Multivariate Sensitivity Analyses Research Articles

Multivariate sensitivity analysis for dynamic models with both random and random process inputs

Dynamic models with both random and random process inputs are frequently used in engineering. However, sensitivity analysis (SA) for such models is still a challenging problem. This paper, therefore, proposes a new multivariate SA technique to aid the safety design of these models. The new method can decompose the SA of dynamic models into a series of SA of their principle components based on singular value decomposition, which will make the SA of dynamic models much more efficient. It is shown that the effect of both random and random process inputs on the uncertainty of dynamic output can be measured from their effects on both the distributions and directions of the principle components, based on which the individual sensitivities are defined. The generalized sensitivities are then proposed to synthesize the information that is spread between the principal components to assess the influence of each input on the entire uncertainty of dynamic output. The properties of the new sensitivities are derived and an efficient estimation algorithm is proposed based on unscented transformation. Numerical results are discussed with application to a hydrokinetic turbine blade model, where the new method is compared with the existing variance-based method.

Effectiveness and cost-effectiveness of Improving clinicians\u2019 diagnostic and communication Skills on Antibiotic prescribing Appropriateness in patients with acute Cough in primary care in CATalonia (the ISAAC-CAT study): study protocol for a cluster randomised controlled trial

BackgroundDespite their marginal benefit, about 60% of acute lower respiratory tract infections (ALRTIs) are currently treated with antibiotics in Catalonia. This study aims to evaluate the effectiveness and efficiency of a continuous disease-focused intervention (C-reactive protein [CRP]) and an illness-focused intervention (enhancement of communication skills to optimise doctor-patient consultations) on antibiotic prescribing in patients with ALRTIs in Catalan primary care centres.Methods/designA cluster randomised, factorial, controlled trial aimed at including 20 primary care centres (N = 2940 patients) with patients older than 18 years of age presenting for a first consultation with an ALRTI will be included in the study. Primary care centres will be identified on the basis of socioeconomic data and antibiotic consumption. Centres will be randomly assigned according to hierarchical clustering to any of four trial arms: usual care, CRP testing, enhanced communication skills backed up with patient leaflets, or combined interventions. A cost-effectiveness and cost-utility analysis will be performed from the societal and national healthcare system perspectives, and the time horizon of the analysis will be 1 year. Two qualitative studies (pre- and post-clinical trial) aimed to identify the expectations and concerns of patients with ALRTIs and the barriers and facilitators of each intervention arm will be run. Family doctors and nurses assigned to the interventions will participate in a 2-h training workshop before the inception of the trial and will receive a monthly intervention-tailored training module during the year of the trial period. Primary outcomes will be antibiotic use within the first 6 weeks, duration of moderate to severe cough, and the quality-adjusted life-years. Secondary outcomes will be duration of illness and severity of cough measured using a symptom diary, healthcare re-consultations, hospital admissions, and complications. Healthcare costs will be considered and expressed in 2021 euros (year foreseen to finalise the study) of the current year of the analysis. Univariate and multivariate sensitivity analyses will be carried out.DiscussionThe ISAAC-CAT project will contribute to evaluate the effectiveness and efficiency of different strategies for more appropriate antibiotic prescribing that are currently out of the scope of the actual clinical guidelines.Trial registrationClinicalTrials.gov, NCT03931577.

Comparison of a New Reduced Toxicity Myeloablative Treosulfan and Fludarabine Preparative Regimen with Myeloablative Busulfan or Melphalan in Combination with Fludarabine in Older Patients with Acute Myeloid Leukemia or Myelodysplastic Syndromes: A Retrospective Matched Pair Analysis of Patients from a Prospective Randomized Trial and the European Blood and Marrow Transplantation Society Registry

