Systematic Review of Health Economic Evaluations Focused on Artificial Intelligence in Healthcare: The Tortoise and the Cheetah

ObjectivesThis study aimed to systematically review recent health economic evaluations (HEEs) of artificial intelligence (AI) applications in healthcare. The aim was to discuss pertinent methods, reporting quality and challenges for future implementation of AI in healthcare, and additionally advise future HEEs.MethodsA systematic literature review was conducted in 2 databases (PubMed and Scopus) for articles published in the last 5 years. Two reviewers performed independent screening, full-text inclusion, data extraction, and appraisal. The Consolidated Health Economic Evaluation Reporting Standards and Philips checklist were used for the quality assessment of included studies.ResultsA total of 884 unique studies were identified; 20 were included for full-text review, covering a wide range of medical specialties and care pathway phases. The most commonly evaluated type of AI was automated medical image analysis models (n = 9, 45%). The prevailing health economic analysis was cost minimization (n = 8, 40%) with the costs saved per case as preferred outcome measure. A total of 9 studies (45%) reported model-based HEEs, 4 of which applied a time horizon >1 year. The evidence supporting the chosen analytical methods, assessment of uncertainty, and model structures was underreported. The reporting quality of the articles was moderate as on average studies reported on 66% of Consolidated Health Economic Evaluation Reporting Standards items.ConclusionsHEEs of AI in healthcare are limited and often focus on costs rather than health impact. Surprisingly, model-based long-term evaluations are just as uncommon as model-based short-term evaluations. Consequently, insight into the actual benefits offered by AI is lagging behind current technological developments.     

Challenges for Economic Evaluations of Advanced Therapy Medicinal Products: A Systematic Review

ObjectivesAdvanced therapy medicinal products (ATMPs) are drugs for human use for the treatment of chronic, degenerative, or life-threatening diseases that are based on genes, tissues, or cells. This article aimed to identify and critically review published economic analyses of ATMPs.MethodsA systematic review of economic analyses of ATMPs was undertaken. Study characteristics, design, sources of data, resources and unit costs, modeling and extrapolation methods, study results, and sensitivity analyses were assessed.ResultsA total of 46 economic analyses of ATMP (from 45 articles) were included; 4 were cell therapy medicinal products, 33 gene therapy medicinal products, and 9 tissue-engineered products. 30 therapies had commercial marketing approval; 39 studies were cost-utility analysis, 5 were cost-effectiveness analysis, and 2 were cost only studies. Four studies predicted that the ATMP offered a step change in the management of the condition and 10 studies estimated that the ATMP would offer a lower mean cost.ConclusionsComparison with historical controls, pooling of data, and use of techniques such as mixture cure fraction models should be used cautiously. Sensitivity analyses should be used across a plausible range of prices. Clinical studies need to be designed to align with health technology assessment requirements, including generic quality of life, and payers should aim for clarity of criteria. Regulators and national payers should aim for compatibility of registers to allow interchange of data. Given the increasing reliance on industry-funded economic analyses, careful critical review is recommended.     

Augmenting a Transitional Care Model With Artificial Intelligence Decreased Readmissions

ObjectivesEvaluate if augmenting a transitions of care delivery model with insights from artificial intelligence (AI) that applied clinical and exogenous social determinants of health data would reduce rehospitalization in older adults.DesignRetrospective case-control study.Setting and ParticipantsAdult patients discharged from integrated health system between November 1, 2019, and February 31, 2020, and enrolled in a rehospitalization reduction transitional care management program.InterventionAn AI algorithm utilizing multiple data sources including clinical, socioeconomic, and behavioral data was developed to predict patients at highest risk for readmitting within 30 days and provide care navigators five care recommendations to prevent rehospitalization.MethodsAdjusted incidence of rehospitalization was estimated with Poisson regression and compared between transitional care management enrollees that used AI insights and matched enrollees for whom AI insights were not used.ResultsAnalyses included 6371 hospital encounters between November 2019 and February 2020 across 12 hospitals. Of the encounters 29.3% were identified by AI as being medium-high risk for re-hospitalizing within 30 days, for which AI provided transitional care recommendations to the transitional care management team. The navigation team completed 40.2% of AI recommendations for these high-risk older adults. These patients had overall 21.0% less adjusted incidence of 30-day rehospitalization compared with matched control encounters, or 69 fewer rehospitalizations per 1000 encounters (95% CI 0.65‒0.95).Conclusions and ImplicationsCoordinating a patient's care continuum is critical for safe and effective transition of care. This study found that augmenting an existing transition of care navigation program with patient insights from AI reduced rehospitalization more than without AI insights. Augmenting transitional care with insights from AI could be a cost-effective intervention to improve transitional care outcomes and reduce unnecessary rehospitalization. Future studies should examine cost-effectiveness of augmenting transitional care models of care with AI when hospitals and post-acute providers partner with AI companies.     

