Habit Formation Intervention to Improve Type 2 Diabetes Self-Management Behaviors: A Feasibility Study

Abstract

Importance: Self-management of lifestyle behaviors is the recommended focus for interventions to address Type 2 diabetes. Habit formation is an effective approach to changing personal behaviors, but evidence of success for Type 2 diabetes is limited.

Objective: To examine the feasibility and preliminary effectiveness of occupation-based habit formation interventions for improving diabetes self-management behaviors.

Design: Single-subject design with multiple participants providing 4 wk of baseline data followed by 10 wk of intervention data.

Setting: Individual telehealth sessions.

Participants: Eight adults (ages 29–75 yr) with Type 2 diabetes, who had access to a telephone and who were not involved in other diabetes-related education or interventions voluntarily, enrolled into the study.

Intervention: Participants engaged in 10 wk of habit formation intervention focused on four diabetes self-management domains: nutrition, blood glucose monitoring, medication management, and physical activity.

Outcomes and Measures: Data gathered included findings on measures of diabetes self-care behaviors and habit formation.

Results: There was a significant change in self-care behaviors for 6 of the 8 participants (p < .05 for 1 participant, and p < .01 for 5 participants). Group changes were statistically significant (p < .001). Habit strength significantly improved for all areas of diabetes self-management (p < .001 for nutrition, blood glucose monitoring, and medication management and p = .001 for physical activity).

Conclusions and Relevance: Findings suggest that the occupation-based intervention was feasible and showed promise for developing self-management behaviors.

What This Article Adds: Habits are considered foundational to occupations, yet application of the science of habit formation is often not well understood by occupational therapists. This study considered the theoretical components of habit formation that have been neglected by prior studies and demonstrates the feasibility and preliminary effect estimates of a habit formation intervention when used with people with Type 2 diabetes.

The study findings suggest that an occupation-based habit formation intervention is feasible and shows promise for improving self-management behaviors for people with Type 2 diabetes.

Approximately 10.5% of people in the United States (i.e., nearly 34.2 million) have diabetes (Centers for Disease Control and Prevention [CDC], 2020), and rates are projected to triple by 2060 (Lin et al., 2018). The total direct and indirect costs of diagnosed diabetes in the United States in 2022 was $412.9 billion, with the medical expenses of people with diabetes being approximately 2.6 times higher than those of people without diabetes (Parker et al., 2023). Diabetes has been associated with an increased risk for other diseases, including hypertension (75%), dyslipidemia (66%), end-stage renal disease (44%), and nontraumatic lower limb amputation (60%; National Institute of Diabetes and Digestive and Kidney Diseases, 2017; Schub & Parks-Chapman, 2017). Furthermore, people with Type 2 diabetes mellitus (T2DM) have a decreased quality of life and increased risk for depression (Schram et al., 2009), with a prevalence of depression as high as 40% among patients with diabetes (Haltiwanger & Galindo, 2013). Investigators have also shown that affected people function poorly in occupational roles, from basic self-care to social participation and physical performance of activities (Feng & Astell-Burt, 2017; Levterova et al., 2018; Li et al., 2011).

Given the high prevalence and significant consequences of T2DM, interventions are vital. Individualized and culturally relevant occupational therapy services have been shown to improve knowledge, attitudes, well-being, blood glucose levels, quality of life, and overall disease management among diverse groups by addressing performance in meaningful activities, from self-care to community involvement (Atler et al., 2018; Haltiwanger & Brutus, 2012; Haltiwanger & Galindo, 2013; Hreha & Noce, 2018; Piven, 2015; Piven & Duran, 2014; Pyatak et al., 2018). However, the available evidence for such interventions suggests low to moderate strength because of small sample sizes, lack of randomized controls, and lack of long-term follow-up. Additionally, there continues to be a lack of clear guidance on the most effective strategies for intervention.

One such intervention that has gained traction in scientific literature is using habit modification and formation strategies to improve Type 2 diabetes self-management (DSM; Phillips et al., 2016). Because people with T2DM are expected to modify unhealthy behaviors while building new knowledge and skills, establishing healthy habits holds promise as an effective intervention strategy (Fritz, 2014), and health behaviors have been deemed more beneficial when they become habits (Marchant et al., 2018; Phillips et al., 2016). Habits allow people to be efficient with self-management behaviors by creating automatic responses while extinguishing prior habits that do not support health (Phillips et al., 2016). Stronger habits have been shown to support the adoption of health behaviors, including eating a healthy diet (Wiedemann et al., 2014), engaging in adequate physical activity (Gardner, Abraham et al., 2012), and medication adherence (Bolman et al., 2011; Phillips et al., 2013). This means that people with T2DM should be trained to develop strong occupational habits, such as eating healthy diets, engaging in regular physical activity, taking medications as prescribed, and monitoring blood glucose consistently (Fritz, 2014; Phillips et al., 2016).

With personal behaviors being one of the most significant risk factors in developing or having uncontrolled T2DM, occupational therapy practitioners could play an important role in addressing health promotion, disease prevention, and self-management through instruction in lifestyle habit modification and formation. However, occupational therapy practitioners are often not part of the diabetes management teams, and the evidence for the effectiveness of occupation-based habit formation interventions is scarce. This study aimed to explore the feasibility and preliminary effectiveness of an occupation-based habit formation intervention in promoting healthy self- management behaviors among adults with T2DM. The interventions incorporated theoretically informed components that have been neglected in prior studies, providing a greater depth of knowledge related to habit formation.

