CBT for anxiety in dementia


We are now facing a rapid increase of the ageing population across all societies. Although in the last decades, mental disorders in older adults received little attention due to the fact that usually health problems encountered by elders were considered a direct consequence of their age, there is now a great need for evidence-based interventions directed at older people.

Anxiety in dementia is very common, with a prevalence ranging from 5 to 21% for anxiety disorders and raising up to 71% for anxiety symptoms. For older individuals suffering from dementia, the most common symptoms of anxiety can present themselves as: motor restlessness, agitation and day/night disturbance and sometimes even aggression. Other symptoms such as increased dependency and behavioural problems might appear in late stages of dementia.

As for treatment, people most often receive psychotropic medication for which we have very little evidence, but we all know the potential serious adverse effects (Moretti et al, 2006).

Due to the fact that there is evidence for the efficiency of cognitive-behavioural therapy for anxiety in older people without dementia, the authors of this study aimed to develop and test a cognitive-behavioural therapy intervention in a single/blind, pilot randomised controlled trial.

Anxiety is common and problematic in dementia, and there are few treatment options available.

Anxiety is common and problematic in dementia, and there are few treatment options available.


The researchers first developed a manual by doing a systematic literature review, an expert review and a consensus conference with 30 people. After this, they tested the intervention with three people who suffered from dementia.

  • The first phase of the intervention was centred on: building a collaborative relationship, psychoeducation about CBT and the excess disability caused by anxiety in dementia, self-monitoring, developing an individualised formulation and identifying goals.
  • The second phase involved application of change processes, which the therapist could adapt according to every individuals’ needs and strengths.
  • The last phase involved work on ending the therapy process and developing a blueprint for the future.

Every patient had a carer and their role was to support the person with dementia in implementing strategies learned during the intervention. The involvement of the carer could range from very little (being there a few times) to being present at all times.

Participants were eligible for this intervention if they:

  • Met DSM-IV criteria for dementia in the mild-to-moderate range;
  • Had clinical anxiety, determined by a score of 11 or above on the RAID (Rating Anxiety in Dementia) Scale;
  • Lived in the community;
  • Had a self-identified carer who was willing to participate in the intervention;
  • Were able to understand and communicate in English;
  • Were willing to engage in therapy involving discussion of thoughts and feelings.

All assessments were administered in week one (baseline), week 15 (follow-up 1) and at 6 months (follow up 2).

They used the RAID Scale for measuring anxiety, the Clinical Services Receipt Inventory for measuring costs, the Cornell Scale for Depression in Dementia for measuring depression. Mood was measured using the Hospital Anxiety and Depression Scale and quality of life was measured using the Quality of Life – Alzheimer’s disease scale. Behavioural disturbance was measured with the Neuropsychiatric Inventory and cognitive function was measured using Mini-Mental State Examination. Researchers also measured the person-carer relationship using the Quality of Caregiver and Patient Relationship.

Participants and their carers were randomly allocated to either CBT or TAU, with an allocation ration of 1:1. The intervention lasted for 10 weeks, with a one hour session every week. Sessions were delivered by four clinical psychologists with experience of working with people with dementia.


  • The level of anxiety was lower in the CBT (17, 95% CI (14-21)) group than in the TAU one (22, 95% CI (17-24)) at baseline. These scores are measured by RAID scale and they are clinically significant, as any score of 11 or above is clinically significant

  • At follow up 1 (after 15 weeks) the level of anxiety was with -4.32 (95% CI -8.21 to -0.43) lower for the CBT group and 3.10 ( 95% CI -6.55 to 0.34 ) for TAU

  • This advantage was maintained at 6 months, with a group score of 4.59 (95% CI -9.34 to 0.15) points lower for the CBT group than for TAU

  • Also the use of anxiolytics and the level of depression were lower in the CBT group at baseline

  • Regarding cognitive functions, quality of life and caregiver-patient relationship; there were no notable differences between the groups at follow up 1

  • Regarding costs, the CBT group had a higher pre-baseline mean total cost from a health and social care perspective, compared with TAU.


The authors concluded that their investigation is feasible for people with mild to moderate dementia and clinically significant anxiety, with differences in anxiety levels that approached significance and maintenance of this differences at 6 months post intervention.

They also added that the therapy was acceptable to people with dementia and their family carers, demonstrated by their willingness to participate in the intervention.

The authors conclude that this pilot provides the evidence required for a large scale RCT to now be conducted.

The authors conclude that this pilot provides the evidence required for a large scale RCT to now be conducted.

Strengths and limitations

First of all, participants and their carers were not masked to group allocation because the intervention was psychosocial, but were asked not to disclose which arm of the study they had been assigned to. It is not about disclosure, it is about them knowing what type of intervention they receive and bias towards tests results.

Second, although the patients were randomly assigned to groups, those in the CBT group had only carers from family members, which means they had a lot of time to support the person caring for. In the TAU group, only 80% of the carers were family members, so we still have 20% of the patients in one group with lower levels of support.

Third, as also acknowledged by the authors, patient in TAU had higher use of both antidepressant and anxiolytic medication, and they were experiencing higher levels of anxiety. Therefore, how do we know that the lower levels of anxiety in the CBT group are a consequence of the intervention or of prior lower levels before the intervention?


This research demonstrated that formulation based CBT may be beneficial for people with dementia and anxiety, having also effects on depression levels and use of medication, but we need larger trials and better methods to investigate the effects of the intervention in order to be sure of its efficiency.

This pilot RCT suggests that a CBT manual may be a feasible approach for treating anxiety in people with dementia.

This pilot RCT suggests that a CBT manual may be a feasible approach for treating anxiety in people with dementia.


Primary paper

Spector A, Charlesworth G, King M, Lattimer M, Sadek S, Marston L, Rehill A, Hoe J, Qazi A, Knapp M, Orrell M. Cognitive–behavioural therapy for anxiety in dementia: pilot randomised controlled trial. The British Journal of Psychiatry Jun 2015, 206 (6) 509-516; DOI: 10.1192/bjp.bp.113.140087 [Abstract]

Other references

Moretti R, Torre P, Antonello R, Pizzolato G. (2006) Atypical neuroleptics as a treatment of agitation and anxiety in Alzheimer’s disease: risks or benefits. Expert Rev Neurother 2006; 6: 705–10. [PubMed abstract]

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+