Rethinking Secondary Cardiovascular Prevention in the Digital Era: Patient Empowerment, Adherence and Optimisation of Guideline-based Care

Abstract

Cardiovascular disease is the leading cause of morbidity and mortality globally. Despite effective treatments, poor adherence limits their long-term benefits. Digital health solutions can enhance clinicians’ ability to optimise guideline-based therapies, improving patient outcomes. Digital tools for remote consultations, monitoring, cardiac device interrogation and clinical decision support systems are now widely available. Digital health monitoring improves care quality, providing value to patients, healthcare professionals, hospitals and governments. This transformation is fostering a future where ‘health’ takes precedence over ‘reactive care’, driven by empowered patients and advanced technologies. Success in this transformation requires not only technology, but also new strategic operational models, optimised workflows and workforce redesign. Health systems that form partnerships with other stakeholders, such as peers, payers, start-ups, life sciences organisations, industries, will be better positioned to improve patient experiences and outcomes. This review examines digital solutions that optimise medical therapy prescriptions, promote patient engagement and address therapeutic inertia in AF, heart failure and coronary artery disease.

Cardiovascular disease (CVD) remains the leading cause of morbidity and mortality worldwide. Despite the availability of effective guideline-based treatment strategies recommended by the European Society of Cardiology and the American College of Cardiology (ACC)/American Heart Association, long-term clinical benefits remain suboptimal, mainly because of poor adherence, therapeutic inertia, fragmentation of care and insufficient achievement of recommended treatment targets.1,2 Indeed, the gap between evidence and real-world implementation remains one of the major unresolved problems in secondary cardiovascular prevention.

Digital health refers to the use of digital, mobile and wireless technologies to support the achievement of health objectives, whereas mobile health (mHealth) is a specific branch focused on mobile devices, such as smartphones and tablets. Due to their widespread availability, these technologies can facilitate symptom monitoring, communication, education, behaviour change support and treatment follow-up through text messages, phone calls, apps, wearable sensors and remote data transmission. In the digital era, these tools may theoretically help close the evidence–practice gap by enabling more continuous patient engagement, more timely treatment adjustment, more structured follow-up and greater personalisation of care.

Graphical Abstract: Management of Cardiovascular Disease: Non-pharmacological Cardiovascular Prevention Strategies in the Digital Era.

The COVID-19 pandemic markedly accelerated the adoption of digital health solutions in cardiovascular care. During that period, digital tools proved useful in preserving continuity of care despite restricted in-person access, and they demonstrated the potential of remote consultation, remote monitoring, digital therapeutic support and clinical decision support systems. Moreover, digital health strategies have demonstrated their cost-effectiveness by streamlining care delivery and optimising healthcare resources.3

Although the uptake of digital solutions has expanded considerably, therapeutic inertia and suboptimal adherence still persist. This apparent paradox indicates that technology alone is not sufficient; rather, digital tools need to be meaningfully integrated into clinical workflows, patient pathways and healthcare systems in order to translate potential into measurable clinical benefit.

The aim of this review is to highlight the role of digital solutions in secondary cardiovascular prevention by promoting adequate patient engagement, improving therapeutic adherence and helping overcome therapeutic inertia in order to facilitate the implementation of optimal guideline-based care. We also explore the use of digital medicine in three specific and clinically relevant contexts: heart failure (HF), AF, and coronary artery disease (CAD), in which digital strategies are increasingly being incorporated into routine practice, although with heterogeneous levels of evidence and implementation maturity.

Interventions to Promote Healthy Behaviour Change and Patient Empowerment

Endorsement of Healthy Behaviours

Health outcomes are significantly influenced by personal lifestyle choices. While strategies that promote behavioural change and self-care can improve healthcare outcomes, starting and maintaining healthy habits can be challenging. Support and empowerment are essential.

Empowerment enables individuals to make self-care decisions while acknowledging their shared responsibility with healthcare professionals. Patients are accountable for their choices and their impact on healthcare.4

Two aspects need to be considered when discussing the endorsement of healthy behaviours: the healthcare provider’s ability to engage in patient-centred communication; and the needs of the patient.

Research shows that effective communication between healthcare providers and patients enhances patient empowerment and leads to better health outcomes.5

A patient-centred communication approach emphasises the importance of the individual’s autonomy in their motivation for behaviour change.

This approach empowers individuals to adopt and sustain healthy behaviours by aligning them with their personal values, rather than by relying on external pressure. Healthcare providers can foster this autonomy by helping patients identify and pursue habits that resonate with their values. Active listening to patients’ needs and understanding their knowledge of the condition and treatments are essential.6 Additionally, tailored interventions that consider cultural, economic and social factors can effectively promote healthy lifestyles.7

Educational interventions, such as improving patient satisfaction and their treatment knowledge, positively impact outcomes by enhancing medication adherence and reducing complications.8,9

In this process, there are two main steps: goal setting and goal striving.6 Goals driven by intrinsic motivation are more likely to be achieved than those based on extrinsic motivation.10 When patients identify the gap between their current and desired states, they are more inclined to develop intrinsic goals.

