Engaging Adolescent Heart Transplant Recipients Through In-App Messaging During Mobile Health Intervention

Abstract

Introduction: Adolescents heart transplant recipients experience difficulty with adherence to immunosuppressive medication leading to increased risk of organ rejection, hospitalization, and mortality. Few interventions have been successful to promote medication adherence in adolescent heart transplant patients as most fail to engage the patient in the behavioral change process and support patient–provider communication. The purpose of this study was to explore the nature and degree of in-app communication between adolescent heart transplant recipients and nursing staff during an asynchronous mobile video directly observed therapy intervention. Methods: A content analysis of 894 in-app messages exchanged between 10 adolescent patients and nurses during a 12-week mobile-based intervention was conducted. Two authors used an inductive, iterative process to guide a thematic analysis of the asynchronous in-app messages with high interrater reliability ranging from 81.5% to 100%. Findings: In-app messages fell under 3 broad content categories: (1) interpersonal support and rapport-building, (2) medically related questions and information, or (3) functional information about the intervention and the mHealth app. Results demonstrated the degree to which interpersonal engagement occurred during the intervention, the nature of these exchanges, and their relationship to medication adherence. Conclusions: The study provided insights into the feasibility and benefits of 2-way communication features of the directly observed therapy intervention in promoting engagement and in improving medication adherence among adolescent heart transplant patients. Continued research and clinical focus on patient engagement and impactful aspects of interpersonal communication could aid in the translation of this intervention into standard clinical care at pediatric transplant centers.

Keywords

pediatric organ transplantation medication adherence mHealth adolescent messaging patient engagement

Introduction

While survival rates after receiving an organ transplant continue to improve for pediatric patients, concerns remain with rates of organ rejection and hospitalization for these patients.^1–3 Medication nonadherence is one of the most significant predictors of late acute rejection, number and frequency of hospitalizations, and even mortality.^1,2 The challenges with and consequences of medication nonadherence experienced by adolescent transplant patients are well-documented.^2,3 Individual studies have reported nonadherence among adolescent transplant recipients as high as 40% to 60%.^2,4

Researchers and healthcare providers have increasingly called for greater use of technology and improvement in mHealth approaches with pediatric transplant patients.^5,6 Metaanalyses addressing efforts to promote adherence in pediatric care have emphasized the need to tailor intervention efforts to the patient, improve patient engagement, provide interpersonal support, and address multiple identified barriers to adherence.^7–10 Critical features facilitate real-time monitoring and observation of medication-taking behavior and include tools for improving patients’ organizational skills, planning, and conscientiousness around the behavior.^10,11

Despite the potential of mHealth adherence interventions, low patient engagement continues to be a significant challenge.^8,9,12 Engagement is a multidimensional concept including behaviors synonymous with system usage as well as psychological processes such as perceived depth of involvement in the behavior change process.¹³ Short et al¹⁴ emphasized microlevel engagement (ie, use of system features and level of user interest and attention when completing activities) as important as macrolevel engagement (ie, motivation for changing behavior) in improving the ability of mHealth interventions to promote behavioral changes. Baysari and Westbrook¹⁵ underlined that mHealth apps must support patients’ active involvement in the coordination of their care. mHealth apps that provide 2-way communication features can improve patients’ reflective motivation, psychological capability, and social opportunities and are, thus, more likely to improve patients’ intentionality to adhere to their regimens.¹⁶

A mobile health application developed by Scene Health Inc. embodies these system requirements to support medication adherence by facilitating an asynchronous mobile video directly observed therapy intervention.¹⁷ This mHealth app also allows patients to record and submit dose-by-dose videos of themselves completing treatment administration at home (eg, taking immunosuppressant medication). Videos are reviewed and approved by a healthcare professional (eg, transplant nurse) through a centralized platform asynchronously. Patients can review and track their progress over time, report any symptoms of illness or side effects, message the reviewing nurse, or communicate with their transplant team. Study staff and nurses can use the app to contact patients, caregivers, and the patient's transplant care team with concerns. The app includes features to support a higher level of engagement via 2-way communication features that, in turn, can improve patient participation in her/his care coordination, and increase their self-efficacy with their treatment regimen.

