Physical activity (PA) is integral to the management of type 2 diabetes mellitus (T2D) and is promoted as first line therapy with diet modification. The benefits of PA for T2D include, but are not limited to, improved glycemic control (via reduction in glycated hemoglobin [HbA1c]), cardiorespiratory fitness, glucose uptake into active muscles, and acute insulin action, as well as reduced visceral adipose tissue and plasma triglycerides,1,2 These benefits also reduce the risk of developing or advancing significant cardiovascular complications3,4 and medication dependence. To achieve these benefits, it is recommended for people with T2D to complete 150 to 210 minutes of moderate intensity PA per week, involving both aerobic and resistance exercise.5–7
Moderate intensity continuous training is the most commonly recommended exercise for people with T2D as it has been shown to provide clinically meaningful physiological adaptations, is associated with a low risk of adverse events including injury, and is tolerable for the majority of the population.8 Higher intensity exercise such as high-intensity interval training (HIIT) has gained popularity in recent years as an alternative to moderate intensity continuous training due to the similar, or occasionally superior, physiological adaptations.9 These adaptations are seen even with a low total HIIT exercise volume.10 This is important, given lack of time is the most commonly reported barrier to PA participation in people with T2D.11,12
Supervised exercise training is a common approach to promote the uptake of and adherence to PA. However, an important issue is adherence to PA when supervision is removed or reduced. The maintenance of sufficient, self-directed PA after supervised training has been investigated in only a few T2D studies, and these suggest declines in sustained participation.13,14 A better understanding of the factors associated with PA maintenance is required to inform the tailoring of exercise programs and increase long-term participation rates. Thus, in people with T2D, the aims of the current study were 2-fold: (1) identify sociodemographic, medical, and health determinants of PA 10 months following a supervised exercise-training program and (2) determine individual-level barriers and enablers of self-directed combined aerobic and resistance exercise.
Materials and Methods
This is a secondary analysis of data collected for the “Exercise for Type 2 Diabetes (E4D)” Trial (ACTRN12615000475549), which was a randomized controlled trial investigating the short- and long-term efficacy, safety, and feasibility of low-volume combined aerobic and resistance HIIT (C-HIIT) in people with T2D. The main clinical outcomes of the E4D Trial are reported elsewhere.15–18 The E4D Trial was prospectively approved by The University of Queensland Human Research Ethics Committee (ethics approval number 2015000164) and adhered to the Declaration of Helsinki principles. All participants provided written informed consent.
E4D Trial Design
More information about the E4D Trial is described elsewhere.15 The 12-month trial involved 69 low active individuals with T2D who were randomly allocated to (1) low-volume C-HIIT (n = 23), (2) combined aerobic and resistance moderate intensity continuous training (C-MICT, n = 23), or (3) waitlist control (CON, n = 23). Phase 1 involved 8 weeks of exercise training at the university, supervised by an Accredited Exercise Physiologist (AEP). Phase 2 involved 10 months of self-directed exercise. Those participants who were initially randomized to the waitlist condition (CON) had 8 weeks of usual care and were then rerandomized to one of the exercise-training conditions for 12 months. Assessment occurred at baseline, after 8 weeks, and again after 12 months.
The C-HIIT group (n = 33) trained for 26 minutes 3 times per week, for a total duration of 78 minutes per week. Each session included a 3-minute aerobic warm-up (50%–60% of heart rate peak [HRpeak]), 4 minutes of high-intensity aerobic exercise (85%–95% HRpeak), 8 × 1-minute intervals of high-intensity resistance exercise (rating of perceived exertion: ≥17; very hard) with 1-minute rest between intervals, and a 3-minute aerobic cool-down (50%–60% HRpeak).
The C-MICT group (n = 30) trained for 52.5 minutes 4 times per week—2 sessions incorporating both aerobic and resistance training and 2 sessions involving aerobic training only, for a total duration of 210 minutes per week. This total duration is consistent with current recommendations.5 For the 2 combination sessions, participants completed 22.5 minutes of moderate intensity aerobic exercise (55%–69% HRpeak), followed by 30 minutes of moderate intensity resistance exercise (rating of perceived exertion: 11–13; fairly light to somewhat hard). For the 2 aerobic-only sessions, participants completed 52.5 minutes of moderate intensity aerobic exercise (55%–69% HRpeak).
During the self-directed phase, participants were advised to continue the exercise dose (duration and intensity) prescribed during the supervised phase and were given written information about exercises they could complete to replicate their allocated supervised training program. Participants were educated on how to progress their exercise (ie, increase aerobic exercise speed and increase resistance exercise weight lifted) to be able to achieve their allocated exercise intensity for the duration of the study. In addition, to support their progress, participants were offered optional monthly supervised “booster” exercise sessions at the university, at the dose identical to their allocated group; combination sessions were delivered for the C-MICT group.
