Are Regular Aerobic Exercisers With Type 1 Diabetes Following Current Physical Activity Self-Management Guidelines? Insights From an Online Survey

Abstract

This study surveyed 102 adults with type 1 diabetes who were a part of online communities for regular exercisers with type 1 diabetes. These individuals, who met or exceeded aerobic exercise guidelines, were evaluated via self-report for their adherence to recommended glucose management strategies before, during, and after physical activity. Although most reported using diabetes-related technologies, 86% of respondents reported having frequent activity-related hypoglycemia. Eighty percent reported using trial-and-error strategies, and only 32% identified their health care professionals as a key source of exercise self-management education. Few followed consensus-based guidelines such as reducing insulin doses before exercise and refraining from exercising within 24 hours after a severe hypoglycemic event.

Key Points

• Within this active cohort, there was a high prevalence of hypoglycemia, with 49% of participants indicating no hypoglycemia symptoms unless blood glucose was <60 mg/dL (<3.3 mmol/L). Eighty-six percent of respondents experienced exercise-induced hypoglycemia, with 18% having severe hypoglycemic events during exercise and 41% having overnight hypoglycemia.

• Before exercise, only 27% of pump users reported adjusting basal insulin in the 60–90 minutes before exercise, and 51% reduced their pre-exercise meal bolus. Forty-two percent engaged their pump’s exercise mode during exercise.

• Ten percent of participants experienced diabetic ketoacidosis after exercise, and only 27% restricted exercise in the 24 hours after having a severe hypoglycemic event, highlighting a potential lack of understanding of exercise guidelines.

• Only 32% of respondents identified their medical team as a key source of information, whereas up to 80% reported relying on trial and error, social media groups, Google searches, and YouTube videos, underscoring the importance of developing online education resources for this population.

Although consensus statements and review articles provide guidance on glucose self-management strategies for exercise for individuals with type 1 diabetes (1–4), significant barriers remain to the successful implementation of these guidelines for patients. Surveys examining knowledge and adherence to diabetes management strategies for exercise highlight numerous challenges such as lack of knowledge of or confidence in managing diabetes around exercise; fear of hypoglycemia, particularly during aerobic activities; or limiting exercise altogether because it can make glucose more difficult to manage (5–12). Although these findings highlight important areas for improvement, they have typically focused on groups of individuals with type 1 diabetes who are failing to meet the 2024 American Diabetes Association recommendation to achieve 150 minutes of moderate to vigorous physical activity per week (13). Although it is well documented that up to 80% of individuals with type 1 diabetes fail to meet this recommendation (14), there is limited research on those who do.

In one recent analysis (15) of 561 physically active adults with type 1 diabetes using real-time continuous glucose monitoring (CGM), an activity monitor, and a study smartphone application, adherence to exercise self-management guidelines was low (∼10%) when taking proactive steps was required (e.g., reducing meal bolus doses before exercise, setting higher temporary targets on an automated insulin delivery [AID] system, or consuming carbohydrates before activity if glucose was below the suggested target range) (16). However, in this study cohort, the type of physical activity (30 minutes of aerobic, resistance, or mixed body-based activities) was imposed by the researchers using instructional videos, and the analysis period was only ∼2 weeks. Moreover, adherence to activity self-management guidelines was inferred based on insulin pump and CGM data downloads rather than provided by the participants through questionnaire.

Prolonged sessions of predominantly aerobic exercise such as walking, running, or cycling are the main types of exercise people with type 1 diabetes report having difficulty with, likely because they are the most common activities done by adults and teens with type 1 diabetes (17–20) and because these forms of activity tend to last longer and appear to have higher rates of blood glucose disposal relative to very intensive or burst-based activities such as weight training or sprinting (1). Gaining more insight into the glucose management strategies used by recreationally active individuals with type 1 diabetes who are experienced with both exercise and diabetes could provide valuable information for developing future guidelines, educational resources, and safe exercise protocols for the broader type 1 diabetes population. Therefore, the aim of this observational survey study was to assess glucose management strategies before, during, and after exercise among recreationally active aerobic exercisers with type 1 diabetes who met or exceeded physical activity guidelines compared with current consensus recommendations.

