ACT2COPE: A pilot randomised trial of a brief online acceptance and commitment therapy intervention for people living with chronic health conditions during the COVID-19 pandemic

Abstract

It is well established that the COVID-19 pandemic increased psychological distress in many populations, particularly for people with chronic health conditions (CHCs). Web-based mental health interventions provide a scalable and cost-effective approach to providing psychological support for people disproportionately affected by the COVID-19 pandemic. The current study aimed to explore the feasibility and acceptability of a one-week web-based psychological intervention (ACT2COPE) and explore preliminary effects of the intervention on reducing depression, anxiety, and stress symptoms, and improving wellbeing and psychological flexibility in adults living with CHCs during the COVID-19 pandemic in Aotearoa (New Zealand). A pilot randomised waitlist-controlled trial explored the acceptability and preliminary efficacy of ACT2COPE among 40 participants (n = 20 in the ACT2COPE group and n = 20 in the waitlist control group). Focus groups and open-ended questions explored usability and acceptability of the intervention as well as levels of engagement and adherence to the intervention. Mixed model ANOVAs explored within and between-group differences in psychological outcomes. Qualitative findings suggested that participants found ACT2COPE acceptable and engaging. Depressive symptoms significantly decreased over time compared to the waitlist group at 4-weeks follow-up (p = .012). No other between-group differences were found. The online ACT2COPE intervention presents a promising, scalable intervention that may improve psychological outcomes in adults living with CHCs during the COVID-19 pandemic. Future research is needed to confirm these findings in a larger and more diverse population and over a longer timeframe.

Keywords:

• Acceptance and commitment therapy
• chronic health condition
• COVID-19 pandemic
• randomised controlled trial
• depression
• digital health intervention
• eHealth
• mental health

COVID-19 first emerged in China in late 2019 before spreading around the world, causing devastating and lasting impacts; including economic recessions, psychological distress, significant social isolation, and over 6.9 million confirmed deaths worldwide as of April 2023 (Wind et al., 2020; World Health Organisation, 2023). The COVID-19 pandemic has heightened the need for accessible interventions to support those living with chronic health conditions (CHCs). In Aotearoa (New Zealand), this includes up to 67% of the adult population and 33% of the child population (Ministry of Health, 2009), similar to many other developed countries, including the United Kingdom (The King’s Fund, 2021). COVID-19 can represent an overwhelming stressor for those living with CHCs, especially considering the increased rates of mental health issues (Ministry of Health, 2009), in particular depression (Barnard et al., 2006; Liu & Tang, 2018; Moussavi et al., 2007) and anxiety (Asmundson & Katz, 2009; El-Gabalawy et al., 2011; Grigsby et al., 2002) among people living with CHCs. While the COVID-19 omicron variant constitutes a moderate illness for most people, those with underlying health conditions and older age are at higher risk of serious complications. People with CHCs may also experience increased physical isolation, health anxiety, and interruptions to usual care (Holmes et al., 2020). Heightened estimation of the risks of COVID-19, with corresponding poorer health linked to COVID-19-specific health anxiety have also been reported (Mertens et al., 2020).

Prolonged isolation and other pandemic-related social, economic, and environmental disruptions have seen significant mental health challenges emerge for individuals living with CHCs. For example, these have included a decline in general health, mood, and activity, and greater psychological distress in women with inflammatory arthritis (Maguire & O’Shea, 2021). In a Canadian study, adults with CHCs reported decreased feelings of belonging, greater health anxieties, loneliness, stress, and increased despair (Pettinicchio et al., 2021). Negative effects on health outcomes have also been recorded in China, where individuals with type 2 diabetes reported decreased glycaemic control during the initial outbreak of COVID-19, potentially due to fewer opportunities to exercise and lack of access to common foods (Banerjee et al., 2020). Delays and reductions in “non-essential” care can also have significant implications for those with CHCs, risking increases in morbidity and mortality (Papautsky et al., 2021).

Due to the reductions in contact with healthcare services and the array of new stressors associated with the ongoing COVID-19 pandemic, there is now a greater need for accessible interventions that address coping skills and self-management of CHCs (Pfefferbaum & North, 2020). Early indications suggest that digital coping skills interventions may result in long-term positive psychological outcomes comparable to face-to-face therapies (Andersson et al., 2014). Importantly, eHealth interventions are not only associated with increased cost- and time-effectiveness, accessibility, and acceptability (Andersson & Titov, 2014) but also present a means of safely offering psychological and social support to those who may need to avoid face-to-face contact during COVID-19 outbreaks. Interventions employing eHealth for CHC populations have become more numerous in recent years. For example, these have included approaches such as positive psychology and mindfulness techniques (Mikolasek et al., 2018) and acceptance and commitment therapy (ACT) (Dindo et al., 2017).

ACT-based interventions seem to be of a strong conceptual fit and of a promising evidence base for people living with CHCs. Interventions utilising multiple components of the ACT methodology have shown improvements in health outcomes, self-management, and mental wellbeing in multiple chronic health conditions, including diabetes, multiple sclerosis, obesity, inflammatory bowel disease, chronic pain, cardiovascular disease (Dindo et al., 2017; Dindo, 2015; Lillis et al., 2009; McCracken et al., 2013; Sheppard et al., 2010), cancer and acquired immunodeficiency disorder (Bassett et al., 2019; Isa et al., 2013; Kashani et al., 2012; Matchim et al., 2011). Systematic reviews and meta-analyses exploring the use of ACT-based interventions for people with CHCs have shown medium effect sizes for reducing pain intensity, depression, and anxiety, along with increased psychological flexibility, self-management, and physical wellbeing (Graham et al., 2016). Online adaptions of these ACT interventions have also been demonstrated to be effective in managing depression for those with a CHC (Brown et al., 2016; Lappalainen et al., 2014).