Using Cost-Effectiveness Analysis to Address Health Equity Concerns

This articles serves as a guide to using cost-effectiveness analysis (CEA) to address health equity concerns. We first introduce the "equity impact plane," a tool for considering trade-offs between improving total health—the objective underpinning conventional CEA—and equity objectives, such as reducing social inequality in health or prioritizing the severely ill. Improving total health may clash with reducing social inequality in health, for example, when effective delivery of services to disadvantaged communities requires additional costs. Who gains and who loses from a cost-increasing health program depends on differences among people in terms of health risks, uptake, quality, adherence, capacity to benefit, and—crucially—who bears the opportunity costs of diverting scarce resources from other uses. We describe two main ways of using CEA to address health equity concerns: 1) equity impact analysis, which quantifies the distribution of costs and effects by equity-relevant variables, such as socioeconomic status, location, ethnicity, sex, and severity of illness; and 2) equity trade-off analysis, which quantifies trade-offs between improving total health and other equity objectives. One way to analyze equity trade-offs is to count the cost of fairer but less cost-effective options in terms of health forgone. Another method is to explore how much concern for equity is required to choose fairer but less cost-effective options using equity weights or parameters. We hope this article will help the health technology assessment community navigate the practical options now available for conducting equity-informative CEA that gives policymakers a better understanding of equity impacts and trade-offs.     

人工智能医疗器械企业质量管理现状调研

目的 了解人工智能生产企业质量管理现状,为人工智能医疗器械(Artificial Intelligence Medical Device,AIMD)产品质量管理研究与标准化提供参考依据.方法 基于YY/T 0287-2017《医疗器械质量管理体系用于法规的要求》、《医疗器械生产质量管理规范独立软件附录》和《Xavier...     

Artificial Intelligence in Nutrients Science Research: A Review

Artificial intelligence (AI) as a branch of computer science, the purpose of which is to imitate thought processes, learning abilities and knowledge management, finds more and more applications in experimental and clinical medicine. In recent decades, there has been an expansion of AI applications in biomedical sciences. The possibilities of artificial intelligence in the field of medical diagnostics, risk prediction and support of therapeutic techniques are growing rapidly. The aim of the article is to analyze the current use of AI in nutrients science research. The literature review was conducted in PubMed. A total of 399 records published between 1987 and 2020 were obtained, of which, after analyzing the titles and abstracts, 261 were rejected. In the next stages, the remaining records were analyzed using the full-text versions and, finally, 55 papers were selected. These papers were divided into three areas: AI in biomedical nutrients research (20 studies), AI in clinical nutrients research (22 studies) and AI in nutritional epidemiology (13 studies). It was found that the artificial neural network (ANN) methodology was dominant in the group of research on food composition study and production of nutrients. However, machine learning (ML) algorithms were widely used in studies on the influence of nutrients on the functioning of the human body in health and disease and in studies on the gut microbiota. Deep learning (DL) algorithms prevailed in a group of research works on clinical nutrients intake. The development of dietary systems using AI technology may lead to the creation of a global network that will be able to both actively support and monitor the personalized supply of nutrients.     

Challenges in Research and Health Technology Assessment of Rare Disease Technologies: Report of the ISPOR Rare Disease Special Interest Group

Background

Successful development of new treatments for rare diseases (RDs) and their sustainable patient access require overcoming a series of challenges related to research and health technology assessment (HTA). These impediments, which may be unique to RDs or also apply to common diseases but are particularly pertinent in RDs, are diverse and interrelated.

Value of Research and Value of Development in Early Assessments of New Medical Technologies

ObjectivesIn early stages of development of new medical technologies, there are conceptually separate but related societal decisions to be made concerning adoption, further development (i.e., technical improvement), and research (i.e., clinical trials) of new technologies. This article presents a framework to simultaneously support these three decisions from a societal perspective. The framework is applied to the 70-gene signature, a gene-expression profile for breast cancer, deciding which patients should receive adjuvant systemic therapy after surgery. The “original” signature performed on fresh frozen tissue (70G-FFT) could be further developed to a paraffin-based signature (70G-PAR) to reduce test failures.MethodsA Markov decision model comparing the “current” guideline Adjuvant Online (AO), 70G-FFT, and 70G-PAR was used to simulate 20-year costs and outcomes in a hypothetical cohort in The Netherlands. The 70G-PAR strategy was based on projected data from a comparable technology. Incremental net monetary benefits were calculated to support the adoption decision. Expected net benefit of development for the population and expected net benefit of sampling were calculated to support the development and research decision.ResultsThe 70G-PAR had the highest net monetary benefit, followed by the 70G-FFT. The population expected net benefit of development amounted to €91 million over 20 years (assuming €250 development costs per patient receiving the test). The expected net benefit of sampling amounted to €61 million for the optimal trial (n = 4000).ConclusionsWe presented a framework to simultaneously support adoption, development, and research decisions in early stages of medical technology development. In this case, the results indicate that there is value in both further development of 70G-FFT into 70G-PAR and further research.     

