Beta Testing Can-Do-Tude: An Online Intervention Using the Principles of Motivational Interviewing for Adolescents with Type 1 Diabetes

Abstract

Objective. This article describes results from a beta test of an online intervention, Can-Do-Tude, that uses the principles of motivational interviewing (MI) and tailored diabetes self-management education for adolescents with poorly controlled type 1 diabetes (T1D). 

Method. Five adolescents (14-17 years) with T1D were recruited to test the intervention. A quantitative descriptive design was used to evaluate feasibility. A qualitative descriptive design was used to evaluate the acceptability and utility of the intervention.

Results. Eighteen of the 20 modules were completed by the teens, a 90% completion rate. Results overall showed that the participants had a positive response to the intervention (M = 4.28, SD = 0.55). Feasibility was measured by a pre-and post-test using the Self-Efficacy for Diabetes Self-Management instrument (Iannotti et al., 2006). A Wilcoxon signed-rank test indicated that the post-test scores were significantly higher than pre-test scores (Z = -2.952, p = .007). Subject’s responses to questions revealed that the teens felt most confident and ready to make a change in behaviors around diet management (M = 8.57, SD = 2.14) but they had little motivation and interest around identifying blood glucose (BG) patterns (M = 3.0, SD = 3.83). Three themes were identified: they are concerned about the impact of diabetes on their health, they want better BG control and they recognize that they need to check their BG more often. 

Discussion. Findings from this first phase of intervention testing support moving to phase 2 testing with an initial test of the intervention in comparison with an appropriate alternative.

Introduction

Type 1 diabetes is a chronic health condition that affects approximately 193,000 American youth under the age of 20 (American Diabetes Association, 2018). Although adults also contend with this condition, metabolic control is worse in adolescents and young adults compared to any other time of life (Miller et al., 2015). Results from the SEARCH for diabetes in youth, a Centers for Disease Control (CDC) and National Institute of Health (NIH) funded study, revealed that youth who have had diabetes from an early age were already showing measurable signs of diabetic complications: peripheral neuropathy, cardiovascular disease, kidney disease and retinopathy (Jaiswal et al., 2018, Jaiswal et al., 2017; Dabelea et al., 2017). This article discusses the results from a beta test of an online intervention—Can-Do-Tude—that uses the principles of motivational interviewing (MI) and tailored diabetes self-management education for adolescents with poorly controlled type 1 diabetes (T1D). A description of the development of Can-Do-Tude and the alpha testing of this intervention have been previously published (Paul, 2015). 

Background

Adolescence is an unsettling time of life. It involves tremendous biological, cognitive and social transitions - changes that result from developments in the endocrine and central nervous systems. These changes help to ready the individual for adulthood and independence. Due to the various changes occurring in these systems, particular behaviors tend to emanate. Adolescence is characterized by risk-taking (impulsivity and sensation seeking) as well as preferences for peer as opposed to parental relationships (Romer, Reyna and Satterthwaite, 2017). Changes in these systems direct the drives and motivations of teens; consequently, these behaviors can potentially place the adolescent at risk for harm. The normal developmental tasks of adolescence are themselves a challenge. When those are coupled with the rigorous responsibilities of managing a chronic illness such as T1D, the demands can become overwhelming.

Health behavior is a critical determinant of an adolescent’s well-being because the health behaviors established during adolescence set the stage for one’s health and well-being later in life. T1D management involves the delicate balance of diet, exercise and insulin injections while closely monitoring blood sugars in order to keep glucose levels under control. This can be especially difficult in adolescence. The Diabetes Control and Complications Trial (DCCT) showed the importance of maintaining good glucose control to delay the onset and progression of diabetic complications (DCCT, 1993 and 1994). Although more recent studies have shown a decline in diabetic complications with the use of intensive insulin therapy (Epidemiology of Diabetes Interventions and Complications [EDIC], 1999), adolescents and young adults are still developing the long-term complications of diabetes (Jaiswal et al., 2018, Jaiswal et al., 2017, Dabelea et al., 2017 and Nathan, 2014). As adolescents with T1D transition into adulthood, metabolic control deteriorates until the mid-twenties (Miller et al., 2015). This deterioration has been linked to risky behaviors such as poor adherence to treatment regimes, insulin misuse and eating disorders (Jaser, Yates, Dumser and Whittemore, 2011; Wasserman, Anderson and Schwartz, 2017). Risk-taking in an adolescent’s life involves a normal process of parental separation, individuation and testing limits, leading to the development and consolidation of one’s identity; however, it can become self-destructive when the risks exceed healthy limits (Romer et al., 2017). Taking risks with one’s chronic illness may have devastating consequences, especially with a disease such as type 1 diabetes.

