Portal de Periódicos da CAPES

Costs associated with treatment of severe combined immunodeficiency—rationale for newborn screening in Sweden

2016; Elsevier BV; Volume: 139; Issue: 5 Linguagem: Inglês

10.1016/j.jaci.2016.10.043

1097-6825

Ann Gardulf, Jacek Winiarski, Moa Thorin, Marianne Heibert Arnlind, Ulrika von Döbeln, Lennart Hammarström,

Neonatal Respiratory Health Research

Severe combined immunodeficiency (SCID) is the most severe form of primary immunodeficiency with profound defects in both the cellular and humoral immune defence.1Puck J. Population-based newborn screening for severe combined immunodeficiency: steps towards implementation.J Allergy Clin Immunol. 2007; 120: 760-768Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar, 2Chan A. Scalchunes C. Boyle M. Puck J. Early vs. delayed diagnosis of severe combined immunodeficiency: a family perspective survey.Clin Immunol. 2011; 138: 3-8Crossref PubMed Scopus (68) Google Scholar, 3Gaspar H.B. Hammarström L. Mahlaoui N. Borte M. Borte S. The case for mandatory newborn screening for severe combined immunodeficiency (SCID).J Clin Immunol. 2014; 34: 393-397Crossref PubMed Scopus (54) Google Scholar The affected newborns are highly susceptible to life-threatening opportunistic bacterial, viral, and/or fungal infections.1Puck J. Population-based newborn screening for severe combined immunodeficiency: steps towards implementation.J Allergy Clin Immunol. 2007; 120: 760-768Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar, 2Chan A. Scalchunes C. Boyle M. Puck J. Early vs. delayed diagnosis of severe combined immunodeficiency: a family perspective survey.Clin Immunol. 2011; 138: 3-8Crossref PubMed Scopus (68) Google Scholar, 3Gaspar H.B. Hammarström L. Mahlaoui N. Borte M. Borte S. The case for mandatory newborn screening for severe combined immunodeficiency (SCID).J Clin Immunol. 2014; 34: 393-397Crossref PubMed Scopus (54) Google Scholar, 4A white paper on the need for newborn (at-birth) screening for severe combined immunodeficiency (SCID) in Europe, November 2013. Available at: http://www.ipopi.org/uploads/NBS%20SCID%20White%20Paper%20FINAL%20designed.pdf. Accessed October 2016.Google Scholar, 5Borte S. von Döbeln U. Hammarström L. Guidelines for newborn screening of primary immunodeficiency diseases.Curr Opin. 2013; 20: 48-54Google Scholar Early detection of SCID in the preinfectious period followed by hematopoietic stem cell transplantation (HSCT) has been shown to dramatically improve survival.1Puck J. Population-based newborn screening for severe combined immunodeficiency: steps towards implementation.J Allergy Clin Immunol. 2007; 120: 760-768Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar This has led to the implementation of national newborn screening (NBS) programs in selected countries and pilot projects ongoing in many additional countries.6IPOPI SCID Newborn Screening Campaign. Available at: http://www.ipopi.org/index.php?page=scid-country. Accessed October 2016.Google Scholar Two studies from the United States and France of real-life costs for HSCT in SCID have been published and both showed that early HSCT resulted in lower treatment-related costs.7Kubiak C. Jyonouchi S. Kuo C. Garcia-Lloret M. Dorsey M.J. Sleasman J. et al.Fiscal implications of newborn screening in the diagnosis of severe combined immunodeficiency.J Allergy Clin Immunol Pract. 2014; 2: 697-702Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar, 8Clement M.C. Mahlaoui N. Mignot C. Le Bihan C. Rabetrano H. Hoang L. et al.Systematic neonatal screening for severe combined immunodeficiency and severe T-cell lymphopenia: analysis of cost-effectiveness based on French real field data.J Allergy Clin Immunol. 