Telephone‐based behavioral activation with mental imagery for depression: A pilot randomized clinical trial in isolated older adults during the Covid‐19 pandemic

Abstract Objectives To shield vulnerable persons, particularly the elderly, during the Covid‐19 pandemic governments around the world have advised to use social distancing and self‐isolation. Social isolation might put older adults at an increased risk for mental health problems such as depression. There is a need for brief, easy‐accessible psychological treatments for depressive symptoms that can be delivered remotely. The aim of this study was to investigate the feasibility, acceptability, and preliminary efficacy of telephone‐delivered Behavioral Activation with Mental Imagery (BA‐MI) for the treatment of depressive symptoms in individuals 65 years and older living in isolation during the Covid‐19 pandemic. Methods In this open‐label pilot randomized clinical trial, N = 41 individuals aged 65 years or older with clinically significant symptoms of depression were randomly assigned to either a BA‐MI treatment condition, or an Attention‐Assessment control condition delivered over the telephone over a 4‐week period. Results Depressive symptoms decreased more in the treatment condition compared to the control condition. At post‐treatment, 2 out of 16 participants in the treatment condition met diagnostic criteria for depression compared to 9 out of 13 in the control condition. Most participants in the treatment condition were satisfied with the treatment and few adverse effects were observed. Conclusions This pilot study suggests that behavioral activation with mental imagery delivered over the telephone is feasible, acceptable, and potentially efficacious for the treatment of depressive symptoms in older individuals living in isolation. Replication in larger samples is needed.


| INTRODUCTION
The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) causing the Coronavirus disease 2019 (Covid-19) became a pandemic in the year 2020 and into 2021. Covid-19-related mortality rates were the highest in older adults particularly those aged 70 years and above.
In Sweden, between April and October 2020, the government urged people 70 years and older to limit physical contact with other people and to stay at home as much as possible, a strategy called social distancing. 1 While these measures are important to shield vulnerable individuals, mental health experts have raised serious concerns over deteriorating mental health during and after the pandemic, 2 particularly among older adults. 3 Indeed, social isolation is related to an increase risk for depression and anxiety disorders among older adults. 4 Thus there is a need to develop and evaluate scalable psychological interventions for mental health problems for this patient group.
Evidence from previous epidemics and the current pandemic indicates that there is an urgent need for effective remotely delivered psychological interventions aimed at improving mental health among older individuals. 5 Research on the SARS epidemic in 2000 indicates that quarantine and isolation may lead to an increase in mental health conditions, such as depression. 6 Depression in older adults increases the risk of mortality and morbidity, 7 reduces quality of life, 8 and leads to functional impairments. 9 A study from Hong Kong during SARS found a nearly 32% increase in suicide rates among older individuals. 10 A recent review showed that social distancing has a negative impact on the mental as well as physical health of older individuals, including increased levels of depression, anxiety, and sleep problems. 5 A recent study in older adults from Sweden found that half of the sample reported staying at home all of the time, and that 44.4% of females and 25.7% of males reported feeling depressed. 11 Current evidence-based treatments for depression in older adults include antidepressant medication, 12 psychological interventions such as Cognitive-Behavioral Therapy (CBT), 13 and physical activity. 14 Previous studies have shown that the majority of older adults prefer psychological treatments to medication, 15 which poses a challenge during the Covid-19 pandemic as such treatments are often delivered face to face. Psychological treatments delivered via the Internet are as efficacious as face to face, 16 but only 3%-4% of the individuals 65 years and above in Sweden use digital applications instead of physical healthcare visits. 17 CBT has been shown to work when delivered via the telephone 18 and has proven to be feasible for older adults, 19 and one study showed that it may even result in lower attrition compared to face-to-face psychotherapy. 20 Due to the wide availability and accessibility of telephones, telephone-based psychological treatment could be an optimal intervention for older adults, particularly during pandemics.
Behavioral Activation (BA) is a brief psychological treatment that aims to (1) increase engagement in adaptive activities, (2) decrease engagement in activities that maintain depression or increase risk for depression, and (3) solve problems that limit access to rewarding activities or that maintain aversive control. 21 Common BA strategies are self-monitoring of activities and mood, planning and scheduling of activities in line with the patients' needs and goals, and problem solving. 21,22 BA is effective in reducing depressive symptoms 23,24 and there is emerging evidence suggesting that BA is also an effective intervention for older adults. 25 BA can be delivered in a brief format, 26,27 and has been shown to be feasible for older individuals in as few as four sessions. 28,29 However, depression is often characterized by low motivation and a lack of energy, which might make it more difficult for patients to initiate engagement in scheduled activities thereby reducing the efficacy of BA interventions. Thus identifying interventions that directly target these motivational difficulties could provide an additional route to promote greater behavioral engagement in scheduled reward activities. One way to facilitate motivation for planned activities might be via prospective mental imagery.
Mental imagery refers to the representation and experience of sensory information without external input. 30,31 By drawing upon prior knowledge and experiences, mental imagery can amplify the anticipation of reward-related emotions. 32,33 Anticipation of the pleasant and rewarding consequences of future behavior, in turn, predicts reward motivation and reward-motivated behavior. [34][35][36] Indeed, it has been shown that simulating engagement in scheduled pleasant and rewarding activities via mental imagery can increase motivation to engage in these activities. [37][38][39] Similar mental imagery interventions have successfully been used by older individuals. 40 The aim of this open-label pilot randomized clinical trial was to investigate the feasibility and preliminary efficacy of a brief telephone-delivered BA intervention, focused on activity scheduling augmented with mental imagery, for the treatment of depressive symptoms in individuals 65 years and older in isolation during the Covid-19-pandemic. We hypothesized that participants randomized to the active treatment condition would show a stronger decrease in depressive symptoms (primary outcome) and improve more on secondary outcomes, compared to participants randomized to an attention-assessment control condition. the Swedish Ethical Review Authority (2020-02079) and all participants provided written informed consent. The study was preregistered with ClinicalTrials.gov (NCT04508868). Note that the initial pre-registered plan was to conduct a full-scale trial, but due to changes in self-isolation restrictions for older individuals in Sweden from 22 October 2020, the recruitment had to be terminated early.

