Health Analytics Research Articles

Antifibrotic therapies reduce mortality and hospitalization among Medicare beneficiaries with idiopathic pulmonary fibrosis.

BACKGROUND: Additional real-world studies are needed to more fully elucidate the effectiveness of antifibrotic treatment in slowing the progression of idiopathic pulmonary fibrosis (IPF). OBJECTIVE: To compare mortality and hospitalization between Medicare beneficiaries with IPF who initiate antifibrotic therapy and those who did not receive treatment. METHODS: A retrospective observational study of Medicare beneficiaries using the 100% Medicare Research Identifiable File was conducted. We included patients aged 67 years and over diagnosed with IPF (≥ 1 inpatient or ≥ 2 outpatient claims with IPF diagnosis) during the study period (January 1, 2010-December 31, 2017). Patients who initiated antifibrotic treatment (pirfenidone or nintedanib) between October 15, 2014 (FDA approval date) and December 31, 2017 (ie, treated patients) were compared with those who did not receive treatment during a historical period (January 1, 2012-October 14, 2014) before the availability of antifibrotics (ie, untreated historical controls). Patients were matched by propensity score, and the outcomes, mortality, and hospitalization (all cause and respiratory related) were compared using a Cox proportional hazards model. RESULTS: We identified 4,641 treated patients and 4,641 propensity score-matched controls who met all study criteria; 352 treated patients who lacked matches were excluded from the study. Cox regression analysis of treated patients vs matched controls showed a significantly lower risk of mortality (HR = 0.62, 95% CI = 0.57-0.68); lower risk of hospitalization (HR = 0.71, 95% CI = 0.67-0.76; HR = 0.70, 95% CI = 0.64-0.76); and lower rate in number of hospitalizations per month (incident rate ratio [IRR] = 0.65, 95% CI = 0.60-0.71; IRR = 0.65, 95% CI = 0.58-0.73). CONCLUSIONS: This study suggests that treatment with antifibrotics may confer a survival benefit and protection against all-cause and respiratory-related hospitalization for IPF patients. DISCLOSURES: This work was sponsored by F. Hoffmann-La Roche/Genentech, Inc. Corral is employed by Genentech, Inc. Reddy, Chang, Broder, and Gokhale are employed by Partnership for Health Analytic Research LLC, a health services research company, which was hired by Genentech to conduct this research. Mooney has received advisory board/consulting fees and research support from Genentech, unrelated to this work. Mooney also reports advisory board/consulting fees and research support from Boehringer Ingelheim; personal fees from Imvaria; and grants from Celgene and Pliant, unrelated to this work. Through their employment with Partnership for Health Analytic Research, Reddy, Chang, Broder, and Gokhale have been compensated to conduct research for AbbVie, Akcea, ASPC, Amgen, AstraZeneca, BMS, Boston Scientific Corporation, Celgene, Eisai, Ethicon, GRAIL, Helsinn, Illumina, Innovation and Value Initiative, Ionis, Jazz, Kite, Novartis, Otsuka, Pathnostics, PhRMA, Prothena, Sage, Verde Technologies, Genentech, Inc., Greenwich Biosciences, Inc., Mirum Pharmaceuticals, Inc., Sanofi US Services, Inc., Sunovion Pharmaceuticals, Inc., and Dompe US, Inc., unrelated to this work. This research was presented as an abstract at CHEST 2020 Annual Meeting (virtual), October 18-21, 2020, and American Thoracic Society 2020 Virtual Meeting, June 2020.

Machine Learning Methods for Predicting Postpartum Depression: Scoping Review.

BackgroundMachine learning (ML) offers vigorous statistical and probabilistic techniques that can successfully predict certain clinical conditions using large volumes of data. A review of ML and big data research analytics in maternal depression is pertinent and timely, given the rapid technological developments in recent years.ObjectiveThis study aims to synthesize the literature on ML and big data analytics for maternal mental health, particularly the prediction of postpartum depression (PPD).MethodsWe used a scoping review methodology using the Arksey and O’Malley framework to rapidly map research activity in ML for predicting PPD. Two independent researchers searched PsycINFO, PubMed, IEEE Xplore, and the ACM Digital Library in September 2020 to identify relevant publications in the past 12 years. Data were extracted from the articles’ ML model, data type, and study results.ResultsA total of 14 studies were identified. All studies reported the use of supervised learning techniques to predict PPD. Support vector machine and random forest were the most commonly used algorithms in addition to Naive Bayes, regression, artificial neural network, decision trees, and XGBoost (Extreme Gradient Boosting). There was considerable heterogeneity in the best-performing ML algorithm across the selected studies. The area under the receiver operating characteristic curve values reported for different algorithms were support vector machine (range 0.78-0.86), random forest method (0.88), XGBoost (0.80), and logistic regression (0.93).ConclusionsML algorithms can analyze larger data sets and perform more advanced computations, which can significantly improve the detection of PPD at an early stage. Further clinical research collaborations are required to fine-tune ML algorithms for prediction and treatment. ML might become part of evidence-based practice in addition to clinical knowledge and existing research evidence.

