Background and aims: This study investigated how patients experience and which outcomes matter to patients and healthcare professionals in the decision to initiate proprotein convertase subtilisin/kexin type 9 inhibitors (PCSK9i) as add-on lipid-lowering treatment (LLT). Methods: We performed a mixed methods study: very high-risk patients qualifying for PCSK9i reimbursement were interviewed about their experiences and preferences. Subsequently, patients using PCSK9i completed an anonymous online survey about their experiences. Additionally, healthcare professionals (HCPs) filled in an online survey about their PCSK9i prescription preferences and perceived patient preferences. Results: We interviewed 25 patients (median [IQR] age 58 [48-65] years, 56 % women, 64 % established cardiovascular disease) at different decision-making stages. The majority (72 %) chose efficacy over side-effects (16 %) and ease of use (12 %) as most important attribute of add-on LLT. Most patients (72 %) prefer shared decision-making. Subsequently, 170 patients using PCSK9i completed a survey (age 64 [56-69], 44 % women, 63 % established cardiovascular disease). Here again, the most important attribute (83 %) in deciding on add-on LLT was efficacy. Almost all (90 %) patients favoured shared decision-making. Of the 59 HCPs (age 44 [40-50], 49 % women, 78 % medical specialist), only 27 % indicated to consider patient preferences when selecting the PCSK9i type. HCPs identified patient characteristics influencing their PCSK9i prescription preferences. Conclusions: For patients and HCPs, efficacy was the most important aspect in choosing a PCSK9i. Even though shared decision-making is recommended by the guidelines and preferred by patients, in clinical practice only a minority of the HCPs apply this. To facilitate shared decision-making, future research should investigate the development and impact of a decision aid for patients.

Objective To provide a comprehensive framework for value assessment of artificial intelligence (AI) in radiology. Methods This paper presents the RADAR framework, which has been adapted from Fryback and Thornbury’s imaging efficacy framework to facilitate the valuation of radiology AI from conception to local implementation. Local efficacy has been newly introduced to underscore the importance of appraising an AI technology within its local environment. Furthermore, the RADAR framework is illustrated through a myriad of study designs that help assess value. Results RADAR presents a seven-level hierarchy, providing radiologists, researchers, and policymakers with a structured approach to the comprehensive assessment of value in radiology AI. RADAR is designed to be dynamic and meet the different valuation needs throughout the AI’s lifecycle. Initial phases like technical and diagnostic efficacy (RADAR-1 and RADAR-2) are assessed pre-clinical deployment via in silico clinical trials and cross-sectional studies. Subsequent stages, spanning from diagnostic thinking to patient outcome efficacy (RADAR-3 to RADAR-5), require clinical integration and are explored via randomized controlled trials and cohort studies. Cost-effectiveness efficacy (RADAR-6) takes a societal perspective on financial feasibility, addressed via health-economic evaluations. The final level, RADAR-7, determines how prior valuations translate locally, evaluated through budget impact analysis, multi-criteria decision analyses, and prospective monitoring. Conclusion The RADAR framework offers a comprehensive framework for valuing radiology AI. Its layered, hierarchical structure, combined with a focus on local relevance, aligns RADAR seamlessly with the principles of value-based radiology. Critical relevance statement The RADAR framework advances artificial intelligence in radiology by delineating a much-needed framework for comprehensive valuation. Keypoints • Radiology artificial intelligence lacks a comprehensive approach to value assessment. • The RADAR framework provides a dynamic, hierarchical method for thorough valuation of radiology AI. • RADAR advances clinical radiology by bridging the artificial intelligence implementation gap.

Since there is no diet that is perfect for everyone, personalized nutrition approaches are gaining popularity to achieve goals such as the prevention of obesity-related diseases. However, appropriate choices about funding and encouraging personalized nutrition approaches should be based on sufficient evidence of their effectiveness and cost-effectiveness. In this study, we assessed whether a newly developed personalized plan (PP) could be cost-effective relative to a non-personalized plan in Denmark. Results of a 10-week randomized controlled trial were combined with a validated obesity economic model to estimate lifetime cost-effectiveness. In the trial, the intervention group (PP) received personalized home-delivered meals based on metabolic biomarkers and personalized behavioral change messages. In the control group these meals and messages were not personalized. Effects were measured in body mass index (BMI) and quality of life (EQ-5D-5L). Costs [euros (€), 2020] were considered from a societal perspective. Lifetime cost-effectiveness was assessed using a multi-state Markov model. Univariate, probabilistic sensitivity, and scenario analyses were performed. In the trial, no significant differences were found in the effectiveness of PP compared with control, but wide confidence intervals (CIs) were seen [e.g., BMI (−0.07, 95