Method

Conceptual Framework

Habits are actions or behaviors that become automatic and effortless responses, triggered by context (Gardner, Lally et al., 2012). Habit formation is the process of developing these habits (Gardner, Lally et al., 2012). The intervention in this study was guided by literature on habit formation and involved education about the phases of habit formation, including the use of implementation intentions, or action plans; the importance of context and contextual modifications; and a need to identify discrete performance situations of simple behaviors as components of occupations.

Study Design

This study used an AB single-subject (N = 1) design to provide control parameters that are appropriate for a feasibility study and to illuminate the anticipated behavior changes (Franklin et al., 2014; Lobo et al., 2017). All study procedures were approved by the Creighton University and University of South Dakota institutional review boards. The trial was registered at ClinicalTrials.gov (NCT05455242). On the basis of recommendations by Julious (2005), we aimed to recruit 12 participants for the study.

Participants

Participants were recruited with the assistance of community diabetes educators. We also sent recruitment emails to targeted sites (e.g., wellness facilities, senior centers) and posted a call for volunteers on social media pages. Interested people contacted the first author (Diana R. Feldhacker), who scheduled individual phone calls with them to discuss the study, answer questions, and verify that they met the inclusion criteria. Participants were included if they were at least 19 yr old, were able to read and write English, were able to participate in virtual or telephone sessions, could be reached by phone or text messaging, had a diagnosis of T2DM, and were willing and able to participate in all study activities. They could not be concurrently involved in other diabetes-related education or interventions, and they had to indicate that they had foundational knowledge of T2DM and DSM (e.g., having taken a diabetes education class or met one-on-one with a diabetes educator).

Measurement and Outcomes

Feasibility Measures

Feasibility was assessed by evaluating the number of participants who were approached about the study versus those who consented to participate, with the feasibility threshold set a priori at 50% on the basis of similar studies. The feasibility threshold for the percent retained through final data collection was set at 80% on the basis of prior, similar studies (e.g., Higgins et al., 2011). Feasibility of recruitment, data collection, outcome measures, and study procedures were assessed through open-ended interviews with participants.

Self-Care Behaviors

Self-care behaviors were assessed with the Summary of Diabetes Self Care Activities (SDSCA; Toobert et al., 2000). The SDSCA is an 11-item self-report questionnaire that requires respondents with T2DM to indicate the frequency with which they performed diabetes self-care behaviors related to diet, exercise, blood glucose monitoring, foot care, and smoking in the previous 7 days (Toobert et al., 2000). The SDSCA has been found to be reliable for assessing DSM with moderate test–retest correlations (mean r = .40) and good criterion validity (Toobert et al., 2000). There is strong evidence for structural validity and internal consistency reliability (α ≥ .70 for all subscales except the foot care subscale; Lee et al., 2020). We used the combined subscales to assess general diet, specific diet, exercise, and blood glucose testing using a global average of Items 1 to 8, reported in days per week.

Habit Formation

Habit formation was assessed with the Self-Report Behavioral Automaticity Index (SRBAI; Gardner, Abraham, et al., 2012), a measure used to assess habit strength and automaticity (Gardner, Abraham, et al., 2012). Respondents indicated how they perform a targeted behavior, on a scale ranging from 1 (strongly disagree) to 7 (strongly agree), for four statements: I “do without thinking,” I “do automatically,” I “do without having to consciously remember,” and I “do before I realize I am doing it” (Gardner, Abraham, et al., 2012, p. 4). The SRBAI has been found to be reliable, valid, and sensitive in measuring the characteristics of habits, including habit–behavior correlation and intention–behavior relationship (Gardner, Abraham, et al., 2012). One criticism of the SRBAI, along with the Self-Report Habit Index (SRHI) on which it was based (Gardner, Abraham, et al., 2012), is that it lacks critical contextual components, which are an inherent defining factor in habit formation (Fritz & Cutchin, 2016; Gardner, 2015). Sniehotta and Presseau (2012) recommended an adaptation of the SRHI by incorporating contextual cues such as “Behavior X in Context Y at Time Z is something . . .” (p. 140). Gardner, Abraham, et al. (2012) acknowledged the need for contextual components, and one was included within their study of validity. They concluded that the SRBAI met the same criteria by which the SRHI was judged and used. We adopted the recommended modification of the SRBAI in this study to fully capture the cue and context dependence of habits.

Procedure

We conducted the study over 13 wk: 4 wk of baseline data gathering (Phase A) and 10 wk of intervention (Phase B), with an overlap of Baseline Week 4 and Intervention Week 1. The SDSCA was administered online weekly for 4 wk in Phase A. In Week 4, the final baseline SDSCA was administered, and the intervention phase of the study began. Each participant was interviewed to establish their occupational profile and identify specific DSM goals. On the basis of the habit formation theory described earlier, education on habit formation and behavior modification was provided. Where indicated, participants and the investigator discussed nutrition issues as they related to diabetes. Participants were guided in setting a context-specific implementation intention for habit formation of a simple occupational habit related to nutrition. This nutrition intention was used to create a target behavior for the SRBAI, which was administered at the end of the first intervention session.