2015; 135: 1589-1593Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar The aim of this study was to use patient-unique, individual, real-life cost and care data to calculate costs for children with SCID undergoing HSCT in Sweden, and to identify types and percentage breakdown of the costs. All children with SCID undergoing HSCT at the Karolinska University hospital in Stockholm during the period 2000 through 2014 were included and divided into 2 groups: those undergoing early HSCT (age ≤6 months, n = 6) and those undergoing late HSCT (age >6 months, n = 8) (Table I). Age at HSCT varied depending on type of SCID, status of the individual child (“infection burden”), and availability of a suitable donor. The age cutoff was set to 6 months or less; if the recommended cutoff age of 3.5 months1Puck J. Population-based newborn screening for severe combined immunodeficiency: steps towards implementation.J Allergy Clin Immunol. 2007; 120: 760-768Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar had been used, it would have resulted in only 1 child in the “Early HSCT group.”Table IPresentation of included children with SCID undergoing HSCT in Stockholm, Sweden, during 15 years (2000 through 2014)GroupChild no.SexBornFirst hospital admission Stockholm∗The Karolinska University Hospital, Huddinge, Stockholm, Sweden.Age at diagnosisSCID diagnosisAge at HSCT (year-month)No. of HSCTsDonorTotal analyzed period regarding costs (pre- and post-HSCT)Status December 31, 2015Early HSCT group1BoySep 2004Sep 20041 moCernunnos deficiency†Siblings.2 mo (2004-10)1Cord blood, unrelated donor5 moDeceased at age 5 mo2BoyAug 2009Nov 20093 moCernunnos deficiency†Siblings.4 mo (2009-12)6 mo (2010-02)2HSCT 1: Maternal stem cellsHSCT 2: Cord blood6 moDeceased at age 9 mo3BoyApr 2000Jul 20003 moUndetermined mutation5 mo (2000-09)1Unrelated donor15 y, 6 moAlive4GirlJan 2012Feb 20121 moADA deficiency‡Siblings.5 mo (2012-06)1Sibling3 y, 10 moAlive5GirlNov 2008Mar 20096 moOmenn syndrome6 mo (2009-06)1Sibling6 y, 9 moAlive6BoyMay 2014Jul 20142 moSCID6 mo (2014-12)1Maternal stem cells1 y, 6 moAliveLate HSCT group7BoyJul 2008Nov 20084 moX-linked SCID7 mo (2009-02)1Cord blood7 y, 1 moAlive8BoyFeb 2012May 20123 moX-linked SCID7 mo (2012-09)1Paternal stem cells3 y, 7 moAlive9BoyAug 2008May 20099 moUndetermined mutation10 mo (2009-06)1 y, 3 mo (2009-11)1 y, 5 mo (2010-01)3HSCT 1: Cord bloodHSCT 2 and 3: Unrelated donor (the same donor)6 y, 7 moAlive10BoyFeb 2004Dec 200410 moX-linked SCID1 y (2005-02)1Cord blood11 yAlive11BoyMar 2010Dec 20109 moX-linked SCID1 y (2011-03)1Unrelated donor7 moDeceased at age 1 y, 4 mo12BoyNov 2004Jan 2006§Initially cared for in his home country.5 moX-linked SCID1 y, 1 mo (2005-12)1Cord blood1 y, 9 moDeceased at age 1 y, 9 mo13GirlAug 2011Feb 20142 y, 5 moArtemis SCID2 y, 7 mo (2014-03)1Unrelated donor1 y, 10 moAlive14GirlNov 2001Jan 20023 moADA deficiency‡Siblings.4 y, 6 mo (2006-05)1Sibling13 y, 11 moAliveThe order of the children is based on age at their first HSCT.∗ The Karolinska University Hospital, Huddinge, Stockholm, Sweden.† Siblings.‡ Siblings.§ Initially cared for in his home country. Open table in a new tab The order of the children is based on age at their first HSCT. Eight children were cared for only in Stockholm, 5 were referred from other hospitals, and 1 from abroad (n = 1). Cost and care data were gathered using the Cost-Per-Patient database. Eighteen types of cost items were identified in the Cost-Per-Patient database (see Table E1 in this article's Online Repository at www.jacionline.org), coded and linked to each child's unique personal identity number. For children cared for at other hospitals, each hospital was contacted, corresponding data retrieved, and the coding system with 18 cost types applied. The cost follow-up period lasted until December 31, 2015, or to the date of death of the child. Two children underwent more than 1 HSCT (Table I); the costs for these additional transplantations have been included in the cost calculations. In total, more than 1.2 million items of individual patient data were collected and more than 35,000 patient-unique, individual cost items were analyzed. In-hospital and outpatient clinic costs for each child were gathered for 2000 through 2015. The costs for the years 2000 to 2014 were adjusted to 2015 values in Swedish kronor (SEK) using an international cost converter.9The Cochrane and Campbell Economic Methods