Emphasising these differences boosts motivation and encourages change.11 Once motivation is identified, the next step is goal setting, which has been proven to effectively alter health behaviours and reduce hospitalisations.12 Goals should focus on a desired outcome, be realistic, require balanced effort (not too easy or too difficult) and foster new behaviours. After setting goals, the next step is to actively pursue and maintain them.

It is essential to support patients by creating an action plan detailing when, where and how to implement new behaviours while identifying potential obstacles and solutions. Patients should be encouraged to develop automatic habits, stay focused on their goals, monitor their progress and engage regularly with their healthcare provider.

Digital interventions for healthy lifestyle management have been defined as the application of advanced innovations to achieve wellness goals. Digital interventions are used in digital wellness programs and information and communication technology frameworks, including communication channels, and they involve instant messaging (text messages), alerts and app notifications.13,14 Digital interventions can motivate and challenge individuals through self-monitoring, goal setting, assessment and generating feedback or recommendations to promote a healthy lifestyle.13,14

An example of digital intervention implementation is the eCHANGE project, which addresses the need for innovative approaches to successful weight-loss maintenance. The overall goal of the eCHANGE project was to enhance knowledge for future weight-loss maintenance practices through the use of digital technology, promoting service- and research-led innovation across all sectors and service levels.15

Participants who tried to self-manage their healthy lifestyle found that the most difficult part was staying motivated; therefore, they needed apps that empowered and inspired them.16 However, despite the widespread use of mobile phones, barriers to digital literacy are common among vulnerable populations, and participants had varying levels of participation in various activities. Researchers using traditional user-centred design methods should routinely measure these communication domains in their end-user samples. Future research should replicate these findings in a larger sample through direct observation, given that persuasive cues may be more effective in providing feedback to those with communication difficulties.

Patient Empowerment

Prevention is essential to reduce the global burden of CVD. In recent years, telemedicine has become a popular solution for managing cardiovascular risk factors.17 Telemedicine is not intended to completely replace the relationship between physician and patient; rather, it is an integrative tool to support patients in managing their health, enabling them to be monitored and educated about risk factor control, particularly in relation to cardiac rehabilitation. For example, a series of systematic reviews and meta-analyses showed that telemedicine management of risk factors, lifestyle and prompting of medication in patients with HF can reduce mortality by 30–35% and hospitalisations by 15–20%.18,19 Telemedicine provides the possibility of constant communication between patients and their care networks, and when integrated with adequate training and patient education, it promotes active involvement and greater adherence of both patients and their support group to the therapeutic programme.

Telemedicine offers significant potential in the follow-up care of cardiopathic patients by enabling real-time therapy optimisation. It allows for continuous monitoring of key parameters, facilitating the early detection of any alterations that may indicate clinical changes. By promptly addressing these changes, telemedicine can prevent the progression of potential complications, reduce hospital admissions and ultimately improve both patient prognosis and quality of life. This proactive approach enhances the precision of care, ensuring timely interventions and minimising the risk of clinical deterioration.18,19

Telemedicine is a valuable tool in prevention, screening, treatment adherence and follow-up, resulting in economic advantages and overcoming logistical and structural barriers, thereby ensuring greater equity in patient care. Furthermore, simplified access to technologies capable of detecting functional parameters can be useful in formulating personalised physical activity programmes.20

Another important area of telemedicine application is cardiac rehabilitation. Examples include programmes such as TeleFIT and TelePEP. The TeleFIT module is a remotely supervised exercise programme and is dependent on the patient’s abilities, while the TelePEP module is a structured behavioural change programme aimed at improving an individual’s lifestyle to reduce cardiovascular risk.21,22 Clinically, both programmes can be used with the support of telemedicine. In the TeleFIT programme, a specialised physiotherapist provides exercises that the patient carries out at home and monitors the parameters. In the TelePEP module, communication and supervision occur through an online app, phone calls and video consultations. This modality may include homework assignments, questionnaires to evaluate progress, educational letters, apps, such as food diaries, or wearable devices, such as accelerometers or multi-parametric watches.

Modalities to Improve Adherence to Guideline-directed Therapy

Several factors may underlie poor adherence to CVD treatments, and multiple strategies should be considered to improve it.23 These strategies include enhancing patient education, implementing medication reminders, reducing therapy costs, simplifying medication dosing regimens and facilitating healthcare accessibility (Figure 1).23

Figure 1: Main Strategies to Improve Medication Adherence.

Given that psychosocial conditions impact adherence rate, these factors must be addressed in clinical practice. Poor health literacy, which is often associated with low education attainment and lower incomes, is another factor that can lead to lower adherence to prescribed treatments.23 Using language appropriate to the patient’s education level and providing simple advice and printed instructions may facilitate understanding of health information and, consequently, improve medication adherence.23 Interventions aimed at promoting patients’ awareness of their disease and empowering them in disease management should also be implemented to improve adherence.24

Drug cost affordability is one of the reasons for non-adherence, accounting for about 20% of non-adherence among US adults aged ≥65 years.25 Improving access to treatment reimbursement and reducing medication costs may help overcome this specific barrier to adherence.

