“…in eHealth psychoeducational research”
“..youth with type 1 diabetes….”
“The purpose of this study was to compare the demographic and clinical characteristics of youth with type 1 diabetes (T1D) on recruitment, participation, and satisfaction with two e-Health psycho-educational programs.”
“Type 1 diabetes (T1D) is a common chronic illness in youth, affecting 1 in 400 youth”
“As youth transition to adolescence and take on greater responsibility for T1D management and decision-making, adherence to diabetes tasks often deteriorates, resulting in family conflict, psychological distress, and poor metabolic control.”
“Psycho-educational programs for youth and family-based programs have shown to be effective in improving psychosocial and diabetes-related outcomes…. However, disseminating and translating research-based programs into clinical care has been challenging due to provider and family time constraints, as well as cost”.
“E-Health technologies, such as the internet, offer a platform for improving the dissemination and accessibility of psycho-educational programs for youth with T1D…. Despite the numerous benefits of web-based programs for youth with chronic illnesses, concerns have been raised about the ‘digital divide’ and internet access for youth of diverse race and ethnicity and youth living in low income families.”
“Challenges of recruiting youth of diverse race and ethnicity for research are well established... [and] Little is known…about the recruitment process and yield of diverse samples in internet research with youth who have a chronic illness.”
“In addition to concerns regarding access to e-Health research, participation in e-Health programs has varied considerably across studies.”
“Psycho-educational programs for youth and family-based programs have shown to be effective in improving psychosocial and diabetes-related outcomes”
“Psycho-educational programs delivered via computer-based internet access have demonstrated efficacy in youth with various chronic illnesses, leading to improved knowledge, symptoms, health outcomes, and quality of life. With respect to youth with T1D, a web-based self-management program with a focus on problem-solving and social networking demonstrated improved self-management and problem solving in youth who completed the program compared to a control group. An internet coping skills training program, developed by our research team, did not demonstrate differential improvements in metabolic control and diabetes-related outcomes compared to an internet diabetes education program, but youth in both groups reported significantly increased self-care autonomy, higher diabetes self-efficacy and improved overall quality of life over time”
Regarding access –
“While the vast majority of youth are online, access is higher in white youth and those who live in high-income families. White youth and youth in high-income families are more likely to have online access at home (96%) and go online more frequently compared to Black youth (92%), Hispanic youth (87%), and youth of low-income families (86%). A positive relationship between socioeconomic status and computer-based internet use has been demonstrated in diverse middle school students and a diverse pediatric clinic population.”
Regarding participation in e-Health Programs –
“….in an e-Health program for youth with asthma, participants did not complete self-monitoring on 60% of study days. In an e-Health program for depression, only 30% of youth completed 50% or more of the program modules. In the e-Health problem-solving program for youth with T1D, the mean number of modules completed was 5.22 (out of 8), with only 63% of youth completing all modules. Participation in e-Health programs for youth typically decreases over the course of the study and higher participation has been associated with outcomes. For example, school-aged youth who had greater participation in an e-Health obesity prevention program demonstrated improved outcomes compared to youth with less participation. A structured environment (ie, school vs home) may improve youth participation in e-Health programs. Youth who participated in a school-based e-Health program on depression had almost a 10-fold higher completion rate for modules and program exercises compared to youth who participated in the same program delivered as open access online.”
“Factors associated with e-Health program participation have begun to be identified. Girls have demonstrated greater participation compared to boys. Increased depressive symptoms may also influence participation, although this effect may vary depending on the characteristics of the program, severity of symptoms, and the type of chronic illness. For example, in one study evaluating an e-Health program for depression treatment, less participation was reported in youth with higher depressive symptoms at baseline. In contrast, in another study evaluating an e-Health depression prevention program, the authors reported that youth with higher depressive symptoms had greater participation in the program.”
“In summary, recruitment, participation, and satisfaction with e-Health programs have the potential to influence e-Health program outcomes and generalizability of results”
“...the purpose of this study was to compare the demographic and clinical characteristics of youth with T1D on recruitment, participation, and satisfaction with two e-Health psycho-educational programs. Specifically, recruitment, participation, and satisfaction were compared by age, gender, race/ethnicity, income, metabolic control, and depressive symptoms.”