Background The best preparative regimen for the growing number of older acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS) patients undergoing allogeneic hematopoietic cell transplantation (HCT) from matched related (MRD) or unrelated donors (MUD) remains undefined. A large randomized phase III trial (MC-FludT.14/L study: ClinicalTrials.gov Identifier: NCT00822393) recently demonstrated that myeloablative intravenous (IV) treosulfan (10 g/m² IV on days -4 to -2) in combination with fludarabine (TreoFlu) improves outcome in older and/or comorbid patients with AML in complete remission (CR) or MDS compared with the reference reduced intensity busulfan (0.8 mg/kg IV in 6-hour intervals on days -4 and -3) and fludarabine (30 mg/m² IV on days -6 to -2 in each study arm) regimen. The beneficial effect of the TreoFlu regimen resulted from a significantly reduced non-relapse mortality (NRM) and translated to improved event-free survival (EFS) and overall survival (OS) (Beelen DW et al The Lancet Haematology, 2019). These results raised the question, how this new regimen compares to broadly applied myeloablative regimens, namely busulfan (0.8 mg/kg IV in 6-hour intervals over 4 days) plus cyclophosphamide (120 mg/kg IV over 2 days) (BuCy) or melphalan (140 mg/m² IV over 1 day or 2 days) plus fludarabine (MelFlu) in older AML and MDS patients. To address this question, we performed a comparative analysis of MC-FludT.14/L study patients treated with the TreoFlu regimen and similar patients of the European Blood and Marrow Transplantation Society (EBMT) registry, who underwent HCT from MRD or MUD after the BuCy or MelFlu regimen between 2010 and 2016. Patients and Methods Inclusion criteria were essentially the same as for the MC-FludT.14/L-study (patient age 50 to 70 years [yrs], primary or secondary AML in CR or MDS, Karnofsky-index ≥ 60%, MRD or MUD, 1st HCT). The study objectives were to compare OS, relapse incidence (RI), and NRM at 2 yrs after HCT between the TreoFlu regimen and the BuCy or MelFlu regimen. A total of 1493 EBMT registry patients (median age 58 yrs, AML: n=1135 [76%], MDS: n=358 [24%]) were identified for the comparison with the 252 MC-FludT.14/L patients (median age 61 yrs, AML: n=174 [69%], MDS: n=78 [31%]). A 1:1 matching method based on propensity scores (PS) with 14 patient-, donor-, and disease-characteristics was used to reduce confounding due to differences between regimens and was performed separately for AML and MDS patients. With the exception of comparison between the TreoFlu and BuCy regimen in AML patients, a significantly higher proportion of patients in the TreoFlu regimen subsets had a HCT-comorbidity index > 2 compared to patient subsets treated with the BuCy or MelFlu regimen. Results For patients with AML, the 2-yrs OS estimate was significantly higher after the TreoFlu compared with the BuCy regimen (76.4%, 95%-confidence interval [95%-CI]: 66.8% - 85.9% vs 49.2%, 95%-CI: 36.4% - 62.1%, p<0.001) and MelFlu regimen (72.7%, 95%-CI: 63.7% - 80.7% vs 58.7%, 95%-CI: 48.3% - 69.1%, p=0.04). This was clearly related to the significantly lower 2-yrs NRM estimate after the TreoFlu compared with the BuCy regimen (3.9%, 95-CI: 0% - 8.2% vs 23.5%, 95%-CI: 13.1% - 33.9%, p<0.001) and MelFlu regimen (6.4%, 95-CI: 1.8% - 11.0% vs 17.5%, 95%-CI: 9.6% - 25.5%, p<0.02). For patients with MDS, the differences of the 2-yrs OS and NRM estimates between patient subsets were similar, but were only significant for the comparison of the OS estimate between the TreoFlu and BuCy regimen (72.0%, 95%-CI: 54.4% - 89.6% vs 30.5, 95%-CI: 6.1% - 54.9%, p<0.01). The adjusted hazard ratios of 2-yrs overall mortality and NRM between regimens including all eligible patients by multivariate sensitivity analysis based on Cox proportional hazards models are given in the table below. Notably, no differences between the respective 2-yrs RI were detectable when comparing these patient subsets. Conclusion In older AML and MDS patients, the new TreoFlu regimen compares favorable to the broadly applied myeloablative BuCy and MelFlu regimens. The substantially lower 2-yrs NRM estimate supports its superior tolerability in this patient population. This large retrospective comparative analysis provides a basis for properly designed randomized trials of the new toxicity reduced myeloablative TreoFlu regimen in comparison with other myeloablative regimens in this target population. Table Disclosures Beelen: Medac GmbH Wedel Germany: Consultancy, Honoraria. Stoelzel:Neovii: Other: Travel funding; Shire: Consultancy, Other: Travel funding; JAZZ Pharmaceuticals: Consultancy. Dreger:MSD: Membership on an entity's Board of Directors or advisory committees, Other: Sponsoring of Symposia; Neovii, Riemser: Research Funding; AbbVie, Gilead, Novartis, Riemser, Roche: Speakers Bureau; AbbVie, AstraZeneca, Gilead, Janssen, Novartis, Riemser, Roche: Consultancy.

