Introduction
Generalised anxiety disorder (GAD) in older adults is associated with increased disability and service usage, alongside reduced quality of life. The rising cost of mental health provision for older adults as a result of increased life expectancy has led to increased focus on the need to provide cost-effective psychological treatments. Group therapies are a potential solution to increase the organisational efficiency of psychological treatments delivered in routine services. Controlled trials of group CBT for older adults with GAD have reported conservative and equivalent outcomes to non-directive psychotherapy, discussion groups and medication. Given that older adults prefer psychological therapy over medication when seeking treatment for GAD, there is an urgent need to develop and test new/modified group psychotherapy treatment protocols for GAD in older adult samples.
The intolerance-of-uncertainty model of GAD has four main features: intolerance-of-uncertainty, positive beliefs about worry, poor problem orientation, and cognitive avoidance (Dugas et al., Reference Dugas, Gagnon, Ladouceur and Freeston1998). Dugas et al. (Reference Dugas, Gagnon, Ladouceur and Freeston1998) describe intolerance-of-uncertainty (negative beliefs about uncertainty and its consequences) as a higher-order process, which drives the other three components, and which together result in the development and maintenance of GAD. Treatment based on the Dugas et al. (Reference Dugas, Gagnon, Ladouceur and Freeston1998) model of GAD has been tested extensively (Dugas and Roubichaud, Reference Dugas and Roubichaud2007). Randomised controlled trial results for working age adults have been encouraging (Dugas and Roubichaud, Reference Dugas and Roubichaud2007). Use of the intolerance-of-uncertainty protocol is in its infancy in the treatment of older adults with GAD.
The aim of the study was to examine the acceptability and feasibility of group cognitive behavioural therapy (CBT) for older adults with GAD based on the intolerance-of-uncertainty treatment protocol (Dugas and Roubichaud, Reference Dugas and Roubichaud2007). A longitudinal mixed methods approach was utilised to explore participant and facilitator experience of the treatment group, and its potential benefits.
Method
Potential participants were referred from older adult community mental health teams (CMHT) and Improving Access to Psychological Therapies (IAPT) services. Screening sessions applied the following inclusion criteria: (1) aged over 65 years, and already in contact with mental health services; (2) GAD diagnosis; (3) willing and able to attend the 12-week group CBT intervention; (4) able to read, write and understand English; (5) no evidence of significant cognitive impairment; and (6) no evidence of weighted risk. There were 23 eligible individuals, of which 87% (n = 20) opted into treatment, and 65% (n = 13) provided informed consent for participation in the research. Of the 13 research participants, 11 completed treatment (85%); reasons cited for drop-out were physical illness (n = 1) and domestic issues (n = 1). One research participant could not be contacted at 8-week follow-up.
Mixed methods enabled the triangulation of findings in order to increase the validity, reliability and credibility of the present study. Quantitative outcomes measures were administered at three main time points (pre-treatment, end of treatment, and 8-week follow-up); qualitative data were collected at the end of treatment.
The primary outcome measure was the Penn State Worry Questionnaire (PSWQ; Meyer et al., Reference Meyer, Miller, Metzger and Borkovec1990) which measures trait worry; for older adults a cut-off score of ≥50 indicates GAD. Secondary outcome measures included a standardised measure of anxiety, the Generalised Anxiety Disorder Scale (GAD-7; Spitzer et al., Reference Spitzer, Kroenke, Williams and Löwe2006) and a measure of depression symptoms, the Patient Health Questionnaire (PHQ-9; Spitzer et al., Reference Spitzer, Kroenke and Williams1999). The Intolerance of Uncertainty Scale (IUS; Freeston et al., Reference Freeston, Ladouceur, Rheaume, Letarte, Gagnon and Thibodeau1994) was also used as a measure of process, as is common in the evaluation of the Dugas and Roubichaud (Reference Dugas and Roubichaud2007) treatment approach.
One-to-one participant interviews were conducted by J.H. using a semi-structured interview used to provide a qualitative overview of factors clients find helpful in treatment. Facilitator feedback on experiences of delivering the protocol was collected via focus groups run by J.H.
Mixed data were analysed in parallel. At an individual level, clinically significant change and reliable improvement was assessed; evidence of both on the primary outcome measure (PSWQ; Meyer et al., Reference Meyer, Miller, Metzger and Borkovec1990) was the recovery criteria adopted. At a group level, effect size calculations were calculated and benchmarked against similar studies. Digitally recorded completer change interviews (n = 11) and facilitator focus groups (n = 2) were transcribed verbatim, and analysed using data-driven thematic analysis.
Treatment
Two clinical psychologists delivered the OWG treatment over 12 × 2-hour weekly sessions. All sessions had a consistent structure, starting with a homework review and a recap on the previous week, and ending with homework setting. The treatment manual was based on the Dugas and Roubichaud (Reference Dugas and Roubichaud2007) treatment protocol, with some older adult modifications. See Table 1 for the OWG session-by-session plan, and links to the Dugas and Roubichaud (Reference Dugas and Roubichaud2007) treatment protocol. Older adult adaptations included planned times for participants to share their knowledge and experiences of living and coping with anxiety and worry (e.g. group discussion exercises) and slower pacing, multi-modal learning and memory aids. Treatment integrity was assessed using a fidelity coding guide.
The Overcoming Worry Group session-by-session plan
Results
Acceptability and feasibility
Average weekly attendance was 87% (range 64–100%). Treatment integrity ratings averaged 94% (range 93–94%). Benchmarked findings found that the treatment had an equivalent opt-in rate (87%) and lower drop-out rate (15%) than group-treatment comparators (Hall et al., Reference Hall, Kellett, Berrios, Bains and Scott2016).
