Theory-based process evaluation of a non-randomised single-arm pilot study investigating time-restricted eating in the treatment of type 2 diabetes—the RESET2 pilot study

Abstract

Background

This study describes a 12-week non-randomised single-arm pilot study, which was conducted to inform the design of a 1-year randomised controlled trial (RCT) on time-restricted eating (TRE) in individuals with type 2 diabetes. Core intervention components were developed to operationalize findings of a needs assessment study and the resulting initial program theory suggesting causal associations influencing TRE adherence. This study assessed the implementation and acceptance of core intervention components and activities, and participant experiences with TRE, and based on the findings, the initial program theory was refined.

Methods

In this non-randomised single-arm study, nineteen individuals with type 2 diabetes completed 8 weeks of TRE with a 10-h eating window, followed by 4 weeks of individualised TRE (core component 1). Participants had two 1-h conversations with healthcare professionals (HCPs) at baseline and after 8 weeks to support TRE initiation (core component 2) and maintenance (core component 3). The process evaluation was conducted using a concurrent mixed methods design with emphasis on qualitative insights. Delivery and implementation of intervention activities, including visits, measurements, conversations, and phone calls, as well as the two-phase design, were evaluated. The conversations were audio-recorded to investigate the extent to which they were conducted as intended. Interviews (by a researcher) at baseline, 8, and 12 weeks explored participants’ experiences, support received, and TRE implementation in daily life. The interviews were audio-recorded. A deductive qualitative content analysis approach was applied.

Results

Intervention activities were delivered with high degree of consistency, and the two-phase design (strict respectively individualised TRE) was effectively implemented. Participants found the TRE intervention highly acceptable, appreciating the relevant information, well-organised trial visits, supportive HCPs, and encouraging conversations. Analysis of participants’ experiences of implementation of TRE generated themes related to routines and habits, support from relatives, social life, and mental and physical challenges. Intervention components such as conversations and individual support and adjustment options were found to facilitate TRE initiation and maintenance. The initial program theory was refined in terms of how the intervention activities supported participants differently and how various contextual factors influenced their experiences and adherence to TRE.

Conclusions

Insights from participants’ experiences with intervention activities and TRE performance in everyday life provided a nuanced understanding of what works, for whom, how, why, and under which circumstances in terms of TRE adherence. These findings will help tailor the following RCT intervention to the target population.

Trial registration

Clinicaltrials.gov, NCT05375695. Registered 27 March, 2022: https://clinicaltrials.gov/study/NCT05375695?term=NCT05375695&rank=1

Key messages regarding feasibility

1. What uncertainties existed regarding the feasibility?

   • Previous studies on time-restricted eating (TRE) interventions in individuals at risk of or with type 2 diabetes have reported significant variability in adherence.
   • The impact of implementing intervention activities to support participants’ retention and adherence is unclear.
2. What are the key feasibility findings?

   • The core intervention activities were effectively delivered and implemented, including the two-phase design with a strict phase followed by an individually tailored phase, as well as the two guided conversations focused on planning and adjusting TRE.
   • Participants found the TRE intervention highly acceptable, appreciating the relevant information, the support from healthcare professionals, and the constructive conversations that helped them initiate and sustain TRE.
   • Participants had diverse experiences with TRE; some found it easy to follow, while others struggled to adhere to the designated eating window. However, intervention activities such as conversations and individual support and adjustment options helped participants in implementing and maintaining adherence to TRE.
3. What are the implications of the feasibility findings for the design of the main study?

   • Participants demonstrated very diverse needs, emphasising that supporting interventions activities must include a variety of options delivered with flexibility in the coming randomised controlled trial.

Background

Type 2 diabetes is a chronic, multifactorial condition in which complications and progression are strongly influenced by glycaemic control 1,2 [1]. In addition to pharmacological treatments, strategies to improve glycaemic control and weight loss often focus on behavioural changes involving diet and physical activity [2–4]. However, maintaining these changes over the long term can be difficult for many individuals [5, 6].

Time-restricted eating (TRE), a form of intermittent fasting, shortens the daily window for eating and drinking, typically without imposing restrictions on the types or amounts of food and beverages consumed [7]. This may make TRE an appealing option for those who find more common diets with energy restriction challenging [8]. Few studies have investigated TRE in individuals with type 2 diabetes, showing promising effects on body weight and glycaemic control [9–17]. However, most of these studies have been short in duration and lacked control groups. Longer-term studies, particularly randomised controlled trials (RCTs), are needed to establish robust evidence for the impact of TRE on glycaemic control and other key outcomes relevant to type 2 diabetes management.

RCTs are widely regarded as the gold standard for establishing the efficacy of interventions, providing evidence in a certain setting for a specific population [18]. High degree of adoption and maintenance of TRE in daily living is important for investigating the effects of TRE and, also, for the potential of TRE as a relevant long-term treatment strategy in future clinical practice. We acknowledge that changing habits takes time, and the TRE intervention should incorporate a variety of activities and practical strategies to support participants in adopting the new behaviour and be transferable to practice afterward to make the intervention clinically relevant [19, 20]. However, as individuals have different needs, capabilities, and circumstances (context), support activities should encompass a range of options delivered with flexibility and thus individually tailored [21]. Also, it is important to recognise that the participants are ‘on their own’ at home while they are (trying to) implement TRE in their everyday life. Process evaluations in trials are increasingly valued for providing critical insights into how interventions are implemented and whether they are delivered as intended in approach, form, and content to effectively trigger causal processes. They also explore participants’ experiences and perspectives on their individual implementation and the promoted changes, offering a deeper understanding of personal processes and contextual influences. Additionally, process evaluations assess how intervention activities may facilitate or hinder participants’ adoption of TRE [18, 22].

This pilot study process evaluation is related to phase 2 in the REStricted Eating Time in the treatment of type 2 diabetes (RESET2) project, which consists of 3 phases: (1) a needs assessment study [23]; (2) a non-randomised single-arm pilot study; and (3) a 1-year RCT. The primary aim of the RESET2 RCT is to investigate the effects of TRE on glycaemic control in individuals with type 2 diabetes. The TRE intervention is designed to assist participants in planning, initiating, and maintaining TRE by providing guidance, optional support activities, and individual adjustments and is based on the needs assessment study [23]. The evaluation of the RESET2 pilot study is framed by a logic model and the process evaluation is using a theory-based approach and is based on an initial program theory (IPT) [18, 22] focusing on hypothesised causal associations, including theories of actions and theories of change, related to TRE adherence. Theory-based evaluation helps understand how contexts influence outcomes, providing insights into what works for whom, how, why, and under which circumstances [18, 22]. This involves investigating participants’ experiences and the impact of intervention activities on TRE implementation and will inform the planning of the RCT.

Study objectives

1. To evaluate the implementation of the intervention by determining the extent to which core components and activities were implemented as intended to activate hypothesised causal processes.

2. To explore participants’ acceptability of the invention’s core components and activities.

3. To explore participants’ experiences with implementing TRE in their everyday lives, focusing on the changes it fostered, including coping strategies, barriers, facilitators, motivation, and impact on physical and mental well-being.

4. To combine findings of objectives 1, 2, and 3 to evaluate what works for whom, how, why, and under which circumstances to initiating and maintaining TRE and refine the IPT accordingly.

5. To suggest adjustments to the intervention design in the coming RCT in terms of new or adjusted components/activities or removal of redundant elements.

Material and methods

Study design and setting

This process evaluation was conducted as part of a non-randomised single-arm pilot study investigating TRE in individuals with type 2 diabetes. A concurrent mixed method design with emphasis on the qualitative component was applied. The study was conducted in Denmark at the Steno Diabetes Center Copenhagen (SDCC) and Hvidovre Hospital between April and November 2022. The study was approved by the Ethics Committee of the Capital Region of Denmark (H-21041830) and registered at Clinicaltrials.gov (NCT05375695). All participants provided informed consent. Further information on methods and feasibility outcomes are being reported elsewhere [24].