More specifically, ACT’s emphasis on accepting distressing but realistic, disease-related thoughts or beliefs rather than trying to modify them (Graham et al., 2016), may be a suitable approach to thoughts and beliefs surrounding COVID-19. In addition, psychological flexibility, one of the pillars of the ACT modality, has also emerged as a potentially important factor contributing to psychological wellbeing during the pandemic. Observational research in healthy populations has identified positive associations between psychological flexibility and wellbeing and inverse associations with anxiety, depression, and COVID-19-related distress (Dawson & Golijani-Moghaddam, 2020). Another study found psychological flexibility mitigated the negative consequences of COVID-19 (Landi et al., 2020). These recent studies support ACT as a promising therapeutic approach to improving psychological wellbeing during the COVID-19 pandemic (Kroska et al., 2020).

This study sought to build on these promising findings, investigating a brief web-based ACT intervention developed for adults living with a CHC in Aotearoa (New Zealand) during the COVID-19 pandemic. The study aimed to explore usability and acceptability of the intervention and to examine preliminary effects of the intervention on depression, anxiety, stress, psychological flexibility and wellbeing compared to a waitlist control group.

2. Methods

2.1. Study design

This study was a pilot randomised controlled trial (RCT) that compared a waitlist control to a brief web-based Acceptance and Commitment Therapy intervention (ACT2COPE) for improving wellbeing in adults living with a CHC. In addition, two focus groups were conducted after the intervention to explore usability and acceptability of the intervention and participants’ experiences and feedback on the website. This study was granted ethical approval on the 22nd of June 2020 by the Auckland Health Research Ethics Committee (AHREC; reference number: AH1362). Recruitment started on the 23rd of June 2020 and was completed by the 2^nd of July 2020. Focus groups were conducted between the 7^th and 9^th of September. The extended Consolidated Standards of Reporting Trials (CONSORT) (Eldridge et al., 2016) and Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines were adhered to and the trial was prospectively registered on the Australian New Zealand Clinical Trials Registry (ACTRN12620000634976).

2.2. Participants

We aimed to recruit 40 participants with 20 in each treatment arm. This sample size was considered adequate for a pilot RCT design where the focus is on feasibility and acceptability rather than efficacy of the intervention. The inclusion criteria were: (1) a diagnosis of a chronic health condition; (2) at least 18 years of age; (3) living in Aotearoa (New Zealand); (4) access to a computer and WiFi; (5) and ability to understand, read, and write in English. Participants were excluded if they were hospitalised during recruitment, were receiving treatment for a mental health condition, completed regular mindfulness or meditation practice, or were diagnosed with COVID-19. Six participants in the intervention group were also invited to join optional focus groups, a sample size deemed appropriate for evaluating the usability of digital interventions (Bardus et al., 2019).

2.3. Procedures

Participants were recruited online through Facebook and Instagram advertisements and support groups, and snowball recruiting was encouraged. Once participants were screened for eligibility and provided informed consent, they were immediately randomised into either the waitlist or intervention group using a web-based protocol of simple randomisation (https://www.randomizer.org/) and given the baseline questionnaire to complete. Participants’ sociodemographic information was collected at baseline only. This included age, ethnicity, sex, living arrangements, marital/relationship status, employment, level of education, diagnosis of CHC, comorbidities, medications, and age when diagnosed. Participants then received an email informing them which treatment condition they were assigned to.

Four weeks after baseline, participants were emailed a link to complete the follow-up questionnaire. Participants received a maximum of two email reminders and three text reminders to complete unfinished questionnaires. Participants were given two $NZD20 vouchers (totaling $NZD40), the first on completion of the intervention and the second on completion of the 4-week questionnaires. Participants in the waitlist-control group also received a link and password to access the ACT2COPE intervention.

In addition, participants allocated to the intervention group were invited to partake in optional 90-minute focus groups using Zoom to further explore their views on the ACT2COPE intervention. Two groups were conducted, with 3 participants per focus group. Both focus groups were facilitated by the first author (KWB), an Aotearoa (New Zealand) European female Master of Health Psychology candidate, and another study author (AB), a female Aotearoa (New Zealand) European female Health Psychology Ph.D. candidate. Both focus group facilitators had experience in facilitating group sessions and focus groups. Participants were informed that KWB was completing the study for her Masters, which involved developing a digital wellbeing intervention for people with CHCs, and had no prior relationship with either facilitator.

The focus groups followed a semi-structured interview schedule devised by KWB and her thesis supervisor (AS). Participants were shown images from the website to refresh their memory and were asked to provide feedback on intervention content, format, acceptability and engagement. After the second focus group, the possibility of data saturation was discussed between KWB, AB, and AS.

