Assessing health care resource use, outcomes, and costs among Medicaid beneficiaries receiving factor IX prophylaxis for hemophilia B
Publication: Journal of Managed Care & Specialty Pharmacy
Volume 30, Number 10
Abstract
BACKGROUND:
Hemophilia B is characterized by a deficiency of clotting factor IX (FIX), leading to excessive bleeding. Hemophilia B is commonly treated using replacement FIX therapy, which may be administered prophylactically or on-demand following a bleeding episode. Previous research has found high health care resource use (HCRU) and costs among Medicare and commercially insured people with hemophilia B (PwHB), with FIX therapy being a primary driver of health care costs.
OBJECTIVE:
To assess HCRU, outcomes, and costs among US Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B.
METHODS:
This study employed a retrospective comparative cohort design to assess HCRU, outcomes, and costs among adult male Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B, relative to a matched comparator population of beneficiaries without bleeding disorders. Nationwide Medicaid claims and enrollment data from 2015 to 2020 were used for this analysis. Adult male PwHB who received FIX prophylaxis, defined as not having identified gaps in FIX therapy exceeding 60-days during a 1-year measurement period, and were continuously enrolled in Medicaid for at least 2 years, were matched 1:4 to comparator beneficiaries without bleeding disorders based on baseline demographic and clinical characteristics. Key measures of HCRU and outcomes included inpatient hospital admissions, outpatient hematologist visits, and bleeding events. Measures of health care costs were assessed among a subset of beneficiaries enrolled in fee-for-service Medicaid.
RESULTS:
PwHB receiving FIX prophylaxis were significantly more likely to have multiple inpatient hospital admissions and had a longer cumulative length of stay per person relative to comparator beneficiaries (30.2 vs 14.8 days, respectively; P = 0.0473). PwHB receiving FIX prophylaxis also had significantly higher rates of bleeding events relative to comparator beneficiaries (0.54 vs 0.02 per person, respectively; P < 0.0001) and outpatient hematologist visits (1.58 vs 0.20 per person, respectively; P < 0.0001). Annual costs among PwHB receiving FIX prophylaxis were significantly higher than costs among comparator beneficiaries ($928,370 vs $34,553 per person, respectively; P < 0.0001) and were overwhelmingly driven by costs associated with FIX therapy.
CONCLUSIONS:
This analysis found higher rates of HCRU and costs among Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B relative to a matched comparator population of beneficiaries without bleeding disorders. Future research should examine hemophilia B costs and outcomes within the context of new treatments with innovative mechanisms of action, such as gene therapies, RNA interference therapies, and antitissue factor pathway inhibitor therapies.
Plain language summary
This study examined health care resource use, outcomes, and costs among Medicaid beneficiaries receiving factor IX prophylaxis for hemophilia B.
Implications for managed care pharmacy
This study found high expenditures associated with factor IX prophylaxis among Medicaid beneficiaries with hemophilia B. Medicaid agencies need to account for these costs when setting capitation rates for managed care pharmacy coverage and may “carve-out” these drugs from the managed care pharmacy benefit package and instead cover them via fee-for-service. Results from this analysis should also be viewed in the context of the potential impact of innovative therapies for the treatment of hemophilia B.
Hemophilia B is characterized by a deficiency of clotting factor IX (FIX), leading to excessive bleeding episodes. Bleeding can occur in the joints, muscles, gastrointestinal tract, and other areas of the body, ranging from mild to life-threatening. These bleeding episodes can result from physical trauma or occur spontaneously. Repeated joint bleeding associated with hemophilia B results in joint arthropathy, which is a serious complication associated with joint pain, deformity, and disability.1 Previous research has found that 2 to 3 bleeding episodes in the same joint may cause irreversible and progressive structural damage that can compromise quality of life for people with hemophilia B (PwHB).2 The World Federation of Hemophilia estimates there are more than 37,998 PwHB worldwide, with more than 6,000 in the United States alone.3
Hemophilia B is treated using replacement, intravenous FIX concentrates to increase plasma FIX levels. Exogenous FIX replacement therapy, including recombinant FIX and plasma-derived FIX, can be administered prophylactically to prevent hemophilia B–related bleeding episodes or on-demand after a bleeding episode. As prophylaxis has the potential to prevent bleeding and provides superior hemostatic benefits relative to on-demand therapy, it is recognized as the standard of care for severe hemophilia B and nonsevere hemophilia B with a severe phenotype.4,5 However, prophylaxis requires regular venous access and frequent FIX therapy administrations, creating significant patient burden. FIX prophylaxis reduces, but does not totally eliminate, bleeding risk among people with severe hemophilia B, and PwHB receiving FIX prophylaxis may still experience joint bleeds and joint arthropathy.2
Previous research has found significant health care resource use (HCRU) and direct costs among PwHB.6-14 PwHB can experience significant indirect costs associated with disability, absenteeism, and caregiver time.6,7 PwHB also have a high risk of bleeding events that can lead to costly inpatient hospital admissions and emergency department visits.6-8 However, previous analyses have consistently found that the majority of costs associated with hemophilia B can be attributed to costs associated with on-demand or prophylactic use of FIX therapy.8,9,12,13 For example, Buckner et al found that commercially insured adult individuals with severe hemophilia B had annual health care costs in excess of $600,000 per person, with more than 95% of these costs associated with FIX therapy pharmacy claims.8
There are several key gaps in the existing literature examining HCRU and costs associated with hemophilia B. First, recent studies in this space have generally focused on commercial-insured and Medicare-insured PwHB. Estimates suggest that 30% of PwHB are covered by public insurance programs, including Medicaid,15,16 and this study seeks to understand the impact in this setting. Second, few analyses have examined costs specifically associated with the use of FIX prophylaxis. This study aims to address these gaps. The objective of this study was to assess HCRU, outcomes, and costs among Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B. Results from this study will have implications for Medicaid administrators, Medicaid managed care pharmacy programs, and other stakeholders in the hemophilia B community.