Research Design and Methods

This 96-question online survey was created via the REDCap (Research Electronic Data Capture) Web application for managing online surveys and databases, and all survey questions were developed by the authors to engage highly active individuals with type 1 diabetes. The complete list of questions included both closed-ended multiple-choice and open-ended questions (Supplementary Appendix S1). Some questions were designed to directly assess adherence to established guidelines (3,4), such as “Do you refrain from exercise if you have had a severe hypoglycemic episode within 24 hours?” Other questions aimed to explore exercise practices discussed in the type 1 diabetes exercise community but not formally assessed in this population, such as “Do you ever use basal insulin with your insulin pump so that your insulin pump can be removed during exercise?” and “Have you worn more than one pump site during exercise as a backup?” The survey content was reviewed by the Institutional Review Board (IRB) of the University of Arkansas for Medical Sciences (IRB #274593). All medical history and survey answers were anonymous and based on recall and self-reported data.

Between September and November 2022, the survey was posted on three Facebook groups targeted to individuals with type 1 diabetes who exercise regularly: Type 1 Run Community, Diabetic Ultra Endurance Athletes, and Type 1 Diabetic Athletes. Although these Facebook groups have a combined membership of ∼20,000 individuals, it was not possible to determine how many individuals are actively engaged in these groups. The groups were selected to direct the survey toward individuals with type 1 diabetes who were performing frequent aerobic exercise in the form of walking, jogging, and/or running. Individuals who may have been primarily participating in high-intensity or anaerobic physical activity (e.g., CrossFit or weightlifting) were not targeted for the survey because the varying glucose responses expected with the different exercise types would have necessitated a heterogeneous management approach and thereby confounded interpretation of the data.

Participants were required to attest to meeting the specific inclusion criteria of having a diagnosis of type 1 diabetes and to participating in walking, jogging, and/or running for exercise to continue and submit the survey. Moreover, participants were permitted to skip individual questions that did not pertain to their current management strategies or if they did not feel comfortable sharing specific information. Surveys could not be submitted, and no answers could be analyzed, unless all 96 survey questions were answered and/or appropriately skipped. Thus, although 176 surveys were initiated, 102 were completed and could be analyzed.

Survey data were analyzed using Microsoft Excel 2016 software. Primary outcomes of interest included overarching safety principles for glucose management with exercise; glucose management strategies used before, during, and after aerobic exercise; and learning resources and barriers to aerobic exercise participation in this population.

The data captured and analyzed for this study are available from the corresponding author upon reasonable request.

Results

Selected survey results on respondents’ reported adherence to aerobic exercise guidelines and strategies in type 1 diabetes are shown in Supplementary Figure S1.

Demographics and Clinical Characteristics of Participants

The main characteristics of the survey cohort (n = 102) are reported in Table 1. On average, respondents (69% female) were 42 ± 12 years of age with a diabetes duration of 20 ± 14 years, and a self-reported A1C of 6.6 ± 1.0% (48.2 ± 10.5 mmol/mol), and about 67% of respondents had an average glucose time in range (TIR; 70–180 mg/dL or 3.9–10.0 mmol/L) in the past 14 days within recommended targets (>70%). Ninety-five percent of respondents reported being followed by an endocrinologist/specialist for their diabetes, and the presence of diabetic retinopathy, peripheral neuropathy, or hypoglycemia unawareness was reported by 5–19% of participants.

Table 1.