Group, CCEMG-EPPI-Centre Cost Converter, version 1.4. Available at: http://eppi.ioe.ac.uk/costconversion. Accessed October 2016.Google Scholar To facilitate international comparisons, the second step was to convert the costs in SEK to euros (EUR) using the mean exchange rate during 2015 as stated by the Swedish Central Bank (1 EUR = 9.9562 SEK). The study was approved by the regional Ethical Committee (no. 2016/488-31/2). The total mean cost per child was EUR 301,832 (range, EUR 144,229-597,817) for the early HSCT group and EUR 423,642 (range, EUR 242,946-616,932) for the late HSCT group. Both the total pre- and post-HSCT costs were lower for the early group (Fig 1, A). The major costs arose during the first year post-HSCT (Fig 1, B). If the child survived, the cumulative cost per child was then relatively stable and increased only marginally over time, especially after the second year post-HSCT. During the 5 years before and after the HSCT, the children in the early group spent totally 663 days in hospital and had to visit the outpatient clinics 302 times; the children in the late group spent totally 1461 days in hospital and had to make totally 787 visits to the outpatient clinic. Overviews of individual in-hospital care days and visits are shown in Fig E1 and Tables E2 and E3 in this article's Online Repository at www.jacionline.org. The year before the HSCT, on average the children in the early group were admitted for 40 days (median, 40 days; range, 17-60 days) and the children in the late HSCT group were admitted for 68 days (median, 62 days; range, 9-161 days). The corresponding data for the first year post-HSCT were for the early group on average 68 days (median, 51 days; range, 29-147 days) and for the late group 92 days (median, 79 days; range, 39-161 days). The percentage breakdown of the hospital cost into the 18 cost items showed that the 4 largest cost items were staff (35%), pediatric intensive care (12%), hotel costs (10%), and common drugs and materials used at the ward (10%). Tests and analyses for clinical immunology, transfusion medicine–related investigations, and clinical microbiology accounted for 9% and individually prescribed drugs for 7% of the total cost (see Fig E2 in this article's Online Repository at www.jacionline.org). The percentage breakdown pattern of the outpatient clinic care cost was somewhat different; staff still accounted for the largest cost fraction (31%), followed by tests and analyses for clinical microbiology (15%), clinical immunology and transfusion medicine–related investigations (11%), and hotel costs (10%) (see Fig E3 in this article's Online Repository at www.jacionline.org). The study was based on very detailed data covering up to 15 years post-HSCT in children with SCID, the longest follow-up yet published. However, not all costs before diagnosis were possible to retrieve, in spite of great efforts. The total costs for the treatment of the children in both groups are therefore, if anything, underestimated. The study showed that early HSCT reduced the total mean cost by approximately 30%. It was also shown that if the child survived, irrespective of early or late HSCT, the main cost was incurred during the first year post-HSCT, while the cumulative cost for most children thereafter increased only marginally. Thus, once a child has passed the first post-HSCT year and survived, the costs for health care were limited. The study also demonstrated important health benefits for the children and the families from early HSCT, such as reduction by more than half of in-hospital days and number of visits to the outpatient clinics. It seems reasonable to assume that this, in turn, leads to improved health-related quality of life for the children and their parents/guardians. For the first time, real-life cost data were broken down into percentages and labeled according to different cost types. This showed that large percentages, 70% of hospital and 46% of outpatient clinic costs, were costs for staff and administration. In comparison, individually prescribed hospital drugs accounted for