Impact of ACT-based Interventions on Diabetes-related Outcomes

A Pilot Study to Examine the Effects of Mindfulness Versus Values-plus-goals Interventions for Adults With Diabetes Treated With Insulin

Individuals with insulin-treated diabetes can experience psychological difficulties associated with living with and managing the condition. Acceptance and Commitment Therapy (ACT) is being increasingly used to treat these psychological difficulties, with research in this area indicating positive psychological and diabetes-related outcomes (Gregg, Callaghan, Hayes, & Glenn-Lawson, 2007; Shayeghian, Hassanabadi, Aguilar-Vafaie, Amiri, & Besharat, 2016). Given the lack of psychology funding in diabetes care provision, a financially feasible theory-based intervention is much-needed (Diabetes UK, 2008). ACT may be the solution as it can be delivered in smaller modules. The study aims to investigate the effectiveness of two online ACT-based interventions (a mindfulness-based intervention [MBI] and a values-plus-goals intervention [VGI]) on wellbeing, diabetes self-management, coping style and glycaemic control among a sample of adults with insulin-treated diabetes. It also aims to examine whether the interventions are associated with changes in diabetes acceptance and valued living, and whether diabetes acceptance and valued living are associated with the aforementioned outcomes. Participants will be recruited from the diabetes outpatient clinics at Ashford and St. Peter's Hospitals NHS Foundation Trust to take part in the study. They will be randomly assigned to take part in either the MBI or VGI, which are both 4-week interventions. Participants will be asked to complete self-report questionnaires to measure their wellbeing, diabetes self-management, coping style, diabetes acceptance and valued living at the beginning of the study, at the end of the intervention and at a 1-month follow-up. Glycaemic control will be measured at the beginning of the study and at a 2-month follow up. It is hypothesised that both interventions will improve diabetes-related outcomes. It is hypothesised that MBI may be associated with increases in acceptance and more positive emotion focused coping, whereas the VGI may be associated with increased valued living and problem-focused/active coping.

