This study has received approval from the Research Ethics Board at the University Health Network (15-9995-BE) and Mount Sinai Hospital (MSH REB 16-0093-E).

The patient inclusion and exclusion criteria are provided in Textboxes 1 and 2, respectively.

Patients will be recruited during regularly scheduled visits to the University Health Network or Mount Sinai Hospital clinics associated with their chronic illnesses. After the clinician determines that the patient is eligible for recruitment, the clinician will ask the patient if they are willing to speak to the study coordinator regarding the study. All eligible participants based on the inclusion and exclusion criteria will be asked to sign a written consent form before being enrolled in the study. After the randomization into the intervention or control groups, the study coordinator will provide the patients in the intervention group and, if appropriate, their caregivers with training on the telemonitoring system. Patients will be compensated Can $24 for their participation in the study, which will cover the cost of parking for one clinic visit.

To perform the randomization, the Web-based computer-generated randomization tool, Research Randomizer (Social Psychology Network, USA, [23]) will be used to generate the sequence of intervention and control group allocation. The study coordinator performing the recruitment will be blinded until the patient has consented to participate. After obtaining consent, the study coordinator will take the top card of the prepared stack of randomization cards and will tell the patient if they are in the intervention or control group. The participants will be block randomized (blocks of 4) and stratified into groups according to their primary condition (HF, COPD, hypertension, and diabetes). The HF group will further be stratified on the basis of New York Heart Association (NYHA) classification (NYHA class 2-3, NYHA class 4), the hypertensive patients will be stratified to those with and without diabetes, and the hypertensive patients will also be stratified to those with and without CKD. The stratification of the hypertensive patients is performed because the coordination of clinical management for the combination of hypertension and CKD and of hypertension and diabetes was established between specialists who were willing to participate in this study.

In addition to asking patients whether they are interested in participating in the study during their regularly scheduled clinic visit, an additional recruitment strategy may be employed to speed up patient enrolment. The research study coordinator, an independent third party who is not part of the clinical care team, will generate an eligible patient mail-out list. This list will be derived from the patient roster at the clinics. The list will be screened and verified by the clinic nurses/physicians to ensure appropriateness as determined by the study inclusion criteria. The research coordinator will mail-out an invitation letter including a study description and consent form. The study coordinator will arrange a time to discuss in person or over the phone with patients interested in learning more about the study.

The primary outcome measure SF-36 was used to conduct a sample size calculation. Assuming a moderate effect size on the general health dimension of SF-36 of 10.5 points [24], a score of 57±21 for the control group [25], 80% power, and Cronbach alpha=.05 (two-sided), the sample size per group was calculated to be 63. Assuming that 15% of patients will be lost to follow-up (including mortality) [26,27], the sample size per group is 73. Hence, 146 patients will be enrolled during the trial and randomized 1:1 into control and telemonitoring groups.

The telemonitoring system, named Medly, will enable patients with complex chronic illnesses, including MCCs, to take clinically relevant physiological measurements with wireless home medical devices and to answer symptom questions on the smartphone (Figure 1). The measurements will be automatically and wirelessly transmitted to the smartphone and then to a data server. Specifically, patients with HF will monitor daily weight (A&D Medical Bluetooth weighing scales), blood pressure/heart rate (A&D Medical Bluetooth blood pressure monitors), and symptoms; patients with COPD will provide self-reported symptoms; hypertensive patients will monitor their blood pressure; patients with diabetes will monitor blood sugar levels (iBGStar Bluetooth blood glucose meters); and patients with CKD will monitor blood pressure. Automated self-care instructions or messages that have been carefully developed with health care specialists will be sent to the patient based on the readings and reported symptoms [28]. If there are signs of deteriorating health of a patient, an alert will be sent to a clinician responsible for the particular chronic condition of concern in the respective specialty clinic(s) as appropriate. All the relevant patient data will be sent to the clinicians via an email alert, and they will be able to access historical and trending data of their patients through a secure Web portal.

The control group will receive usual care, and the intervention group will receive usual care plus the telemonitoring intervention. Both groups will be followed for 6 months. Each intervention study participant will be provided with a smartphone. This phone will be loaded with the app and a limited data plan to enable data transfer. Patients will also be provided with relevant home medical monitoring devices (eg, weighing scales and blood pressure monitors for patients with HF, and blood pressure monitors for hypertensive patients). Patients will be asked to take measurements and record symptoms at a specified frequency at their homes. Home measurement frequency depends on the specific chronic condition(s). Patients with HF and COPD are generally asked to take daily measurements, whereas patients with hypertension and CKD take blood pressure readings several times per week as specified by the algorithm developed in conjunction with specialists. Patients with diabetes will be asked to follow the frequency of blood glucose measurements as recommended by their endocrinologist. Patients will be notified when they should take measurements at home through the smartphone app. Patient-specific baseline information will be entered into the telemonitoring system dashboard, such as an individual target weight range, target blood pressure range, medication list, and COPD action plan, before the patient starts using Medly at home.

The primary outcome will be health status as measured with SF-36 because this metric will be relevant to all chronic conditions under investigation. A secondary outcome measure will be the cost (from the health system and patient perspectives). The cost of health care will include hospitalizations, emergency department (ED) visits, clinic visits, medications, and the telemonitoring program as relevant to the different chronic illnesses. Disease-specific subanalyses on cost will also be conducted.