Analysis of the cost-effectiveness of treatment strategies for CML with incorporation of treatment discontinuation

AbstractThe cost of tyrosine kinase inhibitors (TKIs) in the treatment of chronic myeloid leukemia (CML) is a substantial economic burden. In Japan, imatinib, dasatinib, and nilotinib are now approved as first-line treatment of CML in chronic phase. Recent “stop TKI” trials have shown that TKIs can be safely discontinued in nearly one-half of patients with sustained deep molecular response (DMR). In this study, we analyzed the cost-effectiveness of a simulated 10 years of CML treatment including stop TKI in both the United States and Japan. We constructed Markov models to compare 4 strategies in which treatment was initiated with imatinib, dasatinib, nilotinib, or any of these TKIs at the physician's discretion. Treatment was switched to another TKI in the case of intolerance or resistance to the initial TKI, and TKIs were discontinued if DMR persisted for 2 years. “Imatinib first” offered 7.34 quality-adjusted life years (QALYs) at the cost of $1 022 148 in the United States (US dollars) and ¥32 526 785 in Japan (Japanese yen). In comparison with imatinib first, the incremental cost-effectiveness ratio per QALY of “dasatinib first” (7.68 QALY, $1 236 052, ¥51 506 254), “nilotinib first” (7.64 QALY, $1 245 667, ¥39 635 598), and “physician's choice” (7.55 QALY, $1 167 818, ¥41 187 740) was $641 324, $696 717, and $666 634 in the United States and ¥54 456 325, ¥23 154 465, and ¥39 635 615 in Japan, respectively. None of the 3 strategies met the willingness-to-pay threshold. The results were robust to univariate and multivariate sensitivity analyses. Imatinib first was shown to be the most cost-effective approach even with the incorporation of stop TKI.

Evaluación Económica de Lenalidomida (Revlimid®) Para el Tratamiento Del Síndrome Mielodisplásico Con Deleción 5q de Riesgo Bajo/Intermedio-1

Background Lenalidomide is an effective treatment for patients with lower risk myelodysplastic syndrome (MDS) del5q. It reduces disease-related transfusion requirements, inhibits the malignant clone and causes a delay in tumor growth. There are also benefits in terms of quality of life and use of other health care resources. Objective To perform an economic evaluation of lenalidomide for transfusion-dependent patients with Low/Intermediate-1-risk MDS del 5q from the perspective of the public health care sector in Mexico. Material and methods We developed a cost-effectiveness analysis using a Markov model of 5 stages: MDS transfusion-dependence, MDS transfusion-independence, complications from transfusion, acute myeloid leukemia, and death. The analyzed period was 5 years. Discount rate of 5% after year 1 was applied. Efficacy was elicited from a phase III clinical trial, MDS-004, which shows lenalidomide benefit in patients achieving transfusion-independence. There were also benefits with lenalidomide related to extended overall survival and decreased risk of acute myeloid leukemia. Unitary cost are from public price list of IMSS, and resources use was calculated from a Delphi panel methodology with clinical hematologists. A multivariate sensitivity analysis using a Monte Carlo technique was performed. Results Over a 5-year period, each patient on lenalidomide avoided 487 days of transfusion, on average. The incremental cost per life-year without transfusion dependence is USD$14,072 (Exchange Rate = 1USD = 17.55 MXN). According to the sensitivity analysis the model is robust. If all expected patients are treated, their cost would be wUSD$19.7 million, which represents 0.18% of the total budget in the public health sector. Conclusions Lenalidomide provides important clinical benefits in the treatment of patients with lower risk myelodysplastic syndromes with del5q, with a limited economic impact and with a cost per life living without blood transfusions is within 3 times of the Mexican GDP (Gross Domestic Product).

Derivative-based generalized sensitivity indices and Sobol’ indices

In uncertainty quantification, multivariate sensitivity analysis (MSA), including variance-based sensitivity analysis, and derivative global sensitivity measure (DGSM) are widely used for assessing the effects of input factors on the model outputs. While MSA allows for identifying the order and the strength of interactions among inputs, DGSM provides only a global effect of inputs by making use of model derivatives. It is interesting to combine the advantages of both approaches and to come up with generalized sensitivity indices (GSIs) from MSA based on model derivatives. First, we derive the mathematical expressions of the total effect and total-interaction effect functionals based on derivatives. Second, we construct minimum variance unbiased estimators (MVUEs) of the total-effect and total-interaction effect covariance matrices, and third, we provide the estimators of the total and total-interaction GSIs as well as their consistency and asymptotic normality. Finally, we demonstrate the applicability of these new results by means of simulations.

Cost-Effectiveness of Urolithiasis Management in Pregnancy.