Artificial Intelligence and Liability in Medicine: Balancing Safety and Innovation

Policy Points

• With increasing integration of artificial intelligence and machine learning in medicine, there are concerns that algorithm inaccuracy could lead to patient injury and medical liability.
• While prior work has focused on medical malpractice, the artificial intelligence ecosystem consists of multiple stakeholders beyond clinicians. Current liability frameworks are inadequate to encourage both safe clinical implementation and disruptive innovation of artificial intelligence.
• Several policy options could ensure a more balanced liability system, including altering the standard of care, insurance, indemnification, special/no‐fault adjudication systems, and regulation. Such liability frameworks could facilitate safe and expedient implementation of artificial intelligence and machine learning in clinical care.

     

医疗人工智能的伦理问题及对策研究

医疗人工智能的极速发展在不断创新医疗卫生业务的同时带来了诸多伦理问题与挑战。对医疗人工智能可能带来的7个方面伦理问题进行探讨,包括道德主体性位置、医生的主体性位置、责任认定、病人自主性、保障隐私权、医疗负担、算法监管等。并提出重构伦理与监管体系;保护数据与公众安全;提升医疗质量与医生地位等建议与对策。     

A Systematic Literature Review of Gaps and Challenges in Value Assessment of Biosimilars: An ISPOR Special Interest Group Report

ObjectivesThis study aims to provide an overview of the gaps and challenges in the value assessment of biosimilars and to identify potential approaches to address them.MethodsA multidisciplinary, international team of biosimilar experts identified gaps and challenges. A systematic review was conducted of the peer-reviewed literature in PubMed, EMBASE, Web of Science Core Collection, EBSCOhost Business Source Complete; and of the gray literature. Preliminary results were presented at ISPOR conferences and this article benefited from 2 review rounds among ISPOR Biosimilar Special Interest Group members.ResultsGiven that a biosimilar is highly similar to its reference biologic, health technology assessment agencies should accept the comparability exercise approved by regulatory authorities and, thus, conduct a price comparison when biosimilar reimbursement is requested for the same indication as the reference biologic. If the reference biologic is not reimbursed or is not the standard of care, a full economic evaluation of the biosimilar versus a relevant comparator needs to be conducted. To date, little consideration has been given to specific challenges, such as how biosimilar value assessment can account for the nocebo effect, potential differences between biologic-naive and biologic-experienced patients, the availability of intravenous and subcutaneous administration forms or different administration devices for the same active compound, value-added services, and the contribution of biosimilars for generating health gain at the population level.ConclusionsThere is a need to gather further insights in the methodology of value assessment for biosimilars, and health technology assessment agencies need to develop more elaborate guidance on biosimilar value assessment in specific circumstances.     

Three Sets of Case Studies Suggest Logic and Consistency Challenges with Value Frameworks

Objective

To assess the logic and consistency of three prominent value frameworks.

Evaluation of a Novel Artificial Intelligence System to Monitor and Assess Energy and Macronutrient Intake in Hospitalised Older Patients

Malnutrition is common, especially among older, hospitalised patients, and is associated with higher mortality, longer hospitalisation stays, infections, and loss of muscle mass. It is therefore of utmost importance to employ a proper method for dietary assessment that can be used for the identification and management of malnourished hospitalised patients. In this study, we propose an automated Artificial Intelligence (AI)-based system that receives input images of the meals before and after their consumption and is able to estimate the patient’s energy, carbohydrate, protein, fat, and fatty acids intake. The system jointly segments the images into the different food components and plate types, estimates the volume of each component before and after consumption, and calculates the energy and macronutrient intake for every meal, based on the kitchen’s menu database. Data acquired from an acute geriatric hospital as well as from our previous study were used for the fine-tuning and evaluation of the system. The results from both our system and the hospital’s standard procedure were compared to the estimations of experts. Agreement was better with the system, suggesting that it has the potential to replace standard clinical procedures with a positive impact on time spent directly with the patients.     