The directly observed mHealth intervention was successfully implemented and piloted with adolescent heart transplant patients from a children's hospital in the southeastern United States, wherein patients and their caregivers reported high levels of intervention acceptability^18,19 Ten adolescent transplant recipients recruited from a large pediatric heart transplant program received the mHealth intervention with 8 completing the 12-week protocol. Specifically, the study involved adolescent heart transplant recipients who exhibited poor medication adherence. The current study was a secondary qualitative analysis that sought to explore the nature and degree of interactions between adolescent heart transplant recipients and nursing staff during the intervention.

Methods

Design

The study was a content analysis of adolescent heart transplant recipients’ and nursing staff's engagement during the prospective, single-group pilot of the 12-week asynchronous directly observed mHealth intervention.¹⁹ The University Institutional Review Board reviewed and approved the study (Study #IRB202000695).

Setting

The study took place in a large pediatric heart transplant program in the southeastern United States. Participants made no research-specific clinic visits during recruitment or enrollment due to concerns with COVID-19 and conducting in-person contact with immunosuppressed patients. All study procedures with patients were accomplished virtually or by telephone. When beginning the intervention and setup of the mobile application, staff from Scene Health Inc. assisted patients with download, installation, and feature demonstration on their mobile devices. Patients and parents completed pre and postintervention interviews including completing measures of key study outcomes. A flow chart detailing the enrollment, training, measurement points, and staff responsibilities during the intervention has been previously published.¹⁸

Sampling

A convenience sampling strategy was used to recruit adolescents ages 11 to 21 years from the transplant clinic that had documented evidence or concern for medication nonadherence during routine, outpatient clinic visits. Recipients who had received a heart transplant more than 6 months prior and were otherwise deemed medically stable by the pediatric transplant team were eligible to participate. Problems with medication adherence were defined as medication level variability index¹ values ≥ 2.0 or by reported concerns by the pediatric transplant care team. Non-English-speaking participants were not included due to the communicative features of the directly observed mHealth app and the availability of only English-speaking nursing staff.

Information Collection

The content, purpose, and context of 894 in-app messages exchanged between 10 patients and reviewing nurses were analyzed using content analysis.²⁰ The first 2 authors developed a codebook a priori that included initial codes based on the Persuasive Systems Design model and other relevant theories and research. They coded the data independently, selecting up to 3 codes for each text since messages between patients and nurses could contain several sentences and sentiments. The authors used an inductive, iterative process for coding the data and creating additional codes to guide a thematic analysis of the full dataset of 894 in-app messages. Disagreements in coding were resolved through iterative discussions. The percentage of agreement was calculated after the coding period with interrater reliability ranging from 81.5% to 100%.

Data Analysis

After finalizing the coding, thematic patterns in the data were examined using SPSS v27.0 (IBM Corp Armonk, NY). Additionally, descriptive statistics were calculated for the number and frequency of messages and resulting content categories. Tests of association were conducted between message sources and content code (Fisher’s exact test and effect size phi, ϕ) and associations between message content categories and intervention outcomes (Spearman's rho, ρ).

Findings

Ten patients initiated the mobile directly observed therapy ( Table 1 ), 2 who ceased participation and became nonrespondent to the reviewing nurses around the third week of the intervention. These 10 patients were majority female, white, had a mean age of 16.60 (.95) years, had private insurance, were hospitalized at the time of the transplant procedure, and 80% had current evidence of mild to severe coronary allograft vasculopathy (CAV), the major pathology of organ rejection in heart transplant recipients.

Table 1.

Patient and Parent Characteristics.

