PRIMCAT-CRC: a data-driven whole disease simulation model for predicting the number of colorectal cancer patients by stage and treatment line in Australia.

OBJECTIVES: Effective healthcare planning, resource allocation, and budgeting require accurate predictions of the number of patients needing treatment at specific cancer stages and treatment lines. The PRedicting the population health economic IMpact of current and new Cancer Treatments Colorectal Cancer (PRIMCAT-CRC) simulation model was developed to meet this requirement for all CRC stages and relevant molecular profiles in Australia. METHODS: Real-world data was used to estimate treatment utilisation and time-to-event distributions. This populated a discrete-event simulation, projecting the number of patients receiving treatment across all disease stages and treatment lines for CRC and forecasting the number of patients likely to utilise future treatments. Illustrative analyses were undertaken, estimating treatments across disease stages and treatment lines over a 5-year period (2022-2026). We demonstrated the model's applicability through a case study introducing pembrolizumab as a first-line treatment for mismatch-repair deficient stage IV. RESULTS: Clinical registry data from 7,163 patients informed the model. The model forecasts 15,738 incident and 2,821 prevalent cases requiring treatment in 2022, rising to 15,921 and 2,871 respectively by 2026. Projections show that over 2022-2026, there will be a total of 116,752 treatments initiated, with 43% intended for stage IV disease. The introduction of pembrolizumab is projected for 706 patients annually, totalling 3,530 individuals starting treatment with pembrolizumab over the forecasted period, without significantly altering downstream utilisation of subsequent treatments. CONCLUSIONS: PRIMCAT-CRC is a versatile tool that can be used to estimate the eligible patient populations for novel cancer therapies, thereby reducing uncertainty for policymakers in decisions to publicly reimburse new treatments.

Cost Effectiveness of Molecular Diagnostic Testing Algorithms for the Treatment Selection of Frontline Ibrutinib for Patients with Chronic Lymphocytic Leukemia in Australia.

BACKGROUND: Clinical indications for ibrutinib reimbursement in Australia should consider the inclusion of patients with chronic lymphocytic leukemia (CLL) harboring prognostically unfavorable TP53/IGHV genomic aberrations. This study assessed the cost effectiveness of five first-line treatment strategies in CLL for young (aged ≤ 65 years), fit patients without significant comorbidities: (1) no testing (fludarabine, cyclophosphamide and rituximab [FCR] for all), (2) test for del(17p) only, (3) test for TP53 gene mutation status, (4) test for TP53 and IGHV gene mutation status and (5) no testing (ibrutinib for all). METHOD: A decision analytic model (decision tree and partitioned survival model) was developed from the Australian healthcare system perspective with a lifetime horizon. Comparative treatment effects were estimated from indirect treatment comparisons and survival analysis using several studies. Costs, utility and adverse events were derived from public literature sources. Deterministic and probabilistic sensitivity analyses explored the impact of modeling uncertainties on outcomes. RESULTS: Strategy 1 was associated with 5.69 quality-adjusted life-years (QALYs) and cost 458,836 Australian dollars (AUD). All other strategies had greater effectiveness but were more expensive than Strategy 1. At the willingness-to-pay (WTP) threshold of 100,000 AUD per QALY gained, Strategy 1 was most cost effective with an estimated probability of 68.8%. Strategy 4 was cost effective between thresholds 155,000-432,300 AUD per QALY gained, and Strategy 5 >432,300 AUD per QALY gained. CONCLUSION: Population targeting using mutation testing for TP53 and IGHV when performed with del(17p) testing specifically in the context of frontline ibrutinib choice does not make a cost-ineffective treatment into a cost-effective treatment.

Comparing modeling strategies combining changes in multiple serum tumor biomarkers for early prediction of immunotherapy non-response in non-small cell lung cancer.