Urolithiasis complicating pregnancy presents a challenge for urology and obstetric teams. The management options of ureteral stenting, percutaneous nephrostomy and ureteroscopy vary significantly with regard to efficacy, complications, impact on quality of life and costs. This analytic model compares these factors to determine an optimal strategy per gestational age at presentation. A decision analytic model was built that compared stenting, percutaneous nephrostomy and ureteroscopy. Outcomes included treatment failure, need for re-treatment and complications stratified by severity. Probabilities and utilities were derived from the literature and costs were derived from institutional charge data. The time horizon was 1 year with disutilities limited to gestation length or recovery time for treatment or complication. Multivariate sensitivity analyses and Monte Carlo analysis were performed to evaluate model robustness. In a simulated cohort of 1,000 pregnant women with urolithiasis ureteroscopy would yield 960 quality adjusted life years, compared to 870 with stenting and 880 with percutaneous nephrostomy. Treatment via stenting or percutaneous nephrostomy would fail more than twice as often as ureteroscopy and more than half of patients would require subsequent definitive management. In this cohort ureteroscopy would save $38.8 million compared to stenting and $17.6 million compared to percutaneous nephrostomy. Given its lower costs and improved outcomes, ureteroscopy was the dominant strategy across all gestational ages and remained dominant in all sensitivity analyses and Monte Carlo simulations. Regardless of gestational age, ureteroscopy was a cost-effective strategy to manage urolithiasis compared to stenting or percutaneous nephrostomy. This result was robust against substantial changes in model parameters.

PCN214 COST-EFFECTIVENESS ANALYSIS OF NIVOLUMAB IN COMBINATION WITH IPILIMUMAB VERSUS SUNITINIB FOR THE FIRST-LINE TREATMENT OF INTERMEDIATE- TO POOR-RISK ADVANCED RENAL CELL CARCINOMA IN FRANCE

Nivolumab in combination with ipilimumab (NIVO+IPI) is the first immuno-oncology combination to demonstrate significant, long-term overall survival (OS) benefit in previously untreated patients with intermediate- to poor-risk advanced renal cell carcinoma (1L RCC) compared with standard of care (sunitinib). The objective of this study was to assess the cost-effectiveness of NIVO+IPI compared with sunitinib in 1L RCC from the French all payers perspective. A three-state partitioned survival model (progression-free disease, progressed disease and death) was developed with a one-week cycle length and lifetime time horizon. The model used patient characteristics, progression-free survival, OS, safety data and utilities (EQ-5D-3L) from the CheckMate-214 trial (NCT02231749). French costs for drug acquisition, drug administration, monitoring, terminal care, travel and adverse events were sourced from published prices. Treatment-specific utilities were calculated using the French value set. Costs and outcomes were discounted by 4.0% annually. Univariate deterministic and multivariate probabilistic sensitivity analyses (DSA, PSA) assessed robustness of the results. Outcomes of interest were total costs, quality-adjusted life-years (QALYs), life years (LYs), the incremental cost-utility ratio (ICUR), and the incremental cost-effectiveness ratio (ICER). NIVO+IPI was associated with higher total QALYs and LYs (3.53 and 4.95) versus sunitinib (2.71 and 3.87) and increased total cost (€160,751 versus €86,596), all respectively. This resulted in an ICUR of €89,793/QALY gained and an ICER of €68,626/LY gained versus sunitinib. Based on the DSA, key model drivers were drug acquisition costs, QALY discount rate, and variation of OS function parameters (95% confidence interval). The PSA confirmed robustness of the model results, with NIVO+IPI having a 62% probability of being cost-effective at a willingness-to-pay (WTP) threshold of €100,000/QALY. Driven by an increased and sustained OS benefit, NIVO+IPI would be a cost-effective treatment when compared with sunitinib for 1L RCC patients in France at a WTP threshold of €100,000/QALY.

PSU19 ROBOTIC ASSISTED GASTRECTOMY COMPARED WITH OPEN RESECTION: A CASE-MATCHED STUDY OF CLINICAL OUTCOMES AND COST-EFFECTIVENESS

In recent years, increasingly sophisticated tools have allowed for more complex robotic surgery. Robotic gastrectomy, however, is adopted in only a few selected centers. The goals of this study were to examine the adoption of robotic gastrectomy (RG) and to compare outcomes and costs between open and robotic gastric resections (OG). This is a case-matched analysis of patients who underwent robotic and open gastric resection performed at Sanchinarro University Hospital, Madrid from November 2011 to February 2017. Outcome parameters included surgical and post-operative costs, quality adjusted life years (QALY), and incremental cost per QALY gained or the incremental cost effectiveness ratio (ICER). A sensitivity analysis was carried out in order to propagate the uncertainty of the estimations to the results of the model. We use a multivariate and stochastic sensitivity analysis performed by 5000 Monte Carlo simulations. The cost-effectiveness plane was used to represent all pairs of solutions of the model. Two groups of demographically similar patients were analyzed: the robotic group (n = 20) and the open surgery group (n = 20). The mean operative costs were higher for RG (€ 8471.99 versus €7300.95; p= 0.008); however, the mean overall costs were lower for RG (€15807.13 vs 20520.38€; p=0.771). Mean QALYs at 1 year for RG (0.744) was higher than that associated with OG (0.706). RG for locally advanced gastric carcinoma is safe, and long-term outcomes are comparable to those patients who underwent open resection. RG resulted in a shorter hospital stay, less blood loss and morbidity comparable with the outcomes of open gastrectomy, and is a cost-effective aproach.