Background Diabetes is a long-term condition which affects the way the body regulates blood sugar in around 4.7 million people in the United Kingdom (Diabetes UK, 2019). There are two types of diabetes: Type 1 is when an individual is unable to produce any insulin, which is a hormone that allows sugar to enter the cells in our bodies, and Type 2 is when an individual no longer responds to insulin (Diabetes UK. 2019). Type 1 Diabetes is treated using insulin therapy and some people with Type 2 Diabetes will also need insulin therapy (DeWitt & Hirsch, 2003). Individuals with diabetes have a greater risk of experiencing heart attacks, stroke and cardiovascular diseases, and complications from the condition such as amputations, sight loss and kidney disease (Diabetes UK, 2019). They also face many psychological challenges associated with managing the condition, such as depression, diabetes-related distress, and negative coping strategies, which can lead to poorer diabetes self-management (Rane, Wajngot, Wändell, & Gåfvels, 2011). According to the annual diabetes prevalence figures published in 2019 by Diabetes UK, around 40% of people with diabetes experience psychological difficulties and up to 65% can experience low mood related to their diabetes. Those with mental health difficulties seeking physical treatment can cost the NHS up to 50% more than those without mental health difficulties (Diabetes UK, 2019). Psychological interventions and Acceptance and Commitment Therapy A range of psychological interventions have been used to treat individuals with diabetes, such as psychoeducation about diabetes self-management, problem-solving approaches and Cognitive Behavioural Therapy, which aims to challenge negative thoughts (Thorpe, Fahey, & Johnson, 2012). However, due to the realistic nature of thoughts related to chronic health conditions at times, approaches that challenge thoughts have limited effectiveness for some individuals (Hofmann, Sawyer, & Fang, 2010). Acceptance and Commitment Therapy (ACT; Hayes, Strosahl, & Wilson, 1999) may be a promising approach for use with the diabetes population. It is a therapeutic approach that is increasingly being used to treat psychological difficulties experienced by individuals with health conditions, as it moves away from attempts to alter internal experiences and promotes value-based living alongside our internal experiences. There is an emerging evidence base for the use of ACT with diabetes. Gregg, Callaghan, Hayes and Glenn-Lawson (2007) demonstrated that it is effective in improving coping styles, diabetes self-management and blood sugar control among adults with diabetes. ACT has been shown to be effective in improving diabetes-related outcomes in other studies as well (e.g. Shayeghian, Hassanabadi, Aguilar-Vafaie, Amiri, & Besharat, 2016). It has been delivered online successfully in the diabetes population through the use of smartphones, suggesting that it is feasible and effective in reducing anxiety, depression and diabetes-related distress (e.g. Nes et al., 2012; NHS Grampian, 2015). Individuals with diabetes use a range of coping strategies to manage the condition (Lazarus & Folkman, 1984). Karekla, Karademas and Gloster (2018) proposed links between a commonly used illness self-regulation model and ACT, which increases our theoretical understanding of how ACT can be helpfully applied to the difficulties experienced by people with diabetes, particularly given that it can be delivered in individual modules (Hayes et al., 1999). ACT promotes more active coping styles through enhancing individuals' values and goals and encouraging committed action (Hayes et al., 1999). ACT's mindfulness and acceptance components promote positive emotion-focused coping such as increased mindful awareness, and learning to live alongside difficult internal experiences. Therefore, learning these ACT skills may result in changes in coping strategies and diabetes-related outcomes. Due to the national lack of psychology funding in diabetes services, many services are not able to provide psychological interventions (Diabetes UK, 2008). There is a need for financially feasible and accessible interventions to support the wellbeing of patients with diabetes. Knowing which components of ACT are most effective would allow services to provide a short and more financially-feasible targeted intervention that is tailored to individuals with diabetes. The study is novel, as no research examining different components of ACT with people with diabetes has been published to date. Aims The present study aims to extend previous research on using ACT within the diabetes population, by taking a modular approach and examining the impact of two online ACT component-based interventions (a values-plus-goals intervention [VGI] and a mindfulness-based intervention [MBI]) on patients with insulin-treated diabetes. In particular, the investigators are interested in improving diabetes-related outcomes (i.e. wellbeing, diabetes self-management, coping skills and glycaemic control) by enhancing diabetes acceptance and values-based living. The research questions are: Do VGI and MBI significantly improve self-reported wellbeing, diabetes self-management, coping skills and glycaemic control in adults with insulin-treated diabetes? If so, are these improvements maintained at follow-up? Does MBI significantly increase diabetes acceptance compared to VGI? Does VGI significantly increase valued living compared to MBI? Is there an association between changes in diabetes acceptance pre-post MBI and wellbeing, diabetes self-management, coping skills and glycaemic control at follow-up? Is there an association between changes in valued living pre-post VGI and wellbeing, diabetes self-management, coping skills and glycaemic control at follow-up? Design A mixed design will be employed (between- and within-subjects). Participants will be randomised to take part in either the mindfulness