Emerging adults with sickle cell disease (SCD) face a 7-fold increase in mortality rates during the transition from pediatric care to adult care [1]. In addition, this subpopulation (aged 17-25 years), experiences the highest rates of hospitalizations, emergency room visits, and hospital readmissions compared with children living with SCD [2]. This alarming increase in mortality and acute care use stems partly from the cumulative effects of SCD leading to organ dysfunction, with the risk of comorbidities such as stroke and silent infarcts worsening as individuals age [3,4]. In addition, during this transition process, emerging adults with SCD experience fragmented care coordination, which increases their vulnerability.

Emerging adults with SCD face shifts in insurance, inconsistent transfer processes, and a lack of comprehensive adult-focused programs [4-6]. For instance, while adherence to chronic transfusion therapy and hydroxyurea therapy can substantially mitigate the risk of neurological complications and other adverse outcomes, inadequate care coordination during the transition to adult care, coupled with suboptimal patient engagement and SCD self-management skills, markedly heighten the risk of neurological complications [6-9]. These complications can exert enduring and profound impacts on physical function, educational attainment, employment, income, and overall health-related quality of life (HRQOL).

Beyond health system challenges, emerging adults face additional psychosocial challenges, including the struggle for financial independence and insurability, which can be complicated by job instability and the loss of dependent coverage on parents’ insurance plans [10]. Transportation barriers may also limit access to necessary medical appointments or therapies, especially for those living in areas with limited adult health care resources [8]. In addition, many emerging adults with SCD experience limited access to mental health care, which can exacerbate issues such as depression, anxiety, and the psychological burden of managing a chronic illness [11]. Social determinants of health, including poverty, housing instability, food insecurity, and lack of educational opportunities, also further complicate the ability of emerging adults with SCD to manage their condition effectively and navigate the health care system [11]. Furthermore, the absence of adequate social support and educational resources compounds these challenges, leaving emerging adults with SCD underprepared and less equipped to navigate their transition effectively [6,12].

Disruption in continuity of care, combined with socioeconomic factors, can lead to worse health outcomes, and an increased risk of complications show the need for effective interventions [11,12]. Community health worker (CHW) programs and mobile health (mHealth) platforms have shown promise in addressing these challenges and improving health outcomes for chronic disease transition patients [13,14]. Research shows that CHW programs enhance chronic disease management and health care navigation, improving outcomes in low-resource settings for conditions such as asthma, hypertension, and diabetes [15,16]. For instance, the IMPaCT CHW program demonstrated that participants in the intervention arm received more timely posthospital care and better discharge communication and experienced fewer readmissions [13].

Similarly, mHealth platforms have been effective in promoting self-management and adherence among both adults [17,18] and emerging adults with chronic conditions, including diabetes and post–liver transplant care [14,19,20]. A randomized controlled trial (RCT) of MD2Me, an internet and mobile phone–delivered disease management intervention using automated SMS text messaging algorithms, showed that emerging adults with type 1 diabetes, inflammatory bowel disease, and cystic fibrosis in the intervention arm significantly improved disease self-management and health-related self-efficacy [14]. This technology holds great promise for adolescents with chronic diseases due to the widespread use of smartphones [21]. Integrating CHW and mHealth initiatives can substantially alleviate health care transition challenges and address psychosocial factors, thus improving care continuity and outcomes for patients with chronic diseases.

Gaps in evidence highlighted in the study by Viola et al [6] demonstrate the need for multisite RCTs focusing on long-term care outcomes. Current research, while promising, offers inconclusive results due to limited sample sizes and methodological constraints. Recognizing these limitations, our study aims to bridge this gap by conducting comprehensive multisite RCTs that evaluate both clinical and psychosocial outcomes, ultimately aiming to enhance chronic disease management and overall health during this critical period. We hypothesize that emerging adults in the CHW or mHealth arm will have increased HRQOL compared with those in the enhanced usual care arm.

Our intervention leverages 2 key frameworks: the social-ecological model for adolescents and young adults readiness to transition (SMART) model and the chronic care model. Effective transition care, as emphasized by the SMART model, requires integrating both preexisting and modifiable factors among patients, parents, and health care providers that influence transition readiness. Modifiable factors, such as knowledge, self-management skills, confidence (or self-efficacy), and social support, are critical targets for enhancing transition readiness, especially for individuals managing chronic conditions. The SMART model has been validated across various populations and has been updated for health equity after this trial was designed, including childhood cancer survivors and their parents and health care providers [11,22,23].