PSU11 ROBOTIC VS LAPAROSCOPIC RIGHT HEMICOLECTOMY: A COMPARATIVE STUDY OF CLINICAL OUTCOMES AND COSTS-EFFECTIVENESS

The costs involved in performing robotic surgery present a critical issue which has not been well addressed yet. The aim of this study is to compare the clinical outcomes and cost-effective outcomes of robotic versus laparoscopic surgery for right-sided colon cancer. This is an observational, comparative study whereby data were abstracted from a database of patients who underwent laparoscopic (LAP) and robotic (ROB) right hemicolectomy from October 2010 to March 2017, at Sanchinarro University Hospital. Outcome parameters included surgical and post-operative costs, quality adjusted life years (QALY), and incremental cost per QALY gained or the incremental cost effectiveness ratio (ICER). A sensitivity analysis was carried out in order to propagate the uncertainty of the estimations to the results of the model. We use a multivariate and stochastic sensitivity analysis performed by 5000 Monte Carlo simulations. The cost-effectiveness plane was used to represent all pairs of solutions of the model. A total of 20 robotic and 200 laparoscopic hemicolectomies were included. The mean operative time was significantly lower in the laparoscopic approach (336 versus 283 min;p= 0.001). The main pre-operative data, overall morbidity, hospital stay and oncological outcomes were similar in both groups. The mean operative costs were higher for robotic surgery (€6423.77 vs €4415.20; p= 0.006) and the mean overall costs were higher too (€ 9455.1 for for robotic and € 8227.5 for the laparoscopic approach; p= 0.835). Mean QALYs at 1 year for robotic group (0.5624) was higher than that associated with OL (0.5066) (p=0.018). At a willingness-to-pay threshold of 20000 € and 30000 €, there was a 64.68% and 71.09% probability that robotic group was cost-effective relative to LAP approach. Robotic right hemicolectomy approach appears to be cost-effective compared with laparoscopic approach.

Economic impact of implementing decennial tetanus toxoid, reduced diphtheria toxoid and acellular pertussis (Tdap) vaccination in adults in the United States

BackgroundIn the United States, persons ≥11 years are recommended to receive one dose of tetanus toxoid, reduced diphtheria toxoid and acellular pertussis (Tdap) vaccine, followed by decennial tetanus- and diphtheria-toxoid (Td) boosters. Many providers use Tdap instead of Td. We evaluated epidemiologic and economic impacts of replacing Td boosters with Tdap. MethodsWe used a static cohort model to examine replacing Td with Tdap over the lifetime of 4,386,854 adults ≥21 years. Because pertussis is underdiagnosed and true incidence is unknown, we varied incidence from 2.5 cases/100,000 person-years to 500 cases/100,000 person-years. We calculated vaccine and medical costs from claims data. We estimated cost per case prevented and per quality-adjusted life year (QALY) saved; sensitivity analyses were conducted on vaccine effectiveness (VE), protection duration, vaccine cost, disease duration, hospitalization rates, productivity loss and missed work. We did not include programmatic advantages resulting from use of a single tetanus-toxoid containing vaccine. ResultsAt lowest incidence estimates, administering Tdap resulted in high costs per averted case ($111,540) and QALY saved ($8,972,848). As incidence increased, cases averted increased and cost per QALY saved decreased rapidly. With incidence estimates of 250 cases/100,000 person-years, cost per averted case and QALY saved were $984 and $81,678 respectively; at 500 cases/100,000 person-years, these values were $427 and $35,474. In multivariate sensitivity analyses, assuming 250 cases/100,000 person-years, estimated cost per QALY saved ranged from $971 (most favorable) to $217,370 (least favorable). ConclusionsOur findings suggest that replacing Td with Tdap for the decennial booster would result in high cost per QALY saved based on reported cases. However, programmatic considerations were not accounted for, and if pertussis incidence, which is incompletely measured, is assumed to be higher than reported through national surveillance, substituting Tdap for Td may lead to moderate decreases in pertussis cases and cost per QALY.

