft-versus-host disease after human leukocyte antigen-identical sibling transplants for adults with leukemia. J Clin Oncol 2008;26:5728–34.
35. Jacobsohn DA, Vogelsang GB. Acute graft versus host disease. Orphanet J Rare Dis 2007;2:35.
36. Jacobsohn DA. Acute graft-versus-host disease in children. Bone Marrow Transplant 2008;41:215–21.
37. Eapen M, Rubinstein P, Zhang MJ, Stevens C, Kurtzberg J, Scaradavou A, et al. Outcomes of transplantation of unrelated donor umbilical cord blood and bone marrow in children with acute leukaemia: a comparison study. Lancet 2007;369:1947–54.