2.4. Intervention

The ACT2COPE intervention is based on Acceptance and Commitment Therapy, with modules teaching different skills from the Hexaflex model to improve psychological flexibility (Hayes, 2004; R. Harris, 2019). By offering the full range of skills from the Hexaflex model and different exercises to build each skill, ACT2COPE accounted for the individual needs of people and what skill or exercise they may find the most helpful. Participants were able to repeat modules as many times as they wanted.

ACT2COPE was designed for use within the COVID-19 pandemic context in Aotearoa (New Zealand). The different exercises and scenarios specifically refer to the pandemic and how participants can cope with stressors related to COVID-19 (e.g., increased isolation or increased health anxiety). Participants could either read the content on the website or listen to an embedded audio recording of the written content to allow for greater accessibility. Each module included a video developed by the research team to consolidate learning and practice ACT-related skills further. See Table 1 for an overview of all the modules. During the 1-week intervention, participants also received an email on days three and five to encourage them to keep using the website. Reminders included messages such as “it’s ok if your thoughts wander, keep practicing” to improve adherence and reduce attrition to the intervention.

Table 1. Outline of each ACT2COPE module

Module Title	Module Description
1. Introduction	Introduced how the intervention might benefit people with CHCs during COVID-19 and gave a brief overview of Acceptance and Commitment Therapy (ACT).
2. Present and Mindful	Discussed being present and mindful and gave strategies for practicing mindfulness daily.
3. Defusion	Explained defusion with metaphors to explain the concept and outlined a few exercises to practice defusion.
4. Observing Mind	Explained what observing the mind means and linked it to mindfulness exercises.
5. Acceptance	Discussed making room for thoughts and feelings rather than struggling to change them.
6. Hopes, Values, and Goals	Explored different types of values and explored current alignment with those values.
7. Committed Action	Explored values-based goal setting.

2.5. Outcomes

2.5.1. Feasibility and acceptability of ACT2COPE website

We were interested in exploring engagement and adherence to the intervention, recruitment methods, and lastly retention to the intervention. Regarding engagement and adherence, participants were asked how many modules they completed, with adherence defined a priori as having completed at least four out of the seven modules. All of the participants who completed the ACT2COPE intervention were asked, “do you have any feedback on this module of the intervention?” after each module and “how helpful did you find this module?” with the following multiple choice answers: (1) I found this module very helpful, (2) I found this module somewhat helpful, (3) I found this module only a little bit helpful, (4) I did not find this module helpful. Additionally, the focus groups explored these questions further and probed participants’ experiences using the website and any feedback for improving the content.

For recruitment methods, we recorded how many people we reached through recruiting via social media and how many individuals visited our website, were eligible and lastly consented to be involved in the study. We also kept a record of participants who dropped out of the study.

2.5.2. Depression, anxiety, and stress

Depression, anxiety, and stress were measured by the brief Depression, Anxiety, and Stress Scale (DASS-21) (Lovibond & Lovibond, 1995). The DASS-21 is a 21-item self-report scale measuring three subscales; depression, anxiety, and stress, with seven items per subscale. Each item is rated on a Likert scale, where 0= “did not apply to me at all”, and 3= “applied to me very much or most of the time”. For each subscale, scores are totalled and multiplied by two to obtain a score between 0 and 42, with higher scores indicating higher symptom severity. For each subscale, scores were grouped into normal, mild, moderate, severe, and extremely severe (Lovibond & Lovibond, 1995). The DASS-21 is reliable and valid in diverse samples, including among people with CHCs, and has been used for ACT-based interventions, with a mean Cronbach’s alpha of 0.91 showing high internal consistency (Antony et al., 1998). The Cronbach’s alpha for this present study was 0.89 at baseline.

2.5.3. Wellbeing

Mental wellbeing was measured by the World Health Organisation Wellbeing Index (WHO-5) (World Health Organisation, 1998). The WHO-5 is a 5-item positively worded scale on which participants rate their response to five statements (e.g., “I have felt cheerful and in good spirits”) on a 6-point Likert scale ranging from 0= “At no time” to 5= “all of the time”. Scores are added together with a range from 0–25, with 0 reflecting the lowest levels of wellbeing and 25 reflecting the highest levels of wellbeing. The WHO-5 is reliable and valid in a systematic review of varying health conditions with a validity of 0.81 (Topp et al., 2015). The Cronbach’s alpha for the present study was 0.74, deemed acceptable for research purposes (Allen & Bennett, 2012).

2.5.4. Psychological flexibility

Psychological flexibility was measured using the Acceptance and Action Questionnaire-II (AAQ-II) (Bond et al., 2011). The AAQ-II is a 7-item self-report measure in which statements (e.g., “I worry about not being able to control my worries and feelings”) are rated on a 7-point Likert scale ranging from 1= “never true” to 7 = “always true”. Total scores can range from 7–49, with low scores indicating high psychological flexibility and higher scores indicating low psychological flexibility. The AAQ-II has a good test-retest reliability of 0.81 and is reliable across multiple samples, including people living with CHCs (Bond et al., 2011). The Cronbach’s alpha for this present study was 0.86.

