ESC HEART FAILURE STUDY DESIGN ESC Heart Failure 2020; 7: 4448–4457 Published online 17 September 2020 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ehf2.12962 Study design of Heart failure Events reduction with Remote Monitoring and eHealth Support (HERMeS) Sergi Yun1,2,3 , Cristina Enjuanes1,2,4, Esther Calero1,2,4, Encarnación Hidalgo1,2,4, Marta Cobo5,6, Pau Llàcer7, José Manuel García‐Pinilla6,8, Álvaro González‐Franco9, Julio Núñez6,10, José Luis Morales‐Rull11, Paola Beltrán2,12, Cristina Delso13, Román Freixa‐Pamias2,14, Pedro Moliner1,2,4, Xavier Corbella2,3,15, Josep Comín‐ Colet1,2,4,16* and the HERMeS trial investigators group 1 Community Heart Failure Program, Departments of Cardiology and Internal Medicine, Bellvitge University Hospital, L’Hospitalet de Llobregat, Barcelona, Spain; 2Bellvitge Biomedical Research Institute (IDIBELL), L’Hospitalet de Llobregat, Barcelona, Spain; 3Department of Internal Medicine, Bellvitge University Hospital, L’Hospitalet de Llobregat, Barcelona, Spain; 4Department of Cardiology, Bellvitge University Hospital, L’Hospitalet de Llobregat, Barcelona, Spain; 5Department of Cardiology, Puerta de Hierro Majadahonda University Hospital, Puerta de Hierro‐Segovia de Arana Health Research Institute (IDIPHSA), Madrid, Spain; 6Biomedical Research Networking Center on Cardiovascular Diseases (CIBERCV), Carlos III Health Institute (ISCIII), Madrid, Spain; 7Department of Internal Medicine, Manises Hospital, Medical Research Institute of Hospital La Fe (IIS La Fe), València, Spain; 8Department of Cardiology, Heart Failure and Familial Cardiomyopathy Unit, Virgen de la Victoria University Hospital (HUVV), Institute of Biomedical Research in Málaga (IBIMA), Málaga, Spain; 9Department of Internal Medicine, Central de Asturias University Hospital (HUCA), Foundation for Health and Biomedicine Research and Innovation of Asturias (FINBA), Oviedo, Spain; 10Department of Cardiology, Clinic of València University Hospital, Biomedical Research Institute of València (INCLIVA), School of Medicine, University of València, València, Spain; 11Department of Internal Medicine, Heart Failure Unit, Arnau de Vilanova University Hospital, Lleida Biomedical Research Institute’s Dr. Pifarré Foundation (IRBLleida), Lleida, Spain; 12Department of Cardiology, Viladecans Hospital, Institut Català de la Salut (ICS), Viladecans, Barcelona, Spain; 13Chronic Care Teams, Primary Care Service (SAP) Delta Llobregat, Foundation University Institute for Research in Primary Health Care Jordi Gol i Gurina (IDIAPJGol), L’Hospitalet de Llobregat, Barcelona, Spain; 14Department of Cardiology, Sant Joan Despí Moisès Broggi Hospital, Consorci Sanitari Integral (CSI), Department of Clinical Sciences, School of Medicine, Autonomous University of Barcelona (UAB), Barcelona, Spain; 15Hestia Chair in Integrated Health and Social Care, School of Medicine, Universitat Internacional de Catalunya, Barcelona, Spain; 16Department of Clinical Sciences, School of Medicine, University of Barcelona (UB), Barcelona, Spain Abstract Aims The role of non‐invasive telemedicine (TM) combining telemonitoring and teleintervention by videoconference (VC) in patients recently admitted due to heart failure (HF) (‘vulnerable phase’ HF patients) is not well established. The aim of the Heart failure Events reduction with Remote Monitoring and eHealth Support (HERMeS) trial is to assess the impact on clinical outcomes of implementing a TM service based on mobile health (mHealth), which includes remote daily monitoring of biomet- ric data and symptom reporting (telemonitoring) combined with VC structured, nurse‐based follow‐up (teleintervention). The results will be compared with those of the comprehensive HF usual care (UC) strategy based on face‐to‐face on‐site visits at the vulnerable post‐discharge phase. Methods and results We designed a 24 week nationwide, multicentre, randomized, controlled, open‐label, blinded endpoint adjudication trial to assess the effect on cardiovascular (CV) mortality and non‐fatal HF events of a TM‐based comprehensive management programme, based on mHealth, for patients with chronic HF. Approximately 508 patients with a recent hospital admission due to HF decompensation will be randomized (1:1) to either structured follow‐up based on face‐to‐face appoint- ments (UC group) or the delivery of health care using TM. The primary outcome will be a composite of death from CV causes or non‐fatal HF events (first and recurrent) at the end of a 6 month follow‐up period. Key secondary endpoints will include components of the primary event analysis, recurrent event analysis, and patient‐reported outcomes. Conclusions The HERMeS trial will assess the efficacy of a TM‐based follow‐up strategy for real‐world ‘vulnerable phase’ HF patients combining telemonitoring and teleintervention. Keywords Chronic heart failure; Telemedicine; mHealth; Outcomes research; Chronic care model; Transitional care Received: 23 May 2020; Revised: 17 July 2020; Accepted: 30 July 2020 *Correspondence to: Josep Comín-Colet, Hospital Universitari de Bellvitge, Department of Cardiology, 19th Floor, Feixa Llarga s/n, 08907 Hospitalet de Llobregat, Barcelona, Spain. Tel: +34932607078. Email: jcomin@bellvitgehospital.cat © 2020 The Authors. ESC Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made. Rationale and design of the HERMeS trial 4449 Introduction across pre‐specified subgroups of patients and were accom- panied by a reduction in hospital costs and improvement in Despite the improvement