| Participants
Participants were approached using advertisements in local newspapers that invited individuals aged 65 years or older living in the county of Västmanland to participate in a telephone-based treatment of depressive symptoms. The inclusion criteria were (1) reporting clinically significant depressive symptoms as defined by scores above 12 on the Montgomery-Åsberg Depression Rating Selfrating Scale (MADRS-S), and/or scores above 9 on the Patient Health Questionnaire 9-item (PHQ-9), and/or scores above 5 on the currently receiving psychological therapy, or currently undergoing pharmacological treatments that commenced less than 1 month ago.
Participant characteristics are described in Table 1. Baseline ratings on primary and secondary outcome measures are described in Table 3.

| MINI
The Mini International Neuropsychiatric Interview 7.0 (MINI), 41 a structured clinical interview, was used to assess the presence/ absence of common psychiatric disorders at enrollment as well as assessing the presence/absence of depression post-treatment.

MADRS-S
The primary outcome measure was the MADRS-S, which is a nineitem questionnaire, used to measure severity of depression the past 2 weeks. 42 The score ranges from 0 to 54, with higher scores indicating higher depression severity.

GDS-15
Depressive symptoms were also measured using the GDS-15. 43 The GDS-15 is a 15-item questionnaire used to identify depression in older individuals with scores ranging from 0-15.

PHQ-9
The PHQ-9 is a nine-item questionnaire used to identify depression as well as measuring severity of depression with scores ranging from 0 to 27, with higher scores indicating higher depression severity. 44

GAD-7
Anxiety symptoms were assessed with the GAD-7, a seven-item questionnaire used to identify generalized anxiety disorder as well as measuring severity of anxiety symptoms. 45 The score ranges from 0 to 21, with higher scores indicating higher anxiety severity.

