Study setting and participants

The ECASeT trial (Ensayo Clínico abierto y Aleatorizado para valorar la Seguridad de la Teleconsulta, frente a la consulta clínica presencial in Spanish) is aimed at testing safety of remote consultations, based on its non-inferiority compared with face-to-face consultation. The ECASeT trial will be conducted in four centers in the healthcare area of Alt Penedès-Garraf in Catalonia (North-East Spain): three secondary hospitals (Hospital Sant Antoni Abat, Hospital Sant Camil, and Hospital Comarcal de l’Alt Penedès) and a rehabilitation center, all part of Healthcare Consortium Alt Penedès Garraf (CSAPG). All study sites belong to the public network of hospitals and primary care centers of the Catalan Health Service and provide universal healthcare to a catchment population of 247,357 inhabitants. Hospital departments from all areas (i.e., locomotive, internal medicine, surgery, gynecology, and pediatrics) are involved in the study.

Participants will be screened from all individuals, irrespective of age and sex, who require follow-up in outpatient consultations of any of the departments involved in the study. Eligibility criteria have been established based on the local guidelines for screening patients for remote consultation [15]. Briefly, to be eligible for remote consultation, patients have to have adequate digital literacy at the physician discretion and technological capacity to use the video conferencing software and undergo a type of follow-up that, due to the moderate complexity of the pathology, does not require physical examinations. Patients followed up in more than three departments, those with visual or hearing impairments that may hamper patient-physician communication, and those enrolled in another clinical trial that requires an experimental intervention during the follow-up will be excluded from the study.

Interventions

Participants will be randomly allocated into one of the two study arms: control (i.e., face-to-face consultation) and remote consultation (including teleconsultation, and video consultation). All patients, irrespective of the consultation type, will receive short text message reminders a few days before the visit.

Participants allocated into the control arm will be scheduled with face-to-face appointments as usual. All phone calls (except reminders for scheduled visits) performed by the physician during the study will be noted in the electronic case report form (eCRF). Per protocol, participants allocated in the control arm can receive up to 25% of physician-doctor interactions by phone (that is, one phone call allowed for three face-to-face appointments).

Participants allocated in the remote consultation arm will be scheduled remote visits, either phone or video consultation. In case of phone visits, the physician will call the main phone number provided by the patient and, in case of not receiving an answer, other phone numbers listed as contact numbers. In case of video consultation, the participant will receive an additional text message with the visit link. When clicking the link, a website with the video consultation system embedded will open in the default browser of the user. The video consultation platform of the physician pops up a message indicating that the patient is in the virtual waiting room. Patients will remain in the virtual waiting room until the physician starts the visit. A technical assistance is readily available for physicians who experience troubles with the system. In case the patient does not show up to the video consultation after a reasonable time, the physician calls the patient by phone. The date and hour of the phone call and the reason for the inability to connect are recorded in the eCRF.

Participants allocated in the remote consultation arm that require face-to-face visits will be excluded from the per-protocol analysis, and the reason for the face-to-face visit will be recorded in the eCRF. Participants allocated in the remote consultation arm and scheduled with a phone visit cannot be visited by video, whereas those scheduled for a video consultation can undergo up to 25% of phone visits. Phone contacts and their reason in patients in the video consultation arm will be recorded in the eCRF.

Non-scheduled visits will be defined as any contact between the patient and the physician that occurs before the date established in the previous follow-up visit. Non-scheduled visits will be performed according to the study arm and type of remote consultation (i.e., phone or video) in which the patient has been allocated and will be considered for the analysis.

Screening, allocation, and blinding

All consecutive patients who visited any of the participating departments between May 2022 and May 2024 will be informed about the study design and objective and will be offered to enroll in the trial. Patients who agree to participate will provide their informed consent, which will be signed in case of a face-to-face screening visit and oral in case of a remote visit. Patients who reject enrollment will be recorded in an anonymous registry. Once signed the informed consent, the investigator will assess the eligibility criteria and record them in the eCRF.

Patients will be randomly allocated in either of the study arms irrespective of the modality of the screening visit. Randomization will be balanced by medical specialty to prevent biases towards a particular diagnosis within study arms. To ensure a minimum number of participants who visited through video consultation, participants allocated to the remote consultation arm will be subsequently randomized to either video consultation or remote modality selection at physician discretion at a 3:7 ratio.

Centralized randomization will be automatically conducted by the randomization module of the REDCAP system used for the eCRF. The allocation will be informed to the administrative service, which will schedule all visits indicated by the physician during the investigated period following the criteria of each study arm. The health condition for which a patient has attended the screening visit will be considered the baseline disease for the study purposes.

Although study participants will be informed about the overall study objectives in the patient information sheet, they will not be aware of the study endpoints. In addition, the statisticians performing the analysis will be blinded to participant allocation. No rules for unblinding have been pre-defined in the study protocol.