Methods
This study employed a retrospective comparative cohort design to assess HCRU, outcomes, and costs among adult US Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B relative to a matched comparator population of beneficiaries without bleeding disorders.
DATA SOURCE
This study used Transformed Medicaid Statistical Information System Analytic Files (TAF) data from 2015 to 2020. These data contain information on every Medicaid beneficiary in all 50 states and territories in the United States, including information on beneficiary eligibility, service utilization, expenditure data, and fee-for-service (FFS) claims and managed care encounters. This analysis used TAF data from 2015 to 2020 because these were the most recent years of TAF data available at the time the study was conducted.
STUDY POPULATION
The analysis included adult male Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B as well as a population of matched comparator beneficiaries without bleeding disorders. We identified all male Medicaid beneficiaries that had at least 2 claims with a hemophilia B diagnosis code (Supplementary Table 1, available in online article) on different days from 2015 to 2020. Each beneficiary was assigned an index date, a 1-year observation period to establish Charlson Comorbidity Index (CCI) status and comparative information, and a following 1-year measurement period (Figure 1). Each beneficiary’s index date was defined as the date of their most recent medical visit with a hemophilia B diagnosis code when the beneficiary was continuously enrolled in Medicaid for at least 2 years following the date of the medical visit. The first year immediately following the beneficiary’s index date was defined as the beneficiary’s observation period, and the 1-year period following the observation period was defined as the beneficiary’s measurement period. Beneficiaries who were dual eligible for Medicaid and Medicare at any point during the 2-year follow-up period were excluded because the TAF data are missing some Medicare-reimbursed claims and encounters for these beneficiaries. PwHB were required to be aged at least 18 years at their assigned index date and were also required to meet the definition for receipt of FIX prophylaxis during their 1-year measurement period to be included in the final study population. The analysis was restricted to patients aged at least 18 years to promote homogeneity among the study population and to support comparisons with previous analyses in this space.

The population of matched comparator beneficiaries included male Medicaid beneficiaries that did not have bleeding disorder diagnoses from 2015 to 2020 (Supplementary Table 1). Each comparator beneficiary was assigned an index date based on the date of a medical visit when the beneficiary was continuously enrolled in Medicaid for at least 2 years following the date of the medical visit. We identified all possible index dates for each comparator beneficiary. The index date ultimately assigned to a given comparator beneficiary was based on the index date of their matched control. The first year immediately following a comparator beneficiary’s index date was defined as the beneficiary’s observation period, and the 1-year period following the observation period was defined as the beneficiary’s measurement period. Comparator beneficiaries were required to be aged at least 18 years at the time of their assigned index date, and beneficiaries who were dual eligible for Medicaid and Medicare at any point during the 2-year follow-up period were excluded.
Each PwHB included in the study cohort was matched to 4 comparator beneficiaries based on the following characteristics: age at index date, state of residence at index date, month and year of index date, CCI score assessed during the 1-year observation period, and FFS vs managed care enrollment during the 1-year measurement period. FFS vs managed care enrollment was assessed during the 1-year measurement period because a beneficiary’s Medicaid delivery system impacts how their costs are captured in the TAF data. These matching characteristics were chosen to promote a high degree of similarity between the 2 beneficiary groups. Additionally, matching on the month and year of the index date ensured that PwHB and matched comparator beneficiaries were assessed over the same time period. Most PwHB had more than 4 potential matched comparator beneficiaries, and in these cases, 4 comparator beneficiaries were randomly selected from the pool of eligible matches. Each comparator beneficiary was only matched to a single PwHB (eg, matching without replacement).