For individuals with SCD, as well as their parents and health care providers, the ability to independently manage their SCD is recognized as a crucial goal during the transition to adult care [6,24-26]. Self-management behaviors, defined as the daily, self-motivated, and collaborative activities required to manage symptoms, are essential for successful disease management. In the context of SCD, these behaviors include self-awareness; seeking emotional support; managing diet and hydration; avoiding alcohol, tobacco, and drugs; and adhering to prescribed treatments such as daily hydroxyurea or regular transfusion and chelation therapy.

Self-management support is a foundational element of the chronic care model [27], which is widely regarded as an effective framework for improving chronic illness care. According to this model, improved health outcomes, including better disease control, a smoother transition to adult care, and enhanced quality of life (QOL), are achieved through the activation and engagement of patients in self-management behaviors, supported by proactive health care teams. In this context, individuals who are more “activated” tend to exhibit higher levels of the modifiable factors outlined by the SMART model, such as increased knowledge, self-efficacy, and social support, all of which are crucial for successful self-care and transition. We chose to target these modifiable factors because patients with SCD and their parents and clinicians agree that it is critical to engage in disease self-management to optimize health during the transition period (Figure 1) [28].

The Community Health Workers and Mobile Health for Emerging Adults Transitioning Sickle Cell Disease Care (COMETS) study is an ongoing pragmatic, multicenter, 3-arm, open-label RCT. A total of 5 pediatric SCD centers located in 4 states (Connecticut, New York, Ohio, and Pennsylvania) are participating in this study. Eligible participants are randomized 1:1:1 across sites to a 6-month CHW program, 6-month mHealth self-management program with tailored SMS text messaging, or to enhanced usual care (control).

The overall goal of this study aims to answer the question “How can community health workers and mobile applications help me to improve my ability to take care of myself (activated patient, self-management), stay connected with my doctors (usual source of care), avoid visits to the ER and hospital (acute care use), and ultimately maximize my quality of life while I am transitioning from my pediatric to adult doctors?”

Eligible participants are young adults aged 17 to 25 years with SCD diagnosis of all genotypes who plan to transition to an adult hematologist within 12 months; have access to a mobile device, tablet, or computer; and can interact with a mobile- or web-based program. Eligible participants are identified by reviewing each recruitment site’s daily hospitalized patients, weekly hematology appointments, and hematology practice registries. Across the 5 pediatric SCD centers, we estimate that there will be 745 eligible patients over the recruitment period, with an equal distribution between male and female patients.

Recruitment strategies include in-person enrollment during appointments or admissions and mailed or emailed communications with a recruitment letter, study flyer, and a recruitment video. During the COVID-19 pandemic, telephone recruitment and electronic consent were added to maintain recruitment goals. Retention strategies include completion of enrollment and follow-up procedures in-person, telephone, or an email or text, automatic reminders, and incentives for effort. Figure 2 provides an overview of the study screening and enrollment process.

Random assignment takes place after enrollment and the completion of the baseline survey. A document including stratification and randomly permutated blocks of unequal sizes (to prevent providers and patients from manipulating the randomization) was used in REDCap’s randomization module. Enrolled participants will be randomly assigned to a treatment arm after being stratified by recruitment site, age (≤21 y and ≥22 y), and disease severity (moderate and severe), resulting in 12 different randomization strata [29,30]. Severe SCD disease will be determined by chart review and defined by the following: (1) any history of stroke, acute chest syndrome, or ≥3 emergency room visits within the 3 years before enrollment, which are all indications for disease modifying therapy, such as daily hydroxyurea or monthly exchange transfusion therapy [31-33]; and (2) receipt of monthly exchange transfusion therapy, defined by ≥2 exchange transfusions in an outpatient setting within 6-month period [33]. Severe disease will be defined as any history of acute chest syndrome, stroke, or ≥3 hospitalizations in the past 3 years [31,32].

This is a nonblinded study because the nature of the intervention prevents patients from being blinded to their study arm.