in pharmacological treatment and patient‐centred outcomes.7,13 other non‐pharmacological approaches, heart failure (HF) re- In order to confirm the results of the iCOR study and to as- mains a huge public health problem in developed countries.1 sess the scalability of this mobile health (mHealth) based Population studies have shown a rising prevalence of HF model of care in other health care areas and settings, we de- that leads to an increase in medical resource use and signed the ‘Heart failure Events reduction with Remote Mon- expenditure.2 A significant proportion of such health expendi- itoring and eHealth Support’ (HERMeS) trial. The HERMeS ture is due to hospital admissions, and therefore, efforts trial is a nationwide, multicentre, randomized study aimed to improve the delivery of chronic care and avoiding to assess the effect of a TM‐based comprehensive manage- readmissions in these patients have been promoted in ment programme on CV mortality and non‐fatal HF events recent years.3 compared with usual care (UC) in patients with chronic HF. Readmission rates due to HF are high and tend to concen- In this study, TM is delivered using mHealth technology that trate in the first few months after hospitalization.4,5 For this allows the combined use of telemonitoring of signs and reason, new multidisciplinary transitional care interventions symptoms and teleintervention by means of VC. have been focused on the so‐called ‘vulnerable phase’ The present article provides a full description of the HER- that comprises the period that begins immediately after MeS trial design and the mHealth‐based TM platform specif- hospital discharge and extends from 3 to 6 months during ically designed for this study. post‐discharge follow‐up.3,6 Transitions of care in this period, including actions such as discharge planning, drug concilia- tion, and education, should ideally start during hospitalization before discharge.5 Telemedicine (TM) has been proposed as Study design one of the key components that may help to improve out- comes in transitional HF care programmes. Potential TM The HERMeS trial is a multicentre, randomized (1:1), con- interventions include daily control of biometric data and trolled, open‐label, blinded endpoint adjudication (PROBE) symptoms (telemonitoring) and telemedical interventions investigator‐initiated study (ClinicalTrials.gov Identifier: such as structured follow‐up based on telephone calls or NCT03663907). videoconference (VC) to provide delivery of evidence‐based Patients will be recruited for 24 months in 10 centres care.7,8 across Spain (Appendices S1, S2, and S3) and followed up Trials assessing separately the efficacy of structured for a fixed period of 6 months. The study protocol was ap- follow‐up (telephone) or telemonitoring have shown mixed proved by the institutional review boards of the coordinating results probably owing to differences in (i) the TM models centre (Bellvitge Biomedical Research Institute, Institut evaluated, (ii) the HF patient populations included, and (iii) d’Investigació Biomèdica de Bellvitge [IDIBELL], Barcelona, the health care contexts.7–12 All these factors may explain Spain) and the recruiting centres. The study will be executed why current clinical practice guidelines do not make specific in accordance with the principles of the Declaration of Hel- recommendations regarding the use of non‐invasive TM in sinki (1996); the International Conference on Harmonization the management of patients with HF.1 Thus, the exact role Good Clinical Practice; and local, national, and international and the potential benefits of delivering HF care by means of regulations, including the legal regulations about personal TM need further evaluation particularly in the setting of data confidentiality (Organic Law 3/2018 of 5 December on new transitional care models. In these new models, delivery the protection of personal data and the guarantee of digital of care based on the combined use of telemonitoring (to rights and, by extension, General Data Protection Regulation provide early detection of worsening and enabling early pre- [EU] 2016/679). All enrolled patients will provide written in- ventive therapeutic interventions) and teleintervention (to formed consent for participation in the study. provide evidence‐based structured follow‐up by means of The trial was designed and will be implemented by the VC) is an appealing approach. HERMeS Steering Committee (Appendix S4). Safety will be re- We piloted this strategy in a single‐centre proof‐of‐ ported by local investigators in accordance with the current concept clinical trial (The insuficiència Cardíaca Optimització legislation that regulates pharmacovigilance in Spain. End- Remota [iCOR] Heart Failure Remote Optimization).7 In this point adjudication for the components of the primary out- small study, we demonstrated that adding a combination of come of the study will be performed by an independent telemonitoring and teleintervention to an existing compre- endpoint committee (clinical endpoint committee [CEC]) hensive HF programme reduced the risk of non‐fatal HF blinded to the group allocation and to the recruiting centre events and the risk of HF and cardiovascular (CV)‐related on an ongoing basis according to the reporting of major readmissions as compared with structured follow‐up based events (death, readmissions, and non‐fatal HF events) made on face‐to‐face appointments. These benefits were seen by the local investigators (Appendix S5). ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 4450 S. Yun et al. Population and recruitment specific clinical pathway that includes pre‐planned structured follow‐up contacts with the health care team using VC. Pa- The main inclusion criteria are (i) age ≥ 18 years, HF diagnosis tients in the UC arm will be followed up according to the according to European Society of Cardiology criteria, (ii) a re- UC of each recruiting centre. Because all the recruiting cen- cent hospital discharge after HF hospitalization (≤30 days), tres have active and mature ongoing HF programmes, no and (iii) written informed consent to participate. Detailed in- maximum number of pre‐planned contacts has been defined, clusion and exclusion criteria are shown in Table 1. particularly in the UC arm. Details of the TM platform and the follow‐up strategies of the two groups will be described later. In both groups, patients and caregivers will receive Randomization and stratification health education to improve their self‐care behaviours. All patients will be trained and encouraged to carry out daily Eligible patients before discharge or attending outpatient self‐monitoring of biometric data (weight, blood pressure clinics, in a period of 30 days from discharge, will be screened [BP], and heart rate [HR]) and symptom self‐monitoring in and invited to participate in the study provided they fulfil in- order to detect new episodes of HF that could be managed clusion and exclusion criteria. Participants will be randomized early with oral diuretic adjustments or intravenous diuretics in a 1:1 ratio to control (UC arm) or intervention group (TM on an outpatient basis without admission. In both groups, arm). Randomization will be stratified at each centre and ac- when an acute decompensation is suspected, nurses and cording to the presence or absence of frailty to ensure bal- physicians will promote diuretic dose adjustments following anced assignment of frail patients to each group. For the specific protocols and algorithms (Appendix S6) and/or ob- purpose of this study, frailty will be defined according to tain the immediate support of an HF specialist, who will the criteria specified in Table 2 in line with the definition used make a decision according to the patient signs and symp- in the iCOR study.7 The randomization process will be carried toms (e.g. hospitalization, face‐to‐face visit, and day hospi- out centrally by a dedicated algorithm of the Research Elec- tal). In case of unplanned contacts because of development tronic Data Capture (REDCap®) platform. of new symptoms or abnormal biometrics, decision making will depend on the judgement of the attending physician. In both groups, unplanned face‐to‐face contacts could be Study assessments and follow‐up made if the clinical situation of the participants warrants it. In both arms, patients and caregivers will be able to con- All the patients included in the study will be followed up for tact health care professionals involved in patient care di- 6 months (Figure 1). Briefly, patients included in the TM rectly by telephone during working hours, depending on arm will be telemonitored and followed up according to a the availability of each HF unit. Table 1 Inclusion and exclusion criteria Inclusion criteria Exclusion criteria • Age ≥ 18 years • Age < 18 years • Patients discharged from an HF hospitalization within • Participation in another clinical trial 30 days of enrolment into the study or in the process of • Moderate or severe cognitive impairment without a competent discharge planning caregiver • HF diagnosis according to ESC criteria • Lack of social support • Written informed consent obtained before any assessment • Institutionalized patients is performed • Life expectancy > 1 year (excluding HF) • Patients receiving oral standard medication for CHF • Institutional‐based or end‐of‐life care • All patients will be eligible regardless of the level of LVEF: • Serious psychiatric illness HFrEF, HFmrEF, and HFpEF • Planned cardiac surgery • Planned heart transplantation or LVAD implant • Patients in haemodialysis programme • Death before hospital discharge • The patient is unable or unwilling to give the informed consent to participate • The patient is considered an unsuitable candidate for this study according to the decision of the local investigator • Unstable patients with signs of fluid overload or low cardiac output CHF, chronic heart failure; ESC, European Society of Cardiology; HF, heart failure; HFmrEF, heart failure with mid‐range ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; LVAD, left ventricular assist de- vice; LVEF, left ventricular ejection fraction. ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 Rationale and design of the HERMeS trial 4451 Table 2 Frailty criteria to stratify the patients included in the study provision of multichannel service and monitoring of patients before randomization through the following: a Frailty criteria • Age ≥ 90 years 1 Patient monitoring of • Age 85–89 years needing caregiver • Barthel index < 90 points at any age • Pfeiffer test ≥ 3 mistakes at any age a Biometric data (weight, HR, and BP); • Patient is or will be planning a Home Care Program (ATDOM b Symptom report: seven questions to capture the wors- programme) ening of the symptoms of heart disease, mainly worsen- ATDOM, (Atenció Domiciliària) home health care. ing of the HF, and one question to capture general a Any participant will be considered as fragile if he or she meets at deterioration (Table 3). The questions are posed to an- least one of the prior criteria. swer ‘yes’ or ‘no’. 