DEFINING RECEIPT OF FIX PROPHYLAXIS
Administrative claims data, including TAF data, do not include indicators to differentiate receipt of FIX prophylaxis from receipt of on-demand FIX therapy among PwHB, and there is not a universally accepted approach to identify FIX prophylaxis using claims data. We classified PwHB as receiving FIX prophylaxis if they did not have any gaps in FIX therapy days supply exceeding 60 days during their 1-year measurement period. This criterion was based off an approach previously used to examine prophylaxis for hemophilia A using administrative claims data.17,18 Both outpatient FIX therapy medical claims and FIX therapy pharmacy claims were included in the assessment of FIX prophylaxis. Inpatient medical claims and outpatient emergency department claims were excluded from the assessment of FIX prophylaxis as utilization in these settings likely reflects on-demand use for bleeding episodes or procedures rather than prophylactic use. The days supply associated with each FIX therapy claim was used to assess the length of gaps between claims. For example, if a PwHB had a FIX therapy claim for 30 days supply on April 1 and a subsequent claim on June 1, then they would have had a 31-day gap between claims. As TAF data do not capture days supply for medical claims, we imputed this based on the mode of days supply for the same drug type calculated from FIX therapy pharmacy claims when possible. For medical claims in which the specific type of FIX therapy could not be identified, days supply was imputed as 28.18
HCRU, OUTCOMES, AND COSTS
Measures of HCRU, outcomes, and costs were assessed for each beneficiary during their 1-year measurement period. These measures were summarized for each group using frequencies and percentages for categorical variables, and means, SDs, medians, and interquartile ranges for continuous variables. Key measures of HCRU included inpatient hospital admissions, length of stay associated with inpatient hospital admissions, emergency department visits, outpatient medical encounters, outpatient hematologist visits, non-FIX therapy prescription fills, and blood transfusions. Outpatient medical encounters included all types of outpatient medical visits (eg, general evaluation and management, psychotherapy, physical therapy, vaccinations, laboratory screenings, etc.). Outcomes of interest included any occurrence of arthroscopies/joint surgeries and bleeding events.
When assessing costs, we examined total costs during beneficiary’s 1-year measurement period, as well as medical costs and before-rebate pharmacy costs. We standardized all costs to 2020 dollars using the medical care component of the US Consumer Price Index. Pharmacy costs were further stratified as FIX therapy costs and non-FIX therapy costs. The cost analysis was restricted to beneficiaries enrolled in FFS Medicaid for the entire duration of their 1-year measurement period because of missing data on costs on managed care encounter records in the TAF data. For FFS beneficiaries, the TAF data include costs paid by Medicaid agencies to providers for individual FFS claims. For managed care Medicaid beneficiaries, the TAF data available to external researchers include costs associated with monthly capitation payments paid by Medicaid agencies to managed care organizations, but they do not include data on costs paid by managed care organizations to providers for individual encounters. This limitation prevented us from calculating the costs associated with specific types of services (eg, FIX therapy pharmacy claims) used by beneficiaries enrolled in managed care.
STATISTICAL ANALYSIS
Differences in baseline characteristics between the hemophilia B prophylaxis group and matched comparator group were assessed by calculating standardized mean differences. Standardized mean differences were calculated based on baseline characteristics, including age at index, month and year of index date, state of residence, enrollment in managed care vs FFS Medicaid, and CCI scores. After confirming baseline characteristics were adequately balanced, statistical tests for matched data were used to compare measures of HCRU, outcomes, and costs between the 2 groups. Continuous measures of HCRU and outcomes (eg, number of inpatient hospital admissions) were compared using generalized linear models that assumed negative binomial distributions and log link functions. Categorical measures of HCRU and outcomes (eg, percent of beneficiaries who had any inpatient hospital admissions) were compared using conditional logistic regression. Cost measures were compared using generalized linear models that assumed γ distributions and log link functions. P values for statistical tests comparing the 2 groups were included in summary tables.
Results
CHARACTERISTICS OF STUDY POPULATION
The final study sample included 83 PwHB meeting all inclusion criteria who were each matched to 4 comparator beneficiaries without bleeding disorders, giving a total study sample of 415 beneficiaries (Table 1). The groups were well balanced in terms of baseline demographic characteristics and CCI scores, and results from the analysis of standardized mean differences did not reveal any differences between the 2 groups (ie, standardized mean differences were calculated as 0 for all baseline characteristics) (Table 2). Most beneficiaries included in the analysis were aged 18-35 at index date (71%) and resided in the Southern or Western regions of the United States (71%). Most beneficiaries had index dates in 2018 (67.5%) because 2020 was the most recent year of data available and index date assignment was based on a beneficiary’s most recent medical visit for hemophilia B when they were continuously enrolled for at least 2 years following the date of the visit (Figure 1). Approximately 25% of the study population was enrolled in FFS Medicaid for the entire duration of their 1-year measurement period and were included in cost analyses that were restricted to the FFS population.
Inclusion criteria | n |
---|---|
Beneficiaries with at least 2 claims with hemophilia B diagnosis codes on different days from 2015 to 2020 | 3,142 |
Male beneficiaries | 2,491 |
Nondual-eligible beneficiaries continuously enrolled in Medicaid for at least 24 months following date of a hemophilia B claim (most recent date selected for beneficiaries with multiple potential dates) | 1,276 |
Adults aged ≥18 years at assigned index date | 434 |
Beneficiaries receiving FIX prophylaxis (defined as having no gaps in FIX therapy coverage >60 days during 12-month measurement period) | 86 |
Beneficiaries with at least 4 matched nondual-eligible comparator beneficiaries (matching based on age, state, CCI score, Medicaid delivery system, and month of index date) | 83 |
CCI = Charlson Comorbidity Index; FIX = factor IX.
