The project aims to investigate the effects of a short-term cbt-based psychotherapy intervention for a diagnostically mixed group of psychiatric inpatients. In a multiple baseline single subject design, 5-10 patients with mixed diagnoses that are treated at any of the inpatient units at the Hospital of Västmanland, Västerås, Sweden, will be offered a short psychotherapeutic intervention. The intervention is cbt based, and consists of focused functional analysis with identification of the main problem to be treated; an experiential exercise called the life line, in which obstacles to living a valued life are investigated, and in which alternative steps to be taken in spite of psychiatric symptoms are formulated; a summary excercise called the pause, in which central principles of the treatment are repeated, and in which a relapse prevention plan is formulated. The intervention will consist of approximately 2-5 sessions. The primary research question is whether such an intervention is effective in terms of causal change in problem areas identified as personally meaningful by the individual patient.

The most severely ill psychiatric patients are treated on inpatient wards. On most Swedish inpatient wards, psychological treatment is not routinely offered. At the same time, research show that psychological treatment often adds a clinically significant effect to routine care. However, it is a great challenge to handle the diagnostically complex picture present in this context. Often, diagnoses are unclear or preliminary, or there are many co-morbid conditions to be treated simultaneously. Available treatment protocols are often developed for an outpatient population, and for a specific diagnostic group. This, combined with cultural and organizational challenges inherent to the inpatient context, makes the implementation of psychological treatment for these patients a great challenge. Factors such as form of care (voluntary or mandatory), the patient's motivation, and other unforeseen events add to the fact that the length of any given episode of inpatient care is difficult to estimate. This is also a fact that a psychological treatment model must adapt to.

There seems to be a need for developing psychological treatment models in a way that makes them suitable for the special challenges that psychiatric inpatient care present. Looking at the existing research so far, combined with clinical experience, transdiagnostic applicability and flexibility in terms of time frame, context and level of functioning in the individual patient seem to be key factors for a potentially successful model. One way of developing such a model is to individually tailor each treatment, with the support of a few well establish principles. Some form of clinical functional analysis is fundamental for most treatment models based on cognitive behavior therapy (CBT), which is in general the form of therapy that has the broadest evidence base for the most common psychiatric conditions. Functional analysis is idiographic in nature, and takes the individual patient's problems as its starting point, no matter the diagnosis. It is a tool that, at least in theory, is applicable to all behavior. This means that even complex psychopathology such as psychosis or suicidality is potentially amenable to analysis and change.

The challenge with using psychological treatment in an inpatient context is to identify and treat some part of a complex condition, in a way that is possible to combine with other forms of care given on the ward. Functional analysis is a flexible tool in this respect. For example, it would be possible to formulate a meaningful plan for behavioral activation or limited exposure, in the context of a complex psychiatric condition such as schizophrenia.

Besides functional analysis as a basic tool, there are interventions from various therapy models that have been shown effective in an inpatient context. One CBT-based treatment model that has been investigated in an inpatient context with promising results is Acceptance & Commitment Therapy (ACT). In short, ACT aims to promote psychological flexibility in relation to disturbing experiences, for example symptoms of a psychiatric disorder. One of the central processes that ACT targets are so called values. Values are defined as freely chosen concepts that are connected to patterns of action that give a sense of meaning, and that are able to coordinate behavior over time. One example of such a concept might be being a present parent. This is connected to a pattern of various behaviors, with each one giving the person a sense of meaning, because they are connected to an overarching quality of action. Values, in this respect, are different from concrete and reachable goals. Values are not reachable but point to a direction for action more generally.

Generally, some form of maintenance plan is also central for change in psychotherapy to last over time. One method for constructing a maintenance plan is to summarize central principles in a form that the patient can take with them when treatment is completed.

Looking at the challenges that have been identified in terms of implementing psychological treatment models in a psychiatric inpatient context, and clinical experiences with using flexible interventions in this context, the purpose of the present project is to investigate the effect of a transdiagnostic short-term psychotherapy for psychiatric inpatients.

The main research question is:

Is individually tailored transdiagnostic short-term psychotherapy based on functional analysis, clarification of values, and the establishment of a maintenance plan effective in an inpatient context, in terms of causal change in personally meaningful problem areas defined by the individual patient?

Method Design The study is designed as an experimental sincle case study, with a multiple baseline across participants. Data will be collected during a baseline period (A) of approximately two days (2-3 data points will be collected daily) in order to establish a stable baseline of at least 5 data points. Thereafter, the intervention (B) will be introduced, and data will be continuously collected 2-3 times daily during treatment. Approximately 5 data points will be collected after treatment is complete. In order to control for measurement effects, the length of the baseline will be varied so that it will be prolonged for some participants.