Emerging tools that can facilitate some of the aforementioned strategies are based on mHealth technology. According to the WHO definition, mHealth consists of ‘medical and public health practice supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants (PDAs) and other wireless devices.’ Several studies have tested the use of mHealth technologies to improve treatment adherence in patients with CVD.

The mHealth tools tested in clinical studies comprise SMS reminders for medication, phone calls, mailed material and electronic medical record feedback. In clinical practice, the benefits, potential drawbacks and patient preference for each method should be considered before implementing these strategies on a case-by-case basis. For instance, smartphone-based approaches often appear to be the preferred method. Both medication reminder systems using tailored text messaging and pillbox organisers have been found to increase medication adherence.26 Interventions that combine medication reminders with health education messages have been found to be more beneficial compared with those using medication reminder messages alone.27 Several studies evaluating the impact of phone reminders on medication adherence among patients with dyslipidaemia and/or hypertension reported improved adherence.27

The use of telemedicine and remote consultation options is another application of technological advancement that may enhance healthcare system accessibility. Electronic prescriptions and home delivery of prescribed medications should also be considered among the innovative solutions to increase medication adherence.

Strategies to Overcome Therapeutic Inertia and Implement Best Practice

Evidence-based guidelines, developed through a critical evaluation of high-quality observational and randomised controlled trial (RCT) data and expert consensus, have emerged as a major strategic tool.28 Numerous recommendations of proven benefit for the prevention and management of CVD are available.29 These guidelines should reduce inappropriate practice and improve treatment efficiency, enabling both expert and non-expert practitioners to achieve better patient outcomes.30

However, during follow-up after discharge from a major cardiovascular event, there is a significant tendency towards non-adherence to guideline recommendations, which can be detrimental to outcomes.31 There are many examples of the existing gaps between guidelines and real-world practice:

• blood pressure levels are often higher than recommended;

• there is suboptimal uptitration and persistence of lipid-lowering treatment, especially regarding combination therapy; thus, LDL cholesterol goal achievement rates are remarkably low in secondary prevention settings;32

• a significant proportion of HF patients do not receive treatment with angiotensin-converting enzyme inhibitors, β-blockers and non-steroidal mineralocorticoid receptor antagonists;33

• the use of sodium–glucose cotransporter 2 inhibitors, glucagon-like peptide 1 receptor agonists and aldosterone antagonists, which have rapidly become cornerstones of kidney and cardiovascular risk-focused care, is low;34

• despite guidelines recommending treatment for depression in coronary heart disease patients, few receive it;35 and

• the rate of smoking cessation, and the prescription rate of antiplatelet agents and heart rate-lowering agents, is low.36

A large gap exists between the cardiovascular secondary prevention knowledge base and its implementation in routine care, highlighting the evidence–practice gap.28,37 This gap includes both under-use, in which proven strategies are not applied, and over-use, in which ineffective or unsupported strategies are used.29,38 Systematic monitoring of care quality and outcomes is essential to identify these gaps.38

The failure to implement guidelines is due to various intrinsic and extrinsic barriers in the healthcare system, including patient-, physician- and system-related factors.36–39 Clinician-related issues, such as inconsistent risk assessments, clinical inertia (failure to adjust therapy when goals are unmet) and limited access to care, contribute significantly to this gap.28,39 Additionally, conflicting guidelines, lack of awareness, resistance to change and perceptions that guidelines are unrealistic for real-world patients, along with time constraints, bureaucratic obstacles and high prescription costs, further hinder implementation.38 The lack of infrastructure for optimal healthcare delivery also represents a significant challenge.

System-level barriers may represent a limitation to the implementation of depression screening and treatment in coronary heart disease patients. These barriers include knowledge gaps, workflow integration issues and lack of ownership.35

Knowledge and practice gaps related to the diagnosis and treatment of the disease have been shown to be another limitation in the implementation of evidence-based recommendations. This has been demonstrated in the management of mitral regurgitation by the ACC Education Needs Assessment and Research group through a set of surveys distributed to primary care and cardiovascular physicians.38

The pervasiveness of this problem in the prevention of cardiovascular morbidity and mortality cannot be overstated.37 Thus, to be effective, practice guidelines must not only clearly delineate which therapies are efficacious, but also address the understanding of and navigation through various barriers necessary for their practical implementation.28

To overcome these gaps, strategies to improve adherence to guidelines in both community and clinical settings should be developed. These strategies should include research to better define optimal treatment in different populations, increased funding for research and guideline dissemination and implementation programmes, focused physician- and patient-facing educational interventions, and systems-based process innovations with health information technology strategies.38

While individual interventions may not be sufficient, comprehensive multifaceted approaches that address barriers at the provider, patient and healthcare delivery levels are likely to offer the greatest chance of success.