Participants
People were eligible for the E4D Trial if they were aged 18–80 years and had a diagnosis of T2D, including an HbA1c of ≥6.0%. The exclusion criteria were per the American College of Sports Medicine’s absolute contraindications to exercise,19 including unstable angina, recent myocardial infarction, coronary artery disease, and uncontrolled, symptomatic heart failure. People were ineligible if they self-reported more than 150 minutes of moderate PA, or 75 minutes of vigorous PA, or any equivalent combination, per week.
Measures
Outcome
Self-reported leisure-time PA time was the outcome and was assessed by an interviewer-guided, written questionnaire at baseline and 12 months. The long form of the International Physical Activity Questionnaire (IPAQ) is a valid and reliable measure of self-reported PA.20,21 The IPAQ consists of 27 items and assesses PA time in the previous 7 days across 4 domains (leisure-time PA, domestic and gardening activities, work-related PA, and transport-related PA). For the present study, variables derived from the IPAQ included leisure time spent in walking and moderate and vigorous intensity activities; total leisure-time PA was used as the outcome variable and calculated as the sum of time spent in walking and moderate and vigorous activity, with vigorous activity weighted by 2, given the higher intensity.
Predictors
An in-person self-administered, written questionnaire completed at baseline was used to gather information on participant age, sex, medication use, cohabitation (living alone vs living together), education (up to secondary school, diploma/certificate, or university degree), and employment status (any paid employment, no paid employment, studying, or retired).
Participants also completed the Physical Activity Enjoyment Scale,22 which has 18-items with responses on a 7-point bipolar Likert scale, with anchor labels that change for each question (eg, I enjoy it vs I hate it, I find it energizing vs I find it tiring, and It’s no fun at all vs It’s lots of fun). An overall enjoyment score was determined by summing all items (possible score range: 18–126), with higher scores indicating greater enjoyment. The Physical Activity Enjoyment Scale has been shown to be internally reliable for use with otherwise healthy adults and adolescents, and Cronbach alpha has been reported as .96.23 The Physical Activity Enjoyment Scale questionnaire was administered at baseline and at the end of the supervised exercise training (8 weeks), with the change score used as the predictor variable for total PA at 12 months.
Clinical health outcomes were assessed after an overnight fast (≥12 h), as well as after 48 hours with no exercise and 24 hours with no caffeine, alcohol, tobacco, and other stimulants. A venous blood sample was used to analyze HbA1c and fasting blood glucose using manufacturer supplied assay kits in an automated analyzer (Randox RX daytona+). Body fat percentage was determined using dual-energy X-ray absorptiometry (Discovery, Hologic Inc). A graded cardiopulmonary exercise test was used to determine peak oxygen uptake (
The number of adverse events was monitored throughout the 8-week supervised exercise-training period (phase 1) via in-person check-ins.
Attendance at the optional monthly “booster” exercise sessions was determined from AEP-reported exercise records. The total number of sessions attended, out of a possible 9, was recorded.
Participant Perceived Barriers and Enablers to Self-Directed Exercise Participation
All E4D Trial participants who attended the 12-month follow-up assessment after the date when ethical clearance was awarded for the current study were invited to participate in a semistructured interview (n = 21 out of 43). The interview assessed their perceived barriers and enablers to self-directed exercise participation during phase 2 of the trial. The interview questions are included in Box 1: the first part asked a general question about factors that impacted self-directed exercise and the second part asked about exercise duration, intensity, injury, resources, and supervision. These interviews were conducted face-to-face at the university, individually, with a member of the research team who was known to the participants as the AEP who supervised the exercise training (E.R.C., female). Interviews were scheduled during the 12-month follow-up assessment. The interviews were audio recorded, with notes taken by the interviewer.
Box 1 Interview questions assessing barriers and enablers to self-directed exercise participation
- 1.What do you think are the 3 main factors to affect your ability to do the exercise-training program at home? Please consider factors that made it easier and factors that made it harder.
- 2.Let me ask about a few specific factors:
- a.In terms of the time the sessions take, how did that influence your ability to do the exercise-training program at home?
- b.In terms of the intensity of the session, how did that influence you? Were you worried about injury?
- c.In terms of resources needed to do your exercise, how did that influence your ability to exercise?
- d.You had supervision for the first 8 weeks of exercise training and then you did not, how did that influence your ability to exercise?
Statistical Analysis
Data from 6 of the participants in the E4D Trial who were initially randomized to waitlist control, but not rerandomized to one of the exercise groups (n = 5 lost to follow-up; n = 1 declined rerandomization), were excluded from these analyses. A further 6 participants did not receive their allocated intervention during the supervised phase (phase 1) due to dropout prior to program start or unrelated medical reasons; their data were also excluded from these analyses. Therefore, pooled data from 57 participants from the E4D Trial (C-HIIT: n = 31 and C-MICT: n = 26) were included in the current analyses (see Figure S1 in Supplementary Materials [available online]).
Statistical analyses were conducted using SPSS version 25 for Windows. Shapiro–Wilk tests, along with visual inspection of the distribution of models’ residuals, were used to assess normality. No variables were required to be transformed prior to analysis. Data are presented as mean (SD) for normally distributed variables and n (%) for categorical variables.
