Introduction

The global burden of mental disorders remains high [1]. Mental disorders often remain undetected or untreated, particularly in patients with mental–somatic multimorbidities [2, 3]. One possibility to overcome this gap are stepped and collaborative care models (SCCMs). The main idea of stepped care is to identify and deliver the least invasive, but most effective treatment, and then stepping up treatment if disease burden reaches a specific threshold [4]. Collaborative multi–professional care has been shown to be an important element for appropriately handling mental [5, 6] and somatic illnesses [7] in patients with mental–somatic multimorbidities.

SCCMs combine these two concepts and have been introduced in various countries and health care settings. The heterogeneity of the SCCMs implemented in a range of countries coupled to contextual specificities, does currently not allow concluding on standard and best implementation modalities of SCCMs for managing mental health problems. Yet, a successful embedment in routine health service provision requires a range of aspects to be considered as observed in implementation research. Implementation research gives insights into the factors affecting the implementation in a real world setting [8]. For instance, in a German project a major difficulty to implement a SCCM model within primary care was collaboration across a regionally wide-spread network [9]. Co-location of somatic and mental health specialists in the same working place [10, 11] is valuable to reach integrated care [12] and can help handling mental health conditions in somatic settings. Further, general hospitals in the United Kingdom successfully implemented routine depression and anxiety screenings in some specialties [13]. However, structural factors, such as the ward organization or staff availability, were factors impeding the mental health screening [13].

Here, we report insights from a recent project conducted in a general hospital setting in Switzerland. SomPsyNet is a healthcare project for SOMatic inpatients to prevent PSYchosocial stress consequences by establishing a stepped and collaborative care NETwork in the canton of Basel-Stadt [14, 15]. The SCCM implemented by SomPsyNet aims to improve the quality of life of patients with mental–somatic multimorbidities [14, 15]. First, ward physicians and nurses as well as patients themselves independently assess the patient’s psychosocial distress using a distress thermometer, which has been adapted from the commonly used distress thermometer in oncology [16]. The score is recorded in each patient’s electronic file; hospital information technology (IT) systems were adapted accordingly. Second, all patients are comprehensively screened for depressive, anxiety, and distressing somatic symptoms using validated and reliable assessment tools. The two-step screening shall be completed within the first 72 hours of hospitalization. If patients are distressed according to the two-step screening, they are offered a psychosomatic consultation providing them clinical assessment and appropriate treatment recommendations according to stepped care. To support the implementation of the first step of the SCCM, each ward had one face-to-face training. Additionally, an online training course was created to introduce SomPsyNet and the SCCM to healthcare professionals.

Organizational changes and the introduction of new systems such as the SCCM are challenging regarding various aspects. Two systematic reviews have summarized structural, financial, and individual barriers to integrating mental health in primary care settings [10, 17]. Leadership, reimbursement, and motivation represented important determinants of integration success [10, 17]. Yet, evidence of the facilitators and barriers encountered when SCCMs are implemented in hospital settings is lacking. Thus, we aimed to assess the facilitators and barriers of the psychosocial distress assessment implementation, which represents the first step of the SCCM in Basel-Stadt. We structured this study around a framework widely used in implementation research, the Tailored Implementation for Chronic Diseases (TICD) framework.

Methods

Data analysis

Content analysis was done in NVivo 12 [27] using the framework method by Gale et al. which conceives the analysis of qualitative analysis along seven steps: 1) transcription, 2) familiarization with the interview, 3) coding, 4) developing a framework, 5) applying the framework, 6) charting data into a framework matrix, and 7) interpreting the data [28]. After transcription of the interviews, NJA familiarized herself with the content by multiple reading of the transcripts. Then, NJA deductively coded the interviews by applying the TICD framework. The TICD domains were split into determinants (themes and subthemes) that were used for line-by-line coding. Decisions were made using the definitions of TICD determinants provided by Flottorp et al. [22]. Fig 1 displays an example of the subtheme “compatibility” within the TICD domain “professional factors”. Example quotes for each TICD determinant mentioned by the interviewees are presented in S1 Table. Afterwards, the data was charted in a data matrix. A summary of each TICD determinant (column) was written for each interviewee (row) and linked to illustrative quotes. This data matrix helped to interpret the data and write a memo for each TICD domain including the identified facilitators and barriers. Additionally, detailed researcher notes and discussion with NJA’s supervisor (KW) supported the analysis and interpretation of the interviews and thus, the rigor of the research, during the analysis. Due to the limited time resources of the interviewees, we did not share the transcripts or findings with the interviewees. NJA translated the example quotes from German to English. The Consolidated Criteria for Reporting Qualitative Research (COREQ-32) guided the reporting [29].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Example of the analysis applying the TICD framework.

