Study ObjectiveUstekinumab was recently approved by the United States U.S. Food and Drug Administration for the treatment of Crohn's disease. In this analysis, we aimed to compare the cost‐effectiveness of ustekinumab, infliximab, or adalimumab for the treatment of moderate‐severe Crohn's disease in patients who failed conventional therapy (i.e., corticosteroids and immunomodulators) but were naïve to tumor necrosis factor antagonists (i.e., biologic drugs).DesignCost‐effectiveness analysis using a hybrid model structure (decision tree and Markov model).Measurements and Main ResultsA decision tree simulated biologic induction, and a Markov model simulated biologic and conventional therapy maintenance. Cycle length was 2 weeks with a discounted 5‐year time horizon and a limited U.S. societal perspective in the base case; results from a payer perspective are also reported. Transition probabilities, direct costs, indirect costs, and utilities were obtained from the literature. To measure relative treatment value (i.e., order of treatment cost‐effectiveness), net monetary benefits were reported for a $150,000 willingness‐to‐pay threshold per quality‐adjusted life‐year in the base case. Infliximab dominated both adalimumab and ustekinumab, with a net monetary benefit (NMB) of $9943 and $29,798, respectively, in the base case. Adalimumab dominated ustekinumab, with an NMB of $19,855. All biologics yielded similar quality‐adjusted life‐years (~3.5), whereas costs varied substantially ($50,510, $54,985, and $72,921 for infliximab, adalimumab, and ustekinumab, respectively). The payer perspective, alternate time horizons, and scenario analyses consistently showed infliximab dominance. One‐way, threshold, and probabilistic sensitivity analyses confirmed the robustness of these results with respect to all parameters. Although biosimilars were not explicitly modeled as comparators, one‐way sensitivity analysis showed that drug acquisition costs could alter relative treatment value but would have to be varied by at least 50%.ConclusionFor moderate‐severe Crohn's disease, infliximab yields significantly more NMBs compared with both adalimumab and ustekinumab. Additional clinical (e.g., empiric dosing, biologic cycling) and quality‐of‐life (e.g., lost productivity, disutility of home injections) research is needed to allow for model frameworks and parameters that more accurately reflect the nuances of Crohn's disease treatment.

No abstract

BACKGROUND There are limited longitudinal studies following up persons with hemophilia (PWH) in adherence to clotting factor treatment and health outcomes. The Hemophilia Utilization Group Studies part Va (HUGS Va) was a two-year observational study of persons with hemophilia A conducted from 2005-2007. Participants from HUGS Va were enrolled to the long-term follow-up study (HUGS LTS) in 2014. OBJECTIVES To compare participants' characteristics between baseline of HUGS Va and follow-up in LTS; and investigate the impacts of changes of participants' characteristics on annualized bleeding rates. METHODS We collected data on sociodemographic and clinical characteristics. Self-reported bleedings were obtained from periodic surveys for 2-years in HUGS Va, and a survey that asked bleedings in the past 6-month in HUGS LTS. Clotting factor dispensing records were collected prospectively for two years in HUGS Va, and retrospectively for six months prior to HUGS LTS enrollment. Annualized bleeding rates and factor dispensing (unit/kg body weight) were calculated. Adherence to factor treatment was determined by the ratio of dispensed clotting factor to clinical recommended factor usage. We classified age at baseline of HUGS Va into three groups: children (aged 2-11 years), adolescents (aged 12-20 years), and adults (aged ≥21 years). The characteristics of participants were compared among the three age groups using Chi-square tests for categorical variables and ANOVA for continuous variables for each study. Annualized factor dispensed and bleeding rates were compared between HUGS Va and HUGS LTS using paired T-tests. RESULTS A total of 74 persons participated in both HUGS Va and LTS with completed data to calculate annualized factor dispensed and bleeding rates were included to the analyses. The mean age was 17.8±11.4 years in HUGS Va, and 26.2±11.5 years in LTS, respectively. The sample had 43% of children, 22% adolescents, and 35% adults at baseline. All adolescents at baseline transitioned to young adults in LTS. Approximately 80% of participants were severe hemophilia. Adults (73%) were less likely to have an entire year of health insurance as compared to children (100%) or adolescents (93.8%, P<0.01) in HUGS Va. Health insurance status was not significantly different among age groups in LTS. In HUGS Va, children had the highest rate of prophylactic treatment (74%), adherence to factor treatment (62%), highest mean annualized factor dispensed (4724±4090 u/kg), lowest rate of self-reported moderate or severe joint pain (31%), and lowest mean annualized bleeding rate (4.2±4.7); while adults had the lowest rate of prophylactic treatment (31%), adherence (28%), lowest mean annualized factor dispensed (2084±1870 u/kg), highest rate of self-reported moderate or severe joint pain (73%), and highest mean annualized bleeding rate (11.9±9.3), all P<0.05 among age group comparisons for these variables. There were 18 (26%) people (22% children, 17% adolescents, and 61% adults) switching from episodic treatment to prophylactic treatment from HUGS Va to LTS; others remained on the same treatment regimens. Adolescents and adults increased in prophylactic treatment (63% to 80% for adolescents, 31% to 69% for adults, respectively), adherence (36% to 75%, 28% to 61%), and mean annualized factor dispensed (3206±2581 to 4931±2945 u/kg, 2084±1870 to 4612±2577 u/kg) from HUGS Va to LTS. Mean annualized bleeding rates were not statistically significant different between HUGS Va (7.3±8.4) and LTS (10.6±22.0, P>0.05). The bleeding rates were not significantly different among age groups in LTS (P=0.06). CONCLUSIONS Although current literature indicated that PWH in transition from childhood to adulthood maybe at high risk of adverse health outcomes due to poor management of their condition with diminishing influence from parents, adolescents showed a significant increase in prophylactic treatment, adherence to factor treatment, and annualized factor dispensed, which may be associated with unchanged annualized bleed rates after they transitioned to adulthood in this study. Adults had a larger increase in dispensed factors than adolescents. Prophylaxis and adherence to clotting factor treatment were associated with a lower bleeding rate. Our current analyses reinforce the importance of prophylaxis and adherence to factor treatment for decreasing bleedings in persons with hemophilia A. Disclosures Nichol: Bayer: Research Funding; CSL Behring: Research Funding; Bioverativ: Research Funding; Shire/Baxter: Research Funding; Pfizer: Research Funding; Novo Nordisk: Research Funding; Genentech: Research Funding. Curtis:Gilead: Honoraria; Pfizer: Research Funding; Novo Nordisk: Honoraria, Research Funding; Shire/Baxter: Research Funding; Bioverativ: Research Funding; National Hemophilia Foundation: Honoraria; CSL Behring: Research Funding; Bayer: Research Funding, Speakers Bureau; Genentech: Honoraria, Research Funding. Ding:Novo Nordisk: Employment; Bioverativ: Research Funding. Aliyev:Genentech: Research Funding. Lou:Bioverativ: Research Funding; Novo Nordisk: Research Funding; Genentech: Research Funding; Bayer: Research Funding; Pfizer: Research Funding; CSL Behring: Research Funding; Shire/Baxter: Research Funding. Ullman:Genentech: Research Funding. Tran:Bioverativ: Honoraria; Novo Nordisk: Honoraria; Bayer: Honoraria; Genentech: Research Funding. Baker:Genentech: Research Funding. Riske:Genentech: Research Funding. Wu:Pfizer: Research Funding; Genentech: Research Funding; Bioverativ: Research Funding; CSL Behring: Research Funding; Bayer: Research Funding; Shire/Baxter: Research Funding; Novo Nordisk: Research Funding.