BACKGROUND: Patients treated with immune checkpoint inhibitors (ICI) are at risk of adverse events (AEs) even though not all patients will benefit. Serum tumor markers (STMs) are known to reflect tumor activity and might therefore be useful to predict response, guide treatment decisions and thereby prevent AEs. OBJECTIVE: This study aims to compare a range of prediction methods to predict non-response using multiple sequentially measured STMs. METHODS: Nine prediction models were compared to predict treatment non-response at 6-months (n = 412) using bi-weekly CYFRA, CEA, CA-125, NSE, and SCC measurements determined in the first 6-weeks of therapy. All methods were applied to six different biomarker combinations including two to five STMs. Model performance was assessed based on sensitivity, while model training aimed at 95% specificity to ensure a low false-positive rate. RESULTS: In the validation cohort, boosting provided the highest sensitivity at a fixed specificity across most STM combinations (12.9% -59.4%). Boosting applied to CYFRA and CEA achieved the highest sensitivity on the validation data while maintaining a specificity >95%. CONCLUSIONS: Non-response in NSCLC patients treated with ICIs can be predicted with a specificity >95% by combining multiple sequentially measured STMs in a prediction model. Clinical use is subject to further external validation.

Quantifying hospital-associated costs, and accompanying travel costs and productivity losses, before and after withdrawing tumour necrosis factor-alfa inhibitors in juvenile idiopathic arthritis.

OBJECTIVE: To quantify differences in hospital-associated costs, and accompanying travel costs and productivity losses, before and after withdrawing TNFi in JIA patients. METHODS: Retrospective analysis of prospectively collected data from electronic medical records of paediatric JIA patients treated with TNFi, which were either immediately discontinued, spaced (increased treatment interval) or tapered (reduced subsequent doses). Costs of hospital-associated resource use (consultations, medication, radiology procedures, laboratory testing, procedures under general anaesthesia, hospitalisation) and associated travel costs and productivity losses were quantified during clinically inactive disease until TNFi withdrawal (pre-withdrawal period) and compared with costs during the first and second year after withdrawal initiation (first and second year post-withdrawal). RESULTS: Fifty-six patients were included of whom 26 immediately discontinued TNFi, 30 spaced and zero tapered. Mean annual costs were €9,165/patient on active treatment (pre-withdrawal) and decreased significantly to €5,063/patient (-44.8%) and €6,569/patient (-28.3%) in the first and second year post-withdrawal, respectively (p< 0.05). Of these total annual costs, travel costs plus productivity losses were €834/patient, €1,180/patient, and €1,320/patient, in the three periods respectively. Medication comprised 80.7%, 61.5% and 72.4% of total annual costs in the pre-withdrawal, first, and second year post-withdrawal period, respectively. CONCLUSION: In the first two years after initiating withdrawal, the total annual costs are decreased compared with the pre-withdrawal period. However, cost reductions were lower in the second year compared with the first year post-withdrawal, primarily due to restarting or intensifying biologics. To support biologic withdraw decisions, future research should assess the full long-term societal cost impacts, and include all biologics.

Implementing competing risks in discrete event simulation: the event-specific probabilities and distributions approach.

Background: Although several strategies for modelling competing events in discrete event simulation (DES) exist, a methodological gap for the event-specific probabilities and distributions (ESPD) approach when dealing with censored data remains. This study defines and illustrates the ESPD strategy for censored data. Methods: The ESPD approach assumes that events are generated through a two-step process. First, the type of event is selected according to some (unknown) mixture proportions. Next, the times of occurrence of the events are sampled from a corresponding survival distribution. Both of these steps can be modelled based on covariates. Performance was evaluated through a simulation study, considering sample size and levels of censoring. Additionally, an oncology-related case study was conducted to assess the ability to produce realistic results, and to demonstrate its implementation using both frequentist and Bayesian frameworks in R. Results: The simulation study showed good performance of the ESPD approach, with accuracy decreasing as sample sizes decreased and censoring levels increased. The average relative absolute error of the event probability (95%-confidence interval) ranged from 0.04 (0.00; 0.10) to 0.23 (0.01; 0.66) for 60% censoring and sample size 50, showing that increased censoring and decreased sample size resulted in lower accuracy. The approach yielded realistic results in the case study. Discussion: The ESPD approach can be used to model competing events in DES based on censored data. Further research is warranted to compare the approach to other modelling approaches for DES, and to evaluate its usefulness in estimating cumulative event incidences in a broader context.

The effect of time before diagnosis and treatment on colorectal cancer outcomes: systematic review and dose-response meta-analysis.