Characteristics | Persons with hemophilia B receiving FIX prophylaxis (n = 83) | Comparator group without bleeding disorders (n = 332) | ||
---|---|---|---|---|
n | % | n | % | |
Age, mean (SD), years | 32.3 (12.5) | 32.3 (12.4) | ||
Age, median (IQR), years | 30 (22-39) | 30 (22-39) | ||
Age group in years | ||||
18-25 | 30 | 36.1 | 120 | 36.1 |
26-35 | 29 | 34.9 | 116 | 34.9 |
36-48 | 12 | 14.5 | 48 | 14.5 |
49+ | 12 | 14.5 | 48 | 14.5 |
Year of index date | ||||
2015 | —a | —a | —a | —a |
2016 | —a | —a | —a | —a |
2017 | 15 | 18.1 | 60 | 18.1 |
2018 | 56 | 67.5 | 224 | 67.5 |
Region | ||||
Northeast | —a | —a | —a | —a |
Midwest | —a | —a | —a | —a |
South | 28 | 33.7 | 112 | 33.7 |
West | 31 | 37.3 | 124 | 37.3 |
Medicaid delivery system | ||||
Fee-for-service | 21 | 25.3 | 84 | 25.3 |
Managed care | —a | —a | —a | —a |
Mixed fee-for-service/managed care | —a | —a | —a | —a |
Charlson Comorbidity Index | ||||
Mean score (SD) | 0.45 (1.21) | 0.45 (1.21) | ||
Myocardial infarction | —a | —a | —a | —a |
Congestive heart failure | —a | —a | —a | —a |
Peripheral vascular disease | —a | —a | —a | —a |
Cerebrovascular disease | —a | —a | —a | —a |
Dementia | 0 | 0.0 | —a | —a |
Chronic pulmonary disease | —a | —a | 27 | 8.1 |
Connective tissue disease-rheumatic disease | 0 | 0.0 | —a | —a |
Peptic ulcer disease | —a | —a | —a | —a |
Diabetes without complications | —a | —a | 22 | 6.6 |
Diabetes with complications | —a | —a | 11 | 3.3 |
Paraplegia and hemiplegia | —a | —a | —a | —a |
Renal disease | —a | —a | 15 | 4.5 |
Cancer | 0 | 0.0 | —a | —a |
Metastatic disease | 0 | 0.0 | 0 | 0.0 |
AIDS-HIV | 0 | 0.0 | 0 | 0.0 |
aDenotes cells with censored n’s ranging from 1 to 10. We are required to censor these cells per the Centers for Medicare and Medicaid Services’s cell suppression policy.
IQR = interquartile range; FIX = factor IX.
HCRU AND OUTCOMES
We found that PwHB were significantly more likely than their matched comparators without bleeding disorders to have multiple inpatient admissions (rate suppressed among PwHB because of Centers for Medicare and Medicaid Services cell suppression policies vs 3.3% among comparator beneficiaries; P = 0.0136) as well as significantly more admissions per person (5.2 per person among PwHB vs 2.18 per person among comparator beneficiaries; P = 0.0019) during the 1-year measurement period (Table 3). PwHB had a shorter mean length of stay per admission among beneficiaries with at least 1 admission relative to comparator beneficiaries (5.1 vs 6.4, respectively; P = 0.2999). However, we found that PwHB had a significantly longer cumulative length of stay per person among beneficiaries with at least 1 admission (3.6 vs 1.5, respectively; P = 0.2341), driven by the significantly greater number of admissions per person among the PwHB group. Rates of emergency department utilization were only marginally different between the 2 groups.
Category | Persons with hemophilia B receiving FIX prophylaxis (n = 83) | Comparator group without bleeding disorders (n = 332) | P |
---|---|---|---|
Inpatient hospital admissions | |||
Number of visits | |||
Beneficiaries with at least 1 visit, n (%) | —a | 33 (9.9) | 0.5491 |
Beneficiaries with multiple visits, n (%) | —a | 11 (3.3) | 0.0136 |
Number per person among full sample, mean (SD) | 0.63 (2.16) | 0.22 (0.92) | 0.0359 |
Number per person among beneficiaries with at least 1 visit, mean (SD) | 5.2 (4.02) | 2.18 (2.08) | 0.0019 |
Length of stay per admission among beneficiaries with at least 1 admission (days) | |||
Mean (SD) | 5.1 (1.97) | 6.4 (4.32) | 0.2999 |
Median (IQR) | 4.6 (3.6-6.1) | 5 (3-8) | N/A |
Cumulative length of stay per beneficiary (days) | |||
Mean among full sample (SD) | 3.64 (13.93) | 1.47 (7.31) | 0.2341 |
Median among full sample (IQR) | 0 (0-0) | 0 (0-0) | N/A |
Mean among beneficiaries with at least 1 admission (SD) | 30.2 (29.6) | 14.8 (18.7) | 0.0473 |
Median among beneficiaries with at least 1 admission (IQR) | 17.5 (6-55) | 7 (4-15) | N/A |
Minimum - maximum cumulative length of stay among beneficiaries with at least 1 admission | 3-89 | 2-83 | N/A |
Beneficiaries with an admission in which primary diagnosis is related to mental health or substance use disorder, n (%) | —a | —a | N/A |
ED visits, n (%) | |||
Beneficiaries with at least 1 visit | 32 (38.6) | 125 (37.7) | 0.8744 |
Beneficiaries with multiple visits | 13 (15.7) | 72 (21.7) | 0.2081 |
Number of visits per beneficiary, n (%) | |||
0 | 51 (61.5) | 207 (62.4) | N/A |
1 | 19 (22.9) | 53 (16.0) | |
2-3 | —a | 42 (12.7) | |
4+ | —a | 30 (9.0) | |
Number per person among full sample, mean (SD) | 1.42 (4.14) | 1.40 (4.04) | 0.963 |
