Cost-effectiveness of a machine learning risk prediction model (LungFlag) in the selection of high-risk individuals for non-small cell lung cancer screening in Spain
Year: 2024
Type: article
Abstract: OBJECTIVE: The LungFlag risk prediction model uses individualized clinical variables to identify individuals at high-risk of non-small cell lung cancer (NSCLC) for screening with low-dose computed tomography (LDCT). This study evaluates the cost-effectiveness of LungFlag implementation in the Spanish setting for the identification of individuals at high-risk of NSCLC. METHODS: A model combining a decision-tree with a Markov model was adapted to the Spanish setting to calculate health outcomes and costs over a lifetime horizon, comparing two hypothetical scenarios: screening with LungFlag versus non-screening, and screening with LungFlag versus screening the entire population meeting 2013 US Preventive Services Task Force (USPSTF) criteria. Model inputs were obtained from the literature and the clinical practice of a multidisciplinary expert panel. Only direct costs (€of 2023), obtained from local sources, were considered. Deterministic and probabilistic sensitivity analyses were performed to assess the robustness of our results. RESULTS: A cohort of 3,835,128 individuals meeting 2013 USPSTF criteria would require 2,147,672 LDCTs scans. However, using LungFlag would only require 232,120 LDCTs scans. Cost-effectiveness results showed that LungFlag was dominant versus non-screening scenario, and outperformed the scenario where the entire population were screened since the observed loss of effectiveness (-224,031 life years [LYs] and -97,612 quality-adjusted life years [QALYs]) was largely offset by the significant cost savings provided (€7,053 million). The resulting incremental cost-effectiveness ratio (ICER) for this strategy of screening the whole population versus using LungFlag was €72,000/QALY, showing that LungFlag is cost-effective. Various were described, such as the source of the efficacy or adherence rates, and other limitations inherent to cost-effectiveness analyses. CONCLUSIONS: Using LungFlag for the selection of high-risk individuals for NSCLC screening in Spain would be a cost-effective strategy over screening the entire population meeting USPSTF 2013 criteria and is dominant over non-screening.
Source: Journal of Medical Economics
Authors: Juan Carlos Trujillo, Joan B. Soriano, Mercè Marzo, Oliver Higuera, Luis Gorospe, Virginia Pajares, María Eugenia Olmedo, Natalia Arrabal, Andrés Flores, José Francisco García, María Crespo, David Carcedo, Carolina Heuser, Milan M. S. Obradović, Nicolò Olghi, Eran N. Choman, Luis M. Seijo
Institutions: Hospital de Sant Pau, Hospital Del Mar, Centro de Investigación Biomédica en Red de Enfermedades Respiratorias, Hospital Universitario de La Princesa, Universidad Autónoma de Madrid, Institut Català d'Oncologia, Hospital Universitario La Paz, Hospital Universitario Ramón y Cajal, Roche (Spain), Health Economics and Outcomes Research (United Kingdom), Roche (Switzerland), Clinica Universidad de Navarra
Language: English
Subfield: Pulmonary and Respiratory Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4405592686