This mixed methods evaluation of a web and mobile–based patient-facing platform used messaging, phone call, and email data as well as semistructured interviews to investigate the experiences of PCs and CTMs of children with medical complexity when using a secure messaging system. A mixed methods evaluation enabled us to assess the objective usage of the platform among both CTMs and PCs. It also helped us understand the subjective experiences of users and how they were able to use and incorporate secure messaging into the care of children with medical complexity. This study is a component of a larger study investigating the feasibility of a web and mobile–based patient-facing platform to improve communication and care coordination for children with medical complexity [6].

Connecting2gether (C2) is a web and mobile–based patient-facing platform developed for children with medical complexity that supports various functions, including secure messaging, health tracking, educational resources, and a shared medical summary. This platform was developed by using adult literature on secure messaging, referring to caregiver models, and customizing features of the online platform to the health care needs of children with medical complexity. The platform was accessible through desktop, tablet, and mobile devices. The secure messaging feature enabled one-to-one communication through an instant messaging interface. PCs could use secure messaging with CTMs that they invited to C2. All platform users received a push notification to their mobile devices and an email alerting them that they had received a message on C2. Attachments (ie, pictures, videos, and documents) could also be sent in C2 secure messaging. Messages could be responded to immediately, within seconds after a message was received, or days after the message was received.

Institutional research ethics approval was obtained at The Hospital for Sick Children (SickKids; 1000060804), Royal Victoria Regional Health Centre (RVH; R18-013), and Credit Valley Hospital (CVH; 973) [17]. Informed consent was obtained from all participants, and all methods were carried out in accordance with relevant guidelines and regulations. All study data were deidentified before analysis.

PCs of children with medical complexity were recruited from Complex Care Programs at SickKids, RVH, and CVH. To be eligible for the Complex Care Program, children must meet at least 1 criterion from each of the following conditions: technology dependence and/or users of high-intensity care (eg, mechanical ventilator, constant medical/nursing supervision), fragility (eg, severe/life-threatening condition, an intercurrent illness causing immediate serious health risk), chronicity (condition expected to last at least 6 more months or life expectancy less than 6 months), and complexity (involvement of at least 5 health care practitioners/teams at 3 different locations or family circumstances that impede their ability to provide day-to-day care of decision-making for a child with medical complexity) [18]. Children with medical complexity were also between 0 and 18 years of age at the time of study initiation. Purposive sampling guided parental participant selection to ensure diversity in role, communication experience, age, ethnicity, and location [19,20].

PCs were eligible to participate if they were English-speaking, had access to the internet and a computer, and were the primary caregiver of a child with medical complexity. CTMs were approached prior to recruitment to ensure it was an appropriate time to engage in research for the families (eg, hospitalization, end-of-life, or PC physical/mental health concerns).

In this study, “NPs” refers to the nurse practitioners of children with medical complexity in the Complex Care Program, and “HCPs” refers to other hospital and community–based health care providers. CTMs comprise both NPs and HCPs together.

Every PC had their assigned Complex Care Program NP on the platform. PCs were also able to invite other members of their child’s care team (eg, CTMs like social workers, patient information coordinators, pediatricians, etc) to use C2. CTMs that registered on C2 were presented with the terms of use of the platform and the study information letter. If interested, they were approached by the study research coordinator (RC) and presented with information about the research study and the opportunity to participate. CTMs that declined to participate in the research study were still able to use C2. PCs and NPs received training before registering on C2 (duration of 30 to 60 minutes), and the training presentation was later made available on C2. In addition, CTMs could set up a disclaimer on C2 if they were away or designate time slots in which they would respond to messages (eg, 8 AM to 4 PM) to aid in setting expectations with PCs.

All research study participants received remuneration for participating in the research study. PCs were given CAD $60 (US $44.59) in gift cards (CAD $20 at baseline and CAD $40 after completing the study), and HCPs that completed the end-of-study questionnaire were entered into a draw for a CAD $100 (US $74.32) gift card. Participants that completed the end-of-study semistructured interview received an additional gift card worth CAD $20 (US $14.86). C2 also had a built-in points system where PCs received a specified number of points when completing a platform activity (ie, accessing educational material). As a usage incentive, PCs received a gift card worth CAD $5 (US $3.72) when they reached predetermined point milestones. NPs also received a CAD $5 (US $3.72) gift card for every 50 messages that they sent through C2.