Number per person among beneficiaries with at least 1 visit, mean (SD) | 3.69 (6.06) | 3.73 (5.91) | 0.9587 |
Minimum number - maximum number among beneficiaries with at least 1 visit | 1-30 | 1-43 | N/A |
Beneficiaries with an ED visit in which primary diagnosis is related to mental health or substance use disorder, n (%) | —a | 11 (3.3) | N/A |
Any outpatient medical encounter | |||
Beneficiaries with at least 1 visit, n (%) | 79 (95.2) | 270 (81.3) | 0.0016 |
Number per person among full sample, mean (SD) | 25.5 (67.6) | 36.5 (79.5) | 0.0884 |
Any outpatient hematologist visits | |||
Beneficiaries with at least 1 visit, n (%) | 40 (48.2) | —a | <0.0001 |
Number per person among full sample, mean (SD) | 1.58 (3.98) | 0.20 (2.74) | <0.0001 |
Non-FIX product prescription fills | |||
Beneficiaries with at least 1 fill, n (%) | 62 (74.7) | 236 (71.1) | 0.4706 |
Number per person among full sample, mean (SD) | 23.0 (39.8) | 26.1 (43.6) | 0.575 |
Blood transfusions | |||
Beneficiaries with at least 1 transfusion, n (%) | —a | 0 (0%) | N/A |
Number per person among full sample, mean (SD) | 0.11 (0.56) | 0 (0) | |
Bleeding events | |||
Beneficiaries with at least 1 event, n (%) | 15 (18.1) | —a | <0.0001 |
Number per person among full sample, mean (SD) | 0.54 (1.73) | 0.02 (0.22) | <0.0001 |
Arthroscopies/joint surgeries | |||
Beneficiaries with at least 1 event, n (%) | 0 (0) | 0 (0) | N/A |
Beneficiaries with multiple events, n (%) | 0 (0) | 0 (0) | |
Number per person among full sample, mean (SD) | 0 (0) | 0 (0) |
aDenotes cells with censored n’s ranging from 1 to 10. We are required to censor these cells per the Centers for Medicare and Medicaid Services’s cell suppression policy.
ED = emergency department; FIX = factor IX; IQR = interquartile range; N/A = not applicable.
Nearly all beneficiaries in the final analytic sample had at least 1 outpatient medical encounter during the 1-year measurement period. We found that the PwHB group had a significantly higher proportion of beneficiaries who had at least 1 outpatient medical encounter (95.2% of PwHB vs 81.3% of comparator beneficiaries; P = 0.0016), but comparator beneficiaries had more than 40% more outpatient medical encounters per person during the 1-year measurement period relative to PwHB (36.5 vs 25.5, respectively; P = 0.0884).
We identified less than 11 beneficiaries in the PwHB group and no beneficiaries in the comparator group that had blood transfusions during the 1-year measurement period. We did not identify any beneficiaries in either group that had arthroscopies or joint surgeries during the 1-year measurement period. As expected, the PwHB group had significantly higher rates of bleeding events (0.54 per person among PwHB vs 0.02 per person among comparator beneficiaries; P < 0.0001) and outpatient hematologist visits (1.58 per person among PwHB vs 0.20 per person among comparator beneficiaries; P < 0.0001) relative to the comparator group.
COSTS
We found that FFS PwHB had average total annual costs of $928,370 per person relative to only $34,553 per person among comparator beneficiaries (P < 0.0001) (Table 4). Total annual costs among FFS PwHB were overwhelmingly driven by pharmacy costs, accounting for more than 98% of total annual costs, and FFS PwHB had significantly higher pharmacy costs per person during the 1-year measurement period relative to comparator beneficiaries (annual pharmacy costs of $913,763 vs $6,505, respectively; P < 0.0001). We also found that more than 99% of pharmacy costs among FFS PwHB were attributable to FIX therapy claims. Conversely, total annual costs among the comparator group were driven by medical costs instead of pharmacy costs. We found that FFS comparator beneficiaries had nearly double mean medical costs per person relative to PwHB (annual medical costs of $28,047 vs $14,607, respectively; P = 0.0519). There were multiple potential reasons why FFS comparator beneficiaries had higher mean medical costs relative to PwHB. For example, although PwHB and comparator beneficiaries were matched on overall CCI scores, they were not matched on individual CCI diagnoses, some of which may be more expensive to treat on average than others. Further, we identified several comparator beneficiaries with high-cost chronic conditions not included in CCI calculations that drove up mean costs among this group. Results comparing baseline beneficiary characteristics, HCRU, and outcomes among FFS PwHB and matched comparator beneficiaries are included in Supplementary Tables 2 and 3. Results comparing baseline beneficiary characteristics, HCRU, and outcomes among FFS PwHB and matched comparator beneficiaries are generally similar to results among the full study population.