intervention or the values-plus-goals intervention. They will complete self-report questionnaires on wellbeing, diabetes self-management, coping skills, diabetes acceptance and valued living at three different time-points: pre-intervention, post-intervention and at a 1-month follow-up. Glycaemic control will be measured pre-intervention and a 2-month follow-up. Sample, setting and recruitment The target sample will be English-speaking patients aged 18 and over, who have a diagnosis of insulin-treated diabetes and are currently accessing NHS diabetes services. Individuals must have an HbA1c value of 64 mmol/mol or higher (indicative of poor glycaemic control). The sample will be recruited from local NHS diabetes services in Ashford and St Peter's Hospitals. Consultant Endocrinologists from both services, Dr Thang Han and Dr Helen Ward, confirmed their interest in the project and granted permission for the investigators to use their services for recruitment and to collaborate on medical aspects of the project. Flyers and posters will be put up in the waiting area so that interested participants can discuss the project with a member of the care team to determine eligibility and be provided with a Participant Information Sheet. Clinicians and nurses will also mention the study to eligible patients at the end of their consultations, where possible, to raise awareness of the study and provide contact details of the researcher via the provision of a paper copy of the Participant Information Sheet and Consent Form. Once eligible participants have agreed to take part, they will be sent an electronic Participant Information Sheet to go through the informed consent process electronically. Participants will be given ample time to consider whether they wish to take part and will have the opportunity to have any questions answered by the CI prior to the consent process. Based on an a-priori analysis on G*Power using an effect size of 0.25, an alpha level of 0.05 and power of 0.8 (Cohen, 1992), a sample size of 24 per group is required for a two-tailed between- and within-factors analysis of variance. In total, the investigators are aiming to recruit at least 25 participants per group to ensure that they will be able to detect effects related to the hypotheses. Procedure The study itself is web-based and will be accessible to patients from home, provided that they can access the Internet. All participants will receive a Qualtrics link to take part in the study. They will be presented with a Participant Information Sheet to read and asked for informed consent before proceeding with the study; they will also have had physical copies of these sheets. They will then be randomly allocated to VGI or MBI through Qualtrics. Before the intervention, participants will be presented with five self-report measures to complete electronically. They will then have access to the four-week intervention, following which they will be asked to complete the same set of five outcome measures. One month after intervention completion, participants will be contacted to complete follow-up measures. Glycated haemoglobin level (HbA1c), a widely-used indicator of glycaemic control, will be recorded before the intervention and at a 2-month post-intervention follow-up, as it is an average blood sugar level measure that is representative of the last 3 months. Participants will also have the opportunity to try the intervention that they were not randomly assigned to, at the end of the study. No further interventions would be available thereafter. Analysis The main analyses will include: (i) Exploring group differences in demographic and clinical factors; (ii) Mixed-design analyses to compare the impact of the interventions on the outcomes of interest across the three time-points (analysis of variance); (iii) Examination of changes in acceptance and valued living within each intervention group (analysis of variance); (iv) Using PROCESS analysis to explore (1) whether changes in well-being in the MBI condition is partially explained by changes in diabetes acceptance and emotion-focused coping style, and (2) whether changes in self-management and glycaemic control in the VGI condition are partially explained by changes in valued living and active coping style. Data collection, storage and security arrangements Only personal information provided by participants and their clinicians to the researcher as part of the research will be accessible to the researcher (i.e. the researcher will not have access to the participant's clinical notes). The researchers and co-investigators will have access to the online consent forms, demographic information and self-report questionnaires, as outlined in the Participant Information Sheet and agreed by participants in their consent form. All data will be anonymised therefore no one will be identifiable. Each participant will be allocated a unique personal identification number which will be presented on all questionnaires, demographic and clinical information sheets, computerised data files, and the study database. A separate password-protected file containing the participant's name, telephone number, e-mail address and study number will be kept. All data will be stored in accordance with the guidelines laid out in the NHS code of confidentiality and EU General Data Protection Regulation (GDPR). All data will be stored on the sponsor's secure computer system (Royal Holloway, University of London). There will be no physical data stored during the study. The file containing identifiable personal information will be destroyed once the study data has been fully analysed and written up. After the study has ended, research data will be stored up until the CI's fulfillment of their Doctorate in Clinical Psychology and for up to 5 years following publication in a scientific journal for audit purposes. Dr Michelle Taylor will be in control of and act as the custodian for the data generated by the study.