	Total, n = 10
Patient and parent characteristics	Mean (SD)
Patient age, years	16.60 (2.95)
Age range (min.–max.)	12-21
Parent age, years	39.29 (3.35)
MLVI values at intervention enrollment	2.85 (1.83)
Median	2.20
Range (min.–max.)	0.95-6.17
MLVI > 2.0, n (%)	6 (60.0)
	N (%)
Gender, female	6 (60.0)
Race, minority	2 (20.0)
Type of insurance at transplant
Medicaid	4 (40.0)
Private	6 (60.0)
Patient hospitalized at transplant	8 (80.0)
Patient in ICU at transplant	4 (40.0)
Coronary allograft vasculopathy
None	2 (20.0)
Mild	6 (60.0)
Moderate to severe	2 (20.0)

Abbreviations: MLVI, medication level variability index; ICU, intensive care unit.

Interpersonal Engagement via In-App Messaging

Ten patients and reviewing nurses sent or received 894 unique messages during the 12-week mHealth intervention (see Table 2 ). Of these, 227 (25.4%) were sent by the patients and 613 by the nurses (68.6%). The remaining 54 in-app messages (6.0%) were sent by Scene Health on-boarding staff to assist patients with the mHealth app setup during days 1 to 2 of the 12-week intervention. The mean number of exchanged messages was 89.30 (17.96) between patients and nurses with 27.50 (10.23) sent by the patients and 61.30 (16.12) sent by the nurses. Patients sent messages most often during the evenings with 108 messages (47.6% of 227) sent between 8 PM and 1 AM. Patients sent a smaller proportion of messages during the mornings with 64 messages between 9 AM and 12 PM (29.5% of 227). The distribution of times coincided with patients’ dosing schedules and interactions with the application including submission of videos.

Table 2.

Messaging Characteristics of the 894 In-App Messages.

	Total, N = 894
Messaging characteristics	Mean (SD)
Number of unique days with messages per patient	48.20 (11.63)
Range of messages sent per patient, days	33-70
	N (%)
Messages sent, total, n	894
Sent by patient	227 (25.4)
Sent by nurse	613 (68.6)
Sent by scene health staff	54 (6.0)
Proportion of days in which message was sent when patient was active in mobile directly observed therapy
Sent by patient	15.4%
Sent overall	65.6%
Timing of messages sent by patient
1:00 AM-8:59 AM	8 (3.5)
9:00 AM-11:59 AM	64 (28.2)
12:00 PM-4:59 PM	36 (15.9)
5:00 PM-7:59 PM	11 (4.8)
8:00 PM-12:59 AM	108 (47.6)

Associations between patient characteristics and message codes and frequencies were examined. The total number of messages and the number sent by patients did not significantly correlate with patient age, preintervention medication level variability index, and time since transplant (all P > .05). The number of messages exchanged and those sent by patients did not significantly differ by patient gender, minority status, insurance status at the time of transplant (public vs private medical insurance), and both hospitalization and ICU medical statuses at time of transplant (all P > .05).

Nature and Content of In-App Messages

The content analysis identified 3 broad categories: (1) interpersonal support and rapport-building, (2) medically related questions and information, or (3) supportive information about the mHealth intervention and functional aspects of the Scene Health app. Descriptions of the text exchanges and examples are presented in Table 3 .

Table 3.

Message Content Codes per Category With Examples.

