Character Strength–Focused Positive Psychotherapy on Acute Psychiatric Wards: A Feasibility and Acceptability Study

The South London and Maudsley NHS Foundation Trust granted ethical approval for this study. A manualized single-session positive psychotherapy intervention focused on service users’ positive experiences and character strengths was developed by one of the authors (S.R.). It incorporated elements and exercises from three session components (“good things,” “identifying a personal strength,” and “using personal strengths”) in a positive psychotherapy for psychosis manual (8, 11). This positive psychotherapy intervention was integrated into a full timetable of psychosocial ward activities at a South London psychiatric hospital. A single-session design was purposefully adopted because of the short ward stay and irregular attendance of service users and in response to the limited provision of psychological interventions on the acute wards in the hospital. The intervention was developed in consultation with eight service users and researchers who emphasized the need to maximize the inclusivity of the intervention on the ward.

Participants were recruited in 2018–2019 from three of the hospital’s adult acute psychiatric wards (one for men, one for women, and one for persons of mixed gender). Ward staff referred service users who were deemed suitable for engagement in the intervention. Service users who showed challenging behaviors, such as violence and aggression, were not referred. In sessions, participants were invited to identify recent positive experiences (“good things”), consider what they had done to make a good thing happen, and link these experiences to a personal character strength they possess; participants then planned an activity for the week based on the strength identified. Sessions consisted of nine components and used A4 picture cards representing Values in Actions character strengths (12) to aid identification of strengths (known in the intervention as a “gallery of strengths”). Sessions had a target time of 60 minutes. Table 1 reports the session components.

A clinical psychologist (S.R.) or a trainee clinical psychologist (S.K.) led sessions with an assistant psychologist or a member of ward staff. Sessions were held weekly on each ward in a designated room. The session format enabled participants to participate in the intervention more than once if they wished. If so, they were encouraged to select character strengths and activities different from the ones previously chosen. Ward staff were encouraged to support service users in carrying out the activities.

Demographic data were accessed from clinical records. The World Health Organization’s ICD-10 codes were used to record a diagnosis block. Feasibility was measured by recording session duration, fidelity to session components, and whether participants could identify a character strength, plan a strengths-based activity, and carry out the strengths-based activity postsession. One week after the session, ward staff who were involved in participants’ care were consulted and clinical records were reviewed to identify whether participants had carried out their strengths-based activity. The intervention’s acceptability was measured by participants self-reporting their mood on a Likert scale ranging from 0 (most negative) to 10 (most positive) pre- and postsession. The aim of measuring acceptability was not to assess whether the session had improved mood but rather to determine whether the session had a negative impact on mood. “Helpfulness” of the session was rated on a scale ranging from 0 (not helpful) to 10 (extremely helpful) postsession. Ratings were collected in the session setting, and participants were invited to reflect on any mood changes, a routine approach adopted for other ward-based psychological interventions. Brief narrative feedback on benefits and challenges was collected in a feedback form. Ratings and feedback were optional, and participants were informed that their care would be unaffected if they opted out of providing feedback.

Statistical analyses were conducted in SPSS23. Paired-samples t tests compared pre- and postsession mood only for participants who provided both ratings. Mean±SD helpfulness ratings and narrative feedback themes were reported. Participants who attended sessions more than once were treated as new participants in analyses.

In total, 70 service users participated in this study, most of whom (N=55, 79%) attended a session once. Twelve participants (17%) attended twice, two participants (3%) three times, and one participant (1%) four times. The participants’ mean age was 37.1±13.3 years (range 18–74), and 46 (66%) were women. Thirty participants (43%) identified as White, and 25 (36%) identified as Black. Twenty-seven participants (39%) had a diagnosis of schizophrenia, schizotypal, or delusional disorder (ICD-10 codes F20–F29); 15 (21%) of adult personality and behavior disorders (codes F60–F69); and 15 (21%) of mood (affective) disorders (codes F30–F39).

Eighteen group sessions were delivered with 89% fidelity to session components (144 of a possible 162 components were delivered in the 18 sessions). Components 1–6 and component 9 were almost always delivered. Component 8 (“overcoming obstacles”) was omitted in nearly half of the sessions because of time constraints. Table 1 reports the details of the session components and the fidelity with which they were delivered.

Fifty-six participants (80%) identified a character strength; of these, 40 (71%) identified a strengths-based activity, and of these, 23 (58%) carried out the activity. Kindness (N=11, 20%), self-regulation (N=6, 11%), creativity (N=6, 11%), love of learning (N=5, 9%), and perseverance (N=5, 9%) were character strengths that were most commonly reported. Fifteen sessions (83%) lasted 60 minutes, and three (17%) were shorter (<60 minutes). Mood and helpfulness data were collected from 23 participants (33%). Mood scores significantly increased postsession (7.2±1.7) compared with presession (5.9±1.8) (t=5.3, df=22, p<0.001), with a large effect size (Cohen’s d=1.1). The mean helpfulness rating score was 8.5±1.5.