ObjectivesAs current health technology assessment (HTA) frameworks do not provide specific guidance on the assessment of medical artificial intelligence (AI), this study aimed to propose a conceptual framework for a broad HTA of medical AI.MethodsA systematic literature review and a targeted search of policy documents was conducted to distill the relevant medical AI assessment elements. Three exemplary cases were selected to illustrate various elements: (1) An application supporting radiologists in stroke-care (2) A natural language processing application for clinical data abstraction (3) An ICU-discharge decision-making application.ResultsA total of 31 policy documents and 9 academic publications were selected, from which a list of 29 issues was distilled. The issues were grouped by four focus areas: (1) Technology & Performance, (2) Human & Organizational, (3) Legal & Ethical and (4) Transparency & Usability. Each assessment element was extensively discussed in the test, and the elements clinical effectiveness, clinical workflow, workforce, interoperability, fairness and explainability were further highlighted through the exemplary cases.ConclusionThe current methodology of HTA requires extension to make it suitable for a broad evaluation of medical AI technologies. The 29-item assessment list that we propose needs a tailored approach for distinct types of medical AI, since the conceptualisation of the issues differs across applications.

Objective Early detection of Parkinson's Disease (PD) progression remains a challenge. As remote patient monitoring solutions (RMS) and artificial intelligence (AI) technologies emerge as potential aids for PD management, there's a gap in understanding how end users view these technologies. This research explores patient and neurologist perspectives on AI-assisted RMS. Methods Qualitative interviews and focus-groups were conducted with 27 persons with PD (PwPD) and six neurologists from Finland and Italy. The discussions covered traditional disease progression detection and the prospects of integrating AI and RMS. Sessions were recorded, transcribed, and underwent thematic analysis. Results The study involved five individual interviews (four Italian participants and one Finnish) and six focus-groups (four Finnish and two Italian) with PwPD. Additionally, six neurologists (three from each country) were interviewed. Both cohorts voiced frustration with current monitoring methods due to their limited real-time detection capabilities. However, there was enthusiasm for AI-assisted RMS, contingent upon its value addition, user-friendliness, and preservation of the doctor-patient bond. While some PwPD had privacy and trust concerns, the anticipated advantages in symptom regulation seemed to outweigh these apprehensions. Discussion The study reveals a willingness among PwPD and neurologists to integrate RMS and AI into PD management. Widespread adoption requires these technologies to provide tangible clinical benefits, remain user-friendly, and uphold trust within the physician-patient relationship. Conclusion This study offers insights into the potential drivers and barriers for adopting AI-assisted RMS in PD care. Recognizing these factors is pivotal for the successful integration of these digital health tools in PD management.

OBJECTIVES:Poor nutrition links to chronic diseases, emphasizing the need for optimized diets. The EU-funded project PREVENTOMICS, introduced personalized nutrition to address this. This study aims to perform a health technology assessment (HTA) comparing personalized nutrition interventions developed through this project, with non-personalized nutrition interventions (control) for people with normal weight, overweight, or obesity. The goal is to support decisions about further development and implementation of personalized nutrition. METHODS:The PREVENTOMICS interventions were evaluated using the European Network for HTA Core Model, which includes a methodological framework that encompasses different domains for value assessment. Information was gathered via [1] different statistical analyses and modeling studies, [2] questions asked of project partners and, [3] other (un)published materials. RESULTS:Clinical trials of PREVENTOMICS interventions demonstrated different body mass index changes compared to control; differences ranged from -0.80 to 0.20 kg/m2. Long-term outcome predictions showed generally improved health outcomes for the interventions; some appeared cost-effective (e.g., interventions in UK). Ethical concerns around health inequality and the lack of specific legal regulations for personalized nutrition interventions were identified. Choice modeling studies indicated openness to personalized nutrition interventions; decisions were primarily affected by intervention's price. CONCLUSIONS:PREVENTOMICS clinical trials have shown promising effectiveness with no major safety concerns, although uncertainties about effectiveness exist due to small samples (n=60-264) and short follow-ups (10-16 weeks). Larger, longer trials are needed for robust evidence before implementation could be considered. Among other considerations, developers should explore financing options and collaborate with policymakers to prevent exclusion of specific groups due to information shortages.

Background and Aims: High risk patients who do not reach LDL-C target currently have three options: 1) addition of PCSK9 monoclonal antibody (mAb), 2) addition of a PCSK9 siRNA, or 3) no additional medication. Shared decision making is considered important in contemporary patient care. The first step is to gain insight in the preferences of patients and healthcare providers, which is the aim of our study.