The intervention was delivered by the first author, who is an occupational therapist with advanced training in habit formation and diabetes management. Interventions occurred once a week for 10 wk. Each weekly session was administered virtually (by telephone or on the Zoom platform) and lasted for approximately 30 to 60 min. Sessions began with the administration of the SDSCA and SRBAI. For habit formation, participants were instructed on ongoing context-specific implementation intentions for participation in DSM occupations. The first and second intervention sessions focused on nutrition. At Intervention Weeks 3, 5, and 7, an additional area of DSM was included: blood glucose monitoring, medication management, and physical activity, respectively. During these weeks, additional SRBAIs were administered as DSM intentions were added. We used instruction and collaborative problem-solving to encourage the participants to independently use their environment for cueing to promote success in habit formation. However, during Intervention Weeks 1, 3, 5, and 7, we used regular reminders through text messaging to provide additional contextual cues for the performance of behaviors related to the areas of DSM. By Week 7, all areas of DSM that were to be addressed in this study had set intentions. Weekly intervention sessions continued to facilitate habit formation through Week 10. Studies of habit formation have found that habit strength (automaticity) increases through the initial practice but plateaus after a certain number of repetitions (Haith & Krakauer, 2018). One of the most robust studies of the timing of habit formation, to date, found that the average number of days for automaticity to be reached was 66 (Lally et al., 2010). Thus, Lally et al. (2010) recommended that behaviors be regularly practiced for about 10 wk for habits to be formed, which is what we followed for our intervention phase. At Week 10, the final data were collected, including the final administration of the SDSCA and SRBAIs for all four DSM areas. At the conclusion of the final intervention, the first author conducted an open-ended interview with each participant to gather feedback on the feasibility components of the study.

Data Analysis

A global average score for Items 1–8 of the SDSCA was computed. A mean score of 0 indicated that the participant did not perform diabetes self-care activities, whereas a score of 7 indicated that the participant performed activities in all areas 7 days/wk. Responses on Item 4 were reverse scored for consistent reflection of positive self-care. SDSCA data were first analyzed visually by graphing baseline and intervention raw scores for each participant (Lobo et al., 2017). The graphed data were analyzed using 2- and 3-SD-band methods to assess the variability and significance of the change in behavior (Bloom et al., 2009; Nourbakhsh & Ottenbacher, 1994; Portney & Watkins, 2015). Baseline (Phase A, control: four data points) and intervention (Phase B: nine data points) data were plotted for each participant. The mean of the baseline scores was calculated, and a horizontal line representing this mean was drawn through the graph. The standard deviation of the baseline scores was calculated and multiplied by 2 and 3. Two- and 3-SD bands were created by drawing a horizontal line above and below the mean, respectively, through the graph (Bloom et al., 2009; Nourbakhsh & Ottenbacher, 1994; Portney & Watkins, 2015). At least two consecutive intervention phase data points falling outside the bands indicated a significant change in that variable for the participant (Gottman & Leiblum, 1974). The probability of this happening by chance is less than the criterion of p < .05 for the 2-SD-band method (Nourbakhsh & Ottenbacher, 1994) and p < .01 for the 3-SD-band method (Bloom et al., 2009).

To determine whether there was a main effect of the intervention on diabetes self-care behaviors, as measured with the SDSCA, we analyzed the combined sample data using Friedman’s repeated measures analysis of variance (ANOVA). The assumptions of independence of observations and continuous distributions were tested and met. We used the robust statistical Monte Carlo procedures to improve the accuracy of results with this small set of data, with 10,000 sample iterations, as is recommended for Friedman’s test with an N ≤ 30. The Monte Carlo method is typically used to provide an unbiased estimate of the exact p value through repeated sampling with replacement (Harrison, 2010). We used the 95% confidence interval. Finally, we conducted orthogonal pairwise post hoc comparisons using Wilcoxon signed-rank tests with Bonferroni adjustment (p = .05, 9 comparisons = 0.006). We computed the effect size (r) by dividing the Z value by the square root of the sample size (√n). Cohen (1992) proposed the following effect sizes: r = .1 (small effect), r = .3 (moderate effect), and r > .5 = (large effect).

The weekly SRBAI score was calculated by average ratings on the four items. A mean score of 1 indicated weak habit strength, whereas a score of 7 indicated strong habit strength. Data were analyzed using Freidman’s repeated measures ANOVA to determine the main effect of interventions on habit formation. The variables measured with the SRBAI during the intervention phase were the four DSM areas: nutrition, blood glucose monitoring, medication management, and physical activity. We conducted orthogonal post hoc pairwise comparisons using the same robust statistical procedures described earlier. Pairwise alphas were as follows: nutrition = .006, blood glucose monitoring = .007, medication management = .01, and physical activity = .02.

Results

Eight people were enrolled in the study (M age = 55.6, SD = 15.9; 75% female). The final sample demographic characteristics and recruitment strategy are outlined in Table 1.

Table 1.

Participant Demographics and Recruitment Strategy

Participant	Age, yr	Race	Gender	Income Range, $	No. in Household, Lives With Whom	Time Since Diagnosed With T2DM, yr	T2DM Current Medication	Location^a	Recruitment Strategy
1	65	Hispanic/ Latino	Male	75,000–99,999	2, spouse	14	Insulin	Rural	PCP
2	52	White	Female	50,000–74,999	2, spouse	5	Semaglutide injection	Rural	PCP
3	31	White	Female	75,000–99,999	1, alone	1	Semaglutide injection	Rural	PCP
4	29	White	Female	NR	3, parents	2	Insulin	Metropolitan	Diabetes educator
5	65	White	Female	20,000–34,999	2, sibling	1	Metformin and dulaglutide injection	Rural	PCP
6	62	White	Female	35,000–49,999	2, sibling	2	Metformin	Rural	PCP
7	75	White	Male	>100,000	2, spouse	14	Metformin	Metropolitan	Wellness program for adults
8	66	White	Female	50,000–74,999	1, alone	20	Metformin, glimepiride, and liraglutide injection	Metropolitan	Wellness program for adults

Note. NR = no response; PCP = primary care physician; T2DM = Type 2 diabetes mellitus.