Reuter et al. applied behavioural and implementation science methods to develop a multi-level implementation strategy for promoting guideline adoption, including problem-solving meetings with clinic staff, educational/motivational videos, electronic health record reminders/decisional support for providers and a shared decision-making tool with several functions for patients.35

Hospital-based systems can enhance medical care and accelerate the adoption of guideline-recommended therapies by starting them before patient discharge.33 Additionally, clinical audits serve as a vital tool to ensure quality of care by systematically reviewing CVD management practices and promoting the integration of evidence-based improvements into daily clinical routines. Through these combined approaches, hospitals can effectively optimise patient outcomes and streamline the implementation of best practices.

In a systematic review of RCTs on guideline implementation in cardiovascular care, 54 trials evaluated unimodal strategies and 30 evaluated multimodal strategies compared with usual care. Key findings include that of the unimodal strategies, 15 focused on provider reminder systems, three on audit and feedback, 15 on provider education, four on patient education, five on promoting self-management and 14 on organisational change. The most significant benefit was seen with organisational change (OR 1.96; 95% CI [1.4–2.75]), followed by patient education, provider education and provider reminder systems. Trials on audit and feedback and patient self-management showed mixed or minimal improvements in physician adherence. Multimodal interventions had similar effect sizes and rankings across strategies.40

A French expert panel provided a practical guide, comprising few therapeutic steps, for the implementation of guidelines for managing secondary prevention patients in routine practice, from hospital discharge up to 1 year after the index event, and focusing on the achievement of target LDL cholesterol levels.31 This approach is an efficient method to fight therapeutic inertia.

In another focus meeting, recommendations from the expert panel included harmonising guidelines to focus on common areas of consensus rather than state-of-the-art science, and applying specific tools intended to improve their quality, such as the Appraisal of Guidelines for Research and Evaluation (AGREE); removing of the boundary between primary and secondary prevention and focusing on level of overall risk; and including professional societies from different specialties in guideline development and implementation to increase ownership and decrease fragmentation among guideline committees.30 This level of cooperation would help ensure that clinical best practice roadmaps contain clear, concise and complementary, rather than contradictory, patient care information.30

Several frameworks have been proposed to guide implementation research (Table 1).32,41–43 They include three selected areas: a better assessment of the most important barriers to evidence-based care; novel community intervention strategies, such as cell phone text messaging and similar; and increasing our understanding of successful implementation and sustainability of improvements.29

Table 1: Recent Research Interventions on Guideline Implementation.

Research	Methods
RE-AIM (2019)42	Cross-cultural, cross-topic framework that focuses on the design, dissemination and implementation process
ACC Acute MI Guidelines Applied in Practice programme (2006)41	Initiative designed to improve the quality of cardiovascular care by bringing the ACC/AHA practice guidelines to the point of care in Michigan, US. The programme consisted of three different projects, involving a total of 33 hospitals in five phases: planning, tool implementation, monitoring of tool use, re-measurement and reporting of results, using a collaborative model, which included a series of learning sessions for staff members. The goal was to identify the highest care priorities for patients with acute coronary syndromes and to incorporate these into the care itself with a standardised set of clinical-care tools, such as admission orders and discharge contracts; the use of such tools is associated with improvement in adherence to guidelines.
AHA Get With The Guidelines-Stroke programme (2024)43	Since its establishment in 2003, this is one of the largest and most important nationally representative voluntary disease registries in the US. It is a continuous quality improvement initiative that collects data on patient characteristics, hospital adherence to guidelines and inpatient outcomes. Sustained increases in both the quality of care and patient outcomes over time have been demonstrated.
GOAL Canada program and the North American ACS Reflective III Pilot (2024)32	They demonstrate improvements in the uptake of non-statin therapies and achievement of LDL cholesterol targets through targeted educational and feedback interventions.

ACC = American College of Cardiology; ACS = acute coronary syndrome; AHA = American Heart Association; GOAL = guideline-oriented approach to lipid-lowering; RE-AIM = reach, effectiveness, adoption, implementation and maintenance.

It remains unclear which implementation strategy can improve physician adherence to guideline recommendation, as well as the cost-effectiveness of dissemination strategies.28

We conclude that rapid adaptation and dissemination of clinical guidelines recommendations in the hospital and outpatient settings are of paramount importance; both the art and the science of medicine should be used, with a multifaceted collaborative, multidisciplinary approach and active engagement of patients, pharmacists, primary care providers, subspecialty providers and healthcare system leaders as key stakeholders to improve the quality of care and to obtain better patient outcomes in CVD (Figure 2).29,33,38,39 Healthcare professionals will need to act locally, with implementation adapted to the economic and cultural setting.

Figure 2: Role of Guideline Implementation in the Quality Circle.

AI = artificial intelligence.

Further research focusing on cardiovascular guidelines is needed to comprehensively evaluate the barriers to developing appropriate and sustainable interventions to improve their implementation.

New Follow-up Solutions in a Digital World

Recent advances in follow-up solutions have emerged primarily in three clinical settings: AF, HF and CAD (Figure 3).

Figure 3: New Follow-up Solutions in a Digital World in Three Clinical Settings.

CIED = cardiac implantable electronic device.