BACKGROUND: This systematic review and meta-analysis aimed to evaluate existing evidence on the relationship between diagnostic and treatment intervals and outcomes for colorectal cancer. METHODS: Four databases were searched for English language articles assessing the role of time before initial treatment in colorectal cancer on any outcome, including stage and survival. Two reviewers independently screened articles for inclusion and data were synthesised narratively. A dose-response meta-analysis was performed to examine the association between treatment interval and survival. RESULTS: One hundred and thirty papers were included in the systematic review, eight were included in the meta-analysis. Forty-five different intervals were considered in the time from first symptom to treatment. The most common finding was of no association between the length of intervals on any outcome. The dose-response meta-analysis showed a U-shaped association between the treatment interval and overall survival with the nadir at 45 days. CONCLUSION: The review found inconsistent, but mostly a lack of, association between interval length and colorectal cancer outcomes, but study design and quality were heterogeneous. Meta-analysis suggests survival becomes increasingly poorer for those commencing treatment more than 45 days after diagnosis. REGISTRATION: This review was registered, and the protocol is available, in PROSPERO, the international database of systematic reviews, with the registration ID CRD42021255864.

Withdrawing biologics in non-systemic JIA: what matters to pediatric rheumatologists?

OBJECTIVE: Approximately one third of children with JIA receive biologic therapy, but evidence on biologic therapy withdrawal is lacking. This study aims to increase our understanding of whether and when pediatric rheumatologists postpone a decision to withdraw biologic therapy in children with clinically inactive non-systemic JIA. METHODS: A survey containing questions about background characteristics, treatment patterns, minimum treatment time with biologic therapy, and 16 different patient vignettes, was distributed among 83 pediatric rheumatologists in Canada and the Netherlands. For each vignette, respondents were asked whether they would withdraw biologic therapy at their minimum treatment time, and if not, how long they would continue biologic therapy. Statistical analysis included descriptive statistics, logistic and interval regression analysis. RESULTS: Thirty-three pediatric rheumatologists completed the survey (40% response rate). Pediatric rheumatologists are most likely to postpone the decision to withdraw biologic therapy when the child and/or parents express a preference for continuation (OR 6.3; p < 0.001), in case of a flare in the current treatment period (OR 3.9; p = 0.001), and in case of uveitis in the current treatment period (OR 3.9; p < 0.001). On average, biologic therapy withdrawal is initiated 6.7 months later when the child or parent prefer to continue treatment. CONCLUSION: Patient's and parents' preferences were the strongest driver of a decision to postpone biologic therapy withdrawal in children with clinically inactive non-systemic JIA and prolongs treatment duration. These findings highlight the potential benefit of a tool to support pediatric rheumatologists, patients and parents in decision making, and can help inform its design.

What matters most to pediatric rheumatologists in deciding whether to discontinue biologics in a child with juvenile idiopathic arthritis? A best-worst scaling survey.

OBJECTIVES: Care for JIA patients has been transformed in the biologics era; however, biologics carry important (although rare) risks and are costly. Flares after biological withdrawal are seen frequently, yet there is little clinical guidance to identify which patients in clinical remission can safely have their biologic discontinued (by stopping or tapering). We examined what characteristics of the child or their context are important to pediatric rheumatologists when making the decision to discuss withdrawal of biologics. METHODS: We conducted a survey including a best-worst scaling (BWS) exercise in pediatric rheumatologists who are part of the UCAN CAN-DU network to assess the relative importance of 14 previously identified characteristics. A balanced incomplete block design was used to generate choice tasks. Respondents evaluated 14 choice sets of 5 characteristics of a child with JIA and identified for each set which was the most and least important in the decision to offer withdrawal. Results were analyzed using conditional logit regression. RESULTS: Fifty-one (out of 79) pediatric rheumatologists participated (response rate 65%). The three most important characteristics were how challenging it was to achieve remission, history of established joint damage, and time spent in remission. The three least important characteristics were history of temporomandibular joint involvement, accessibility of biologics, and the patient's age. CONCLUSIONS: These findings give quantitative insight about factors important to pediatric rheumatologists' decision-making about biologic withdrawal. In addition to high quality clinical evidence, further research is needed to understand the perspective of patients and families to inform shared decision-making about biologic withdrawal for JIA patients with clinically inactive disease. Key Points ● What is already known on this topic-there is limited clinical guidance for pediatric rheumatologists in making decisions about biologic withdrawal for patients with juvenile idiopathic arthritis who are in clinical remission. ● What this study adds-this study quantitatively examined what characteristic of the child in clinical remission, or of their context, are most important to pediatric rheumatologists in deciding whether to offer withdrawal of biologics. ● How this study might affect research, practice or policy-understanding of these characteristics can provide useful information to other pediatric rheumatologists in making their decisions, and may guide areas to focus on for future research.