Participants

Inclusion criteria were ≥ 18 years, diagnosed with type 2 diabetes and related complications, HbA_1c > 53 mmol/mol, BMI ≥ 25 kg/m², and a habitual eating window ≥ 12 h/day. Participants were recruited at regular control visits at two different areas (SDCC and Hvidovre Hospital) to include different socioeconomic backgrounds with approximately 75% from SDCC and the remaining from Hvidovre Hospital. Participants who expressed interest in the study were contacted by a healthcare professional (HCP) to have more information about the study. During participant recruitment, we made an additional effort to encourage participants to involve a relative (family member or friend) in the TRE intervention. This aspect was further explored in the interviews, based on our hypothesis that support from relatives would be an important factor in adherence to the intervention. Details on eligibility are being reported elsewhere [24].

Sample size

Sixteen participants were considered sufficient to obtain insights into the intervention feasibility in terms of recruitment, retention, intervention fidelity, acceptability, and adherence in a non-randomised single-arm pilot study [25]. To account for potential drop-out, 20 participants were included. Participants’ acceptance of the intervention components and adherence in terms of experiences with implementing the intervention in everyday life were investigated by interviews. In this regard, we adopted the concept of information power (rather than data saturation) [26] by interviewing each participant three times to investigate individual perspectives and processes in-depth in the framework of a program theory. Furthermore, all participants belonged to the specified target group, and at the same time to enable diversity related to contextual factors, participants were recruited from different locations. In terms of clinical measurements, it was not the intention of this pilot study to evaluate the clinical effects with statistical power. Clinical parameters were measured only for the purpose of testing the feasibility of the data collection methods and the effects are reported as preliminary results.

Initial program theory and intervention description

High study retention and adherence is decisive for assessing the clinical effects of TRE in the RESET2 RCT, and the RESET2 pilot study focused on testing and evaluating the TRE intervention to gain knowledge of and improve retention and adherence. The intervention was developed in a co-creation process involving a needs assessment study [23] which together with established theories on health behaviour change [19] informed the development of an IPT. The IPT suggests key causal associations expected to influence TRE initiation, intervention retention rates, and degree of adherence to TRE which are presented in Table 1. Based on the IPT, three core intervention components were formulated:

1. First, it was important for participants to gather lived experiences with a reduced eating window to discover potential challenges related to TRE. These insights helped in aiding optimal planning of the eating window and to discuss and prioritise adjustments and support in the longer-term perspective and underpins why the intervention had a two-phased design, one phase with strict TRE followed by a phase with individual adjustments and support (core component 1).

2. Second, participants received individual guidance and support to plan and initiate TRE in 1-h conversations with an HCP at baseline (core component 2).

3. Third, participants discussed challenges and needs they experienced during the initial 8 weeks of strict TRE in 1-h conversations with an HCP to facilitate maintaining TRE by implementing individually chosen adjustments and support (core component 3).

Table 1.

Initial program theory of the RESET2 pilot study

Initial program theory:	If…	Then…	Because…a	Relation to intervention core components and/or effects of TRE:
1	HCP uses a whole-person approach and draws on participants’ previous experiences and includes a perspective on daily life, e.g. psychological, and social factors when providing individual guidance and support …	Planning and initiating TRE feel easier and more manageable by individual participants…	The process is tailored to individual needs and preferences	2
2	HCP uses a whole-person approach and draws on participants’ previous experiences and includes a perspective on daily life, e.g. psychological, and social factors when providing individual guidance and support …	Individual participants feel supported and confident in maintaining TRE…	The process is tailored to individual needs and preferences	2 + 3
3	HCP uses a whole-person and dialogue-based nonjudgmental communication approach when explaining the TRE concept and intervention …	It facilitates making sense of TRE in everyday life for individual participants…	The process is tailored to individual life circumstances	2
4	HCP uses a whole person approach and engages in dialogue-based nonjudgmental communication …	Participants feel safe and dare to talk about difficult issues related to TRE (which enhances the possibilities to obtain tailored and relevant support)…	A trusting relationship is created	2 + 3
5	Participants obtain personal experiences in initiating and implementing strict TRE in everyday life …	Individual participants feel/understand what the most challenging aspects are to them in following TRE in everyday life…	They obtain personal experiences from their everyday life empowering them to choose the most helpful and relevant support option(s)	1
6	Dialogue on individual choice of support and adjustments are based on participants’ personal challenges …	Individual adjustment and support activities feel relevant and motivating for continued TRE	The process promotes reflections and learnings	1 + 3
7	Participants join peer support sessions …	They feel genuinely supported in terms of feeling understood and more knowledgeable	They experience to mirror and model other participants as well as being mirrored and modelled by other participants	3
8	Participants experience improved health benefits during intervention …	They feel encouraged/motivated to continue TRE	It makes sense to them/they feel moving in the right direction	Clinical effects
9	Participants involve family and/or relatives in their TRE implementation …	They experience being supported practically in terms of adjustment of plans, habits and structures	It feels like a joint goal to accommodate TRE in everyday life	1 + 2 + 3 + changes related to TRE implementation
10	Participants involve family and/or relatives in their TRE implementation …	They experience being supported mentally	They share thoughts and feelings with people they know	1 + 2 + 3 + changes related to TRE implementation
11	Participants implement individually chosen adjustments such as occasionally having a day off from TRE …	They generally feel it easier to maintaining TRE and/or retaining in a TRE intervention	They feel less guilt when deviating from the eating window	3

HCP Healthcare professional, TRE Time-restricted eating

^aSupporting theories: Empowerment, Social Cognitive Theory, Self-efficacy, Self-determination Theory, Sense of Coherence, Theory of Planned behaviour, Trans-Theoretical Model of Change, Normalisation Process Theory, Social Norms Theory

For core intervention components 2 and 3, the conversations were based on a whole-person approach and dialogue-based non-judgmental communication. These three core components were the focus of the process evaluation of the pilot study as illustrated in the logic model in Fig. 1. The logic model also connects the pilot study to the subsequent RCT, which is informed by the findings of the pilot study. The outcomes of the process evaluation align with the objectives outlined above and illustrated in Fig. 1.

Fig. 1.

Logic model of RESET2 pilot study. HCP, healthcare professional; RCT, randomised controlled trial; TRE, time-restricted eating. *Described elsewhere [24]

The TRE intervention consisted of two phases: an 8-week strict TRE period with a fixed, self-selected, 10-h eating window (a daily interval for eating and drinking between 6 am and 8 pm), followed by a 4-week TRE period incorporating individually chosen adjustments and support options (Fig. 1 + Fig. 2). These options included occasional adjustments in the eating window (e.g., occasionally rescheduling of the eating window to fit daily life, and days off TRE), energy and caffeine-free beverages outside the eating window, and support from HCPs, peers, and relatives (Fig. 1 + Fig. 2). Participants participated in two 1-h conversations with HCPs at baseline (visit 1), and after 8 weeks (visit 2).

Fig. 2.

The RESET2 pilot study design. EW, eating window; TRE, time-restricted eating

Data collection

All study data were collected and managed through REDCap electronic data capture tools (V.13.7.14, Vanderbilt University, TN, USA) hosted at the Capital Region of Denmark [27, 28].

Output related to the implementation of the intervention

The degree to which appointments, measurements, questionnaires, and conversations were conducted as intended.

From case-report forms, we collected data on dates for conducted visits and phone calls and conducted measurements and tests (body weight/bioimpedance, waist/hip circumference, HbA_1c, blood pressure, and food preference task), completed questionnaires, and notes from conducted conversations (Figs. 1 and 2).

Core component 1

We collected data on dates for visits 1, 2, and 3 for all participants in the case-report forms (Figs. 1 and 2).