https://doi.org/10.1371/journal.pone.0285395.g001

Results

Ten nurses and eight physicians participated in the interviews. Three nurses knew the interviewer (NJA) from a first interview in 2020 while 15 interviewees, replacing others who could not be re-interviewed, did not know her. Table 1 shows average interviewee characteristics. The mean age of the interviewed nurses was higher than the one of the physicians, which translates into a longer professional experience. Further, we conducted more interviews with female than with male nurses who worked in various departments at the three included hospitals. The interviews lasted on average 29 min (range 20–44 min).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics of the interviewees, affiliated institutions, and interview duration (N = 18).

https://doi.org/10.1371/journal.pone.0285395.t001

Interviewees mentioned facilitators and barriers in all seven TICD domains. Capacity for organizational change and social, political, and legal factors were discussed less comprehensively than factors from other domains. Table 2 presents an overview of the facilitators and barriers cited by the interviewees. We provide a summary of all TICD domains in the following paragraphs. Additionally, S1 Table shows example citations of each determinant.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Overview of facilitators and barriers within the Tailored Implementation for Chronic Diseases (TICD) domains.

https://doi.org/10.1371/journal.pone.0285395.t002

Discussion

This implementation study explored different facilitators and barriers related to the introduction and operation of a psychosocial distress assessment by physicians and nurses within a SCCM in a Swiss general hospital setting. While integrating the assessment in preexisting IT systems and daily processes (e.g., admission interview) at the hospital supported the introduction, major barriers were identified in the domains of guideline factors, individual healthcare professional factors, and incentives and resources.

Integrating patients’ psychosocial distress assessment into preexisting hospital processes and IT systems reduced additional staff efforts. Integrated IT systems also fostered the collaborative care implementation in somatic health settings in previous studies [11]. The possibility to tailor processes to the specificities of wards may increase the motivation to assess patients’ psychosocial distress. Adaptability enhances commitment, and thus, sustainability [30]. Further, leadership support fosters commitment of staff members [11] as observed in our study.

Interviewees questioned the usefulness of a psychosocial distress assessment tool that is based on subjective judgment and “gut feeling”, although intuition plays an important role in clinical settings [31], especially for nurses [32]. The intuitive conclusions need to be corroborated by objective assessments as proposed by a previous study [31], which is important, as shown in delirium research, where subjective assessments are associated with misclassifications [33]. This explains the interviewees’ desire to increase objectivity and standardization with examples, checklists, or questionnaires and thus, support the psychosocial distress assessment.

Additionally, insufficient knowledge about and awareness of mental health has been described in several studies [17, 21]. Mainly physicians mentioned lack of awareness of and familiarity with the psychosocial distress assessment and the entire SCCM. One important reason for particularly junior physicians highlighting this may be their frequent rotation in general hospitals, which made adequate information about the assessment and the SCCM a challenge. This may hinder its sustainability as observed in another implementation study in mental health [34].

Lastly, time constraints are widespread when implementing integrated care [10, 17, 35], and healthcare professionals face many competing priorities. Typically, the focus on somatic health is to the detriment of mental health conditions [10, 17, 21]. The psychosocial distress assessment was perceived to be of low priority in general hospitals, negatively affecting the SCCM implementation including the patient assessment. This might be reinforced by missing observability of the assessment’s benefits on healthcare professionals’ daily work or the senior physician’s potential lack of interest.

Strengths and limitations

This study adds value to implementation science in general hospital settings by highlighting important facilitators and barriers of a time-constraint setting that have to be accounted for when implementing routine psychosocial distress assessment. The inclusion of three different hospitals and different wards increases the generalizability of the findings. However, the findings may not apply to wards with short hospital stays, such as surgical wards or emergency departments.

Some limitations need to be considered when interpreting the findings. First, the recruitment strategy may have led to limited views on the SCCM and its first step, the psychosocial distress assessment. The view of participating nurses and physicians may differ from other healthcare professionals’ views. Nonetheless, the interviewees mentioned major critical factors. Second, interviewees may give socially desirable answers. Letting healthcare professionals choose the interview location and mentioning the opportunity to suggest improvements may have reduced this bias. Further, some interviews were of comparatively short duration due to multiple competing tasks of the interviewees, limiting the possibility to cover topics of interest in-depth. To counteract this, we shared topics of interest and the interview guide ahead of the interview. Third, not the entire SCCM had been implemented at the time of the interviews. Only one physician had initiated a consultation with a mental health specialist because of the SCCM, at the time of the interview. Therefore, we can only conclude on the subjective psychosocial distress assessment by the healthcare professionals and not on subsequent consultations for mental conditions. Fourth, this study only includes the healthcare professionals’ opinions and not those of patients, mental health specialists, or hospital management. Insights on the patient factors, capacity for organizational change, and social, political, and legal factors domains are limited. Finally, we used a deductive approach, which allowed structuring the analysis of facilitators and barriers as observed by other implementation research. This reduces the likelihood of identifying new themes. However, the analysis captured new factors such as the turnover rate relevant in acute care settings already mentioned by others [23].