The following predictor variables were dichotomized prior to analyses: highest level of education (categorized as tertiary educated or not), polypharmacy (concomitant use of 5 or more prescribed medications or not), retirement status (retired or not), and cohabitation (living together or living alone).
Pearson rank and point biserial correlations were used to examine associations between each of the predictor variables and total PA (outcome variable) at 12 months. Correlation coefficients were interpreted as: <.1, weak or small association; .1 to .5, moderate association; and >.50, strong or large association.25 Variables that had an association (P < .20) were included in the multiple regression analysis to examine the effects of each predictor after adjustment for the other variables in the model. Independent t or χ2 tests were used to compare participants who completed the 12-month assessment and those who were lost to follow-up, as well as those who were invited to attend the semistructured interview, and those who were not.
The data analysis of the individual interviews followed the process for reflexive qualitative thematic analysis; this facilitates recursive engagement with the data, enabling deep understanding and interpretation.26 The first author (E.R.C.) read all the transcripts to gain an overview of the content and generated initial codes to represent the various barriers (factors increasing exercise difficulty) and enablers (factors increasing exercise ease) described by participants. The codes were then examined and sorted into preliminary themes. Draft themes and example codes were discussed by E.R.C. and N.W.B. to verify that the themes reflected the content and perceived meanings. These authors then discussed and refined the conceptual structure and presentation of analysis. A title and narrative for each theme was prepared by E.R.C. and reviewed by N.W.B., with resolution via consensus-based discussion. The percentage of participants reporting each common barrier and enabler was calculated. Participant responses were compared for the 2 supervised training conditions (ie, C-HIIT and C-MICT) and for those who, at 12 months, met the exercise guidelines for people with T2D (ie, ≥210 min·wk−1 total PA) and those who did not (determined from IPAQ).
Results
Table 1 shows the characteristics of the included participants at baseline and 8 weeks (ie, postsupervised training). Fourteen participants (C-HIIT: n = 6, C-MICT: n = 8) were lost to follow-up during the self-directed exercise phase (phase 2; see Figure S1 in Supplementary Materials [available online]). Factors associated with dropout were low attendance at the optional “booster” exercise sessions (2 [2] vs 7 [3], P < .001), younger age (55.6 [6.1] y vs 61.9 [8.7] y, P = .016), working/studying full-time (n = 24 vs n = 12, P = .044), and a lower number of adverse events during the supervised exercise phase (1.1 [1.5] vs 2.3 [1.6], P = .021; see Table S1 in Supplementary Materials [available online]).
Participant Characteristics at Baseline and 8 weeks (n = 57)
| Variable | Baseline | 8 weeks | Change |
|---|---|---|---|
| Demographics | |||
| Female, n (%) | 23 (40.4) | ||
| Age, y | 60.4 (8.6) | ||
| Duration of diabetes, y | 10.8 (7.8) | ||
| Tertiary educated, n (%) | 36 (63.2) | ||
| Retired, n (%) | 18 (31.6) | ||
| Living together, n (%) | 43 (75.4) | ||
| Polypharmacy,a n (%) | 27 (47.4) | ||
| Physical activity participation, min·wk−1 | |||
| Walking | 43.5 (58.9) | ||
| Moderate physical activity | 14.0 (52.5) | ||
| Vigorous physical activity | 5.1 (17.0) | ||
| Weighted total physical activityb | 67.7 (145.3) | ||
| Clinical outcomes | |||
| HbA1c, % | 8.8 (1.9) | 8.4 (1.9) | −0.4 (1.3) |
| HbA1c, mmol·mol | 72.7 (21.1) | 68.0 (21.0) | −4.7 (13.6) |
| Fasting blood glucose, mmol·L | 9.1 (2.9) | 8.6 (2.5) | −0.4 (2.6) |
| Body fat, % | 39.5 (7.4) | 39.1 (7.5) | −0.5 (1.3) |
| | 24.3 (5.9) | 24.9 (5.9) | 0.4 (3.9) |
| Exercise enjoyment | |||
| Physical Activity Enjoyment Scale scorec | 93 (18) | 96 (23) | 8 (20) |
Abbreviations: HbA1c, glycated hemoglobin;
aPolypharmacy defined as concomitant use of 5 or more prescribed medications. bSum of time spent in walking and moderate and vigorous activity, with vigorous activity weighted by 2 given the higher intensity. cPossible scores range from 18–126, with higher scores indicating more enjoyment.
At 12 months, the average weighted total PA across all participants (pooled data from both training conditions, n = 43) was 209.4 (174.3) minutes per week (Table 2). Participants allocated to C-HIIT reported an average of 260.9 (193.0) minutes per week of weighted total PA, while participants allocated to C-MICT reported an average of 134.2 (92.2) minutes per week.