^aLocation was assigned as defined by Health Resources and Services Administration (2022): metropolitan, >50,000 people in urban core; micropolitan, 10,000–49,999 people in urban core; and rural = areas outside of micropolitan and metropolitan areas.

Feasibility

The overall recruitment rate was 72.7%, with 62.5% of participants recruited through their primary care provider, 12.5% recruited through a diabetes educator, and 16.7% recruited through an adult wellness program. Eleven people agreed to participate, and 3 withdrew after consent and before data collection because of concerns about time commitment. All participants completed the study activities (no attrition), attended 100% of the weekly intervention sessions, and completed the open-ended interview after the final intervention. Both recruitment (50%) and retention (80%) targets for feasibility were met. Data were collected for diabetes self-care and habit formation outcomes for 100% of participants. All participants reported that they found the SDSCA easy to understand and relevant. Most participants reported that the items on the SRBAI were not easy to distinguish from each other. During the intervention, the level of engagement varied among participants. Two participants were self-directed and goal-oriented, requiring little interaction with the interventionist during weekly sessions (20–25 min/wk), whereas 6 required more guidance to set goals and modify their behavior (60 min/wk). All participants required assistance in setting implementation intentions consistent with the theoretical principles of habit formation.

Overall, feedback on study procedures was strongly positive. Seven of the participants specifically reported that the positive and supportive interactions with the therapist were key to their success. This interaction included direct weekly coaching, individualized follow-up emails after sessions to provide further information and resources, and reminders by text messages to complete tasks. All participants reported that they felt that 10 wk was an appropriate duration and that telehealth was an appropriate service delivery method for the intervention. Five of the 8 participants stated that they could not participate in this study if they had been required to attend weekly face-to-face sessions for 10 wk.

Self-Care Behaviors

The graphed SDSCA data can been seen in Figure A.1 (available online with this article at https://research.aota.org/ajot). For each graph, the lowest line signifies the baseline mean, and the middle and highest lines indicate 2- and 3-SD bands around the mean. There was a significant change in self-care activities among 5 of the 8 participants (effect of intervention on this variable was significant at p < .01 for 63% of the study participants). For one person (Participant 7), there was a clinically significant improvement in this variable at p < .05 (2-SD-band criterion). No significant change in this variable was detected for 2 participants (Participants 2 and 4). Participant 3 experienced an improvement in the outcome variables early in Intervention Week 3 followed by a slight decline and subsequent significant improvement in Week 7. The average time from the beginning of the intervention phase to significant change at p < .05 was 6.17 wk and 7.8 wk at p < .01.

There was a main effect of intervention on self-care behaviors as measured on the SDSCA, χ²(12) = 71.21, p < .001. Orthogonal pairwise comparisons indicated significant differences in mean scores between baseline and all weeks in the intervention except Week 3. There was a large effect of intervention, r = .89. See Table 2 for orthogonal pairwise comparisons.

Table 2.

Wilcoxon Signed-Rank Test Pairwise Comparison From Average Baseline to Intervention Weeks for SDSCA

Statistic	Comparison of Average Baseline With Intervention Week
Statistic	2	3	4	5	6	7	8	9	10
Z	−2.52	−2.38	−2.52	−2.52	−2.52	−2.52	−2.52	−2.52	−2.52
p	.004*	.007	.004*	.004*	.004*	.004*	.004*	.004*	.003*
r	.89	.84	.89	.89	.89	.89	.89	.89	.89

Note. Significance values are based on 10,000 sampled tables. SDSCA = Summary of Diabetes Self-Care Activities.

*p ≤ .006.

Habit Formation

We evaluated mean ranks for each of the four intervention areas—nutrition, blood glucose monitoring, medication management, and physical activity—using the SRBAI. Measures of central tendency indicated a positive trend of improved automaticity for each area of DSM. There was a main effect of intervention on habit formation for each DSM area as measured on the SRBAI: nutrition, χ²(9) = 58.23, p < .001; blood glucose monitoring, χ²(7) = 40.02, p < .001; medication management, χ²(5) = 26.46, p < .001; and physical activity, χ²(3) = 14.06, p = .001. Orthogonal pairwise comparisons indicated significant differences in median scores between the initial intervention week and Intervention Weeks 3–10 for nutrition; Weeks 6, 7, 8, and 10 for blood glucose monitoring; and Weeks 8–10 for physical activity. There was a large effect of intervention on each of the four areas of intervention: nutrition, r = .89; blood glucose monitoring, r = .84; and physical activity, rs = .80, .78, and .84 respectively. There were no significant differences in median scores for medication management. See Table 3 for orthogonal pairwise comparisons.

Table 3.