Implications for practice and future research

Based on the Expert Recommendations for Implementing Change (ERIC) compilation [36], we propose several approaches to overcome the three most important barriers observed in our psychosocial distress assessment implementation as part of an SCCM in general hospital settings.

Checklists and examples should be made available so to make psychosocial distress assessments more consistent across healthcare professionals. For the assessment to be sustainable, healthcare professionals need to be aware of the benefits to their patients and why a psychosocial distress assessment is necessary [10, 30, 37], even if it is subjective. This can be achieved through regular personal training, e.g., offered for healthcare professionals starting a new position, particularly important in settings with a high turnover rate. The training should clearly demonstrate the evidence base and the required action to assess psychosocial distress. Clear guidance coupled with repeated training may promote standardized approaches and thus, reduce subjectivity in the assessment. Additionally, readily available online training should be easily accessible, for instance by linking the training to the assessment in the patient file. Educational strategies are widely used [38] and seem to positively affect care, patient health, and health systems in nursing [39] and particularly when implementing collaborative care into somatic health care settings [11]. Further, reminders integrated directly to the IT system increase the awareness and save time.

In this time sensitive setting, the interviewees reported low priority of the patient’s psychosocial distress assessment. Especially physicians emphasized thereby the importance of the senior physician’s interest in mental health. Senior physicians should act as champions or opinion leaders who support the implementation and positively influence the SCCM uptake. Other research saw that physician champions helped the implementation of collaborative care into primary care [10].

The results of this study were shared and discussed with the SomPsyNet project team. They will complement quantitative data, e.g., the percentage of patients who were assessed for psychosocial distress by physicians and nurses, and will altogether support tailoring the SomPsyNet intervention to implementation realities. For instance, the provision of the psychosocial distress assessment could be restricted to only one healthcare professional.

Future research should focus on three main aspects. First, staff turnover was important in an emergency department study using the TICD [23]. We agree with the authors that this is a major determinant in a hospital setting in general. Future implementation research should therefore focus on how implementation can be guaranteed despite high turnover rates and how barriers related to the turnover rate can be sustainably overcome to better recognize mental health and see SCCM as part of hospital operations.

Second, we found that literature on how healthcare professionals perceive subjective assessments of patients’ mental health is scarce. While our findings show that objective assessments are preferred to subjective ones, we suggest investigating experiences of different healthcare professionals in different settings to better understand the reasons for this.

Third, other stakeholders’ perspectives are essential, particularly those of patients. Other research has found that patients wish to separate mental and somatic health spatially [11, 40]. An additional barrier for psychosocial distress assessment was the perceived stigmatization [41]. These factors may impede the mental health project implementation in general hospitals. Hence, research should focus on patients’ experiences with and preferences about integrating mental health into somatic settings.

Conclusion

Subjectivity of the assessment, lack of awareness due to high turnover rates, and low priority of the assessment due to time constraints posed major challenges when implementing a psychosocial distress assessment of patients admitted to general hospitals, especially for physicians. We suggest providing regular training for new employees, providing feedback on healthcare professional performance and the observed patient benefits, and appointing champions or opinion leaders to address these challenges in implementing a routine assessment for mental health in general hospitals. Furthermore, the integration of the psychosocial distress assessment into current hospital processes and systems is a necessary factor to facilitate the implementation in a time-constrained setting.