Given that a major cause of poor metabolic control in teenagers stems from poor decision making and a lack of adherence to treatment regimens, adolescents need opportunities to develop intrinsic motivation to manage their diabetes. Motivational interviewing (MI) is a “collaborative conversation style for strengthening a person’s own motivation and commitment to change” (Miller and Rollnick, 2013, pg. 12). MI has also become a topic of interest in the diabetes behavioral field because it helps patients to become interested in self-management and to develop plans for action. A general goal of MI is to support and enhance a person’s self-efficacy by focusing on providing opportunities that help individuals assess for themselves what might be important or possible and how change might potentially be achieved (Powell, Hilliard and Anderson, 2014). In numerous studies, motivational interviewing has been demonstrated to be a potentially effective method for facilitating positive behavioral changes in adolescents with T1D (Stranger et al., 2013; Channon, Smith, & Gregory, 2003; Channon et al., 2007; Knight et al., 2003; Viner et al., 2003). MI interventions that promote adherence to one’s diabetes management tasks rather than improving glycemia have demonstrated the greatest results in youth with T1D (Powell, et al., 2014).

Methods

Teens were eligible to test the intervention if they were 13 – 17 years of age, had T1D for more than one year, spoke English (the intervention was in English) and had Internet access. Teens were incentivized to participate with an offer of $5 per week or $20 for completion of the intervention. 

Beta testing of Can-Do-Tude involved a multi-method approach. A quantitative descriptive design was used to evaluate the feasibility of Can-Do-Tude by examining the acceptability (retention and satisfaction). Feedback was provided via a satisfaction survey with regards to usability, purposefulness and recommended improvements. There was a one-group, pre-and post-test to examine, in a limited way, diabetes self-management self-efficacy and intervention implementation. 
Teens were asked questions throughout the intervention. Data was collected from the teens’ responses to those questions and a qualitative descriptive design was utilized to evaluate them. Answer choices and texts were evaluated for the frequency of same answers and the consistency of themes. Summative content analysis was used to quantify and interpret the responses. 

The Intervention

Can-Do-Tude is a 4-week online intervention that uses MI techniques and tailored diabetes self-management education. The intervention was designed using Qualtrics™ survey software and was password protected. Each week focuses on a different topic: improving blood glucose (BG), managing the effect of exercise on BG, managing the effect of food on BG, and recognizing and managing trends seen with BG. For each weekly topic, the participant is assessed (on a scale of 0-10) on their motivation—importance, readiness and confidence—to change behaviors relative to their diabetes self-management. Based on the assessment, the teen would then be taken to a tailored module for their particular level of motivation for the week. This allowed the participant to have an individualized intervention. 

Results

Five adolescents between 14 and 17 years of age were recruited from a diabetes clinic and a diabetes summer camp in Northern Arizona to beta test the intervention. Feasibility with a focus on acceptability (retention and satisfaction) of Can-Do-Tude was evaluated. Implementation was measured, in a limited way, by assessing diabetes self-management self-efficacy. 

Retention of the intervention was measured by the number of participants who logged into Can-Do-Tude each week and the number who completed all four modules of the intervention. Of the five participants, three completed all four weeks of the intervention and two completed 3 weeks of the intervention. One of the individuals who completed only 3 of the weeks of the study stopped the week that the winter holiday break began; the other stopped with the start of school after their summer break. An e-mail reminder had been sent to the two participants encouraging them to finish but neither responded nor logged in to continue. To indicate feasibility, it was desirable to have 75% of the modules completed for all four weeks of the intervention by the five teens. Between the 5 teens there were a total of 20 modules to complete. The participants logged into and completed 18 of the 20 modules, a 90% overall completion rate.

