Pharmacy-based Opportunistic Atrial Fibrillation Screening Integrated With Emergency Services (NUMAPLUS-FA)

Overview

This study evaluates the feasibility and fidelity of an opportunistic atrial fibrillation (AF) screening programme conducted in 38 community pharmacies in Seville (Spain), using the Kardias® 6-lead portable ECG device, coordinated with the emergency medical services centre Salud Responde (CES 061) and primary care physicians. Eligible participants are individuals aged 65 years or older without a prior AF diagnosis who attend a participating pharmacy for any routine purpose. The screening consists of a 30-second ECG recording performed by a trained community pharmacist. Positive or indeterminate results are transmitted in real time to Salud Responde for clinical validation and referral to primary care. The study is evaluated using the RE-AIM framework (Reach, Effectiveness, Adoption, Implementation, Maintenance). The primary outcome is protocol fidelity, measured by the Implementation Adherence Grade (IAG). The study is part of the NUMAPLUS project, funded by INTERREG VI-A España-Portugal (POCTEP 2021-2027), co-financed by the European Regional Development Fund (ERDF).

STUDY DESIGN This is a prospective, single-cohort, type-1 effectiveness-implementation hybrid study conducted within the NUMAPLUS project (INTERREG VI-A España-Portugal, POCTEP 2021-2027, co-financed by the European Regional Development Fund). This design was selected because external evidence on the diagnostic accuracy of the Kardias® device is sufficient to justify its deployment, while uncertainty persists regarding the organisational conditions, care pathways and economic incentives required for its sustainable integration into the Andalusian community pharmacy network. The primary focus is on implementation outcomes; clinical outcomes related to atrial fibrillation screening are collected as secondary endpoints. The study is reported in accordance with the Standards for Reporting Implementation Studies (StaRI) guideline. SETTING The study is conducted in 38 community pharmacies in the province of Seville affiliated with the Real e Ilustre Colegio Oficial de Farmacéuticos de Sevilla (RICOFS), selected by voluntary participation following an open invitation. Eligibility criteria for pharmacies required stable internet connectivity for access to the Numaplus application, completion of pre-study training, and absence of concurrent participation in another cardiovascular screening programme. Priority was given to pharmacies with 24-hour or 12-hour opening schedules. Geographic coverage targets at least three urban health zones per province and at least one semi-urban or rural zone per healthcare management area. The study articulates three coordination levels: (1) the community pharmacy as the point of patient recruitment and screening; (2) Salud Responde (CES 061) as the clinical validation, risk stratification and real-time coordination node; and (3) the primary care physician as the responsible party for definitive diagnosis and initiation of treatment. TRAINING OF PARTICIPATING PHARMACISTS Prior to the recruitment period, all enrolled pharmacists completed a structured training programme accredited by the Escuela Andaluza de Salud Pública (EASP) and the Agencia de Calidad Sanitaria de Andalucía (ACSA). The programme was delivered entirely online in asynchronous modality between 23 March and 13 April 2026 (10 hours total) through the EASP Moodle platform. It comprised four learning units: (1) epidemiological basis and scientific rationale for opportunistic AF screening; (2) detection protocol, radial pulse palpation technique, and interpretation of Kardias® device outputs; (3) data entry in the Numaplus platform, risk classification and patient communication within the pharmacy-Salud Responde circuit; and (4) ethical and legal framework, informed consent procedures, and data protection requirements. Certification required completion of all practical activities and a minimum score of 7 out of 10 on the final knowledge assessment. INTERVENTION The screening protocol comprises five sequential operational phases aligned with the RE-AIM framework: Phase 1 (Reach): Opportunistic invitation during dispensing or consultation, without prior appointment. The pharmacist records the monthly denominator of unique patients aged 65 years or older using pharmacy management software. Reasons for refusal are recorded anonymously. Phase 2 (Adoption): Following written informed consent, the case is opened in the Numaplus application. The pharmacist records current symptoms (palpitations, chest pain, dyspnoea, syncope), oral anticoagulant therapy status, and the CHA₂DS₂-VA score (0-6), adopted per the 2024 ESC Guidelines for AF Management in replacement of CHA₂DS₂-VASc. Phase 3 (Implementation): The Kardias® 6-lead portable device is applied for a 30-second ECG recording (maximum 3 attempts). The device algorithm classifies the result as: Suspected AF / Sinus rhythm / Bradycardia / Tachycardia / Unclassified. Protocol fidelity is operationalised through the Implementation Adherence Grade (IAG). A performance-based financial incentive of 20 euros per complete and valid case is applied. Phase 4 (Effectiveness): In case of suspected or indeterminate AF, the Numaplus platform automatically transmits a PDF containing the ECG tracing and clinical data to Salud Responde; simultaneously, the pharmacist places a direct call. The Salud Responde nurse applies a structured triage protocol: emergency 061 activation (heart rate \>110 or \<60 bpm or alarm symptoms), medium priority referral (haemodynamically stable suspected AF, primary care appointment within 72 hours), or low priority referral (unclassified tracing, scheduled ECG at primary care). Definitive AF diagnosis is confirmed by the primary care physician. Phase 5 (Maintenance): Six-month telephone follow-up conducted by Salud Responde, assessing anticoagulation initiation and adherence, health-related quality of life (visual analogue scale 0-10), and major adverse events (ischaemic stroke, major bleeding, death). STATISTICAL ANALYSIS The primary analysis is precision-based, targeting 385 participants (operational target: 380, 10 per pharmacy). Categorical variables are reported as proportions with 95% confidence intervals (Clopper-Pearson); the IAG is estimated with an exact binomial 95% CI. Factors associated with low fidelity are explored by logistic regression. Missing data are handled by multiple imputation (MICE, 20 imputations) if below 15% under the MAR assumption. All analyses are conducted in R version 4.3 or higher.