Value in Health最新文献

Objectives: The rate of sickness absence in Norway is at its highest point since 2009 and policy makers need tools to make informed decisions on high-value interventions to address sick leave. Using trial-linked registry data, multi-state modelling, and decision-analytic modelling, we assessed the cost-effectiveness of two return-to-work (RTW) interventions for individuals with musculoskeletal and psychological disorders in Norway.

Methods: Using data from 166 individuals in a randomized trial, we developed a decision-analytic model to compare two multi-domain RTW interventions: O-ACT (outpatient acceptance and commitment therapy) and I-MORE (inpatient multimodal occupational rehabilitation). The probabilistic model was informed using trial-based input parameters, including transition probabilities, healthcare costs, production loss, and health-related quality of life to project long-term costs and quality-adjusted life years (QALY) over a 25-year time horizon for each intervention.

Results: Modelled outcomes were consistent with the trial outcomes, showing that I-MORE led participants to RTW more quickly. However, assuming a healthcare perspective and a cost-effectiveness threshold of $50,000 per QALY, I-MORE was not considered cost-effective in 98% of our simulations (probabilistic ICER: $356,447 per QALY gained) compared to O-ACT. In contrast, when accounting for production loss, I-MORE not only became cost-effective but was projected to be more beneficial and less costly compared to O-ACT.

Conclusion: Under current Norwegian benchmarks for cost-effectiveness, I-MORE would not be considered cost-effective unless production loss was included. Our findings emphasize the key role of a broader societal perspective in economic evaluations, which although is being considered, is currently not recommended in Norwegian guidelines.

Objectives: Increasingly, discrete choice experiments (DCEs) are conducted online, with little consideration of the digitally-excluded, who are unable to participate. Policymakers or others considering online research data need clarity about how views might differ across this 'digital divide'. We took tasks from an existing online DCE designed to elicit social preferences for health and wellbeing outcomes. We aimed to explore: i) How telephone interview participants answered a series of choice tasks taken from an online DCE; and ii) Whether and how decision-making for these tasks differed between digitally-excluded and non-excluded participants.

Methods: Semi-structured telephone interviews with members of the public (n=27), recruited via an existing social research panel. Data were analysed thematically to identify key approaches to decision making.

Results: Twelve participants were classed as 'digitally-excluded', and 15 as 'digitally non-excluded'. Responses were similar between the two samples for the majority of choice tasks. We identified three approaches used to reach decisions: (1) simplifying; (2) creating explanatory narratives; and (3) personalising. Whilst these approaches were common across both samples, understanding the exercise appeared more challenging for the digitally- excluded sample.

Conclusions: This novel study provides some assurance that the participants' views over the choice tasks used are similar across the digital divide. The challenges we identify with understanding highlight the need to be careful examining the views held by the digitally-excluded. If online data are to inform policy-making, it is essential to explore the views of those who cannot participate in online DCEs.

Objectives: Low-value care refers to medical services whose benefits do not outweigh the costs and potential harm. This study estimates the prevalence, distribution, and associated costs of 24 low-value care services within the German public healthcare system.

Methods: Large-scale retrospective observational study using statutory health insurance data provided by the Techniker Krankenkasse (TK) spanning from 2018 to 2021 covering approximately 11.1 million insured individuals annually. The prevalence of 24 low-value service indicators, which were identified through a systematic review and expert consultations was calculated. To address uncertainties in distinguishing between appropriate and low-value care, both broad (potential overestimation) and narrow definitions (potential underestimation) were applied to all suitable indicators, providing a range within which the true extent of low-value care is expected to lie.

Results: Between 2019 and 2021, 1.6 million patients were identified as having received at least one low-value service, using the 24 indicators. Of all 10.6 million delivered services (cases) examined, on average per year, 1.1 million cases (broad definition) and 0.43 million cases (narrow definition) were classified as low-value, corresponding to 10.4% and 4.0%, respectively. Costs incurred by the identified services were approximately 15.5 million euros (broad definition) and 9.9 million euros (narrow definition) annually.

Conclusions: Despite the limitations of German statutory health insurance data, considerable low-value care was found within several of the 24 low-value-indicators. The findings highlight the necessity for targeted interventions to mitigate low-value care in Germany, guiding healthcare policy and practice to enhance quality and safety effectively.

Objectives: The direct medical costs associated with Alzheimer's disease (AD) in the United States have been estimated to be over $360 billion, but this value does not reflect the substantial financial burden on unpaid caregivers and society. We estimated the economic burden of unpaid caregivers and patient productivity loss due to AD across all disease severity stages to better understand the indirect impacts of AD.

Methods: We performed a narrative literature review to identify estimates of unpaid caregiver burden and market productivity loss. Additionally, we leveraged a published algorithm to estimate non-market productivity loss due to AD. Patient-level estimates were scaled to the population based on AD prevalence in the United States (approximately 12.5 million), weighted by disease severity.

Results: The total annual indirect costs of unpaid caregiving and of market and non-market productivity loss of AD increased with severity: $36,934 for mild cognitive impairment due to AD, $65,565 for Mild AD, $103,717 for Moderate AD, and $145,250 for Severe AD (2024 USD). Considering the current distribution of prevalent patients across severity stages, the total annual indirect cost was estimated at $832 billion, which includes $599 billion in unpaid caregiving costs and $233 billion in productivity losses.

Conclusions: Conventional cost estimates, which do not consider unpaid caregiver burden and patient productivity loss, significantly underestimate the total AD burden. Both elements should be incorporated into cost estimates and value assessments to best capture the total indirect impact of AD and the value of new therapies.