Sarcopenia Is a Clinically Relevant and Independent Predictor of Health Outcomes after Chimeric Antigen Receptor T-Cell Therapy for Lymphoma

Sarcopenia Is a Clinically Relevant and Independent Predictor of Health Outcomes after Chimeric Antigen Receptor T-Cell Therapy for Lymphoma

Dasatinib Treatment Patterns after Pleural Effusion Among Patients with Chronic Myeloid Leukemia

Dasatinib Treatment Patterns after Pleural Effusion Among Patients with Chronic Myeloid Leukemia

9.G. Workshop: Looking beyond COVID-19: Putting people at the centre in the digital health era

The COVID-19 pandemic has given rise to the use of several technologies to control disease transmission. Whilst these applications are intended to empower people to be informed, alerted or to be able to pursue a so-called return to normality their use has also raised several questions around the way in which data is being gathered and used across countries. However, the potential of big data and real time data in better managing health and disease as well as in leaving nobody behind may have also been underestimated to date. This roundtable will permit a deep dive and follow up discussion to some of the points raised in the plenary debate on Capturing the breadth and depth of the digital health era - beyond the COVID-19 pandemic.The workshop will kick off with three lightning presentations:COVID-19 contact tracing apps: Have they really been useful in the European Region?Clayton HamiltonThe EU Digital Vaccination Certificate - Where do freedom of movement, equity, and public health coincide? - EU representative tbcThe use of Big Data to improve people's health: Big Brother watching or genuine potential to improve health and well-being on a massive scale - David NovilloThe COVID-19 pandemic has accelerated the use of new and existing technologies to help support the national public health response and mitigate the impacts on the continuity of essential health services. While many digital applications employed in the context of COVID-19 have been designed to inform and empower individuals, or to enable a gradual relaxation of public health and social measures, their continued use beyond the scope of the pandemic has raised questions as to the gathering and use of data by governments and its potential to be repurposed. At the same time, the potential of big data and real-time data to improve the quality of care and disease management and in working towards universal health coverage have also been underestimated to date. Addressing the underlying issues of public trust for digital technologies and data sharing, appropriate regulation and accountability mechanisms to protect the privacy of individuals, and the development of health workforce competencies in using digital tools are also part of the complex picture of barriers and enablers to putting people at the centre of their own health and well-being in the digital era.This roundtable will involve a “technical deep dive” and follow up of the discussions and points raised in the plenary debate “Capturing the breadth and depth of the digital health era — beyond the COVID-19 pandemic”. The session will start with three lightning presentations to contextualize the application of digital technologies during the pandemic and will be followed by reflections and questions posed to a panel of civil society representatives in a discussion facilitated by the European Observatory on Health Systems and Policies.Speakers/Panelists Clayton Hamilton Digital Health Flagship, WHO/Europe Ioana-Maria Gligor European Reference Networks and Digital Health DG SANTE, EC David Novillo Data, Metrics and Analytics, WHO/Europe Dineke Zeegers Paget EUPHA Caroline Costongs EuroHealthNet Paolo Benanti Moral Theology, Bioethics, and Neuroethics, Gregorian Pontifical University, Rome, Italy Carlos Martins Faculty of Medicine of Porto, PortugalKey messages As we have learned from COVID-19, the use of several technologies to help overcome the pandemic also raises questions around the way data is being gathered and used.

A Two-Echelon Responsive Health Analytic Model for Triggering Care Plan Revision in Geriatric Care Management

Due to the increasing ageing population, how can caregivers effectively provide long-term care services to meet the older adults’ needs with finite resources is emerging. In addressing this issue, nursing homes are striving to adopt smart health with the internet of things and artificial intelligence to improve the efficiency and sustainability of healthcare. This study proposed a two-echelon responsive health analytic model (EHAM) to deliver appropriate healthcare services in nursing homes under the Internet of Medical Things environment. A novel care plan revision index is developed using a dual fuzzy logic approach for multidimensional health assessments, followed by care plan modification using case-based reasoning. The findings reveal that EHAM can generate patient-centred long-term care solutions of high quality to maximise the satisfaction of nursing home residents and their families. Ultimately, sustainable healthcare services can be within the communities.

Real-World Tumor Response of Palbociclib Plus Letrozole Versus Letrozole for Metastatic Breast Cancer in US Clinical Practice.

BackgroundLimited information exists regarding tumor response to palbociclib plus an aromatase inhibitor (AI) versus AI alone in real-world practice.ObjectiveTo evaluate the real-world tumor response of palbociclib plus letrozole (PAL+LET) versus LET alone as first-line treatment for patients with hormone receptor-positive/human epidermal growth factor receptor 2-negative metastatic breast cancer (HR+/HER2‒ MBC) in routine US clinical practice.Patients and MethodsThis retrospective analysis included HR+/HER2‒ MBC patients who initiated PAL+LET or LET as first-line treatment between February 2015 and September 2018 in the Flatiron Health Analytics database. Patients were followed until December 2018. Real-world best tumor response (rwBTR) was determined based on physicians’ assessment of radiologic evidence for change in burden of disease.ResultsOf the 1383 eligible patients who initiated PAL+LET or LET as first-line therapy in the Flatiron database, 968 patients had ≥ 1 tumor response assessment (662 received PAL+LET and 306 received LET). The rwBTR rate (complete response+partial response) in the first-line setting was 59.8% in the PAL+LET group and 39.2% in the LET group (odds ratio 2.31 (95% CI 1.75‒3.04), P < 0.0001). After 1:1 propensity-score matching, the rwBTR rate was 58.6% in the PAL+LET group versus 39.1% in the LET group (odds ratio 2.21 (95% CI 1.50‒3.25), P < 0.0001).ConclusionsThis real-world analysis demonstrated that HR+/HER2‒ MBC patients were more likely to respond to PAL+LET compared to LET. These findings further showed the effectiveness of PAL+LET therapy in the real-world setting and support the combination as a standard of care for MBC.Study RegistrationPfizer; NCT04176354; registered November 25, 2019Supplementary InformationThe online version contains supplementary material available at 10.1007/s11523-021-00826-1.

Making Open Source Project Health Transparent

We explore the Community Health Analytics for Open Source Software project and how it plays an integral role in the automation of key measures to make the state of open source readily observable.