2.6. Data analyses

2.6.1. Qualitative analysis

The focus groups were recorded and transcribed, and the feedback from the focus groups and module feedback was collated into NVivo (a qualitative analysis software) for coding and analyses (Zamawe, 2015). The qualitative data from the module feedback and focus groups were examined using qualitative content analysis, a structured approach of systematically organising qualitative data to find key concepts and patterns in the data using an existing theoretical framework or model which guides the coding (Hsieh & Shannon, 2005; Schreier, 2012). Deductive content analyses was used to explore participant experiences of the intervention and their views on ACT-related concepts, including what they did and did not like, and what tools they used or planned to use after the intervention ended, as well as, any changes or improvements suggested for future iterations of this intervention, in line with a more positivist/realist epistemology. The analysis aimed to explore four research questions framed around exploring the usability and acceptability of the web-based intervention and ACT-related concepts; (1) “what did participants like about the intervention?”, (2) “what did participants dislike about the intervention?”, (3) “how did participants respond to the tools and activities in the intervention?”, and (4) “what would participants change or like to see in future versions of this intervention?”. KWB and AB co-coded the data and any discrepancies in codes or themes were resolved with the wider research team.

2.6.2. Quantitative analysis to explore preliminary effects of the intervention

Pearson’s correlations explored the relationships between psychosocial measures at baseline. A series of 2 (groups) x 2 (time) mixed-model Analysis of Variance (ANOVAs) were used to assess the differences in psychosocial outcomes between the two treatment groups over time. An additional analysis of covariance test (ANCOVA) was conducted to control for psychological flexibility (due to a trend towards group differences at baseline) and anti-depressant medication at baseline. Means, SDs, and 95% CIs are presented with the analyses. Effect sizes were calculated using partial eta squared (η_p²).

3. Results

3.1. Participant recruitment and demographics

Participants were recruited online through advertising on social media including Facebook and Instagram. The advertising ran for 10 days during June 2020 and reached 2837 people, resulting in 69 visits to the intervention website which contained the Participant Information Sheet and sign up process. From advertising the study on social media, 91 interested individuals completed an online screening questionnaire via a secure research platform (REDCap) to confirm eligibility, of which 40 were eligible and completed the online consent form.

Of the 91 individuals assessed for eligibility, 46 were eligible, 40 completed the consent form and baseline questionnaire and were randomised. Three participants were lost to follow-up (all from the ACT2COPE intervention group). Figure 1 shows the CONSORT diagram representing the flow of participants through the study. The demographic and clinical characteristics of the study population are summarised in Tables 2 and 3. Overall, participants were aged between 18 to 69 years and had a mean age of 39.20 years (SD = 14.77). The majority (87.5%) of the sample was Aotearoa (New Zealand) European, and there were 5% Māori (indigenous New Zealanders), 2.5% Indian, and 5% other ethnicities. Most of the sample was female (97.5%), currently in paid employment (62.5%), living with friends or family (87.5%), and in a relationship (67.5%). Education levels were mixed but the majority had completed tertiary education (60%).

Figure 1. CONSORT diagram and flow of participants through the study.

Table 2. Demographic Variables at Baseline

Demographic variable	Waitlist control (N=20)	Intervention (N=20)	Total sample (N=40)
Age in years M(SD)	36.55 (14.53)	41.85 (14.90)	39.20 (14.77)
Sex
Female	19 (47.5%)	20 (50%)	39 (97.5%)
Male	1 (2.5%)	0 (0%)	1 (2.5%)
Ethnicity
NZ European	18 (45%)	17 (42.5%)	35 (87.5%)
Māori	1 (2.5%)	1 (2.5%)	2 (5%)
Indian	1 (2.5%)	0 (0%)	1 (2.5%)
Other	0 (0%)	2 (5%)	2 (5%)
Studying
Yes	4 (10%)	4 (10%)	8 (20%)
No	16 (40%)	16 (40%)	32 (80%)
Level of education
Did not complete high-school	1 (2.5%)	2 (5%)	3 (7.5%)
Completed high-school	3 (7.5%)	2 (5%)	5 (12.5%)
Started undergraduate degree	5 (12.5%)	3 (7.5%)	8 (20%)
Completed tertiary education (e.g., Bachelors)	11 (27.5%)	13 (32.5%)	24 (60%)
Employment status
Employed	14 (35%)	11 (27.5%)	25 (62.5%)
Unemployed	6 (15%)	9 (22.5%)	15 (37.5%)
Living arrangements
Living alone	3 (7.5%)	2 (5%)	5 (12.5%)
Living with friends/family	17 (42.5%)	18 (45%)	35 (87.5%)
Relationship status
Single	8 (20%)	5 (12.5%)	13 (32.5%)
In a relationship	12 (30%)	15 (37.5%)	27 (67.5%)