Other secondary outcome measures will include combined hospitalization for any reason or death from any cause within 6 months after enrolment, hospitalization, mortality, days in hospital, number of ED and clinic visits, and medications. Secondary outcome measures will also include general quality of life as measured by the EuroQol 5D-RL, anxiety and depression as measured with the Hospital Anxiety and Depression Scale [29], and patients’ self-efficacy to self-manage their disease as measured with the Self-Efficacy for Managing Chronic Disease Scale [30]. See Table 1 for the assessment schedule.

The disease-specific outcome measures (generally obtained as standard of care) are listed below:

Number of hospitalizations, days in hospital, number of ED and clinic visits, and medications will be determined through the hospital electronic medical records and a manual chart review of all participants’ clinical records. Number of hospitalizations, days in hospital, and ED visits will be verified by patient participants through self-report with the help of a questionnaire. The cost of the intervention will be tracked, including equipment costs and human resources for clinical support, technical support, and program management. Standard cost values for a day in hospital, ED visit, etc will be used to estimate net cost savings or expenditures. Values for pre- and posttrial health outcome measures (eg, blood test results) will be obtained through hospital clinical records or through the Medly database for symptoms. Pre- and follow-up questionnaires (baseline, 1-month, and 6-months) will be administered to both control and intervention groups that contain the validated survey tools listed above. Participants will be able to complete the questionnaires at the clinic or complete it at home. Participants who choose to complete the questionnaire at home will be given a prepaid self-addressed envelope to mail the questionnaire back to the study team. The 1-month questionnaire will be given to participants upon enrolment with instructions to complete the questionnaire 1 month after starting the trial and to mail it back to the study team using the prepaid envelope. The study coordinator will call each participant in the 1-month period to remind participants to complete the questionnaire. Usage data, in terms of adherence to taking measurements and use of the different features, will also be data mined from the servers.

In addition to the quantitative metrics, all participants will be asked basic questions such as their self-care practice and their thoughts on telemonitoring at the start of the study. A subgroup of the intervention arm will also be interviewed individually to assess their experiences and perceptions regarding the use of Medly at the end of the study. The number of patients that will be interviewed will depend on when saturation of information is reached, typically 15-20 participants. The interviews will be conducted in a quiet and private space within a clinic (eg, consultation room or education room). Before the interview, participants will be informed that notes will be taken and they will be audiotaped for data analysis. The clinicians involved in the trial will also be asked to participate in interviews to determine their perceptions of Medly.

For the primary analyses, posttrial data and change scores will be compared between the control and intervention groups using independent Student t test and Mann–Whitney test (for normally and not normally distributed data, respectively) based on intention-to-treat. Paired Student t test and Wilcoxon signed rank test will be performed to compare baseline and poststudy data within the control and telemonitoring groups. Secondary analyses seeking to assess the impact of the telemonitoring system over time by incorporating the 1-month time point will be analyzed using general linear mixed model procedures. Randomization of study participants should yield an equal distribution of possible confounding variables. However, adjustments may be required for unexpected variations in the baseline characteristics between groups (eg, age and gender). Similarly, depending on the combinations of diseases of the participants, subgroup analyses may be performed. All statistical analyses will be performed using the statistical software app SPSS (IBM Corporation, USA).

The cost evaluation will be performed by comparing any savings through reductions in hospitalization, emergency, and clinic visits through the use of the intervention. This will further include amortizing the equipment costs and factoring in additional clinical, technical, and management resources. An analysis of indirect costs will also be performed, including such factors such as days absent from work.

Interview data will be analyzed using a conventional content analysis approach [37]. Two researchers will analyze the transcripts independently and will discuss the themes until a consensus is reached. The software program NVivo will be used to help organize the themes (QSR International, Doncaster, Victoria, Australia).

This RCT aims to investigate the impact of a smartphone-based telemonitoring system on patients with complex chronic conditions, including those with MCCs. The mixed methods design of the trial will enable triangulation from a variety of data sources, including questionnaires, chart reviews, and interviews.

Evidence from previous trials indicate that positive health outcomes are achievable from telemonitoring, particularly for patients who have the most severe conditions. However, very few clinical trials have been conducted to investigate the impact of telemonitoring across MCCs [17-22]. This is largely because of the following three reasons:

First, it is difficult to develop a telemonitoring platform that can be tailored to effectively manage several combinations of chronic conditions because of the disparate needs of different chronic conditions. We have used our experience of over 10 years of designing, developing, and evaluating apps for chronic disease management using a user-centred design process to develop the Medly platform.

Second, determining appropriate outcome measures that are appropriate across MCCs is difficult because many metrics are disease specific. For example, reduction in hospitalization may be the primary goal of telemonitoring of patients with HF but would not be a suitable measure for chronic diseases such as hypertension. Therefore, health status as measured by SF-36 was chosen as the primary outcome measure because it is relevant to all interested chronic conditions.

Third, performing a trial of patients with MCCs requires coordination of several specialities that are typically not integrated and do not have ideal communication channels between them. For our clinical trial, we have clinical specialists as part of the research team who will be managing the patient participants. These clinical specialists are very familiar with the telemonitoring system and have already worked together in the past to coordinate care for certain complex patients.

Patients with multiple chronic illnesses account for the largest expenditure of health care dollars and require the most support to manage their complex conditions. This RCT will be among the first to provide evidence of the impact of telemonitoring on costs and health outcomes of complex patients who may have MCCs. We anticipate that our health app for the management of MCCs will be a cost-effective and scalable tool that will improve health outcomes and quality of life, while empowering and reassuring patients and their informal caregivers.