Review of medical diagnostics via data mining techniques

Data mining is one of the most popular analysis methods in medical research. It involves finding patterns and correlations in previously unknown datasets. Data mining encompasses various areas of biomedical research, including data collection, clinical decision support, illness or safety monitoring, public health and inquiry research. Health analytics frequently uses computational methods for data mining, such as clustering, classification, and regression. Studies of large numbers of diverse heterogeneous documents, including biological and electronic information, provided medical and health studies.

Predictive Monitoring-Impact in Acute Care Cardiology Trial (PM-IMPACCT): Protocol for a Randomized Controlled Trial.

BackgroundPatients in acute care wards who deteriorate and are emergently transferred to intensive care units (ICUs) have poor outcomes. Early identification of patients who are decompensating might allow for earlier clinical intervention and reduced morbidity and mortality. Advances in bedside continuous predictive analytics monitoring (ie, artificial intelligence [AI]–based risk prediction) have made complex data easily available to health care providers and have provided early warning of potentially catastrophic clinical events. We present a dynamic, visual, predictive analytics monitoring tool that integrates real-time bedside telemetric physiologic data into robust clinical models to estimate and communicate risk of imminent events. This tool, Continuous Monitoring of Event Trajectories (CoMET), has been shown in retrospective observational studies to predict clinical decompensation on the acute care ward. There is a need to more definitively study this advanced predictive analytics or AI monitoring system in a prospective, randomized controlled, clinical trial.ObjectiveThe goal of this trial is to determine the impact of an AI-based visual risk analytic, CoMET, on improving patient outcomes related to clinical deterioration, response time to proactive clinical action, and costs to the health care system.MethodsWe propose a cluster randomized controlled trial to test the impact of using the CoMET display in an acute care cardiology and cardiothoracic surgery hospital floor. The number of admissions to a room undergoing cluster randomization was estimated to be 10,424 over the 20-month study period. Cluster randomization based on bed number will occur every 2 months. The intervention cluster will have the CoMET score displayed (along with standard of care), while the usual care group will receive standard of care only.ResultsThe primary outcome will be hours free from events of clinical deterioration. Hours of acute clinical events are defined as time when one or more of the following occur: emergent ICU transfer, emergent surgery prior to ICU transfer, cardiac arrest prior to ICU transfer, emergent intubation, or death. The clinical trial began randomization in January 2021.ConclusionsVery few AI-based health analytics have been translated from algorithm to real-world use. This study will use robust, prospective, randomized controlled, clinical trial methodology to assess the effectiveness of an advanced AI predictive analytics monitoring system in incorporating real-time telemetric data for identifying clinical deterioration on acute care wards. This analysis will strengthen the ability of health care organizations to evolve as learning health systems, in which bioinformatics data are applied to improve patient outcomes by incorporating AI into knowledge tools that are successfully integrated into clinical practice by health care providers.Trial RegistrationClinicalTrials.gov NCT04359641; https://clinicaltrials.gov/ct2/show/NCT04359641International Registered Report Identifier (IRRID)DERR1-10.2196/29631

Healthcare Utilization and Costs in Commercially Insured Patients with AL Amyloidosis

Objectives: Amyloid light chain (AL) amyloidosis is a rare, progressive, and typically fatal disease caused by extracellular deposition of misfolded immunoglobulin light chains (LCs). Soluble toxic aggregates and deposited fibrils (amyloid) lead to progressive failure of vital organs, including the heart, kidneys, and nervous system, causing significant morbidity and mortality. Its effects on resource utilization are poorly understood. The aim of this study was to estimate healthcare utilization and costs among patients with AL amyloidosis.Methods: In this cross-sectional study, for each calendar year (CY) 2007-2015, patients with AL amyloidosis were identified in the Truven MarketScan® Commercial and Medicare Supplemental databases if they had: (1) ≥1 inpatient or 2 outpatient claims consistent with AL amyloidosis [International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes: 277.30 or 277.39; International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes: E85.4x, E85.8x, or E85.9x] in any diagnosis field within that CY; and (2) AL treatment (eg, chemotherapy, hematopoietic stem cell transplant [HSCT]) following diagnosis. This algorithm was used to identify patients as there is no diagnosis code specific to AL amyloidosis. Exclusion criteria were age <18 and lack of enrollment on January 1 of ID year. Full-year enrollment was not required after exploratory analysis suggested most patients with partial year enrollment likely died during the year. We reported descriptive statistics on healthcare costs and utilization by CY and overall. Some patients may have been included in multiple CYs at different points in the course of their disease. Costs were adjusted to 2015 US$. All analyses were done using SAS®.Results: There were 7,326 patients identified overall (between 368 and 1,080 unique patients per year), with a mean (SD) age 63.6 (12.1), 45% of whom were female. All US regions were represented. Hospitalization was common: 50.1% (n=3,670) were admitted at least once during the year and 11.3% (827) were admitted 3 or more times. Utilization decreased over time (57.0% in 2007 and 46.2% in 2015 had any hospital admission). Among admitted patients, mean (SD) length of stay (LOS) was 14.7 (19.5) days; 3.7% of patients had HSCT, with mean (SD) LOS of 21.0 (12.6) for the transplant. Patients were seen for outpatient care with a mean (SD) of 45.2 (36.2) times per year. Mean (SD) [median] total annual cost was $101,855 (148,965) [51,939] for all patients. Medical costs were $88,801 (144,491) [38,936], with $50,126 (86,232) [22,100] accrued in the outpatient setting, $37,909 (100,892) [802] inpatient, and $766 (3,650) [0] emergency department. Outpatient pharmacy costs were $13,054 (26,606) [3,501]. Costs increased over time ($92,866 in 2007 to $114,030 in 2015).Conclusions: Patients with AL amyloidosis use substantial healthcare resources. Half of patients with AL amyloidosis were admitted to the hospital each year for an average stay of more than 2 weeks. Visits to laboratories, offices, and other outpatient sites occur almost twice a month. The total cost of this care is more than $100,000 per patient per year and rising. New therapies aimed at improving organ response have the potential to reduce disease burden and healthcare utilization. DisclosuresQuock:Prothena Biosciences Inc: Employment, Other: Stockholder. Yan:Partnership for Health Analytic Research, LLC: Employment. Chang:Partnership for Health Analytic Research, LLC: Employment. Broder:Partnership for Health Analytic Research, LLC: Employment.