To assess the methodological quality of cost-effectiveness analyses (CEA) of nivolumab in combination with ipilimumab, we conducted a systematic literature review in the first-line treatment of patients with recurrent or metastatic non-small cell lung cancer (NSCLC), whose tumors express programmed death ligand-1, with no epidermal growth factor receptor or anaplastic lymphoma kinase genomic tumor aberrations. PubMed, Embase, and the Cost-Effectiveness Analysis Registry were searched, in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The methodological quality of the included studies was assessed by the Philips checklist and the Consensus Health Economic Criteria (CHEC) checklist. 171 records were identified. Seven studies met the inclusion criteria. Cost-effectiveness analyses differed substantially due to the applied modeling methods, sources of costs, health state utilities, and key assumptions. Quality assessment of the included studies highlighted shortcomings in data identification, uncertainty assessment, and methods transparency. Our systematic review and methodology assessment revealed that the methods of estimation of long-term outcomes, quantification of health state utility values, estimation of drug costs, the accuracy of data sources, and their credibility have important implications on the cost-effectiveness outcomes. None of the included studies fulfilled all of the criteria reported in the Philips and the CHEC checklists. To compound the economic consequences presented in these limited number of CEAs, ipilimumab's drug action as a combination therapy poses significant uncertainty. We encourage further research to address the economic consequences of these combination agents in future CEAs and the clinical uncertainties of ipilimumab for NSCLC in future trials.

Immunotherapy offers a distinctive mechanism of action compared to traditional treatments, arising from additional value dimensions that may not be captured in standard health technology assessments. Cancer patients may have the expectation that immunotherapy provides durable, long-term survival gains. Moreover, some patients may be willing to take a 'risk' to undergo immunotherapy to achieve better survival outcomes. We reviewed quantitative methods that explored patients' risk preferences in their non-small cell lung cancer (NSCLC) treatment choices, in PubMed (MEDLINE), from January 1, 2015, until July 1, 2022. The consideration of a value dimension ('hope') based on patients' risk-seeking preferences is specifically addressed for the valuation of immune checkpoint inhibitors in NSCLC. We reported that the quantitative methods that aim to measure patients' risk preferences or 'hope' empirically are emerging. Value assessments should not only comprise survival improvements for the mean or median patient but also consider methods that reflect durable, long-term overall survival gains for risk-seeking patients. However, the published evidence for incorporating 'hope' based on patients' stated preferences for uncertain treatment profiles is not strong, and future research could strengthen this evidence base. We encourage further research on the development and validation of quantification methods to incorporate 'hope' and risk preferences of patients treated with immunotherapy for NSCLC and beyond.

Abstract Background Complex regional pain syndrome (CRPS) is a chronic pain condition of an extremity. While achieving pain relief in CRPS is challenging, esketamine infusions can accomplish pain relief for several weeks post-infusion in a subgroup of CRPS patients. Unfortunately, CRPS esketamine protocols are very heterogeneous in advice on dosage, administration and treatment setting. Currently, no trials are available that study differences between intermittent and continuous esketamine infusions for CRPS. With the current situation of bed shortages, it is difficult to admit patients for several consecutive days for inpatient esketamine treatments. In this study, we investigate whether 6 intermittent outpatient esketamine treatments are not inferior to a continuous 6-day inpatient esketamine treatment in establishing pain relief. In addition, several secondary study parameters will be assessed in order to investigate mechanisms responsible for pain relief by esketamine infusions. Furthermore, the cost-effectiveness will be analyzed. Methods In this RCT, the primary objective is to demonstrate that an intermittent esketamine dosing regimen is non-inferior to a continuous esketamine dosing regimen at 3 months follow-up. We will include 60 adult CRPS patients. The inpatient treatment group receives a continuous intravenous esketamine infusion for 6 consecutive days. The outpatient treatment group receives a 6-hour intravenous esketamine infusion every 2 weeks for 3 months. Esketamine dose will be individually tailored and is started at 0.05 mg/kg/h and can be increased to a maximum of 0.2 mg/kg/h. Each patient will be followed for 6 months. The primary study parameter is perceived pain intensity, measured by an 11-point Numerical Rating Scale. Secondary study parameters are conditioned pain modulation, quantitative sensory testing, adverse events, thermography, blood inflammatory parameter, questionnaires about functionality, quality of life and mood and costs per patient. Discussion If our study reveals non-inferiority between intermittent and continuous esketamine infusions, these findings can be beneficial to increase the availability and flexibility of esketamine infusions through outpatient treatments. Furthermore, the costs of outpatient esketamine infusions could be lower than inpatient esketamine infusions. In addition, secondary parameters may predict response to esketamine treatment. Trial registration ClinicalTrials.gov Identifier NCT05212571, date of registration 01-28-2022. Protocol version: Version 3, February 2022.