Economic Evaluation of Systemic Treatments for Advanced Melanoma: A Systematic Review

BackgroundMany high cost treatments for advanced melanoma have become available in recent years. National health technology assessment agencies have raised concerns regarding uncertainty in their clinical and cost-effectiveness.ObjectiveThe aim of this systematic review is to identify economic evaluations of treatments for advanced melanoma and review model assumptions, outcomes, and quality as preparation for a health technology assessment.MethodsA search of Embase, MEDLINE, EconLit, and the Cochrane Database was conducted. Only studies using decision-analytic models were included. Two authors independently completed full-text review and data extraction.ResultsFifteen studies were identified. There were major differences in the structural assumptions underpinning the models. There was general agreement in study conclusions, although the predicted costs and quality-adjusted life years for each treatment varied. BRAF monotherapy (vemurafenib, dabrafenib) or BRAF/MEK combination therapy (BRAF monotherapy with cobimetinib or trametinib) has not been shown to be cost-effective in any jurisdiction. PD-1 inhibitors (pembrolizumab, nivolumab) are consistently found to be cost-effective compared with ipilimumab, although their cost-effectiveness compared with chemotherapy is not established. Combination therapy with nivolumab and ipilimumab is unlikely to be cost-effective in any setting. One study including all agents found that none of the new treatments were cost-effective relative to chemotherapy. Publication of the study in a health economics journal is associated with better reporting of and higher-quality assessment than those published in clinical journals.ConclusionDespite differences in model structures and assumptions, the conclusions of most included studies were consistent. Health technology assessment has a key role in maximizing value from high-cost innovative treatments. Consideration should be given to divestment from BRAF/MEK inhibitors and ipilimumab in favor of reimbursement of PD-1 inhibitors.     

Advancing Value Assessment in the United States: A Multistakeholder Perspective

Rising costs without perceived proportional improvements in quality and outcomes have motivated fundamental shifts in health care delivery and payment to achieve better value. Aligned with these efforts, several value assessment frameworks have been introduced recently to help providers, patients, and payers better understand the potential value of drugs and other interventions and make informed decisions about their use. Given their early stage of development, it is imperative to evaluate these efforts on an ongoing basis to identify how best to support and improve them moving forward. This article provides a multistakeholder perspective on the key limitations and opportunities posed by the current value assessment frameworks and areas of and actions for improvement. In particular, we outline 10 fundamental guiding principles and associated strategies that should be considered in subsequent iterations of the existing frameworks or by emerging initiatives in the future. Although value assessment frameworks may not be able to meet all the needs and preferences of stakeholders, we contend that there are common elements and potential next steps that can be supported to advance value assessment in the United States.     

Reported Challenges in Health Technology Assessment of Complex Health Technologies

ObjectivesWith complex health technologies entering the market, methods for health technology assessment (HTA) may require changes. This study aimed to identify challenges in HTA of complex health technologies.MethodsA survey was sent to European HTA organizations participating in European Network for HTA (EUnetHTA). The survey contained open questions and used predefined potentially complex health technologies and 7 case studies to identify types of complex health technologies and challenges faced during HTA. The survey was validated, tested for reliability by an expert panel, and pilot tested before dissemination.ResultsA total of 22 HTA organizations completed the survey (67%). Advanced therapeutic medicinal products (ATMPs) and histology-independent therapies were considered most challenging based on the predefined complex health technologies and case studies. For the case studies, more than half of the reported challenges were “methodological,” equal in relative effectiveness assessments as in cost-effectiveness assessments. Through the open questions, we found that most of these challenges actually rooted in data unavailability. Data were reported as “absent,” “insufficient,” “immature,” or “low quality” by 18 of 20 organizations (90%), in particular data on quality of life. Policy and organizational challenges and challenges because of societal or political pressure were reported by 8 (40%) and 4 organizations (20%), respectively. Modeling issues were reported least often (n = 2, 4%).ConclusionsMost challenges in HTA of complex health technologies root in data insufficiencies rather than in the complexity of health technologies itself. As the number of complex technologies grows, the urgency for new methods and policies to guide HTA decision making increases.     

Distributional Cost-Effectiveness Analysis of Health Technologies: Data Requirements and Challenges

Governments and health technology assessment agencies are putting greater focus on and efforts in understanding and addressing health inequities. Cost-effectiveness analyses are used to evaluate the costs and health gains of different interventions to inform the decision-making process on funding of new treatments. Distributional cost-effectiveness analysis (DCEA) is an extension of cost-effectiveness analysis that quantifies the equity impact of funding new treatments. Key challenges for the routine and consistent implementation of DCEA are the lack of clearly defined equity concerns from decision makers and endorsed measures to define equity subgroups and the availability of evidence that allows analysis of differences in data inputs associated with the equity characteristics of interest. In this article, we detail the data gaps and challenges to build robust DCEA analysis routinely in health technology assessment and suggest actions to overcome these hurdles.     

人工智能在超声医学影像的应用研究进展

近年来,伴随着云计算机的发展,以及互联网+的不断深化,人工智能已逐渐出现在生活的各个领域,而深度学习技术的出现和大数据的不断发展也助推着人工智能向人工智能+医疗的方向快速发展,并取得了突破性进展.超声影像检查作为重要的医疗细分领域,也搭上了人工智能这趟列车.文章梳理了人工智能在超声医学影像的应用现状及研究进展,...