Population-Based Screening Using Low-Dose Chest Computed Tomography: A Systematic Review of Health Economic Evaluations.

BACKGROUND: Chest low-dose computed tomography (LDCT) is a promising technology for population-based screening because it is non-invasive, relatively inexpensive, associated with low radiation and highly sensitive to lung cancer. To improve the cost-effectiveness of lung cancer screening, simultaneous screening for other diseases could be considered. This systematic review was conducted to analyse studies that published evidence on the cost-effectiveness of chest LDCT screening programs for different diseases. METHODS: Scopus and PubMed were searched for English publications (1 January 2011-22 July 2022) using search terms related to screening, computed tomography and cost-effectiveness. An additional search specifically searched for the cost-effectiveness of screening for lung cancer, chronic obstructive pulmonary disease or cardiovascular disease. Included publications should present a full health economic evaluation of population screening with chest LDCT. The extracted data included the disease screened for, model type, country context of screening, inclusion of comorbidities or incidental findings, incremental costs, incremental effects and the resulting cost-effectiveness ratio amongst others. Reporting quality was assessed using the 2022 Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist. RESULTS: The search yielded 1799 unique papers, of which 43 were included. Most papers focused on lung cancer screening (n = 40), and three were on coronary calcium scoring. Microsimulation was the most commonly applied modelling type (n = 16), followed by life table analysis (n = 10) and Markov cohort models (n = 10). Studies reflected the healthcare context of the US (n = 15), Canada (n = 4), the UK (n = 3) and 13 other countries. The reported incremental cost-effectiveness ratio ranged from US$10,000 to US$90,000/quality-adjusted life year (QALY) for lung cancer screening compared to no screening and was US$15,900/QALY-US$45,300/QALY for coronary calcium scoring compared to no screening. DISCUSSION: Almost all health economic evaluations of LDCT screening focused on lung cancer. Literature regarding the health economic benefits of simultaneous LDCT screening for multiple diseases is absent. Most studies suggest LDCT screening is cost-effective for current and former smokers aged 55-74 with a minimum of 30 pack-years of smoking history. Consequently, more evidence on LDCT is needed to support further cost-effectiveness analyses. Preferably evidence on simultaneous screening for multiple diseases is needed, but alternatively, on single-disease screening. REGISTRATION OF SYSTEMATIC REVIEW: Prospective Register of Ongoing Systematic Reviews registration CRD42021290228 can be accessed https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=290228 .

A systematic review of methodological considerations in time to diagnosis and treatment in colorectal cancer research.

Research focusing on timely diagnosis and treatment of colorectal cancer is necessary to improve outcomes for people with cancer. Previous attempts to consolidate research on time to diagnosis and treatment have noted varied methodological approaches and quality, limiting the comparability of findings. This systematic review was conducted to comprehensively assess the scope of methodological issues in this field and provide recommendations for future research. Eligible articles had to assess the role of any interval up to treatment, on any outcome in colorectal cancer, in English, with no limits on publication time. Four databases were searched (Ovid Medline, EMBASE, EMCARE and PsycInfo). Papers were screened by two independent reviewers using a two-stage process of title and abstract followed by full text review. In total, 130 papers were included and had data extracted on specific methodological and statistical features. Several methodological problems were identified across the evidence base. Common issues included arbitrary categorisation of intervals (n = 107, 83%), no adjustment for potential confounders (n = 65, 50%), and lack of justification for included covariates where there was adjustment (n = 40 of 65 papers that performed an adjusted analysis, 62%). Many articles introduced epidemiological biases such as immortal time bias (n = 37 of 80 papers that used survival as an outcome, 46%) and confounding by indication (n = 73, 56%), as well as other biases arising from inclusion of factors outside of their temporal sequence. However, determination of the full extent of these problems was hampered by insufficient reporting. Recommendations include avoiding artificial categorisation of intervals, ensuring bias has not been introduced due to out-of-sequence use of key events and increased use of theoretical frameworks to detect and reduce bias. The development of reporting guidelines and domain-specific risk of bias tools may aid in ensuring future research can reliably contribute to recommendations regarding optimal timing and strengthen the evidence base.