Physical Activity Participation at 12 Months
| Variable, min·wk−1 | All | C-HIIT | C-MICT |
|---|---|---|---|
| N = 43 | n = 24 | n = 19 | |
| Walking | 65.2 (69.4) | 50.3 (54.5) | 86.9 (84.4) |
| Moderate physical activity | 74.5 (132.4) | 93.2 (165.8) | 47.3 (51.9) |
| Vigorous physical activity | 30.2 (44.2) | 58.7 (55.7) | 0 (0) |
| Weighted total physical activitya | 209.4 (174.3) | 260.8 (198.3) | 134.2 (96.0) |
Abbreviations: C-HIIT, combined aerobic and resistance high-intensity interval training; C-MICT, combined moderate intensity continuous training. Note: Data are presented as mean (SD).
aSum of leisure time spent in walking and moderate and vigorous activity, with vigorous activity weighted by 2 given the higher intensity.
Table 3 shows the univariable and multivariable associations between the predictor variables and total PA (outcome variable) at 12 months. There were moderate negative univariable relationships between age (r = −.256, P = .158), allocation to C-HIIT (r = .360, P = .043), reduction in fasting blood glucose during the supervised training phase (r = −.359, P = .043), and polypharmacy (r = −.238, P = .191) with total PA at 12 months. There were moderate positive univariable relationships between the number of adverse events during the supervised training phase (r = .432, P = .014) and increase in exercise enjoyment during the supervised phase (r = .344, P = .127) with total PA at 12 months. There was no relationship between sex; retirement status; education level; cohabitation; number of optional “booster” exercise sessions attended; or change in HbA1c, body fat percentage, or relative
Regression Models for Associations Between Predictor Variables and Total Physical Activity (Minutes per Week) at 12 Months (n = 43)
| Univariable | Multivariable | ||||||
|---|---|---|---|---|---|---|---|
| r | P | R2 | P | β | P | ||
| Model | .453 | .019* | |||||
| Number of adverse eventsa | .432 | .014* | 0.363 | .059 | |||
| Allocation to C-HIIT | .360 | .043* | 0.477 | .027* | |||
| Change in fasting blood glucosea | –.359 | .043* | –0.245 | .252 | |||
| Change in exercise enjoymenta | .344 | .127 | 0.376 | .067 | |||
| Age | –.256 | .158 | –0.063 | .782 | |||
| Polypharmacy status (<5 medications) | –.238 | .191 | 0.098 | .631 | |||
| Retirement status (retired) | –.214 | .241 | |||||
| Sex (female) | –.181 | .320 | |||||
| Educational level (tertiary) | .181 | .323 | |||||
| Change in body fat percentagea | –.166 | .364 | |||||
| Cohabitation status (living together) | .161 | .377 | |||||
| Number of optional “booster” sessions attended during self-directed phase | .127 | .496 | |||||
| Change in relative VO2peaka | .049 | .789 | |||||
| Change in HbA1ca | –.030 | .872 | |||||
Abbreviations: C-HIIT, combined aerobic and resistance high-intensity interval training; HbA1c, glycated hemoglobin; VO2peak, peak oxygen consumption. Note: Boldface indicates variables that had a univariable association (P < .20) with total PA at 12 months and were included in the multivariable analysis.
aFrom baseline to after the 8-week supervised phase. Study allocation coded as 0, C-MICT, and 1, C-HIIT. Polypharmacy status coded as 0, use of ≥ 5 prescribed medications, and 1, use of < 5 prescribed medications. Retirement status coded as 0, not retired, and 1, retired. Sex coded as 0, males, and 1, females. Educational level coded as 0, not tertiary educated, and 1, tertiary educated. Cohabitation coded as 0, living alone, and 1, living together.
*Statistical significance (P ≤ .05).
A multiple regression was used to predict total PA at 12 months, with independent variables of number of adverse events during the supervised phase, allocation to C-HIIT, reduction in fasting blood glucose during the supervised phase, increase in exercise enjoyment, age, and polypharmacy (Table 3). The regression statistically significantly predicted total PA at 12 months (R2 = .453, P = .019), with only allocation to C-HIIT an independent predictor (β = 0.477, P = .027). The number of adverse events (β = 0.363, P = .059) and increase in exercise enjoyment (β = 0.376, P = .067) during supervised training were borderline independently significant.
All participants invited to interview postprogram accepted (n = 21; C-HIIT: n = 13, C-MICT: n = 8). There were no differences in the age, sex, duration of diabetes, education level, retirement status, or cohabitation status of those who were invited to complete the interviews and those who were not. The average interview duration (median [interquartile range]) was 13 minutes and 43 seconds (11:72–15:21). When asked about the main factors impacting self-directed exercise, the most commonly reported barriers to self-directed exercise were lack of access to specialized equipment (reported by 57%) and competing time demands (43%). Other barriers included health problems/injury (24%), low motivation (19%), and lack of positive reinforcement (eg, not seeing health changes; 10%). The most commonly reported enablers were flexibility to schedule exercise around other commitments (reported by 57%) and social support (48%). Other enablers included access to a gym (48%) and positive reinforcement (eg, self-observed improvements in cardiorespiratory fitness and strength; 15%). Barriers and enablers were similar for participants in C-HIIT and C-MICT, and for participants who met the exercise guidelines at 12 months (ie, ≥210 min·wk−1 total PA) and those who did not.
