European Heart Journal - Digital Health · 2023 · 19 citations · 15 references
MedicineHealth InformaticsPharmacotherapyClinical Decision SupportGeneral Internal MedicinePrecision MedicineBlood Pressure ControlBlood Pressure ManagementClinical Trial Data CollectionClinical TrialsHypertension ManagementRandomized Controlled TrialsActive ControlBlood Pressure MonitoringBlood Pressure Self-managementHome Blood Pressure MonitoringHypertension Self-managementPrimary HypertensionTreatment Protocol EvaluationTrial Feasibility Assessment
Aims: Artificial intelligence-driven small data platforms such as CURATE.AI hold potential for personalized hypertension care by assisting physicians in identifying personalized anti-hypertensive doses for titration. This trial aims to assess the feasibility of a larger randomized controlled trial (RCT), evaluating the efficacy of CURATE.AI-assisted dose titration intervention. We will also collect preliminary efficacy and safety data and explore stakeholder feedback in the early design process. Methods and results: In this open-label, randomized, pilot feasibility trial, we aim to recruit 45 participants with primary hypertension. Participants will be randomized in 1:1:1 ratio into control (no intervention), home blood pressure monitoring (active control; HBPM), or CURATE.AI arms (intervention; HBPM and CURATE.AI-assisted dose titration). The home treatments include 1 month of two-drug anti-hypertensive regimens. Primary endpoints assess the logistical (e.g. dose adherence) and scientific (e.g. percentage of participants for which CURATE.AI profiles can be generated) feasibility, and define the progression criteria for the RCT in a 'traffic light system'. Secondary endpoints assess preliminary efficacy [e.g. mean change in office blood pressures (BPs)] and safety (e.g. hospitalization events) associated with each treatment protocol. Participants with both baseline and post-treatment BP measurements will form the intent-to-treat analysis. Following their involvement with the CURATE.AI intervention, feedback from CURATE.AI participants and healthcare providers will be collected via exit survey and interviews. Conclusion: Findings from this study will inform about potential refinements of the current treatment protocols before proceeding with a larger RCT, or potential expansion to collect additional information. Positive results may suggest the potential efficacy of CURATE.AI to improve BP control. Trial registration number: NCT05376683.
15
Mohammad H. Forouzanfar, Ashkan Afshin, Lily Alexander et al. · The Lancet · 2016 · 7.8K citations · Full text
Bin Zhou, Rodrigo M. Carrillo‐Larco, Goodarz Danaei et al. · The Lancet · 2021 · 3.7K citations · Full text
Medicine, Chronic Disease Control, Clinical Epidemiology +16
CONSORT 2010 statement: extension to randomised pilot and feasibility trials
Sandra Eldridge, Claire Chan, Michael J. Campbell et al. · BMJ · 2016 · 3.5K citations · Full text
Why we need a small data paradigm
Eric B. Hekler, Predrag Klasnja, Guillaume Chevance et al. · BMC Medicine · 2019 · 295 citations · Full text