Pharmacological treatment patterns in patients with juvenile idiopathic arthritis in the Netherlands: a real-world data analysis.

OBJECTIVE: To investigate medication prescription patterns among children with JIA, including duration, sequence and reasons for medication discontinuation. METHODS: This study is a single-centre, retrospective analysis of prospective data from the electronic medical records of JIA patients receiving systemic therapy aged 0-18 years between 1 April 2011 and 31 March 2019. Patient characteristics (age, gender, JIA subtype) and medication prescriptions were extracted and analysed using descriptive statistics, Sankey diagrams and Kaplan-Meier survival methods. RESULTS: Over a median of 4.2 years follow-up, the 20 different medicines analysed were prescribed as monotherapy (n = 15) or combination therapy (n = 48 unique combinations) among 236 patients. In non-systemic JIA, synthetic DMARDs were prescribed to almost all patients (99.5%), and always included MTX. In contrast, 43.9% of non-systemic JIA patients received a biologic DMARD (mostly adalimumab or etanercept), ranging from 30.9% for oligoarticular persistent ANA-positive JIA, to 90.9% for polyarticular RF-positive JIA. Among systemic JIA, 91.7% received a biologic DMARD (always including anakinra). When analysing medication prescriptions according to their class, 32.6% involved combination therapy. In 56.8% of patients, subsequent treatment lines were initiated after unsuccessful first-line treatment, resulting in 68 unique sequences. Remission was the most common reason for DMARD discontinuation (44.7%), followed by adverse events (28.9%) and ineffectiveness (22.1%). CONCLUSION: This paper reveals the complexity of pharmacological treatment in JIA, as indicated by: the variety of mono- and combination therapies prescribed, substantial variation in medication prescriptions between subtypes, most patients receiving two or more treatment lines, and the large number of unique treatment sequences.

The societal impact of implementing an at-home blood sampling device for chronic care patients: patient preferences and cost impact.

BACKGROUND: Diabetes mellitus, cardiovascular diseases, chronic kidney disease, and thyroid diseases are chronic diseases that require regular monitoring through blood tests. This paper first investigates the experiences of chronic care patients with venipuncture and their expectations of an at-home blood-sampling device, and then assesses the impact on societal costs of implementing such a device in current practice. METHODS: An online survey was distributed among chronic care patients to gain insight into their experience of blood sampling in current practice, and their expectations of an at-home blood-sampling device. The survey results were used as input parameters in a patient-level monte carlo analysis developed to represent a hypothetical cohort of Dutch chronically ill patients to investigate the impact on societal costs compared to usual care. RESULTS: In total, 1311 patients participated in the survey, of which 31% experience the time spent on the phlebotomy appointment as a burden. Of all respondents, 71% prefer to use an at-home blood-sampling device to monitor their chronic disease. The cost analysis indicated that implementing an at-home blood-sampling device increases the cost of phlebotomy itself by €27.25 per patient per year, but it reduces the overall societal costs by €24.86 per patient per year, mainly due to limiting productivity loss. CONCLUSIONS: Patients consider an at-home blood-sampling device to be more user-friendly than venous phlebotomy on location. Long waiting times and crowded locations can be avoided by using an at-home blood-sampling device. Implementing such a device is likely cost-saving as it is expected to reduce societal costs.

Modeling strategies to analyse longitudinal biomarker data: An illustration on predicting immunotherapy non-response in non-small cell lung cancer.

Serum tumor markers acquired through a blood draw are known to reflect tumor activity. Their non-invasive nature allows for more frequent testing compared to traditional imaging methods used for response evaluations. Our study aims to compare nine prediction methods to accurately, and with a low false positive rate, predict progressive disease despite treatment (i.e. non-response) using longitudinal tumor biomarker data. Bi-weekly measurements of CYFRA, CA-125, CEA, NSE, and SCC were available from a cohort of 412 advanced stage non-small cell lung cancer (NSCLC) patients treated up to two years with immune checkpoint inhibitors. Serum tumor marker measurements from the first six weeks after treatment initiation were used to predict treatment response at 6 months. Nine models with varying complexity were evaluated in this study, showing how longitudinal biomarker data can be used to predict non-response to immunotherapy in NSCLC patients.