Data sources

To reduce the workload of participating investigators during visits, variables recorded in the eCRF will be limited to those that cannot be accurately retrieved from electronic health records (EHR), which will include both local hospital records and the dataset of the Catalan Health Surveillance System (CHSS), a central database of the Catalan Health System. By the time of submitting the study protocol to the ethics committee, all four study sites have separate and different EHRs. Throughout the following year, all the systems are to be unified into a single one and all the data from the different systems will be migrated. These EHR collect all the data related to the care received in the centers, clinical course, prescription, diagnostic images, laboratory tests, and administration, among others. The CHSS systematically collects and stores healthcare data on health status and resource utilization from all healthcare providers. The CHSS was originally designed for healthcare planning and has an automated validation system to identify internal inconsistencies between variables. Since the CHSS registries are used for control of capitation payments, they undergo external audits to ensure provider payment accuracy [16].

Outcomes and variables

Table 1 summarizes the definitions and data sources of primary and secondary outcomes assessed in this trial. Complications and adverse reactions will be considered only if related to the baseline disease (i.e., the health condition for which the first study visit was scheduled). To prevent interference with routine practice, the investigator will have the option of recording the reference disease in a free-text field in the eCRF. The option of completing a field in English or Spanish linked to the ontologies of the ICD10 will also be available. If this field is not filled in, the ICD-10 code will be extracted from the free-text field. The investigator will also record the stage (if applicable) of the baseline disease and its severity.

Table 1 Study outcomes and data sources

The follow-up period planned by the investigator for the baseline disease will be retrieved from the local EHR. Other variables collected from EHRs at baseline will include sex, age, area of residence, the number of active prescriptions at the end of the screening visit, and the comorbidity burden, summarized using the Adjusted Morbidity Groups (AMG). Briefly, the AMG is a case-mix tool that generates a summary index of health risk by considering all chronic conditions and recent acute diagnostic codes [17]. The tool allows stratifying patients into mutually exclusive health-risk groups and has shown a high predictive capacity of healthcare and resource use outcomes, including mortality, hospitalization, visits to primary care, polypharmacy, and overall expenditure [18, 19].

For homogeneity regarding the rate of preventable hospitalizations, we considered all hospitalizations due to any of the ICD-10 diagnostic codes listed as ambulatory care sensitive conditions [20] managed in the participating departments, including conditions related to ischemic heart disease (I20, I240, I248, I249), asthma (J45, J46), chronic obstructive pulmonary disease (COPD) (J20, J41, J42, J43, J44, J47), heart failure (I110, I50, J81), seizures (G40, G41, R56, O15), diabetes (E100-E108, E110-E118, E120-E128, E130-E138, E140-E148), high blood pressure (I10, I119), and iron-deficiency anemia (D501, D508, D509).

Usability will be assessed by passive and active surveillance. Study investigators will record non-attendances, as well as all technical incidences that occurred during the visit in an unstructured field of the eCRF. For active surveillance, a trained interview panel will contact by phone a subset of 100 participants randomly selected from all patients who have completed the final visit. Interviewers will administer the computer system usability questionnaire (CSUQ), a validated questionnaire aimed at measuring the users’ satisfaction with the usability of computer systems in scenario-based usability studies [21].

The users’ satisfaction with the quality of care received will be assessed using a validated tool encompassed within the satisfaction survey plan of the Catalan Health Service [22]. The questionnaire consists of 32 items that combine binary and multiple choice questions, 10-point assessment scales, and 2 open questions. An interview panel will administer the questionnaire by phone to a subset of 300 participants (100 per arm), randomly selected from all participants who have completed the final visit.

Participant timeline

Figure 1 summarizes the participant timeline. All participants, irrespective of the study arm, will be scheduled follow-up visits at a frequency determined by the physician based on the diagnostic and therapy of the baseline disease and will receive a reminder (either SMS or phone call) before each visit. Physicians can schedule and cancel appointments at their discretion. At least three visits should be scheduled in the study setting: the baseline visit (i.e., screening visit), intermediate visits (pre-defined by the physician based on the follow-up needed for managing the baseline condition), and the final visit (i.e., either the follow-up visit closest to 12 ± 2 months after enrollment or before in case of discharge or early discontinuation).

Fig. 1

Participant timeline

aThe frequency of the follow-up visits will be determined by the physician based on the diagnostic and therapy of the baseline disease and will receive a reminder (either SMS or phone call) before each visit. bEither the follow-up visit closest to 12 ± 2 months after enrollment or before in case of discharge or early discontinuation

In addition to voluntary withdrawal, the following situations will lead participants in the remote consultation arms to discontinue the study: a change in their clinical status that requires face-to-face consultation and not following the procedures for remote consultation adequately. Patients allocated in the face-to-face arm will discontinue in case their health status discourages moving to the hospital for follow-up visits. The reasons for ending the study will be recorded in the eCRF.

The study will be temporarily interrupted if the local authorities or healthcare center managers decide to ban or limit face-to-face consultations because of the rapid spread of an infectious agent. This interruption can be partial in case banning face-to-face consultations does not affect all hospital services. The study will resume when face-to-face consultations are restored, and there is enough healthcare activity for adequate recruitment in all study arms. In case of a temporal interruption, the study will be automatically extended for a time that is at least equivalent to the interruption time.