Opioid Utilization Patterns in United States Patients with Sickle Cell Disease

Opioid Utilization Patterns in United States Patients with Sickle Cell Disease

Access to Care for Medicaid and Commercially-Insured United States Patients with Sickle Cell Disease

Access to Care for Medicaid and Commercially-Insured United States Patients with Sickle Cell Disease

VTE Recurrence and Safety of Anticoagulants Among Patients with Cancer Treated for Venous Thromboembolism

Introduction:Anticoagulation (AC) is effective for the treatment of venous thromboembolism (VTE) in cancer patients, but is associated with an increased risk of bleeding. Currently, LMWH monotherapy is the recommended treatment for cancer patients with VTE; however, emerging data from Medicare patients suggests that rivaroxaban had a significantly lower risk of VTE recurrence and similar risk of major bleeding compared to LMWH or warfarin. In this study, we compared the effectiveness and risk of major bleeding associated with these anticoagulants in a younger cohort of patients from a commercial claims database.Methods: Truven Health Analytics MarketScan databases were used to find patients aged ≥18 newly diagnosed with cancer who then had their first VTE (deep vein thrombosis (DVT) or pulmonary embolism (PE); excluding upper VTE) between 1/1/2013 - 9/30/2016 and AC start ≤ 7 days. Patients were categorized into LMWH, warfarin, or rivaroxaban cohorts based on index AC. The warfarin cohort included patients who started LMWH and warfarin ≤ 7 days of their first VTE. Inverse probability of treatment weights based on propensity score were used to adjust for treatment differences in the comparison of LMWH vs. rivaroxaban, warfarin vs. rivaroxaban, and LMWH vs. warfarin. Patients were observed until the earliest of end of eligibility or data end. VTE recurrences were defined conservatively as hospitalizations with a primary diagnosis of VTE ≥7 days after the first VTE. Major bleeding events were identified using validated criteria (Cunningham et al., 2011) and evaluated while patients were on treatment. Kaplan-Meier rates at 6 and 12 months were used to compare VTE recurrence between cohorts. Time on treatment was approximately 30-50% of total eligibility time, so rates at 3 and 6 months were used to compare major bleeding between cohorts. Cox proportional hazard models were used to compare VTE recurrence and major bleeding risk between cohorts.Results: A total of 12,457 patients (LMWH: N=4,313; warfarin; N=4,774; rivaroxaban: N=3,370) were included. After weighting the cohorts, baseline demographic and clinical characteristics were well balanced. Mean patient age ranged from 61.3 to 63.6 years. Median observation time with LMWH was shorter than with rivaroxaban (6.8 vs. 8.3 months, p<0.001) and with warfarin (7.4 vs. 9.8 months, p<0.001). Time on treatment was significantly shorter as well (Figures 1A-1C). VTE recurrence rates were significantly lower with rivaroxaban than with LMWH users at 6 months (8.7% vs. 11.7%; p=0.003) and 12 months (11.9% vs.14.7%; p=0.006), respectively, with a hazard ratio (HR [95%CI]) of 0.83 (0.73-0.96; p=0.010; Figure 1A). Major bleeding rates for rivaroxaban and LMWH users were similar at 3 months (3.2% vs. 3.5%; p=0.592) and 6 months (4.4% vs. 4.9%; p=0.438), respectively [HR 0.91 (0.71-1.17; p=0.455]. VTE recurrence rates for rivaroxaban and warfarin users were similar at 6 months (8.2% vs 8.8%; p=0.530) and 12 months (11.3% vs 11.6%; p=0.675), respectively [HR 0.95 (0.83-1.09; p=0.456; Figure 1B)]. Major bleeding rates for rivaroxaban and warfarin users were also similar at 3 months (3.2% vs. 2.8%; p=0.199) and 6 months (4.2% vs. 3.8%; p=0.362), respectively [HR 1.08 (0.86-1.37; p=0.500)]. Recurrence rates were significantly lower for warfarin users than for LMWH users at 6 months (10.2% vs. 12.4%; p=0.006) and 12 months (13.3% vs. 15.3%; p=0.011), respectively, but the rate of recurrence over the entire follow up was not significantly different between the two cohorts [HR of 0.91 (0.81-1.02; p=0.103; Figure 1C)]. Major bleeding rates for warfarin and LMWH users were similar at 3 months (3.1% vs. 3.7%; p=0.114) and 6 months (4.3% vs. 5.1%; p=0.109), respectively [HR of 0.86 (0.68-1.08; p=0.187)].Conclusion: In this large cohort of cancer patients treated for VTE, time on treatment was shorter for patients treated with LMWH than oral agents. Rivaroxaban was associated with a significantly lower rate of VTE recurrence than LMWH and a similar rate of VTE recurrence to warfarin; major bleed rates were similar for all. While LMWH has been associated with greater reductions in recurrent VTE compared to warfarin in trials, poor persistence on LMWH in real world settings may explain its lack of comparative effectiveness in this analysis. [Display omitted] DisclosuresKhorana:Bayer: Consultancy, Honoraria; Amgen: Consultancy, Honoraria, Research Funding; Sanofi: Consultancy, Honoraria; Janssen Scientific Affairs, LLC: Consultancy, Honoraria, Research Funding; Leo: Consultancy, Honoraria, Research Funding; Halozyme: Consultancy, Honoraria; Pfizer: Consultancy, Honoraria; Roche: Consultancy, Honoraria. Milentijevic:Janssen Scientific Affairs, LLC: Employment, Equity Ownership. McCormick:Janssen Scientific Affairs, LLC: Research Funding. Laliberté:Janssen Scientific Affairs, LLC: Research Funding. Crivera:Janssen Scientific Affairs, LLC: Employment, Equity Ownership. Lefebvre:Janssen Scientific Affairs, LLC: Research Funding. Lejeune:Janssen Scientific Affairs, LLC: Research Funding. Rozjabek:Janssen Scientific Affairs, LLC: Other: Fellowship supported by Janssen Scientific Affairs, LLC. Schein:Janssen Scientific Affairs, LLC: Employment, Equity Ownership, Other: Own in excess of $10,000 of J&J stock. Streiff:Roche: Research Funding; CSL Behring: Consultancy, Research Funding; Janssen Scientific Affairs, LLC: Consultancy, Research Funding; Portola: Research Funding.