Cost category | Persons with hemophilia B receiving FIX prophylaxis (n = 21) | Comparator group without bleeding disorders (n = 84) | P | ||
---|---|---|---|---|---|
Mean (SD) | Median (IQR) | Mean (SD) | Median (IQR) | ||
Total annual costs per person | $928,370 ($508,739) | $748,876 ($643,768 - $1,006,962) | $34,553 ($58,229) | $8,037 ($1,476 - $49,361) | <0.0001 |
Annual medical costs per person | $14,607 ($30,929) | $1,511 ($522 - $16,198) | $28,047 ($52,782) | $4,763 ($1,295 - $37,444) | 0.0519 |
Annual pharmacy costs per person | $913,763 ($502,273) | $748,571 ($640,069 - $1,006,440) | $6,505 ($22,207) | $108 ($0 - $1,329) | <0.0001 |
Annual FIX therapy pharmacy costs per person | $907,535 ($499,945) | $747,146 ($627,458 - $1,006,440) | $0 ($0) | $0 ($0 - $0) | N/A |
Annual non-FIX therapy pharmacy costs per person | $6,227 ($12,134) | $115 ($0 - $7,196) | $6,505 ($22,207) | $108 ($0 - $1,329) | 0.5316 |
Pharmacy costs, including FIX therapy and non-FIX therapy costs, reflect before-rebate costs.
FIX = factor IX; IQR = interquartile range.
Discussion
This study was the first to our knowledge to examine HCRU, outcomes, and costs, among adult US Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B. We found that PwHB had significantly higher rates of inpatient hospital admissions and significantly longer cumulative lengths of stay per person relative to matched comparator beneficiaries without bleeding disorders. We also found that average annual costs among PwHB were nearly 30 times higher than average annual costs among matched comparator beneficiaries. Consistent with previous research among commercial and Medicare-insured PwHB, annual costs among PwHB were overwhelmingly driven by FIX therapy pharmacy claims, with these costs alone accounting for more than 97% of total annual costs among the PwHB group on average.8,19 Meanwhile, pharmacy costs accounted for less than 20% of the total annual costs among matched comparator beneficiaries.
Results from this analysis have important implications for state Medicaid programs. Previous estimates suggest that more than 30% of PwHB in the United States are covered by public insurance programs, including Medicaid.15 However, hemophilia B is a rare disease, and costs borne by PwHB are unlikely to exert a major impact on state Medicaid budgets overall given the relatively small population of PwHB covered by each state. For example, previous research estimated that FIX products accounted for nearly $238 million in Medicaid spending in 2019, but total Medicaid service spending in federal fiscal year 2019 reached nearly $600 billion.20,21 Despite the relatively small overall budget impact associated with hemophilia B, Medicaid agencies still need to account for high-cost FIX therapy when setting capitation rates for Medicaid managed care pharmacy benefits. Several states have chosen to “carve-out” FIX products from their managed care pharmacy benefit package and instead cover these drugs via the state’s FFS delivery system, potentially because of the high costs associated with FIX products.16
Findings from this study among Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B are generally consistent with previous research from Buckner et al examining HCRU and costs among commercial and Medicare-insured persons with severe hemophilia B. Consistent with the Buckner et al analysis, this analysis found that PwHB generally had higher rates of HCRU relative to comparator beneficiaries without bleeding disorders. However, the differences in HCRU between the 2 populations observed in this study were at times less stark than the differences observed in the Buckner et al analysis, potentially because of higher baseline rates of HCRU among adult Medicaid beneficiaries without bleeding disorders relative to the comparator patients included in the Buckner et al analysis.
Buckner et al also found that PwHB had vastly higher mean annual costs relative to comparator beneficiaries without bleeding disorders (eg, mean annual costs of $632,088 among persons with severe hemophilia B relative to $7,546 among comparator beneficiaries), and costs among people with severe hemophilia B were overwhelmingly driven by pharmacy costs. There are several potential reasons why annual costs among PwHB in the present study were higher than costs identified by Buckner et al. First, Buckner et al classified persons with severe hemophilia B as individuals with at least 6 FIX therapy claims during a 1-year period regardless of the amount of time in between claims or the number of days supply associated with each claim, whereas our definition of FIX prophylaxis accounted for the days supply associated with each claim and only included individuals who did not have gaps in FIX therapy coverage exceeding 60 days during a 1-year period. Defining FIX prophylaxis based on persistent FIX therapy coverage may have resulted in our study population including individuals who received greater cumulative FIX product volumes relative to the severe hemophilia B population included in the Buckner et al analysis. Second, the Medicaid population included in our study may have included individuals who were heavier or sicker on average, and thus required greater FIX therapy dosing, relative to the population included in the Buckner et al analysis. Finally, our study leveraged claims data through December 2020, although the Buckner et al analysis only leveraged claims data through February 2019, and higher costs among the PwHB population included in our study may reflect more recent market trends in FIX product utilization and associated costs.
Despite high costs associated with FIX therapy, FIX replacement therapy remains a vital treatment for PwHB. On-demand and prophylactic FIX replacement therapy revolutionized the treatment of hemophilia B and improved the quality of life and life expectancy for PwHB.22,23 FIX prophylaxis is the recommended treatment for all persons with severe hemophilia B without inhibitors. FIX prophylaxis significantly reduces bleed deaths and complications related to bleeding among PwHB and may offset long-term medical costs associated with joint surgery and disability that may be incurred in the absence of available therapy. Previous research has found that postponing the start of prophylaxis following an individual’s first joint bleed increases the long-term risk of developing arthropathy.24 Economic analyses have also found that the cost-effectiveness of FIX therapy is comparable with that of other orphan drugs.25 Improved disease management associated with FIX prophylaxis may also improve quality of life and reduce indirect costs among PwHB.