When asked about the specific aspects of self-directed exercise, lack of access to exercise resources such as exercise machines (reported by 71%), and lack of supervision (57%) were commonly reported as barriers. The short exercise session duration was identified by 25% of C-HIIT participants as enabling adherence. Exercise intensity was not seen as relevant by participants in either training condition.
Discussion
This secondary analysis of data from the E4D Trial15 identified being allocated to low-volume C-HIIT as the only significant independent determinant of total leisure-time PA 10 months following an 8-week supervised exercise program in people with T2D. At 12 months, participants allocated to C-HIIT reported 120 minutes more weighted total PA per week than those allocated to C-MICT (260.8 [198.3] min·wk−1 vs 134.2 [96.0] min·wk−1, respectively). The multivariable analyses indicated that the combination of being younger, taking less than 5 prescribed medications, and having more exercise-related adverse events, a greater reduction in blood glucose, and a greater increase in exercise enjoyment during the supervised exercise-training phase, was associated with more total PA at 12 months. The main barriers to self-directed exercise reported by participants were lack of access to specialized equipment and competing time demands. The main enablers were flexibility to schedule exercise around other commitments and social support. These factors should be considered when designing and delivering support for sustained exercise for people with T2D.
To our knowledge, this is the first study to identify C-HIIT (vs C-MICT) supervised exercise training as a determinant of self-directed total PA at 12 months in people with T2D. This group was prescribed one-third the duration of exercise than the C-MICT group (78 and 210 min·wk−1, respectively). Only those prescribed C-HIIT met the exercise guidelines of 210 minutes per week5 at 12 months. However, those assigned to C-HIIT reported doing moderate intensity physical activities and walking, as well as HIIT, during the 10 months of self-directed exercise phase. This is in line with findings from the Generation 100,27 FITR Heart,28 and Small Steps for Big Changes29 studies where participants randomized to HIIT chose to engage in a combination of activities with different intensities, when self-directing exercise.
The reasons for C-HIIT (vs C-MICT) predicting total PA at 12 months follow-up are, however, not clear. It may be that the C-HIIT participants developed higher self-efficacy for PA during the supervised training phase than the C-MICT participants, and this enabled them to better maintain self-directed exercise. The shorter duration per week for C-HIIT than C-MICT (approximately one-third) may have been seen as more achievable therefore promoting ongoing PA. More work is needed to examine intraindividual changes associated with C-HIIT, and other formats of exercise training, and how these affects self-directed exercise adherence.
While not independent significant predictors in the regression model, we identified a combination of sociodemographic, medical, and health factors associated with PA at 12 months. The finding that more exercise-related adverse events during the supervised exercise-training phase was positively associated with total PA was contrary to expectations. Previous studies have reported adverse events result in a cessation of exercise (temporarily or permanently) in people with T2D30,31; but our study has indicated the opposite at follow-up. That is, experiencing adverse events during the exercise program did not dissuade people with T2D from participating in self-directed PA during the following 10 months. This may have been a result of the way adverse events were managed during the supervised training phase; the AEP was able to provide education and adjust the exercise prescription, which may have consequently reduced the participants’ fear of such events in the future and increased their confidence to continue with exercise. Also, participants who had experienced adverse events were encouraged to interpret them as temporary setbacks to be problem solved rather than a reason to discontinue exercise, and this cognitive process may have been maintained by participants when self-directing exercise.
Exercise enjoyment is an important antecedent to exercise behavior, with higher levels of enjoyment resulting in a more positive attitudes toward exercise and consequently a more positive intention to exercise.32 This was reflected in the current study; a positive change in exercise enjoyment during the supervised training phase was part of the combination of factors positively associated with total PA at 12 months. Previous research has demonstrated that enjoyment of exercise is related to the specialized supervision and support, the context of exercise, or participation in C-HIIT.33
A greater reduction in fasting blood glucose during the supervised exercise training was also one of the combined factors positively associated with total PA at 12 months. Positive health changes have been reported previously as a motivator for exercise in people with T2D.34 These results are also consistent with the participant interview data in the current study, which indicated that positive reinforcement such as health changes were enablers of self-directed PA participation. However, this was not a significant independent predictor in the regression model, suggesting that improvements in glycemic control are not the most important factor impacting exercise maintenance and perhaps should not be a primary or sole focus when evaluating exercise participant achievements.
The concomitant use of less than 5 prescribed medications was associated with more total PA at 12 months. This finding is in line with previous work showing an inverse association between number of prescribed medications and exercise.35 This may be due to the link between polypharmacy and the presence of multimorbidity, in addition to potential medicine interactions and side effects, which may interfere with exercise participation. In line with this, participants’ interview data indicated health problems as a barrier to self-directed exercise.