Do Health Technology Assessment organisations consider manufacturers' costs in relation to drug price? A study of reimbursement reports.

INTRODUCTION: Drug reimbursement decisions are often made based on a price set by the manufacturer. In some cases, this price leads to public and scientific debates about whether its level can be justified in relation to its costs, including those related to research and development (R&D) and manufacturing. Such considerations could enter the decision process in collectively financed health care systems. This paper investigates whether manufacturers' costs in relation to drug prices, or profit margins, are explicitly mentioned and considered by health technology assessment (HTA) organisations. METHOD: An analysis of reimbursement reports for cancer drugs was performed. All relevant Dutch HTA-reports, published between 2017 and 2019, were selected and matched with HTA-reports from three other jurisdictions (England, Canada, Australia). Information was extracted. Additionally, reimbursement reports for three cases of expensive non-oncolytic orphan drugs prominent in pricing debates in the Netherlands were investigated in depth to examine consideration of profit margins. RESULTS: A total of 66 HTA-reports concerning 15 cancer drugs were included. None of these reports contained information on manufacturer's costs or profit margins. Some reports contained general considerations of the HTA organisation which related prices to manufacturers' costs: six contained a statement on the lack of price setting transparency, one mentioned recouping R&D costs as a potential argument to justify a high price. For the case studies, 21 HTA-reports were selected. One contained a cost-based price justification provided by the manufacturer. None of the other reports contained information on manufacturer's costs or profit margins. Six reports contained a discussion about lack of transparency. Reports from two jurisdictions contained invitations to justify high prices by demonstrating high costs. CONCLUSION: Despite the attention given to manufacturers' costs in relation to price in public debates and in the literature, this issue does not seem to get explicit systematic consideration in the reimbursement reports of expensive drugs.

Factors associated with care- and health-related quality of life of caregivers of children with juvenile idiopathic arthritis.

OBJECTIVE: This study investigates the relationship of child, caregiver, and caring context measurements with the care-related quality of life (CRQoL) and health-related quality of life (HRQoL) of caregivers of children with juvenile idiopathic arthritis (JIA). METHODS: We performed a cross-sectional analysis of baseline data on caregivers of children with JIA from Canada and the Netherlands collected for the "Canada-Netherlands Personalized Medicine Network in Childhood Arthritis and Rheumatic Diseases" study from June 2019 to September 2021. We used the CRQoL questionnaire (CarerQoL), adult EQ-5D-5L, and proxy-reported Youth 5-Level version of EuroQoL (EQ-5D-5L-Y) to assess caregiver CRQoL, caregiver HRQoL, and child HRQoL, respectively. We used a multivariate analysis to assess the relationship between both caregiver CRQoL and HRQoL and patient, caregiver, and caring context measurements. RESULTS: A total of 250 caregivers were included in this study. Most of the caregivers were from the Netherlands (n = 178, 71%) and 77% were females (n = 193). The mean CarerQoL scores was 82.7 (standard deviation (SD) 11.4) and the mean EQ-5D-5L utility score was 0.87 (SD 0.16). Child HRQoL and employment had a positive relationship with both caregiver CarerQoL and EQ-5D-5L utility scores (p < 0.05), while receiving paid or unpaid help had a negative relationship with both scores (p < 0.05). CONCLUSION: Our findings indicated that to understand the impact of JIA on families, we need to consider socio-economic factors, such as employment and support to carry caregiving tasks, in addition to child HRQoL.

Machine Learning Methods in Health Economics and Outcomes Research-The PALISADE Checklist: A Good Practices Report of an ISPOR Task Force.