Table 3. Clinical Characteristics at Baseline

Clinical characteristics	Waitlist control (N=20)	Intervention (N=20)	Total sample (N=40)
Age at diagnosis M(SD)	26.0 (17.37)	29.85 (15.26)	27.93 (16.25)
Number of CHCs M(SD)	3.60 (2.48)	2.90 (1.80)	3.25 (2.17)
Number of medications M(SD)	3.35 (3.56)	4.10 (4.23)	3.73 (3.88)
Diagnosis
Immune	6 (15%)	7 (17.5%)	13 (32.5%)
Inflammatory	10 (25%)	7 (17.5%)	17 (42.5%)
Pain	4 (10%)	2 (5%)	6 (15%)
Metabolic	0 (0%)	3 (7.5%)	3 (7.5%)
Other	0 (0%)	1 (2.5%)	1 (2.5%)
Comorbid diagnosis
Yes	16 (40%)	14 (35%	30 (75%)
No	4 (10%)	6 (15%)	10 (25%)
Currently on anti-depressants
Yes	4 (10%)	4 (10%)	8 (20%)
No	16 (40%)	16 (40%)	32 (80%)
Medications
Anti-depressant	0 (0%)	2 (5.5%)	2 (5.5%)
Immunosuppressant	4 (11.1%)	5 (13.9%)	9 (25%)
Pain	6 (16.7%)	2 (5.5%)	8 (22.2%)
Asthma	2 (5.5%)	2 (5.5%)	4 (11%)
Cardiovascular	1 (2.8%)	1 (2.8%)	2 (5.6%)
Other	4 (11.1%)	7 (19.4%)	11 (30.5%)

The age at which participants were diagnosed with their CHC ranged from 1 to 64 years, with a mean age of 27.93 years (SD = 16.25). Medical diagnoses were grouped into five categories, with inflammatory conditions (e.g., inflammatory bowel disease, arthritis) being the most common (42.5%), followed by immune conditions (e.g., coeliac and multiple sclerosis) at 32.5%, pain conditions (e.g., chronic regional pain syndrome and migraines) at 15%, metabolic conditions (e.g., type 2 diabetes and hypothyroidism) at 7.5%, and other conditions (e.g., scoliosis and nephrotic syndrome) at 2.5% of the sample. Most participants (75%) had multiple CHCs with a mean of 3.25 comorbid conditions (SD = 2.17). Despite an exclusion criterion of “currently receiving mental health treatment”, 20% of participants were taking anti-depressants during this study, as anti-depressants are commonly used as part of pain management for many CHCs (Mercier et al., 2013). Medications were grouped into six categories: anti-depressants (n = 10), immunosuppressants (n = 19), pain relief (n = 34), asthma (n = 18), cardiovascular (n = 10), and other medications (n = 55). The total number of medications participants were taking ranged from 0 to 17, with a mean of 3.73 medications per participant (SD = 3.88).

Six participants who completed the ACT2COPE intervention also participated in the focus groups, with three participants in each group. Participants were aged between 23 to 64 years and had a mean age of 33.83 years (SD = 15.96). Five out of the six participants were female. Five of the six participants were Aotearoa (New Zealand) European, and one was Indian. Half of the sample had comorbid health conditions, with the number of health conditions ranging from one to six (M = 2.33, SD = 1.97). Inflammatory conditions were the most common (three out of six), followed by immune conditions (two out of six), and other conditions (one out of six).

3.2. Baseline psychological measures

Based on the classifications of the DASS-21, where higher scores indicate higher severity, the total sample had moderate levels of depression (M = 14.75, SD = 9.67) and anxiety (M = 10.85, SD = 6.40) and mild levels of stress (M = 15.85, SD = 8.73) at baseline. The mean wellbeing scores (M = 9.82, SD = 4.17) and psychological flexibility (M = 24.58, SD = 8.60) indicated average levels at baseline.

3.3. Engagement and adherence

Self-reported adherence to the intervention was measured to explore engagement and acceptability, with nearly half (42.5%) of participants reported to have completed all seven modules. Overall, 22 participants (50.5%) met the a priori adherence level of completing at least four modules (See Figure 2). Six participants did not complete this question.

Figure 2. Self-reported number of modules completed for each participant at follow-up.

Note. Error bars represent 95% CI.

3.4. Qualitative analyses exploring usability and acceptability of ACT2COPE

An overview of the qualitative results, differentiating major and minor themes, can be seen in Table 4. Overall, there was a higher proportion of positive feedback than negative for the ACT2COPE intervention. The findings indicated that the intervention content and ACT-related concepts were generally viewed as acceptable, engaging, and useful. The findings from focus group participants and the open-ended questions on module feedback have also provided important feedback on necessary improvements to the ACT2COPE intervention.

Table 4. Summary of qualitative themes, with example quotations, and their corresponding prevalence in both the focus groups and module feedback responses. Themes are organised by how many focus group participants reported the theme

