This was a before-and-after quasi-experimental intervention study.

ValCrònic was a 5-year (2011-2016) telemonitoring program in 4 Spanish health centers in the Valencia Region (population 5 million), situated on the Mediterranean coast, within the health departments of Sagunto and Elche: Sagunto health center, Sagunto Port health center II, Elx-El Raval health center, and Santa Pola health center. In December 2013, two additional health centers were incorporated: Elx-Altabix health center and Elx-San Fermin health center. The program was led by primary care services in collaboration with the referral hospital and other institutions in the health sector, with the participation of more than 150 professionals. In the Spanish public health system, primary health centers from the same area have the same referral hospital where patients are admitted. Since the participating health centers belonged to 2 different areas, 2 referral hospitals collaborated in the study in order to collect all hospital admissions and emergency department visits.

Development of the technical procedures and protocols began in April 2011, and participants were recruited from February 2012 to February 2015 for a 1-year telemonitoring intervention. We compared the clinical outcomes before and after the intervention. The program ended in February 2016.

The study included people at high risk for rehospitalization or an emergency department visit with 1 or more of the following conditions: heart failure, COPD, type 2 diabetes mellitus, and arterial hypertension. Heart failure and COPD are the most frequent causes of nonscheduled hospital admissions, while diabetes and hypertension consume a large amount of health care resources in the field of primary care due to their high prevalence. Furthermore, we selected these chronic diseases because patients can easily measure their indicators, and which were also sensitive to changes introduced as part of the intervention. We excluded patients who did not sign the informed consent form, did not have a telephone, or had a life expectancy of less than 1 year (based on their physician’s opinion).

The Kaiser Permanente model [19] recommends stratifying the risk of patients for hospitalizations or emergency department encounters according to the Community Assessment Risk Screen (CARS) [20]; this screen can identify high-risk patients and has been validated in the Valencia Region by the Polibienestar Research Institute at the University of Valencia [21]. To improve the validity of the scale, we complemented the result obtained (high risk or not) with the clinical judgement of the participant’s usual physician.

We used a consecutive, nonprobability sampling method in the primary care setting. In addition, we actively recruited patients who met the inclusion criteria (according to the data in their eHealth records), by means of written invitations to participate in the program.

The ValCrònic program is preventive and based on innovative approaches to chronicity such as the chronic care model [22] and the Kaiser Permanente model [19]. The program features continuous telemonitoring of patients with chronic conditions and multiple comorbidities. It is led mainly by primary care but operates in collaboration with hospital services (especially the general medicine service, among others), primary care- and hospital-based emergency services, and the home hospitalization unit.

The intervention consisted of several components: participants measured their own vital signs related to the conditions included in the study and automatically entered them into their eHealth record via a wireless electronic device (tablet) provided to them for the study. In addition, an automated alert system was set up to promptly detect any alterations, which promoted health professionals’ proactive involvement. Participants also received health education and awareness interventions tailored to their conditions through informative videos available in the tablet.

On the basis of different possible combinations of the 4 diseases included in the program, we defined 8 individualized care programs and monitored each condition from the patient’s home, based on the following indicators: (1) heart failure (blood pressure, heart rate, symptom questionnaire, and weight gain), (2) COPD (COPD Assessment Test [CAT] questionnaire [22], oxygen saturation), (3) heart failure plus COPD, (4) heart failure plus diabetes (capillary glycemia), (5) diabetes plus COPD, (6) COPD plus arterial hypertension (blood pressure), (7) heart failure plus COPD plus diabetes, and (8) diabetes plus arterial hypertension.

Participants received different tools for self-measurement according to their diseases (Figure 1), along with videos instructing them in how to use them: a set of scales, a blood pressure monitor, a glucometer, and a pulse oximeter. In addition, we used health questionnaires for COPD and heart failure. Participants with COPD also completed the CAT questionnaire [23]. All information was recorded and automatically sent to the health center using the eHealth record functionalities included on the tablet (Figure 2). The frequency of the self-measurements was established individually with each patient, but health information was usually submitted on a weekly basis.

The self-reports were entered in the eHealth records in a similar way to how they would be entered in face-to-face visits to the center. This was possible due to the existence of a unique patient identifier and eHealth record for each patient within the Abucasis health information system (version II; Valencia Health Agency), which was used by the professionals in primary and specialized care in this study. Several months of prior technical work was necessary to ensure compatibility between the data sent from patients’ homes and the eHealth records. Changes were also implemented in the clinical records used for hospital admissions in every health department to enable identification of ValCrònic patients in the event they went to an emergency department or were admitted to hospital.