Advances in machine learning (ML) and artificial intelligence offer tremendous potential benefits to patients. Predictive analytics using ML are already widely used in healthcare operations and care delivery, but how can ML be used for health economics and outcomes research (HEOR)? To answer this question, ISPOR established an emerging good practices task force for the application of ML in HEOR. The task force identified 5 methodological areas where ML could enhance HEOR: (1) cohort selection, identifying samples with greater specificity with respect to inclusion criteria; (2) identification of independent predictors and covariates of health outcomes; (3) predictive analytics of health outcomes, including those that are high cost or life threatening; (4) causal inference through methods, such as targeted maximum likelihood estimation or double-debiased estimation-helping to produce reliable evidence more quickly; and (5) application of ML to the development of economic models to reduce structural, parameter, and sampling uncertainty in cost-effectiveness analysis. Overall, ML facilitates HEOR through the meaningful and efficient analysis of big data. Nevertheless, a lack of transparency on how ML methods deliver solutions to feature selection and predictive analytics, especially in unsupervised circumstances, increases risk to providers and other decision makers in using ML results. To examine whether ML offers a useful and transparent solution to healthcare analytics, the task force developed the PALISADE Checklist. It is a guide for balancing the many potential applications of ML with the need for transparency in methods development and findings.

Are patients with cancer at higher risk of COVID-19-related death? A systematic review and critical appraisal of the early evidence.

BACKGROUND: Early reports suggested that COVID-19 patients with cancer were at higher risk of COVID-19-related death. We conducted a systematic review with risk of bias assessment and synthesis of the early evidence on the risk of COVID-19-related death for COVID-19 patients with and without cancer. METHODS AND FINDINGS: We searched Medline/Embase/BioRxiv/MedRxiv/SSRN databases to 1 July 2020. We included cohort or case-control studies published in English that reported on the risk of dying after developing COVID-19 for people with a pre-existing diagnosis of any cancer, lung cancer, or haematological cancers. We assessed risk of bias using tools adapted from the Newcastle-Ottawa Scale. We used the generic inverse-variance random-effects method for meta-analysis. Pooled odds ratios (ORs) and hazard ratios (HRs) were calculated separately. Of 96 included studies, 54 had sufficient non-overlapping data to be included in meta-analyses (>500,000 people with COVID-19, >8000 with cancer; 52 studies of any cancer, three of lung and six of haematological cancers). All studies had high risk of bias. Accounting for at least age consistently led to lower estimated ORs and HRs for COVID-19-related death in cancer patients (e.g. any cancer versus no cancer; six studies, unadjusted OR=3.30,95%CI:2.59-4.20, adjusted OR=1.37,95%CI:1.16-1.61). Adjusted effect estimates were not reported for people with lung or haematological cancers. Of 18 studies that adjusted for at least age, 17 reported positive associations between pre-existing cancer diagnosis and COVID-19-related death (e.g. any cancer versus no cancer; nine studies, adjusted OR=1.66,95%CI:1.33-2.08; five studies, adjusted HR=1.19,95%CI:1.02-1.38). CONCLUSIONS: The initial evidence (published to 1 July 2020) on COVID-19-related death in people with cancer is characterised by multiple sources of bias and substantial overlap between data included in different studies. Pooled analyses of non-overlapping early data with adjustment for at least age indicated a significantly increased risk of COVID-19-related death for those with a pre-existing cancer diagnosis.

The risk of contracting SARS-CoV-2 or developing COVID-19 for people with cancer: A systematic review of the early evidence.

BACKGROUND: The early COVID-19 literature suggested that people with cancer may be more likely to be infected with SARS-CoV-2 or develop COVID-19 than people without cancer, due to increased health services contact and/or immunocompromise. While some studies were criticised due to small patient numbers and methodological limitations, they created or reinforced concerns of clinicians and people with cancer. These risks are also important in COVID-19 vaccine prioritisation decisions. We performed a systematic review to critically assess and summarise the early literature. METHODS AND FINDINGS: We conducted a systematic search of Medline/Embase/BioRxiv/MedRxiv/SSRN databases including peer-reviewed journal articles, letters/commentaries, and non-peer-reviewed pre-print articles for 1 January-1 July 2020. The primary endpoints were diagnosis of COVID-19 and positive SARS-CoV-2 test. We assessed risk of bias using a tool adapted from the Newcastle-Ottawa Scale. Twelve studies were included in the quantitative synthesis. All four studies of COVID-19 incidence (including 24,181,727 individuals, 125,649 with pre-existing cancer) reported that people with cancer had higher COVID-19 incidence rates. Eight studies reported SARS-CoV-2 test positivity for > 472,000 individuals, 48,370 with pre-existing cancer. Seven of these studies comparing people with any and without cancer, were pooled using random effects [pooled odds ratio 0.91, 95 %CI: 0.57-1.47; unadjusted for age, sex, or comorbidities]. Two studies suggested people with active or haematological cancer had lower risk of a positive test. All 12 studies had high risk of bias; none included universal or random COVID-19/SARS-CoV-2 testing. CONCLUSIONS: The early literature on susceptibility to SARS-CoV-2/COVID-19 for people with cancer is characterised by pervasive biases and limited data. To provide high-quality evidence to inform decision-making, studies of risk of SARS-CoV-2/COVID-19 for people with cancer should control for other potential modifiers of infection risk, including age, sex, comorbidities, exposure to the virus, protective measures taken, and vaccination, in addition to stratifying analyses by cancer type, stage at diagnosis, and treatment received.