Enhanced usual care, standardized across all sites through transition or transfer of care checklist, is administered by the social work team at each recruitment site. During the intervention development phase, all recruitment site investigators reviewed and decided on the elements to be included in this checklist based on transition elements at their institution. The checklist is reviewed at the 12-month time point in the trial with the participant during a scheduled hematology visit. Elements of the checklist include (1) patient seen by the pediatric provider, with the parent outside the examination room at least once, (2) a social work consult to screen and address sociodemographic risk factors, (3) information on health insurance availability provided to patient, (4) adult hematologist identified, (5) adult primary care provider identified, (6) medical release signed, and (7) medical record viewable or sent to adult provider. Elements of the checklists are based on existing checklists by organizations such as the Transfer of Care Checklist developed by Got Transition or Center for Health Care Transition Improvement [34] and those developed by the National Institute for Child Health Quality [35]. All study participants received standardized enhanced usual care.

The CHW program is modeled after the IMPaCT program developed by IMPaCT Care, Inc [13,36,37]. Components of the CHW program include (1) development of patient-centered goals and individualized action plan around self-care, symptom tracking, and transition to adult care; (2) provision of information, skills, and tips; and (3) tailored peer support using telephone calls or visits.

To identify the components of the CHW intervention, web-based work group meetings were held to define the role and training components [38]. Training topics included local hospital and community knowledge, SCD and transition to adulthood education, basic clinical research training, foundational CHW skills and concepts, and trial-specific aims and resources.

Training components included IMPACT’s 2-week training course, onboarding and orientation at their respective sites, and Collaborative Institutional Training Initiative for clinical research. With an emphasis on ongoing support and education, CHWs meet weekly with a trial supervisor and their site supervisor to review caseloads and address any concerns. Monthly group huddles are held to support peer-to-peer learning for both CHWs and their supervisors. CHWs are expected to contact participants weekly during the 6-month intervention period to remain consistent with other successful protocols [39,40].

Participants randomized to the mHealth arm will download an enhanced version of iManage [41] called iManage SCD, developed by Crosby et al in collaboration with adolescents and young adult patients with SCD [28,38]. Key components of iManage SCD include (1) development of patient-centered goals around self-care, symptom tracking, and the transition to adult care; (2) provision of information, skills, and tips to support these goals; (3) virtual peer support, where users can encourage others, form teams, and interact with fellow youth with SCD; and (4) daily symptom tracking and visual monitoring of goal completion. iManage SCD is an mHealth application accessible via smartphone, tablet, or web and has been shown to be highly feasible and beneficial for emerging adults with SCD.

Enhancements to the iManage application were identified through the convening of a multi-stakeholder group, including an application development firm, clinicians, researchers, and young adults with SCD across the 5 identified recruitment sites. The development process of the enhanced application has been described [28]. Interviews with stakeholders explored 8 topics, including (1) social behaviors and entertainment, (2) treatment, (3) pain impact, (4) pain management, (5) SCD knowledge, (6) lifestyle and goals, (7) communication and support, and (8) envisioning the future. Final components of the application include health behavior challenges, tracking of medications and symptoms, an educational material library, discussion boards, and medical summary. User engagement was tracked through analytic tools integrated within the mHealth app [27].

In addition, daily tailored texting (THRIVE 2.0) [42], based on patient-defined goals were delivered to participants randomized to the mHealth arm of the study. Preliminary studies have found that THRIVE 2.0 improves health-promoting behaviors such as sunscreen use among adolescent and young adult cancer survivors. These SMS text messages were tailored around modifiable factors and delivered twice a week. Two types of messages were delivered, educational and interactive. Messages focused on SCD knowledge, skills and efficacy, relationships, self-management, school or work, and coping.

Participants will receive training on how to use and customize iManage SCD during enrollment and have access to a helpdesk for support. The content on iManage SCD was continuously monitored by research staff. Participants were able to access the application after the 6-month active intervention period.

Both intervention arms include the same core components, with one delivered by a CHW and the other through a digital platform. Both interventions were designed to support self-management by facilitating patient-centered goal setting, providing tailored information and skill-building, and offering peer support. In the mHealth intervention, these components are adapted to be delivered via the app, mirroring the CHW’s role in engaging participants through structured content and interactive features.

The primary outcome of the COMETS study will be the change in HRQOL. The secondary outcomes are acute care use (hospitalizations and emergency department visits), patient activation, self-management behaviors, and successful transfer to adult hematology care.