Demographics and Clinical Characteristics of Participants (n = 102)

Characteristic	Value
Age, years	42 ± 12
BMI, kg/m2	24.2 ± 4.0
Female sex	69 (69/100)
Race  Caucasian  African American  Asian	94 (95/101) 5 (5/101) 1 (1/101)
Country  United States  Australia  Canada  United Kingdom  Other	78 (78/100) 7 (7/100) 6 (6/100) 5 (5/100) 4 (4/100)
Distance/week, miles (km)	22.5 ± 14 (36.2 ± 22.5)
Minutes/week	301.6 ± 193.4
Exercise ≥4 days/week	78 (79/101)
Type 1 diabetes duration, years	19.8 ± 14
A1C, % (mmol/mol)	6.6 ± 1.0 (48.2 ± 10.5)
TIR in the past 14 days, %  ≤50  51–60  61–70  71–80  >80	3 (3/97) 18.6 (18/97) 11 (11/97) 26.8 (26/97) 40 (39/97)
CGM use	97 (99/101)
MDI insulin regimen	26 (27/102)
Insulin pump therapy	74 (75/102)
Open-loop pump with CGM*	24 (18/75)
AID system*	75 (56/75)
Not specified*	1 (1/75)
Presence of retinopathy	14 (14/101)
Presence of peripheral neuropathy	5 (5/100)
Hypoglycemia unawareness	19 (19/100)

Data are % (x/n), with x representing the number of responders with a given characteristic and n representing the number of responses received to the question, or mean ± SD.

*Percentage of individuals who indicated which pump therapy they were using; one respondent did not answer.

Of the 102 survey respondents, 97% reported CGM use, with a high percentage of participants (74%) using an insulin pump compared with 26% using a multiple daily injection (MDI) insulin regimen. Seventy-five percent of pump wearers reported using an AID system, 24% used an open-loop pump with CGM, and 1% (one user) did not specify. Not all pump users continued wearing their pump during exercise; 57% reported removing their pump for exercise at least some of the time, and 22% reported removing their pump for exercise >50% of the time.

Consistent with the expected active cohorts from the three surveyed Facebook groups, 78% of respondents reported exercising ≥4 days/week, with an average 302 ± 193 minutes/week of aerobic exercise and a total of 22.5 ± 14 miles (36.2 ± 22.5 km) of combined walking, jogging, and/or running per week.

Exercise Safety

Supplementary Figure S2 demonstrates key results regarding respondents’ exercise preparation and safety. Regarding general safety guidance, only about half of respondents (52%) carried or wore medical identification, and only 20% reported having a glucagon kit accessible during exercise.

Despite the high use of CGM systems that featured hypoglycemia alerts and alarms (97% of CGM users had low alerts on, and 60% had the blood glucose fall rate alert on) and low rate of diagnosed hypoglycemic unawareness (19%), symptoms of hypoglycemia did not begin until blood glucose fell to <70 mg/dL for ∼90% of respondents (Supplementary Figure S2). Eighty-six percent of respondents reported a history of stopping exercise to treat hypoglycemia (a CGM value <70 mg/dL [<3.9 mmol/L]), with 18% reporting at least one severe hypoglycemic event during exercise. Moreover, only 27% reported that they avoided exercise for 24 hours after a severe hypoglycemic event (<50 mg/dL [<2.8 mmol/L]) had occurred, which is advised according to guidelines (3). Among the subset of individuals with diagnosed hypoglycemia unawareness, 21% reported severe hypoglycemia during exercise (slightly higher than the 18% in the overall group), and 21% said they restricted exercise if they had had severe hypoglycemia in the past 24 hours (less than the 27% in the overall group).

In relation to hyperglycemia management, 73% reported that they would not check for ketones in blood or urine if their blood glucose was >250 mg/dL (>13.9 mmol/L) prior to exercise, which is recommended in several guidelines (3,4,21,22), and 10% reported developing diabetic ketoacidosis (DKA) after exercise. Additionally, although a large proportion of pump users (66%) reported an infusion set becoming dislodged or failing to operate during exercise, which could lead to unexpected hyperglycemia and DKA, only 20% of respondents stated that they have worn a backup infusion set during exercise.