Aim: To understand the financial impact of health system adoption of novel heart failure medications under US alternative payment models (APMs). Materials and methods: This study used a decision tree model to assess the financial impact of health system adoption of sacubitril/valsartan to treat acute decompensated heart failure (ADHF). A comparator scenario modeled current health care utilization and cost for treating hospitalized ADHF patients with angiotensin-converting-enzyme inhibitors (ACEi) or angiotensin-receptor blockers (ARB). The study then measured the impact of adopting sacubitril/valsartan to treat ADHF on health system economic outcomes. Differences in treatment efficacy were based on the PIONEER-HF clinical trial. The financial impact of changes in patient outcomes under the sacubitril/valsartan and ACEi/ARB arms was assessed across three APMs: the Medicare Shared Savings Program, Bundled Payments for Care Improvement, and fee-for-service payments adjusted according to the Hospital Readmission Reduction Program. Results: Sacubitril/valsartan reduced re-hospitalizations after an initial ADHF admission by 46.3% for individuals aged 18-64 years and 23.4% for individuals aged 65 years. Health systems' financial benefit of adopting sacubitril/valsartan was $740 per ADHF case per year (PCPY). Savings were larger for patients aged 65 years ($803 PCPY) compared to those <65 years ($653 PCPY). The majority of the health system financial benefit came from changes in APM bonus and penalty reimbursements. Valuebased payments from the Hospital Readmission Reduction Program ($1,190 financial gain PCPY) and the Bundled Care Payment Improvement Initiative ($645 financial gain PCPY) produced larger financial benefits than participation in the Medicare Shared Savings Program ($253 financial gain PCPY). Limitations: The model uses clinical trial data, which may not reflect real-world outcomes. Further, the financial implications were modeled based only on three widely used APMs. Conclusion: Sacubitril/valsartan adoption decreased hospitalizations and led to a positive net financial impact on health systems after accounting for APM bonus payments.

included same-day costs, costs for GERD-related procedures and any LF-related complications that occurred within 3 months after LF at a patient level. The UR of LF was calculated as the number of LF episodes divided by the total number of GERD patient-years throughout study period. ReSultS: 8,233 LF episodes among 8,197 patients were included in the analysis (7, and 1,093 Medicare-insured [MI]). The mean (standard deviation