Alerts were programmed into the system to automatically display self-measurement values outside the normal range established for each participant. The primary care nurse who was usually in charge of managing the patient’s care was responsible for following up on these alerts by means of the relevant action protocol. Every day, when the nurse opened the eHealth record app, they received information about the patients for whom an alert had been raised in the previous 24 hours or over the weekend. In the event of an emergency, the standard face-to-face procedure was followed.

Upon receipt of the self-measurements in the eHealth center, it was up to the nurse to decide whether to call participants to check the values, ask them to come to the health center, go to their home, or consult a doctor. Before program launch, all the doctors, nurses, and administrative staff at the participating centers received specific training on the study process, the program, and the devices.

In addition, all participants received group training from their primary care nurses or doctors regarding the use of the monitoring devices and software apps. This training included content aimed at improving patient knowledge of self-care. In addition, participants received individualized training in their homes from technical personnel regarding the use of telemedicine devices. Patients received the equipment in their homes and had access to technical assistance via telephone or in-person visits from the company in charge of managing the operation of the hardware and software (eHealth department of Telefónica España, SA, Madrid, Spain). The communication protocols between monitoring devices, tablets, and eHealth records were developed and implemented jointly by the eHealth center, Abucasis, and Telefónica eHealth technicians.

A senior management committee of 9 people was set up to monitor the activities of the ValCrònic program, composed of representatives from the Regional Health Ministry, including the General Health Care Directorate, the Health Area Directorate, the 2 managers of the participating health departments (Sagunto and Elx), the head of Information Technology Systems, and the technical program coordinator; and representatives of Telefónica, who were in charge of the telemedicine devices used. Meetings were held on a quarterly basis. A scientific committee was also set up, composed of the technical program coordinator, representatives of the professionals of the 2 participating health departments, and representatives of Telefónica.

We recorded the weight and heart rate of patients, the proportion of patients with poor control of systolic (≥140 mmHg) or diastolic (≥ 90 mmHg) blood pressure, and the proportion of patients with poor control of hemoglobin A_1c (HbA_1c) (≥8%). We also recorded visits to primary care- or hospital-based emergency services due to an exacerbation of the target diseases. Finally, we recorded unscheduled (emergency) hospital admissions. In addition, we calculated indicators of clinical relevance to the ValCrònic program: absolute risk reduction, relative risk reduction, and number needed to treat to prevent a harmful outcome.

We collected demographic data (age, sex) from all the participants, as well as the indicators for each condition: weight (kg), heart rate (beats/min), blood pressure (mmHg), capillary glycemia (mg/dL), and HbA_1c from a venous blood sample (%). Although other variables were collected during the study period, we did not include them, such as the results of the questionnaires regarding signs and symptoms, in the analyses reported in this paper.

All the information was recorded in the eHealth records either on-site or remotely from the self-measurements made by patients in their homes. We followed and monitored all participants for 1 year to manage their disease.

The Committee of Ethics and Clinical Research at the Elx Health Department approved this study. We complied with the procedures and regulations of Law 15/1999 on the Protection of Personal Data of December 13, 1999 (Spanish Data Protection Agency). All participants signed a statement of informed consent and committed to taking care of the equipment loaned to them.

Compared with the preintervention year, during the intervention year, people with at least one of 4 chronic diseases (hypertension, diabetes mellitus, COPD, or heart failure) who participated in the ValCrònic program had better weight, heart rate, blood pressure, and glycemic control. In addition, primary care emergency and hospital emergency visits were decreased, despite the participants being a year older and a year further along in their disease evolution.

In participants with diabetes, participation in ValCrònic was associated with improved disease control, reducing the proportion of patients with HbA_1c ≥8% by 44% in the year of monitoring. Similarly, previous telemonitoring studies in patients with diabetes reported improved control of HbA_1c and fewer admissions, although with more visits to a doctor (primary care or specialist) [6,9].

The ValCrònic program also reduced the proportion of participants with poorly controlled systolic and diastolic blood pressures, by 10% and 44%, respectively. A review by Verberk et al [5] found that participants receiving telemedicine experienced a greater reduction in blood pressure than the usual-care control group. As for the use of health care resources, ValCrònic led to a 51.9% reduction in visits to the primary care emergency department and a 32.3% reduction in visits to the hospital emergency department. Emergency admissions fell by 33.2% and admissions due to worsening of the conditions being monitored fell by 23%. The literature contains encouraging examples of studies reporting that telemonitoring resulted in reduced use of sociohealth resources [24] and reduced mortality (by 24%) and readmissions (by 28%) in patients with heart failure, especially for New York Heart Association classes III and IV [25]. Telephone support only reduced mortality due to relapses, but not overall mortality, while heart rate monitoring decreased the risk of hospital admissions due to heart failure by 43% [25]. Visits to emergency services also decreased, but contact with primary care increased, and patient satisfaction and quality of life improved [7,24]. In COPD, most studies have been in people aged over 65 years and with forced expiratory volume in the first second of expiration of 27% to 43% (Global Initiative for Chronic Obstructive Lung Disease class 2-4). Although those authors reported reductions in readmissions and emergency visits, there were no differences in mortality rate, quality of life, or satisfaction [8,26,27].

