The study takes place at the HT service in Lucerne and agglomeration (approximately 235,000 residents) offered by Lucerne Psychiatry since 2007. In 2020, 234 patients in the whole psychiatric spectrum were treated in their domestic environment. In average, treatment lasted for 41 days. At least once a day, patients are visited by one member of the interdisciplinary treatment team, which consists of 12 psychiatric nurses, 3 psychologists, and 4 medical doctors. At the beginning, for each patient a core team, consisting of 3 specialists from the respective professional groups, draws up an individual treatment plan together with the patient. Many of the daily visits are handled by a member of the core team. However, because the entire team works in shifts, patients are also visited by other team members. Through daily rapport, all team members are involved in the treatment ensuring treatment consistency.

Patients can participate in several group therapies at the inpatient hospital of Lucerne Psychiatry. A 24 h emergency service is provided by the team. In case of a psychiatric emergency during HT, a bed in the inpatient hospital of Lucerne Psychiatry is provided. Patients can stay there for up to 7 days with daily visits by the HT team. If inpatient care is needed for more than 1 week, patients are fully admitted to hospital. During HT, patients get a sickness certification with the possibility of starting a therapeutic work trial in their regular employment or in a protected work environment. Patients are supported in the search and initiation of a suitable follow-up treatment, e.g., outpatient psychotherapy.

The sample consists of patients assigned to the acute psychiatric HT in Lucerne. They are referred to HT by institutions and clinical practitioners. Minimal age for treatment is 18 years and patients with all psychiatric diagnoses are treated. As a prerequisite for treatment, patients must be able to distance themselves from acute danger to themselves or others. Furthermore, patients fulfilling the following inclusion criteria are eligible for the study: patients admitted to the examined HT, agree to participate voluntarily. Informed consent is given as documented by signature. Excluded are patients unable to follow the procedures of the study due to language problems. Study participation does not influence the treatment and participation is not compensated.

Patients are assigned to the acute psychiatric HT. Once the referral is received, the project leaders send a flyer and template of informed consent by mail to the assigned patients to inform them about the study. The mail is followed by a phone call to personally inform them about the study. Patients can be included in the study, if their verbal and written consent is obtained. Patients have time to decide whether to participate or not (at least 24 h before). Time to start of treatment varies depending on current capacity. In 2019 the average time to start of treatment was 8 days.

In case of participation, the link to the first set of questionnaires is sent by mail (paper versions are available if requested), which they fill out until the start of treatment (Fig. 1).

Flowchart of study procedures. BSCL Brief Symptom Checklist, WHO‑5 World Health Organization–Five Well-Being Index, PATHEV Patient Questionnaire on Therapy Expectation and Evaluation, SWE General Perceived Self-Efficacy Scale, EMOREG Fragebogen zur Emotionsregulation, PSTB The Bern Post Session Report, WAI-SR Working Alliance Inventory—Short Revised, Well-being self-constructed measurement on a rating scale from 1 (very bad) to 10 (very well) based on the common question in psychiatric practice about the current state of well-being

The study protocol and procedures were approved by the Swiss ethics committee Nordwest- und Zentralschweiz (EKZN: 2020-02736).

The first set of questionnaires aims to assess patient’ characteristics as potential predictors for treatment outcome (focal point of first research question). The following sociodemographic data are assessed: age, gender, social living situation, relationship status, highest degree, job situation before treatment, previous inpatient hospitalization, and previous HT. Symptoms are measured with the German version of Brief Symptom Checklist (BSCL) [26]. BSCL consists of 53 items rated on a 5-point Likert scale. Well-being is assessed with the five items of the German version of World Health Organization–Five Well-Being Index (WHO-5) on a 6-point Likert scale [27]. Therapy expectation and evaluation are measured with the German version of Patient Questionnaire on Therapy Expectation and Evaluation (PATHEV) [28], which consists of 11 items rated on a 5-point Likert scale. Ten items of self-efficacy are assessed with the German version of General Perceived Self-Efficacy Scale (SWE) [29] on a 4-point Likert scale and emotion regulation is measured with the “Fragebogen zur Emotionsregulation” (EMOREG) [30] with 26 items on a 6-point Likert scale. Clinical diagnosis is also included. It is carried out as part of the clinical assessment and subsequent treatment by specialists (psychiatrists, psychologists).

For the second focal point, therapeutic relationship is assessed with three items on a 7-point Likert scale of the German version of Bern Post Session Report (PSTB) [31]. Well-being is measured with a self-constructed daily measurement on a numeric rating scale from 1 (very bad) to 10 (very well) based on the common question in psychiatric practice about the current state of well-being. Both questionnaires are assessed daily. The profession and function (e.g., member of the core team) of the visiting team member as well as the type of the treatment (amount of emergency calls, participation in group therapy) for each participant visit is documented separately.

Working alliance is measured weekly with the German version of Working Alliance Inventory—Short Revised (WAI-SR) [24]. WAI-SR includes 12 items rated on a 5-point Likert scale.

For both research questions, BSCL and WHO‑5 are measured weekly throughout, at the end of treatment and 3 months following treatment. Changes in their scores over time represent the primary outcome as a measure of treatment success.

At the end of treatment patients fill out WAI-SR, EMOREG and SWE in addition to BSCL and WHO‑5. Three months after the end of the treatment, the project leaders contact the participants by phone. They receive a last set of questionnaires (BSCL, WHO‑5, EMOREG and SWE; Table 1 for the schedule of assessments).

A power-analysis was carried out to determine the sample size. We calculated with a F-test for analysis of variance (ANOVA), repeated measures, within factors, one group and three measurement-points (baseline, post and follow-up). The power analysis (G * Power 3.1.9.4) [32] showed a sample size of 43 for an effect size of f = 0.25 with an α error of 0.05 and actual power of 0.95. The effect size was chosen based on findings in psychiatric research [33].

In 2019, around 30% of the registered patients did not start treatment. In addition, the treatment duration was less than a week in 12% of the treatments. Finally, a discontinuation rate of 30% was anticipated for the follow-up according to previous literature [34, 35]. To compensate for these foreseeable losses, the final sample size was adjusted to at least 82. Drop-outs will be analyzed. Random missing values will be supplemented with the “multiple imputation” method, categorical missing values are supplemented by last observation forward as indicated.

A general linear mixed model (GLMM) [36] was initially planned. Experience with recruitment indicates that the planned sample size is unlikely to be achieved, which has implications for statistical model. Therefore, treatment success as main outcome is evaluated with a repeated measures analysis of variance (ANOVA [37]) model. Data will be analyzed by SPSS Statistics (version 26, IBM, Armonk, NY, USA) [38].

Treatment success is defined by the comparison of baseline and post measurements of symptoms (BSCL) and well-being (WHO-5) as dependent variables. Further, regression analysis is carried out to assess the predictive value of patient level characteristics (baseline level of BSCL and WHO‑5, SWE, EMOREG, sociodemographic data, diagnosis and PATHEV) as independent variables for treatment success as dependent variable.

Concerning the therapeutic relationship, two separate linear regression analyses will be carried out: one based on weekly measurements (WAI-SR) and one based on daily measurements (PSTB). The therapeutic relationship functions as the independent variable in both analyses. Treatment success as defined above and daily well-being act as dependent variables.

To explore the assumption of an association between the relationship to a treatment team in general (WAI-SR) and the relationship to the individual members of the team on each visit (PSTB), a linear regression analysis is carried out with the WAI-SR as a dependent and the PTSB as independent variable.