Category and code	Example quotes, singular, that is, not drawn from a continuous conversation
Interpersonal rapport and support
Praise	Nurse: “Keep up the great work taking your medicine to be healthier.” Nurse: “Thank you for sending videos. Great work!”
Rapport building	Nurse: “Way to go taking your medicines today! Have a great day :)” Nurse: “Have a great weekend!” Nurse: “What are your plans for the thanksgiving? I can see that your household would be busy with you and folks cooking and baking for tonight.” Patient: “I did have a fun even though I could not bowl it was fun to see some friends again”
Empathy	Nurse: “Good morning, [patient]. Are you feeling better? … Please recover fully. We are sending you positive vibes.” (patient reported being ill in previous videos) Nurse: “I really do not know your situation but I am here to listen if you need one.”
Gratitude	Nurse: “Thank you!” Patient: “Thanks ”
Medically-related
Medical problems	Patient: “I did not video myself last night or this morning because I was in so much pain I could not even move” Patient: “I’ve been sick for a few days now, with a cold I think.” Patient: “Yes, I read that about [medication], raising the blood pressure. I think I’m just having withdrawals, and I will try to just wait until the [next appointment].” Patient: “but the reason why I didn’t take my medicine earlier is because I’m not feeling good if I would’ve took the pills I would have probably threw them up”
Patient education	Nurse: “Looking at your videos, it seemed that you are having a hard time swallowing pills. Have you tried other means of taking the pills like using a straw?” Nurse: “I will suggest that you get a travel size pill storage case so you can take your medications while you are out of the house.”
DOT intervention-related
Requesting communication	Nurse: “Hi [patient], what is going on? Are you ok, you have not submitted any videos today. Let me know if you are having any problems. Thanks.” Nurse: “…. please let me know if you are still having side effects.”
Feedback	Nurse: “Please continue the great work you are doing. We are still aiming for 100% adherence rate.” Nurse: “… 30 days and counting!” (patient took medication twice a day consistently for 30 days)
Functional Information Exchange	Nurse: “All you need to do is to capture the capsules in the cup for maybe 5 seconds.” (instructions for showing medications on video) Nurse: “Please do not forget to send your videos.” Nurse: “Please continue to take your daily medications and communicate with your care team.”
Questions or request for information	Nurse: “Hey [patient], are you OK? What happened with this morning dose?” Patient: “Do I need to say what my pills are each time?”
Information exchange or answers	Patient: “Oh I am sorry I forgot to mention I added [allergy medication] to my night pills” Nurse: “I made the adjustments to reflect 1 mg [medication] in the morning and 1 mg [medication] for the evening dose. Thank you!” Patient: “It is all good I forgot to let you know”

Interpersonal support and rapport-building messages included Praise or complimentary statements for something done well. Rapport Building referred to friendly, medically unrelated messages such as jokes, or stories about family members and plans around holidays and weekends. These also included messages from nurses expressing empathy in response to patients’ sharing personal challenges or medical difficulties. These messages were coded as Empathy. Lastly, simple messages like those acknowledging a prior message or a simple thank you were coded as messages of Gratitude.

Medically related messages referred to the exchange of medical information or nurses’ attempts to promote overall healthy behaviors or increase patient knowledge of their medication condition. Medical Problems were evident in messages containing reporting of symptoms or other medically related patient experiences including side-effects of medications. Nurses also offered information or advice about beneficial health behaviors and ways to reduce specific barriers to medication adherence. Patient Education messages included suggestions that patients monitor when they needed a prescription refill, and consider different medication organization tools or practice methods to reduce ingestion issues when taking medication.

In-app messages were also sent to assist the patients through the mHealth intervention and to provide functional information about the intervention. Requesting Information were messages containing requests for information directly. Feedback included messages acknowledging when patients completed directly observed therapy-related intervention activities. Examples included the nurses telling patients their adherence rate or that a specific medication was seen in the video and counted. Functional Information were messages informing the patient about the intervention process including the use of the app, intervention procedures, and reminders to continue taking the medication and demonstrate this behavior via video. Messages sometimes contained questions about procedures, the app, and the intervention that were coded as Questions. The messages that followed answers to these questions were considered Information Exchange.