The independent ethics committee can interrupt the study at any time in the advent of risk for participants, and the study sponsor can do it as well because of financial issues.

Data management and analysis

All data recorded in the eCRF will be stored in a REDCap platform hosted in a local server. The REDCap data will be transferred to a database, which will also contain the data extracted from the local EHR and the CHSS registries. All registries will be linked through the identification number of the Catalan Health Service.

Data recorded in the eCRF will be monitored for completeness and quality by a professional monitor working at the research office of the Hospital acting as sponsor. We will schedule periodic visits to assess the adherence of investigators to the study protocol and good clinical practices. The clinical monitor will release reports at the end of each visit.

A statistical analysis plan, with detailed analyses for each study outcome, will be performed after the last patient has attended the final visit and before the closure of the database.

The primary objective will be assessed using a non-inferiority analysis of the cumulative incidence of complications of the baseline disease between remote consultation (video and teleconsultation arms) and face-to-face consultation, using a non-inferiority margin of 2%. The analysis will be based on a modified binomial test to assess the non-inferiority of an experimental intervention vs. a control group in a three-arm trial [23]. The primary analysis will be conducted on a per-protocol study sample, which will include all participants who have finished the study and have not been withdrawn because of non-allowed visits using modalities other than scheduled. The intention-to-treat population, used for sensitivity analyses, will consist of all randomized patients completing at least one visit. Since the analysis will be based on an event rate, missing data will not be imputed.

The primary analysis will be replicated with patients stratified according to the following characteristics: hospital department, diagnostic, disease duration (chronic vs. acute or sub-acute), anatomic-therapeutic treatment groups, and GMA health risk group. The clinical and demographic characteristics of study participants will be described and compared between excluded/withdrawn participants.

The statisticians will be blinded for the allocation arm. The type I error will be set at 2.5% for the primary analysis and 5% for the rest of the analyses. All analyses will be conducted using R software (version 4.0 or higher) [24]. Data collected during the ECASeT trial will not be re-used.

Sample size considerations

The sample size was estimated based on the primary endpoint of cumulative incidence of complications related to the baseline disease. The sample size analysis was computed with R software [24], following the approach described by Steen A. Julious for non-inferiority trials [25]. Although there is little available information regarding this outcome in the general population of chronic patients, we expect 10% of events in face-to-face consultations [26,27,28] and 12% in the remote consultation arm (based on routine practice of the research team). Assuming a type-I error of up to 2.5% and a statistical power of 80%, 1068 subjects per arm should be analyzed to reject the non-inferiority hypothesis with a non-inferiority margin of 2% and assuming 10% of participants lost to follow-up.

Altogether, the participating centers schedule 252,000 routine follow-up visits during a period equivalent to the study, corresponding to 121,400 different patients. Therefore, the recruitment of the estimated 2136 patients is considered within the capabilities of the participating centers.

Owing to the scarcity and heterogeneity of available data about the primary endpoint, a blinded interim analysis will be conducted when 50% of the total sample has been recruited to revise the assumptions used to calculate the sample size, as proposed in the E9 Guide of the ICH on Statistical Principles for Clinical Trials [29].

Ethical study conduct

We will collect only the data necessary to achieve the study objectives. All data will be handled according to the Spanish Organic Low 03/2018 for the protection of personal data and assurance of digital rights (LOPD-GDD) and the General Data Protection Regulation 2016/679 on data protection and privacy for all individuals within the European Union and the local regulatory framework regarding data protection. No situations classified at high risk of violation of the rights and freedoms of patients according to the LOPD-GDD are foreseen. Nevertheless, considering that the study will include video conferencing, we will perform the risk analysis report requested by the LOPD-GDD. The data generated during the study will be stored for 5 years after the study ends unless the Independent Ethics Committee or the local regulatory framework establishes otherwise.

The study protocol has been approved by the Independent Ethics Committee of the Bellvitge Hospital (L’Hospitalet de Llobregat, Spain) (ref. PR215/21). All changes made to the study design after protocol approval will be notified to the Independent Ethics Committee and documented as an amendment in the ClinicalTrials.gov registry. The scientific committee and the sponsor are responsible for communicating all amendments that have been approved by the Independent Ethics Committee to the rest of the investigators, as well as, patients already enrolled in the trial if deemed necessary.

When offered to enter the study, all participants will be provided with a patient information sheet (Additional file 2). All patients who accept to participate in the study will provide informed consent before entering the study. In case the first visit is remote, consent will be oral, otherwise, written. Participants younger than 18 years will sign an informed assent, and their legal guardian will provide the corresponding consent.

The members of the scientific committee commit to public dissemination of the trial results. Therefore, upon trial completion and analysis, a report will be submitted in a peer-reviewed, open-access journal, regardless of the trial results. The scientific committee will be responsible for making decisions regarding authorship, which will be in agreement with the authorship criteria of the International Committee of Medical Journal Editors. The report will also be sent to local authorities responsible for healthcare planning.