The Whole Systems Demonstrator, which started in 2008, is the largest telemonitoring experience in Europe, involving 3230 patients in 179 primary care groups. Investigators have found a reduction in mortality and secondary resource consumption in telemonitored patients, but at a higher cost [28]. Giamouzis et al emphasized the need to define the profile of the patient who can benefit from the intervention and for how long [29]. ValCrònic did not enlist additional health care staff to run the program, and we included participants at high risk, often with several comorbidities, as occurs with this type of patient with chronic diseases. Unlike previous studies, this study did not focus exclusively on 1 specific disease, making this program more pragmatic and more generalizable to the realities of clinical practice.

The clinical benefits obtained show that telemonitoring of patients provides an additional benefit to the medical and nonmedical measures used to date, and thus is another option for treating patients with chronic disease. Table 2 shows that the number needed to treat associated with the program are very favorable compared with the results of other common health interventions. EMPA-REG investigators found that it was necessary to treat 39 patients with empagliflozin to prevent 1 cardiovascular event [30]; in the LEADER study, 66 patients treated with liraglutide were needed to prevent 1 major cardiovascular event [31]; and in the SUSTAIN-6 study, 45 patients treated with semaglutide were needed [32]. To prevent 1 death in the 4S study, 30 patients treated with simvastatin during 5.4 years were needed [33], and when using ramipril, 56 patients treated for 5 years were needed [34]. Thus, ValCrònic is well within the normally accepted parameters for implementation in clinical practice [35].

Regarding the opinion of the participants, a previously published report described high satisfaction among ValCrònic participants [36]. We also highlight the peace of mind reported by patients and their families, stemming from the knowledge that health professionals were remotely monitoring any changes in glycemia, blood pressure, and oxygen saturation, and that they would follow up if necessary.

The limitations of this study are mainly methodological. We dichotomized participant risk (high risk or not) using the CARS scale and, to improve its validity, we complemented the result obtained with the clinical judgement of the professional who usually treated the patient. There are numerous risk assessment scales, but their validity is usually limited, so we believed it was important that health care professionals assisted in the final selection based on their clinical experience and knowledge of the patient’s sociohealth environment.

It is possible that included participants would have been more predisposed to the use of telemedicine. In any case, the high number of participants included suggests that many elderly patients or their family members feel comfortable handling these devices. Elsewhere [36], our group described the degree of satisfaction of patients and their families with the ValCrònic program. We also conducted a survey in patients who opted not to participate, finding no substantial differences in overall satisfaction, although for very different reasons, not only a disinclination to use the health monitoring devices.

As we used a before-and-after study design, we lacked a parallel control group, and the results of this study have to be interpreted with caution. Secular trends or sudden changes might have made it difficult to attribute observed changes to the intervention. However, since it was carried out in two specific health areas, information regarding potential interventions carried out outside the program is available, and there was no interference in this regard. The fact that participants’ diseases had evolved for an additional year, and that they were a year older at study end, adds even more value to the good results obtained.

Another factor to take into account is the Hawthorne effect, or the feeling of being observed. In all intervention studies, there is a potential additional beneficial effect due to people knowing that they are participating in a study. However, given the high risk in the included participants, with more than 20% of them having been hospitalized in the previous year, it is difficult to imagine that a benefit like the one described could occur purely as a result of the Hawthorne effect.

Without a clear understanding of the economic implications of telemonitoring interventions, it will be difficult to establish informed national policies regarding reimbursement for these programs. Cost-effectiveness analyses of this program will be published.

Compared with the preintervention year, during the intervention year, people with at least one of 4 chronic diseases (hypertension, diabetes mellitus, COPD, or heart failure) at high risk for rehospitalization or an emergency department visit and who participated in the ValCrònic telemonitoring program had better weight control, reduced blood pressure and glycemia, and made fewer visits to primary care- or hospital-based emergency services due to disease exacerbation. In addition, hospitalizations due to an exacerbation of a chronic disease decreased.