When coding messages, initial interrater reliability was determined through agreement and Cohen's κ ( Table 4 ). Across all possible categories of codes, agreement ranged from 81.5% to 99.2% with total agreement across all messages being 56.2% (894 messages each with 3 possible codes). Codes with the highest agreement were Medical Problems (N = 888, 99.2%), Patient Education (N = 865, 96.5%), and Requesting Communication (N = 859, 96.1%). Cohen's κ across all codes ranged from .012 to .843. Gratitude was the exception with 100% agreement between raters given the definition of the code was limited to simple thank you or similar wording.

Table 4.

Message Content Codes per Category, Inter-Rater Reliability, Agreement, and Proportion Sent by Groups, N = 894.

Qualitative code	Total N = 894	Cohen's κ	Rater agreement	Sent by patients N = 227	Sent by nurse N = 613	Sent by on-boarding team N = 54
Interpersonal rapport and support	N (%)		N (%)	N (%)	N (%)	N (%)
Praise	210 (23.5%)	.843	94.5%	0 (0.0%)	197 (32.1%)	13 (24.1%)
Rapport building	318 (35.6%)	.598	81.5%	73 (32.2%)	241 (39.3%)	4 (7.4%)
Empathy	37 (4.1%)	.291	93.6%	0 (0.0%)	37 (6.0%)	0 (0.0%)
Gratitude	146 (16.3%)	1.00	100%	40 (17.6%)	97 (15.8%)	9 (16.7%)

Medically related
Medical problems	6 (0.7%)	.585	99.2%	6 (2.6%)	0 (0.0%)	0 (0.0%)
Patient education	39 (4.4%)	.396	96.8%	0 (0.0%)	37 (3.7%)	2 (3.7%)

Intervention related
Requesting communication	15 (1.7%)	.139	96.1%	1 (0.4%)	12 (2.0%)	2 (3.7%)
Feedback	174 (19.5%)	.012	84.5%	0 (0.0%)	157 (25.6%)	17 (31.5%)
Functional information and protocol	176 (19.7%)	.668	89.4%	0 (0.0%)	165 (26.9%)	11 (20.4%)
Questions or request for information	174 (19.5%)	.827	94.7%	13 (5.7%)	160 (26.1%)	1 (1.9%)
Information exchange or answers	218 (24.4%)	.610	84.7%	160 (70.5%)	56 (9.1%)	2 (3.7%)

Message Content by Patients and Nursing Staff

The most common themes identified in messages (N = 894) across the patients and nurses were rapport-building (N = 318, 35.6%), praise (N = 210, 23.5%), and information exchange (N = 218, 24.4%). Patients most often sent messages providing answers to questions or information (N = 160, 70.5%), rapport-building messages (N = 73, 32.2%), and messages of gratitude (N = 40, 17.6%). Patients were significantly more likely to send messages about medical problems (P < .001, ϕ = .139), and to provide information (P < .001, ϕ = .632). Importantly, patients and nurses were similarly likely to send messages coded as requesting communication (P = .204, ϕ = .055), expressing gratitude (P = .529, ϕ = .022), and building rapport (P = .065, ϕ = .066).

The nature of messages from nurses (N = 613) were most often related to rapport-building (N = 241, 39.3%), giving praise (N = 197, 32.1%), providing functional information (N = 165, 26.9%), asking questions or requesting information (N = 160, 26.1%), and providing feedback to patients (N = 157, 25.6%). Results from Fisher's exact tests indicated that nurses were significantly more likely to send the following types of messages: praise (P < .001, ϕ = .337), feedback (P < .001, ϕ = .292), education (P < .001, ϕ = .131), functional information (P < .001, ϕ = .301), questions (P <.001, ϕ = .224), and expressions of empathy (P < .001, ϕ = .131).