only 7% of the total cost. The drugs used for HSCT are considered “expensive drugs” and there is an ongoing debate in Sweden where it is claimed that the cost of drugs within health care is unproportionally high. However, the results from this study highlight the fact that this is not likely the case for HSCT in SCID. In the French real-life study, social care and welfare benefits were also estimated and amounted to a mean of EUR 8836 over a 1-year post-HSCT period.8Clement M.C. Mahlaoui N. Mignot C. Le Bihan C. Rabetrano H. Hoang L. et al.Systematic neonatal screening for severe combined immunodeficiency and severe T-cell lymphopenia: analysis of cost-effectiveness based on French real field data.J Allergy Clin Immunol. 2015; 135: 1589-1593Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar If costs such as production losses for the parents/guardians, travel costs, and accommodation costs had been included in the present study, the cost for the HSCT process would have been higher, especially in the late HSCT group, where the children were admitted to hospital for longer periods. Costs are related to the days admitted to hospital and number of visits to outpatient clinics. The implementation of an NBS program might help to reduce the number of in-hospital care days and outpatient clinic visits and thus lead to a decrease in health care costs. With NBS programs children with SCID are diagnosed at birth and the life-saving HSCT can be provided much earlier in the preinfectious period. The results from the present study may be a rational to support a decision to implement an NBS program for SCID in Sweden. It is concluded that early HSCT of children with SCID substantially reduced health care costs, number of in-hospital care days, and visits to the outpatient clinic. It is furthermore concluded that for children with SCID undergoing HSCT in Sweden, the cost of drugs is not a major cost issue. We are grateful to all hospital economists who have provided the data, especially Marianne Modin at Karolinska University Hospital, Huddinge, Stockholm, to the heads of the participating clinics, and to Professor Tore G. Abrahamsen at the University Hospital in Oslo, Norway. Fig E2Costs for in-hospital care for children with SCID undergoing HSCT. The figure shows types and percentage breakdown of costs for in-hospital care for children with SCID undergoing HSCT.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E3Costs for outpatient clinic visits for children with SCID undergoing HSCT. The figure shows types and percentage breakdown of costs for outpatient clinic visits for children with SCID undergoing HSCT.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table E1The 18 cost items identified and used in the cost analysisCost codeType of costsExamples of cost items included in the cost analysis1Real estate costs (HC)Eg, insurances for the building, devices, and furnishings of major public areas; annual depreciation; repairs and renovations2Hotel costs (HC)Eg, costs for cleaning, laundry (children's bed clothes and medical clothes for staff), furnishing of rooms for caring for the children, food and drinks to children and guardians3Staff (HC)Salaries4Common drugs and materials used at the pediatric ward or outpatient clinics (HC)Common drugs and materials acquired for the ward/outpatient clinic and used for all children irrespective of diagnosis, eg, cough drops, nasal sprays, and antipyretics, or, eg, medical supplies and protective devices/materials for staff5Medical service (HC)Transportation of patients within the hospital, external special transports of samples6Individually prescribed drugsEg, immunosuppressive drugs, antibiotics, antiviral drugs, antifungals, high-dose polyclonal intravenous IgG and valproate7Admission to the pediatric intensive care unit, including operating theater and postoperative care8ImagesX-ray, magnetic resonance imaging, computed tomography scan, other images9-17Tests/analysis: Clinical physiology, neurophysiology, immunology and transfusion medicine, chemistry, microbiology, pathology, cytology, histology, genetics, metabolic