Core components 2 and 3

The content, flow, and approach of the conversations at visits 1 and 2 were described in protocols, developed in accordance with the IPT (Table 1). All conversations were audio-recorded (except for one participant, who did not want conversations or interviews recorded, and field notes were used instead) (Figs. 1 and 2).

Intervention acceptability among participants

Telephone interviews after visits 1 and 2, as well as interviews at visit 3, followed semi-structured interview guides and were conducted by an experienced qualitative researcher (AV). Questions covered participants’ experiences with the following pre-defined topics: (1) visit structure; (2) information level; (3) measurements and questionnaires; (4) HCP’s approach; (5) telephone calls; and (6) conversations including tools (Fig. 1). All interviews were audio-recorded.

Participants’ experiences of implementing TRE in everyday life

The interview at visit 3 also explored how participants implemented TRE, its impact on everyday life, and how their unique circumstances influenced this process. Specifically, the interview addressed the challenges faced, coping strategies, barriers, facilitators, motivation, and impact on physical and mental well-being. A semi-structured interview guide was used, and all interviews were audio-recorded.

Data analysis

Description of study population

Baseline data are reported as mean (standard deviation (SD)) (if data were normally distributed) or median (quartiles 1 and 3 (Q1, Q3)) (if data were not normally distributed).

Output related to the implementation of the intervention

The degree to which appointments, measurements, questionnaires, and conversations were conducted as intended.

The extent to which appointments, measurements, questionnaires, and conversations were completed or conducted is reported in percentage (number of completed or conducted appointments/measurements/questionnaires/conversations divided by total appointments/measurements/questionnaires/conversations during the study × 100) (Table 2).

Table 2.

The degree to which appointments, measurements, questionnaires, and conversations were conducted as intended

Extent to which appointments were kept (%)	Visit 1: 19/19 = 100% Visit 2: 19/19 = 100%* Visit 3: 19/19 = 100% Phone calls between visit 1 and 2: 37/38 = 97%
Extent to which clinical measurements/tests and questionnaires were completed (%)	V1, V2, and V3: Body weight/bioimpedance: 57/57 = 100%* V1 and V3: Questionnaires 38/38 = 100% Waist and hip circumference 38/38 = 100% Blood sample (HbA1c): 36/38 = 95% Blood pressure 38/38 = 100% Food preference task 38/38 = 100%
Extent to which the conversations were conducted (%)	Conversation at visit 1: 19/19 = 100% Conversation at visit 2: 19/19 = 100%*

^*All were conducted in person, expect for two who wished for online meetings at visit 2

Core component 1

The duration of the two phases was calculated and reported as mean (SD) days and weeks (Table 3).

Table 3.

Core component 1: duration of phases

	Days	Weeks
Phase 1 – Strict TRE	55.3 (2.0) days	7.9 (0.3) weeks
Phase 2 – Individually adjusted TRE	27.7 (3.4) days	4.0 (0.5) weeks

Data are mean (standard deviation)

Core components 2 and 3

A student assistant reviewed audio recordings from conversations conducted at visits 1 and 2. For each participant, the assistant documented how the conversation unfolded and assessed whether the tasks/topics were addressed as outlined in the protocol. The tasks/topics that should be addressed are presented in left columns in Tables 4 and 5, which functioned as checklists in the analysis. For each task/topic, we evaluated the extent to which the tasks/topics were addressed as intended (protocol) across participants by calculating the percentage of coverage (i.e., number of participants for whom the task/topic was covered divided by total number of participants × 100).

Table 4.

Core component 2: individual guidance and support to plan and initiate TRE taking a whole person/person-centred approach

Task/topic	Procedure (protocol)	Degree to which the task/topic was addressed as intended in the conversation
Presentation of the study design	The HCP provides a schematic overview of the RESET2 pilot study design, including a timeline outlining the three visits and a brief description of each visit’s purpose and measurements. Additionally, a detailed overview of each visit, featuring icons representing all measurements and a timeframe for the visit is presented. Both the study overview and visit-specific overviews are included in the participant folder, which is given at the screening visit following project inclusion	18/19 = 95%*
Presentation of TRE concept	The HCP presents a figure with two 24-h clocks: the first illustrating a typical eating window (e.g. 7 am to 11 pm), and the second depicting a 10-h eating window (e.g. 9 am to 7 pm) to illustrate a restricted eating window	18/19 = 95%*
Presentation of the background of studying TRE in terms of the potential effects	The HCP provides a two-page information sheet: the front page displays a simple overview of potential effects with icons and a few headlines, while the reverse side offers a more detailed explanation, including potential mechanisms associated with TRE. With emphasis on the potential effects of TRE, the HCP first explains the simple version, and the participant then decides whether he/she would like the HCP to explain the detailed version as well. Ensuring the level of information matches each participant’s interests and needs is essential. The participant is given the sheet to take home	19/19 = 100%
Identification of the 10-h eating	The purpose of this task is to identify the most appropriate 10-h eating window for each participant. The dialogue focuses on the following topics: ✓ Exploration of the participant’s past experiences with habit changes Example of questions: – Have you had previous experiences with changing habits, particularly eating habits? – What exactly did you do? And how did it work? ✓ Exploration of the participant’s specific motivation for participating in the project Example of questions: – What let you to participate in the RESET2 pilot project? – What do you hope to get out of participating? ✓ The HCP and the participant explore the participant’s daily life using the ‘My day’ dialogue tool (Fig. 3), visualising a 24-h schedule where activities such as eating, drinking, work, sleep, medication, social activities, and exercise are illustrated to explore current habitual living and facilitate optimal scheduling of the eating window. Key notes are written on the tool (either by the participant or HCP in collaboration), the participant keeps the tool in his/her folder, and a copy is kept by the HCP Eventually, the participant, in collaboration with the HCP, identifies his/her new eating window Example of questions: – Can you describe a typical day for you using the ‘My day’ tool? – If you now look at your ‘My day’, how will a 10-h eating window fit your everyday living best? And why?	19/19 = 100% 19/19 = 100% 19/19 = 100%
Exploration of the participant’s possible worries and challenges	Throughout the conversation, the HCP provides opportunities to discuss any possible concerns or challenges the participant might have related to initiating and maintaining a restricted eating window: Example of questions: – Do you have any questions at the outset? (allows space to address any initial concerns) – What are your first thoughts on following a restricted eating window? – When might it be challenging to follow a restricted eating window? And how can these challenges be addressed? – Is there anything that concerns you about eating and drinking within a restricted eating window? Furthermore, the participant is instructed to measure their fasting blood sugar regularly and return to HCP in case of low blood sugar readings	19/19 = 100%
Exploration of possible involvement of relatives	During the initial telephone contact, the potential participant is encouraged to bring a relative to study visits. If a participant wishes to involve a relative in the project, the HCPs provide a letter for the participant to share with their relative Throughout the study, the HCPs continue to encourage the participant to include his/her relatives by inviting them to attend study visits with a special focus on participants who are not bringing relatives to visits nor feeling supported at home. This topic was also addressed at the conversations, referring to what has been discussed in advance	19/19 = 100%
Summary of the conversation	At the end of the conversation, the HCP and participant collaboratively complete a summary sheet covering key topics: motivation for participation, anticipated effects, past experiences with lifestyle changes, eating window schedule, daily routines, potential challenges, benefits, and concerns related to the eating window The participant chooses whether he/she or the HCP fill out the summary sheet The participant receives a copy of their completed summary sheet, which serves as a reference point for follow-up calls at weeks 2 and 5	19/19 = 100%

HCP Healthcare professional, TRE Time-restricted eating

^*One participant did not accept audio-recordings of conversations and interviews, and notes were written by the project workers during interviews and conversations; also, completed forms/sheets were available for this participant. However, for some of the tasks/topics, it was not possible to determine whether the task/topic was addressed as intended in the documentation available for this participant

Table 5.