Mobile Directly Observed Therapy Adherence and In-App Message Content

The proportion of messages containing each coded theme was calculated as patients varied in the number of messages sent and received. Among the 10 patients, the number of escalations per the protocol were significantly associated with the proportion of the messages that were requests for information (ρ = .632, P = .05) and questions (ρ = 855, P = .002). Both types of messages were required by the escalation protocols, including reminders sent to patients when multiple video submissions demonstrating medication taking were not completed. Conversely, greater medication adherence observed during the intervention was associated with greater exchange of praise (ρ = .604, P = .065, moderate effect size although not statistically significant) and gratitude messages (ρ = .728, P = .017). Adherence rates as observed in the videos were also negatively correlated with the proportion of messages related to requests for communication (ρ = −.824, P = .003), functional information (ρ = −.742, P = .014), questions (ρ = −.884, P < .001). Given the very small sample of patients, no other correlations were significant or had only a moderate effect size (ρ < 0.500, P > .10).

Discussion

The content analysis of in-app messages exchanged between patients and nursing staff offers an in-depth and rich exploration of an innovative 2-way communication feature of the intervention and demonstrated the degree to which interpersonal engagement occurred during the intervention, the nature of these exchanges, and its relationship to medication adherence.

This mHealth intervention^18,19 was unique because it included features and functionalities that target multiple and modifiable risk factors for nonadherence and promote sustained behavioral change.^9,12,21 Previous studies involving mHealth interventions for pediatric transplant patients have mainly focused on the delivery of notifications as reminders to take medications, often in the form of an alarm on a mobile phone or other smart device.²¹ These interventions offer limited engagement and produce little to promote health behaviors and medication adherence. Unlike 1-way notification-based interventions that narrowly focus on reminding patients to take their medications, apps that provide 2-way communication can improve patients’ motivations, self-efficacy, and intentionality in adhering to their medication regimens.¹⁶ Previous research has identified difficulties with or anxiety around communication with transplant teams as a possible barrier to medication adherence for adolescent transplant recipients.²² In line with this, other research has shown that providing patients with a means to communicate with their care team, leads to better perceptions of and better satisfaction with doctor communication, and, consequently, improved adherence.¹⁶ Research has also shown that interventions that facilitate patient–provider communication can enhance patients’ perceptions of collaboration, empathy, and therapeutic alliance and, thus, help strengthen medication adherence.²³

The mobile directly observed therapy mHealth app provides system features that constitute key dimensions in the Persuasive Systems Design model²⁴ that are critical for improving engagement, and for facilitating behavior change. In addition to primary task and credibility support, the mHealth app includes interactive dialog and social support features that motivate users to engage in and monitor their medication-taking behavior. Combined with direct observation of medication-taking behaviors, in-app messaging facilitated greater patient interaction with transplant care team members, enabled the exchange of more personalized feedback to patients, provided an active path for encouraging positive behaviors, and was possibly key in helping patients improve organizational skills and conscientiousness around their medication. Patients often reported medical problems through the app as they experienced them, indicating a level of comfort in communicating with their medical staff. A significant proportion of the in-app messages contained instances of rapport-building language, praise, and exchanges of medically important information. Nurses often provided praise, and feedback, and asked important questions to patients as prompts for taking their medication.

The volume of and popularity of in-app messaging implies positive perceptions about the 2-way communication and support features of the mHealth app. This finding was further bolstered by the previously reported results of the pilot study¹⁹ that showed patients who completed the 12-week protocol submitted 1211 videos of a possible 1344—demonstrating a 90.1% adherence rate. This adherence rate would have been approximately 82.9% if the videos not submitted by the 2 patients who dropped out from the study prematurely in the third week had been counted.

Limitations in this content analysis, the data, and the overall study should also be considered. Given the purpose of the original pilot study, the sample of patients was small and from a single pediatric transplant center, thus leaving open the possibility of considerable sampling error and lack of generalizability of results. Much communication from the patient to the nurses occurred within videos submitted as evidence of medication adherence, and these were not coded. Many of the messages from the nurses may have been in response to patient communication contained in these videos. There was no control group so comparisons with patients not provided messaging could not be made. The current analysis focused on messaging via the mHealth app and potential causal relationships between in-app messaging and adherence, video submission rates, and patient factors were not examined. Future research should and will include a more robust trial of mobile directly observed therapy that would include a control group to further explore the role of 2-way messaging in adherence outcomes among adolescent heart transplant recipients.