A total of 36 PCs of children with medical complexity registered on C2. Other platform members in the study included 7 NPs and 59 HCPs, 35% (n=21) of whom were pediatricians.

During the study period, the RC sent 1818 secure messages to all platform users. PCs sent 1133 (median 30; IQR 13-46) messages, NPs sent 615 (median 74; IQR 37-119.5) messages, and HCPs sent 113 (median 2; IQR 1-4.5) messages. Moreover, 85.45% (1257/1471) of secure messages sent during the study period were responded to within 24 hours, and 92.45% (1360/1471) within 72 hours. PCs replied fastest to HCPs and slowest to the RC. The RC had the fastest response times (16 minutes) compared to all other platform users (Table 1).

The messaging response was 84.53% (1295/1532) for all secure messages over the study period. Most secure messages from PCs (261/280, 93.2%) that required a response from the NP received a response. In messages from PCs to HCPs that required a response, 73% (45/62) received a response.

The majority (2987/3679, 81.19%) of secure messages sent on C2 were sent on weekdays, while 17.09% (n=629) of all messages were sent during evenings, and 1.71% (n=63) were sent on weekends.

Among the PCs who had phone calls with their care team during the study period (21/36, 58%), there were, on average, 4 (range 1-10) phone calls with their care team (Figure 1). PCs that sent emails to their care team during the study period sent an average of 5 (range 1-19) emails. On average, PCs sent 17 secure messages to both NPs and HCPs on C2.

Due to technical limitations of the platform, conversations on C2 could only include 2 people (ie, the parent and 1 other user). Email allowed individuals to have any number of people within each email thread. Accordingly, 73% (100/137) of the emails contained 2 participants, 19% (n=26) contained 3, 5% (n=7) contained 4, and 3% (n=4) contained 5 participants.

The use of secure messaging as a means of communication in health care is a relatively novel concept in the pediatric literature. To our knowledge, this is the first mixed methods study investigating the use and perceptions of a secure messaging system housed in a web and mobile–based patient-facing platform for PCs and CTMs of children with medical complexity. Our findings demonstrate that secure messaging was highly used, allowed for diverse topics of conversation, and enhanced the PC-CTM relationship. This paper also demonstrated the utility of secure messaging as a significant collaborative tool between CTMs and parental caregivers, as demonstrated by the large volume of messages exchanged, as an addition to existing forms of communication. In addition to medical information and administrative conversations, secure messaging has surrounded the sharing of educational resources, proactive check-ins initiated by CTMs, and general conversation about the child/family life. Our discussion of this paper will focus on 3 key areas: the clinical utility of secure messaging, the enhancement of the role of the PC with secure messaging, and the workflow and risks of secure messaging.

There is limited pediatric literature to date showing the clinical utility and role of SM; however, there has been noted enthusiasm about the exploration of this topic [11,24,25]. There is mixed evidence in the literature of how secure messaging could be used clinically. Only 1 study with children and adolescents with sickle cell disease found that messaging was used for proactive check-ins and sharing of educational information [14], whereas a study with pediatric surgical patients did not [26]. Considering the current literature, our findings suggest that children with higher care needs and chronic conditions, such as those with medical complexity and those with sickle cell disease, benefit from the use of secure messaging as it allows for proactive check-ins from their CTMs as well as continued education surrounding their medical conditions. For children with medical complexity, there is a need for constant check-ins from CTMs, and secure messaging enables this. In particular, for these children and their families, there is a clinical utility in accessible secure messaging, as it enables prompt clinical response to the high acuity issues we frequently see in this population. Our finding suggesting that secure messaging could be a modality to support parental education is especially salient as previous research has shown that PCs of children with medical complexity often wish to learn more about their child’s condition(s) and how to provide care at home [27,28].

Taken together, we suggest that all the noted benefits of secure messaging can contribute to reducing the burden of care coordination for PCs of children with medical complexity. The four defining characteristics of care coordination include (1) family-centeredness; (2) planned, proactive, and comprehensive focus; (3) promotion of self-care skills and independence; and (4) emphasis on cross-organizational relationships [4]. Indeed, improving the connection and ease of communication between PCs and CTMs increases the opportunity for partnership, family-centered care, and independence, while the finding of proactive check-ins by CTMs aligns well with a planned, proactive, and comprehensive focus.