Forty-nine participants (70%) provided narrative feedback. Participants stated that the session was helpful and that they enjoyed identifying positive experiences. Participants liked component 5 (“gallery of strengths”), with some asking for copies of pictures to put on their walls; others reported enjoying connecting and sharing experiences. One participant reported, “[I] really enjoyed the session, felt it was really good for people in acute crisis” (participant 17). Another said, “I enjoyed connecting with other group members” (participant 55). Some participants reported wanting to discuss negative emotions: “[I] can see it being helpful for other patients but prefer to talk about negative emotions, as these are not discussed at home” (participant 36). Several participants reported that they did not like using numerical ratings for mood and helpfulness, and they did not provide such data.

The aim of this study was to evaluate the feasibility and acceptability of a single positive psychotherapy session delivered in a group format on acute psychiatric wards. This was a novel setting for positive psychotherapy, and high completion rates for session components, identification of strengths, and strengths-based activities indicated feasibility of this intervention in this setting. The positive effect of the intervention on mood and high helpfulness ratings indicated that it is acceptable to and has a positive impact on service users with complex and severe mental health conditions who are typically hard to engage in psychological interventions. The observed improvements in mood were consistent with findings of positive psychotherapy research on depression reduction in psychosis and from positive psychotherapy exercises that increase happiness (10). The prominence of character strengths such as kindness highlighted by the service users was consistent with results from research in populations of individuals with psychosis (7), and the activity completion rates were consistent with research indicating that positive exercises can be carried out on psychiatric wards (10).

Narrative feedback in this study about negative emotions was consistent with previous critiques of positive psychology, such as an overemphasis on happiness and the individual, which may overlook underlying structural determinants of mental health and well-being (13, 14). This observation underscores an important consideration for positive psychotherapy in acute and crisis settings. Delivering positive psychotherapy to service users who are significantly distressed and overwhelmed by negative emotions may be challenging. If service users feel that they are unable to think positively, they may develop feelings of guilt that can exacerbate original difficulties (15). Future research in acute and crisis settings might therefore investigate how positive psychotherapy might balance validation of distress with promotion of the positive benefits of engaging with positive experiences and character strengths.

A strength of this study was its adaptation of positive psychotherapy to acute and crisis psychiatric settings, which included individuals who had severe and complex mental health conditions. Limitations of this study included challenges of collecting numerical data, lack of standardized measures, and collecting data in sessions, which might have introduced bias. The relatively small proportion of participants who completed pre- and postmeasures reflected both the general difficulty of collecting data on acute psychiatric wards and the fact that irregular attendance is typical in this setting. Consultation with service users in the study design was limited, and future studies could involve service users more in study design. Use of staff reports and clinical records to collect data, rather than participant self-reports, may have been another limitation. Service users can lack trust in health care services, so it is possible that the participants may have complied with the intervention and postintervention activity because they wanted to appear well or wanted to be discharged rather than because they were interested in receiving positive psychology treatments.

Future research might test the effectiveness of this stand-alone character strengths–focused positive psychotherapy intervention in a larger study with a control group and might include more self-report data from service users about their activities and their views about attending the group session. If possible, it would be helpful to assess the impact of the intervention on service users who attend more than one session to see whether any further changes may occur. Fidelity to session components might be improved with more time spent planning the sessions. Reviewing session timings might address the lower completion rates for components 7 and 8, and the potentially associated moderate rate of participants carrying out the strengths-based activity. Providing additional individual support to service users might improve completion rates. Alternatively, future studies that wish to retain the single-session format might test a shorter version of the intervention with fewer components or going over the intervention components in more than one session.

The findings of this study indicate that positive psychotherapy is feasible and acceptable on acute psychiatric wards and that service users with severe and complex mental health conditions find the intervention helpful.

The authors thank the service users and clinical staff involved in this study and the Biomedical Research Centre Service User Advisory Group at the Institute of Psychiatry, Psychology & Neuroscience, King’s College London, for their consultation on this study. Dr. Slade acknowledges the support of the Center for Mental Health and Substance Abuse, University of South-Eastern Norway, and the NIHR Nottingham Biomedical Research Centre. The authors thank Sonia Kotecha, Tiffany Rameswari, Holly Smith, Emilia Woch, and Natalie Yap for clinical and research assistance.