“It’s Scary, but I Handle It”: A Qualitative Study on Severe Hypoglycemia Experiences and Reasons for Declining to Participate in a Community Paramedicine Trial to Prevent Hypoglycemia Among People With Diabetes

Abstract

OBJECTIVE

Severe (level 3) hypoglycemia is a life-threatening complication of diabetes associated with substantial risk of recurrence and long-term morbidity. This study examined how adults with diabetes interpreted severe hypoglycemic events and why they declined participation in a community paramedicine program (Diabetes Rescue, Engagement, and Management [Diabetes-REM]) designed to prevent recurrence.

RESEARCH DESIGN AND METHODS

We conducted semistructured qualitative interviews with adults with type 1 or type 2 diabetes who experienced severe hypoglycemia requiring emergency medical services and declined enrollment in Diabetes-REM. Interviews explored hypoglycemia management experiences and reasons for nonparticipation and were explored using thematic analysis.

RESULTS

Sixteen adults (44% female; mean age 60 years, range 25–83) participated. Four interrelated themes emerged. First, participants normalized glucose variability as a baseline condition of diabetes and interpreted severe events as situational and correctable when a cause was identifiable, or as inevitable when it was not, rather than as evidence of increased risk. Second, participants relied on prevention systems encompassing habitual routines, caregiver involvement, and glucose-monitoring technologies. Third, severe episodes were frightening and disruptive but typically prompted adjustments to personal management over pursuit of additional clinical support. Finally, reasons for declining Diabetes-REM included time constraints (56%), perceived lack of benefit (25%), and privacy concerns or discomfort with home visits (25%).

CONCLUSIONS

Among adults who declined a community paramedicine hypoglycemia prevention intervention, severe events were absorbed into existing self-management frameworks rather than viewed as signals of increased risk. Programs targeting hypoglycemia recurrence should account for patients’ risk perceptions and existing management systems, positioning support as complementary rather than corrective.

Introduction

Severe or level 3 hypoglycemia is a serious and potentially lethal complication of diabetes capable of causing cognitive impairment, seizures, and loss of consciousness that necessitate assistance from others (1). Severe hypoglycemia also contributes to significant long-term morbidity, including increased risks for cardiovascular disease, dementia, and death (2,3). These events impose substantial psychosocial burdens, often described by patients as frightening and highly disruptive to daily life (4,5). Despite guidelines promoting personalized glycemic management (1), many patients who experience severe hypoglycemia do not receive timely follow-up care to prevent recurrence (6,7). After emergency medical services respond to an episode, patients may decline transport to the emergency department due to concerns about cost, inconvenience, or the belief that further care is unnecessary (8). Consequently, precipitating causes often remain unaddressed, and these patients face nearly double the risk of recurrent severe hypoglycemia within 30 days compared with those who receive hospital evaluation (8). Indeed, even among patients who are treated in the emergency department or hospital, hypoglycemia often remains unaddressed and its recurrence unmitigated (6,8). Accordingly, these patients remain at elevated risk of recurrence without adequate follow-up.

To bridge this gap in follow-up care, the Diabetes Rescue, Engagement, and Management (Diabetes-REM) program was developed for adults with diabetes in a predominantly rural region of southeast Minnesota where access to specialty endocrinology care and formal diabetes self-management education/support (DSMES) is limited (9). This intervention leverages community paramedics with advanced training in management of low-acuity and chronic health conditions (10) to provide home visits and telehealth check-ins after a hypoglycemic emergency, offering education, assessing glucose management, facilitating resource access, and coordinating with primary care and diabetes specialty clinicians. Community paramedicine has demonstrated effectiveness in managing chronic conditions and reducing acute care episodes (11–14), making it uniquely suited to addressing the complex needs of individuals at high risk of recurrent hypoglycemia. Despite the program’s promise (13), initial recruitment revealed that a notable proportion of eligible patients declined participation. Understanding the reasons behind nonparticipation is crucial for improving the acceptability and accessibility of such interventions.