The report is a secondary analysis of baseline and participation data from a randomized controlled clinical trial
“A convenience sample was recruited from 4 university-affiliated clinical sites, which included Children’s Hospital of Pennsylvania, University of Arizona, University of Miami, and Yale University… Youth and parents were approached in the clinic setting and informed consent/assent was obtained by trained research personnel. Demographic data were completed by parents at enrollment and email communication was subsequently established with youth. Youth were sent a link to a password-protected data collection website, and parents were notified of this communication.”
Participants had a single logon identity based upon their email which was used to access the data collection and intervention websites. After enrollment, notice to complete study tasks and data collection was via automated emails. Other contact with study participants (e.g. additional reminders about overdue data collection) was via email or phone. Occasionally youth were met at clinic visits.
Psychosocial outcomes were self-assessed through online questionnaires. Clinical outcomes were collected by research staff from the medical records. Participation was assessed by programming which tracked number of intervention sessions completed by each unique user.
Participants were provided with a unique logon/username based upon their email address. The applications could be accessed from any computer. Most participants logged on from home; a few logged on from school, a library, or another location. There were separate websites for data collection and the intervention – both were accessed at appropriate times using the same logon information. Participants were sent a gift card to Target each time they completed online data collection ($20 gift card at baseline). No compensation was given for participation in the intervention.
“Youth were sent a link to a password-protected data collection website, and parents were notified of this communication. Internet sites were password protected with all data encrypted and stored on a secure server with hardware and software firewalls.”
“Upon completion of baseline data collection, an automated email was sent to youth and their parent/guardian to identify their group assignment and provide a link to the appropriate program. A unique password was provided to each participant and they were instructed to change this password the first time they logged onto the program.”
“The current study is a secondary analysis of data from a clinical trial evaluating the effect of an internet coping skills training program (TEENCOPE) compared to an internet diabetes health education program (Managing Diabetes) for youth with T1D.”
“Each program consisted of 5 sessions with content tailored to adolescents with T1D. TEENCOPE used a cast of ethnically diverse characters (youth with T1D) and a graphic novel format to model common problematic social situations (i.e., parent conflict) and different coping skills to solve the problems. Managing Diabetes used visuals and a highly interactive interface that allowed adolescents to actively problem-solve diabetes self-management situations.”
A detailed description of the development of the coping skills training website is available in: Whittemore, R., Grey, M., Lindemann, E., Ambrosino, J., & Jaser, S. (2010). An internet coping skills training program for teens with type 1 diabetes. Computers, Informatics, and Nursing, 28, 103-111.
Descriptions of both the coping skills training and the internet diabetes health education programs are available in Grey, M., Whittemore, R., Liberti, L., Delamater, A., Murphy, K., Faulkner, M. A comparison of two internet programs for adolescents with type 1 diabetes: Design and methods. Contemporary Clinical Trials, 2012 Mar 29,
“Youth were contacted by phone after the first session was released to ensure that youth had received the email and were able to access the program. If youth did not complete a session within one week, weekly emails, phone, or postcard reminders were sent. Parents were contacted by email if youth did not complete a lesson after 3 weeks.”
There were no co-interventions
There were no changes to trial outcomes once the trial commenced
Participants were blinded to their intervention assignment until baseline data collection was complete. Participants found out their assignment when intervention sessions were released. The investigators, data analysts, and clinicians were blind to assignment. All data collectors were completely blind to assignment with the exception of a research associate at the parent site who collected clinical data from the medical chart and who could potentially review study records to see the group to which a study participant had been assigned. This R.A. minimized her access to group assignment in order to remain unbiased.
Participants did not know which was the “intervention of interest”
Imputation techniques to deal with attrition and missing values are not applicable. The study was a cross-sectional analysis of study data. “Chi-square or ANOVA analyses were used...” to compare demographic and clinical characteristics of teens with T1D on recruitment and participation rates. Imputation rates were used in the evaluation of the outcomes of the programs.