Wilcoxon Signed-Rank Test Pairwise Comparisons of Initial With Subsequent Intervention Weeks for SRBAI

DSM Area on the SRBAI	Initial Intervention Week/Subsequent Intervention Week Comparisons
Nutrition	Initial Week 1/Subs. Week 2: Z = −2.38 p = .008 r = .84	Initial Week 1/Subs. Week 3: Z = −2.53 p = .004* r = .89	Initial Week 1/Subs. Week 4: Z = −2.52 p = .006* r = .89	Initial Week 1/Subs. Week 5: Z = −2.52 p = .004* r = .89	Initial Week 1/Subs. Week 6: Z = −2.52 p = .003* r = .89	Initial Week 1/Subs. Week 7: Z = −2.52 p = .004* r = .89	Initial Week 1/Subs. Week 8: Z = −2.52 p = .004* r = .89	Initial Week 1/Subs. Week 9: Z = −2.53 p = .003* r = .89	Initial Week 1/Subs. Week 10: Z = −2.53 p = .004* r = .89
Blood glucose monitoring			Initial Week 3/Subs. Week 4: Z = −2.20 p = .015 r = .78	Initial Week 3/Subs. Week 5: Z = −2.21 p = .017 r = .78	Initial Week 3/Subs. Week 6: Z = −2.37 p = .007** r = .84	Initial Week 3/Subs. Week 7: Z = −2.38 p = .007** r = .84	Initial Week 3/Subs. Week 8: Z = −2.38 p = .006** r = .84	Initial Week 3/Subs. Week 9: Z = −2.39 p = .009 r = .84	Initial Week 3/Subs. Week 10: Z = −2.38 p = .007** r = .84
Medication management					Initial Week 5/Subs. Week 6: Z = −2.23 p = .014 r = .79	Initial Week 5/Subs. Week 7: Z = −2.23 p = .016 r = .79	Initial Week 5/Subs. Week 8: Z = −2.26 p = .014 r = .80	Initial Week 5/Subs. Week 9: Z = −2.26 p = .014 r = .80	Initial Week 5/Subs. Week 10: Z = −2.23 p = .012 r = .79
Physical activity							Initial Week 7/Subs. Week 8: Z = −2.26 p = .015*** r = .80	Initial Week 7/Subs. Week 9: Z = −2.21 p = .016*** r = .78	Initial Week 7/Subs. Week 10: Z = −2.37 p = .009*** r = .84

Note. Significance values are based on 10,000 sampled tables. DSM = diabetes self-management; SRBAI = Self-Report Behavioral Automaticity Index; Subs. = subsequent.

*p ≤ .006; **p ≤ .007; ***p ≤ .02.

Discussion

The purpose of this study was to evaluate the feasibility of a habit-formation approach to increasing the automaticity of DSM behaviors and thereby improve diabetes self-care. Our study advances knowledge about the application of habit science to everyday occupation in the following ways. First, our results indicated a significant improvement in group scores of diabetes self-care behaviors, as measured on the SDSCA, and automaticity of habit formation for nutrition, blood glucose monitoring, medication management, and physical activity. This is similar to the findings of prior research reports that intentional development of stronger habits for healthy behaviors improved health and well-being, including dietary behavior (Gardner et al., 2014; Wiedemann et al., 2014), physical activity (Gardner, Abraham et al., 2012; Marchant et al., 2018), and medication adherence (Badawy et al., 2020; Bolman et al., 2011; Phillips et al., 2013).

Another important contribution of this work is that the success of the current intervention suggests an important role for occupational therapy as part of the DSM care team. Despite positive findings in prior studies, occupational therapy practitioners are not often part of the chronic disease management team (Ahmed et al., 2015). For example, Hwang et al. (2009) found that only 13.7% of community-dwelling older adults with diabetes received occupational therapy services. Instead, people typically receive diabetes education services through a diabetes educator or nurse in consultation with a dietician. Traditional diabetes education has been shown to be helpful in improving people’s knowledge of diabetes (Rise et al., 2013). However, the degree to which this knowledge is incorporated into daily life is unclear (Weller et al., 2017). During their occupational profile interviews, participants in this study reported a disconnect with typical diabetes education services, which were most often group or one-on-one sessions with a diabetes educator. The use of occupational therapy intervention for habit formation in this study seemed to help bridge this gap through interventions that facilitated the application and integration of diabetes education into areas of daily life, including nutrition choices for meal preparation and grocery shopping, monitoring blood glucose, managing medications as prescribed, and engaging in physical activity. Those occupation-based habit formation intervention strategies were consistent with the most updated Consensus Report recommendations, which states that “personalized and comprehensive methods are necessary to promote effective self-management required for day-to-day living with diabetes” (Powers et al., 2020, p. 1639).

Another key result of the study is the degree to which our results align with those of existing studies on the timing of habit formation. For example, Keller et al. (2021) demonstrated that it took a median of 59 days for participants who formed habits to reach peak automaticity. This result was similar to Lally et al.’s (2010) finding of an average of 66 days to reach peak automaticity. Lally et al. recommended that behaviors be practiced for approximately 10 wk for habits to be formed. That length of time proved to be appropriate for participants in this study. When reviewing single-subject analyses for SDSCA data, individual significance was noted after an average of 6.17 wk (median = 6 wk) of intervention (p < .05, 2-SD-band method) and 7.8 wk (median = 9 wk) of intervention (p < .01, 3-SD-band method), which is consistent with findings in these prior studies.

Pairwise comparison of self-care data indicated a statistically significant improvement between baseline scores and all intervention weeks except Week 3. This is consistent with the theoretical model of habit formation by Haith and Krakauer (2018), which suggests an initial steep rise in skill and habit across a continuum of ongoing practice. For habit formation outcomes (measured using the SRBAI), pairwise comparisons indicated a significant difference between the initial intervention week of the DSM area and subsequent intervention weeks after 2 wk of intervention for nutrition, 3 wk for blood glucose monitoring, and 1 wk for physical activity. These findings are consistent with a steep initial learning phase in the model proposed by Haith and Krakauer (2018).