Cost-effectiveness analysis of universal adult immunization with tetanus-diphtheria-acellular pertussis vaccine (Tdap) versus current practice in Brazil

BackgroundA pertussis outbreak occurred in Brazil from 2011 to 2014, despite high coverage of whole-cell pertussis containing vaccines in early childhood. Infants were the most affected. This study aimed to evaluate the cost-effectiveness of introducing universal adult vaccination with Tdap into the National Immunization Program in Brazil. MethodsEconomic evaluation using a dynamic model to compare two strategies: (1) universal vaccination with single dose of Tdap at 20 years of age and (2) current practice (only pregnant women pertussis vaccination). The health system perspective was adopted. Temporal horizon was 10 years. Discount rate of 5% was applied to costs and benefits. Vaccine effectiveness (VE) was obtained from a population-based observational study. Epidemiological, resource utilization and cost estimates were obtained from the Brazilian Health Information Systems. The primary outcome was cost per life year saved (LYS), based on life expectancy at birth in Brazil in 2015. Univariate and multivariate sensitivity analysis were performed. ResultsAdult vaccination with VE of 82.6% and coverage of 40%, at price of US$7.01 per dose, and assuming herd protection would avoid 167 infant deaths by pertussis, saving 12,325 years of life and costing a total of US$105495891.61, from the health system perspective. The universal immunization would result in ICER of US$8459.13. The results were highly sensitive to disease incidence. ConclusionsThe results suggest that universal adult vaccination with Tdap would not be a cost-effective intervention for preventing pertussis cases and deaths in infants in Brazil.

Cost-Effectiveness of Active Surveillance, Radical Prostatectomy and External Beam Radiotherapy for Localized Prostate Cancer: An Analysis of the ProtecT Trial

Despite increasing emphasis on value based care, to our knowledge the cost-effectiveness of prostate cancer management options has not been compared using prospective clinical trial data. The ProtecT (Prostate Testing for Cancer and Treatment) trial demonstrated no difference in survival in patients randomized to active surveillance, external beam radiotherapy or radical prostatectomy. We compared cost-effectiveness among the arms of ProtecT. Using a Markov model we compared the cost-effectiveness of active surveillance, radical prostatectomy and external beam radiotherapy based on ProtecT outcomes, specifically 6-year quality of life data and 10-year oncologic data. Costs were based on 2017 Medicare reimbursement while utility values were assigned using the literature. Univariable and multivariable sensitivity analyses were performed. Six years after randomization the mean costs per patient were $12,143 for active surveillance, $17,781 for radical prostatectomy and $29,238 for external beam radiotherapy. The incremental cost-effectiveness ratio relative to active surveillance was $127,752/QALY for radical prostatectomy and $381,894/QALY for external beam radiotherapy. Ten years after randomization radical prostatectomy ($5,627/QALY) and external beam radiotherapy ($78,291/QALY) were more cost-effective than active surveillance. The model was sensitive to the metastasis rate on active surveillance with a threshold of 2.4% at 10 years, below which active surveillance was more cost-effective than radical prostatectomy. On multivariable sensitivity analysis at 10 years using a willingness to pay threshold of $100,000/QALY the most cost-effective strategy was radical prostatectomy in 45% of model microsimulations, external beam radiotherapy in 30% and active surveillance in 25%. Although active surveillance represents a cost-effective strategy to manage localized prostate cancer during the initial several years after diagnosis, the relative cost-effectiveness of treatment emerges with extended followup.

PMD8 COST-EFFECTIVENESS OF THERMOABLATION VERSUS SURGERY FOR TREATMENT OF LIVER METASTASIS

To perform the cost-effectiveness analysis of thermoablation versus surgery for liver metastasis treatment (lesions smaller than 3.5 cm). In the present economic evaluation, overall survival and progression-free survival probabilities for surgery and thermoablation were based on a systematic review. The model considered as horizon the Brazilian’s population average lifetime (75 years old), while the initial study population age was set at 50 years. In the health state transition model, following health states were considered: stable disease, progression and death. The costs included: thermoablation = US$ 3,328.89; surgery = US$ 7,043.22; post-procedure follow-up = US$ 190.84; thermoablation adverse events = US$ 40.26; surgery adverse events = US$ 202.11; pre-progression monitoring after thermoablation = US$ 1,120.86; pre-progression monitoring after surgery = US$ 2,666.87; palliative care after thermoablation = US$ 8,036.85; palliative care after surgery = US$ 4,0598.58. Thermoablation was superior to surgery (dominated) and presented a cost reduction of US$ 10,804.32 and incremental effectiveness of 1% when compared to surgery. The multivariate probabilistic sensitivity analysis showed that after 100,000 iterations, thermoablation had the lowest ICER regardless of the willingness to pay threshold. In the studied scenario, thermoablation was considered dominant over surgery and should be considered for the treatment of liver metastasis in Brazil.

Platelet to Lymphocyte Ratio and Glucocorticoid Resistance in Newly Diagnosed Primary Immune Thrombocytopenia: A Retrospective Cohort Study.