Atrial Fibrillation

AF is a major global health challenge, affecting approximately 1% of the general population, rising to 5% among those aged ≥65 years. When left untreated, AF contributes to approximately 15% of all strokes, and carries independent associations with HF, cognitive decline and increased mortality. The clinical benefits of early detection are substantial: they enable personalised risk factor modification, consideration of ablation for symptomatic patients and, most importantly, initiation of oral anticoagulant (OAC) therapy, which reduces stroke risk by 65% and mortality by 25%.44

Digital health interventions have transformed AF management across three integrated clinical domains: early detection through screening and continuous monitoring, stroke risk stratification with optimised clinical pathways, and OAC adherence with safety monitoring. Despite clear guideline recommendations, large detection gaps persist. Estimates suggest that up to 30% of AF cases remain undiagnosed until a thromboembolic complication occurs.44 Digital tools address this through structured pathways that progress from opportunistic screening with smartphone apps and smartwatch photoplethysmography (PPG) to diagnostic confirmation via patch ECG or 12-lead ECG, automated CHA₂DS₂-VASc risk stratification and seamless OAC initiation with follow-up adherence monitoring.

The advent of artificial intelligence (AI) and machine learning has revolutionised AF detection capabilities. Deep learning models trained on massive ECG databases now identify AF from single-lead recordings with sensitivity ranging from 95% to 98%, and specificity of 90–96%, often matching or exceeding expert cardiologist performance. Perhaps most remarkably, the Mayo Clinic’s AI-ECG platform powered by convolutional neural networks trained on more than 180,000 ECG recordings can detect an ‘AF fingerprint’ even during sinus rhythm, predicting future AF development with an area under the receiver operating characteristic curve of 0.87.45 This capability marks a paradigm shift from reactive detection to truly predictive cardiology, although challenges remain, including false positives in low-prevalence populations and the need for specialist over-reading of algorithm outputs.

User-grade wearable devices have dramatically democratised AF screening.46 The landmark Apple Heart Study enrolled more than 419,000 participants and demonstrated that irregular pulse notifications from PPG-based smartwatch algorithms carried a positive predictive value of 84% when confirmed by subsequent ECG patch monitoring.47 Similarly, the Fitbit Heart Study across 455,000 users validated PPG detection algorithms with 98% sensitivity and 99.6% specificity.48 Devices such as AliveCor’s KardiaMobile offer single-lead ECG recording with 98.0% sensitivity and 97.4% specificity, bringing hospital-grade diagnostics into everyday use.

For high-risk populations, implantable loop recorders (ILRs) and insertable cardiac monitors (ICMs) provide continuous, long-term monitoring. The REVEAL-AF study found that ICMs detected AF in 29.3% of cryptogenic stroke patients over 18 months, compared with only 3.0% with conventional follow-up.49 The LOOP study, a landmark RCT involving 6,004 elderly patients, demonstrated that ILR screening yielded a threefold increase in AF detection and achieved a 20% relative reduction in stroke risk.50

OAC therapy, while delivering a 60–70% relative risk reduction for stroke prevention, is affected by suboptimal real-world adherence that substantially diminishes effectiveness. Digital health platforms targeting medication adherence through multimodal strategies combining patient education, reminders, symptom tracking and bidirectional provider communication have shown promising results. The SUPPORT-AF 2 trial randomised 1,103 patients starting direct OACs to either a comprehensive digital platform or standard care, demonstrating superior medication adherence, higher proportion of days covered and a 34% relative reduction in the composite endpoint of stroke, systemic embolism, major bleeding and all-cause mortality at 12 months.51

Despite these advances, important challenges persist. PPG struggles with motion artefacts and rhythm misclassification, particularly yielding low positive predictive values (<50%) in low-prevalence populations. AI algorithms face issues of external validity, signal-quality dependency and workflow burden, given that 10–20% of outputs require expert over-reading. Successful clinical integration demands automated triage systems that efficiently direct cases to specialist review for therapy initiation.

The integration of multimodal data streams combining wearable sensors, genetic risk scores, circulating biomarkers and imaging parameters promises unified predictive models for comprehensive AF risk assessment. Closed-loop AI systems that deliver real-time insights to directly inform therapeutic adjustments represent the next frontier in personalised AF management.

Heart Failure

More than 64 million people worldwide have HF, and it is estimated that more than 10% of individuals >70 years old are affected.52 Achieving optimal guideline-directed medical therapy (GDMT) for HF results in improvement of symptoms, quality of life and overall prognosis.52 Digital solutions can support clinicians in enhancing the optimisation of GDMT prescription rates in various ways.

Digital consultations are transforming healthcare by enabling remote accessibility and convenience.53 Patients can connect with healthcare providers from any location, eliminating the necessity of in-person visits. This is made possible through multimodal communication channels, including video and audio calls and text messaging, which offer flexibility in interactions. The IMPLEMENT-HF pilot RCT demonstrated that a virtual pharmacist–clinician team delivering evidence-based pharmacotherapy recommendations in a non-cardiology setting could safely optimise GDMT.54

Digital remote monitoring involves the continuous collection of real-time data from various sources to track and analyse health metrics. This approach enables clinicians to monitor patients’ health from any location. The range of cardiac implantable electronic devices is expanding, including ILRs, pulmonary artery pressure monitors, pacemakers and defibrillators.55 These devices incorporate sensors for continuous monitoring and provide real-time data on cardiac activity, thereby enhancing patient care pathways to optimise GDMT prescriptions.