Limitations

A primary limitation of this study is the small sample size. However, single-subject studies allow participants to serve as their own comparison during data collection, thus controlling for many confounding variables that can affect outcomes (Franklin et al., 2014; Lobo et al., 2017). We also replicated the study across multiple participants, which strengthened our trust in our conclusions (Lobo et al., 2017). This design also proved useful in furthering our understanding of the process of behavioral change (Franklin et al., 2014). Despite the strengths of choosing a single-subject design, generalizability is limited. We cannot conclude with certainty that habit formation interventions are an effective intervention for people with T2DM in general. Future studies, with larger, more diverse samples to improve external validity, are needed (Janosky, 2005; Lobo et al., 2017). An attempt to overcome this limitation was the use of robust statistical Monte Carlo procedures during analysis to improve the accuracy of the results for combined data.

A second limitation is the repeated exposure of participants to outcomes measures, which can affect participant response (Kratochwill et al., 2010). However, replicating the study across multiple participants provided some control for this effect. All assessments relied on participant self-report or self-rating, which posed inherent potential bias.

Finally, this study was limited by a short duration and lack of follow-up. The intervention was 10 wk long, as guided by habit formation theoretical models. However, no follow-up with the participants was conducted. A long-term follow-up would have provided insights into whether behavioral changes endured over time.

Implications for Occupational Therapy Practice

Overall, the prevalence of T2DM continues to rise. This rise has significant implications for individual and population health because of increasing medical expenses, the risk of other diseases, decreased quality of life, and poor functioning in occupational roles. With personal behaviors being one of the most significant risk factors, there is a need for interventions that address health promotion, disease prevention, and self-management.

▪ The theoretical principles of habit formation were supported in the outcomes of this study. Occupational therapists who want to promote habit formation should understand and follow these principles.

▪ Occupational therapists should advocate for inclusion in the DSM interprofessional team and use studies that support habit-formation approaches as evidence to support this role.

▪ Those in academia who prepare occupational therapy students for future practice should incorporate content on the science of habit formation and its application in the service of occupational engagement.

Conclusion

We demonstrated the feasibility of the occupation-based habit formation intervention approach used in this study. This intervention is promising for promoting diabetes self-care and behavior change among people with T2DM. We recommend larger randomized trials with long-term follow-up to better understand the treatment effects of this approach. This and future research have the potential to further build the evidence for the inclusion of occupational therapy practitioners on diabetes management teams.

Supplemental Material

Supplementary material for Habit Formation Intervention to Improve Type 2 Diabetes Self-Management Behaviors: A Feasibility Study

Supplementary material, sj-pdf-1-aot-10.5014_ajot.2023.050351.pdf for Habit Formation Intervention to Improve Type 2 Diabetes Self-Management Behaviors: A Feasibility Study by Diana R. Feldhacker, Moses N. Ikiugu, Heather Fritz, William E. Schweinle and Hongmei Wang in The American Journal of Occupational Therapy

References

Ahmed

Ware

Visca

Bareil

Chouinard

M. C.

Desforges

, et al . Gogovor, A. (2015). The prevention and management of chronic disease in primary care: Recommendations from a knowledge translation meeting. BMC Research Notes, 8, 571. https://doi.org/10.1186/s13104-015-1514-0

Atler

K. E.

Schmid

A. A.

Klinedinst

T. C.

Grimm

L. A.

Marchant

T. P.

Marchant

D. R.

, & Malcolm

M. P.

(2018). The relationship between quality of life, activity and participation among people with type 2 diabetes mellitus. Occupational Therapy in Health Care, 32, 341–362. https://doi.org/10.1080/07380577.2018.1522017

Badawy

S. M.

Shah

Beg

, & Heneghan

M. B.

(2020). Habit strength, medication adherence, and habit-based mobile health interventions across chronic medical conditions: Systematic review. Journal of Medical Internet Research, 22, e17883. https://doi.org/10.2196/17883

Bloom

Fischer

, & Orme

J. G.

(2009). Evaluating practice: Guidelines for the accountable professional (6th ed.). Pearson Education.

Bolman

Arwert

T. G.

, & Völlink

(2011). Adherence to prophylactic asthma medication: habit strength and cognitions. Heart and Lung: The Journal of Acute and Critical Care, 40, 63–75. https://doi.org/10.1016/j.hrtlng.2010.02.003

Centers for Disease Control and Prevention. (2020). National diabetes statistics report, 2020: Estimates of diabetes and its burden in the United States. U.S. Department of Health and Human Services.

Cohen

(1992). Statistical power analysis. Current Directions in Psychological Science, 1, 98–101. www.jstor.org/stable/20182143. https://doi.org/10.1111/1467-8721.ep10768783

Feng

, & Astell-Burt

(2017). Impact of a type 2 diabetes diagnosis on mental health, quality of life, and social contacts: A longitudinal study. BMJ Open Diabetes Research and Care, 5, e000198. https://doi.org/10.1136/bmjdrc-2016-000198

Franklin

R. D.

Allison

D. B.

, & Gorman

B. S.

(Eds.). (2014). Design and analysis of single-case research. Psychology Press. https://doi.org/10.4324/9781315806402

10.

Fritz

(2014). The influence of daily routines on engaging in diabetes self-management. Scandinavian Journal of Occupational Therapy, 21, 232–240. https://doi.org/10.3109/11038128.2013.868033

11.

Fritz

, & Cutchin

M. P.

(2016). Integrating the science of habit: Opportunities for occupational therapy. OTJR: Occupation, Participation and Health, 36, 92–98. https://doi.org/10.1177/1539449216643307

12.