A barrier to communication between PCs of children with medical complexity and their CTMs is a perceived lack of partnership, as PCs do not feel acknowledged for their expertise and cannot contribute to their child’s care plan [6]. PCs in our study noted that secure messaging allowed them to play a more active role in their child’s care, allowed for more informal conversations with their child’s health care team, and ultimately improved their perceived partnership with their child’s health care team. This finding is supported in adult literature; a study of women with breast cancer demonstrated that the use of an interactive communication tool allowed participants to feel more empowered to participate in their health care and improved their relationship with their health care team [29].

A common concern in the literature surrounding real-time communication and secure messaging is that patients may overwhelm CTMs by sending many messages, significantly increasing their workload [30,31]. In this study, PCs and NPs received training on the appropriate use of C2 prior to registering. Research on the use of patient portals in primary care clinics demonstrates that our efforts to provide clear and structured training addressed any concerns on how to communicate with CTMs appropriately [32]. Our research adds to the literature demonstrating that patients do not overuse secure messaging systems [33]. The findings of this paper dispel the concerns of secure messaging overuse because secure messaging did not add significantly to the workload for CTMs, and it can be constricted to weekday daytime hours.

Both PCs and CTMs noted limitations that affected their use of the secure messaging system on C2. These limitations, including additional log-in steps associated with secure messaging and difficulty identifying which communication method to use, are similar to those reported by Hsiao et al [34], who implemented a secure messaging platform in the care of pediatric patients with respiratory diseases. CTMs in our study felt that having the secure messaging system integrated with their EMR would greatly reduce any barriers to access and the additional log-in steps required. Indeed, previous research has demonstrated a high uptake of secure messaging systems when integrated into a patient portal and EMR [35]. In this study, CTMs were required to both communicate by secure messaging and copy the conversation into the patient’s chart. Secure messaging as a part of the EMR would thus lead to reduced redundancy in separate health information platforms, as all information would be housed in a single system. In addition, if they are separate systems, other people who are not using the separate platform would not have access to the secure messages that are important aspects of their child’s care. The integration of secure messaging into the EMR also enables messaging with patients to become a part of a CTM’s standard workflow and includes the messaging as a permanent record in the EMR [36]. Moreover, having secure messaging as a core function within an existing EMR would enhance health care delivery. Specifically, this would allow the provider to respond to secure messages with access to accurate and real-time information including up-to-date medication lists, upcoming appointment times, and hospitalizations or emergency department encounters. Ultimately, including secure messaging in the EMR blends well into the workflow of CTMs and leads to family-centered care, wherein all care providers and PCs have access to the same information. Future research should focus on health-related outcomes (for instance, hospitalizations and emergency department visits) associated with integrating messaging into the EMR, as well as perceptions of HCPs on how integration affected their clinical workflow.

This study has several limitations. First, most of our relatively small PC sample who used secure messaging on the web-based patient-facing platform were highly educated and had high household incomes. Therefore, our results may not be representative of families with lower education, health literacy, or income. However, previous research has shown that economically disadvantaged families still access the internet regularly and are interested in receiving hospital communication through electronic means [12]. Further, the platform was only available in English, and we did not collect PC immigration status. Using secure messaging may present different benefits and challenges to families who are new to Canada and those who do not speak English fluently. There was a possible participation bias, as PCs who chose to use the platform for 6 months and participate in the research study likely already had positive perceptions of and were comfortable using technology such as smartphones, the internet, and computers. However, we attempted to mitigate this bias in the semistructured interviews by recruiting high and low platform users. Finally, this study’s primary focus was on the evaluation of platform usage. Further analysis will include understanding the individual-level characteristics and utilization patterns of secure messaging.

Our study demonstrates that secure messaging was an effective and additive form of communication for PCs of children with medical complexity and their CTMs that had a high uptake among users. Through a web and mobile–based patient-facing platform, secure messaging provided the opportunity for continued patient education, proactive check-ins from CTMs, and casual conversations about family and child life, all of which contributed to PCs feeling improved connections with their child’s health care team. Therefore, we suggest that secure messaging can improve communication between PCs of children with medical complexity and their care team to reduce the burden of care coordination and ultimately improve the experience of care delivery. Future research should focus on a more rigorous evaluation of secure messaging in the care of children with medical complexity and how it can affect health-related outcomes, especially when integrated into the EMR.