Management Before Exercise

Because of the associated risk of hypoglycemia during exercise from a recent insulin bolus, sometimes referred to as having elevated (or positive) insulin on board (IOB), participants were asked if they understood the idea of reducing IOB before exercising. A majority of participants (93%) reported understanding this concept, and further questions queried how they made practical adjustments before exercise in terms of carbohydrate replacement, glucose targets, and insulin delivery (Supplementary Figure S3A).

Over half of respondents (60%) highlighted that they considered a pre-exercise target glucose of 125–180 mg/dL (6.9–10.0 mmol/L) to be optimal, which reflects current guidelines, with 11% targeting above and 20% targeting below these thresholds. To try to achieve a pre-exercise glucose target, carbohydrate consumption was the most frequently used pre-exercise strategy. Ninety-two percent of respondents reporting eating carbohydrates immediately before exercise onset, with 15 g being the amount commonly consumed (59%) (Supplementary Figure S3A).

When examining insulin strategies used in preparation for exercise, pre-exercise meal bolus reductions were made on occasion by 51% of individuals, but the percentage of bolus insulin reduction varied from 10 to 100%. This finding was similar to that in the subset of individuals diagnosed with hypoglycemia unawareness, of whom 53% reduced their pre-exercise meal bolus. For those using an MDI insulin regimen, 24% reported reducing basal insulin for exercise. For pump users, basal rate reductions were made by 65% of individuals in anticipation of exercise, with most reductions either between 21 and 50% or ≥50% (42 and 43% of respondents, respectively). However, the time of basal reduction initiation widely varied among participants (at exercise onset 28%, 30 minutes before exercise 37%, 60 minutes before exercise 27%, and ≥90 minutes before exercise 12%).

Management During Exercise

Regardless of insulin therapy modality, 55% of respondents felt that a target glucose of 125–180 mg/dL (6.9–10.0 mmol/L) during exercise was optimal, whereas 41% felt that the target should be lower, and 2% felt that the target should be >180 mg/dL (>10 mmol/L). The frequency of glucose monitoring via CGM ranged from every 5 minutes to every 60 minutes (Supplementary Figure S3B). To maintain glucose management during exercise, 83% of respondents routinely consumed carbohydrates to prevent hypoglycemia, with most (51%) consuming ∼20 g/hour, while the remainder consumed more than this amount. If hypoglycemia did occur with exercise, 62% of individuals reporting resuming exercise immediately after their glucose level was >70 mg/dL. Thirty-four percent of respondents reported using short bouts of sprinting or high-intensity intervals to prevent hypoglycemia during aerobic activity.

Although 76% of all insulin pump users wore an AID system comprising a connected CGM, infusion pump, and AID algorithm, only 57% of these users reported continuing automated (i.e., “closed-loop”) functionality during exercise, with others reverting to open-loop mode (i.e., traditional insulin pump functionality). Among all AID users, 42% reported using their pump’s exercise function, and 15% of users maintained the usual AID closed-loop functionality without adjusting targets for exercise.

In relation to hyperglycemia, as shown in Supplementary Figure S3B, if glucose was >270 mg/dL (15.0 mmol/L) during exercise, 78% of responders said they would administer a correction bolus (<25% of the usual correction 31%, 26–50% of the usual correction 31%, 51–75% of the usual correction 11%, and >75% of the usual correction 6%), which again would be considered appropriate on some, but not all, occasions, according to guidelines (3,4,21,22).

Management After Exercise

Results of management after exercise are summarized in Supplementary Figure S3C. Thirty-five percent of respondents reported that their glucose levels continue to drop after exercise, whereas 65% reported that their glucose rose (or “spiked”) sharply.

In reference to the immediate risk of post-exercise hypoglycemia, 81% of respondents reported consuming carbohydrates and protein within 1 hour of exercise, and 90% within 90 minutes, at least in part to help mitigate hypoglycemia risk. Seventy percent of respondents gave their usual bolus insulin for post-exercise meals, whereas 26% reduced the amount of post-exercise meal boluses, and 4% did not given any bolus when food was consumed after exercise. Of those who wore an insulin pump and also used a temporary setting designed to limit insulin delivery during exercise (e.g., reduced their basal rate or set a higher temporary target if on an AID system), 17% resumed their usual insulin delivery before completing exercise, 57% immediately after completing exercise, 20% within 1 hour after completing exercise, and 6% >1 hour after completing exercise.