The MANAGE-HF study found that HF treatment was optimised in 74% of the 585 alert cases and 54% of the 3,290 weekly alerts.56 In the OptiLink HF study, the intervention arm had 0.37 changes in GDMT every 6 months as a result of digital remote monitoring.57 Similarly, the CHAMPION study showed that when pulmonary artery pressure monitor readings were made remotely available to investigators, GDMT was changed more frequently in the remote group compared with the control group, which relied on symptoms and daily weights.58 Additional significant studies addressing both invasive and non-invasive multiparameter-guided management for HF are summarised in Tables 2–4.58–79

Table 2: Haemodynamic-guided Heart Failure Management.

Trial	Study Design	Device	Monitoring	Baseline GDMT/Diuretics Use	Intervention GDMT/Diuretics Use
COMPASS-HF (2008)59	Double arm	Chronicle	PA/RV pressure	ACEI/ARBs 85% BB 83% OD 93%	NA
HOMEOSTASIS (2010)60	Single arm	HeartPOD	LA pressure	ACEI/ARBs 95% BB 92% MRA 46% LD 100% TD 19%	ACEI/ARBs 97%* BB 97%* MRA 49% LD 95% TD 16%
REDUCEhf (2011)61	Double arm	Chronicle	PA/RV pressure	ACEI/ARBs 92% BB 96% OD 93%	NA
CHAMPION (2011)58	Double arm	CardioMEMS	PA pressure	ACEI/ARBs 79% BB 91% MRA 43% LD 92% TD 35%	ACEI/ARBs 76% BB 88% MRA 44% LD 90% TD 15%
LAPTOP-HF (2015)62	Double arm	HeartPOD	LA pressure	NA	NA
SIRONA I (2020)63	Single arm	Cordella	PA pressure	NA	NA
GUIDE-HF (2021)64	Double arm	CardioMEMS	PA pressure	ACEI/ARBs 64% ARNI 29% BB 89% MRA 48% SGLT2I 1% LD 93% TD 13%	ACEI/ARBs 57% ARNI 29% BB 87% MRA 48% SGLT2I 8% LD 93% TD 18%
SIRONA II (2022)65	Single arm	Cordella	PA pressure	ACEI/ARBs 27/13% ARNI 39% SGLT2I 17% BB 77% MRA 54% LD 95–97%	NA
MONITOR-HF (2023)66	Double arm	CardioMEMS	PA pressure	ACEI/ARBs 21/15% ARNI 46% SGLT2I 7% BB 85% MRA 81% LD 95% TD 6%	ACEI/ARBs 10/14% ARNI 60% SGLT2I 31% BB 82% MRA 84% LD 91% TD 6%
VECTOR-HF (2023)67	Single arm	V-LAP	LA pressure	ACEI/ARBs 7% BB 93% ARNI 73% MRA 53% SGLT2I 23% OD 97%	ACEI/ARBs 10% BB 96% ARNI 75% MRA 50% SGLT2I 36% OD 97%
PROACTIVE-HF (2024)68	Single arm	Cordella	PA pressure	ACEI/ARBs 14% ARNI 44% SGLT2I 57% BB 80% MRA 67% OD 93%	NA

Bold = statistically significant increase in the intervention arm of the trial (p<0.05). *Statistically significant increase (p<0.05) in dosage but not in prevalence of use. ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor blocker; ARNI = angiotensin receptor–neprilysin inhibitor; BB = β-blocker; GDMT = guideline-directed medical therapy; LA = left atrium; LD = loop diuretic; MRA = mineralocorticoid receptor antagonist; NA = not available; OD = oral diuretic; PA = pulmonary artery; RV = right ventricle; SGLT2I = sodium–glucose cotransporter 2 inhibitor; TD = thiazide diuretic.

Table 4: Non-invasive Multiparameter-Guided Heart Failure Management.

Trial	Study Design	Device	Baseline GDMT/Diuretics Use	Intervention GDMT/Diuretics Use
Antonicelli et al. (2008)75	Double arm	Telemonitoring	NA	Significant increase in BB and MRA
TEMA-HF 1 (2012)76	Double arm	Mobile phone-based telemonitoring + IoT devices	ACEI 28% ARBs 7% BB 27% MRA 24% LD 2% TD 2%	ACEI 25%* ARBs 8% BB 28%* MRA 16% LD 2%* TD 2%
TIM-HF2 (2018)77	Double arm	Telemonitoring + IoT devices	ACEI/ARBs 82% ARNI 6% BB 92% MRA 58% LD 94% TD 25%.	NA
Artanian et al. (2020)78	Double arm	Mobile phone-based telemonitoring	ACEI/ARBs 67% ARNI 14% BB 86% MRA 62%	Higher proportion of optimal GDMT doses, with quicker dose optimisation
Romero et al. (2023)79	Double arm	Telemonitoring + IoT devices	ACEI/ARBs/ARNI 94% BB 100% MRA 70% SGLT2I 0%	Higher GDMT optimisation

Bold = statistically significant increase in the intervention arm of the trial (p<0.05). *Statistically significant increase (p<0.05) in dosage but not in prevalence of use. ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor blocker; ARNI = angiotensin receptor–neprilysin inhibitor; BB = β-blocker; GDMT = guideline-directed medical therapy; IoT = Internet of things; LD = loop diuretic; MRA = mineralocorticoid receptor antagonist; NA = not available; OD = oral diuretic; SGLT2I = sodium–glucose cotransporter 2 inhibitor; TD = thiazide diuretic.