Feasibility in regard to the acceptability of the intervention was also measured using a satisfaction questionnaire. The survey was incorporated into the last module of the intervention, week 4. However, because two of the participants did not complete week 4 of the intervention, there are only three completed satisfaction surveys. Results are shown in Table 1. Participants were asked to rate their responses on a 5-point Likert-type scale, ranging from “not at all” (1) to “very much” (5). The mean scores are reported for each question and a higher mean indicates a higher level of satisfaction. Results showed that the participants, overall, had a positive response to the intervention (M = 4.28, SD = 0.55). The teens reported that the intervention was helpful in managing diabetes (M = 4.67, SD = 0.58); enjoyable (M = 4.0, SD = 1.0); interesting (M = 3.33, SD = 0.58); easy to use (M = 5.0, SD = 0); that the time spent doing the intervention was worthwhile (M = 4.33, SD = 0.58); and that they were more likely to do something different in the management of their diabetes after completing the intervention (M = 4.33, SD = 0.58). The one question that fell short of the desired “mostly” to “very much” satisfaction level asked the participants if the intervention was interesting to them. The participants had no recommendations for the improvement of the intervention Can-Do-Tude. One teen stated that they saw no need for improvements and another said that it was easy to use.

Feasibility related to implementation was measured by a pre-and post-test using the Self-Efficacy for Diabetes Self-Management instrument (Iannotti, et. al., 2006). The instrument was built into the first week’s module and the last module, week 4. Again, since only 3 of the 5 teens completed week 4, only 3 of the instruments can be used for analysis. Analyses of the questions pre-and post-intervention are summarized in Table 2. Participants were asked to rate their responses on a 5-point Likert-type scale, ranging from “not sure at all” (1) to “completely sure” (5). The mean scores were reported for each question and a higher mean indicated a higher level of diabetes self-management self-efficacy. Of the 10 questions, 8 of the items showed improvement in diabetes self-management self-efficacy after completion of the intervention (n = 3). Participants reported improvements in the ability to adjust insulin correctly when eating (M = 4.33, SD = 0.58), to choose healthful foods when going out to eat (M = 4.0, SD = 1.0), to exercise even when they didn’t really feel like it (M = 3.67, SD = 1.15) and to adjust insulin or food accurately based on exercise (M = 4.33, SD = 0.58). They also reported improvements in the ability to talk to their doctor or nurse about any problems with diabetes management (M = 5, SD = 0). When asked about managing diabetes the way the health care team wanted them to, the participants responded favorably (M = 3.67, SD = 1.15). The responses indicated improvements in finding ways to deal with feelings of frustration about diabetes (M = 3.0, SD = 1.73) as well as identifying things that could get in the way of managing diabetes (M = 4.0, SD = 0). There was no change between pre-and post-test scores when asked if they would check blood sugars even when they were really busy (M = 2.67, SD = 1.15). One item had a small decrease in score; it asked whether they were able to manage their diabetes even when they felt overwhelmed (M = 3.33, SD = 1.15). A Wilcoxon signed-rank test indicated that the post-test scores were statistically significantly higher than pre-test scores (Z = -2.952, p = .007).

Qualtrics™ allows for many types of data collection such as short text responses and question types including pick, group and rank, drill down, rank order, heat map and hot spots. Participant’s responses to these questions revealed that the teens felt most confident and ready to make a change in behaviors around diet management (M = 8.57, SD = 2.14), checking BG (M = 8.43, SD = 2.21) and exercise (M = 8.0, SD = 2.37). They had little motivation and interest in identifying patterns with BG (M = 3.0, SD = 3.83), Figure 1. Three themes were identified by all five of the participants, (see Figure 2): they are concerned about the impact of diabetes on their health, they want better BG control and they recognize that they need to check their BG more often. Many of the reasons given for not checking more often were that they did not always have their equipment on them, that they were too busy or rushed to check, and that they needed reminders from parents. Three of the teens made smart goals to check more often.