Regulatory authorization of oncology drugs, including immune-checkpoint inhibitors, is often based on enhanced efficacy and acceptable toxicity profiles, investigated in randomized, open-label clinical trials. Regulatory approval decisions of the United States (US) Food and Drug Administration (FDA) and the European Medicines Agency (EMA) are frequently compared and contrasted, specifically based on review requirements, and time to approval or refusal decisions. We reviewed databases of the US FDA, the EMA and Clinicaltrials.gov, from January 1, 2015 until December 31, 2021, and analyzed regulatory approvals for immune-checkpoint inhibitors in the treatment of non-small cell lung cancer (NSCLC). We specifically focused on time to approval duration of each immune-checkpoint inhibitor, and considerations of patient-reported outcomes (PROs) by each regulatory agency. Despite similarities in the regulatory pathways and methods used for immune-checkpoint inhibitor approvals, NSCLC indications that stood out in terms of outcome divergence were mainly first-line drugs for treatment naïve patients. The US FDA was quicker to reach approval decisions, when compared with the EMA. The US FDA and the EMA both recognize the value of PROs as important patient-centered endpoints. Policy statement: There are several regulatory structures in the US and Europe that aim to leverage the latest clinical trial evidence and speed up the regulatory approval processes. In our study, the preponderance of outcome differences in approvals were not influenced by the expedited drug development and access programs. Increased harmonization and collaboration on the PRO measurement and validation are encouraged among these agencies to improve the efficiency of regulatory decisions in the future.

Background: The need to assess the effectiveness and value of interventions involving digital health and health and welfare technologies is becoming increasingly important due to the rapidly growing development of these technologies and their areas of application. Systematic reviews of scientific literature are a mainstay of such assessment, but publications outside the realm of traditional scientific bibliographic databases-known as gray literature-are often not included. This is a disadvantage, particularly apparent in the health and welfare technology (HWT) domain. Objective: The aim of this article is to investigate the significance of gray literature in digital health and HWT when reviewing literature. As an example, the impact of including gray literature to the result of two systematic reviews in HWT is examined. Methods: In this paper, we identify, discuss, and suggest methods for including gray literature sources when evaluating effectiveness and appropriateness for different review types related to HWT. The analysis also includes established sources, search strategies, documentation, and reporting of searches, as well as bias and credibility assessment. The differences in comparison to scientific bibliographic databases are elucidated. We describe the results, challenges, and benefits of including gray literature in 2 examples of systematic reviews of HWT. Results: In the 2 systematic reviews described in this paper, most included studies came from context-specific gray literature sources. Gray literature contributed to the overall result of the reviews and corresponded well with the reviews' aims. The assessed risk of bias of the included studies derived from gray literature was similar to the included studies from other types of sources. However, because of less standardized publication formats, assessing and extracting data from gray literature studies were more time-consuming and compiling statistical results was not possible. The search process for gray literature required more time and the reproducibility of gray literature searches were less certain due to more unstable publication platforms. Conclusions: Gray literature is particularly relevant for digital health and HWT but searches need to be conducted systematically and reported transparently. This way gray literature can broaden the range of studies, highlight context specificity, and decrease the publication bias of reviews of effectiveness of HWT. Thus, researchers conducting systematic reviews related to HWT should consider including gray literature based on a systematic approach.

INTRODUCTION:Remote patient monitoring (RPM) of heart failure patients has the potential to reduce healthcare resource use and costs, but current evidence has been inconclusive. This study aims assess the impact of RPM of heart failure patients with an implantable cardioverter defibrillator on medical resource use, direct medical costs, quality-adjusted life years (QALYs), and travel time of patients, and to estimate its commercial headroom in the Netherlands and Germany. METHODS:Data from the REMOTE-CIED randomized controlled trial were used to calculate differences in length of hospital stay, outpatient clinic visits, telephone consults, emergency room visits, and travel time between patients on in-clinic follow-up and RPM in the Netherlands, Germany, and France. Incremental cardiac-related healthcare costs and QALYs were calculated and used to calculate the commercial headroom of RPM in the Netherlands and Germany. The impact of imputation, parameter, and case-mix uncertainty on these outcomes was explored using probabilistic analysis. RESULTS:Length of hospitalization, number of unscheduled admissions, and number of outpatient visits were lower in the remote monitoring group in all three countries. Number of hospital admissions was higher, and number of calls was lower in the Netherlands and Germany but not in France. Costs were lower in both the Netherlands (-€1041, 95% confidence interval (CI): -€3308, €1005) and Germany (-€2865, 95% CI: -€7619, €1105), while incremental effectiveness differed: -0.003 (95% CI: -0.114, 0.107) QALY in the Netherlands and +0.086 (95% CI: -0.083, 0.256) in Germany. Commercial headroom was estimated at €881 (95% CI: -€5430, €7208) in the Netherlands and €5005 (95% CI: -€1339, €11,960) in Germany. DISCUSSION:RPM was found to result in reduced medical resource use and travel time. Whether it is cost saving or cost effective strongly depends on the costs of remote monitoring. TRIAL REGISTRATION NUMBER AND TRIAL REGISTER:ClinicalTrials.gov: NCT01691586.