Costs Associated with Productivity Loss Among U.S. Patients Newly Diagnosed with Multiple Myeloma Receiving Oral Versus Injectable Chemotherapy

BackgroundThe use of novel Multiple Myeloma (MM) drug agents has been associated with improved therapeutic outcomes and survival; however, MM continues to pose a significant economic burden on both patients and healthcare systems. Evaluating the economic implications and outcomes of available treatments is important to improve efficiency of care.Patients with MM may experience productivity loss, including lost days from work or inability to work due to MM symptoms or to undergoing treatment. Although direct costs of illness have been well described in the literature, indirect costs associated with MM are understudied. Here, we examine the extent of workplace productivity loss among U.S. adult patients with MM and associated costs. These outcomes were compared among patients who received injectable versus oral MM therapy.MethodsAnalyses were conducted using the Truven Health Analytics MarketScan Commercial Claims and Encounters with Medicare Supplemental Coordination of Benefits and Health and Productivity Management Databases. Subjects at least 18 years of age diagnosed with MM between January 1, 2008 and December 31, 2014 were included in our study and followed until December 31, 2015 (Figure 1). Subjects were analyzed one year prior to, and one year following first MM diagnosis. Patients receiving only oral therapy were compared with those who received injectable therapy.Productivity loss was evaluated using several key variables found within workplace absenteeism (WAB), short-term disability (STD), and long-term disability (LTD) claim files. Workplace productivity loss (WPL), an aggregate of absenteeism days, was calculated using claims for WAB, STD, and LTD benefits. Cost of productivity loss was calculated for each patient by multiplying the number of absenteeism days (as indicated in the aggregate WPL measure) by the average daily wage for all occupations, as reported by the U.S. Bureau of Labor Statistics (Table 1).Absenteeism days and costs were compared among study groups using multi-variable zero-inflated Poisson regression. The following variables were used as covariates in the model: age, index year, gender, Charlson comorbidity index, and plan type. An interrupted time-series analysis was also conducted to evaluate changes in productivity losses in the pre- vs. post-diagnosis periods, and to compare these among the treatment groups.ResultsThe final study cohort included 299 patients with newly diagnosed MM, of whom 73 received oral therapy and 226 received injectable therapy. Baseline characteristics of the study groups are depicted in Table 2. In the zero-inflated Poisson regression, treatment type was a significant predictor of productivity loss (Figure 2). Patients who received injectable therapy missed an average of 110 work days in the one year post-diagnosis, compared with 88 for patients receiving only oral therapy (difference of 22 days, 95% CI: 19 - 25, p<0.001). Treatment type was also a significant predictor of cost of lost productivity in this model (Figure 3). Patients who received injectable therapy experienced productivity loss valued at $18,329, compared with patients who only received oral drug therapy ($14,467). The difference in valuated productivity losses between study cohort groups was also statistically significant ($3,862, 95% CI: $3,518 - $4,205, p<0.001).An interrupted time-series analysis showed a significant increase in missed work days per month following the initial diagnosis of MM for both treatment groups. Patients receiving injectable therapy experienced an immediate increase of 6.9 lost productivity days per month (95% CI: 5.1, 8.7, p<0.001), versus patients receiving oral therapy only (4.4 days per month; 95% CI: 2.7, 6.2, p<0.001). The difference in these values was not statistically significant, although a strong trend was observed (2.5 day per month; 95% CI: -0.1, 5.0, p=0.057).ConclusionsIn conclusion, patients newly diagnosed with MM face significant losses in productivity. Injectable MM therapy users experience significantly more utilization of disability benefits and incur higher productivity costs, compared to oral MM therapy users. Further studies elucidating the nature of the differences between parenteral and non-parenteral chemotherapy users are needed. [Display omitted] DisclosuresMerola:Millennium Pharmaceuticals, Inc.: Consultancy. Yong:Millennium Pharmaceuticals, Inc., Cambridge, MA, USA, a wholly owned subsidiary of Takeda Pharmaceutical Company Limited: Employment. Noga:Millennium Pharmaceuticals, Inc., Cambridge, MA, USA, a wholly owned subsidiary of Takeda Pharmaceutical Company Limited: Employment. Shermock:Millenium Pharmaceuticals: Consultancy.