Digital solutions encompass a broad range of technologies that include remote monitoring, treatment and consultations. While early RCTs and observational cohort studies have shown promising results, further large-scale studies are needed to improve the digital solutions available for HF treatment.

Coronary Artery Disease

CAD remains the leading cause of mortality worldwide, accounting for approximately 9 million lives annually. Despite advances in revascularisation and evidence-based pharmacotherapy, secondary prevention outcomes disappoint, with 15–25% of patients having recurrent events within 5 years after acute coronary syndrome. This persistent burden stems from multiple failures: medication discontinuation rates reaching 30–50% within 1 year, inadequate lifestyle modification with only 20–30% achieving multi-domain risk factor control, and widespread therapeutic inertia in GDMT optimisation.80

Digital health interventions for CAD have evolved far beyond the pioneering text-messaging RCTs, maturing into sophisticated multimodal platforms that integrate wearable biosensors, AI-driven risk profiling, telerehabilitation programmes, remote monitoring systems and behavioural nudging technologies.

Contemporary wearable biosensors have transformed episodic clinical assessments into continuous, real-world cardiovascular phenotyping. Modern devices seamlessly combine PPG for heart rate variability and oxygen saturation monitoring, accelerometry for comprehensive physical activity and sleep pattern tracking, bioimpedance for fluid status assessment, and ECG for arrhythmia and ischaemic change detection, to deliver unprecedented ambulatory cardiovascular surveillance.

Hybrid telerehabilitation models represent another important advancement in secondary prevention. A recent meta-analysis demonstrated that digital cardiac rehabilitation programs, including app-based telehealth interventions, achieve clinical outcomes comparable to those of traditional centre-based cardiac rehabilitation, while also showing higher patient adherence and significant improvements in quality of life.81

AI now enables precision risk profiling for personalised therapy titration. Machine learning algorithms predict residual cardiovascular risk to guide LDL cholesterol target intensification, balance bleeding versus ischaemic risk for optimal antithrombotic duration, and individualise blood pressure goals based on patient-specific factors.82

The foundational digital intervention trials established proof of concept. The TEXT ME trial delivered four weekly text messages to patients, achieving 11% LDL cholesterol reduction, 8 mmHg systolic blood pressure lowering, increased physical activity and 50% smoking cessation, and more than 90% of participants rated messages as useful and appropriately frequent.83

Tekkeşin et al. demonstrated that daily smartphone coaching across 483 high-risk patients reduced 10-year atherosclerotic cardiovascular disease scores by 2.7% through comprehensive risk factor control.84 Li et al.’s randomised study of patients found that self-management smartphone applications increased guideline-recommended medication use by 41% alongside superior blood pressure and LDL cholesterol achievement.85

A comprehensive meta-analysis synthesising 18 studies (12,345 patients) has provided the strongest outcome evidence to date that mobile health technologies significantly outperformed traditional disease management in reducing all-cause hospitalisations (RR 0.68; 95% CI [0.50–0.91]), cardiac-related hospitalisations (RR 0.55; 95% CI [0.44–0.68]) and emergency department visits (RR 0.37; 95% CI [0.26–0.54]). However, no significant mortality benefit was observed (RR 1.72; 95% CI [0.64–4.64]) and major adverse cardiovascular event (MACE) reduction fell short of statistical significance (RR 0.68; 95% CI [0.40–1.15]).86

These findings reveal critical implementation gaps that temper enthusiasm. Intervention heterogeneity from simple SMS text messages to complex platforms complicates comparative effectiveness research. Most trials remain underpowered for mortality and MACE endpoints, while 6–24-month follow-ups prove insufficient for lifelong secondary prevention. Digital literacy barriers also threaten equitable implementation.

Future research priorities include comparative effectiveness trials identifying optimal intervention components for specific patient phenotypes, adequately powered pragmatic studies demonstrating mortality/MACE reduction and patient-centred outcomes research prioritising quality of life, treatment satisfaction and shared decision-making preferences.

Critical Appraisal of Digital Solutions: Limitations and Implementation Barriers

Digital solutions offer important opportunities for secondary cardiovascular prevention, but their growing availability should not be interpreted as equivalent to established clinical effectiveness. A critical appraisal of the field shows that current evidence remains heterogeneous, with substantial differences in study population, intervention design, monitored parameters, comparator groups, follow-up duration and endpoints. As a result, the digital health literature is often difficult to compare across studies and only partially translatable into routine practice.