2 Generation and management of warning alarms notified to Telemedicine group: the PIRENe platform the professionals assigned to each patient in case one of Follow‐up and treatment of patients in the TM arm will the following situations occurs: be based on the PIRENe platform. The PIRENe platform (Platform for the Provision of Tele‐Intervention, Remote a Out‐of‐range biometric data (according to preset ranges Monitoring and Empowerment to people with specified in Appendix S7). advanced/complex chroNic [CV] disease based on electronic b Any alarm symptom among the answers to the question- health [eHealth]) is a comprehensive solution for the care naire (one ‘yes’ answer in the questionnaire generates a and monitoring of chronic patients, modelled and tested in warning alarm in the platform). chronic HF patients (Figure 2). This platform allows the c Absence of biometric measurements on any given day. FIGURE 1 Flow chart of the study. CV, cardiovascular; HF, heart failure; PROMs, patient‐reported outcome measures; TM, telemedicine; UC, usual care. ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 4452 S. Yun et al. FIGURE 2 The PIRENe platform (Platform for the Provision of Tele‐Intervention, Remote Monitoring and Empowerment to people with advanced/ complex ChroNic [CV] Disease based on e‐Health). Table 3 Daily symptom telemonitoring gateway assigned to each patient, as well as to retrieve bio- metric values and responses to the symptoms questionnaire Symptom report Answers to be assessed by the nursing and medical staff; and (ii) a 1. My feet are more swollen than usual. Yes/No household patient interface (app version for smartphone 2. I feel more fatigued, tired or with a Yes/No sensation of choking. with Android operating system of the PIRENe platform). This 3. I had bad nights because of shortness Yes/No is the gateway to the TM platform for receiving and sending of breath or sensation of choking. information (4G or WiFi). By means of the designed app, pa- 4. I needed more pillows to breathe better Yes/No at night lying on my bed. tients and caregivers will manage the daily monitoring of 5. I had to sleep seated because of shortness Yes/No biometric data retrieved from peripheral devices (scale and of breath or sensation of choking when lying BP/HR monitor) by Bluetooth, and symptoms will be re- on the bed. 6. I felt weaker or more dizzy than usual. Yes/No trieved from questionnaires available in the app. Planned 7. I had more breast pain than usual. Yes/No VCs will also be launched through the app (4G or WiFi); and 8. In general, I feel worse than usual. Yes/No (iii) monitoring devices: digital scale and BP/HR monitor with Bluetooth connection provided to the patients. Linking of data between medical devices, gateway, and the clinical This will allow medical professionals to contact the patient workstation will be performed according to national regula- proactively and thus be able to act earlier. tions for the of handling personal data. At the beginning of 3 Follow‐up via teleconsultation (videoconferencing, audio follow‐up, each patient randomized to the TM arm will be conferencing, mailing, and management of incoming and provided with the equipment detailed in Appendix S8. outgoing calls) between professionals and As mentioned earlier, the follow‐up of patients is based on patients/caregivers a standardized clinical pathway. The assessments performed at each visit (evaluation schedule) are shown in Table 4. The platform is composed of three key technical elements: (i) a health care web application interface for health care pro- Usual care group fessionals (PIRENe Platform website). This application is a Patients randomized to the UC arm will be followed up for clinical workstation designed to store basic patient informa- 6 months, with no maximum number on planned face‐to‐face tion. It allows videoconferencing, management of agendas contacts, and will receive the standard of care considered UC (patient and health care professionals), and the specific at each centre. As mentioned before, all participating centres ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 Table 4 Evaluation schedule a Rationale and design of the HERMeS trial Variables V1 Baseline V2 Week 3 V3 Week 5 V4 Week 7 V5 Week 9 V6 Week 11 V7 Month 3 V8 Month 4 V9 Month 5 V10 Month 6 Demographic x Aetiology of HF x CV risk factors x x x Co‐morbidity x x x Treatment x x BP, HR, weight x x x x x x x x x x Height, BMI x NYHA x x x x x x x x x x Signs and symptoms x x x x x x x x x x EKG x x x b Echocardiography x x b Laboratory values x x x EQ‐5D, VAS x x Barthel, Pfeiffer, FRAIL x Socio‐economic variables x EHFSCBS x x x Satisfaction x x Events x x x x x x x x x x Technological use x x a The initial visit will be the only face‐to‐face visit of the study. The rest of the visits will be by videoconference in the TM arm. b In the first visit, the last data available before the visit will be recorded; and in the final visit, the last data available before the visit will be recorded, when available (non‐mandatory data). BMI, body mass index; BP, blood pressure; CV, cardiovascular; EKG, electrocardiogram; EHFSCBS, European Heart Failure Self‐care Behaviour scale; EQ‐5D, European Quality of Life 5 Dimensions; FRAIL, fatigue, resistance, ambulation, illnesses, loss of weight; HF, heart failure; HR, heart rate; NYHA, New York Heart Association; TM, telemedicine; VAS, visual analogue scale. ESC Heart Failure 2020; 7: 4448–4457 