Conclusions

The current study has a number of implications for research on the promotion of adherence health behaviors and engagement of adolescent heart transplant recipients within mHealth interventions. The mobile directly observed therapy intervention and mHealth app included features and functionalities fostering patient engagement and interpersonal support while addressing multiple risk factors for nonadherence potentially improving clinical outcomes. In-app messages and exchanges demonstrated this and the ability of a transplant care team to implement a mHealth intervention that facilitated patient engagement, interaction, and encouragement. To improve scalability, content identified in the messages and exchanges can be used to create a database of predeveloped messages. Medical staff using the provider-facing platform to review patient-submitted videos can respond using prerecorded messages from the patient's own transplant team or other individuals within the patient's social support system. Asynchronous and centralized review can thus reduce providers’ time and effort in engaging and encouraging patient adherence and health behaviors. Continued research and clinical focus on patient engagement and impactful aspects of interpersonal communication could aid in the translation of this intervention into standard clinical care at pediatric transplant centers. Future mixed-methods research should examine specific patient engagement factors and patient perceptions within the mHealth approaches. All these efforts may lead to a positive impact on medication nonadherence especially among high-risk adolescent populations. Overall, this content analysis and exploration of patient engagement in a mHealth intervention provided insights regarding patient engagement and promotion of medication adherence in a sample of high-risk adolescent heart transplant recipients.

Footnotes

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Florida State University Office of the Vice President of Research COVID-19 Pandemic Funding.

Declaration of Conflicting of Interest

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Drs Gupta and Killian serve on the Scientific Advisory Board for Scene Health Inc. The other authors declared no potential conflicts of interest with the research, authorship, and/or publication of this article.

ORCID iD

Michael O. Killian

References

Shemesh

Bucuvalas

Anand

, et al. The Medication Level Variability Index (MLVI) predicts poor liver transplant outcomes: A prospective multi-site study. Am J Transplant. 2017;17(10):2668‐2678. doi:10.1111/ajt.14276

Oliva

Singh

Gauvreau

Vanderpluym

Bastardi

Almond

. Impact of medication non-adherence on survival after pediatric heart transplantation in the USA. J Heart Lung Transplant. 2013;32(9):881‐888. doi:10.1016/j.healun.2013.03.008

Killian

Schuman

Mayersohn

Triplett

. Psychosocial predictors of medication non-adherence in pediatric organ transplantation: A systematic review. Pediatr Transplant. 2018;22(4):e13188. doi:10.1111/petr.13188

Eaton

Gutierrez-Colina

Quast

, et al. Multimethod assessment of medication nonadherence and barriers in adolescents and young adults with solid organ transplants. J Pediatr Psychol. 2018;43(7):789‐799. doi:10.1093/jpepsy/jsy016

Foster

Pai

Zelikovsky

, et al. A randomized trial of a multicomponent intervention to promote medication adherence: The Teen Adherence in Kidney Transplant Effectiveness of Intervention Trial (TAKE-IT). Am J Kidney Dis. 2018;72(1):30‐41. doi:10.1053/j.ajkd.2017.12.012

Shah

Badawy

. Telemedicine in pediatrics: Systematic review of randomized controlled trials. JMIR Pediatr Parent. 2021;4(1):e22696. doi: 10.2196/22696

Duncan

Annunziato

Dunphy

LaPointe Rudow

Shneider

Shemesh

. A systematic review of immunosuppressant adherence interventions in transplant recipients: Decoding the streetlight effect. Pediatr Transplant. 2018;22(1):10.1111/petr.13086. doi:10.1111/petr.13086

Wyatt

Bayless

Krauskopf

Gaylord

. Using mHealth applications to promote self-managed health behaviors among teens. J Pediatr Nurs. 2021;59:164‐172. doi: 10.1016/j.pedn.2021.04.025

Badawy

Radovic

. Digital approaches to remote pediatric health care delivery during the COVID-19 pandemic: Existing evidence and a call for further research. JMIR Pediatr Parent. 2020;3(1):e20049. doi:10.2196/20049

10.