A critical evaluation of the current evidence base highlights significant limitations that constrain generalisability and clinical applicability. First, substantial heterogeneity exists between interventions, ranging from simple text-messaging programs (e.g. TEXT ME trial) to sophisticated AI-driven adaptive platforms, complicating meta-analytic synthesis and identification of optimal configurations for specific patient populations. Second, most trials remain underpowered for hard clinical endpoints (mortality, MACE), with limited sample sizes and event rates yielding wide confidence intervals; pragmatic trials enrolling >10,000 participants with multi-year follow-up are urgently needed. Third, follow-up duration (typically 6–24 months) is insufficient to assess sustained behaviour change essential for lifelong secondary prevention.87,88 Last, profound concerns surround data privacy, ethics and technical interoperability. Wearable devices and smartphone apps continuously collect sensitive physiological, behavioural and geolocation data, often stored in cloud-based systems with cross-jurisdictional flows. Current consent processes rarely meet ethics standards, resembling lengthy terms of service rather than meaningful informed consent, while patients may feel compelled to participate when recommended by clinicians or incentivised by healthcare systems. Cybersecurity vulnerabilities expose data to breaches, enabling identity theft, insurance discrimination and fraud.89,90

Table 3: Cardiac Implantable Electronic Device-based Multiparameter Telemonitoring of Patients with Heart Failure.

Trial	Study Design	Device	Baseline GDMT/Diuretics Use	Intervention GDMT/Diuretics Use
TRUST (2010)69	Double arm	Single- or dual-chamber ICD	ACEI/ARBs 51% BB 80%	NA
ECOST (2013)70	Double arm	Single- or dual-chamber ICD	NA	NA
IN-TIME (2014)71	Double arm	Dual-chamber ICD or CRT-D	ACEI/ARBs 92% BB 91% OD 95%	NA
TELECART (2016)72	Double arm	CRT-D	ACEI/ARBs 80% BB 89% LD 92%	NA
REM-HF (2017)73	Double arm	Single- or dual-chamber ICD or CRT-P or CRT-D	ACEI/ARBs 91% BB 91% MRA 52% OD 77%	NA
RESULT (2020)74	Double arm	Single- or dual-chamber ICD or CRT-D	ACEI/ARBs 96% BB 90% MRA 96% LD 89% TD 34%	NA

ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor blocker; BB = β-blocker; CRT-D = cardiac resynchronisation therapy with defibrillator; CRT-P = cardiac resynchronisation therapy with pacemaker; GDMT = guideline-directed medical therapy; LD = loop diuretic; MRA = mineralocorticoid receptor antagonist; NA = not available; OD = oral diuretic; TD = thiazide diuretic.

AI algorithms, trained predominantly on data from well-resourced populations, risk perpetuating bias against underrepresented groups (e.g. ethnic minorities, low socio-economic status), with inadequate external validation and explainability hindering clinical trust. Fragmented ecosystems lacking standardisation between devices, electronic health records and hospital information systems impede scalability and real-world integration. The digital divide exacerbates inequities: older adults, rural residents, low-income individuals and those with limited digital literacy face barriers to access, potentially widening rather than closing health disparities.91 Cost-effectiveness analyses are sparse and methodologically limited, often failing to capture downstream savings (e.g. reduced hospitalisations) against upfront infrastructure and training costs.92 Implementation challenges include clinician workload (alert fatigue, data overload), workflow redesign, reimbursement gaps, medico-legal uncertainties and the need for multidisciplinary change management.

Conclusion

Digital solutions represent a broad and evolving group of interventions that include remote monitoring, remote treatment support, remote consultation, digital education and data-driven decision support. In secondary cardiovascular prevention, these strategies may enhance patient engagement, improve therapeutic adherence and help reduce therapeutic inertia, thereby facilitating the implementation of guideline-based care.

At the same time, the current evidence should be interpreted with caution. Although digital tools are promising and several small-to-medium studies have shown favourable effects on process indicators and intermediate outcomes, the evidence base is still limited by heterogeneity of interventions, variable implementation and a relative lack of large randomised studies with hard clinical endpoints. Therefore, the development of digital cardiovascular medicine will require not only technological innovation, but also rigorous validation, workflow integration and sustainable healthcare organisation.

Particular attention should be paid to data privacy, cybersecurity, ethics, interoperability, clinician workload, reimbursement and the digital divide, all of which may influence real-world effectiveness. Addressing these barriers will require a collaborative effort involving patients, healthcare professionals, technology developers, healthcare systems, policymakers and regulatory agencies.

Looking ahead, digital consultation, telemonitoring, wearable-based follow-up, digital rehabilitation and AI-supported decision systems are likely to become increasingly relevant components of cardiovascular prevention. Their true value, however, will depend on whether they can deliver equitable, clinically meaningful and cost-effective improvements in care. Future research should therefore focus on pragmatic implementation strategies and robust outcome evaluation to determine how digital solutions can best be incorporated into routine secondary prevention pathways in AF, HF, CAD and beyond.