4453 DOI: 10.1002/ehf2.12962 4454 S. Yun et al. have active integrated HF programmes in place delivering (i) Statistical methods nurse‐based coordination of care, (ii) education to patients and caregivers, (iii) pro‐active up‐titration of HF Sample size disease‐modifying drugs, (iv) cardiac rehabilitation and other According to our previous data,7 the expected event rate usual interventions provided in HF programmes, and (v) open for the primary endpoint (CV death or non‐fatal HF event) access clinics to treat HF worsening without hospital admis- in the control group would be 49% after 6 months of sion. The participating centres represent a wide range of HF follow‐up. To obtain a reduction of 25% in the rate of models deployed in different settings (tertiary, secondary, the primary endpoint in the group managed with TM, and and community hospitals and home‐based primary care assuming an alpha risk = 0.05 and a beta error of 20%, programmes), coordinated by different specialties (cardiol- we need to recruit 508 analysable patients (254 patients ogy, internal medicine, and primary care) in the various Span- in each arm). We estimate that a 25% reduction in the pri- ish regions with differences in the organization of health care. mary endpoint would drop the expected event rate of this In both arms, unplanned contacts will be permitted in the endpoint from 49% in the control group (UC) to a 36% in event of clinical deterioration. the TM group. Thus, in our opinion, this reduction is clini- cally relevant. Data collection Analysis of primary and secondary outcomes The medical history, relevant clinical and demographic infor- Details on the final statistical analyses will be defined in a mation, a physical examination, laboratory tests, functional, dedicated complete statistical analysis plan after the last pa- cognitive, and socio‐family evaluation, and the technology tient’s last visit and before the database lock. skills of patients included in the study (Appendix S9) will be obtained at baseline and at the final study visit. Clinical a Time to first event analysis:The incidence of events in events will be prospectively captured by investigators and re- the two study groups (TM and UC) will be described ported through a dedicated formulary of the electronic case using Kaplan–Meier survival functions, which will be report form. These events will be allocated and validated by compared using the log rank test. Additionally, we will CEC members during the study. All data will be registered in use Cox proportional risk regression models to obtain the REDCap® platform. the hazard ratio of each of the events under study by comparing TM (as exposure to study) with UC as a refer- ence group. Outcomes and event adjudication b Recurrent event analysis:A bivariate negative binomial regression analysis will be performed to determine the The primary outcome will be a composite of death from CV effect of the intervention on CV hospitalizations and causes or non‐fatal HF events (first and recurrent) at the non‐fatal HF events (first and recurrent). Coefficients end of the 6 month follow‐up period as previously used in from this method will be estimated by accounting for other studies.14,15 A non‐fatal HF event is defined as a new the positive correlation among the recurrent outcome episode of worsening of symptoms and signs consistent with and death as a terminal event, by linking the two simul- acute decompensated HF requiring intravenous decongestive taneous equations (readmission count and death) with therapy (e.g. diuretics) either on an outpatient basis (e.g. shared frailty.16 In addition, each patient’s follow‐up day‐case HF hospital or at home) or in the emergency depart- time will be used as an offset in the models to account ment (<24 h), or requiring unplanned hospital admission for differences in the follow‐up. By using this methodol- (>24 h) or complicating the course of a non‐CV admission. ogy, the potential for bias due to death as informative Key secondary endpoints will include components of the censoring is minimized, an issue commonly seen in acute primary event analysis (all‐cause, CV and HF death, and/or HF studies. Risk estimates from this method are pre- hospitalization), days spent in the hospital (length of stay sented as incidence rate ratios and 95% CIs. for readmission), emergency visits, patient‐reported outcome measures such as self‐care behaviour and health‐related The subgroup analyses for the primary event will be previ- quality of life (European Quality of Life 5 Dimensions), and ously specified in the final statistical analysis plan, based on patient‐reported experience measures. relevant clinical characteristics such as the left ventricular As stated previously, the CEC will adjudicate all the events ejection fraction (LVEF), age, or gender, among others. Once reported during the HERMeS trial to be used for the analyses defined, the interactions between the subgroups and the of the primary and secondary endpoints (Appendix S10), ac- type of management (TM vs. UC) in the primary event will cording to pre‐specified criteria (Appendix S11). be evaluated. ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 Rationale and design of the HERMeS trial 4455 Discussion calls or VC. These virtual pre‐planned encounters may allow nurses educating the patient and the caregiver in order to The HERMeS trial is an unique clinical trial since it (i) inte- promote patient empowerment and adequate self‐care be- grates mHealth as a management tool in patients with HF haviours. Moreover, remote monitoring is a source of individ- combining two key elements of TM (telemonitoring and ual biometric and symptoms data that can be leveraged to structured teleintervention); (ii) evaluates the impact of create opportunities for drug optimization, because they the simultaneous use of these two modalities of TM in the may facilitate the decision‐making process of up‐titration of most ‘vulnerable phase’ of patients with HF; (iii) integrates HF disease‐modifying drugs in these patients.7,8 The target mHealth in pre‐existing HF programmes with a high standard population that can benefit most from this management of care and covering a broad range of care models and set- strategy is the one that has a higher risk of presenting new tings; and (iv) includes real‐world patients in terms of co‐ HF (fatal or non‐fatal) events. Early detection means early morbidities, frailty, or range of LVEF. treatment, thus slowing disease progression and reducing Beyond telemonitoring and teleintervention, the PIRENe clinical events and health care costs.7,17,18 As previously men- platform allows promoting self‐care and self‐efficacy of tioned, the iCOR study7 demonstrated the positive impact patients by providing them with up‐to‐date information of combined remote monitoring and teleintervention on about the individual progression of their own biometrics HF patients during the transition between discharge and and self‐perceived health status. From our previous pilot early post‐discharge ambulatory care. Therefore, in o ur experience, we believe that the care delivered with the mo- study, we built a mHealth platform capable of providing dality of TM used in the present study will translate into the combination of these two modalities of TM during the so‐- an improvement of outcomes. Particularly, combining, from called ‘vulnerable phase’ of HF where most adverse events one side, teleintervention by VC to deliver structured concentrate in these patients.5 pre‐planned interventions and, from the other side, In the current health care context of aging and multiple co‐ telemonitoring of signs and symptoms, may be an effective morbidities, these models of remote care have many advan- strategy especially for those patients with HF in the tages. In contrast with usual HF programme models, where so‐called ‘vulnerable phase’ , the period that encompasses only highly selected patients are candidates to be managed discharge from an acute HF admission and the following early in hospital‐based units and delivery of care is based on post‐discharge period. Besides, from the subgroup analyses face‐to‐face encounters, these new remote management of the iCOR study,7 we believe that the benefit will extend transitional care models delivered using mHealth platforms to all HF patients in the ‘vulnerable phase’ regardless of allow extending a high standard of care to real‐world patients LVEF,13 presence of frailty, depressive symptoms, or co‐ who are often expelled from specialized hospital care. Inter- morbidities. The purpose of the HERMeS trial will also include estingly, in our previous study, we showed that the efficacy validation of the main results of the iCOR study in a nation- of TM combining telemonitoring and teleintervention was wide ‘vulnerable phase’ HF patient’s cohort and confirm the similar in all patients regardless of age, frailty, co‐morbidities, robustness of the results among different clinical subgroups and ejection fraction.7,13 This approach will reduce patient in- of patients and diverse health care settings. According to equalities in access to good quality of care. The organization all these aspects, the HERMeS trial will address the of care around the natural place for patients (home) by current gap of knowledge regarding the right strategy in the means of TM will engage productive interactions with pri- managed care in early post‐discharge period or ‘vulnerable mary care professionals. This fact will promote a truly inte- phase’ in patients with HF and will allow the generalization grated care in these patients and will raise de bar of quality of the results found in the iCOR study. Additionally, it will of care among primary care professionals. help to define the profile of patients that could benefit most The development that we propose can transform the cur- from a mHealth‐based management strategy combining rent management process in symptomatic patients with HF, telemonitoring and teleintervention in integrated care particularly in those in the ‘vulnerable phase’ of the disease. settings. This may eventually translate in new recommendations in In the care process focused on transitions of care, there clinical practice guidelines regarding the use of TM in the are potential advantages of using a modality of TM that managed care of patients with HF and subsequently in the combines telemonitoring and teleintervention. Firstly, it design of prevention policies or procurement of public health allows remote monitoring of biometric data and/or symp- systems. The methods and approaches we evaluate in the toms to enable the early detection and monitoring of decom- setting of HF management may find a broader application pensation and other clinical events that would otherwise lead elsewhere in cardiology and other areas of medicine involving to readmission; secondly, it establishes a channel of commu- the management of chronic conditions. Our study may help, nication with patients from their home to conduct structured through