The VGI is a four-week intervention, adapted from an existing online value-affirmation and goals intervention (Kingston & Ellett, 2014) and an existing 'Living Well with Physical Health Conditions' programme handbook. Participants are first introduced to the concept of values as understood in Acceptance and Commitment Therapy. They are then asked to complete a card-sort task, which involves organising ten life domains according to their perceived level of importance. Participants will be asked to choose one values that is the most important and meaningful to them and to write for up to 10 minutes about it, thinking about why it is meaningful to them. Following this, participants will be given information on how to set value-based 'SMART' goals and asked to choose one to focus on over the next month. Participants will be sent a reminder e-mail every week over the 4 weeks, asking them to review their progress towards the SMART goal they set themselves.

The MBI is a four-week intervention, where participants are introduced to the concept of mindfulness and acceptance as understood in ACT (Hayes et al., 1999). They will be asked to complete a 10-minute health-focused mindfulness meditation exercise by listening to a recording of a script focusing on cultivating mindful awareness of internal experiences, particularly in relation to diabetes, and practising making room for any difficult internal experiences. This will be followed by some prompts to promote reflections related to their diabetes following the exercise. They will be asked to practise this exercise 3-4 times a week over a four-week period. As a manipulation check, participants will be sent a reminder e-mail every week over the four weeks to record how many times they practised the exercise over the past week.

Diabetes Self-Management Questionnaire (DSMQ; Schmitt et al., 2013) is a 16-item measure of self-care activities associated with glycaemic control. Each item is scored on a 4-point Likert scale from 'applies to me very much' (score of 3) to 'does not apply to me' (score of 0). The measure is composed of four subscales: 'Glucose Management', 'Dietary Control', 'Physical Activity' and 'Health-Care Use'. Schmitt et al. (2013) have shown that the DSMQ has good internal consistency (0.84) and significant convergent correlations with parallel scales of self-management and HbA1c levels.

Well-Being Questionnaire (W-BQ; Bradley, 1994) is a 22-item measure of psychological well-being specifically designed for use with people with diabetes, focusing on cognitive symptoms rather than somatic symptoms, as the latter may overlap with the physical condition of diabetes. Items are scored on a four-point Likert scale where respondents indicate how much each item applies to them in the past few weeks from 'not at all' (score of 0) to 'all the time' (score of 3). The measure is composed of four subscales: Depression, Anxiety, Energy (fatigue) and Positive Well-being. The scale has shown satisfactory construct validity and internal consistency (0.7-0.88), and has been shown to be a useful tool in evaluating new interventions (Bradley & Lewis, 1990).

Brief COPE (Carver, 1997) is a 28-item widely-used measure of an individual's range of coping responses. Each coping statement is scored on a 4-point Likert scale from 'I've been doing this a lot' (score of 4) to 'I haven't been doing this at all' (score of 1). There are 14 scales overall: 'Self-distraction', 'Active coping', Denial', 'Substance use', 'Use of emotional support', 'Use of instrumental support', 'Behavioural disengagement', 'Venting', 'Positive reframing', 'Planning', 'Humour', 'Acceptance', 'Religion', and 'Self-blame'. The scale has good internal consistency and concurrent validity (Carver, 1997).

Diabetes Acceptance Scale (DAS; Schmitt et al., 2018) is a 20-item measure of diabetes acceptance, composed of four subscales 'Integration', 'Motivation', 'Avoidance/Defence' and 'Distress'. Each item is scored on a four-point Likert scale, where 'never true for me' (score of 0) and 'always true for me' (score of 3). The sum score can range from 0 to 60 with higher scores indicating higher diabetes acceptance.Schmitt et al. (2018) demonstrated that factorial and criterion validity were high, as well as internal reliability (0.96).

Valued Living Questionnaire (VLQ; Wilson, Sandoz, Richards & Roberts, 2010) is a 20-item measure of value-based living in the past week across ten life domains (Family, Relationships, Parenting, Friendship, Work, Education, Recreation, Spirituality, Citizenship, and Physical Self-care). Each domain is rated on a ten-point Likert scale of importance and consistency, from 'not at all important/consistent' (score of 1) to 'extremely important/consistent' (score of 10). Wilson et al. (2010) have shown satisfactory internal construct validity and internal reliability (0.74).

Glycated haemoglobin level (HbA1c), a widely-used indicator of glycaemic control, will be recorded before the intervention and at a 2-month post-intervention follow-up, as it is an average blood sugar level measure that is representative of the last 3 months.

Inclusion Criteria: Diagnosis of type 1 diabetes or type 2 diabetes treated with insulin Recent HbA1c value of 64 mmol/mol or higher At least 18 years of age Fluent in written and spoken English Access to the Internet to access the intervention Exclusion Criteria: People with diabetes treated with diet and/or tablets alone Non-English speakers Aged under 18 years No Internet access Severe and enduring mental health problems Presence of another serious and/or life-threatening physical condition Cognitive and/or visual impairment Currently accessing psychological treatment