Our analysis also included a comparison of participants who completed the 12-month assessment and those who were lost to follow-up during the self-directed exercise phase. Low attendance at the optional “booster” exercise sessions was positively associated with dropout. Complete withdrawal of professional exercise supervision and subsequent reduction in exercise participation has been reported in previous lifestyle intervention trials in people with T2D.34,36 Leehey et al37 utilized once monthly supervised sessions, in addition to weekly phone calls, to improve likelihood of maintenance of a lifestyle intervention in people with T2D, and comorbid obesity and chronic kidney disease. Therefore, an element of specialized follow-up support after supervised exercise training could be incorporated into program plans to improve maintenance rates. Participants in the current study who were working/studying full-time were also more likely to dropout. The most commonly reported barrier to PA participation in people with T2D is lack of time,11,12 and competing time demands were also identified in the interview component of the current study as a key barrier to self-directed exercise. Therefore, it is not surprising that full-time work/study commitments were associated with study dropout. A potential advantage of C-HIIT is that the low-volume protocol requires one-third of time commitment required of the current exercise guidelines.5 It may be useful to overcome this time-related barrier. The current study demonstrated a higher dropout rate (loss to follow-up at 12 mo) in the C-MICT (31%) than C-HIIT (19%) group. Additional support, such as problem-solving potential situations when exercise might not be maintained in specific situations (eg, busy holiday or work periods) and developing related coping strategies, may be useful for reducing exercise dropout in people with T2D, irrespective of the volume of exercise prescribed.
Interview data in the current study indicated that lack of access to specialized equipment and lack of supervision were key participant perceived barriers to self-directed exercise, while flexibility to schedule exercise and social support were key enablers. These factors have been reported previously by people with T2D as influences of exercise adherence.34,36 Lack of access to specialized equipment was an issue specifically for the resistance training component of the program of the current study. We attempted to demonstrate and encourage participants to use body weight and basic free weights (rather than machines) for resistance training in the final stages of the 8-week supervised training phase as part of the transition to self-directed exercise; at follow-up, participants described a perception that resistance training meant that they needed to be in a gym using specific equipment. Also, participants reported difficulties with managing progression of the intensity of self-directed exercise. To overcome this, exercise practitioners could focus on initial prescription of exercises based on the actual availability of equipment and facilities to the participant, irrespective of potential availability during supervised training sessions, as well as provide education and resources about how to progress exercises over time.
The other reported participant perceived barrier to self-directed exercise was competing time demands, which has been identified extensively in the literature.11,12,14 In support of this, a commonly reported enabler was flexibility to schedule exercise around other commitments, rather than fixed, center-based appointments with an exercise practitioner that required effort and forethought to reschedule. Exercise practitioners could educate clients on flexible scheduling as part of planning for exercise self-regulation. Habit is a recognized influence on the maintenance of exercise behavior,38 so a delicate balance between flexibility and regularity may be needed to promote exercise maintenance.
The other key enabler to self-directed exercise in the interview assessment of the current study was social support from a family member or friend. Support is a consistently identified correlate of self-care behaviors such as exercise in people with T2D.34,39,40 The average age of participants in the current study was 60 years, and our previous research has demonstrated that otherwise healthy adults in this age group are at risk of low levels of support for PA and that all types of exercise support and many sources of exercise companionship decrease over time.41,42 People with a chronic condition such as T2D may require more exercise support than otherwise healthy people. Practitioners could work with exercise clients to plan for social support when prescribing exercise programs, including identifying different types (eg, emotional, network, informational, and material) and sources (eg, family, friends, and exercise groups) of support.
This study has several limitations. The study design precludes conclusions about causality. Moreover, the interviewer was known to the participants as the person who supervised the exercise training during phase 1, which may have influenced their responses. Participants who did not complete the 12-month assessment did not take part in the semistructured interviews, so it is unknown whether the perceived barriers and enablers to self-directed exercise differed for those participants.
Conclusions
This is the first study to identify supervised low-volume C-HIIT as an independent determinant of PA 10 months following an 8-week supervised exercise program in people with T2D. The combination of younger age, taking less than 5 prescribed medications, and a greater number of exercise-related adverse events, greater reduction in blood glucose, and greater increase in exercise enjoyment during the 8-week supervised training phase, also reported more PA at follow-up. Access to social support, specialized equipment, flexible scheduling, and management of competing time demands may be important for sustained self-directed activity after supervised training. These factors should be considered when designing and delivering long-term exercise support for people with T2D.
Acknowledgments
The authors would like to acknowledge the trial participants for donating their time, Mr Gary Wilson for his technical support, and Diabetes Queensland for their assistance with participant recruitment. Data Availability: Data are available upon reasonable request. Ethics Approval: The E4D Trial was prospectively approved by The University of Queensland Human Research Ethics Committee (ethics approval number 2015000164) and adhered to the Declaration of Helsinki principles. Patient Consent: All participants provided written informed consent before participating in the study. Trial Registration: Australian New Zealand Clinical Trials Registry Identifier: ACTRN12615000475549. Author Contributions: Concept and design: Cox, Gajanand, Keating, Brown, Coombes, Burton. Data acquisition: Cox, Gajanand, Keating, Brown, Coombes, Burton. Data analysis: Cox, Gajanand, Keating, Brown, Coombes, Burton. Data interpretation: Cox, Gajanand, Keating, Brown, Coombes, Burton. Drafting of manuscript: Cox, Burton. Critical feedback: Gajanand, Keating, Brown, Coombes. Final version of the manuscript: Cox, Gajanand, Keating, Brown, Coombes, Burton. Cox is the guarantor of the work and accepts full responsibility for the work and/or conduct of the study, had access to the data, and controlled the decision to publish.