Research question	Themes	Frequency	Example quotes
Suggested changes to future versions	1. Module restructuring*	• Reported by all focus group participants (6 of 6) •Reflected in 2 of 137 module feedback responses	“It is the biggest of all modules, module six… And so I don’t know whether maybe you wanted to even either integrate with other modules, or split this module in half and then add an extra module maybe.” (Female, 29 years)
	2. Re-wording modules*	•Reported by 5 of 6 focus group participants •Reflected in 4 of 137 module feedback responses	“For me, it would [be helpful not to use the word mindfulness] because we’ve had it slammed in training and we’re just all so put it off, but for people that aren’t familiar with that, you know, it is a trendy word and a trendy thing.” (Female, 64 years)
	3. Improved accessibility*	•Reported by 5 of 6 focus group participants •Reflected in 2 of 137 module feedback responses	“And it would be really, really cool if it could be converted into an app somehow. I found that obviously, I got the reminder on my phone and then was like, ah, I have to go to my laptop and set all that up sort of thing.” (Female, 29 years)
	4. Tailoring the intervention*	•Reported by 5 out of 6 focus group participants •Reflected in no module feedback responses	“And so, get if you like, could choose when you wanted a reminder, maybe that could be quite cool” (Female, 29 years)
	5. No changes	•Reported by 3 out of 6 focus group participants •Reflected in no module feedback responses	“I wouldn’t change a thing.” (Female, 29 years)
	6. Videos and animations	•Reported by 2 out of 6 focus group participants •Reflected in 2 out of 137 module feedback responses	“So either just a voice-over or keep the animation but just maybe a more lifelike form of animation.” (Female, 29 years)
Likes	1. Content was helpful and relevant*	•Reported by all focus group participants (6 of 6) •Reflected in 33 out of 137 module feedback responses	“I like the fact that obviously it’s targeted at people with chronic illnesses who tend to be a bit more vulnerable anyway, but it was like a nice screenshot. And if people felt like they needed extra help or anything during this pandemic, instead of just downloading a random app from the App Store, like they got a really, really nice selection of different helping tools.” (Female, 29 years)
	2. Time and length of modules were achievable*	•Reported by all focus group participants (6 of 6) •Reflected in 7 out of 137 module feedback responses	“I thought the length of the little video slots was just about right. It didn’t like, eat into your day, didn’t feel like ‘oh, I need to schedule your time for this’. But it was still long enough to kind of get that message across.” (Female, 39 years)
	3. Clear layout and format*	•Reported by all focus group participants (6 of 6) •Reflected in 5 out of 137 module feedback responses	“It was really clear. I’ve done a few of these. And this was one of the better ones that I’ve ever done, just within terms of what we’ve covered, how it was covered, how it was laid out, step by step through it.” (Female, 64 years)
	4. Videos reinforced content*	•Reported by all focus group participants (6 of 6) •Reflected in 3 out of 137 module feedback responses	“ … it kept your concentration, and it kind of, in some ways, stopped your mind from wandering elsewhere. So you kind of really focus your attention on what you were doing. Kind of like a visual mantra.” (Female, 25 years)
	4. Helpful reminder*	•Reported by 5 out of 6 focus group participants •Reflected in 18 out of 137 module feedback responses	“The breathing was something I was already familiar with, so it was kind of a nice reminder. But that also, because I was familiar with it made it a nice segue into the progressive muscle relaxation, which I knew of but hadn’t practiced before.” (Female, 25 years)
Dislikes	1. Dislike of certain content or activities*	•Reported by all focus group participants (6 of 6) •Reflected in 10 out of 137 module feedback responses	“When I try and visualise, I only see black. And I just find it irritating when it says, you know, do this do that, and it didn’t work. But that’s, you know, some can visualise some can’t.” (Female, 64 years)
	2. Animations were distracting	•Reported by 3 out of 6 focus group participants •Reported by 1 out of 137 module feedback responses	“I think it was a good way to convey the message because I’m quite a visual person. But that kind of worked counterintuitively here because I found it distracting instead of engaging.” (Female, 25 years)
Experience with tools and activities	1. Increased ACT skills*	•Reported by 5 out of 6 focus group participants •Reflected in 5 out of 137 module feedback responses	“I liked how it created a little bit of awareness about all the different things that actually go on in your body, things that you don’t actually notice normally.” (Female, 23 years)
	2. Intention to use tools in the future*	•Reported by 4 out of 6 focus group participants •Reflected in 17 out of 137 module feedback responses	“I’d say I’d use breathing again definitely” (Female, 29 years)
	3. Felt more relaxed and calm*	•Reported by 4 out of 6 focus group participants •Reflected in 10 out of 137 module feedback responses	“I think I felt quite relaxed after particularly the first two. It was like ‘oh yeah, I kind of didn’t realise I was a bit stressed beforehand, and yeah, a similar response to the first one just felt quite aware and relaxed afterward” (Female, 39 years)
	4. Unpleasant reactions	•Reported by 3 out of 6 focus group participants •Reflected in 9 out of 137 module feedback responses	“I just found that side of it difficult that if you’re observing your thoughts all the time and what’s popping into your head, sometimes I’ve found that makes me focus on pain when I don’t want to.” (Female, 64 years)

Note: *Major theme, defined as being reported by four or more focus group participant

3.5. Quantitative analyses exploring preliminary effects of ACT2COPE

3.5.1. Correlations between psychological outcomes at baseline

Results demonstrated higher psychological inflexibility scores were positively correlated with increased depression scores at baseline, r(38)= .693, p < .001. Higher anxiety scores were also positively correlated with higher psychological inflexibility scores, r(38)= .455, p = .003. Similarly, higher stress scores were positively correlated with higher psychological inflexibility scores, r(38)= .592, p < .001. Results also demonstrated an association between psychological flexibility and wellbeing scores, r(38)= −.501, p < .001, where higher psychological inflexibility scores were negatively correlated with wellbeing scores. See Table 5.