Core component 3: individual support and adjustments to maintaining TRE

Task/topic	Procedure (protocol)	Degree to which the task/topic was addressed as intended in the conversation
Presentation of the purpose and agenda of the conversation	The HCP provides a schematic overview of visit 2 featuring icons representing measurements and a timeframe for the visit	18/19 = 95%a
Exploration of participant’s experiences of the first 8 weeks	Before visit 2, the HCP reviews the summary sheet from the conversation at visit 1, including notes added from the follow-up calls at weeks 2 and 5 The HCP facilitates a dialogue on the participant’s experiences with TRE over the last 8 weeks Examples of questions: – So, would you like to describe, how the first 8 weeks have been like for you? (And additional specific questions tailored to the participant’s summary sheet)	19/19 = 100%
Exploration of possible challenges	The HCP presents an illustration named ‘When is it difficult to follow TRE?’, which shows a person in the centre surrounded by icons representing situations from life, where it may be difficult to follow TRE, e.g. work, leisure time, fitting meals and drinking to everyday life, sport, transport, holidays, celebrations, medication intake, worries and stress, social activities. Furthermore, the illustration supports an in-depth dialogue and may encourage the participant to recall additional situations/experiences with TRE, to be addressed	19/19 = 100%
Exploration of how participants perceive being supported	The HCP asks if and how the participant receives support from relatives during the TRE intervention with focus on how the participant perceives being supported or not	19/19 = 100%
	HCP presents a sheet with three sentences for the participant to complete: – For me good support has been/is … – I have been lacking support when … – Less good support has been when …	6/18 = 33%
Presentation of individual adjustments and support options	The HCP presents an illustration outlining individual adjustment and support options available to choose among for the remaining 4 weeks. Next, the HCP explains the separate options in more detail. Each option is presented with easy-to-understand figures: ✓ Overview of four types of adjustments in terms of timing of the eating window ✓ Calorie and caffeine-free beverages ✓ Peer support groups ✓ Further involvement of relatives ✓ Extra telephone calls from HCPs Based on the participant’s experiences with TRE, and explorations of their challenges, the HCP and participant discuss and collaboratively identify which adjustment and support options that are most relevant to the individual participant, which were then selected for implementation in the final 4 weeks of the intervention.b	19/19 = 100% 19/19 = 100% 19/19 = 100% 19/19 = 100% 19/19 = 100% 19/19 = 100%
Summary of the conversation	At the end of the conversation, the HCP and participant collaboratively complete a summary sheet detailing the chosen adjustment and support options, as well as the participant’s expectations for incorporating these adjustments into their everyday life in coming 4 weeks	19/19 = 100%

HCP Healthcare professional, TRE Time-restricted eating

^aOne participant did not accept audio-recordings of conversations and interviews, and notes were written by the project workers during interviews and conversations; also, completed forms/sheets were available for this participant. However, for some of the tasks/topics, it was not possible to determine whether the task/topic was addressed as intended in the documentation available for this participant

^bThe adjustment and support options are thoroughly described previously along with information about who choose the specific options, and how the options helped in maintaining TRE [25]

Intervention acceptability among participants

A student assistant carefully listened to recordings of the telephone interviews conducted after visits 1 and 2, summarising participants’ responses into the pre-defined topics outlined in the data collection section. Additionally, the assistant transcribed the visit 3 interviews. A deductive content analysis [29–31] was performed across all three interviews using the pre-defined topics as framework. This examined participants’ reception and experiences of the intervention activities, focusing on visit structure, information provided, measurements and questionnaires, the HCP’s approach, telephone calls, and the two conversations. Following categorisation, the data were synthesised to present a comprehensive overview of the acceptability of the TRE intervention.

Participants’ experiences on implementing TRE in everyday life

Visit 3 interviews were transcribed verbatim by a student assistant. A deductive content analysis [29–31] was conducted using the pre-defined topics that were also guiding the interview. The qualitative analysis framework was further developed in iterations to generate the final themes by two researchers in collaboration. ADT did the initial coding and synthesis, and further analyses was done by ADT and AV in collaboration.

Refinement of the IPT

To refine or adjust the causal processes suggested in the IPT, data analysis explored how the intervention activities supported participants differently and how various contextual factors influenced participants’ experiences and adherence to TRE. Using data from multiple sources, including recordings of interviews and conversations with participants, and data on adherence to eating window understandings of what works, for whom, how, why, and under which circumstances are suggested, providing a nuanced view of intervention outcomes.

Results

Participant characteristics

We included 10 women and 10 men with type 2 diabetes. Marital status was distributed as follows: married (n = 10), in a relationship (n = 3), widowed (n = 2), single (n = 1), and divorced (n = 3). Ten participants lived with a partner, 2 lived with a partner and children, 2 lived with children, and 5 lived alone. Highest educational level attained was as follows: basic school (n = 1); vocational education and training (n = 3); short education (n = 1); intermediate education (n = 12); and higher education (n = 2). On average (SD), the participants were 66 [8] years old, had a BMI of 36.7 (7.8) kg/m², HbA_1c of 62 [9] mmol/mol, and a diabetes duration of 15 [6] years. Median (Q1, Q3) habitual eating window was 14.5 (13.8, 16.0) h [24]. More information can be found elsewhere [24].

Recruitment, retention, and adherence

Following feasibility results are being described in more details elsewhere [24]:

Recruitment

Recruitment took place between April and August 2022. Fifty-two participants responded to study advertisements from clinicians, from which 26 were pre-screened over telephone [24]. Twenty-two participants attended in-person screenings at study sites; two were deemed ineligible due to low HbA_1c levels, resulting in the inclusion of 20 participants.

We recruited four participants from Hvidovre Hospital, and the remaining 16 from SDCC. Difficulties in procedures related to booking and collecting blood samples and reserving rooms for testing and conversations at Hvidovre Hospital led to confusion and additional workload for HCPs.

Retention

Nineteen out of the 20 participants completed the 12-week intervention resulting in a retention rate of 95%. One participant cancelled the baseline visit and did not want to continue in the study [24].

Adherence

Participants reported a median eating window of 10.0 h (9.8, 10.0), and a median reported adherence to the eating window across the 12 weeks of 94% (86, 98) [24].

Delivery and implementation of intervention components

Findings related to the delivery and implementation of the intervention components are presented in Tables 2–5. In summary, the overall degree of delivering intervention activities in terms of conduction of visits, measurements/tests, questionnaires, conversations, and phone calls as scheduled was very high with only one missed telephone call and two failed HbA_1c measurements due to laboratory errors (Table 2). A few appointments were rescheduled due to sickness. The intervention activities were delivered by the same two HCPs across both settings.

Core component 1

As reported in Table 3, the two phases of strict and individualised TRE were well implemented with average durations of 7.9 (0.3) weeks, and 4.0 (0.5) weeks, respectively.

Core component 2

Table 4 outlines the key tasks and topics that were to be addressed during conversations between individual participants (and a relative, if accompanied) and the HCP at visit 1. It includes a brief description of what each task/topic entailed and how it was intended to be conducted, alongside an evaluation of the extent to which these tasks/topics were addressed in practice. All major tasks were conducted as intended with all participants (Table 4).

Core component 3

Table 5 presents the major tasks and topics that were to be addressed during conversations between participants (and a relative, if accompanied) and the HCP at visit 2. It includes a short description of what and how the tasks and topics were intended to be conducted together with the extent to which they were in practice. All topics (except one) were addressed as intended (Table 5). However, the use of the tool to exploring details about how participants perceived support was lower (33%). The tool presents three sentences for the participant to complete (details in Table 5). However, the HCPs found these questions to be too counterintuitive and were only using it with the first six participants. Consequently, the remaining participants were asked about how they needed or were receiving support from relatives in a more open-ended manner.

Intervention acceptability

Based on interviews at three timepoints (baseline, 8 weeks, and 12 weeks), participants reported a very high acceptability of the TRE intervention.