disorders, medical physics18Overhead costs (HC)Costs not possible to refer to any of the above-mentioned costsTwenty-four hour costs set yearly by the hospital economists are marked HC. Open table in a new tab Table E2Annual in-hospital care days and outpatient visits 5 years before and 5 years after HSCT for a group of children with SCIDNo. of children admitted to hospitalNo. of days admittedNo. of children visiting outpatient clinicsNo. of visits at outpatient clinicsEarly group (n = 6)Late group (n = 8)Early groupLate groupEarly group (n = 6)Late group (n = 8)Early groupLate groupYear −501∗Child no. 14 in Table I. of 8010501∗Child no. 14 in Table I. of 802Year −401∗Child no. 14 in Table I. of 8030000Year −3000001∗Child no. 14 in Table I. of 801Year −201∗Child no. 14 in Table I. of 801501∗Child no. 14 in Table I. of 809Year −16 of 68 of 82395444 of 67 of 86163HSCTYear +16 of 68 of 84107384 of 68 of 8148295Year +22 of 4†Two children in the Early group (no. 1 and no. 2 in Table I) died during the first year post-HSCT; therefore, data for 4 children were available at follow-up year +2 and onward.4 of 6‡Two children in the Late group (no. 11 and no. 12 in Table I) died during the first year post-HSCT; therefore, data for 6 children were available at follow-up year +2 and onward.14353 of 3§The cost follow-up was concluded on December 31, 2015, and at this time point 2 children had died during the first year post-HSCT (no. 1 and no. 2 in Table I) and 1 child (no. 6 in Table I) had just concluded his first post-HSCT year. Of these reasons, data for 3 children were available.6 of 6‡Two children in the Late group (no. 11 and no. 12 in Table I) died during the first year post-HSCT; therefore, data for 6 children were available at follow-up year +2 and onward.65125Year +303 of 60213 of 36 of 615118Year +400003 of 34 of 69101Year +500002 of 24 of 6473Total6631461302787Early, Children aged ≤6 mo at HSCT (n = 6); Late, children aged >6 mo at HSCT (n = 8).∗ Child no. 14 in Table I.† Two children in the Early group (no. 1 and no. 2 in Table I) died during the first year post-HSCT; therefore, data for 4 children were available at follow-up year +2 and onward.‡ Two children in the Late group (no. 11 and no. 12 in Table I) died during the first year post-HSCT; therefore, data for 6 children were available at follow-up year +2 and onward.§ The cost follow-up was concluded on December 31, 2015, and at this time point 2 children had died during the first year post-HSCT (no. 1 and no. 2 in Table I) and 1 child (no. 6 in Table I) had just concluded his first post-HSCT year. Of these reasons, data for 3 children were available. Open table in a new tab Table E3Outpatient clinic visits for each child for the entire cost follow-up period for children before and after HSCT for a group of children with SCIDGroupHSCTChildYear −5Year −4Year −3Year −2Year −1Year +1Year +2Year +3Year +4Year +5Year +6Year +7Year +8Year +9Year +10Year +11Year +12Year +13Year +14Year +15TotalEarly groupNo. 10∗Four children died during the first post-HSCT year.No. 20∗Four children died during the first post-HSCT year.No. 3172100112454481075383No. 4737409396No. 52755156532113No. 6105464Late groupNo. 795219144422106No. 84383226100No. 9331194727117118No. 104837292228124180No. 1133∗Four children died during the first post-HSCT year.6No. 12143∗Four children died during the first post-HSCT year.17No. 1339512110No. 14201927532430281418169155251The children are presented in the same order as in Table I. The children have been followed regarding care and costs until December 31, 2015, and therefore the follow-up periods vary.∗ Four children died during the first post-HSCT year. Open table in a new tab Twenty-four hour costs set yearly by the hospital economists are marked HC. Early, Children aged ≤6 mo at HSCT (n = 6); Late, children aged >6 mo at HSCT (n = 8). The children are presented in the same order as in Table I. The children have been followed regarding care and costs until December 31, 2015, and therefore the follow-up periods vary.