References
- 1.↑
Colberg SR, Sigal RJ, Fernhall B, et al. Exercise and type 2 diabetes: The American College of Sports Medicine and the American Diabetes Association: joint position statement. Diabetes Care. 2010;33(12):e147–e167. doi:
- 2.↑
Thomas DE, Elliott EJ, Naughton GA. Exercise for type 2 diabetes mellitus. Cochrane Database Syst Rev. 2006;10(3):CD002968. doi:
- 3.↑
Balducci S, Iacobellis G, Parisi L, et al. Exercise training can modify the natural history of diabetic peripheral neuropathy. J Diabetes Complications. 2006;20(4):216–223. doi:
- 4.↑
Zoppini G, Targher G, Zamboni C, et al. Effects of moderate-intensity exercise training on plasma biomarkers of inflammation and endothelial dysfunction in older patients with type 2 diabetes. Nutr Metab Cardiovasc Dis. 2006;16(8):543–549. doi:
- 5.↑
Hordern MD, Dunstan DW, Prins JB, Baker MK, Singh MA, Coombes JS. Exercise prescription for patients with type 2 diabetes and pre-diabetes: a position statement from exercise and sport science Australia. J Sci Med Sport. 2012;15(1):25–31. doi:
- 6.
Kanaley JA, Colberg SR, Corcoran MH, et al. Exercise/physical activity in individuals with type 2 diabetes: a consensus statement from the American College of Sports Medicine. Med Sci Sports Exerc. 2022;54(2):353–368. doi:
- 7.↑
Sigal RJ, Armstrong MJ, Bacon SL, et al. Physical activity and diabetes. Can J Diabetes. 2018;42(suppl 1):S54–S63. doi:
- 8.↑
Decker ES, Ekkekakis P. More efficient, perhaps, but at what price? Pleasure and enjoyment responses to high-intensity interval exercise in low-active women with obesity. Psychol Sport Exerc. 2017;28:5. doi:
- 9.↑
Weston KS, Wisloff U, Coombes JS. High-intensity interval training in patients with lifestyle-induced cardiometabolic disease: a systematic review and meta-analysis. Br J Sports Med. 2014;48(16):1227–1234. doi:
- 10.↑
Ramos JS, Dalleck LC, Borrani F, et al. Low-volume high-intensity interval training is sufficient to ameliorate the severity of metabolic syndrome. Metab Syndr Relat Disord. 2017;15(7):319–328. doi:
- 11.↑
Egan AM, Mahmood WAW, Fenton R, et al. Barriers to exercise in obese patients with type 2 diabetes. QJM. 2013;106(7):635–638. doi:
- 12.↑
Thomas N, Alder E, Leese GP. Barriers to physical activity in patients with diabetes. Postgrad Med J. 2004;80(943):553. doi:
- 13.↑
Dunstan DW, Daly RM, Owen N, et al. Home-based resistance training is not sufficient to maintain improved glycemic control following supervised training in older individuals with type 2 diabetes. Diabetes Care. 2004;28(1):3. doi:
- 14.↑
Sigal RJ, Kenny GP, Boulé NG, et al. Effects of aerobic training, resistance training, or both on glycemic control in type 2 diabetes: a randomized trial. Ann Intern Med. 2007;147(6):357–369. doi:
- 15.↑
Gajanand T, Cox ER, Keating SE, et al. Low-volume combined aerobic and resistance high-intensity interval training in type 2 diabetes: a randomised controlled trial. BMJ Open Sport Exerc Med. 2024;10(4):e002046. doi:
- 16.
Cox ER, Gajanand T, Keating SE, et al. Effect of low-volume combined aerobic and resistance high-intensity interval training on vascular health in people with type 2 diabetes: a randomised controlled trial. Eur J Appl Physiol. 2024;10:5473. doi:
- 17.
Cox ER, Gajanand T, Burton NW, Coombes JS, Coombes BK. Effect of different exercise training intensities on musculoskeletal and neuropathic pain in inactive individuals with type 2 diabetes—preliminary randomised controlled trial. Diabetes Res Clin Prac. 2020;164:108168. doi:
- 18.↑
Torquati L, Gajanand T, Cox ER, et al. Effects of exercise intensity on gut microbiome composition and function in people with type 2 diabetes. Eur J Sport Sci. 2022;10:5436. doi:
- 19.↑
Pescatello LS, Arena R, Riebe D, Thompson PD. ACSM’s Guidelines for Exercise Testing and Prescription. 9th ed. Wolters Kluwer/Lippincott Williams & Wilkins; 2014:456.