Supporting information

Acknowledgments

We thank Jana Gerold for her methodological support and all the interviewees for taking their time to participate in the study and giving us insights into their experiences. We are grateful for the valuable support of the SomPsyNet Consortium. Additionally to the authors of this manuscript, the SomPsyNet Consortium members are:

Marco Bachmann (Department of Psychiatry and Psychosomatics, Bethesda Hospital, Basel, Switzerland), Gabriele Bales (University of Basel, Geriatric Psychiatry, Department of Geriatric Medicine FELIX PLATTER, 4055 Basel, Switzerland), Klaus Bally (Centre for Primary Health Care, University of Basel, Switzerland), Stefano Bassetti (Division of Internal Medicine, University Hospital and University of Basel, Basel, Switzerland; Department of Clinical Research, University Hospital and University of Basel, Basel, Switzerland), Reto Baumgartner (Social Insurance Institution Basel-Landschaft, Binningen, Switzerland), Johannes Beck (Clinic Sonnenhalde, 4125 Riehen, Switzerland), Virginie Bourquin (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), David Büchel (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Andreas Dörner (St. Claraspital, Medical clinic, Basel, Switzerland), Lukas Ebner (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Jennifer Erb (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Peter Ettlin (Foundation Rheinleben, Basel, Switzerland), Elvira Fasel (Department of Psychiatry and Psychosomatics, Bethesda Hospital, Basel, Switzerland), Noélie Fiechter (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Lavinia Flückiger (Department of Health Canton Basel-Stadt, Division of Addictions, Basel, Switzerland), Johanna Fremmer (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Alexander Frick (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Simon Fuchs (Department of Health Canton Basel-Stadt, Medical Services, Basel, Switzerland), Lavinia Giamboni (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Florian F. Grossmann (Department of Medicine, Division of Nursing, University Hospital Basel, Basel, Switzerland), Anja Hermann (Department of Medicine, Division of Nursing, University Hospital Basel, Basel, Switzerland), Matthew Hotopf (Department of Psychological Medicine, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, United Kingdom), Christian G. Huber (University Psychiatric Clinics (UPK), Department of Psychiatry and Psychotherapy, Basel, Switzerland), Lydia Isler-Christ (Sevogel-Apotheke, Basel, Switzerland; Baselstädtischer Apotheker-Verband, Basel, Switzerland), Christina Karpf (Department of Health Canton Basel-Stadt, Division of Prevention, 4001 Basel, Switzerland), Maria C. Katapodi (Department of Clinical Research, University of Basel, Basel, Switzerland; University of Michigan School of Nursing, Ann Arbor, MI USA), Robert C. Keller (Swiss Heart Foundation, Bern, Switzerland), Sabrina Klimmeck (University Hospital of Basel, Basel, Switzerland), Melinda Kress (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Yvonne Künstle (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Undine E. Lang (University Psychiatric Clinics (UPK), Department of Psychiatry and Psychotherapy, Basel, Switzerland), Yasmin Liechti (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Cécile Longoni (Gyn. Social Medicine and Psychosomatics, University Hospital and University of Basel, Basel, Switzerland), Sherado Mazander IV-Stelle Basel-Stadt, Basel, Switzerland), Daria Meier (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Alexander Minzer (Swiss Academy for Psychosomatic and Psychosocial Medicine (SAPPM), Reiden, Switzerland), Lara Riedo (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland; Department of Health Canton Basel-Stadt, Division of Prevention, 4001 Basel, Switzerland), Francisca Schiess (Centre of Self-Help Basel, Basel, Switzerland), Lisa Schiess (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Felix Schirmer (Vereinigung der psychosomatisch tätigen Aerztinnen und Aerzte der Region Basel, Basel, Switzerland), Nadine Schur (Institute of Pharmaceutical Medicine (ECPM), University of Basel, 4056 Basel, Switzerland), Peter Schwob (Psychotherapists Association of Basel VPB, Basel, Switzerland), Sonja Seelmann (Division of Internal Medicine, University Hospital and University of Basel, Basel, Switzerland), Gayoung Son (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Thomas Steffen (former cantonal medical officer; Department of Health Canton Basel-Stadt, Medical Services, Basel, Switzerland), Friedrich Stiefel (Liaisonpsychiatrischer Dienst, University Hospital Lausanne, Lausanne, Switzerland), Marion Tegethoff (Institute of Psychology, RWTH Aachen University, Aachen, Germany), Shannon Timoney (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Corinne Urech (Gyn. Social Medicine and Psychosomatics, University Hospital and University of Basel, Basel, Switzerland), Thomas von Allmen (Department of Health Canton Basel-Stadt, Health Care, Basel, Switzerland), Lilly-Sophie Walzer (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Sybille Werner (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Andrea Wetz (Rheumaliga beider Basel, Basel, Switzerland), Dragana Weyermann (Patientenstelle Basel, Basel, Switzerland), Raffaela Widmer (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Viktoria Yarkova (St. Claraspital, Medical clinic, Basel, Switzerland), Christoph Zäh (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland), Diana Zwahlen (Department of Psychosomatic Medicine, University Hospital and University of Basel, 4031 Basel, Switzerland).