Delayed-onset overnight hypoglycemia was reported by 41% of individuals. Multiple responses were accepted as strategies to reduce this risk, including eating a carbohydrate-containing snack at bedtime (61%), reducing basal insulin delivery (pump) or through a reduced basal insulin dose (MDI) overnight (46%), or waking up in the middle of the night to check blood glucose and have a snack, if necessary (34%).

To mitigate post-exercise hyperglycemia, 31% of respondents used a prolonged walk or cooldown after exercise to prevent or limit blood glucose spikes. If a glucose spike occurred after exercise, 87% indicated that they would give an insulin correction bolus (usual correction 61%, half of usual correction 26%).

Barriers to and Facilitators of Exercise

As shown in Supplementary Figure S4A, the most common source of learning about diabetes management with exercise was personal trial and error (80%), followed by social media groups (46%), medical teams (32%), online searches for advice (28%), and/or online videos (9%). The most frequently reported barriers to performing additional exercise (Supplementary Figure S4B) were lack of time (65.5%), increase in glycemic variability (35.7%), and fear of hypoglycemia (27.4%).

Discussion

Despite surveyed participants with type 1 diabetes exceeding current physical activity recommendations of 150 minutes/week of moderate to vigorous physical activity (2), the results from this online survey suggest that even highly active exercisers with type 1 diabetes face similar barriers to exercise self-management as the general type 1 diabetes population and that activity-related hypoglycemia remains particularly prevalent in this population. Overall, we also found that there is poor adherence to current guideline-based strategies for managing glycemia before, during, and after prolonged, predominantly aerobic, activities, including most individuals not reducing insulin levels for exercise and not avoiding physical activity after developing severe hypoglycemia and/or severe hyperglycemia with DKA. Moreover, pump disconnection was common even with those using modern AID systems, and ketone testing was typically not done by a majority of individuals even when activity-associated hyperglycemia ensued.

The survey responders in this observational self-report study were likely engaged in considerably more physical activity per week than the general type 1 diabetes population (6,20,23–27), averaging >300 minutes/week of predominantly aerobic exercise, and largely used current diabetes technologies such as CGM and advanced AID systems. Although the use of newer diabetes-related technologies can help to improve overall glucose management, as evidenced in this cohort with a low mean A1C of ∼6.6% and with 67% of individuals achieving >70% TIR, exercise remains a persistent glucose management challenge. Almost all of the survey respondents reported needing to stop exercise because of hypoglycemia, at least on occasion, and all appeared to also struggle with overnight post-exercise hypoglycemia.

It is notable that nearly one in five individuals (18%) reported having had a severe hypoglycemic event during exercise and that 10% of participants reported an episode of DKA precipitated by exercise. Although severe hypoglycemic events could not be adjudicated and a direct causal relationship with DKA could not be confirmed, these are normally infrequent, if not rare, events; therefore, further consideration should be given to the rates found in this group versus in other type 1 diabetes population–based surveys (28). This finding is possibly related to the sample size, self-reporting, and relative homogeneity of participants (primarily Caucasian, in good glycemic management, and using a CGM and/or AID system).