Epidemiology of AL Amyloidosis in a US Commercially Insured Population

Introduction: Amyloid light chain (AL) amyloidosis is a rare, progressive, and typically fatal disease caused by extracellular deposition of misfolded immunoglobulin light chains. There is a paucity of epidemiological studies of AL amyloidosis, and many existing studies are not recent. The objective of the current study was to provide an up-to-date estimate of the prevalence and incidence of AL amyloidosis in the US.Methods: This was a retrospective, cross-sectional study using 2007-2015 data from the Truven MarketScan® Commercial and Medicare Supplemental claims databases. As there is no diagnosis code specific to AL amyloidosis, the following algorithm was used to identify patients. Adults ≥18 years old with AL amyloidosis were identified if they (1) had at least one inpatient claim or two outpatient claims consistent with AL amyloidosis [International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes: 277.30 or 277.39; International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes: E85.4x, E85.8x, or E85.9x] in any diagnosis field within each calendar year and (2) received one AL-specific treatment (eg, chemotherapy, hematopoietic stem cell transplant [HSCT]) on or after the first amyloidosis diagnosis in the study period. Prevalence was calculated as the number of AL patients divided by the number of enrollees on June 30 of each calendar year and reported as per million person-years (PMPY). Incidence was calculated as the number of patients with AL amyloidosis who were disease free and enrolled in a health plan for 1 year prior, divided by the number of enrollees enrolled from July 1 of the previous year to June 30 of each calendar year and reported as PMPY. A sensitivity analysis requiring enrollees to have continuous enrollment in each calendar year (and in the year prior for incident cases) found similar estimates.Results: Between 2007 and 2015, 7,326 prevalent patients with AL amyloidosis were identified, ranging from 368 to 1,080 cases per year. The mean (SD) age of prevalent cases was 63 (12) years and was stable from 2007 to 2015. The prevalence of AL amyloidosis increased significantly during the study period: from 15.5 in 2007 to 40.5 cases PMPY in 2015, an annual percentage change (APC) of 12% (P < 0.001) (Figure 1). The incidence of AL amlydoisis ranged from 9.7 to 14.0 cases PMPY, with no statistically significant increase (APC:3.1%; P= 0.114) (Figure 2). Prevalence and incidence were generally higher in males than females.Conclusion: The pattern of an increase in AL amyloidosis prevalence, coupled with stable incidence, is consistent with earlier diagnosis over time that extends life, although our study could not determine whether this mechanism was responsible for the observed change. Our database included commercially insured patients only and so may not precisely reflect the US population. Cases were identified using claims, not clinical methods, which might overestimate or underestimate prevalence and incidence. Nonetheless, extrapolating from these data, there are at least 12,000 adults in the US currently living with AL amyloidosis, and the number seems likely to rise in the future. [Display omitted] DisclosuresQuock:Prothena Biosciences Inc: Employment, Other: Stockholder. Yan:Partnership for Health Analytic Research, LLC: Employment. Chang:Partnership for Health Analytic Research, LLC: Employment. Broder:Partnership for Health Analytic Research, LLC: Employment.

Real World Clinical and Economic Burden of Patients with Hodgkin Lymphoma (HL) Who Fail Frontline (FL) Therapy in the US

Introduction: Hodgkin lymphoma (HL) is a hematologic malignancy for which frontline treatment options and outcomes have remained unchanged for several decades. HL is considered to have a favorable prognosis although failure rates after FL therapy range from 10% to 30% depending on stage. Patients who fail go on to bear the burden of additional therapy after FL. Little real-world evidence exists on the burden of disease for patients who fail FL (FLF) therapy for healthcare providers to contextualize the value of novel therapies. This study aimed to identify FLF patients and compare healthcare costs between HL patients with and without FLF.Methods: This retrospective analysis used the MarketScan Commercial and Medicare Supplemental administrative claims databases to select adults aged >18 years with ≥ 1 inpatient or 2 outpatient claims with a HL diagnosis (ICD-9 201.xx excluding Hodgkin paragranuloma, Hodgkin sarcoma and lymphocytic-histiocytic predominance) from 1/1/2010 to 9/30/2015. Eligible patients were required to have continuous plan enrollment for > 6 months prior to the diagnosis, no prior cancer, and initiate a multi-agent regimen indicative of curative intent.Patients were in the FLF cohort if they had one of the following failure events: initiate a new chemotherapy after discontinuing FL (gap >60 days with no chemotherapy), switch to a new treatment, or having a stem cell transplant (SCT). Patients were in FLF+XRT cohort if they also received radiation after completing FL chemotherapy. Index dates for the FLF and FLF+XRT cohorts were the earliest date of the failure events, and were randomly assigned for non-FLF patients based on distribution of intervals between FL start date and index date for the FLF cohort. They had to have > 30-day post-index continuous enrollment and followed until date of disenrollment, death, or end of the study period.Annual healthcare costs were measured at 12-month post-index periods among patients with > 12-months post-index continuous enrollment. Costs were measured during FL and the entire follow-up period (reported as per patient per month [PPPM]).Results: 795 patients met the selection criterion (mean age 38.5 years, 47.5% female). 67.8% of patients 18-44 years old, 26.4% 45-64 years old, and 5.8% ≥65 years old. Most resided in urban areas (81.9%), and the mean pre-index Charlson Comorbidity Index score of study patients was 2.9±1.9. The two most frequent coded comorbidities during FL were symptoms involving the respiratory system (41.3%) and cardiovascular (22.9%) in both cohorts. Of the patients in the FL+XRT cohort, 60.1% received XRT after FL discontinuation, 21.8% discontinued and restarted therapy, 10.4% switched to a new therapy, and 7.8% went on to a SCT directly after FL.The PPPM costs for FLF cohort during the follow up was $12,494 (N=77, Table 1) and the annual all-cause cost of failure was $160,839 (N=33, Table 2) higher than patients who did not fail FL. The PPPM costs in FLF+XRT cohort (N=193) was $11,918 (vs $7,772) (Table 1). The annual all-cause cost in was $68,431 (N=94) higher than patients in non-FLF cohort (Table 2). PPPM costs during FL in FLF+XRT was $21,676 (vs $16,396) (Table 1) driven by outpatient services (50.5%) and inpatient admissions (48.6%). The FLF+XRT (vs non-FLF) cohort had a significantly shorter duration of FL therapy (average of 112 vs 131 days, p<0.05).The economic burden of failure varied by type (Table 3). The annual costs of failure due to restarting, switching, SCT, or radiation were $269,189, $46,951, $231,824, and $55,989, respectively. Similar trends in cost were observed among patients with ≥24 months post-index enrollment (n=200).Conclusions: HL patients who fail FL represent a substantially higher economic burden compared to those without FLF, not accounting for stage at diagnosis. We did not assess the indirect costs or societal impact of FLF for the population and therefore this is an underestimate of the overall burden. More research is needed to determine optimal treatment that could reduce the risk of progression, the need for treatment after completion of FL and enhance long-term clinical and economic outcomes. [Display omitted] DisclosuresStraus:Received consulting fee from Seattle Genetics for involvement in the research: Consultancy. Bonafede:Employee of Truven Health Analytics, an IBM Company, that received a research contract to conduct this study with and on behalf of Seattle Genetics, Inc: Employment. Feliciano:Seattle Genetics: Employment, Equity Ownership. Cai:Truven Health Analytics: Employment, Other: I am an employee of Truven Health Analytics, an IBM company which received funding from Seattle Genetics to conduct this study. Noxon:Truven Health Analytics, an IBM company which received project funding from Global Blood Therapeutics: Employment. Princic:Truven Health Analytics, IBM Company: Employment. Josephson:Seattle Genetics: Employment, Equity Ownership. Richhariya:Seattle Genetics: Employment, Equity Ownership.