Real-world data reveals the complexity of disease modifying anti-rheumatic drug treatment patterns in juvenile idiopathic arthritis: an observational study.

OBJECTIVE: Pharmacological treatment is a cornerstone of care for children with juvenile idiopathic arthritis (JIA). The objective of this study is to evaluate prescription patterns of conventional and biologic disease modifying anti-rheumatic drugs (c-DMARDs and b-DMARDs) for patients with JIA. METHODS: We conducted a retrospective cohort study of children diagnosed with JIA at a rheumatology pediatric clinic. Eligibility criteria were defined as children and youth newly diagnosed with enthesis-related arthritis, polyarticular, or oligoarticular JIA between 2011 and 2019, with at least one year of observation. Data on c-DMARDs and b-DMARDs prescriptions were obtained from electronic medical charts. We used descriptive statistics, Kaplan-Meier survival methods, and Sankey diagrams to describe treatment prescription patterns. RESULTS: A total of 325 patients with JIA were included, with a median observation time of 3.7 years. The most frequently prescribed c-DMARD and b-DMARD were methotrexate and etanercept, respectively. Within the first year of rheumatology care, 62% and 21% of patients had a c-DMARD and a b-DMARD prescribed, respectively. These proportions varied greatly by JIA subtype. Among the 147 (147/325, 45%) patients that had at least one b-DMARD prescribed, 24% were prescribed a second, and 7% a third-line of b-DMARD. A total of 112 unique treatment sequences were observed, with c-DMARD monotherapy followed by the addition of either a b-DMARD (56%) or another c-DMARD (30%) being the two most prevalent patterns in this cohort. CONCLUSION: We observed a variety of treatment trajectories, with many patients experiencing multiple treatment lines, illustrating the complexity of the overall JIA treatment path.

Disparities in quality of life, social distress and employment outcomes in Australian cancer survivors.

PURPOSE: To examine how socio-demographic, comorbidities and information needs influence quality of life (QoL) outcomes of survivors of breast, colorectal, or prostate cancer, non-Hodgkin lymphoma or melanoma. METHODS: Cross-sectional postal survey with eligible participants identified through a population-based cancer registry. QoL outcomes were assessed by EQ-5D-5L, social difficulties index (SDI) and, for those employed at diagnosis, current employment. Regression analyses explored associations between outcome variables and cancer type, age, time since diagnosis, residential location, socio-economic disadvantage, comorbidities and unmet information needs. Mediation analyses examined whether comorbidities and information needs explained relationships between outcome variables and socio-economic disadvantage. RESULTS: 2115 survivors participated. Mean EQ-5D-5L scores (mean = 0.84) were similar to population averages and SDI scores were low for the entire sample (mean = 3.80). In multivariate analyses, being aged over 80, greater socio-economic disadvantage, comorbidities and unmet information needs decreased EQ-5D-5L scores. Higher SDI scores were associated with socio-economic disadvantage, comorbidities and unmet information needs. Not being employed was associated with being aged over 50, more comorbidities and socio-economic disadvantage. Comorbidities but not information needs partially mediated the impact of socio-economic disadvantage on EQ-5D-5L and SDI accounting for 17% and 14% of the total effect of socio-economic disadvantage respectively. Neither comorbidities nor information needs mediated the association between socio-economic disadvantage and employment outcomes. CONCLUSIONS: To improve quality of life, survivorship care should be better tailored to address the needs of individuals given their overall health and impact of comorbidities, their age and type of cancer and not simply time since diagnosis.