BackgroundIn China, evidence regarding to the association between platelet to lymphocyte ratio (PLR) and glucocorticoid (GC) resistance in participants with primary newly identified immune thrombocytopenia (ITP) is limited. We aimed to investigate whether PLR is independently linked with GC-resistant ITP.Material/MethodsWe non-selectively and consecutively collected 154 newly diagnosed ITPs. The start enrollment time and the end enrollment time were from March 2013 to June 2017. The independent and dependent variables were PLR measured at diagnosis and GC non-response. Other variables involved in the present work can be summarized as demographic data and factors that were correlated with PLR reported by published studies. Univariate and multivariate binary logistic regression model and sensitivity analysis were used to evaluate the associations between PLR and GC resistance.ResultsAfter adjusting covariates, PLR level was negatively associated with GC non-response [odds ratio (OR)=0.89, 95% confidence intervals (CI): 0.80 to 0.98], and supported by propensity score matching model (OR=0.74, 95%CI: 0.57 to 0.96]. Nonlinearity of PLR and GC resistance was observed whose inflection point was 5.08 (by 2-piecewise model). The OR and 95%CI on both sides of inflection point were 3.14 (0.81 to 12.21) and 0.81 (0.69 to 0.95), respectively. Subgroup analysis showed no significant differences from subgroups.ConclusionsThreshold effect on PLR and GC resistance is observed. When PLR is larger than 5.08, a unit increase of PLR is independently associated with 19% reduction of GC resistance.

Cost-effectiveness of omalizumab for the treatment of moderate-to-severe uncontrolled allergic asthma in the United States

Objective: Uncontrolled asthma is associated with considerable clinical burden and costs to payers and patients. US economic models evaluating biologics using data from clinical trials demonstrate high incremental cost-effectiveness ratios (ICERs), but the cost-effectiveness based on real-world treatment patterns is unknown. This analysis used real-world evidence to assess the cost-effectiveness of adding omalizumab to standard of care (SOC).Methods: A Markov model was applied to track patients’ health states in 2-week cycles, comparing costs and treatment effects of SOC alone versus SOC + omalizumab over a lifetime (US payer perspective). Outcomes included exacerbation events, life years, quality-adjusted life years (QALYs), total costs, and an ICER. Patient characteristics, exacerbations, patient-reported outcomes, and work productivity were derived from the real-world PROSPERO (Prospective Study to Evaluate Predictors of Clinical Effectiveness in Response to Omalizumab) study. Published literature informed mortality, exacerbation-related disutility, and unit costs. Sensitivity analyses assessed model robustness.Results: Over a lifetime horizon, omalizumab was associated with an increase of 2.0 QALYs at a cost of $US 148,319 in patients with uncontrolled asthma (ICER of $75,319/QALY gained) and a reduction in exacerbations of 6.0 events/patient. Accounting for responder status improved the ICER ($70,505/QALY); incorporating indirect costs further reduced the ICER. One-way and multivariate sensitivity analyses confirmed that the base case outcome was robust to variation in inputs.Conclusions: Based on real-world outcomes, omalizumab may be cost-effective for uncontrolled asthma from the US payer perspective. Including broader evidence on treatment discontinuation, caregiver burden, and oral corticosteroid reduction from real-world studies may better reflect the effects and value of omalizumab for all healthcare stakeholders.

Five-layer border dressings as part of a quality improvement bundle to prevent pressure injuries in US skilled nursing facilities and Australian nursing homes: A cost-effectiveness analysis.

The BORDER III trial found that five‐layer silicone border dressings effectively prevented pressure injuries in long‐term care, but the value of this approach is unknown. Our objective was to analyse the cost‐effectiveness of preventing facility‐acquired pressure injuries with a quality improvement bundle, including prophylactic five‐layer dressings in US and Australian long‐term care. Markov models analysed the cost utility for pressure injuries acquired during long‐term care from US and Australian perspectives. Models calibrated outcomes for standard care compared with a dressing‐inclusive bundle over 18 monthly cycles or until death based on BORDER III outcomes. Patients who developed a pressure injury simulated advancement through stages 1 to 4. Univariate and multivariate probabilistic sensitivity analyses tested modelling uncertainty. Costs in 2017 USD and quality‐adjusted life years (QALYs) were used to calculate an incremental cost‐effectiveness ratio (ICER). Dressing use yielded greater QALYs at slightly higher costs from perspectives. The US ICER was $36 652/QALY, while the Australian ICER was $15 898/QALY, both of which fell below a willingness‐to‐pay threshold of $100 000/QALY. Probabilistic sensitivity analysis favoured dressings as cost‐effective for most simulations. A quality improvement bundle, including prophylactic five‐layer dressings, is a cost‐effective approach for pressure injury prevention in all US and Australia long‐term care residents.