Visit structure

Overall, participants found the three visit days well organised, with minimal waiting time and time effectively spent during each visit. Many also expressed that the project was conducted professionally.

It is professional, everything was well-prepared, and everything is running as planned. No waiting time. This is how it should be. (Man, 61 years, interview after visit 1)

Information level

All participants reported feeling well informed throughout the study, noting no lack of information. They found the written materials clear, concrete, and easy to understand. Upon inclusion at the screening visit, participants received the written materials (overview of study visits, personal calendar, checklists, etc.) in a folder, which was highly appreciated as a helpful reminder and for preparation for future visits. During visits, participants felt that instructions and explanations for measurements and conversations were thorough. After completing visits 1 and 2, they felt well prepared for the next stages of the study.

I found the [information material] to be comprehensive and it made me want to participate (…) I did not think anything was missing. (Woman, 74 years, interview after visit 1)

Measurements/tests and questionnaires

Participants reported no issues related to measuring waist/hip circumference, body weight/bioimpedance, blood pressure, or blood sample.

Regarding the food preference task, some participants noted that it felt a bit long-lasting. Some found the task enjoyable and fun, and a few considered it challenging, particularly when faced with choosing between two disliked food options and being unable to go back if they made an unintended choice. Some highlighted that it was easy to discern the ‘right’ option, thereby favouring healthier over less healthy choices.

The food preference test is perceived as long (…) There were many of the same questions, which feels a bit annoying when answering the same questions multiple times. (Woman, 58 years, interview after visit 1)

Participants appreciated the option to complete the questionnaires at home before visits 1 and 3, though some expressed no preference for completing them at home or during the visit. All participants noted that there were too many similar questionnaires with repetitive questions, which led to irritation for some. Participants generally appreciated the option to pause the questionnaires and return to them when they were ready to continue.

I had the time and energy to do it from home, so I did. I could just as well have done it on-site. Having the option is a positive thing. (Woman, 59 years, interview after visit 1)

It’s really nice because I consider myself a slow reader. It would irritate me to have to complete too many questionnaires here, as I would feel like I’m wasting others’ time because I can’t read as fast. (Woman, 69 years, interview at visit 1)

The approach of HCPs

Participants consistently described their interactions with HCPs as positive and supportive. Well-prepared and friendly HCPs, who respected and took participants seriously, contributed to a comfortable, safe, open, and informal atmosphere during visits. Participants appreciated having adequate time for questions and found HCPs’ communication clear and approachable. Overall, participants felt that interactions with HCPs were productive and meaningful, enhancing their commitment to the study.

Very positive [about the conversation with HCPs]. She was very pleasant to talk to. I felt that they [HCPs] were well-prepared. (Woman, 69 years, interview after visit 1)

Positive, calm, relaxed, and a good atmosphere. A very pleasant person, which means a lot. (Man, 76 years, interview after visit 1)

Telephone calls

Participants found the phone calls with HCPs valuable, as they offered a space to share concerns and discuss TRE experiences. For some participants, the phone calls served as reminders and encouraged to maintain commitment. Although a few participants felt that the calls were not essential, many still saw them as a valuable source of support, especially if they continued TRE beyond the 12 weeks. Some participants preferred reaching out by phone when they had questions or needed clarification.

I have felt good about the fact that you [HCPs] have called… it makes you reflect on it more. (Woman, 63 years, conversation at visit 2)

I could do without those [phone calls]. I didn’t feel that I needed you to call me. I could have called myself. (Man, 76 years, conversation at visit 2)

Conversations

Visit 1

Conversation: Participants found conversations about planning of their eating window with HCPs supportive and constructive. They felt confident and prepared to initiate TRE in their daily lives. Most participants viewed the conversation as a meaningful use of time.

We created a good schedule for it, so it went fine. I feel that I received the information I needed to complete it. (Woman, 69 years, interview after visit 1)

That was probably the only concern I had before the project. I had calculated the ten hours and could see that it might be a problem. I had quite a bit of restlessness in my head before I arrived … [HCPs] were good at grounding me and making me realise that it wasn’t too bad not having my usual breakfast. I received support, no doubt about it. (Man, 62 years, interview after visit 1)

‘My day’ tool: All participants appreciated the opportunity to discuss daily routines and habits using the ‘My day’ tool [32], which helped them choose an eating window that was a practical and personal fit for their everyday lives (Fig. 3). Some felt that talking through individual timing was crucial for being able to choose their eating window. Others had pre-selected their eating window before the visit, yet they expressed that the conversation helped solidify their choice, and in a few cases, they adjusted the eating window consequently. Some participants preferred to fill in the tool themselves, as they felt that writing helped strengthen their memory. Others valued the option of having HCPs complete it, allowing them to concentrate more fully on the conversation.

It [‘My day’ tool] helped uncover some things about my habits. It was essential. (Man, 61 years, interview after visit 1)

Fig. 3.

The healthcare professional and the participant explore the participant’s daily life using the ‘My day’ tool visualising a 24-h schedule where activities such as meals, beverages, work, sleep, medication, social activities, and exercise/activities are illustrated to explore current habitual living and facilitate optimal scheduling of the eating window [32]

‘Potential effects’ tool: Participants found the overview including both a simple and a detailed version, helpful and relevant. While some used the detailed version sparingly, they appreciated having both options for a more comprehensive understanding of the potential effects of TRE, and what they might expect during participation. Several participants found the information useful for discussion with partners and relatives at home.

It [the overview of potential effects] was good; I think a lot was explained. It’s nice to take it home (…) It’s hanging on my fridge so I can see it (…) It’s good for talking with relatives about. (Woman, 63 years, interview after visit 1)

Visit 2: 

Conversation: Participants appreciated the opportunity to openly discuss experiences and challenges with the eating window, facilitated by the ‘When is it difficult to follow TRE’ tool. Overall, they found the adjustment and support options relevant, with HCPs presenting them clearly through helpful overviews and illustrations. The individualised adjustments boosted participants’ confidence in managing the eating window, particularly during upcoming vacations. They felt they could still adhere to TRE, even if they took a day off or adjusted the timing, as these changes were prearranged, preventing them from feeling as they were ‘cheating’. Several participants also mentioned that the conversation helped renew or strengthen their motivation to continue with the intervention.

There were solutions for potential challenges [in the conversation]. (Man, 62 years, interview after visit 2)

…. an equal conversation (…) We mostly talked about adjustments in relation to my vacation. I had been wondering how I could fit it in, and the conversation gave me a sense of calm and peace of mind. It gave me a better opportunity to do well. (Woman, 76 years, interview after visit 2)

The [conversation] worked well (…) I wished for more phone calls (…) I feel heard (…) I still really enjoy it [TRE] and would like to continue (…) I found it helpful to put things into words. (Woman, 58 years, interview after visit 2)

Summary sheets: At the end of each conversation, HCPs and participants collaboratively completed summary sheets. Participants found this tool valuable for retaining information and summarising key points. Some preferred filling it out themselves, noting that writing helped reinforce their memory, while others appreciated the option of having HCPs completing it, allowing them to focus on the conversation. Many valued the ability to refer to the sheet later, especially for discussing benefits, concerns, and challenges with relatives.

When you see it summarised, you remember it (…) It’s really important. (Man, 69 years, interview after visit 1)

It worked well; she convinced me to write it down myself (…) Something happens when you write it yourself, you remember it better. (Man, 62 years, interview after visit 1)

Analysis of participants’ experiences with implementing TRE in their everyday life

In summary, participants had very different experiences with following TRE. Some were very enthusiastic and had positive experiences during the 12-week period and were considering or had decided to keep on with TRE in the future. Others were more neutral or reluctant and were looking forward to having their ‘normal’ life back even though they were not regretting having participated. In between, some were more ambivalent and were seeing positive as well as less advantageous aspects of following TRE and expressed that they might wanted to follow TRE with some modifications going forward.