Implicit in many hypoglycemia mitigation approaches is the assumption that a severe hypoglycemic episode functions as a teachable moment: a disruptive event that heightens perceived vulnerability and openness to additional support. However, less is known about how individuals themselves interpret severe hypoglycemia, particularly when they decline structured follow-up care. Prior work on recruitment in self-management interventions has identified factors that may hinder participation, including competing responsibilities, perceptions that the intervention is not needed, and concerns about privacy and inconvenience (15,16). In the specific context of a hypoglycemia-focused community paramedic program, additional nuances may play a role, including patients’ feelings about managing independently, their emotional responses to severe events, and comfort with a health care professional entering their home (17,18). Understanding how patients conceptualize risk, responsibility, and the role of external support is essential for designing interventions that align with lived experience.

To better understand these dynamics, this qualitative study explored the lived experiences of adults who declined participation in Diabetes-REM after a severe hypoglycemic event. While prior work has documented that patients often decline postevent interventions, little is known about the cognitive and experiential frameworks through which they interpret severe hypoglycemia or how these framings shape decisions about engaging with structured support. This study examined participants’ hypoglycemia management strategies and rationales for nonparticipation to inform the design of future interventions better aligned with patient perspectives and needs.

Research Design and Methods

Study Design

We conducted a qualitative study using a descriptive phenomenological approach to understand lived experiences and decision-making related to severe hypoglycemia and participation in a clinical trial of a community paramedic-led DSMES program among people with diabetes. Eligible participants were adults residing in southeastern Minnesota who were eligible for and offered participation in the Diabetes-REM study but had declined participation. This study was approved by the Mayo Clinic Institutional Review Board (Rochester, MN). The study was preregistered with ClinicalTrials.gov (identifier: NCT04874532), and its protocol was previously published (23). All participants provided informed consent.

Participants and Sampling

Participants were community-dwelling adults (age ≥18 years) diagnosed with type 1 or type 2 diabetes who had experienced severe hypoglycemia requiring medical attention. Details of the Diabetes-REM study eligibility criteria are described in the trial protocol (9). Briefly, eligible participants for the Diabetes-REM trial were identified using two complementary approaches: 1) a real-time report generated by FirstWatch, a quality tool used by Mayo Clinic Ambulance, including all emergency medical service encounters with documented glucose <54 mg/dL or where oral glucose, intravenous dextrose, or intramuscular glucagon were administered, and 2) a monthly report from the Mayo Clinic electronic health record identifying adults with diabetes treated in a Mayo Clinic or Mayo Clinic Health System emergency department or hospital for hypoglycemia. Exclusion criteria were not being paneled to a Mayo Clinic or Mayo Clinic Health System primary care clinician, living outside of the community paramedic service’s catchment area in southeastern Minnesota, cognitive impairment, residency in a long-term-care facility or enrollment in hospice, participation in another care coordination or disease management program, and presence of advanced or terminal illness. Eligible individuals were contacted by the study coordinator to explain the Diabetes-REM study and offer enrollment. Individuals who declined participation in the Diabetes-REM trial were then offered to participate in a one-time telephone interview. Oral informed consent was obtained, and the interview was scheduled for a time convenient for the participant.

Data Collection

Interviews were conducted by telephone using a semistructured guide with open-ended questions designed to explore participants’ experiences with hypoglycemia, their diabetes management practices, perceptions of risk, and specific reasons for declining participation in Diabetes-REM. Interviews lasted 30–60 min, were audio recorded, and were professionally transcribed verbatim. To ensure confidentiality, identifying information was removed, and each participant was assigned a unique identifier.

Data Analysis

We conducted a thematic analysis using both inductive and deductive coding approaches. Four research team members (E.H.G., G.L.H., K.T., and D.J.) independently reviewed all transcripts to familiarize themselves with the data. A preliminary codebook was developed by the lead analyst (E.H.G.), and transcripts were then independently coded line by line by the research team. Team members met regularly to discuss coding, resolve discrepancies, and iteratively refine the codebook. NVivo software was used to systematically organize and analyze codes, identifying broader themes and subthemes related to hypoglycemia management and reasons for nonparticipation. Consistent with established conventions for thematic analysis (19), we report findings using qualitative descriptors (e.g., most, several) rather than frequency counts, except where specific responses were directly elicited from all participants and lent themselves to meaningful quantification (e.g., reasons for declining Diabetes-REM participation).