Gardner

(2015). A review and analysis of the use of ‘habit’ in understanding, predicting and influencing health-related behaviour. Health Psychology Review, 9, 277–295. https://doi.org/10.1080/17437199.2013.876238

13.

Gardner

Abraham

Lally

, & De Bruijn

G. J.

(2012). Towards parsimony in habit measurement: Testing the convergent and predictive validity of an automaticity subscale of the Self-Report Habit Index. International Journal of Behavioral Nutrition and Physical Activity, 9, 102. https://doi.org/10.1186/1479-5868-9-102

14.

Gardner

Lally

, & Wardle

(2012). Making health habitual: The psychology of ‘habit-formation’ and general practice. British Journal of General Practice, 62, 664–666. https://doi.org/10.3399/bjgp12X659466

15.

Gardner

Sheals

Wardle

, & McGowan

(2014). Putting habit into practice, and practice into habit: A process evaluation and exploration of the acceptability of a habit-based dietary behaviour change intervention. International Journal of Behavioral Nutrition and Physical Activity, 11, 135. https://doi.org/10.1186/s12966-014-0135-7

16.

Gottman

J. M.

, & Leiblum

S. R.

(1974). How to do psychotherapy and how to evaluate it. Holt, Rinehart & Winston.

17.

Haith

A. M.

, & Krakauer

J. W.

(2018). The multiple effects of practice: Skill, habit and reduced cognitive load. Current Opinion in Behavioral Sciences, 20, 196–201. https://doi.org/10.1016/j.cobeha.2018.01.015

18.

Haltiwanger

E. P.

, & Brutus

(2012). A culturally sensitive diabetes peer support for older Mexican-Americans. Occupational Therapy International, 19, 67–75. https://doi.org/10.1002/oti.320

19.

Haltiwanger

E. P.

, & Galindo

(2013). Reduction of depressive symptoms in an elderly Mexican-American female with type 2 diabetes mellitus: A single-subject study. Occupational Therapy International, 20, 35–44. https://doi.org/10.1002/oti.1338

20.

Harrison

R. L.

(2010). Introduction to Monte Carlo simulation. AIP Conference Proceedings, 1204, 17–21. https://doi.org/10.1063/1.3295638

21.

Health Resources and Services Administration. (2022). Defining rural population . https://www.hrsa.gov/rural-health/about-us/what-is-rural

22.

Higgins

J. P. T.

Altman

D. G.

Gøtzsche

P. C.

Jüni

Moher

Oxman

A. D.

, et al . Cochrane Bias Methods Group; Cochrane Statistical Methods Group. (2011). The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. British Medical Journal, 343, d5928. https://doi.org/10.1136/bmj.d5928

23.

Hreha

, & Noce

(2018). Implementation of a multidisciplinary diabetes self-management training class in acute rehabilitation. Occupational Therapy in Health Care, 32, 412–421. https://doi.org/10.1080/07380577.2018.1525509

24.

Hwang

J. E.

Truax

Claire

, & Caytap

A. L.

(2009). Occupational therapy in diabetic care-areas of need perceived by older adults with diabetes. Occupational Therapy in Health Care, 23, 173–188. https://doi.org/10.1080/07380570902950259

25.

Janosky

J. E.

(2005). Use of the single subject design for practice based primary care research. Postgraduate Medical Journal, 81, 549–551. https://doi.org/10.1136/pgmj.2004.031005

26.

Julious

S. A.

(2005). Sample size of 12 per group rule of thumb for a pilot study. Pharmaceutical Statistics, 4, 287–291. https://doi.org/10.1002/pst.185

27.

Keller

Kwasnicka

Klaiber

Sichert

Lally

, & Fleig

(2021). Habit formation following routine-based versus time-based cue planning: A randomized controlled trial. British Journal of Health Psychology, 26, 807–824. https://doi.org/10.1111/bjhp.12504

28.

Kratochwill

T. R.

Hitchcock

Horner

R. H.

Levin

J. R.

Odom

S. L.

Rindskopf

D. M.

, & Shadish

W. R.

(2010). Single-case design technical documentation. What Works Clearinghouse. https://ies.ed.gov/ncee/wwc/Document/229

29.

Lally

Van Jaarsveld

C. H. M.

Potts

H. W. W.

, & Wardle

(2010). How habits are formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40, 998–1009. https://doi.org/10.1002/ejsp.674

30.

Lee

E. H.

Chae

, & Kim

C. J.

(2020). Patient-reported outcome measures for diabetes self-care: A systematic review of measurement properties. International Journal of Nursing Studies, 105, 103498. https://doi.org/10.1016/j.ijnurstu.2019.103498

31.

Levterova

B. A.

Orbetzova

Levterov

Dimitrova

, & Todorov

(2018). Assessment of the impact of type 2 diabetes on the quality of life by Audit of Diabetes-Dependent Quality-of-Life (ADDQoL-19). Biotechnology, Biotechnological Equipment, 32, 1627–1634. https://doi.org/10.1080/13102818.2018.1532319

32.

T. C.

Lee

Y. D.

Liu

C. S.

Chen

C. C.

C. I.

, & Lin

C. C.

(2011). Disease-specific quality-of-life measures as predictors of mortality in individuals living with type 2 diabetes. Journal of Psychosomatic Research, 70, 155–160. https://doi.org/10.1016/j.jpsychores.2010.07.003

33.

Lin

Thompson

T. J.

Cheng

Y. J.

Zhuo

Zhang

Gregg

, & Rolka

D. B.

(2018). Projection of the future diabetes burden in the United States through 2060. Population Health Metrics , 16 , 9. https://doi.org/10.1186/s12963-018-0166-4

34.