See Figure 1 – Consort Flowdiagram
See Table 1 - Baseline Demographics
“A total of 518 youth were approached, and 63% of eligible teens enrolled in the study, with 22% refused (n=112) and 15% passively refused (n=78)”. …“A total of 320 youth enrolled in the study.”
“Participation in the internet programs was high, with 78% of youth completing at least 4 of 5 sessions, 13% completing 1-3 sessions, and 9% completing no sessions. The mean number of sessions completed was 4.08 (+1.64) across both groups. There was no significant difference in participation between groups; teens in TeenCope participated (completed at least 80% of sessions) at a rate of 81.9%, whereas those in Managing Diabetes participated at a rate of 74.4% (χ2 = 2.63, P = .11).”
“…the purpose of this study was to compare the demographic and clinical characteristics of youth with T1D on recruitment, participation, and satisfaction with two e-Health psycho-educational programs. Specifically, recruitment, participation, and satisfaction were compared by age, gender, race/ethnicity, income, metabolic control, and depressive symptoms.”
“Chi-square analyses were used to test for demographic differences by recruitment category (see Table 2). There were no significant differences in recruitment category by gender (χ2 = 3.97, P = .14) or age group (11-12 vs. 13-14) (χ2 = 2.33, P = .31). There was, however, a significant difference for race/ethnicity (χ2 = 22.48, P < .001) with respect to enrollment. White youth were more likely to enroll and less likely to passively refuse, and non-White youth were less likely to enroll and more likely to passively refuse. There was also a significant difference by income (χ2 = 25.16, p < .001), with teens from the lowest income category (annual income < $40,000) less likely to enroll and more likely to passively refuse than teens from the higher income categories (annual income > $40, 000). It is important to mention that data were not available on race/ethnicity and income for all recruitment categories; data were unavailable on the race/ethnicity and income of youth who refused at the point of contact.”
“Results of the Chi-square analyses to test for demographic differences by participation are provided in Table 3. There was no significant difference by gender (χ2 = 1.02, P = .31) or race/ethnicity (χ2 = 3.15, P = .37). There was a significant difference for income (χ2 = 12.64, P = .002), with those in the lowest income category (annual income < $40,000) less likely to participate, and those in the highest income category (annual income > $80,000) most likely to participate. There was a trend toward significance in participation by age group (χ2 = 3.14, P = .08), with those in the younger group (age 11-12) somewhat more likely to complete 4 or more sessions than those in the older group (age 13-14). There was no significant difference in participation for metabolic control; adolescents who had an A1C below the recommended cutoff (< 8%) were no more likely to participate than those above the cutoff (χ2 = 0.23, P = .63). Lastly, depressive symptoms (i.e., CDI score) were significantly related to participation (χ2 = 3.87, P = .05); youth who scored above the clinical cutoff on depressive symptoms were less likely to complete 4 or more sessions than those who scored in the normal range.”
“Satisfaction was high with both programs, with no significant difference between groups. The mean satisfaction score was 3.97 (+ .71) for TeenCope and was 3.89 (+.56) for Managing Diabetes. There were significant gender, race/ethnicity, and income differences, in that girls (t= 3.28; P=.001), non-White youth (t= 2.42; P=.02), and low-income youth (F=3.80; P=.02) reported higher satisfaction. There was no difference in satisfaction by age or depressive symptoms.”
“This study provides important new information regarding the role of demographic factors in the study of e-Health programs, but there are several limitations. First, it is important to acknowledge that many youth in our study needed prompts and reminders from research staff to achieve a high level of participation. Further, we did not have data on race/ethnicity and income for the families who did not enroll in the study. Finally, while the sample was diverse, race/ethnicity and income were highly correlated; only 23% of the high income youth were non-White, making it difficult to determine the relative effect of race/ethnicity and income on participation and satisfaction.”
Registration: Clinicaltrials.gov NCT00684658
Funding: NINR R01NR0094009