Data and Resource Availability

Deidentified data can be made available by the corresponding author upon reasonable request. Due to the sensitive nature of the data, access will be granted to qualified researchers who provide appropriate funding to cover the costs associated with data deidentification and sharing and receive institutional ethics review.

Results

The study included 16 participants (44% female) age 25–83 years (mean [SD], 59.7 [14.8] years). Participants had either type 1 diabetes (62.5%) or type 2 diabetes (37.5%). Overall, 68.8% used a continuous glucose monitor (CGM) for glycemic monitoring, 37.5% used an insulin pump, and 31.3% used both. Participants were initially identified for inclusion in Diabetes-REM on the basis of having an ambulance call for hypoglycemia (62.5%) or after an emergency department visit or hospitalization for hypoglycemia (37.5%). Overall, 56.0% of participants had at least one emergency department visit or hospitalization for hypoglycemia in the prior year before consenting to take part in the interview. Participant demographics and clinical characteristics are presented in Table 1. Thematic analysis of interviews revealed four interrelated themes: 1) glycemic instability as a normalized feature of diabetes management, including the reframing of severe hypoglycemia as situational or unavoidable rather than indicative of increased risk; 2) layered systems of vigilance used to detect and prevent lows, encompassing embedded routines, caregiver involvement, technology, and escalation safeguards; 3) acute fear and vulnerability during severe episodes followed by adjustments to self-management and compensatory behaviors; and 4) rationales for declining participation in the Diabetes-REM intervention, including broader attitudes related to formal hypoglycemia prevention education and support.

Table 1.

Demographic and clinical composition of participants (N = 16)

Characteristic	Value
Age, years, mean (SD)a	59.7 (14.8)
Age, years, n (%)
18–25	1 (6.3)
26–45	2 (12.5)
45–64	5 (31.3)
65–75	6 (37.5)
≥76	2 (12.5)
Sex, n (%)
Male	9 (56.3)
Female	7 (43.8)
Ethnicity, n (%)
Non-Hispanic	15 (93.8)
Hispanic	1 (6.3)
Race, n (%)
White	14 (87.5)
Asian	1 (6.3)
Black	1 (6.3)
Highest education level, n (%)
College degree	3 (18.8)
Some college	9 (56.3)
High school	4 (25.0)
Diabetes type, n (%)
Type 1	10 (62.5)
Type 2	6 (37.5)
HbA1c, %, mean (SD)b	6.6 (1.1)
Diabetes technology use, n (%)a
CGM	11 (68.8)
Insulin pump	6 (37.5)
Both CGM and insulin pump	5 (31.3)

^aAt the time of interview.

^bMost recent value before the interview as extracted from the electronic health record.

Theme 1: The Persistent but Normalized Threat of Hypoglycemia

Participants described hypoglycemia within the broader context of living with highly variable blood glucose levels. Across interviews, participants emphasized glycemic instability as a defining feature of diabetes management and positioned hypoglycemia, while disruptive and something to be avoided, as an expected and recurrent part of daily life. Within this framing, severe hypoglycemic episodes were characterized not as evidence of higher risk, but as situational events embedded in an already unpredictable chronic condition.

Participants frequently highlighted the unpredictability of blood glucose fluctuations, describing the experience as “pretty much just like a roller coaster…I can never get it right” (P008). Others described diabetes as “just the daily battle” (P002) or characterized their blood sugar levels as simply “not stable at all” (P011). For some, CGM heightened this awareness: “[My CGM] just really opened my eyes to how crazy and up and down my blood sugars go” (P004). Within this broader instability, hypoglycemia was described as a particularly challenging and ever-present concern; for many, an expected occurrence was embedded in daily life rather than a rare event.