Lobo

M. A.

Moeyaert

Baraldi Cunha

, & Babik

(2017). Single-case design, analysis, and quality assessment for intervention research. Journal of Neurologic Physical Therapy, 41, 187–197. https://doi.org/10.1097/NPT.0000000000000187

35.

Marchant

Chevance

, & Boiché

(2018). Intention and automaticity toward physical and sedentary screen-based leisure activities in adolescents: A profile perspective. Journal of Sport and Health Science, 7, 481–488. https://doi.org/10.1016/j.jshs.2016.08.006

36.

National Institute of Diabetes and Digestive and Kidney Diseases. (2017). Type 2 diabetes. U.S. Department of Health and Human Services. https://www.niddk.nih.gov/health-information/diabetes/overview/what-is-diabetes/type-2-diabetes

37.

Nourbakhsh

M. R.

, & Ottenbacher

K. J.

(1994). The statistical analysis of single-subject data: a comparative examination. Physical Therapy, 74, 768–776. https://doi.org/10.1093/ptj/74.8.768

38.

Parker

E. D.

Lin

Mahoney

Ume

Yang

Gabbay

R. A.

ElSayed

N. A.

, & Bannuru

R. R.

(2023). Economic costs of diabetes in the U.S. in 2022. Diabetes Care, dci230085. https://doi.org/10.2337/dci23-0085

39.

Phillips

L. A.

Cohen

Burns

Abrams

, & Renninger

(2016). Self-management of chronic illness: The role of ‘habit’ versus reflective factors in exercise and medication adherence. Journal of Behavioral Medicine, 39, 1076–1091. https://doi.org/10.1007/s10865-016-9732-z

40.

Phillips

L. A.

Leventhal

, & Leventhal

E. A.

(2013). Assessing theoretical predictors of long-term medication adherence: Patients’ treatment-related beliefs, experiential feedback and habit development. Psychology and Health, 28, 1135–1151. https://doi.org/10.1080/08870446.2013.793798

41.

Piven

E. F.

(2015). Activity and occupational demands of type two diabetes: The voice of Mexican-American older adults. Physical and Occupational Therapy in Geriatrics, 33, 34–52. https://doi.org/10.3109/02703181.2014.980961

42.

Piven

, & Duran

(2014). Reduction of non-adherent behaviour in a Mexican-American adolescent with type 2 diabetes. Occupational Therapy International, 21, 42–51. https://doi.org/10.1002/oti.1363

43.

Portney

L. G.

, & Watkins

M. P.

(2015). Foundations of clinical research: Applications to practice. F. A. Davis.

44.

Powers

M. A.

Bardsley

J. K.

Cypress

Funnell

M. M.

Harms

Hess-Fischl

, et al . Uelmen, S. (2020). Diabetes self-management education and support in adults With type 2 diabetes: A consensus report of the American Diabetes Association, the Association of Diabetes Care and Education Specialists, the Academy of Nutrition and Dietetics, the American Academy of Family Physicians, the American Academy of PAs, the American Association of Nurse Practitioners, and the American Pharmacists Association. Diabetes Care, 43, 1636–1649. https://doi.org/10.2337/dci20-0023

45.

Pyatak

E. A.

Carandang

Vigen

C. L. P.

Blanchard

Diaz

Concha-Chavez

, et al . Peters, A. L. (2018). Occupational therapy intervention improves glycemic control and quality of life among young adults with diabetes: The Resilient, Empowered, Active Living with Diabetes (REAL Diabetes) randomized controlled trial. Diabetes Care, 41, 696–704. https://doi.org/10.2337/dc17-1634

46.

Rise

M. B.

Pellerud

Rygg

L. O.

, & Steinsbekk

(2013). Making and maintaining lifestyle changes after participating in group based type 2 diabetes self-management educations: A qualitative study. PLoS One, 8, e64009. https://doi.org/10.1371/journal.pone.0064009

47.

Schram

M. T.

Baan

C. A.

, & Pouwer

(2009). Depression and quality of life in patients with diabetes: A systematic review from the European depression in diabetes (EDID) research consortium. Current Diabetes Reviews, 5, 112–119. https://doi.org/10.2174/157339909788166828

48.

Schub

, & Parks-Chapman

(2017). Quick lesson: Diabetes mellitus, Type 2. CINAHL Information Systems.

49.

Sniehotta

F. F.

, & Presseau

(2012). The habitual use of the Self-Report Habit Index. Annals of Behavioral Medicine, 43, 139–140, 141–142. https://doi.org/10.1007/s12160-011-9305-x

50.

Toobert

D. J.

Hampson

S. E.

, & Glasgow

R. E.

(2000). The Summary of Diabetes Self-Care Activities measure: Results from 7 studies and a revised scale. Diabetes Care, 23, 943–950. https://doi.org/10.2337/diacare.23.7.943

51.

Weller

S. C.

Baer

Nash

, & Perez

(2017). Discovering successful strategies for diabetic self-management: A qualitative comparative study. BMJ Open Diabetes Research and Care, 5, e000349. https://doi.org/10.1136/bmjdrc-2016-000349

52.

Wiedemann

A. U.

Gardner

Knoll

, & Burkert

(2014). Intrinsic rewards, fruit and vegetable consumption, and habit strength: A three-wave study testing the associative-cybernetic model. Applied Psychology: Health and Well-Being, 6, 119–134. https://doi.org/10.1111/aphw.12020

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.87 MB