Table 5. Pearson’s Correlations of Psychological Outcomes at Baseline

	1.	2.	3.	4.	5.
1. Depression	-
2. Anxiety	.417**	-
3. Stress	.727**	.606**	-
4. Psychological inflexibility	.693**	.455**	.592**	-
5. Wellbeing	−.626**	−.417**	−.487**	−.501**	-

Note: ** p < .01

3.5.2. Exploring the effect of ACT2COPE on outcome measures

Three 2 (time: baseline, 4-week follow-up) x 2 (group: intervention, waitlist) mixed ANOVAs were conducted to test whether participants allocated to the intervention group would demonstrate reductions in depression, anxiety, and stress scores, and improvements in wellbeing and psychological flexibility scores from baseline to four-weeks follow-up when compared to the waitlist control group. Results suggested that the intervention significantly improved (i.e., reduced) depression scores (F(1, 35) = 7.00, p = .012, η_p² = .17) from baseline (M = 15.80, SD = 8.97) to four-week follow-up (M = 8.24, SD = 6.96), relative to the waitlist control group from baseline (M = 13.70, SD = 10.49) to four-weeks follow-up (M = 11.80, SD = 11.24). Anxiety (F(1, 35) = 1.64, p = .209, η_p² = .05), stress (F(1, 35) = 1.12, p = .297, η_p² = .03), psychological flexibility (F(1, 35) = 3.47, p = .071, η_p² = .09), and wellbeing (F(1, 35) = 1.37, p = .250, η_p² = .04) scores did not significantly improve from baseline to four-week follow-up relative to the waitlist group. The findings did not change after adding anti-depressant use (n = 8) as a covariate or after controlling for psychological flexibility at baseline. See Table 6.

Table 6. Comparisons between groups in outcome measures over the study period

Outcome variable	Intervention group mean (SD)	Waitlist control mean (SD)	Marginal mean difference ACT2COPE versus waitlist (95% CI)	Mean difference P value	Group-by-time interaction P value	Partial eta squared (ηp^2)
Depression scores					.012*	.17
Baseline	15.80 (8.97)	13.70 (10.49)	2.10 (−4.13 to 8.33)	.499
Four-weeks	8.24 (6.96)	11.80 (11.24)	−3.57 (−9.94 to 2.81)	.264
Anxiety scores					.209	.05
Baseline	10.20 (5.87)	11.50 (6.99)	−1.30 (−5.43 to 2.83)	.528
Four-weeks	8.00 (7.48)	11.80 (8.48)	−3.80 (−9.19 to 1.59)	.161
Stress scores					.297	.03
Baseline	15.40 (8.31)	16.30 (9.32)	−0.90 (−6.55 to 4.75)	.749
Four-weeks	13.53 (7.73)	15.00 (9.77)	−1.47 (−7.23 to 4.49)	.619
Wellbeing scores					.250	.04
Baseline	9.95 (4.50)	9.70 (3.92)	0.25 (−2.45 to 2.95)	.852
Four-weeks	11.41 (3.84)	10.50 (5.26)	0.91 (−2.21 to 4.04)	.557
Psychological flexibility scores					.071	.09
Baseline	24.65 (8.99)	24.50 (8.42)	0.15 (−5.43 to 5.73)	.057
Four-weeks	22.59 (5.30)	24.80 (9.86)	−2.21 (−7.64 to 3.21)	.413

Note: * p < .05

4. Discussion

The current pilot randomised trial of a brief online ACT-based intervention demonstrates the acceptability and usability of the ACT2COPE intervention for adults living with CHCs during the COVID-19 pandemic, filling a significant gap in the literature. Qualitative results indicated that participants found ACT2COPE engaging and useful, appreciating the helpful and relevant content, brevity and flexibility of modules, the clear layout, and the inclusions of video content. In terms of their experience with tools, participants reported an increase in ACT-based skills, feeling calmer and more relaxed, and an intention to continue using the tools in the future. Reductions in depressive symptoms, even during ongoing COVID-19 restrictions, at four weeks were also evident, demonstrating potential clinical utility of ACT2COPE. However, no significant changes were observed in wellbeing or psychological flexibility. Qualitative results also highlighted areas for future improvement to improve engagement, including different animation styles, slight changes to module content, and options for tailoring and personalisation. Our findings suggest that an online ACT-based intervention is acceptable, conceptually relevant, and demonstrates preliminary efficacy in reducing depressive symptoms for people living with CHCs during the COVID-19 pandemic and times of increased restrictions.

In line with previous research, ACT2COPE supports and provides further evidence that brief ACT-based eHealth interventions can effectively improve mental health outcomes in CHC populations (Graham et al., 2016; Köhle et al., 2021; Mikolasek et al., 2018). These results are encouraging as this format allows for greater accessibility and scalability with less demand on public health resources during times of increased demand and public health restrictions (Andersson & Titov, 2014; Wind et al., 2020). Previous online ACT interventions vary in length, from 1 day to 12 weeks (Brown et al., 2016), with some having a time commitment of a full-day workshop or 60 minutes each week for multiple weeks (Köhle et al., 2021; Räsänen et al., 2016). Interventions that were similar in length to ACT2COPE did not include all aspects of the ACT Hexaflex model, instead focusing on just one part of the model, such as mindfulness (Cavanagh et al., 2013; Howarth et al., 2016). Therefore, ACT2COPE is novel in its integration of all six components of the Hexaflex in a brief format.