This study is ongoing; therefore, the statistical analyses have not been performed.

Sample size calculations are based on our primary end point of change in HRQOL and health care use for each participant over their study participation [62]. Details of the sample size calculations are presented in Multimedia Appendix 1.

Analyses will be conducted in Stata (version 18.0; StataCorp), with 2-sided tests of hypothesis and a P value <.05 as the criterion for statistical significance.

To compare the effectiveness of 2 self-management support interventions (CHWs and mHealth) versus enhanced usual care on the primary outcomes (HRQOL and acute care use), we will first compare groups at baseline using chi-square tests for categorical variables and 1-way ANOVA or Kruskall-Wallis, as appropriate, for continuous variables. We will use descriptive summaries and graphical displays to evaluate the distribution of and relationship between variables. When required, transformations to achieve normality will be considered.

The initial analysis will use 1-way ANOVA to test the hypothesis that the 3 treatment groups showed no change from the end of the intervention to the baseline in the primary and secondary outcomes. If the hypothesis of equal group means is rejected (there was a change detected), 3 post hoc tests will be applied (with Bonferroni adjustment for multiple testing) to identify pairwise differences between treatment groups. We postulate that a difference in change of 10 points in PedsQL SCD total between groups will be clinically meaningful.

Next, we will construct longitudinal models that include baseline, 6-, 12-, and 18-month measurements as outcomes. Longitudinal models account for correlation of measures within emerging adult over time; use all data collected; allow for and can be used to adjust for dropout or nonadherence to assigned treatment; can adjust for differences in patient characteristics not balanced by randomization; can include indicator variables for clinical site; and permit continuous, count, and binary outcomes. We will build models using generalized estimating equations (GEEs) or quasi-least squares (QLS) [63]; QLS is based on GEE but allows for implementation of the Markov correlation structure that is appropriate if participants intermittently miss appointments. Mixed effects models may also be considered. For example, mixed models can be used to relax the assumption of constant variance that is required for QLS and GEE; they will, therefore, be useful if residual diagnostics for GEE or QLS suggest an increase in the variance of outcomes over time.

The longitudinal models will include indicator variables for visit, group, and time-by-visit group interaction terms (if appropriate, time will be included as a continuous variable). If the time-by-group interaction terms differ significantly from 0, this will indicate that the change over time in outcomes differs significantly between the treatment groups. In sensitivity analysis to examine the impact of treatment intensity, we will also evaluate the effects of the intervention stratified by intervention dosage (frequency of interaction with the CHW, or mHealth or texting program).

Next, we will assess whether patient activation, self-management behaviors, biologic markers, and transfer to adult care are mediators of intervention treatment effects. This analysis focuses on the role of patient activation and self-management behaviors at 6 or 12 months as mediators of the effect of the intervention on the study outcomes. We will perform the mediation analysis using structural equation modeling; this will allow us to estimate the total effect of the treatment on outcomes as a sum of direct effects and indirect effects that are mediated by patient activation, self-management behaviors, biologic markers, or transfer to adult care. To evaluate the role of each potential mediating variable, we will calculate the proportion of the total effect that is mediated, the ratio of the indirect effect to the direct effect, and the ratio of the total effect to the direct effect. We also evaluate percentile and bias-corrected bootstrapped SEs and CIs for the effects.

We will next perform an exploratory analysis to identify individual and family factors that moderate intervention treatment effects. To test the hypothesis that adults with SCD and high disease severity will moderate the effect of intervention on outcome measures at 6, 12, and 18 months, we will first evaluate whether the impact of the intervention on overall changes (6 m minus baseline) differs between the high versus low severity groups. This will be tested by building a regression model with changes (end of intervention minus baseline) as the outcome variable, and that includes indicator variables for treatment groups, severity group (high vs low), and group-by-severity interaction terms. If the interaction term for a particular treatment group differs significantly from 0, this will indicate that the impact of that intervention differs according to (is moderated by) severity of disease. Next, to evaluate moderation in longitudinal models, we will fit models with change since baseline at 6, 12, and 18 months as the outcomes; these models will also include indicator variables for treatment groups, indicator variables for severity group, and time-by-severity interaction terms. If the interaction terms differ significantly from 0, this will indicate that the change since baseline depends on both treatment group and severity status (ie, that the impact of treatment is moderated by severity of disease). We anticipate that this model will be adequate to evaluate interactions; however, it may be necessary to consider more complex models, by modifying the longitudinal models for analysis of our primary outcome variable to include time by severity group, severity group by intervention, and severity by time by treatment group (3-way interaction) models. Finally, to explore the moderating effects of individual and family demographic characteristics on treatment effects, we will convert all patient and family demographic variables into dichotomous variables and evaluate potential moderation using the approach just described for severity status.