Maldonado

West-Thielke

Dew

, et al. Meeting report: Consensus recommendations for a research agenda to address immunosuppressant nonadherence in organ transplantation. Clin Transplant. 2018;32(9):e13362. doi: 10.1111/ctr.13362

11.

Duettmann

Naik

Zukunft

, et al. eHealth in transplantation. Transpl Int. 2021;34(1):16‐26. doi: 10.1111/tri.13778

12.

Geramita

DeVito Dabbs

DiMartini

, et al. Impact of a mobile health intervention on long-term nonadherence after lung transplantation: Follow-up after a randomized controlled trial. Transplantation. 2020;104(3):640‐651. doi:10.1097/TP.0000000000002872

13.

Yardley

Spring

Riper

, et al. Understanding and promoting effective engagement with digital behavior change interventions. Am J Prev Med. 2016;51(5):833‐842. doi: 10.1016/j.amepre.2016.06.015

14.

Short

DeSmet

Woods

, et al. Measuring engagement in eHealth and mHealth behavior change interventions: Viewpoint of methodologies. J Med Internet Res. 2018;20(11):e292. doi: 10.2196/jmir.9397

15.

Baysari

Westbrook

. Mobile applications for patient-centered care coordination: A review of human factors methods applied to their design, development, and evaluation. Yearb Med Inform. 2015;10(1):47‐54. doi: 10.15265/IY-2015-011

16.

Chiang

Guo

Amico

Atkins

Lester

. Interactive two-way mHealth interventions for improving medication adherence: An evaluation using the behaviour change wheel framework. JMIR Mhealth Uhealth. 2018;6(4):e87. doi: 10.2196/mhealth.9187

17.

Holzman

Zenilman

Shah

. Advancing patient-centered care in tuberculosis management: A mixed-methods appraisal of video directly observed therapy. Open Forum Infect Dis. 2018;5(4):ofy046. doi: 10.1093/ofid/ofy046

18.

Killian

Clifford

Gupta

. Directly observed therapy to promote medication adherence in paediatric heart transplant recipients. Cardiol Young. 2021;31(12):2048‐2050. doi:10.1017/S1047951121002109

19.

Killian

Clifford

Lustria

MLA

Skivington

Gupta

. Directly observed therapy to promote medication adherence in adolescent heart transplant recipients. Pediatr Transplant. 2022;26(5):e14288. doi: 10.1111/petr.14288

20.

Hsieh

Shannon

. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):1277‐1288. doi: 10.1177/1049732305276687

21.

Marcolino

Oliveira

JAQ

D'Agostino

Ribeiro

Alkmim

MBM

Novillo-Ortiz

. The impact of mHealth interventions: Systematic review of systematic reviews. JMIR Mhealth Uhealth. 2018;6(1):e23. doi: 10.2196/mhealth.8873

22.

Simons

McCormick

Mee

Blount

. Parent and patient perspectives on barriers to medication adherence in adolescent transplant recipients. Pediatr Transplant. 2009;13(3):338‐347. doi:10.1111/j.1399-3046.2008.00940.x

23.

Pringle

Fawcett

. Facilitating the therapeutic alliance between pharmacists and patients to improve medication adherence. In Prescott DS, Maeschalck CL, Miller SD (eds), Feedback-informed treatment in clinical practice: Reaching for excellence. Washington DC: American Psychological Association; 2017: 299‐320.

24.

Oinas-Kukkonen

Harjumaa

Persuasive systems design: Key issues, process model, and system features. Comm Assoc Infor Sys. 2009;24(28):485‐500. doi:10.17705/1CAIS.02428