Against this backdrop, participants drew a clear distinction between everyday hypoglycemia and severe events requiring outside intervention. Typical lows were described as common and inevitable, while severe episodes were framed as unusual, situational, and attributable to identifiable triggers, such as fasting, exercise, medication changes, or dosing miscalculations. Severe events occurred in a range of circumstances, including during sleep, while driving, and during physical activity, such as mowing the lawn, and were not confined to a particular time of day. One participant summarized a hospitalization bluntly as, “I guessed wrong,” (P010) on an insulin dose. Even when severe episodes had required emergency assistance, participants often minimized their broader significance, incorporating them into a narrative of constant learning and adjustment rather than escalating risk. One participant emphasized that severe episodes had occurred “only twice,” attributing them to specific circumstances he had since learned to recognize (P007). At the same time, glucose variability was described as “having a mind of its own” (P014); some lows were simply unavoidable, with “no rhyme or reason” (P012).

Theme 2: Detecting and Preventing Lows—Vigilance, Technology, and Shared Responsibility

Participants described highly individualized systems for detecting and preventing hypoglycemia. These strategies reflected planning, vigilance, and reliance on technology and family members. Although most expressed moderate to high confidence in their ability to manage diabetes overall, their accounts revealed the substantial cognitive, logistical, and relational work required to prevent and respond to hypoglycemic events (Table 2).

Table 2.

Subthemes and representative quotations related to detecting and preventing lows

Subtheme	Quotations
Embedded prevention routines	“I have—well, liquid glucose, or some candy in my purse. I even have some stashed in my vehicle [laughter].” (P015) “As I said, orange juice. Meat and cheese, high-protein sandwiches. Sometimes a little bit of candy will work, but my glucose tabs have been the biggest. My husband’s got a system set down pat to get it back up.” (P011)
Hypoglycemia symptom recognition and awareness	“It really depends on how fast it’s dropping—I get, like, really warm or shaky and sweaty, just confused sometimes, or I’ll just get really weak.” (P008) “I don’t really start feeling anything until it probably gets in the 40s or 30s. What happens is I just start getting really sweaty and just weak. If it gets really low, I get really confused.” (P004)
Family and caregiver involvement in detection and response	Interviewer: “How do you feel when your blood glucose is low?” Participant: “I don’t know. All right I guess. My wife says, ‘Check your glucometer.’ She says, ‘You’re kinda acting goofy.’” (P001)
Benefits and limitations of diabetes technology	“It’s very, very easy to handle when you have the [application] on the pump or on the phone. Look at the pump all the time. It tells you, obviously, it dings when you go high. It dings when you go low, so you know what to look for.” (P007) “The pump has made a big difference. I’ve had the pump, I wanna say, 10 years, maybe more. It was harder to control my diabetes without the pump because I was taking shots maybe once or twice a day. I wasn’t always taking the right amount of insulin. I would take too much insulin, which would cause me to have a low blood glucose. The pump has helped me to stay level.” (P012) Interviewer: “[The CGM] has been a great tool?” Participant: “Well, there was one time when I knew I was having an insulin reaction, but the continuous glucose monitor was telling me I was 88. I did a fingerstick blood glucose, and it said I was at, I think it was 38. That was kind of a major difference there.” (P015)
Rescue preparedness, glucagon, and thresholds for escalation	Interviewer: “Have you needed to go to the [emergency room] or hospital before because of low blood glucose?" Participant: “I have. They’ve done nothing, just watched. They didn’t do anything. I just sat on a gurney for a few hours, 4 or 5 hours I think it was. They didn’t do anything. They just watched it.” (P011) Interviewer: “When you went in, did they have any idea what happened or what caused it?” Participant: “No. It’s the emergency room. You know how they are. Get in, get out.” (P013) “[The ambulance] just showed up and gave me a bag of juice and stuffed a couple cookies down my neck, and I was fine.” (P003)

Carrying fast-acting carbohydrates at all times was nearly universal among participants, who described keeping glucose tablets, juice, or candy on hand and checking glucose at predictable points (e.g., before driving, during exercise, before bed) as reflexive practices developed over years of lived experience. Beyond these embedded routines, participants provided detailed accounts of how they recognized hypoglycemia, though awareness was highly variable. A subset of participants described episodes in which they did not recognize lows until afterward; for example, one participant passed out while mowing the lawn with no preceding symptoms (P014). Others described cognitive changes that impaired self-response, including confusion, fatigue, and inability to think clearly. Diminished hypoglycemia awareness over time was a recurrent concern, with one participant describing, “I’m finding it harder to tell when I’m low…I could be 40 before I really feel the symptoms” (P012). These accounts underscored the importance of external detection systems when physiologic warning signs were unreliable.

Family members were frequently incorporated into participants’ systems for detecting and responding to hypoglycemia, particularly when symptoms were subtle, awareness was diminished, or episodes occurred during sleep. Caregivers recognized physical signs, provided fast-acting carbohydrates, and in many cases, engaged in digital monitoring through remote access to CGM readings, functioning as an additional layer of surveillance and response. CGMs and insulin pumps were also central tools in many participants’ approaches to self-management. Devices were described as providing reassurance, early warning, and improved glycemic awareness and control. However, reliance on technology did not eliminate uncertainty. Discrepancies between CGM readings and fingerstick values were also described, with some participants reporting clinically significant differences that may have contributed to delayed recognition of hypoglycemia. Notably, interviews did not systematically assess whether devices were functioning correctly at the time of the index severe episode; participants’ accounts focused on their lived experiences rather than technical device performance. Not all participants had access to advanced devices such as CGMs and insulin pumps, highlighting variation in access. Others described barriers to initiating new technology, including time constraints for required training.

Beyond day-to-day prevention, participants described varying levels of preparedness for severe hypoglycemia. While nearly all carried fast-acting carbohydrates, fewer had ready-to-use glucagon available; for some it had never been prescribed, and for others, it had expired due to cost or perceived low likelihood of need. Glucagon occupied an ambiguous position, recognized as an emergency medication but not uniformly integrated into routine preparedness. Emergency services were similarly described as a last resort, with emergency department visits often characterized as involving observation rather than active intervention.

Theme 3: Fear and Recalibration After Severe Hypoglycemia

Although participants generally framed hypoglycemia as manageable, descriptions of severe episodes revealed a notable emotional impact (Table 3). Fear, anger, and frustration were common emotions, particularly when events involved confusion, loss of awareness or consciousness, or emergency services. However, these emotional reactions rarely led participants to define themselves as high risk. Instead, severe episodes were often followed by behavioral recalibration and renewed vigilance.

Table 3.

Subthemes and representative quotations related to the emotional impacts of severe hypoglycemic events and subsequent behavioral modifications

Subtheme	Quotation
Fear and vulnerability associated with severe episodes	“My heart was pounding, which I know to be signs of low blood glucose. Normally, I would catch that, but because I was so out of it, I didn’t…it dawned on me, ‘Oh my god.’ I was trying to tell [my daughter], ‘Call 911. Tell ’em I have diabetes.’ That’s what I kept telling her because I couldn’t think of what else to say.” (P006) Interviewer: “Do you remember feeling any symptoms?” Participant: “Well, I knew I was getting confused, and I carry candy and glucose in my purse…but at that point of time, when I’m that confused, I don’t think about that I have it with me. Now I’m making sure I check before I get behind the wheel.” Interviewer: “That’s very scary.” Participant: “Yeah, definitely, and I mean I could’ve caused an accident or hurt someone, and I did not want to do that.” (P015)
Postepisode recalibration and compensatory behaviors	“I’ve been avoiding [insulin] ’cause I don’t like going low.” (P008) “I have found that I can be like 200, and I can go for a walk, and it will drop off down to 80 or 90. Well, the other day it was 200 when I went out…it was 92. I thought, ‘No, I’m just gonna go home. I’m not gonna do the 12-block walk,’ ’cause I know that if it was 92, it could drop to 65 in no time at all.” (P010) “Well, I don’t feel like I will [have another severe low] because I’m really trying to make sure things are more on an even level now…and I did cut back on my nighttime basal units, like I said, on my pump, so I haven’t been bottoming out during the night.” (P015) “Unfortunately, I overeat then [trying to bring my blood glucose back up] ’cause it’s not coming in too fast, and then it goes up into the 300 or 400s. Kind of a vicious cycle.” (P004)

Fear and Vulnerability Associated With Severe Episodes

Although participants often normalized severe hypoglycemia in retrospect, their accounts of the episodes themselves revealed moments of acute fear and perceived loss of control. Severe lows were described as disorienting and terrifying, particularly when altered mentation left participants unable to think clearly or communicate; one participant described knowing something was wrong but being unable to do anything other than urge her daughter to call 911 (P006). Fear was often tied to the possibility of being alone, with participants expressing ongoing anxiety about having an episode without a caregiver present. Importantly, fear was typically episodic rather than enduring, tending to peak during or immediately after severe episodes and recede as routines were reestablished.

Postepisode Recalibration and Compensatory Behaviors

Despite describing fear, confusion, and vulnerability during severe episodes, participants rarely framed these events as evidence that they could not manage diabetes on their own. Instead, severe hypoglycemia often prompted recalibration: adjustments to routines, dosing, or safety behaviors intended to reduce recurrence risk, including more frequent glucose checks, modified insulin timing, and intentionally higher glucose targets before driving or sleep. Recalibration appeared to reinforce self-efficacy rather than diminish it.

Theme 4: “We Just Have to Do It Ourselves”: Reasons for Nonparticipation in a Community Paramedicine Hypoglycemia Prevention Program

The single most common reason cited for declining to participate in the Diabetes-REM study was related to lack of time and competing priorities (56.3% of participants). One participant stated simply, “It would take up too much time” (P016). Another described reluctance to commit to recurring engagement: “I didn’t wanna be tied up…for a whole month” (P010). For others, competing life stressors were more substantial.

A substantial subset of participants (25.0%) framed their decision to decline as a reflection of long-standing self-management skills and confidence in their ability to handle diabetes independently. Lived experience was described not simply as familiarity but as expertise: “No, there is nothing they could teach me about diabetes now. I’ve picked up on every scenario that you might have, from low blood sugars to high blood sugars, and everything in between” (P014) (Table 4).

Table 4.

Reasons for declining participation in the Diabetes-REM program

Reason	Percentagea	Representative quotations
Competing priorities and time constraints	56.3	“The reason I was really busy. I didn’t have time to do nuthin’.” (P001) “It just sounds to me like something that would take up too much time.” (P016) “I didn’t wanna be tied up with anything like that for a whole month.” (P010)
Does not recall being asked to participate	37.5	“I don’t remember what it entailed or anything. Did I say that I wasn’t interested? I don’t know what to say. Maybe I just didn’t feel like it that day.” (P004) “I don’t recall anybody talking to me about that.” (P007)
Lack of perceived need	25.0	“Yeah, but I’m not interested. I’ve had diabetes for so long. I pretty much know how it works.” (P002) “I just felt like it may be something that I really didn’t need since I’ve had this for so long. I know what foods to eat. I know what I need to do. I feel like I’m experienced enough in this that I can take pretty good care of myself.” (P015)
Privacy concerns	25.0	“Realistically, I’m kind of private. As far as information and stuff like that, I just don’t want to really deal with people like that.” (P009) “I just don’t like people coming into my home. This is my private area…I get anxiety when people come into my house. I don’t know why.” (P008)

^aPercentages do not sum to 100% as some participants gave multiple reasons.

Some participants also described hypoglycemia prevention through a lens of personal responsibility, characterizing the role of health care institutions as relatively futile: “What are they going to do?…We just have to do it ourselves” (P010). Participation in a structured intervention like Diabetes-REM was viewed as unnecessary—not reflecting opposition to program providers but a perception that additional education or monitoring would not meaningfully change outcomes.

Reluctance to engage in a home-based program for privacy reasons emerged as another barrier (25.0%). Participants expressed discomfort with having medical personnel enter their homes, citing concerns about personal space; scrutiny of their lifestyle; and, for one participant recruited during the coronavirus disease 2019 pandemic, infection risk, “For people to come to my house right now, with the COVID stuff that’s going on, I don’t let anybody in” (P011).

A substantial subset of participants (37.5%) did not recall declining participation or reported a lack of clarity about the program’s purpose. Representative quotations and percentages for reasons for nonparticipation are shown in Table 4.

Conclusions

Among adults with diabetes who recently experienced severe hypoglycemia but declined participation in a community paramedicine DSMES program offered as part of a research study, severe events were rarely interpreted as markers of heightened vulnerability. Instead, participants drew on two complementary framings that, while seemingly in tension, served a common function: attenuating perceived risk. When a precipitating cause could be identified (e.g., a dosing error, an unusually strenuous activity, a missed meal), the episode was cast as situational and correctable, a lesson learned rather than a sign of ongoing fragility. When no clear cause was apparent, the event was attributed to the inherent unpredictability of glucose variability, rendering it an inevitable feature of the condition rather than a personal failure or a harbinger of future episodes. In both cases, severe events were absorbed into an existing narrative of diabetes as fundamentally unstable but manageable rather than prompting a reassessment of personal risk. This dual framing helps explain why participants who had recently required emergency care nevertheless declined structured follow-up support.

A core assumption underlying many postevent interventions is that severe hypoglycemia functions as a teachable moment (i.e., a disruptive event that increases a patient’s desire for additional clinical support). However, in our study, rather than serving as inflection points, severe events were typically incorporated into an existing framework of lived experience and self-acquired expertise. Participants distinguished between routine lows and rare severe episodes, framing the latter as episodic miscalculations or the result of identifiable triggers rather than evidence of inadequate self-management. This interpretation attenuated perceived future risk and the perceived necessity of additional intervention.

Participant accounts of individualized, layered self-management (e.g., access to fast-acting carbohydrates, caregiver involvement, CGM) reflected both competence and burden. The cognitive and logistical effort required to prevent and mitigate lows was substantial, echoing prior work on the demands of diabetes self-regulation (20,21). Impaired hypoglycemia awareness was a recurrent concern, and external detection systems, such as CGMs and family members, appeared to compensate when physiological signals were unreliable, consistent with prior qualitative work in similar populations (18,22–25). While CGM use is strongly recommended for high-risk patients (26,27) and reduces hypoglycemia risk among individuals with both type 1 and type 2 diabetes (28,29), it does not eliminate severe events, even among users of automated insulin delivery systems (30,31). This limitation was encountered directly by participants in our study.

Severe episodes were described as frightening and disruptive, particularly when associated with confusion, impaired driving, loss of consciousness, or being alone—findings consistent with prior literature on the psychosocial impacts of severe hypoglycemia (4,18,32). Ensuing fear of hypoglycemia shaped subsequent self-management behaviors to some degree, a phenomenon documented in previous studies (33). However, rather than undermining participants’ sense of competence, fear more often prompted recalibration or adaptations. This aligns with evidence that fear of hypoglycemia can drive defensive behaviors, sometimes at the expense of optimal glycemic control (33–35). Importantly, among participants in our study, recalibration appeared to reinforce self-efficacy rather than diminish it. Severe events were interpreted as feedback or signals to refine self-management behaviors rather than as indications of fragility requiring additional, external oversight. This reframing has direct implications for intervention design: if severe events tend to consolidate rather than destabilize patients’ sense of competence, then programs framed as corrective may land poorly with the very patients they aim to reach.

Despite well-established recommendations for glucagon access among patients at risk for severe hypoglycemia, our participants described it as peripheral to their actual response strategies. Most instead relied on fast-acting carbohydrates even during severe episodes, consistent with prior reports of low glucagon use (36). Glucagon prescriptions were sometimes allowed to expire on the basis of cost or perceived low need, and emergency department visits were characterized as offering observation rather than intervention beyond what participants could provide themselves. This pattern likely reinforces, rather than challenges, confidence in existing self-management systems and points to a meaningful gap between guideline-recommended preparedness and lived practice.

Against this backdrop, decisions to decline Diabetes-REM participation reflected an internally coherent calculus driven by confidence in existing routines, limited perceived added benefit, and practical barriers, including logistical concerns, privacy considerations, competing responsibilities, or discomfort with home visits. These findings align with broader literature on barriers to community-based care delivery (14,16) and on the role of perceived necessity, trust, and contextual constraints in recruitment to self-management interventions (37). The implication is not that such programs should be abandoned but that their framing matters. Interventions positioned as corrective measures for inadequate self-management are unlikely to engage patients who, in their own assessment, are already managing well. Conversely, interventions positioned as additive supports, with explicit acknowledgment of the substantial self-directed work that patients are already doing, and that engage family members and caregivers as partners in detection and response (38) may better align with how this patient population actually experiences and manages their condition. Critically, our findings suggest that nonparticipation is not driven by avoidance or disengagement but, rather, reflects an internally coherent assessment of risk, competence, and need. This distinction has direct implications for how future programs are framed, delivered, and evaluated.

Strengths and Limitations

This study had several strengths. We interviewed adults with both type 1 diabetes and insulin-treated type 2 diabetes who had recently experienced severe hypoglycemia and were offered, but declined, a structured DSMES intervention. The sample included individuals with varying durations of disease, levels of technological use, and social contexts. This diversity allowed for exploration of both shared themes and meaningful variation in decision-making.

Several limitations also warrant consideration. First, all participants were recruited from a single health system in the Upper Midwest region of the U.S., which may limit generalizability to other settings. Second, recall bias may have influenced descriptions of both severe events and decisions to decline, particularly when time had elapsed between the emergency episode and interview. Third, interviews did not systematically assess the functionality of diabetes technology devices (e.g., CGMs, insulin pumps) at the time of the index severe hypoglycemic event; future studies examining the relationship between device performance and severe hypoglycemia in real-world settings would complement these findings. Finally, as with all qualitative research, findings are interpretive and intended to illuminate mechanisms and meanings rather than estimate prevalence.

Implications for Practice

First, variability in hypoglycemia awareness and response across participants reflects broader inconsistencies in clinical education and communication about hypoglycemia risk. Specifically, discussions about hypoglycemia occur in only a minority of clinical encounters, despite their importance for diabetes management (7,8). Patients with prior severe hypoglycemia are at highest risk of recurrence, and conversations should explicitly address this rather than treating each event as isolated. In addition, the fear and uncertainty associated with severe events underscore the psychological dimensions of hypoglycemia management, suggesting the importance of emotional support as a component of clinical care (39). Second, the mixed understanding about and use of glucagon point to a clear target for education of health care professionals and people with diabetes and their caregivers that covers not only access and availability but also the situational judgment required to recognize when emergency intervention is warranted (40). Finally, addressing practical barriers to participation in DSMES programs, such as perceived invasiveness, time constraints, and competing priorities, is essential.

In conclusion, adults who declined a community paramedicine DSMES intervention after severe hypoglycemia did so for reasons rooted in self-efficacy, normalized risk perceptions, and practical constraints rather than indifference or lack of concern. Severe events, while frightening, were typically incorporated into existing self-management frameworks rather than experienced as catalysts for seeking additional support. Future efforts to prevent recurrent hypoglycemia should center patient autonomy and frame structured support as complementary to, rather than a replacement for, the considerable expertise that people with diabetes develop through lived experience.

Funding Statement

This effort was funded by the National Institutes of Health (NIH) National Institute of Diabetes and Digestive and Kidney Diseases (grant R03DK127010 [to. R.G.M.]) and the NIH National Institute on Aging (grant P30AG097158). R.G.M. is an investigator at the University of Maryland Institute for Health Computing, which is supported by funding from Montgomery County, MD, and the University of Maryland Strategic Partnership MPowering the State, a formal collaboration between the University of Maryland, College Park, and the University of Maryland, Baltimore.

The study contents are the sole responsibility of the authors and do not necessarily represent the official views of the NIH.