Current consensus guidelines consider a recent severe hypoglycemic event in the past 24 hours to be a contraindication to moderate to intense exercise because of the increased risk of a more serious episode during a subsequent bout of exercise (29). Despite this guidance, approximately two-thirds of individuals reported that they performed exercise within 24 hours of an episode of severe hypoglycemia (typically defined as a confirmed glucose <50 mg/dL [<2.8 mmol/L] or a glucose level that required help from another individual for treatment). Regarding hyperglycemia, consensus guidelines recommend checking blood or urine ketones before exercise if blood or interstitial glucose is unexpectedly elevated to a value >250 mg/dL (>13.9 mmol/L). However, 74% of respondents indicated that they were not checking for ketones in this situation; in fact, in those who monitored ketone levels regularly when unexpected hyperglycemia ensured, some (11%) reported exercising with elevated ketones already present in their blood or urine. Potential barriers to increasing the use of ketone testing include a lack of access to blood ketone meters or urine ketone testing supplies and/or limited insurance coverage of these products. Hopefully, part of this concern may be ameliorated in the future with the current development of continuous glucose/ketone monitoring systems (30). Regardless, when taken together, these findings suggest potential educational opportunities to help improve exercise safety.

While we presume that educational programs would be helpful to increase compliance with guidelines, it is also a possibility that the high motivation to exercise in active individuals with type 1 diabetes is an actual barrier to guideline adherence because, in some situations (e.g., a recent episode of severe hypoglycemia or having hyperglycemia with elevated ketone levels), following the guidelines would limit their ability to perform exercise. Additionally, it is possible that having both more experience with exercise and more episodes of exercise-related hypoglycemia may lead to a higher level of false confidence in glucose self-management. These findings are consistent with a prior study that suggested that, in adults, having a higher proportion of physical activity sessions resulting in hypoglycemia resulted in less concern about activity-related hypoglycemia, simply because participants were somewhat accustomed to it (i.e., they accepted it as part of being physically active and believed they could manage it reasonably, rather than having to fear it) (31).

This survey also serves to highlight the prevalence of hypoglycemia unawareness in this population of active individuals with type 1 diabetes, which is in line with the recent findings of the real-world T1DEXI (Type 1 Diabetes Exercise Initiative) study (32). We found that, although 19% of respondents reported having a diagnosis of hypoglycemia unawareness, approximately half of all survey respondents did not feel symptoms of hypoglycemia beginning until blood glucose levels were already <60 mg/dL (3.3 mmol/L), indicating that hypoglycemia unawareness in this population is both underdiagnosed and underreported. Additionally, we found that those with a diagnosis of hypoglycemia unawareness did not increase protective behaviors or make insulin adjustments compared with those without diagnosed hypoglycemia unawareness. Consequently, the use of CGM and predictive alerts for hypoglycemia during exercise are critical to prevent severe hypoglycemia events (21). However, given that our survey showed that up to 40% of the respondents who used CGM did not use the predictive glucose fall alert, we have identified another area of opportunity for education to alert individuals to times of increased risk of hypoglycemic events.

Our survey demonstrated that, although carbohydrate intake before, during, and after exercise was a frequently used strategy to minimize the risk of hypoglycemia, adjunctive insulin reductions were underutilized. These results mimic the recent findings of Jacobs et al. (16), who found that more proactive measures to prevent activity-related hypoglycemia, such as insulin dose reductions, were uncommon. Although postprandial exercise was recently shown to be safe, effective, and feasible if appropriate precautions are taken (33), a 25–50% reduction in insulin dosing for the pre-exercise meal is still typically needed based on clinical investigation (34–36). By contrast, nearly half of respondents in this study reported that they did not routinely reduce prandial insulin before prolonged aerobic exercise. To minimize IOB during exercise, guidelines suggest a 50–80% basal rate reduction for insulin pump users and/or the setting of a higher temporary target 60–90 minutes before exercise for those on an AID systems, an approach that is associated with lower hypoglycemia risk both during and after exercise (3,4,21,22).

Our survey showed that, despite 93% of respondents reporting that they understood the concept of reducing IOB, only 65% of pump users consistently reduced their basal rate before exercise, and even fewer (27%) made this change >60 minutes in advance of physical activity. Moreover, the majority of respondents who regularly performed a basal rate reduction before exercise reported using a <50% reduction in their usual basal rate rather than the recommended 50–80% reduction (3). For people on an MDI insulin regimen, a 20% basal insulin reduction is recommended for increased physical activity (21); however, the majority (76%) of respondents on MDI therapy reported that they never change their basal insulin dose for exercise. This failure to reduce basal insulin may be related to the inflexibility of using newer long- and ultra-long-acting basal insulins, which are difficult to titrate for exercise without compromising glucose levels at other times of the day (37).