Multivariate Ensemble Sensitivity Analysis for Super Typhoon Haiyan (2013)

Abstract Ensemble sensitivity is often a diagonal approximation to the multivariate regression, leading to a simple univariate regression. Comparatively, the multivariate ensemble sensitivity retains the full covariance matrix when computing the multivariate regression. The performances of both univariate and multivariate ensemble sensitivities in multiscale flows have not been thoroughly examined, and the demonstration of the latter in realistic applications has been sparse. A high-resolution ensemble forecast of Typhoon Haiyan (2013) is used to examine the performances of the two ensemble sensitivities. Compared to the multivariate sensitivity, the univariate sensitivity overestimates the forecast metric, especially at higher levels. To increase the predicted Haiyan’s intensity, multivariate ensemble sensitivity gives initial perturbations characterized by a warming area around the center of the storm, an increased moisture area around the eyewall, a stronger primary circulation around the radius of maximum wind, and stronger inflow at low levels and stronger outflow at high levels. Perturbed initial condition experiments verify that the predicted response from the multivariate sensitivity is more accurate than that from the univariate sensitivity. Therefore, the ability of multivariate sensitivity to provide more accurate predicted responses than the univariate sensitivity has been demonstrated in a realistic multiscale flow application.

A sequential experimental design for multivariate sensitivity analysis using polynomial chaos expansion

ABSTRACT Multivariate output sensitivity analysis has gained much attention when the output of the computational model is a vector. A preferable strategy to deal with the multivariate output issue is the covariance decomposition approach based on the polynomial chaos expansion (PCE) metamodel. However, since the PCE construction depends on the quality of experimental design to some extent, the selection of design points is significant in determining the accuracy of the sensitivity estimator. In this article, a PCE-based sequential experimental design is proposed to estimate the multivariate output sensitivity index. In this method, the optimal design point is sequentially selected to minimize the determinant of covariance matrix of the sensitivity estimator. To validate the performance of the proposed method, several numerical examples are presented, which show that the sequential design approach performs better than other prevalent methods in terms of accuracy and robustness.

Comparative Cost-effectiveness of Surgery, Collagenase Clostridium Histolyticum, and Penile Traction Therapy in Men with Peyronie’s Disease in an Era of Effective Clinical Treatment

BackgroundTraditionally, surgery has been considered the gold standard treatment for Peyronie’s disease (PD). Less-invasive alternatives, such as collagenase Clostridium histolyticum (CCH) and traction therapy, have been proposed and proven effective. AimTo compare cost-effectiveness of management options for PD. MethodsA Markov analytic model was created to compare the cost-effectiveness of treatment with a novel traction device, RestoreX (RXPTT), vs CCH vs surgery. Outcomes were derived from single-institution, prospective data of 63 men treated with RXPTT, 115 with CCH, and 23 with plication or incision and grafting. Costs were based on 2017 Medicare reimbursement and utility values from the literature. Main Outcome MeasuresModel outcomes included complications for each treatment arm, as well as the probability of success, which was defined as ≥20% improvement in curvature. Univariable and multivariable sensitivity analyses were performed to test the robustness of the model. ResultsOverall success rates were 96% (surgery), 66% (CCH), and 48% (RXPTT). At 10 years after treatment, RXPTT was the most cost-effective, with mean costs per patient of $883 (RXPTT), $11,419 (surgery), and $33,628 (CCH). CCH and surgery both resulted in a gain of quality adjusted life years (QALYs) relative to RXPTT (9.44 and 9.36 vs 9.27, respectively). Sensitivity analysis demonstrated greater cost-effectiveness for surgery if lower (≤46%) rates of postoperative erectile dysfunction or length loss (≤3%). CCH became more cost-effective at lower costs (≤$16,726) or higher success rates (≥76%). On multivariable sensitivity analysis at a willingness to pay threshold of $100,000/QALY, the most cost-effective strategy was RXPTT in 49%, surgery in 48%, and CCH in 3% of simulations. At a willingness to treat threshold of $150,000/QALY, the most cost-effective treatment option was RXPTT in 33%, surgery in 55%, and CCH in 12% of simulations. Clinical ImplicationsIn an era of value-based care, this model can guide cost-effective treatment selection on the basis of provider, patient, and payer characteristics. Strengths & LimitationsThe current study represents the first cost-effectiveness comparison of treatment modalities for PD and is strengthened by prospective data collection, large CCH and traction sample sizes, and robust sensitivity analyses. Consistent with cost-effective models, the model is limited by assumptions and may not apply to all scenarios. ConclusionsRXPTT represents a more cost-effective method for achieving ≥20% curvature improvement compared with surgery or CCH. Depending on treatment goals, rate of surgical complications, and willingness to pay threshold, surgery and CCH may become more cost-effective in select scenarios.Wymer K, Kohler T, Trost L. Comparative Cost-effectiveness of Surgery, Collagenase Clostridium Histolyticum, and Penile Traction Therapy in Men with Peyronie’s Disease in an Era of Effective Clinical Treatment. J Sex Med 2019;16:1421–1432.