- 20.↑
Craig CL, Marshall AL, Sjostrom M, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–1395. doi:
- 21.↑
Wanner M, Probst-Hensch N, Kriemler S, Meier F, Autenrieth C, Martin BW. Validation of the long international physical activity questionnaire: influence of age and language region. Prev Med Rep. 2016;3:250–256. doi:
- 22.↑
Kendzierski D, DeCarlo KJ. Physical activity enjoyment scale: two validation studies. J Sport Exerc Psychol. 1991;13(1):50–64.
- 23.↑
Motl RW, Dishman RK, Saunders R, Dowda M, Felton G, Pate RR. Measuring enjoyment of physical activity in adolescent girls. Am J Prev Med. 2001;21(2):110–117. doi:
- 24.↑
Collado-Mateo D, Lavín-Pérez AM, Peñacoba C, et al. Key factors associated with adherence to physical exercise in patients with chronic diseases and older adults: an umbrella review. Int J Environ Res Public Health. 2021;18(4):2023. doi:
- 25.↑
Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. L. Erlbaum Associates; 1988.
- 26.↑
Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3(2):77–101. doi:
- 27.↑
Stensvold D, Viken H, Steinshamn SL, et al. Effect of exercise training for five years on all cause mortality in older adults—the Generation 100 study: randomised controlled trial. BMJ. 2020;371:m3485. doi:
- 28.↑
Taylor JL, Holland DJ, Keating SE, et al. Short-term and long-term feasibility, safety, and efficacy of high-intensity interval training in cardiac rehabilitation: the FITR heart study randomized clinical trial. JAMA Cardiol. 2020;5(12):1382–1389. doi:
- 29.↑
Jung ME, Locke SR, Bourne JE, et al. Cardiorespiratory fitness and accelerometer-determined physical activity following one year of free-living high-intensity interval training and moderate-intensity continuous training: a randomized trial. Int J Behav Nutr Phys Act. 2020;17(1):933. doi:
- 30.↑
Johansen MY, MacDonald CS, Hansen KB, et al. Effect of an intensive lifestyle intervention on glycemic control in patients with type 2 diabetes: a randomized clinical Trial. J Am Med Assoc. 2017;318(7):637–646. doi:
- 31.↑
Cox ER, Coombes JS, Keating SE, Burton NW, Coombes BK. Not a painless condition: rheumatological and musculoskeletal symptoms in type 2 diabetes, and the implications for exercise participation. Curr Diabetes Rev. 2019;15:3504. doi:
- 32.↑
Kwan BM, Bryan AD. Affective response to exercise as a component of exercise motivation: attitudes, norms, self-efficacy, and temporal stability of intentions. Psychol Sport Exerc. 2010;11(1):71–79. doi:
- 33.↑
Jung ME, Bourne JE, Little JP. Where does HIT fit? An examination of the affective response to high-intensity intervals in comparison to continuous moderate- and continuous vigorous-intensity exercise in the exercise intensity-affect continuum. PLoS One. 2014;9(12):e114541. doi:
- 34.↑
Schmidt SK, Hemmestad L, MacDonald CS, Langberg H, Valentiner LS. Motivation and barriers to maintaining lifestyle changes in patients with type 2 diabetes after an intensive lifestyle intervention (The U-TURN Trial): a longitudinal qualitative study. Int J Environ Res Public Health. 2020;17(20):7454. doi:
- 35.↑
Bielemann RM, Silveira MPT, Lutz BH, et al. Objectively measured physical activity and polypharmacy among Brazilian community-dwelling older adults. J Phys Act Health. 2020;17(7):729–735. doi:
- 36.↑
Wycherley TP, Mohr P, Noakes M, Clifton PM, Brinkworth GD. Self-reported facilitators of, and impediments to maintenance of healthy lifestyle behaviours following a supervised research-based lifestyle intervention programme in patients with type 2 diabetes. Diabet Med. 2012;29(5):632–639. doi:
- 37.↑
Leehey DJ, Collins E, Kramer HJ, et al. Structured exercise in obese diabetic patients with chronic kidney disease: a randomized controlled trial. Am J Nephrol. 2016;44(1):54–62. doi:
- 38.↑
Kwasnicka D, Dombrowski SU, White M, Sniehotta F. Theoretical explanations for maintenance of behaviour change: a systematic review of behaviour theories. Health Psychol Rev. 2016;10(3):277–296. doi:
- 39.↑
Arda Sürücü H, Büyükkaya Besen D, Erbil EY. Empowerment and social support as predictors of self-care behaviors and glycemic control in individuals with type 2 diabetes. Clin Nurs Res. 2018;27(4):395–413. doi:
- 40.↑
Qiu SH, Sun ZL, Cai X, Liu L, Yang B. Improving patients’ adherence to physical activity in diabetes mellitus: a review. Diabetes Metab J. 2012;36(1):1. doi:
- 41.↑
Smith GSE, Moyle W, Burton NW. Frequency of physical activity done with a companion: changes over seven years in adults aged 60+ living in an Australian capital city. J Aging Health. 2023;35(9):736–748. doi:
- 42.↑
Smith GSE, Ware RS, Moyle W, Burton NW. Social support for physical activity over 9 years in adults aged 60-65 years at baseline. J Aging Phys Act. 2023;31(5):776–785. doi:
