D‐ACTnow—A diabetes‐specific group‐based psychological intervention for young adults with type 1 diabetes and elevated diabetes distress: Results from a single‐arm pilot study

Abstract

Aim

To test the feasibility of a group‐based psychological intervention for young adults with type 1‐diabetes and elevated diabetes distress.

Methods

Young adults (18–30 years), screened for diabetes distress using the Type 1 Diabetes Distress Scale (T1‐DDS), participated in six group sessions led by a psychologist and a diabetes nurse. The intervention was based on Acceptance and Commitment Theory (ACT) and focused on reducing diabetes distress. Feasibility was assessed through trial recruitment, intervention completion, intervention acceptability and preliminary estimates of effect size (T1‐DDS, HbA_1c, continuous glucose measurement), using descriptive statistics, t‐test and systematic text condensation of interviews.

Results

During May 2023–December 2023, 22 young adults were recruited; 20 completed the intervention with high acceptability. Participants (85% women, mean age 23.6 years) showed significant reductions in diabetes distress (T1‐DDS: 1.2, p < 0.0001). No changes were found in HbA_1c or glucose time‐in‐range. Qualitative analyses highlighted benefits of sharing experiences with peers, gaining clarity and practicing coping strategies, supporting the intervention's feasibility and potential effectiveness.

Conclusions

The intervention is feasible and acceptable with encouraging diabetes distress improvements. Minor adjustments will optimize the intervention for the upcoming RCT.

What's new?

• Group‐based psychological interventions can improve diabetes distress in adults, but knowledge on the effect of group sessions led by psychologists and diabetes nurses in young adults is missing.

• The D‐ACTnow intervention was feasible and acceptable and showed reductions in diabetes distress post‐intervention.

• Findings suggest that the combination of a psychologist and a diabetes nurse in the delivery of group‐therapy sessions may be a feasible and effective way to reduce diabetes distress in young adults.

1. INTRODUCTION

Diabetes distress refers to the ‘negative emotional or affective experience resulting from the challenges of living with the demands of diabetes’. ¹ A high level of diabetes distress (e.g., high level of worry, feeling that diabetes takes too much energy, despair, powerlessness and a guilty conscience) can impair quality of life and is associated with poor diabetes management and elevated long‐term blood glucose levels (HbA_1c). ²

A Norwegian nationwide registry‐based study that included >10,000 participants with type 1 diabetes (T1D) found that 22% experienced high levels of diabetes distress. ³ Among younger Australians (13–19 years old), 36% with T1D experience moderate to high levels of diabetes distress. ² A national survey among Danish diabetes patients revealed that 38% of those aged 19–39 years experienced high diabetes distress. ⁴ For young people with T1D, diabetes distress is often linked to concerns about their future and frustration with adherence to a complex disease regimen, which does not comply with their needs and wishes for whom they want to be and what they want to do during the early years of adulthood (educational choices, social relations and gatherings, identity, diabetes technology). ⁵ Emerging adulthood is characterized by major life transitions and increasing responsibility for diabetes self‐management, which may shape diabetes distress differently than in older adults and highlights the need for age‐specific research and interventions. ⁶

Although elevated diabetes distress is not a psychiatric disorder, monitoring diabetes distress and making appropriate care available is essential, as recommended by ADA/EASD guidelines. ⁷ , ⁸ Psychological interventions tailored to diabetes, particularly those utilizing cognitive‐behavioural therapeutic approaches, can be effective in reducing diabetes distress, ⁹ with group formats showing greater efficacy than individual consultations. ¹⁰ A large‐scale U.S. study, EMBARK, found that group‐based interventions reduced diabetes distress and HbA_1c among adults with T1D, and that interventions led by psychologists showed the most significant benefits. ¹¹

No studies have explored the effect of group‐based psychological interventions among young adults (18–30 years) with T1D, and there is a lack of specific tools designed to reduce diabetes distress in this group, for whom diabetes distress may be triggered by other factors (education, identity, new social relations) than for the adult population (+30 years old). Acceptance and commitment theory (ACT) is an evidence‐based behavioural therapy that has shown promising results in the management of chronic diseases. ¹² ACT is designed to help people enhance psychological flexibility. This means, learning to increase awareness of experiences (i.e., mindfulness), accepting uncomfortable thoughts, emotions, bodily sensations or uncomfortable situations as a natural part of life, and committing to behaviours that are consistent with personal values. ¹² Nevertheless, emotional distress often decreases using ACT because of enhanced acceptance and value‐based living. ¹² , ¹³

The aim of this pilot study was to test the feasibility of a group‐based psychological ACT‐informed intervention, (referred to as D‐ACTnow) in young adults with T1D (D‐ACTnow) aimed at reducing diabetes distress. This pilot aims to investigate trial recruitment, intervention acceptability and preliminary results regarding its effectiveness. Results will inform the design of a subsequent RCT study, and qualitative user feedback will refine the intervention.

2. MATERIALS AND METHODS

2.1. Participants and recruitment

Participants were recruited between May 2023 and December 2023. Inclusion criteria: being diagnosed with type 1 diabetes >12 months, receiving care at one of two participating diabetes outpatient clinics; aged 18–30 years; mean item score of ≥ 2 on the Type 1 Diabetes Distress Scale (T1‐DDS), indicating moderate to high diabetes distress ¹⁴ ; and proficient in Danish. Exclusion criteria: pre‐screening procedures conducted by a psychologist excluded participants with psychiatric diagnosis that may compromise participation in the intervention; cognitive impairments such as brain injury and complex mental health problems that are better suited for individual treatment, as determined by the psychologist.

Different recruitment methods were employed. During routine consultations, clinical staff (physicians, optometrists, nurses and dietitians) informed eligible patients showing reduced wellbeing related to diabetes and referred interested candidates for a screening interview. Interested patients received an information flyer. In addition, posters were displayed in the clinics, and posts for recruitment were shared on social media. A clinical psychologist or nurse screened respondents against inclusion/exclusion criteria and obtained written informed consent.

2.2. Study design

This pilot study is a pre–post uncontrolled single‐arm intervention study conducted between May 2023 and April 2024. The intervention was tested in three groups of 6–8 individuals across two diabetes outpatient clinics. Recruitment for group one and group two started in May 2023, and once we had enrolled enough participants for a group, the intervention was initiated. The first group started in September 2023 and the second group in November 2023. Recruitment for the third group started after initiation of the first group, and the third group started their intervention in January 2024. The study follows the Consolidated Standards of Reporting Trials (CONSORT) 2010 statement for randomized pilot and feasibility trials ¹⁵ and was conducted in accordance with good scientific practice and the Declaration of Helsinki.

2.3. D‐ACTnow intervention

2.3.1. Development

D‐ACTnow is a newly developed diabetes‐specific group‐based psychological intervention focused on reducing diabetes distress. D‐ACTnow primarily uses principles from ACT theory, ¹² but also incorporates elements from cognitive‐behavioural therapy and health education principles, both of which are found beneficial in the management of diabetes and diabetes distress. ¹⁶ , ¹⁷ , ¹⁸ Thus, the intervention is not an ACT intervention, but inspired by ACT theory, which serves as an informing framework to design the intervention components. The components of the intervention were primarily developed at Steno Diabetes Center Odense (SDCO) by clinicians (psychologists, diabetologists and nurses) specialized in diabetes care and ACT. The intervention also integrated three selected tools from an existing diabetes distress intervention, called REDUCE, targeting adults with moderate to high diabetes distress. ¹⁹ During the development process, early and ongoing testing of the elements occurred through user‐involvement workshops and individual consultations. Both current and former patients provided feedback regarding the intervention, including text explanations, videos, as well as specific exercises and tools.

2.3.2. Intervention

D‐ACTnow focuses on raising participants' awareness of their thoughts, emotions and beliefs, while also helping them to challenge and change unhelpful thoughts, empowering them with new strategies to manage diabetes and enhance coping skills. D‐ACTnow connects diabetes self‐care to participants' personal values, works on self‐compassion to foster motivation, meaning and the foundation for long‐term changes and habit formation.

D‐ACTnow consists of six sessions, each lasting 2 h, every second week, facilitated by a psychologist and a nurse specialized in diabetes. All sessions had group leaders who completed a 2‐day training course conducted by a psychologist with expertise in ACT and diabetes psychology to ensure a consistent approach throughout the intervention. The intervention follows a manual that guides group leaders in delivering the content consistently. All sessions had specific themes working around the different sources of diabetes distress. Each session followed a structured framework consisting of manualized slideshow presentations and in‐group activities; for example, image card exercises, reflection questions and home practice. Each session, starting from the second session, began with a small mindfulness exercise to foster awareness and create focus. The content of the individual sessions is outlined in Table 1 (described in more detail in Data S1).

TABLE 1.

D‐ACTnow intervention content.

Session	Topic	Purpose
1	Introduction	Create a safe atmosphere that promotes learning and exchange of experiences. Discuss desires and expectations for the group intervention and have participants share their own diabetes stories
2	Knowledge on diabetes distress	General knowledge about diabetes distress and identification of personal sources of diabetes distress. From session 2, participants work with mindfulness and attention training through short mindfulness‐based exercises, ending each session by identifying focus points for continued work at home
3	Youth life and values	Youth life and diabetes, including dilemmas, challenges, and experiences of being a young adult with diabetes. Participants identify personal values and link them to diabetes‐related actions
4	Blood sugar	Experience sharing about challenges with blood sugar and ad hoc knowledge dissemination. Challenge and change unhelpful thoughts about blood sugar and risk. Identification of concerns related to low blood sugar
5	Relationships	Invisible illness and interaction with others, including family, social networks and healthcare professionals. Challenge and change unhelpful thoughts about social involvement Exercise‐based work on desires to involve others in life with diabetes and concrete strategies for handling communication about diabetes with others
6	Closing	Exercise‐based work with a positive view of diabetes (‘What has diabetes given me’). Review and conclusion of the group intervention

2.4. Measures

Feasibility of the pilot study was assessed through trial recruitment, intervention completion, intervention acceptability and preliminary estimates of effect size to estimate the sample size needed for the future RCT study. Measurements were taken at baseline and at follow‐up, approximately 11 weeks after baseline data collection. Trial recruitment, completion and acceptability were assessed quantitatively and qualitatively. Preliminary effect sizes were assessed quantitatively using self‐reported data from questionnaires and clinical measures.

2.4.1. Recruitment, completion and acceptability

Recruitment is depicted in a flow diagram (Figure 1). Participants were profiled by gender, age, education, diabetes duration, pump and sensor use, diabetes‐related complications and mental/somatic comorbidities. Intervention acceptability was assessed by field observations during the 6 group sessions and through 13 single interviews with participants and one focus group interview with group leaders. Study completion was defined as the proportion of participants who completed all or five out of the six group sessions, as well as the attrition rate, with a pre‐defined feasibility criterion of < 20.0%. Survey completion at both baseline and follow‐up was also assessed, with a pre‐defined feasibility criterion of >85.0%.

FIGURE 1.

Recruitment flow.

2.4.2. Outcome measures

The primary outcome measure, diabetes distress, was assessed using the Type 1 Diabetes Distress Scale (T1‐DDS), which contains 28 questions ranging from one (not a problem) to six (a very serious problem). ²⁰ , ²¹ In addition to a total diabetes distress score (the average of 28 items), the scale includes seven subscales covering different sources of diabetes distress (see Table 4). A score of 2.00–2.99 is considered moderate diabetes distress, while a score above 3.00 is regarded as high diabetes distress. ¹⁴ Secondary outcomes were (HbA_1c) and percentage of time with glucose value in range (%TIR 3.9–10.0 mmol/L), the Type 1‐Diabetes Distress Assessment System (T1‐DDAS) subscale ‘Worries about technology’, ²² diabetes‐specific quality of life using the DAWN‐2 Impact of Diabetes Questionnaire—7 items (DIDP) ²³ , ²⁴ and psychological flexibility using the Acceptance and Action Questionnaire (AAQ‐II). ²⁵ %TIR was obtained via electronic upload from glucose sensors and provided a more detailed insight into glycaemic variability. These data were only collected for patients who had a sensor at baseline (90%). HbA_1c values were obtained from clinic records if available. T1‐DDAS is a newly developed scale for measuring diabetes distress. ²² The diabetes technology subscale includes three questions. The scale has been validated in English, but not in Danish. We conducted a forward–backward translation of the three questions to ensure meaning was preserved across languages. The three questions are answered on a 5‐point Likert scale from 1 to 5. A score above 2.00 is considered moderate concerns about diabetes technology.

TABLE 4.

Change in primary and secondary outcomes from baseline to follow‐up (N = 18).

	Baseline	Follow‐up	Change	Effect size d	p a	MCID b
	Mean (SD)	Mean (SD)	Mean (SD)			N (%)
Total diabetes distress score	3.1 (±0.7)	2.3 (±0.6)	−0.7 (±0.5)	1.2	<0.0001	15 (83)
Powerlessness	4.1 (±0.9)	2.8 (±0.7)	−1.3 (±0.7)	1.6	<0.0001
Management distress	3.3 (±1.1)	2.5 (±0.9)	−0.8 (±0.6)	0.8	<0.0001
Hypoglycaemic distress	2.8 (±1.1)	2.0 (±0.8)	−0.75 (±1.0)	0.8	0.006
Negative social perceptions	3.0 (±1.2)	2.2 (±0.8)	−0.76 (±0.9)	0.8	0.002
Eating distress	3.2 (±1.1)	2.3 (±0.8)	−0.9 (±0.7)	0.9	<0.0001
Physician distress	2.1 (±1.2)	2.1 (±1.2)	0.01 (±0.6)	0.0	0.921
Friend/family distress	2.8 (±1.2)	2.2 (±0.8)	−0.6 (±0.9)	0.6	0.004
Worries about technology	2.7 (±0.9)	2.3 (±0.6)	−0.3 (±0.8)	0.5	0.08
Psychological flexibility	30.3 (±8.7)	27.1 (±7.3)	−3.3 (±7.0)	0.4	0.06
Diabetes‐related quality of life	3.1 (±1.0)	3.5 (±0.9)	+0.4 (±1.3)	0.4	0.21
HbA1c in mmol/mol (NGSP %) c	60 (7.7%), (±13.6)	60 (7.7%), (±13.6)	−0.06 (±7.2)	0.0	0.97
Average % Time‐in‐Range c	57.3 (±24.1)	55.9 (±22.5)	+1.4 (±12.8)	0.1	0.67

Note: Diabetes distress measured using T1‐DDS (min. score: 1–max. score: 6), moderate diabetes distress (2.0–2.9), high diabetes distress (≥ 3.0). Worries about technology, measured using the T1DDAS sub scale with 3 items, (min. score: 1—max. score: 5). Psychological flexibility assessed using The Acceptance and Action Questionnaire (AAQ‐2) with 7 items (range 7–49), higher scores indicate less flexibility. Diabetes‐related quality of life assessed using the DAWN‐2 Impact of Diabetes Profile (DIDP‐7 item scale), composite score 1–7.

Abbreviation: SD = Standard deviation.

^a Tested using paired t‐test.

^b MCID, Minimal clinically important difference: −0.19.

^c Missing data for one participant (n = 17).

^d Tested using Cohen's d.

3. DATA ANALYSIS

Quantitative data were analysed using SAS version 9.4 and include descriptive statistics (frequency, percentage, mean and standard deviation (SD)) and dependent samples t‐test, used to compare T1‐DDS diabetes distress mean scores, the seven subscales and the secondary outcomes, between baseline and follow‐up. We also examined the percentage of participants who decreased their mean total T1‐DDS score by the minimal clinically important difference (MCID) > 0.19. ²⁶ Mean, SD, significance level (p ≤ 0.05) and Cohen's d effect size were reported for all outcomes. Qualitative data were analysed using NVIVO version 14. Interviews were transcribed verbatim, and systematic text condensation—as described by Malterud et al. ²⁷ —was used to explore the experiences of participating in the intervention. The systematic text condensation consisted of four steps: (1) two researchers read the material; (2) transcripts were divided between the two researchers and each identified and developed meaning units, followed by a comparison of their meaning units and subsequent adjustments; (3) the two researchers systematically abstracted all meaning units; (4) both researchers re‐conceptualized the data and developed subthemes, which were combined into the final themes. Group session observations were conducted by one researcher, based on an inductive approach, writing down observations made as well as immediate thoughts about the observations. Observation notes were analysed ad hoc, and the analysis shaped while the data were collected. We chose this approach to continuously refine the intervention through observations and to guide the interviews. The final qualitative analysis was thus primarily based on interview data, but observational data were used to support and discuss interview findings.

4. ETHICS

Participants received oral information during screening interviews and written information via Digital Post, where they also provided electronic informed consent before participation and data collection. Participation was voluntary and could be withdrawn at any time without consequences for standard care. The regional research ethics committee was consulted prior to study start (journal no. S‐20222000 – 160). The study is registered in the regional research project registry in the Region of Southern Denmark (journal no. 23/42284).

5. RESULTS

5.1. Trial recruitment and acceptability

5.1.1. Recruitment

During the recruitment period, 65 young adults with T1D were referred to a pre‐screening interview with the aim of inclusion in D‐ACTnow (see Figure 1). Clinicians primarily referred the young adults from diabetes outpatient clinics (83.1%). 14 young adults were excluded because they were not experiencing significant diabetes distress, did not want to participate in a group or were not suitable for group participation. During the visitation interviews, 15 young adults were additionally excluded because they did not meet the criteria for diabetes distress (T1‐DDS ≥ 2.0) and one young adult was found unsuitable for group sessions. Lastly, 13 young adults were excluded for other reasons, despite being assessed as eligible for inclusion. In total, 22 young adults were included in the trial.

5.1.2. Baseline characteristics

Table 2 presents baseline characteristics of participants who completed the intervention. It shows a predominance of female participants (85.0%), with a mean age of 23.6 years (SD 3.6). All included participants were using a continuous glucose monitoring system (sensor), and 55.0% were using an insulin pump. More than half of the participants had at least one other mental health condition (such as anxiety, depression, bipolar disorder, autism or ADHD), and 20.0% reported having diabetes‐related complications. The average HbA_1c at baseline was 60 mmol/mol (7.6%) with 65.0% of participants having an HbA_1c > 53 mmol/mol.

TABLE 2.

Baseline characteristics (n = 20).

Variable	N (%)/mean (SD)
Women	17 (85.0%)
Age, mean (SD)	23.6 (±3.6)
Single	11 (55.0%)
Short education a	10 (50.0%)
Diabetes duration in years, mean (SD)	10.9 (±8.0)
Pump	11 (55.0%)
Sensor	20 (100%)
Diabetes complications b	4 (20.0%)
Mental illness c	11 (55.0%)
Somatic illness	3 (15.0%)
HbA1c in mmol/mol (NGSP %), mean (SD)	60 (7.6%), (±13.4)
% Time‐in‐Range, mean (SD)	58.1, (±23.6)

Abbreviation: SD = standard deviation.

^a Short higher education (under 3 years), secondary education or elementary school.

^b Retinopathy or neuropathy.

^c Anxiety, depression, bipolar disorder, autism or ADHD.

5.1.3. Completion

Of the 22 participants who initiated the intervention, 20 completed the intervention, while two dropped out (attrition rate was thus 9.1%).

Among the 20 young adults who completed the intervention, five attended all six sessions, 10 attended five sessions, and five attended three to four sessions (see Table 3). The average number of sessions attended was 4.9 (±0.9). Analyses of baseline data revealed no significant differences in diabetes distress (T1‐DDS) at baseline between those who completed the intervention and those who dropped out. A total of 90% of those completing the intervention provided both baseline and post‐intervention survey responses.

TABLE 3.

Activity.

Participants completing D‐ACTnow (n = 20)	Mean (SD)	Range
Number of sessions completed	4.9 (±0.9)	3–6
Number of sessions completed, n (percentage)
6 sessions	5 (25%)
5 sessions	10 (50%)
3–4 sessions	5 (25%)
Session duration in weeks	10.3 (±0.6)	10–11

Abbreviation: SD = standard deviation.

5.1.4. Participants' acceptability

The qualitative analysis resulted in four themes related to participation in D‐ACTnow: (1) articulating thoughts and feelings makes it easier to understand and manage diabetes; (2) dividing diabetes into smaller parts provides clarity about personal challenges and strengths; (3) meeting others with diabetes offers comfort, support and the feeling of not being alone; (4) testing new strategies in practice helps establish new coping strategies and habits. Observations and interviews provided insights into the participants' experiences with the intervention and helped adjust the format and content of D‐ACTnow. The four themes are described in more detail in the following sections.

Articulating thoughts and feelings

Dialogue and visual exercises helped participants find language for their diabetes experience and express their thoughts and feelings about the condition:

You physically get to unload these problems, instead of just keeping them in your head. And I think that helps a lot of people. It helps me a lot to be a bit more visual, rather than having everything running through my head. (Pt11, female)

The pre‐printed quotes on dialogue cards, along with the opportunity to hear others' experiences and the group leaders' normalization of common mental challenges related to diabetes, enabled participants to develop their own language and understanding of diabetes. Additionally, the use of images and metaphors, such as describing diabetes as a constant and unpredictable burden, contributed to a deeper understanding of the disease:

The psychologist saying it was like having an extra full‐time job really stuck with me afterward. The idea of saying you have an extra full‐time job and not knowing whether you were on a night shift or a day shift, or when you had to show up. I really liked that reference. (Pt3, female)

Dividing diabetes into smaller parts

Through visual dialogue exercises and discussing life with diabetes, participants were able to break down diabetes into smaller components, which gave them a better understanding of what was truly overwhelming and what was less impactful in their lives with diabetes:

It's not the whole of diabetes that's a problem, but maybe this specific category or area that is the biggest challenge right now. (Pt3, female)

This approach provided a sense of control and a better ability to manage the aspects that were causing challenges:

I just think there were a lot of different things that could be overwhelming. But then you found out that it wasn't everything that was overwhelming, just certain things. So, it doesn't feel so overwhelming. There are actually some things that work just fine. (Pt4, female)

Meeting others with diabetes

Meeting other young people with diabetes contributed to three central elements in managing life with the condition—sharing experiences, feeling safe and supported and feeling a sense of belonging. Exchanging experiences allowed participants to both give and receive useful ideas for managing the challenges of living with diabetes. It was meaningful for participants to receive help and to pass on their own experiences:

It was mostly the same challenges people had. And then maybe someone had cracked the code and could offer a tip on how it was done. (Pt10, male)

Being in a supportive environment made it easier for participants to allow themselves to be emotionally affected:

Personally, I think it hit me harder to share than I had thought. I've always seen myself as a tough cookie, and then I just became ‘soft as butter’. It was a bit strange. I also thought, ‘This is a sign that it's been a safe environment to be in. (PT13, female)

The support and encouragement from group members increased motivation for participants to try new strategies and behave differently:

It's also about the fighting spirit and a push from them [the other participants] that makes it go in a positive direction. Because if they have also worked on improving their lives and achieving their goals, it's like, ‘Oh cool, I can do it too. If they can, then I can. (Pt11, female)

The support and exchange of experiences between group members created a sense of solidarity, as participants realized they were not the only ones facing challenges with diabetes:

It was nice to meet others with diabetes, and to feel like you could relate to them. Also, just knowing that you're not the only one struggling with diabetes, but that everyone there understands what you're going through. (Pt2, female)

Testing new strategies in practice

Support from the group leaders in formulating new behavioural strategies, along with encouragement from group members, allowed participants to gain experience in managing diabetes in new ways. Working on small steps for behaviour change and reflecting on these steps within the group helped make change more manageable and created the foundation for establishing new, helpful habits. As one participant expressed regarding their work with taking insulin at lunch:

I had a few weeks where I set reminders on my phone. I also put notes in my lunchbox when I was at school. But I don't do that anymore. Now, I just do it a little on my own. (Pt9, female)

Patients also worked on spotting diabetes‐specific behaviours and work in helpful and value‐based directions:

It's about stopping up and reflect on what I am actually doing right now. Am I fleeing or fighting right now? And is this actually the right strategy or should I try to do something else? (Pt13, female)

5.2. Outcome data

The average score for diabetes distress decreased from 3.1 (±0.7) to 2.3 (±0.6) (1.2, p < 0.0001), corresponding to approximately one standard deviation (see Table 4). This suggests that the target group experienced a shift from high to moderate levels of diabetes distress. Additionally, 83.0% of participants reported a clinically significant reduction in diabetes distress (T1‐DDS: 0.19) from baseline to follow‐up.

The results also demonstrated a significant reduction in all subscales of diabetes distress, except for the subscale related to physician distress (0.0, p < 0.921). It should be noted that this subscale had the lowest average score at baseline. The most positive reductions in diabetes distress were observed in the subscales powerlessness (1.6, p < 0.0001), eating distress (0.9, p < 0.0001) and management distress (0.8, p < 0.0001).

Secondary outcomes are presented in Table 4. There were no statistically significant changes from baseline to follow‐up for TIR (0.1, p < 0.67) and HbA_1c (0.0, p < 0.97), as well as for psychological flexibility (AAQ‐2) (0.4, p < 0.06), diabetes‐related quality of life (DIDP‐7) (0.4, p < 0.21) and worries about diabetes technology (T1DDAS sub scale) (0.5, p < 0.08).

6. DISCUSSION

This pilot study evaluated the feasibility and potential effectiveness of a group‐based psychological intervention for young adults with T1D and moderate to high diabetes distress. To our knowledge, this is the first study investigating a group‐based and ACT‐informed intervention delivered by a psychologist and a diabetes nurse targeting this population.

We tested feasibility by assessing recruitment, intervention completion and acceptability. Quantitative and qualitative results suggest that the D‐ACTnow intervention is feasible. Recruitment strategies, acceptability and completion are employable in the upcoming RCT study with minor changes.

Participants had a baseline diabetes distress score of 3.1, indicating recruitment reached the appropriate population. Combining pre‐intervention consultations with eligibility screening proved feasible. However, we did not reach gender balance (85% women), unlike other studies showing a more equal distribution of gender (42%–48% men) and no diabetes distress differences between genders. ⁵ This may be due to recruitment strategies, as more women are typically referred to psychologists. Moreso, recruitment was time‐consuming and challenging as several eligible young adults were excluded for different reasons (see Figure 1). Future recruitment efforts could benefit from adopting a range of different recruitment methods, such as referrals from physicians and nurses, posters in clinics, online advertising and phone calls.

Both pre‐defined feasibility criteria for intervention completion were met. The intervention had a dropout rate of 9.1%, which was considered satisfactory, and 90% of completers provided post‐intervention data. Participation in the intervention was high, with participants attending an average of 4.9 out of 6 sessions. Only 25% participated in all six sessions. Consequently, despite the target group being relatively challenged (55% had a co‐occurring psychiatric disorder), D‐ACTnow successfully retained participants in the intervention. Event dropout was lower than in similar trials ²⁸ and attention should be given to mechanisms that may foster adherence in the upcoming RCT study.

Participants expressed satisfaction with the intervention, especially valuing a safe space for meaningful discussions, developing language around diabetes challenges and trying new management strategies. This aligns with other ACT‐informed diabetes interventions. ¹⁹ , ²⁹ However, field notes revealed that some sessions were rushed due to many exercises, and participants wanted more space for sharing personal experiences, suggesting future sessions should prioritize more time for facilitated group discussions. Observations and interviews indicated that participants benefit more from sharing experiences under group leader guidance than from traditional lectures. Similar findings are reported elsewhere, suggesting D‐ACTnow leverages peer‐support mechanisms. ¹⁹ , ³⁰ The intervention was co‐facilitated by a psychologist and a diabetes nurse, which may enhance clinical depth and integration of psychological and medical perspectives. However, this dual‐provider model may limit adoption in routine outpatient settings where access to psychological expertise is constrained. Workforce capacity, competing clinical demands and funding structures may pose challenges to sustainability beyond the research context. While the model was feasible within the present study, its scalability in standard diabetes services requires careful consideration. Previous interventions delivered exclusively by diabetes nurses, such as the REDUCE study ¹⁹ have demonstrated significant reductions in diabetes distress. However, participants in the D‐ACTnow study appeared to have greater psychological vulnerability, including higher baseline levels of diabetes distress and comorbid mental health conditions, compared with those enrolled in the REDUCE study. This increased psychological complexity may necessitate the involvement of a psychologist to adequately address participants' mental health needs.

Effectiveness was primarily assessed by changes in diabetes distress via the T1‐DDS scale. Analyses showed a statistically significant and clinically relevant short‐term reduction in diabetes distress (1.2, p < 0.0001), with participants shifting from high to moderate levels of diabetes distress. Improvements were found in six of seven T1‐DDS subscales, with the largest reductions in powerlessness, eating distress and management distress. We also observed positive, but non‐significant, changes in diabetes technology‐related worries and psychological flexibility. No significant changes were found in HbA_1c or %TIR. These results align with the REDUCE ¹⁹ and T1‐REDEEM ³¹ studies, which also found reductions in T1‐DDS scores. However, both studies involved older participants (almost twice as old), and their interventions were not ACT‐informed. Notably, D‐ACTnow participants also had higher baseline T1‐DDS scores (3.1 compared to 2.6 and 2.9, in the REDUCE and T1‐REDEEM studies), and more than half of the participants reported a psychiatric diagnosis, suggesting that the intervention targeted a particularly diabetes distressed and potentially more vulnerable group. Kortegaard et al. also found diabetes distress reductions in young adults (18–23 years old) participating in group intervention, using PAID‐20 (from 41.3 to 29.1). ³² Another ACT‐informed T1D study in adults showed similar results. ¹¹ The absence of measurable changes in psychological flexibility (conceptualized as the central therapeutic mechanism in ACT theory) raises the possibility that other mechanisms contributed more to the observed reduction in diabetes distress. Elements such as perceived group cohesion, normalization of diabetes‐related challenges, shared lived experiences and peer support may have functioned as key mechanisms. Opportunities to exchange experiences with others facing similar disease‐related burdens may reduce feelings of isolation, enhance emotional validation and foster adaptive coping independently of ACT‐specific processes. Other reasons might be the relatively short intervention period (6 sessions spread over 3 months), which may not be sufficient to improve psychological flexibility. Several participants also expressed difficulties understanding the questionnaire, leading to uncertainty whether participants answered the questionnaire as intended.

Future research should therefore incorporate mediation analyses or additional comparison arms and control groups to examine whether changes in psychological flexibility statistically account for improvements in diabetes distress and to differentiate between ACT‐specific effects and group‐based mechanisms.

6.1. Strengths and limitations

The study provided deeper insights into evidence on group‐based psychological interventions for young adults with T1D and diabetes distress using both qualitative and quantitative data collection. The intervention was thoroughly co‐designed with relevant healthcare professionals and people with T1D, through user‐involvement workshops and individual consultations. This approach will make it more likely that the intervention can be adopted to other clinical settings. Our study had some limitations: The study was not a randomized controlled trial and included only a few participants. Hence, we are not able to compare the results to standard care. Secondly, 85% of included participants were women, which may limit generalizability to young men with T1D. Thirdly, the outcome data were measured immediately after the final group session, and the long‐term effects on glycaemic control and psychosocial outcomes are accordingly uncertain. Fourth, although a detailed manual and training were provided to diabetes nurses and psychologists, full adherence to the intervention protocol cannot be guaranteed, and the intervention may have been delivered with minor variations in practice. Fifth, we did not collect data on participants' mental health support from external services, which, if used by some participants, might affect the results of our study. Finally, DDAS has not been validated in Danish, and it is therefore uncertain if some meanings were lost in the translation.

7. CONCLUSIONS

This pilot study showed promising results, indicating the intervention is ready for rigorous testing in a larger RCT after minor adjustments. Trial completion and intervention acceptability were sufficient, and improvements in diabetes distress are encouraging, which suggest potential for a group‐based psychological intervention for young adults with T1D and moderate to high diabetes distress. The preliminary findings suggest that further research should focus on examining whether the intervention's effect is sustained over time, and which intervention components are most important in reducing diabetes distress. This will be critical to better understand the long‐term benefits and the optimal delivery of the intervention for young adults managing both the psychological and physical challenges of T1D.

FUNDING INFORMATION

The authors received no funding from an external source.

CONFLICT OF INTEREST STATEMENT

None.

Supporting information

Data S1: The D‐ACTnow Intervention.