big data analysis of biometric data, to refine predic- follow‐up after discharge, either by the means of telephone tion models, redesign clinical protocols, and prompt earlier preventive interventions.19 ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 4456 S. Yun et al. This project represents an important step in linking tech- Conflict of interest nology with clinical science and incorporates mHealth as an- other tool for the follow‐up of patients with HF. Previous The authors declare that they have no conflicts of interest rel- studies have evaluated mobile technology by using it in the evant to the content of this manuscript. form of either telephone calls or electronic messages.7,9,11,17 However, the evaluation of the efficacy of mHealth combin- ing both remote monitoring and teleintervention by VC on outcomes and disease progression in the ambulatory man- Funding agement of HF20 particularly in patients with HF in the ‘vulnerable phase’ has not previously been addressed. The The HERMeS trial is funded by an unrestricted research grant HERMeS trial will provide the adequate framework to address by Novartis. this gap of knowledge. In summary, this project has the potential for significant advances in clinical management, resulting in direct improve- Supporting information ments to the lives of many thousands of HF patients. Additional supporting information may be found online in the Supporting Information section at the end of the article. Conclusions Appendix S1. The HERMeS trial participating centres. Appendix S2. The HERMeS trial participating centres (collab- The HERMeS trial will assess the efficacy of TM combining orating investigators). telemonitoring and teleintervention in real‐world HF Appendix S3. The HERMeS trial participating centres (distri- patients with recent hospitalization. We believe that this bution in Spain). mHealth‐based strategy of delivering managed care will Appendix S4. The HERMeS steering committee. translate into a significant reduction in mortality or hospital Appendix S5. The HERMeS clinical endpoint committee (CEC). readmissions in these high‐risk patients in the ‘vulnerable Appendix S6. Algorithm in case of suspected HF decompensa- phase’ of the disease. tion. Appendix S7. Pre‐set ranges of altered biometric data from which warning alarms will be generated on the PIRENe plat- Acknowledgements form. Appendix S8. Key technical elements of the PIRENe platform We want to acknowledge the technical support and contribu- and technological equipment provided to patients random- tions of Vodafone Spain in the design and conduct of this ized in the telemedicine arm. study. The PIRENe platform was developed in public partner- Appendix S9. Functional, cognitive and frailty assessment, ship with Fundació i2cat (i2cat Foundation) and with Everis patient‐reported outcome measures (PROMS) and Spain. We thank the Biostatistics Unit of Bellvitge Biomedical patient‐reported experience measures (PREMS) performed Research Institute, IDIBELL, for all the technical support. We in the trial. thank CERCA Programme / Generalitat de Catalunya for insti- Appendix S10. The HERMeS clinical endpoint committee tutional support. In addition, it should be noted that the (CEC) adjudication workflow. study has the support of the Spanish Society of Cardiology Appendix S11. The HERMeS clinical endpoint committee (SEC) and the Spanish Society of Internal Medicine (SEMI) (CEC) adjudication criteria for events. and their respective Heart Failure subgroups. References 1. Ponikowski P, Voors AA, Anker SD, heart failure: The Task Force for the di- and expenditure in patients with chronic Bueno H, Cleland JGF, Coats AJS, Falk agnosis and treatment of acute and heart failure: a population‐based analy- V, González‐Juanatey JR, Harjola VP, chronic heart failure of the European So- sis of 88 195 patients. Eur J Heart Fail Jankowska EA, Jessup M, Linde C, ciety of Cardiology (ESC). Eur Heart J 2016; 18: 1132–1140. Nihoyannopoulos P, Parissis JT, Pieske 2016; 37: 2129–2200. 3. Albert NM, Barnason S, Deswal A, B, Riley JP, Rosano GMC, Ruilope LM, 2. Farré N, Vela E, Clèries M, Bustins M, Hernandez A, Kociol R, Lee E, Paul S, Ruschitzka F, Rutten FH, van der Meer Cainzos‐Achirica M, Enjuanes C, Ryan CJ, White‐Williams C, American P. 2016 ESC Guidelines for the diagnosis Moliner P, Ruiz S, Verdú‐Rotellar JM, Heart Association Complex Cardiovascu- and treatment of acute and chronic Comín‐Colet J. Medical resource use lar Patient and Family Care Committee ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962 Rationale and design of the HERMeS trial 4457 of the Council on Cardiovascular and 10. Chaudhry SI, Mattera JA, Curtis JP, Skouri H, van der Meer P, Jankowska Stroke Nursing, Council on Clinical Spertus JA, Herrin J, Lin Z, Phillips CO, EA. Rationale and design of the Cardiology, and Council on Quality of Hodshon BV, Cooper LS, Krumholz HM. AFFIRM‐AHF trial: a randomised, Care and Outcomes Research. Transi- Telemonitoring in patients with heart double‐blind, placebo‐controlled trial tions of care in heart failure: a scien- failure. N Engl J Med 2010; 363: comparing the effect of intravenous fer- tific statement from the American 2301–2309. ric carboxymaltose on hospitalisations Heart Association. Circ Heart Fail 11. Koehler F, Winkler S, Schieber M, and mortality in iron‐deficient patients 2015; 8: 384–409. Sechtem U, Stangl K, Böhm M, Boll H, admitted for acute heart failure. Eur J 4. Comín‐Colet J, Verdú‐Rotellar JM, Vela Baumann G, Honold M, Koehler