Participants Approximately 5-10 participants will be included. The number needed is dependent on the variability of the dependent variable, the length of the baselines, possible trends in baseline and intervention phases, the size of the change that can be expected, and how fast the effect on the dependent variable can be reached after introducing treatment. Meta-analyses of the effect of psychological treatment for inpatients have shown small to medium effects. Even if the modality of treatment and diagnoses vary in studies included in meta-analyses, and compared to the project being described here, it seems reasonable when including patients to expect a quite limited (while clinically significant) effect. It is estimated that at least five patients will be needed to establish sufficient experimental control in the project being described here. However, more might be needed, or fewer if the baselines are stable enough, and the effect of treatment is clear.

Context Data will be collected on the inpatient wards of the Hospital of Västmanland, Västerås, Sweden, where the main researcher Mårten Tyrberg has his clinical position. In the hospital, there are 6 inpatient wards, each with 10-15 beds. Patients suitable for inclusion will be identified on routine rounds, where the main researcher participates regularly.

Outcome measures Outcome measures and dependent variables are the Personal Questionnaire (PQ), the Bull's-Eye Values Survey (BEVS) and objectively observable behaviors such as as-needed medication, amount of self-harm or degree of activity. PQ is a procedure for constructing reliable and personally relevant questionnaires for rating mental health. In an assessment interview, one or more so called illness statements are formulated, that describe the central problem with the patient's own words (for example, "I can't stand my anxiety"). For each illness statement an improvement statement is formulated, that is meant to describe some degree of improvement (for example "At times, I can handle my anxiety"), and a recovery statement that is meant to describe a situation where treatment is no longer needed (for example "I can handle my anxiety in a good way most of the time"). Every statement is paired with each other statement on a total of three cards. The order in which the statements are presented is varied on the three cards. The cards are presented for the patient, and for each card they get to choose the statement that best describes their present condition. 1-4 points are obtained for each illness statement, higher points representing worse mental health. BEVS is an instrument that measures, on a scale from 0-28, to what extent a patient is living according to their values. On a visually presented dart board, the patient marks with a cross somewhere on the board to indicate degree of values-based living, a cross closer to the middle (the bull's-eye) representing better health. BEVS has shown satisfying reliability and validity. Objectively observable behaviors such as use of as-needed medication can be tracked in the patients charts, degree of activity can be measured using an actigraph, and amount of self-harm can be tracked in patient charts.

Intervention, transdiagnostic short-term psychotherapy The treatment is a flexible application of clinical functional analysis and a few chosen interventions that in clinical experience have shown to work well with the patient group under study (psychiatric inpatients with complex symptoms), each being adjustable to suit the kind and degree of problems being presented. The first step is functional analysis and case conceptualization, in the form of a clinical interview. This is based on four focus questions that are helpful in conducting a quick analysis and isolating a treatable problem: What are you seeking? What have you tried? How has it worked? What has it cost you? The questions aim to focus in on the patient's main problem, as well as previous strategies that they have tried, in terms of their long- and short-term functionality, and in relation to central values. The second step is a simplified version of an experiential exercise called the lifeline. Potentially, this exercise can function as a concretization, where the findings of the functional analysis can be put into action, resulting in a plan for behavior change that is connected to personally meaningful life areas. The lifeline exercise is conducted as an exercise where both therapist and patient stand up and use the physical properties of the room to clarify life directions in a concrete way. Aids such as post-its are often used to concretize obstacles in one's valued direction, where key words describing symptoms are noted. According to clinical experience, this concrete way of working has been particularly helpful for patients with complex diagnoses, including some cognitive deficits, as it helps them register and remember central principles of the treatment (such as the need for acceptance in relation to experiences that can't be changed). The exercise has also proven adjustable to different kinds of problems. The third step is a repetition and a summary of the previous two steps, where an exercise called the pause is given to the patient, in the form of a printed card that the patient gets to keep. This is used as a maintenance plan. The exercise consists of the following steps: pause and slow your breathing; use the observer perspective; experience thoughts and feelings as they are; formulate a direction and act.

Data analysis Data will be analyzed using visual analysis, focusing on differences in outcome measures between A and B phases concerning central location, trend and variability. Statistical analysis will be used in the form of Kendall's Tau to analyze trends in A and B phases, and non-overlap of all pairs (NAP) to analyze differences between A and B phases.

Inclusion Criteria:

• Patient treated at any of the inpatient wards at Hospital of Västmanland, Västerås, Sweden, with any diagnosis or combination of diagnoses.
• Cognitive level such that informed consent can be given.
• A stable baseline with regards to the chosen dependent variable.

Exclusion Criteria:

• Other ongoing psychological treatment on the ward.
• Cognitive deficits that prevent the giving of informed consent.
• Unstable baseline.