Participant experience
Four main participant themes emerged from treatment completers regarding acceptability and feasibility, as follows.
Enjoyable
Many of the participants (10/11) described the treatment as a pleasant and social experience: ‘I’ve enjoyed it, I think some of the time it was just meeting people as well’ (Participant 8).
Better in a group than expected
Almost half of the participants (5/11) described coping better with group-based treatment than expected: ‘I thought I might not be able to do that and yet I did do that, and went to all 12 of them’ (Participant 5).
Supportive facilitators
Facilitators were described as supportive and patient by most participants (9/11): ‘If you didn’t understand you just had to say and they went over it again’ (Participant 11).
‘Why invent worries!’
The hypothetical worry exposure task was not liked by many of the participants (5/11): ‘We had one week it was, I forget what it was titled and you think about going into a care home. One week it was a bit oh, made you go a bit like that. I thought I don’t know whether I like that’ (Participant 7).
The facilitator experience
Four main themes emerged from facilitator focus groups concerning the acceptability and feasibility of treatment.
OK together
Group delivery with other older adults was described by facilitators as an acceptable and ‘normalising’ treatment format: ‘People were very clear they liked being in a group with older people. It was something about similar stage and all that kind of thing’ (Psychologist 2).
Too much paperwork for some
Both facilitators suggested that they felt there were too many handouts for some of the participants: ‘One person felt and a few people agreed itʼs too many handouts’ (Psychologist 1).
Familiar co-facilitator helped
Previous experience of co-delivery was described as a factor which increased the feasibility of delivery: ‘I think it worked well because you and I have worked together a lot. So, it made facilitation easy’ (Psychologist 2).
Structure helps
Facilitators described the regular structure of the weekly protocol as a positive/helpful aspect of delivery: ‘I like the overall format. The familiarity, we start off the same and it pretty much ends the same. I think people respond quite well to that’ (Psychologist 1).
Potential clinical benefits
At an individual level of analysis, 5/13 participants met the worry (PSWQ) recovery criteria post-treatment, increasing to 7/13 at follow-up. At a group level, worry effect sizes were large at end of treatment, d = 2.04 (95% CI: 0.70, 3.38) and follow-up, d = 2.02 (95% CI: 0.69, 3.36). Benchmarking suggested that the intervention had a larger end of treatment worry effect size than group treatment comparators (Hall et al., Reference Hall, Kellett, Berrios, Bains and Scott2016).
Data triangulation – acceptability and feasibility
Quantitative findings indicated acceptability of the protocol: the opt-in rate (87%), drop-out rate (15%), average attendance (87%) and average homework completion rate (73%). This converged with qualitative findings, which also indicated that the protocol was acceptable for both participants (themes – ‘enjoyable’ and ‘better in a group than expected’) and facilitators (‘OK together’).
Similarly, qualitative and quantitative findings were concordant, and indicated both feasibility of delivery and factors which increased feasibility of delivery: treatment integrity (94%), facilitator theme ‘structure helps’, and facilitator theme ‘familiar co-facilitator’.
Discussion and conclusion
This proof-of-concept study provides evidence concerning the group treatment of GAD in older adults, and represents an important contribution to an under-researched area. The adapted protocol (the Overcoming Worry Group, OWG) based on the intolerance-of-uncertainty model appears an acceptable and feasible treatment, and preliminary outcomes observed exceed those reported in previous trials of group CBT for older adults (Hall et al., Reference Hall, Kellett, Berrios, Bains and Scott2016). Group delivery appears a potentially cost-effective intervention for older adults with GAD, although clinical effectiveness and cost-effectiveness need to be considered in parallel and equipoise.
Whilst the small sample size limits the generalisability of findings, it was appropriate to the main study aims around acceptability and feasibility. In order to explore clinical effectiveness, an increased sample size, alongside a study design with an active comparator would be indicated. In addition, as follow-up was relatively short, further studies assessing the clinical effectiveness of the amended protocol should include a longer follow-up period to examine treatment durability.
This study suggests the OWG is an acceptable and feasible treatment option for older adults with GAD. The protocol therefore shows promise as a treatment option for services looking to provide a potentially cost-effective therapeutic offer to older adults living with generalised anxiety disorder.
Acknowledgements
The authors thank all the participants in this study. The authors would like to express their gratitude to Mark Knowles and staff at Sheffield IAPT, for the assistance and support provided with patient recruitment.
Author contributions
The study was conceived and designed by J.H., S.K., S.S. and M.B. S.S. and M.B. conducted screening and facilitated the intervention, H.S. assessed treatment integrity and assisted with recruitment. J.H. collected data, conducted analyses and produced interpretations of the data. S.K. supervised the study, the facilitators, and assisted with the preparation of the manuscript. The final manuscript was approved by all.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflicts of interest
Dr Jo Hall, Dr Stephen Kellett, Dr Manreesh Bains, Heather Stonebank and Dr Shonagh Scott have no conflicts of interest with respect to this publication.
Ethical statement
Authors have abided by the Ethical Principles of Psychologists and Code of Conduct as set out by the APA. The study received NHS ethical approval (reference no. 15/YH/0137) alongside permission to proceed from Sheffield Health and Social Care NHS Foundation Trust (reference no. ZQ13) and Sheffield University research ethics committee (reference no. 177057).
Comments
No Comments have been published for this article.