Improving psychological flexibility is a key goal of any ACT intervention and is associated with improved mental health outcomes and reduced COVID-19 distress (Dawson & Golijani-Moghaddam, 2020; Kroska et al., 2020; Pakenham et al., 2020). Previous studies have found a mediating relationship between psychological flexibility and improved psychological outcomes in ACT interventions (Lin et al., 2018; Trompetter et al., 2015). However, ACT2COPE did not improve psychological flexibility. Previous studies that did find a mediating relationship included interventions spanning multiple weeks (Lin et al., 2018), indicating that the brevity of ACT2COPE may have limited improvements to psychological flexibility. An alternative mediating factor could be increased self-awareness. In the qualitative data, participants reported noticing an increased awareness of their thoughts and emotions after completing ACT2COPE, growing their ACT-based coping tools. Increased self-awareness has been associated with improved mental health outcomes, including depressive symptoms (Gu et al., 2015), and thus may be an interesting avenue to explore in future research.

Similarly, the qualitative findings from this study were in line with previous eHealth studies reporting participants’ wishes for tailored intervention content, the ability to set reminders, and mobile accessibility (Boggiss et al., 2021; Liverpool et al., 2020). Research has also indicated that tailored and personalised interventions may also benefit wellbeing outcomes, such as sleep, stress, and physical symptoms related to somatisation in adult samples (Moe-Byrne et al., 2022), highlighting the importance of tailoring content. The current study also meets the needs outlined in previous eHealth studies of having varied activities, a combination of audio, video, and text information, and being shorter in length (Bendelin et al., 2011).

Further, the knowledge gained from this study has many promising clinical and theoretical implications. eHealth interventions are known to provide accessible and effective psychological support to many different populations. However, most have focused on being tailored to one patient population per intervention (Beatty & Lambert, 2013; Graham et al., 2016). ACT2COPE has shown that psychological benefits can occur for people with a diverse range of CHCs from a broad intervention. As ACT is grounded in functional contextualism (David & Mogoase, 2015), this study adds to the growing literature that ACT can be an effective treatment modality in a COVID-19 pandemic context for vulnerable populations.

The current study has several strengths. These include incorporating outcomes assessing both feasibility and preliminary efficacy as well as a qualitative element to gather deeper feedback on the ACT2COPE intervention. Other strengths include the intervention’s novelty, brevity, and flexibility. However, several limitations need to be discussed, such as the generalisability of results. Due to the sample having a high proportion of Aotearoa (New Zealand) European women, creating a homogenous sample, the results are disproportionality skewed. This sample skew is typical in studies with chronically ill populations, as females have a higher proportion of CHCs and are more likely to participate in eHealth interventions, creating a self-selection bias (Donkin et al., 2012). Another limitation of this study, and many eHealth interventions, is that of self-reported adherence to the intervention. While many eHealth interventions often fail to report adherence (Donkin et al., 2011), a systematic review of 48 eHealth interventions found an average adherence rate of 50% (Kelders et al., 2012). The current study had an adherence rate of 55%, which is in line with this previous research. However, adherence was self-reported, which is vulnerable to social-desirability bias. Therefore adherence rates in the current study may be inflated (Van de Mortel, 2008). In addition, the AAQ-II has shown to have some limitations in its construct and face validity that could have impacted the results of this study (Tyndall et al., 2019). Given the short time frame of this study, small sample size and lack of diversity in terms of sex and ethnicity, further research and replication are needed to confirm these findings and determine underlying mechanisms. Future research could include a larger, more diverse sample with a longer follow-up for examining the maintenance of effects. A design that allows for manipulation of intervention length could provide knowledge about the length of intervention required to produce beneficial psychological outcomes.

Overall, ACT2COPE provides a novel and promising mental health intervention for people living with CHCs. This study extends current knowledge and offers further evidence for the feasibility and efficacy of brief online interventions. These interventions have the capacity for more extensive reach and accessibility for many hard-to-reach populations while reducing resource burdens on patient populations. As such, the ACT2COPE intervention is currently being adapted to be tested in a fully powered randomised controlled trial.

Clinical Trial Registration

ACTRN12620000634976

What is already known on the subject?

• For people living with chronic health conditions, the COVID-19 pandemic added barriers to accessing psychological care, and increased isolation and psychological distress.

• Psychological interventions that are scalable, accessible, and cost-effective are needed to support people living with chronic health conditions.

What does this study add?

• Early findings demonstrate ACT2COPE as a promising, acceptable, and scalable intervention.

• A web-based Acceptance and Commitment Therapy intervention (ACT2COPE) reduced depressive symptoms in those living with a chronic health condition, compared to a waitlist control group during the COVID-19 pandemic.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Additional information

Notes on contributors

Kate Wallace-Boyd

Kate Wallace-Boyd is a Health Psychologist working at the Massey Psychology Clinic in Auckland, New Zealand.

Anna Lynette Boggiss

Dr Anna Boggiss is a postdoctoral researcher and an Intern Psychologist in Health Psychology at the Department of Psychological Medicine, University of Auckland, New Zealand.

Sian Ellett

Sian Ellett is currently completing postgraduate study in Health Psychology at the Department of Psychological Medicine, University of Auckland, New Zealand.

Roger Booth

Associate Professor Roger Booth is an immunologist and teaches in the health psychology and medical postgraduate programmes at the University of Auckland, New Zealand.

Rebecca Slykerman

Dr Rebecca Slykerman is a neuropsychologist and Senior Lecturer at the Department of Psychological Medicine, University of Auckland, New Zealand.

Anna Sofia Serlachius

Dr Anna Serlachius is a Senior Lecturer in Health Psychology at the Department of Psychological Medicine, University of Auckland, New Zealand.