This research aims to compare the effectiveness of 2 interventions, a CHW intervention and an mHealth intervention, to enhanced usual care alone. Our study seeks to address the significant gap in research focused on transition to adult care, while examining the impact of such interventions on chronic disease management, health care use, or health outcomes. The COMETS trial is one of the first multisite RCTs to assess the effectiveness of these interventions specifically for young adults with SCD.

Despite the under-studied effectiveness of mHealth and CHW programs in improving the transition to adult care for emerging adults with SCD, this research hypothesizes that incorporating either a CHW or an mHealth app will improve HRQOL and reduce acute care use. This trial aims to identify effective interventions to support young adults with SCD as they navigate the complexities of adult care, addressing gaps in current research. Positive results could translate into significant cost savings for payers and health care systems by reducing avoidable emergency department visits and hospitalizations, informing funding agencies such as Health Resources and Services Administration and State Title V programs dedicated to improving care for children with special health care needs. With the increase in interest and clinical recognition of both CHW programs [64,65] and mHealth apps [66-69], our study aims to support the growing body of research to aid in integrating these interventions into standard care models potentially leading to more personalized, accessible, and efficient care. This increase in mHealth app use and support for CHW programs in not only patients with SCD but also among patients with other chronic diseases is important to acknowledge. These interventions were chosen for their potential cost-effectiveness and scalability. Ultimately, demonstrating measurable improvements in health outcomes and cost savings will be key to integrating these personalized and accessible interventions into standard transition care models, leading to more efficient care and sustained improvements in health outcomes for young adults with SCD and other populations with chronic diseases.

The study uses several key strengths to enhance its effectiveness and scalability. First, the involvement of stakeholders, particularly young adult coinvestigators, during the development phase ensures that research is aligned with their real-world experiences. This provides a comprehensive understanding that can inform implementation. In addition, building upon the established IMPACT program [40], the study leverages past research and implementation insights for further refinement and enhancement. In response to the limitations posed by COVID-19, such as reduced in-person recruitment opportunities, the study incorporates diverse communication methods to increase accessibility and engagement. Finally, the use of robust patient registries allows for real-time monitoring and identification of patients with SCD who meet study criteria, facilitating a streamlined recruitment process even with pandemic-related constraints.

SCD interventions have anticipated limitations that must be acknowledged, particularly when engaging young adults. Emerging adults are often at pivotal time in their lives where they might be transitioning into college or the workforce. This stage of life is inherently demanding and can make it challenging for young adults to adhere to treatment plans and prioritize participation in SCD programs. In addition, there is a widespread distrust of the health care system among this demographic, often rooted in historical and systemic inequities, which can be exacerbated by experiences of discrimination and stigma in both health care settings and broader society [70]. These factors together can lead to feelings of alienation and reluctance to engage with and adhere to SCD treatment plans or seek necessary care [70]. Additional limitations stemmed from the population being emerging adults, specifically surrounding the reliance on self-reporting and its potential bias. Recall bias is a challenge because adolescents and young adults with SCD may not accurately remember the past, or due to social desirability bias, they may answer questionnaires with what they think they should say or what they believe is socially desirable. To mitigate these challenges and bias, we carefully chose questionnaire designs and collected input and consulted adolescents and young adults with SCD and other stakeholders who could speak to these challenges. Furthermore, addressing these challenges requires targeted strategies that build trust through culturally competent care, increase accessibility by aligning program delivery with young adults’ schedules, and provide support systems that acknowledge and alleviate the unique pressures faced during this life transition.

In conclusion, the results of this study will contribute significantly to filling the gap in transition care research for emerging adults with SCD, guiding future intervention strategies, and providing valuable evidence for health care systems, payers, and policy makers. The potential impacts of these interventions extend beyond clinical outcomes, offering the possibility of improved psychosocial well-being, enhanced patient self-efficacy, and a more successful transition to adult care for individuals with SCD.