Participants’ experiences of and ease with TRE implementation were influenced by how following TRE had (1) entailed difficulties/benefits related to changing (daily) routines and habits, (2) affected social life, (3) involved support from partners and relatives, (4) felt appealing and promoted physical and mental changes, and (5) included intervention activities that felt supportive. In the following, the analysis of data related to these themes is presented.

Changing everyday routines and habits entailed more planning, difficulties, and benefits

Initiating and maintaining TRE in everyday life required different efforts in terms of type and degree of changed behaviours and were not easy for all. However, it also provoked changes that participants appreciated. Some experienced that initiating TRE was not that hard, and for those, it seemed that everyday routines were not affected so much. For others, more substantial efforts were required and the changes disrupted enjoyable habits.

I still miss my evening snacks. Tea and cake. It can also be fruit. I can survive without it, but it’s difficult. And when he [my husband] takes out his tea and cookies, I push it to the other side of the table because I don’t want it right in front of me – it’s too tempting. [After being presented with the option of drinking a cup of tea outside the eating window at visit 2] (…) That would actually make it worse. Then I would miss the cake even more. (Woman, 69 years, interview at visit 2)

Breakfast imposed a significant challenge for participants who were leaving home before the eating window opened and/or felt hunger in the morning. They were either skipping breakfast or postponing it to the eating window opened, which for those who went out meant ‘food on the go’. That was also the case for participants who went home late in the evening due to hobbies, duties, etc. Food on the go implied changes to more convenient, and for some also denser, more filling options, for example, rye bread was preferred for breakfast or dinner. These changes required more planning than before and were harder to manage in the beginning; however, it also imposed a feeling of loss. For example, porridge for breakfast was missed, but the thing that most participants missed in the morning was coffee. Several participants expressed that if they could just have their morning coffee then it would be easy/easier to postpone or do without breakfast.

[It has] been hardest in the morning because I was only ‘allowed’ to drink water. If I could drink my coffee and use it as a substitute instead of food, then I could easily wait until 10 am to eat. (Woman, 59 years, interview at visit 3)

Participants who were used to eat dinner later in the evening had to eat earlier, while those who typically had dinner earlier adjusted to eating closer to the end of the eating window to avoid feeling hungry later in the evening. In general, these changes did not impose challenges for most participants. Also, participants generally reported that their overall dietary habits remained unchanged, with no significant changes in the types of food they consumed. However, some made changes to denser, more filling food to eat close to the end of the eating window.

It’s (...) when I got better at planning and bringing something more sensible, something a bit denser to eat when I was heading out [for an activity with my child], so I could eat before 6 pm. Because sometimes when I forgot, I just grabbed something on the way, which might not be as filling as if I had brought two slices of rye bread or something from home. (Woman, 54 years, interview at visit 3)

On busy days, some shifted what they normally ate for lunch and dinner, as lunch was easier to prepare before the closing of the eating window, and some implemented this as a new routine.

Many participants were drinking more water during TRE, either they replaced other beverages like soda, tea, or coffee, or increased water intake outside the eating window. In contrast, intake of alcohol was decreased for some due to changes in their evening routines. Also, some reduced their evening intake of sweets and snacks because their eating window closed before their usual snacking time, or they felt less desire for sweets. However, few participants increased their intake of cakes or candy during the eating window as compared to what they used to eat during the day.

Adhering to TRE and having shift work were challenging, and more time was needed to identify the best eating window.

At first, we talked about having it from 7 am to 5 pm, (…), but then I moved it from 8 am to 6 pm, and on the days I had evening shifts, it was extended to 8 pm (…) It just became too messy (…) And then I decided to have it from 10 am to 8 pm (…) Yeah, how can you even make it work? (...) But it’s definitely a challenge for those who have rotating shifts, I have no doubt about that. (Woman, 59 years, interview at visit 3)

Social life was challenged

Retaining a normal social life was challenging while conducting TRE. Some compromised the eating window from time to time, especially when being out with friends and during weekends. Some expressed that it was easier when the dinner was held at home (their own) because they were more in charge of time.

We have a very rich social life with friends and family, and when we’ve been out, I haven’t been able to stick to the eating window, because when you sit down for dinner at half past seven, there’s really nothing you can do about it. (…) And we go out to dinner with friends about once a week. When we visit friends, I don’t want to impose any restrictions on them. (Man, 74 years, interview at visit 3)

Social togetherness with family members in everyday life was also impacted as, for example, eating breakfast with the rest of the family was often not possible if they were up early in the morning.

Of course, there’s a problem on weekends because we have breakfast together on Saturdays and Sundays. And that clashes with the fact that my daughter only sleeps until half-past seven and then she’s hungry. So, if I want to sit and eat with her, that’s when it has to happen. (Man, 55 years, interview at visit 3)

Having tea, coffee, and snacks together in the evening was also hindered and participants found this to be a major drawback if TRE should be conducted in the long term.

The social aspect of eating together - now, I can only speak for myself - but I can imagine that this is something many people might not want to do because they can’t eat with their spouse [tea in the evening] or their children [breakfast], or whatever it may be (…) The cost of it would simply be too high to bear. (Man, 55 years, interview at visit 3)

Joining parties or events were highly affected as these often occurred during evenings and participants either stopped eating or expanded the eating window that day. In both cases, it felt uncomfortable and disturbing.

Finally, travelling and vacations posed challenges both regarding schedules and changed routines due to holidays. However, some experienced that it was easier than expected to maintain TRE during a vacation.

We were going on vacation shortly after we started. It was a bit exciting because how could I stick to it? But it turned out to be just as easy [as in the everyday life]. (Man, 69 years, interview at visit 3)

Involving partners, family, and relatives—more or less

For participants who lived with a partner, the partner was generally also eating dinner in the eating window. The wives of participating men adapted their cooking or mealtime (if the wife did the cooking which most often was the case) to the new schedule. This was highly appreciated by the participating men, but at the same time, it was seen almost as a natural and necessary foundation for their participation. For participating women who had a husband, the women cooked themselves and their husbands adapted to the new schedule. These women were not feeling supported in other aspects and were seeing participation in TRE as their own project and were reluctant to discuss and involve their husbands more. However, most participating women were not living with a partner. Instead, some involved friends, partners, or other family members who supported and adapted to the eating window when eating together. This was of great importance although such support was often not daily. However, a participant’s sister followed TRE regularly and she also joined the intervention visits, which was highly appreciated by the participant.

Concerning involving friends and family members such as grown-up children and grandchildren in their participation in TRE, most chose to be open about it without imposing requests for adaptations on them. However, a few participants preferred not to tell others about their participation. Some were seeing their children and grandchildren regularly and they experienced support, acceptance, and respect although it was difficult from time to time to fit the eating window into the family plans. One participant explained how she was advised to take care of herself for example by ‘serving herself in the fridge’ during family events.

They [the children] know about it, but they haven’t really reacted to it. The only thing was when we were at my daughter’s place for their birthday, she said: “Mom, if you need to eat something before the [eating window] closes, just go to the fridge and find something yourself” because she didn’t want to get involved in it. (Woman, 69 years, interview at visit 3)

Some participants mentioned that TRE was easier to implement if you were not involved with many people in daily life. However, involving others also felt supportive, especially when those around viewed their participation as cool and expressed interest in the process and its effects. This sense of encouragement and understanding contributed positively to their experience.

They think I’m cool (…) And my friend immediately said when I told her that my blood sugar had dropped, “Maybe this is how you should eat for the rest of your life”. (Woman, 58 years, interview at visit 3)

They are really good friends, so they fully understand (…) They are even quite excited to hear how the project has gone. (Man, 58 years, interview at visit 3)

Personal attitude and physical and mental experiences with TRE

Participants had different expectations to TRE, ranging from curious and excited to trying a new approach to cautious and hesitant. However, most were hoping for weight loss and better blood sugar control but also simply wanted to contribute to the project.