K, Heart Fail 2019; 21: 1651–1658. E, Clèries M, Bustins M, Mendoza L, Gelbrich G, Kirwan BA, Anker SD, 16. Xu X, Hardin JW. Regression models for Badosa N, Cladellas M, Ferré S, Telemedical Interventional Monitoring bivariate count outcomes. Stata J 2016; Bruguera J, Working group of the Inte- in Heart Failure Investigators. Impact of 16: 301–315. grated Program for Heart Failure Man- remote telemedical management on 17. Koehler F, Koehler K, Deckwart O, agement of the Barcelona Litoral Mar mortality and hospitalizations in ambu- Prescher S, Wegscheider K, Kirwan BA, Integrated Health Care Area, Spain. Effi- latory patients with chronic heart fail- Winkler S, Vettorazzi E, Bruch L, Oeff cacy of an integrated hospital‐primary ure: the telemedical interventional M, Zugck C, Doerr G, Naegele H, Störk care program for heart failure: a monitoring in heart failure study. Circu- S, Butter C, Sechtem U, Angermann C, population‐based analysis of 56,742 pa- lation 2011; 123: 1873–1880. Gola G, Prondzinsky R, Edelmann F, tients. Rev Esp Cardiol (Engl Ed) 2014; 12. Inglis SC, Clark RA, Dierckx R, Prieto‐ Spethmann S, Schellong SM, Schulze 67: 283–293. Merino D, Cleland JG. Structured tele- PC, Bauersachs J, Wellge B, Schoebel C, 5. Comín‐Colet J, Enjuanes C, Lupón J, phone support or non‐invasive Tajsic M, Dreger H, Anker SD, Stangl K. Cainzos‐Achirica M, Badosa N, Verdú telemonitoring for patients with heart Efficacy of telemedical interventional JM. Transitions of care between acute failure. Cochrane Database Syst Rev management in patients with heart fail- and chronic heart failure: critical steps 2015; 10: CD007228. ure (TIM‐HF2): a randomised, con- in the design of a multidisciplinary care 13. Jiménez‐Marrero S, Yun S, Cainzos‐ trolled, parallel‐group, unmasked trial. model for the prevention of rehospitali- Achirica M, Enjuanes GA, Farre N, Verdú Lancet 2018; 392: 1047–1057. zation. Rev Esp Cardiol (Engl Ed) 2016; JM, Linas A, Ruiz P, Hidalgo E, Calero E, 18. Grustam AS, Severens JL, De Massari D, 69: 951–961. Comín‐Colet J. Impact of telemedicine Buyukkaramikli N, Koymans R, Vrijhoef 6. Feltner C, Jones CD, Cené CW, Zheng on the clinical outcomes and healthcare HJM. Cost‐effectiveness analysis in ZJ, Sueta CA, Coker‐Schwimmer EJ, costs of patients with chronic heart fail- telehealth: a comparison between home Arvanitis M, Lohr KN, Middleton JC, Jo- ure and mid‐range or preserved ejection telemonitoring, nurse telephone sup- nas DE. Transitional care interventions fraction managed in a multidisciplinary port, and usual care in chronic heart fail- to prevent readmissions for persons with chronic heart failure programme: a ure management. Value Health 2018; 21: heart failure: a systematic review and sub‐analysis of the iCOR randomized 772–782. meta‐analysis. Ann Intern Med 2014; trial. J Telemed Telecare 2020; 26: 19. Frederix I, Caiani EG, Dendale P, Anker 160: 774–784. 64–72. S, Bax J, Böhm A, Cowie M, Crawford 7. Comín‐Colet J, Enjuanes C, Verdú‐ 14. McMurray JJV, Solomon SD, Inzucchi J, de Groot N, Dilaveris P, Hansen T, Rotellar JM, Linas A, Ruiz‐Rodriguez P, SE, Køber L, Kosiborod MN, Martinez Koehler F, Krstačić G, Lambrinou E, González‐Robledo G, Farré N, Moliner‐ FA, Ponikowski P, Sabatine MS, Anand Lancellotti P, Meier P, Neubeck L, Parati Borja P, Ruiz‐Bustillo S, Bruguera J. Im- IS, Bělohlávek J, Böhm M, Chiang CE, G, Piotrowicz E, Tubaro M, van der Velde pact on clinical events and healthcare Chopra VK, de Boer RA, Desai AS, Diez E. ESC e‐Cardiology Working Group Po- costs of adding telemedicine to multidis- M, Drozdz J, Dukát A, Ge J, Howlett sition Paper: overcoming challenges in ciplinary disease management JG, Katova T, Kitakaze M, Ljungman digital health implementation in cardio- programmes for heart failure: results of CEA, Merkely B, Nicolau JC, O’Meara vascular medicine. Eur J PrevCardiol a randomized controlled trial. J Telemed E, Petrie MC, Vinh PN, Schou M, 2019; 26: 1166–1177. Telecare 2016; 22: 282–295. Tereshchenko S, Verma S, Held C, 20. Wagenaar KP, Broekhuizen BDL, 8. Inglis SC, Clark RA, Dierckx R, Prieto‐ DeMets DL, Docherty KF, Jhund PS, Jaarsma T, Kok I, Mosterd A, Willems Merino D, Cleland JG. Structured tele- Bengtsson O, Sjöstrand M, Langkilde FF, Linssen GCM, Agema WRP, phone support or non‐invasive AM, DAPA‐HF Trial Committees and In- Anneveldt S, Lucas CMHB, Mannaerts telemonitoring for patients with heart vestigators. Dapagliflozin in patients HFJ, Wajon EMCJ, Dickstein K, Cramer failure. Heart 2017; 103: 255–257. with heart failure and reduced ejection MJ, Landman MAJ, Hoes AW, Rutten 9. Cleland JG, Louis AA, Rigby AS, fraction. N Engl J Med 2019; 381: FH. Effectiveness of the European Janssens U, Balk AH, TEN‐HMS Investi- 1995–2008. Society of Cardiology/Heart Failure gators. Noninvasive home 15. Ponikowski P, Kirwan BA, Anker SD, Association website telemonitoring for patients with heart Dorobantu M, Drozdz J, Fabien V, ‘heartfailurematters.org’ and an failure at high risk of recurrent admis- Filippatos G, Haboubi T, Keren A, e‐health adjusted care pathway in pa- sion and death: the Trans‐European Khintibidze I, Kragten H, Martinez FA, tients with stable heart failure: results Network‐Home‐Care Management Sys- McDonagh T, Metra M, Milicic D, of the ‘e‐Vita HF’ randomized con- tem (TEN‐HMS) study. J Am Coll Cardiol Nicolau JC, Ohlsson M, Parhomenko A, trolled trial. Eur J Heart Fail 2019; 2005; 45: 1654–1664. Pascual‐Figal DA, Ruschitzka F, Sim D, 21: 238–246. ESC Heart Failure 2020; 7: 4448–4457 DOI: 10.1002/ehf2.12962