Clinical and Economic Burden of Transfusion-Dependent β-Thalassemia in Adult Patients in the United States

Introduction: β-thalassemia is a genetic blood disorder in which defective β-globin synthesis leads to ineffective erythropoiesis. Patients with transfusion-dependent β-thalassemia require lifelong management with blood transfusions and iron chelation therapy. Complications related to iron overload can cause significant morbidity and mortality, and it is well understood that the chronic management of this disease presents a substantial burden to patients in terms of healthcare visits and costs. However, the specific clinical and economic burden has not been well characterized in the USA.Objectives: The objectives of this study were to analyze the clinical and economic burden of current disease management for patients with transfusion-dependent (TD) β-thalassemia in the USA over a 1-year time period.Methods: This retrospective matched cohort study was conducted using the Truven Health MarketScan® Research Databases (Truven Health Analytics, Ann Arbor, MI). The MarketScan® Databases provide detailed utilization, outcomes, and cost data for healthcare services performed in inpatient and outpatient settings, and contains de-identified data from > 110 million patient lives. Adult patients with TD β-thalassemia and matched controls were identified between Jan 1, 2011 and Sep 30, 2015. Patients were required to be ≥ 18 years old at the index date with ≥ 12 months of continuous enrollment post index date. TD β-thalassemia patients were required to have ≥ 2 claims with a β-thalassemia (ICD-9-CM 282.44, ICD-10-CM D56.1) or hemoglobin E-β thalassemia (ICD-9-CM 282.47, ICD-10-CM D56.5) diagnosis, and ≥ 8 transfusion events to be defined as TD. Patients were excluded if they had ≥ 2 claims with any sickle cell diagnosis. Controls were required to have ≥ 1 medical claim and were excluded if they had a diagnosis of any thalassemia or sickle cell condition. Patients with TD β-thalassemia were matched to controls in a 1:5 ratio using exact matching based on age, sex, index year, payer type, and census division. Comorbidities were not used in matching because patients with TD β-thalassemia have a high burden of disease-associated comorbidities. Index date was the first claim for a β-thalassemia diagnosis in patients with TD β-thalassemia or the first medical claim for controls. Outcomes examined were disease burden such as comorbid conditions, healthcare utilization and costs specific to the management of TD β-thalassemia, as well as general healthcare utilization and costs during the 12 months post index date. Chi-squared tests and Fisher's exact test were used for categorical variables and the Wilcoxon-Mann-Whitney test was used for continuous outcomes.Results: Fifty patients with TD β-thalassemia met selection criteria and were matched to 250 controls. Mean age was 34.7 ± 9.1 years for both patients and controls. Patients with TD β-thalassemia had higher rates of diabetes (18% vs 6%, P = 0.009), cardiac disease (30% vs 2%, P < 0.001), hypogonadism (22% vs 7%, P = 0.002), osteoporosis (16% vs 1%, P < 0.001), and hypothyroidism (12% vs 4%, P = 0.045) compared with controls. On average, patients with TD β-thalassemia required 17 transfusions per year, 23 days apart. Mean number of outpatient visits was significantly higher in the TD β-thalassemia cohort compared with matched controls; however, inpatient and ER visits were not significantly different (Table 1). Mean total healthcare costs among the TD β-thalassemia cohort were USD 128,062 ± 62,260, compared with USD 5,438 ± 11,855 in controls (P < 0.001). Medical and medication costs were significantly higher in the TD β-thalassemia cohort compared with matched controls (Table 2). Among patients with TD β-thalassemia, total costs were primarily driven by chelation and transfusion costs, with other costs displayed in the figure. The majority of patients with TD β-thalassemia received oral chelators (72%), whereas 30% of patients received injectable chelators; 14% of patients received both oral and injectable chelators.Conclusions: Ongoing comprehensive treatment for TD β-thalassemia poses a substantial clinical and economic burden as compared with matched controls. Patients with TD β-thalassemia had significantly higher healthcare resource utilization, medication costs, and total costs compared with matched controls, and required frequent visits to their healthcare provider, which poses a burden to their quality of life and work productivity. [Display omitted] DisclosuresSheth:Novartis, Celgene, ApoPharma, Bluebird Bio: Consultancy. Weiss:Celgene: Employment; Zocdoc: Other: Equity ownership (options have not vested yet); Zocdoc: Other: Husband's employment. Parisi:Celgene: Employment, Equity Ownership. Ni:Celgene: Employment.