Many participants expected to feel hunger during evenings, nights, and mornings. Some reported overeating up to the closure of the eating window. However, for most, this subsided after a few weeks due to discovering not suffering from hunger.

Yes, something I have discovered is that I’m probably not as hungry as I thought. I used to believe I was hungry in the morning and that I absolutely needed my breakfast, but I’ve realised that I’m not actually that hungry - I can skip it. In the beginning, I remember that I tended to overeat, right? (Man, 62 years, interview at visit 3)

As I told you, during the first few days, I would binge eat [haha], just at the thought of having to wait a long time until I could eat again. (Woman, 54 years, interview at visit 3)

While most participants felt less hunger and felt motivated by it, a few felt hunger throughout the entire study period. Where some were accepting feeling hunger and associated it with other beneficial effects, others struggled more and looked forward to having their normal life back.

So, what I’ve learned is that I can skip eating in the evening if it’s really necessary, but it’s not the optimal solution for me because then I go to bed hungry. And that’s okay for this period, since I chose to do it, but it’s not something I could continue doing for years. I wouldn’t be able to keep going to bed hungry every night for years (…) So in that sense, I know that when this project is over, I’ll go back to eating in the evening again. (Woman, 69 years, interview at visit 3)

Experiencing the beneficial effects of following TRE was important for participants’ continued motivation. Several observed decreased blood glucose levels and some felt and measured weight loss. Others reported that health benefits improved sleep, joints, digestion, and mood as well as higher energy levels.

Some participants felt calmer during the evenings as they no longer focused on what to snack and some even felt less desire for sweets and snacks during the study period.

I thought my sweet tooth would stand out more than it actually has (…) My evenings have been calmer because I didn’t constantly have to think about what to eat next or what I was craving, because I simply didn’t have to. (Woman, 74 years, interview after visit 3)

Few participants who suffered from severe physical or emotional challenges due to other conditions or difficult life circumstances were struggling more than other participants in sustaining their eating windows in everyday life as they were used to eating to compensate.

Supporting intervention principles, components, and activities

Overall, it was highly valued by participants that there were no restrictions in terms of type and amounts of food in the eating window, which also made the approach attractive to them initially. However, following TRE still encompassed difficulties which intervention activities and support were attempting to remedy. How the different options supported the participants are being published elsewhere [24].

Refined program theory

A refined program theory relating intervention activities and contextual factors to adherence-related outcomes via mechanisms of change was generated based on the findings (Table 6). Overall, in terms of adherence to TRE throughout the intervention, participants can roughly be classified by three distinct groups (Table 6). The first group was highly self-motivated and found TRE easy to implement and sustain across the 12-week intervention. These were mainly males who were married to partners who also followed TRE making them feel supported and less drawn to food temptations outside the eating window. The second group encountered some challenges with TRE during the intervention, and these were people with a rich social life, women with non-supporting partners, and people with smaller children. However, support and individual adjustments assisted them in adhering to TRE on most days. The third group consistently struggled to adhere to TRE despite available support and adjustments, which they appreciated but found only partly helpful. Stress, depression, co-morbidities, and other personal challenges made adherence particularly difficult.

Table 6.

Refined program theory of the RESET2 pilot study

If…	Then…	Because…
Participants get individual conversations, and the HCPs are competent in using a whole-person, nonjudgmental approach encompassing previous experiences and everyday life	Initiating TRE and choosing relevant individual support and adjustments are eased	Participants feel prepared and confident	Relates to IPT 1, 2, 3, 4, 6
Participants are motivated and have partners who also followed TRE	Implementing and sustaining TRE are eased	They feel supported and less drawn to food temptations outside the eating window	Relates to IPT 9, 10
Participants are following TRE and experience health benefits such as lowered fasting blood glucose and weight loss	Sustaining TRE is eased	They feel TRE makes sense and intrinsic motivation is evoked or reinforced	Relate to IPT 8
Participants have a rich social/family life and individual adjustment options such as having energy-free beverages outside the eating window and occasionally having days off from TRE are available	Sustaining TRE is eased	They feel part of a social setting without feelings of guilt and failure	Relates to IPT 11
Participants lack social support in their everyday life and support such as telephone calls with HCPs and participation in peer support group meeting are available	Sustaining TRE is eased	It made them feel cared for, understood, and supported	Relates to IPT 2, 4, 6, 7
Participants experience stress, depression, co-morbidities, or other personal challenges	Sustaining TRE is challenging	They use food as a coping mechanism during difficult times and/or planning is disturbed and unmanageable

HCP, healthcare professional; IPT, initial program theory; TRE, time-restricted eating

Adjustments in the RCT design based on findings from the pilot study

Based on the findings from this theory-based process evaluation and participants’ experiences with individual support and adjustment options [24], we have gained new insights that have shaped the design of the RCT. Below are the five most prominent adjustments:

First, we recognise the importance of participants consuming energy-free beverages outside the eating window. In the pilot study, all participants opted to include energy-free beverages after the strict TRE period, emphasising the relevance of this adjustment. Morning coffee and the ability to socialise with family and colleagues were highly valued by most participants. In the RCT, participants will have the option to drink coffee and tea without milk, cream, or sugar, as well as other energy-free beverages before the eating window opens (after waking up) and energy-free, caffeine-free beverages after the window closes (before bedtime). In the pilot study, we completely excluded caffeine-containing beverages outside the eating window as this may impact metabolism. However, offering the participants the option to have their morning coffee with caffeine is likely more beneficial and important for adherence.

Second, during the pilot study, participants had the option to occasionally reschedule the 10-h eating window to conclude after 8 pm after the strict TRE period. However, their reasons for choosing this option were similar to another option of taking days off from TRE [24]. Consequently, this option will not be included in the RCT.

Third, in this study, HCPs found the structured questions at visit 2 regarding participants’ perceived support from relative counterintuitive. Instead, they preferred a more flexible approach to discussing support from relatives. For the RCT, we will continue to allow HCPs to address this topic in a way that aligns with participants’ experiences.

Fourth, participants reported varying levels of support from relatives, ranging from highly supportive partners adhering to the eating window to limited support from others. After the strict TRE phase, we introduced a tool called ‘Our joint plan’, designed to facilitate problem-solving with relatives. Despite limited support from some partners and relatives, participants chose not to use the tool during the pilot study. In the RCT, this tool will be introduced from the start for participants randomised to the TRE group. It will be included in the participant folder, and HCPs will remind participants of this resource during telephone calls if they express a lack of support.

Finally, participants consistently found the questionnaires too repetitive and time-consuming. In response, we critically revised the questionnaire set, reducing its length by half. We then asked four participants from the pilot study to test the revised version to ensure it addressed concerns about repetition and time burden. Their feedback confirmed the improved feasibility and acceptability of the updated questionnaires.

Discussion

This theory-based process evaluation of the RESET2 non-randomised single-arm pilot study demonstrates that the TRE intervention is feasible and acceptable for individuals with type 2 diabetes. Intervention activities, including visits, measurements, conversations, and phone calls, were delivered as intended. The findings highlight the feasibility of recruiting and retaining participants, the high adherence for most of the participants to the intervention, and the positive reception of the intervention components.

Perspectives of the findings

In a previous systematic scoping review, we examined the feasibility of TRE in individuals with overweight, obesity, prediabetes, or type 2 diabetes [21]. Among the 28 included studies, we found that recruitment was feasible, retention rates were high (~ 92% on average), and the degree of adherence to the eating window was similarly high (~ 86%). However, we observed significant variability in adherence levels, suggesting that TRE can be challenging for some participants—consistent with the findings of the present study.

The review also underscored the importance of allowing individual flexibility in defining the eating window, including energy-free beverages outside the window, and receiving support from family, friends, or HCPs. Additionally, participants reported difficulties aligning the eating window with social activities and adapting to changes in routines and habits, reflecting the broader challenges of sustaining TRE in daily life as also found in the present study.