The treatment landscape for hemophilia B continues to evolve with new FIX therapies and other treatments on the horizon.26 Gene therapies are one such innovation that hold the promise to improve clinical outcomes and reduce long-term costs to the health care system associated with hemophilia B.27 Two gene therapies have already been approved by the US Food and Drug Administration for use in adult PwHB, with others in development.28 Ultimately, PwHB will need to work with their clinical care providers to determine which treatment options will work best for their needs. It will also be critical for Medicaid administrators to identify coverage options for hemophilia B gene therapies to ensure that Medicaid beneficiaries have equitable access to these treatments, including setting expectations on Medicaid managed care plans. For example, some state Medicaid agencies have implemented separate payment policies in which a hospital is provided an extra payment for a cell or gene therapy product in addition to their diagnosis-related group base payment for hospital services.29 Other states are testing value-based payment policies to ensure adequate reimbursement and patient access for existing cell and gene therapies.30 In the future, these policies may be extended to cover new gene therapies for hemophilia B to ensure that Medicaid beneficiaries have equitable access to a variety of treatment options.
LIMITATIONS
This study was subject to several limitations. First, the administrative claims data used for this analysis do not include indicators to differentiate on-demand FIX therapy from FIX prophylaxis. We classified FIX prophylaxis receipt using an approach previously developed to identify prophylaxis among persons with hemophilia A using administrative claims data, but the sensitivity and specificity of this approach have not been validated. Second, the administrative claims data used for this analysis do not include indicators for hemophilia B severity (eg, factor levels) so measures of HCRU and costs could not be adjusted for hemophilia B severity. Individuals with severe hemophilia B may develop inhibitors and experience higher rates of HCRU and costs relative to other PwHB; however, the development of inhibitors among PwHB is relatively rare and our analysis did not include any individuals with evidence of inhibitors (as assessed by examining claims for inhibitor bypassing agents).30 Third, this analysis included a relatively small sample of adult Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B. Future research should seek to assess HCRU and costs associated with FIX prophylaxis for hemophilia B among pediatric PwHB. Fourthly, the TAF data available for external researchers do not include data on costs associated with managed care encounter records, which required us to restrict the cost analysis for this study to a smaller subset of beneficiaries who were enrolled in FFS Medicaid for the entire duration of their 1-year measurement period. Finally, results from this analysis may have been impacted by changes in HCRU associated with the COVID-19 public health emergency.
Conclusions
This analysis found high rates of HCRU and costs among adult US Medicaid beneficiaries receiving FIX prophylaxis for hemophilia B relative to a matched comparator population of Medicaid beneficiaries without bleeding disorders. These results are aligned with previous research among commercial and Medicare-insured PwHB showing that the majority of costs associated with hemophilia B can be attributed to FIX therapy. Future research should examine hemophilia B costs and outcomes in the context of innovative gene therapies for hemophilia B and should also consider whether Medicaid beneficiaries and other vulnerable populations have equitable access to these treatments.
ACKNOWLEDGMENTS
The authors would like to acknowledge Mary-Lacey Reuther, Robert Rouse, and Vidhi Desai from CSL Behring as well as Tara Ahi from Manatt Health Strategies for their contributions to the study.
Supplementary Material
Supplemental Material
- Download
- 192.81 KB
REFERENCES
1.
Valentino LA. Blood-induced joint disease: The pathophysiology of hemophilic arthropathy. J Thromb Haemost. 2010;8(9):1895-902.
2.
Gringeri A, Ewenstein B, Reininger A. The burden of bleeding in haemophilia: Is one bleed too many? Haemophilia. 2014;20(4):459-63.
3.
World Federation of Hemophilia Report on the Annual Global Survey 2021. World Federation of Hemophilia; 2022. Accessed September 10, 2023. https://www1.wfh.org/publications/files/pdf-2324.pdf
4.
Hart DP, Matino D, Astermark J, et al. International consensus recommendations on the management of people with haemophilia B. Ther Adv Hematol. 2022;13:20406207221085200.
5.
Srivastava A, Santagostino E, Dougall A, et al. WFH guidelines for the management of hemophilia, 3rd edition. Haemophilia. 2020;26 Suppl 6:1-158.
6.
Chen SL. Economic costs of hemophilia and the impact of prophylactic treatment on patient management. Am J Manag Care. 2016;22(5 Suppl):s126-33.
7.
Chen CX, Baker JR, Nichol MB. Economic burden of illness among persons with hemophilia B from HUGS Vb: Examining the association of severity and treatment regimens with costs and annual bleed rates. Value Health. 2017;20(8):1074-82.
8.
Buckner TW, Bocharova I, Hagan K, et al. Health care resource utilization and cost burden of hemophilia B in the United States. Blood Adv. 2021;5(7):1954-62.
9.
Guh S, Grosse SD, McAlister S, Kessler CM, Souice JM. Health care expenditures for Medicaid-covered males with haemophilia in the United States, 2008. Haemophilia. 2012;18(2):276-83.
10.