With increasing use of MDI connected insulin pens (sometimes called “smart pens”) and AID systems, clinicians and individuals with type 1 diabetes will need to establish a stronger consensus on what defines IOB, as displayed on these devices. This may be accomplished by using so-called netIOB, a measure of both basal and bolus insulin adjustments that may be more appropriate especially for managing diabetes during exercise (38).

Importantly, our results highlight that most active individuals with type 1 diabetes seek out, trust, and use mostly online resources (e.g., social media, search engines, and videos) in addition to or instead of their health care team for education on exercise management (Supplementary Figure S4A). Our finding of a high percentage of individuals who were not following published consensus guidelines underscores the importance of developing and maintaining updated, accessible, and user-friendly online learning resources for both people with type 1 diabetes and their health care professionals (HCPs).

Doing so will require two distinct approaches to address both the complexity of content and its delivery. A recent Australian survey found that people with type 1 diabetes would prefer having access to a structured educational program, whereas HCPs would prefer having access to e-learning programs because of their time constraints (39). To improve adherence to guidelines, future collaborative efforts should strive to create simplified, accessible, and user-friendly guides to exercise and develop structured educational programs. These programs should be scalable via a teach-the-teacher model, similar to the DAFNE (Dose Adjustment for Normal Eating) program (40), which has been proven effective for carbohydrate counting. Programs are needed that can be initiated from diabetes onset (such as the 4T [Teamwork, Targets, Technology, and Tight Control] program) (41) or added at any time after diagnosis as needed (42,43). These programs should be tailored as needed for people with type 1 diabetes at all stages of exercise participation and performance and for their HCPs, as well as for school teachers and sports coaches.

Limitations

This study had several limitations. Generalizability of our results may be limited by the study’s relatively small sample size, respondent demographics (including limited ethnic diversity, predominantly female participants, and excellent glycemic control), and the higher proportion of respondents using AID systems compared with the broader population of people with type 1 diabetes. Moreover, the survey relied on self-reported responses, which could not be independently validated. The online-only survey format may have excluded groups of active individuals with type 1 diabetes who do not engage in online platforms. This limitation introduces an inherent bias, particularly in findings that highlight a high reliance on social media as a source of exercise information. Furthermore, the survey did not address adherence to guidelines for anaerobic forms of exercise or for high-intensity interval training. Discrepancies in the wording of questions within the survey, as well as variations in participants’ understanding and interpretation of the questions, may have influenced the results related to adherence to exercise safety guidelines. Finally, we acknowledge that recall bias and the generalized wording of certain questions may have led participants to provide broader, less precise responses, potentially limiting the accuracy and specificity of the results.

Conclusion

This online survey of highly active and engaged endurance exercisers with type 1 diabetes using CGM and pump therapy demonstrated a considerable lack of adherence to current exercise management guidelines that are designed to minimize activity-related hypoglycemia and hyperglycemia. This survey study also revealed that respondents made insufficient adjustments for insulin dosing before, during, and after exercise, resulting in a high prevalence of hypoglycemic events and overall safety concerns. Most individuals surveyed did not identify their HCPs as key resources for education regarding diabetes and exercise, and many reported using trial-and-error strategies and nonregulated online information resources to determine their diabetes self-management strategies for exercise. Further support is needed for even highly active individuals with type 1 diabetes, as well as increased engagement with health care teams to ensure that people with type 1 diabetes who exercise are exposed to evidence-based guidelines. With additional engagement, new approaches can be developed to assess individuals’ knowledge and meet their ongoing educational needs.

This article contains supplementary material online at https://doi.org/10.2337/figshare.28212833.