We understand that different behavioural interventions appeal to different people. Even among those who agree to participate in a TRE intervention, not everyone can adhere to it, and some may find it challenging. This is largely due to varying competencies, motivation, experiences, and environmental circumstances, which influence individuals’ ability to follow an intervention. These life conditions play a key role in terms of the possibility for a mechanism of change to be triggered by the intervention resources. While options for adjustments and support may help some, they may not be effective for others. Through the RESET2 needs assessment study [23], we developed an intervention designed to be inclusive and broadly applicable, aiming to reach as many individuals as possible, though it may not meet the needs of all.

The IPT proposed several causal associations expected to influence the initiation of TRE, intervention retention rates, and adherence to TRE (as shown in Table 1). These causal processes were supported by the findings, indicating that our thorough work in the needs assessment study was both relevant and valuable for the intervention and the participants in the pilot study. Participants demonstrated very diverse needs, emphasising that support activities must include a variety of options delivered with flexibility.

These findings underscore the need for further refinement and elaboration of the intervention to better address these difficulties in the upcoming RCT. However, several of the contextual factors influencing adherence were not related to nor supported by the intervention activities. We found that the life conditions of some participants caused difficulties in adhering to TRE. These challenges were related to social activities in the evenings, having long-term physical and mental challenges, managing breakfast routines with children, and early wake-up times (often due to early work schedules) and therefore missing their morning coffee until opening of the eating window. This shows the importance of also exploring contextual factors of importance of adherence beyond the intervention to get a more comprehensive understanding of what influences adherence. Such insights can be used either to realise for whom the intervention might not be suitable or to adjust it to embrace and support the challenges the participants face. The influence of contextual factors and the various types of support will be further explored in the RCT by interviewing participants after the intervention. Regarding the contextual influence of the intervention setting, the findings emphasise the importance of organisational support and clear communication to enable effective planning and coordination of intervention activities.

Strengths and limitations

This study demonstrates several strengths. First, our theory-based process evaluation is grounded in a robust pilot study design informed by a comprehensive needs assessment study [23]. A key strength of process evaluations is that they often include talking to people involved in the intervention (e.g. participants and HCPs) providing valuable qualitative insights that enrich the understanding of both the intervention activities and implementation, and its outcomes [18, 22, 33]. This approach enhanced the probability that the intervention was tailored to address the specific needs and preferences of target group. Secondly, another key strength of this evaluation lies in the substantial amount of qualitative data collected through interviews and conversations. All recordings were carefully transcribed, coded, and analysed in depth, allowing for a nuanced understanding of the participants’ perspectives. This rigorous methodology has generated significant knowledge about participants’ experiences with TRE and their acceptability of the intervention. Furthermore, how the intervention activities were implemented was also analysed, and by integrating all these insights, the design of the RCT is founded on solid and participant-centred data, enhancing the relevance and potential impact of the RCT. Process evaluations contribute to the transparency and reproducibility of research, enhancing its overall generalisability and applicability in a clinical practice [18, 34]. Lastly, by combining qualitative and quantitative data, and suggesting causal association and contextual impact, we have gained valuable detailed insights into what works for whom, how, why, and under which circumstances, providing a nuanced understanding of the TRE intervention [23, 24]. This knowledge has been incorporated into the design and implementation of the RESET2 RCT [35], and will be further explored to develop a more tailored and relevant intervention to the target population.

However, the study has its limitations. First, the limited number of participants in our pilot study may restrict the breath and generalisability of the findings. Recruiting participants from two hospital sites with distinct patient populations enhanced response diversity, potentially strengthening the study’s generalisability. However, we did not account for other specific contextual factors in the recruitment process. In this study, the male participants all had very supportive partners, whereas the female participants either lived alone or had less supportive partners. As a result, we lack perspectives from, for example, men living alone and women with supportive partners. In the ongoing RESET2 RCT, we aim to recruit 160 participants and expect to capture a more diverse range of contextual factors. Secondly, the qualitative nature of much of the data can introduce subjectivity, as the interpretation of participants’ experiences and feedback may be influenced by researcher biases or perspectives. Thirdly, limited generalisability may exist due to the specific population included in the pilot study—individuals with complicated type 2 diabetes treated at the outpatient clinics at SDCC and Hvidovre Hospital—making it uncertain whether the findings apply to other populations or settings. However, many of the findings have been reported in other studies examining TRE experiences [12, 36–45]. Fourthly, all intervention activities were delivered by the same two HCPs across both hospital setting. As a result, this study does not provide insights into whether these activities would be feasible for other HCPs to conduct in accordance with a whole-person, non-judgmental approach. Additionally, it remains unclear what level of training or education would be necessary for less experienced personnel to effectively deliver the intervention. Finally, adherence data were self-reported based on participant recall, which is known to be influenced by memory biases and inaccuracies. However, these self-reports were verified through end-of-intervention interviews, allowing us to cross-check the data and increase confidence in the reported adherence.

Conclusions

In this theory-based process evaluation of the RESET2 pilot study, focusing on factors influencing adherence to TRE, we found that intervention activities, including visits, measurements, conversations, and phone calls, were delivered as intended. The two-phase design, starting with a strict TRE phase followed by an individually adjusted phase, was effectively implemented. Key conversation tasks at baseline and after 8 weeks were carried out as intended, with minor adjustments to the protocols. Interviews at baseline, 8 weeks, and 12 weeks showed that participants found the TRE intervention highly acceptable, appreciating the relevant information, the support from HCPs, and the constructive conversations that helped them initiate and sustain TRE. Participants’ varied experiences with TRE were described in five themes: (1) changing everyday routines and habits entailed more planning, difficulties, and benefits; (2) social life was challenged; (3) involving partners, family, and relatives—more or less; (4) personal attitude and physical and mental experiences with TRE; and (5) supporting intervention principles, components, and activities.

Causal associations relating contextual factors and intervention activities to adherence to TRE through proposed mechanisms are suggested.

Insights from the pilot study have deepened our understanding of the TRE intervention and have informed the design and implementation of the following (now ongoing) RESET2 RCT. The findings of the pilot study will be further explored in the RCT in terms of how the intervention works for various participants and contribute to assessing the clinical relevance of TRE and its nuances among people with type 2 diabetes.

Abbreviations

HCP

Healthcare professional

IPT

Initial program theory

Q1, Q3

Quartiles 1 and 3

RCT

Randomised controlled trial

RESET2

REStricted Eating in the Treatment of type 2 diabetes

SD

Standard deviation

SDCC

Steno Diabetes Center Copenhagen

TRE

Time-restricted eating

Authors’ contributions

JSQ, KF, and NFH conceptualised the study and obtained funding. All authors contributed to the design of the study. ADT, AV, LJ, and GSH collected data. ADT and AV analysed and interpreted data. ADT wrote first draft of the manuscript. AV contributed to writing the manuscript. All authors reviewed and edited manuscript and approved the final version of the manuscript.

Funding

Open access funding provided by Copenhagen University ADT, NB, GSH, LJ, NFH, JSQ, and AV are employed at Steno Diabetes Center Copenhagen, a hospital and research institution under the Capital Region of Denmark, which is partly funded by a grant from the Novo Nordisk Foundation. This study was supported by funds from the Novo Nordisk Foundation (NNF200C0065836), the Danish Diabetes Association, and the Danish Diabetes Academy which is funded by the Novo Nordisk Foundation (NNF17SA0031406). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by Ethics Committee of Capital Region of Denmark (H-21041830).

Consent for publication

Not applicable.

Competing interests

KF has received research grants from Novo Nordisk A/S, Unilever A/S, and AstraZeneca A/S; holds shares in Novo Nordisk A/S and ChemoMetec A/S; and is currently employed by Novo Nordisk A/S. JSQ has received funding from Novo Nordisk A/S for other studies. ADT, NB, GSH, LJ, NFH, and AV report no potential conflicts relevant to this article.