BADS-SF
Changes in avoidance and activation was measured using the Behavioral Activation for Depression Scale-Short Form (BADS-SF). 46

| Mental imagery ability
Psi-Q

| Iatrogenic effects
NEQ Adverse effects were measured using the Negative Effects Scale (NEQ)-short form, a 20-item questionnaire used to assess adverse and unwanted effects of psychological treatments. 49 In the analysis, only the items rated by the participant as probably caused by the treatment were included. Adverse effects were also monitored by the therapists throughout the interventions.

| Feasibility and acceptability
Feasibility was assessed by recording recruitment and dropout rates.
Time was measured for each session, in the treatment group as well as in the control group. Acceptability was measured with the question "overall, how satisfied were you with the treatment you received?", with ratings ranging from 1 to 5 representing "very dissatisfied" to "very satisfied", with 3 representing a neutral alternative, "neither satisfied nor dissatisfied".

| Interventions
The treatment manual as well as the manual for the control group was adapted into web-based versions on the system Entermedic.

| Procedure
Interested participants contacted the research unit and received information about the study. A psychologist or research nurse screened the participants concerning age, subjective symptoms of PELLAS ET AL.

| Randomization and blinding
An independent statistician generated a 1:1, 6-block randomization sequence. The statistician prepared sequentially numbered opaque envelopes to conceal allocation until the moment of randomization.
The corresponding author was responsible for evaluating eligibility and enrolling subjects in the study groups. The baseline questionnaires were scored by a research nurse and blinded to the psychologist until after the structured clinical interview. All subsequent selfratings were performed by the participants at home and sent by mail, and were scored by a research nurse and thereby blinded to the psychologists. A structured clinical interview was performed at week 4/post-treatment by the patient's therapist.

| Measures
The MADRS-S was assessed weekly. Acceptability and adverse effects were assessed at post-treatment. All other measures were assessed at baseline and post-treatment.

| Analysis plan
To test the effect of the intervention condition on the primary outcome measure, a repeated-measures ANOVA was conducted with condition (BA-MI vs. control) as the between subject factor and time

| Primary outcome measure
The means and 95% confidence interval for each group on MADRS-S across the different time points are shown in Figure 2. Means, standard deviations, and between-group effect size posttreatment on the primary outcome measure are shown in Table 3.

| Secondary outcome measures
Means, standard deviations, and effect sizes of secondary outcome measures at baseline and post-intervention are presented in Table 3.
Post-intervention, 2 out of 16 participants in the treatment group still met depression criteria according to the MINI, which is lower than the control group, where 9 out of 13 participants still met criteria, but not statistically significant according to Fisher's exact test (two-tail) p = 0.073.

| Iatrogenic effects
Nineteen participants in the treatment group completed the NEQ and 4 (21%) reported adverse effects that they attributed to the treatment. The most frequent adverse effects were increased anxiety, more unpleasant feelings, that unpleasant memories resurfaced, that they didn't always understand their treatments, that the treatment didn't produce positive results, and that became afraid that other people would find out about their ongoing treatment. No serious adverse effects or serious deterioration was reported by the therapists during the interventions.  In general, the results of this study should be interpreted with caution, given the small sample size that may lead to inflated effect sizes. 56 The number of individuals fulfilling depression criteria according to MINI decreased from 16 to 2 in the treatment group, and from 13 to 9 in the control group, which is promising given the brevity of the intervention.
Although the majority of participants (67%) in the treatment group reported being "very satisfied" or "somewhat satisfied" with the treatment, five participants reported being "neither satisfied nor dissatisfied", and one was "very dissatisfied" with the treatment. Four participants reported adverse effects on the NEQ that they attributed to the treatment, which include experiences of anxiety, un-

CONFLICT OF INTERESTS
The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.