Quality of Care in United States Children with Sickle Cell Anemia

Quality of Care in United States Children with Sickle Cell Anemia

Healthcare Costs Associated with Venous Thromboembolism in Cancer Patients Treated with Anticoagulants

Introduction: Venous thromboembolism (VTE) is highly prevalent in patients with active cancer and is associated with higher healthcare costs. Although LMWH is currently recommended for the treatment of VTE in cancer patients, oral anticoagulants warfarin and rivaroxaban are commonly used for treatment. Recent evidence among Medicare cancer patients with VTE showed that rivaroxaban is associated with lower healthcare costs than LMWH and similar costs to warfarin. This study compares healthcare costs of cancer patients treated with rivaroxaban or warfarin and LMWH in younger patients from a commercial insurance claims database.Methods: Medical and pharmacy claims from the Truven Health Analytics MarketScan databases in 1/1/2013 - 9/30/2016 were analyzed. Newly diagnosed cancer patients aged ≥18 were selected if they had a first VTE (deep vein thrombosis (DVT) or pulmonary embolism (PE); excluding upper VTE) diagnosis occurring within 30 days prior to or any time after their first cancer diagnosis and ≥1 dispensing of an anticoagulant agent within 7 days after their VTE diagnosis. Based on the first anticoagulant agent received, patients were classified into one of the following cohorts: LMWH, warfarin, or rivaroxaban. The warfarin cohort included patients who started LMWH and warfarin within 7 days of their first VTE. Inverse probability of treatment weights (IPTW) based on propensity score were used to adjust for differences between treatment cohorts. Baseline characteristics were evaluated during the 6-month period prior to the index VTE. VTE-related costs were identified with a primary or secondary diagnosis of DVT or PE, and were evaluated for the entire follow-up period, starting from the initiation of the anticoagulant therapy until end of eligibility or end of data, whichever was earlier. Healthcare costs per-patient-per-year (PPPY) were evaluated and compared using mean cost differences (CD). Significance in cost differences between cohorts and 95% confidence intervals (CI) were estimated using nonparametric bootstrap procedures.Results: A total of 12,457 patients (LMWH: N=4,313; warfarin: N=4,774; rivaroxaban: N=3,370) were included. After weighting the cohorts, baseline demographic and clinical characteristics were well balanced. The mean age of patients ranged from 61.3 to 63.6 years and approximately half of all patients were women. Across cohorts, 25%-28% of patients were diagnosed with PE, 47%-48% were diagnosed with DVT, and 24%-27% were diagnosed with both PE and DVT. Between 71% and 76% of patients had their first VTE within one year of their cancer diagnosis. At first cancer diagnosis, 80%-82% of patients had solid tumors. The percent with cancers at high-risk for VTE (stomach, pancreas, brain) varied from 7%-14%. The mean Charlson comorbidity index ranged from 2.9-3.0. During the 6-month baseline period, healthcare costs were similar across cohorts. Over the follow-up period, patients treated with rivaroxaban had significantly lower VTE-related healthcare costs when compared with patients in the LMWH cohort (Table 1). VTE-related hospitalization, emergency room (ER) visit, outpatient visit, pharmacy, imaging, and other costs were all significantly lower for patients treated with rivaroxaban than for patients treated with LMWH (all P <0.03). VTE-related healthcare costs were $17,847 for the rivaroxaban cohort and $28,273 for the LMWH cohort (CD= -$10,426 [-$13,461; -$7,981]). VTE-related healthcare costs were similar between the rivaroxaban cohort and the warfarin cohort at $14,880 for the rivaroxaban cohort and $14,911 for the warfarin cohort (CD [95% CI] = -$32 [-$2,618; $2,417] (Table 1)). Higher VTE-related pharmacy costs in the rivaroxaban cohort (CD= $832 [$630; $1,082]) were offset by lower VTE-related outpatient visit costs (CD= -$1,370 [-$1,977; -$915]).Conclusion: This real-world analysis based on a commercial claims database showed that cancer patients treated with rivaroxaban after an initial VTE incurred significantly lower VTE-related healthcare costs than patients treated with LMWH and similar costs to patients treated with warfarin. [Display omitted] DisclosuresStreiff:CSL Behring: Consultancy, Research Funding; Janssen Scientific Affairs, LLC: Consultancy, Research Funding; Portola: Research Funding; Roche: Research Funding. Milentijevic:Janssen Scientific Affairs, LLC: Employment, Equity Ownership. McCormick:Janssen Scientific Affairs, LLC: Research Funding. Laliberté:Janssen Scientific Affairs, LLC: Research Funding. Crivera:Janssen Scientific Affairs, LLC: Employment, Equity Ownership. Lefebvre:Janssen Scientific Affairs, LLC: Research Funding. Lejeune:Janssen Scientific Affairs, LLC: Research Funding. Rozjabek:Janssen Scientific Affairs, LLC: Other: Fellowship supported by Janssen Scientific Affairs, LLC. Schein:Janssen Scientific Affairs, LLC: Employment, Equity Ownership, Other: Own in excess of $10,000 of J&J stock. Khorana:Pfizer: Consultancy, Honoraria; Halozyme: Consultancy, Honoraria; Janssen Scientific Affairs, LLC: Consultancy, Honoraria, Research Funding; Sanofi: Consultancy, Honoraria; Bayer: Consultancy, Honoraria; Amgen: Consultancy, Research Funding; Leo: Consultancy, Honoraria, Research Funding.