Discussion

Five teens were utilized to beta test Can-Do-Tude. Although the sample size was small, their responses to the testing is nevertheless very promising. Participants completed most of the modules, were satisfied with the intervention and had an overall increase in diabetes self-management self-efficacy. 

Though promising, there are lessons learned from this beta test – as is the purpose of such testing. First, participants should not be enrolled into the intervention if it will coincide with major events such as the beginning of a holiday break or the beginning of a school year. In the future, the recruitment time frame will not occur during major holidays or other important events that would interfere with a participant’s ability to complete the intervention.

Secondly, the Can-Do-Tude intervention was well received by those who completed the satisfaction survey but one of the items surveyed reported lower than desirable satisfaction. The item surveying whether participants found the intervention ‘interesting’ reported that they thought it was interesting “sometimes”. Although the intervention was developed to engage the teen as much as possible, it is apparent that future edits to the intervention need to be made in order to improve the experience for teens. In addition, since the creation of the intervention, teens have moved more to mobile delivery technology as opposed to computer-delivered technology. Can-Do-Tude needs to be adapted for mobile delivery.

Lastly, qualitative data was analyzed to get a sense of how the participants would use and respond to the intervention. Participants were intentional in their responses and used the intervention as it was designed to be used. Qualitative results provide insight into what the teens were concerned about and their motivation in regards to diabetes behaviors: they are very concerned about their health and the impact diabetes may have on it and they want to improve their BG control to protect their health. They recognize that the number one thing that they could do differently to improve their diabetes control is to check BGs more often. Diabetes self-management self-efficacy demonstrated weaknesses in them checking their blood sugars when really busy and finding ways to deal with diabetes tasks when overwhelmed. These results have been similarly found in another study (Ye, Jeppson, Kleinmaus, Kliems, Schopp and Cox, 2017). Many of the reasons given for not checking BG more often were that they did not always have their equipment with them, they were too busy or rushed to check, or that they needed reminders from parents. Improvements to Can-Do-Tude will consider these results and will include MI language that involves conversations surrounding changes to behavior that may increase BG checks and improvements in BG that would therefore protect their health.
The teens felt that the intervention was easy to use. Furthermore, they felt that the intervention was helpful in managing their diabetes. Additionally, most felt that the time spent doing the intervention was worthwhile and that they would do something different about managing their diabetes. The fact that participants rated these items positively is indeed encouraging; and if only for this reason, this innovative study warrants further testing.

This was a beta test of an online intervention for adolescents with T1D. Results from beta testing will be used to improve and refine the intervention. Findings from this first phase of intervention testing support moving to phase 2 testing with an initial test of the intervention in comparison with an appropriate alternative. A small randomized trial (with a sample size of 30) will be used to identify the intermediate outcome of diabetes self-management self-efficacy and the intervention’s final outcome, improved adherence to diabetes self-management work that will keep adolescents with type 1 diabetes healthy. 

Paul, L. & Moreno-Perez, A. (Winter 2019). Beta Testing Can-Do-Tude: An Online Intervention Using the Principles of Motivational Interviewing for Adolescents with Type 1 Diabetes. Online Journal of Nursing Informatics (OJNI), 23(1). Available at http://www.himss.org/ojni

Citation: Paul, L. & Moreno-Perez, A. (Winter 2019). Beta Testing Can-Do-Tude: An Online Intervention Using the Principles of Motivational Interviewing for Adolescents with Type 1 Diabetes. Online Journal of Nursing Informatics (OJNI), 23(1). Available at http://www.himss.org/ojni

The views and opinions expressed in this blog or by commenters are those of the author and do not necessarily reflect the official policy or position of HIMSS or its affiliates.

Online Journal of Nursing Informatics

Powered by the HIMSS Foundation and the HIMSS Nursing Informatics Community, the Online Journal of Nursing Informatics is a free, international, peer-reviewed publication that is published three times a year and supports all functional areas of nursing informatics.

Read the Latest Edition