Objective: To provide an introduction to the uses of generative Artificial Intelligence (AI) and foundation models, including large language models (LLMs), in the field of health technology assessment (HTA).

Methods: We reviewed applications of generative AI in three areas: systematic literature reviews, real world evidence (RWE) and health economic modeling.

Results: (1) Literature reviews: generative AI has the potential to assist in automating aspects of systematic literature reviews by proposing search terms, screening abstracts, extracting data and generating code for meta-analyses; (2) Real World Evidence (RWE): generative AI can facilitate automating processes and analyze large collections of real-world data (RWD) including unstructured clinical notes and imaging; (3) Health economic modeling: generative AI can aid in the development of health economic models, from conceptualization to validation. Limitations in the use of foundation models and LLMs include challenges surrounding their scientific rigor and reliability, the potential for bias, implications for equity, as well as nontrivial concerns regarding adherence to regulatory and ethical standards, particularly in terms of data privacy and security. Additionally, we survey the current policy landscape and provide suggestions for HTA agencies on responsibly integrating generative AI into their workflows, emphasizing the importance of human oversight and the fast-evolving nature of these tools.

Conclusions: While generative AI technology holds promise with respect to HTA applications, it is still undergoing rapid developments and improvements. Continued careful evaluation of their applications to HTA is required. Both developers and users of research incorporating these tools, should familiarize themselves with their current capabilities and limitations.

Objective: There is uncertainty around whether, and under what circumstances, there is value in embedding economic considerations into multi-arm, multi-stage (MAMS), adaptive, and adaptive-platform trial designs. This systematic review was conducted to assess the analytical methods and factors that are considered when incorporating health economic analyses when designing and modifying MAMS adaptive, and adaptive-platform trials.

Methods: The review searched for health economic analyses, including planned analyses, of interventions assessed through MAMS adaptive, and adaptive-platform trials. The search for articles was conducted in EMBASE, MEDLINE, Web of Science, Scopus, and ClinicalTrials.gov electronic databases from their inception to 7^th August 2023. The screening for articles was conducted by two blinded reviewers who followed a predetermined screening process. A narrative synthesis was conducted on the methods used in the analysis and how the results informed the trial designs and modifications.

Results: The review included 17 articles of which four were results of economic evaluations while 13 were economic evaluation protocols. No trial reported using pre-trial economic evaluations to inform the trial designs. In 14 articles it was possible to estimate the costs and benefits of the interventions at the interim analysis stages. However, there were only five interim cost-effectiveness analyses and three of these informed decisions to drop or maintain trial arms.

Conclusions: Health economics is being embedded in some MAMS adaptive, and platform-adaptive trials to inform trial modifications. However, the use of economic evidence is limited, both by design and circumstance, despite its potential important to adoption decisions.

Objective: To understand the role of alternative prices and financing mechanisms on a payer's budget impact and the manufacturers' risks and returns for gene therapies.

Methods: This paper uses fundamental economic principles to interpret the implications of alternative pricing mechanisms in terms of the manufacturer's appropriation share of the value and how alternate financing mechanisms alter it. It demonstrates these concepts by studying the financial impacts for a payer and the manufacturer across alternative pricing and financing mechanisms that could be used by the US Centers for Medicare and Medicaid Services (CMS) to pay for gene therapy for sickle cell disease (SCD).

Results: Unlike value-based and manufacturer-set monopoly prices, an effective monopoly price can be derived to guarantee monopoly profits for manufacturers during their exclusivity period, thereby providing a high appropriation share and substantially lowering price and budget impact for a payer. For SCD gene therapy, the 10-year budget impact for CMS would range from $8.6 Billion-$12.8 Billion under a value-based price, $10.2 Billion-$15.2 Billion under a monopoly price, but reduce to $7.7 Billion under an effective monopoly price. The latter price would still fetch over 50% of the total surplus to the manufacturer while mitigating their risk of sales volume.

Conclusion: I show that significant budget impacts for funding gene therapy are not mitigated across alternative financing mechanisms at any given price. The price determines most of the budget impact. The option of a patent buyout may help negotiate down prices to effective monopoly prices.

Objective: Current real-world health care cost information is needed to project future expenditures and inform policy. We estimated adults' 2019 health care costs in the United States (US) by age, sex, race/ethnicity, geographic region, and comorbidity.

Methods: We aggregated and summarized health care costs in 2021 US dollars using claims data derived from Optum's de-identified Clinformatics® Data Mart database, which includes inpatient, outpatient, and prescription claims for commercial and Medicare Advantage beneficiaries nationwide.

Results: A total of 9,227,901 adults were included in the analysis. The largest group represented was 71-75 years old (13%), female (53%), White (68%), received care in the South (41%), and had commercial health insurance (56%). There was a positive relationship between health care cost and age. Females had a 1.3-fold multiplicative increase in costs compared to males (95% CI 1.33-1.34). There were 92.5% of individuals who had health claims in the Northeast, 89.6% in the Midwest, 88.9% in the South, 77.1% in the West, and 12.7% with unknown geographic region. Patients with severe renal failure, heart failure, or metastatic cancer incurred the highest mean yearly costs ($139,844, $113,031, and $85,299, respectively). Metastatic cancer and severe renal failure were associated with a 5.3-fold multiplicative increase in costs compared with not having these conditions, after adjusting for potential confounders (95% CI, 5.26-5.41 and 4.98-5.16, respectively).

Conclusions: We identified patient characteristics and medical conditions that are associated with high health care cost burden and could benefit from tailored interventions. We provided detailed cost estimates to aid health care modeling, cost projection, and cost-minimizing interventions.