Gater A, Thomson TA, Strandberg-Larsen M. Haemophilia B: Impact on patients and economic burden of disease. Thromb Haemost. 2011;106(3):398-404.
11.
Tencer T, Friedman HS, Li-McLeod J, Johnson K. Medical costs and resource utilization for hemophilia patients with and without HIV or HCV infection. J Manag Care Pharm. 2007;13(9):790-8.
12.
Tortella BJ, Alvir J, McDonald M, et al. Real-world analysis of dispensed IUs of coagulation factor IX and resultant expenditures in hemophilia B patients receiving standard half-life versus extended half-life products and those switching from standard half-life to extended half-life products. J Manag Care Spec Pharm. 2018;24(7):643-53.
13.
Eldar-Lissai A, Hou Q, Krishnan S. The changing costs of caring for hemophilia patients in the U.S.: Insurers’ and patients’ perspectives. Value Health. 2015;18(3):A304.
14.
Li N, Sawyer EK, Maruszczyk K, et al. Economic burden of hemophilia B in the US: A systematic literature review. J Drug Assess. 2019;8(Suppl 1):28.
15.
Centers for Disease Control and Prevention. Males with hemophilia registry report 2014-2017. Published September 3, 2019. Accessed October 22, 2023. https://www.cdc.gov/ncbddd/hemophilia/communitycounts/registry-report-males/health-insurance-coverage.html
16.
National Bleeding Disorders Foundation. Medicaid. Accessed October 24, 2023. https://www.hemophilia.org/advocacy/state-priorities/medicaid-state
17.
Croteau SE, Cook K, Sheikh L, et al. Health care resource utilization and costs among adult patients with hemophilia A on factor VIII prophylaxis: An administrative claims analysis. J Manag Care Spec Pharm. 2021;27(3):316-26.
18.
Thornburg CD, Adamski K, Cook K, et al. Health care costs and resource utilization among commercially insured adult patients with hemophilia A managed with FVIII prophylaxis in the United States. J Manag Care Spec Pharm. 2022;28(4):449-60.
19.
Burke T, Asghar S, O’Hara J, Chuang M, Sawyer EK, Li N. Clinical, humanistic, and economic burden of severe haemophilia B in adults receiving factor IX prophylaxis: Findings from the CHESS II real-world burden of illness study in Europe. Orphanet J Rare Dis. 2021;16(1):521.
20.
Hernandez I, Rowe D, Gellad WF, Good CB. Trends in the use of conventional and new pharmaceuticals for hemophilia treatments among Medicaid enrollees, 2005-2020. JAMA Network Open. 2021;4(6):e2112044.
21.
Annual Medicaid & CHIP Expenditures | Medicaid. Accessed October 23, 2023. https://www.medicaid.gov/state-overviews/scorecard/annual-medicaid-chip-expenditures/index.html
22.
Henderson W. A brief history of hemophilia treatment | Hemophilia News Today. Accessed November 21, 2023. https://hemophilianewstoday.com/social-clips/brief-history-hemophilia-treatment/
23.
National Bleeding Disorders Foundation. History. National Hemophilia Foundation. Accessed November 21, 2023. https://www.hemophilia.org/bleeding-disorders-a-z/overview/history
24.
Fischer K, van der Bom JG, Mauser-Bunschoten EP, et al. The effects of postponing prophylactic treatment on long-term outcome in patients with severe hemophilia. Blood. 2002;99(7):2337-41.
25.
Thorat T, Neumann PJ, Chambers JD. Hemophilia burden of disease: A systematic review of the cost-utility literature for hemophilia. J Manag Care Spec Pharm. 2018;24(7):632-42.
26.
CVS Caremark. What’s next for hemophilia treatments? CVS Health Payor Solutions. Accessed December 18, 2023. https://payorsolutions.cvshealth.com/insights/whats-next-hemophilia-treatments
27.
Pipe SW, Leebeek FW, Recht M, et al. Gene therapy with etranacogene dezaparvovec for hemophilia B. N Engl J Med. 2023;388(8):706-18.
28.
Otto MA. FDA approves second gene therapy for hemophilia B. Published April 26, 2024. Accessed June 6, 2024. https://www.medscape.com/viewarticle/fda-approves-second-gene-therapy-hemophilia-b-2024a100086v
29.
Ofengeym Y, Dworkowitz A, Fiori A. Accessing cell and gene therapies: Insights on coverage, reimbursement and emerging models. Accessed October 24, 2023. https://www.manatt.com/insights/white-papers/2023/accessing-cell-and-gene-therapies-insights-on-cove
30.
Santoro C, Quintavalle G, Castaman G, et al. Inhibitors in hemophilia B. Semin Thromb Hemost. 2018;44(6):578-89.
Information & Authors
Information
Published In

Journal of Managed Care & Specialty Pharmacy
Volume 30 • Number 10 • October 2024
Pages: 1095 - 1105
PubMed: 38923896
Copyright
Copyright © 2024, Academy of Managed Care Pharmacy. All rights reserved.
History
Published online: 26 June 2024
Published in print: October 2024
Authors
Funding Information
This research was funded by CSL Behring LLC.
Metrics & Citations
Metrics
Citations
Export citation
Select the citation format you wish to export for this article or chapter.
Cited By
There are no citations for this item