Participants and procedure
The study was conducted as a nationwide cross-sectional online survey among psychotherapists, both licensed and in training, across all therapeutic approaches recognized in Germany. The therapeutic approaches in Germany are: cognitive behavioral therapy (CBT), psychoanalysis (PA), depth psychology (DP), systemic therapy (ST). Inclusion criteria were (a) being a licensed psychotherapist or psychotherapist in training and (b) giving informed consent to participate in the survey. The study was approved by the ethics committee of the Catholic University Eichstätt-Ingolstadt (number: 122–2022). Data collection took place between February and April 2023.
Our recruitment approach aimed to reflect the reality of the German psychotherapeutic care system as accurately as possible. Therefore, all regional Psychotherapists’ Chambers (“Psychotherapeutenkammern”), in which licensed psychotherapists need to be registered, were asked to forward the online survey link to their members. After a follow-up, commitments from eight out of 12 chambers were received. To include psychotherapists in training we used a random sampling approach. Training institutions in each of the 16 federal states were asked to forward the survey to their trainees. Given the absence of an official comprehensive list of all registered training institutions in Germany, we made a concerted effort to compile a thorough inventory of training institutes across the federal states (up to January 2023). This was achieved by utilizing the websites of psychotherapist chambers (Bavaria, Berlin, Bremen, Hessen, Niedersachsen, Nordrhein-Westfalen, Saarland, Schleswig-Holstein). In cases where lists were outdated or unavailable, additional searches were conducted on the official websites of states (Baden-Württemberg, Brandenburg, Hamburg, Mecklenburg-Vorpommern, Rheinland-Pfalz, Sachsen) as well as the German Association of Psychotherapists (DPtV, Sachsen-Anhalt. Thüringen). The final list comprised 271 institutions across all federal states and therapeutic approaches.
In each of the federal states, a random selection of 10% (in total, n = 33) of the institutions was contacted to forward the survey to their trainees. In case an institution denied distributing the survey to their trainees, another institution for this federal state was randomly selected. At the end of the recruitment process, a sum of 45 training institutions was contacted, of which 27 distributed the survey link to their trainees. The 10% target could not be achieved in 5 federal states (Bavaria, Bremen, Hessen, Nordrhein-Westfalen, Rheinland-Pfalz). In addition, all regional associations of statutory health insurance physicians (“Kassenärztliche Vereinigungen”) and three professional associations of psychotherapists that operate across therapeutic approaches and throughout Germany (“Berufsverbände”, Association of Psychological Psychotherapists in the professional association of German psychologists, BDP-VPP; Federal Association of Contract Psychotherapists, BVVP; German Association of Psychotherapists, DPtV) were requested to distribute the survey. Two out of three requested professional associations published the survey link on their homepage and five out of 17 regional associations of statutory health insurance physicians (Bremen, Hamburg, Niedersachsen, Westfalen-Lippe, Thüringen) forwarded the survey to their members. Members of the national bodies could be licensed psychotherapists and psychotherapists in training. Members of associations of statutory health insurance physicians were licensed.
All contacted institutions received detailed information about the study by phone and e-mail and distributed the survey information, link, and QR-Code electronically via e-mail, internal newsletter, and/or a notice on their homepage and intranet. Four training institutions placed an announcement (printed version of the tender text) on their bulletin board.
A total of 624 psychotherapists clicked on the survey link, of whom 51 denied consent or did agree and dropped out before answering to the items for experience regarding patients with climate change-related concerns. Thus, we analyzed the responses of the remaining participants (N = 573). Dropouts after the consent page were not excluded from subsequent analyses as participants dropped out at different stages of the survey, and itemwise analyses were conducted (see 24, for a similar approach). To ensure the robustness of this approach, we contrasted participants with more and less than 10% missing values across the survey on all items. This yielded no significant differences in terms of therapists’ characteristics, experiences with patients with climate change-related concerns, or views on the topic.
Measures
The survey comprised 37 items, of which 24 items administered to all participants and 13 items (focusing on climate change-specific reactions) presented only to those reporting that they had already treated patients with climate change-related concerns (i.e., therapists with experience). Items were newly developed for this study and based on a large-scale survey on experience, attitude, and knowledge of MHPs with climate change topics raised by their clients [24]. To ensure comprehensibility and relevance of all items, the survey was piloted by five psychotherapists in training before circulation. The survey was provided online using the survey tool Qualtrics. The complete survey is presented in Appendix A.
At the beginning of the survey, socio-demographic and work-related information about the participants was collected in ten items about: age (year of birth), gender (female/male/diverse), level of training (trainee vs. licensed), therapeutic approach (CBT, PA, DP, ST, other), work experience (number of years working as a therapist including time as trainee, number of weekly treatment sessions), and practice setting (private practice, hospital, outpatient clinic, other). Additionally, engagement in climate or pro-environmental advocacy groups was assessed dichotomously. The degree of pro-environmental behavior in everyday life was assessed on a 4-point scale (1 = in no area of everyday life, 4 = in almost all areas of everyday life). Thereafter, participants were assigned to one of two paths, depending on whether they had already encountered patients expressing climate change-related concerns. Path A (for therapists with experience) collected information about the number of such patients (seen in the last 12 months), the patients’ socio-demographic characteristics as well as their cognitive, emotional, physiological, and behavioral reactions. Predefined answer options for cognitive styles (e.g., rumination), physiological (e.g., racing heart) and behavioral reactions (e.g., crying) were formulated according to our current knowledge of human stress response and the recent literature on climate change-related emotions [17, 19, 26]. In addition, their expression of feelings related to climate change-related concerns were collected in free-text format. In path B, therapists without experience were asked whether they expected to encounter patients with climate change-related concerns in the future. In the first two parts, for most items multiple responses were allowed (therapists’ practice setting and therapeutic approach; patients’ age in years, educational degree, assumed family status, most frequent assigned diagnoses, cognitive styles, feelings, physiological and behavioral reactions). In the last part of the survey, all participants (therapists with and without experience) answered 12 questions regarding their views on climate change-related concerns (a) in relation to mental health, (b) on how to deal with them in therapy and (c) whether they felt well equipped or wished for additional training and resources on the topic. Therapists answered on a 4-point scale (1 = I do not agree at all, 4 = I fully agree).
Statistical analysis
All analyses were performed itemwise because the survey did not employ a forced choice format and participants dropped out at different stages of the survey. This means that we included the number of participants who had answered the respective items (indicated by n/N for all frequencies reported; see 24, for a similiar approach).
Descriptive statistics were used to describe items presented in the three parts of the survey, using frequencies or mean values. The free text answers related to the patients’ feelings were mostly given in one word per option (e.g., anxiety, fear, anger). The answers were categorized inductively [27] and analyzed descriptively.
For contrasting therapists with and without experience with regard to their views on climate change-related concerns in therapy, all items presented in the last part of the survey were dichotomized in 0 = disagreement (on the 4-point scale: 1 = I do not agree at all, 2 = I rather disagree) and 1 = agreement (on the 4-point scale: 3 = I rather agree, 4 = I fully agree). Group differences were computed using chi-square test, t-tests, and Mann-Whitney-U-test, depending on the type of data. All tests were two-tailed with α = 0.05. Bonferroni-Holm correction was performed within each thematic group of items asking about therapists’ views (i.e., views on consequences of climate change-related concerns for mental health, views on how to address climate change-related concerns in therapy and views on required resources for addressing climate change-related concerns in therapy). Data were analyzed using SPSS statistics, version 28.