Current Practices and Barriers of Family-focused Care of Patients with Severe Mental Illness and Their Children: A Survey Among Czech Psychologists and Psychiatrists

This study was conducted after obtaining approval from the ethics committees of Masaryk University (EKV-2021-108). The survey was anonymous, and respondents were informed about the study, privacy, and terms at the beginning of the survey. The respondents filled out the questionnaire after providing informed consent.

We created a study questionnaire to capture the specific context of the Czech healthcare system. The development of the questionnaire followed a multi-step process (see Figure 1).

FIGURE 1:

Process of development of the questionnaire.

First, relevant literature was reviewed, including studies on family-focused practices and existing research on the attitudes and barriers experienced by mental health professionals working with adult patients (21,22,23,24,25,26). Inspiration was taken from the Family-Focused Mental Health Practice Questionnaire (27). We also considered the legal and institutional specifics of the Czech healthcare system, including the structure of training and specialization.

Second, we formulated an initial set of items covering five thematic areas: demographic and work characteristics, current practice, personal attitudes, perceived barriers (professional, systemic, and organizational), and recommendations for practice.

Third, the 31-item version of the questionnaire was reviewed by the research team and pilot-tested with four mental health professionals (two psychologists and two psychiatrists) from different clinical backgrounds.

Fourth, based on the pilot feedback, several items were revised for clarity and three new questions were added to the demographic section. This resulted in the final 34-item version of the questionnaire, which included both factual items (e.g., demographic and work-related characteristics) and multi-statement scales targeting attitudes, perceived barriers, and recommended practices.

Different types of response formats were used depending on the character of the information being collected. This choice was informed by existing literature and aimed at ensuring that both subjective attitudes and objective characteristics could be appropriately captured. A detailed description of the specific response formats is provided below. The number of questions associated with each thematic area is summarized in Table 1. However, only a subset of items relevant to the study aims is reported and analyzed in this paper.

For the questionnaire topics, see Table 1.

Most questions in the demographic and work characteristics section used fixed-response formats such as multiple choice or categorical options. The current practice section included questions about frequency, and the answer options were: Always (in 75–100% of cases), Often (in 50–75% of cases), Sometimes (25–50%), Rarely (5–25%), and Never (0% of cases). One question was in the multiple choice format (“Which family member do you discuss your patient’s illness with?”). The section of personal attitudes and perceived barriers contained 21 items with five-point Likert-type questions (1 – agree, 2 – rather agree, 3 – neither agree nor disagree, 4 – rather disagree, 5 – disagree). One question was in a yes-no format (“Would you like to see a change in the legislation framework according to the Norwegian model?”) and one in multiple-choice format (“Why has the child not been invited to a joint meeting with you?”). The recommendation for practice section had five-point Likert-type questions and multiple-choice questions.

Based on the National Register of Health Service Providers, we contacted via email all healthcare facilities that provide psychiatric or clinical psychological care to adult patients with mental disorders (this included representatives from large psychiatric hospitals as well as small outpatient clinics or individuals). A total of 1100 medical institutions and private practices were sent a request to participate in the study during February 2023. The follow-up set of requests to increase the respond rate was sent in the 1st week in April 2023, and the data collection ended on April 21, 2023.

Quantitative data were analyzed using SPSS 27 software with descriptive statistics.

We conducted exploratory factor analysis (EFA) to examine the underlying structure of the questionnaire and to assess whether the data supported the theoretically proposed thematic groupings. The initial item allocation was based on prior literature and conceptual considerations and was refined based on the EFA results.

Specifically, EFA was conducted on a subset of 21 items related to attitudes and perceived barriers in order to identify underlying dimensions within the data. Since no standardized instrument was available in the Czech context, EFA was selected as an appropriate method for uncovering latent constructs based on participants’ responses. Prior to conducting EFA, the suitability of the data was assessed using the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity. EFA was conducted using Principal Component Analysis (PCA) with Varimax rotation (Kaiser normalization). Six components with eigenvalues above 1.0 were extracted, and factor retention was guided by the scree plot and interpretability. However, the sixth component was weakly defined and lacked theoretical interpretability; thus, the final solution retained five components that were both statistically robust and conceptually meaningful. Factor loadings of ≥ .40 were considered significant, and items with high cross-loadings were reviewed for possible exclusion. The five retained components explained 63.0% of the total variance and reflected a coherent factor structure aligned with the underlying constructs of attitudes and perceived barriers related to discussing parenting in mental health care.

The components were labeled based on the thematic content of the highest-loading items, informed by both statistical outcomes and theoretical relevance. Two components captured attitudes and beliefs (Awareness of the negative impact of SMI on children and Workforce attitudes regarding involving children), while three reflected perceived barriers (Fear of losing rapport, Feeling incompetent to work with children, and Feeling confident in parenting topics). The use of Varimax rotation, which assumes orthogonal (i.e., uncorrelated) factors, was supported by both the statistical results and the theoretical distinctiveness of the identified dimensions.

Nonparametric tests, specifically the Mann-Whitney U tests, were used to detect differences in attitudes between respondents based on professional role and parental status, as the data were primarily ordinal (e.g., Likert-type scales) and did not meet the assumptions required for parametric tests (such as normal distribution). We approached the investigation of differences based on respondents’ professional roles and parental status drawing on insights from previous international studies (e.g. 18, 22). Due to the limited sample size, however, we refrained from conducting more granular analyses, such as those segmented by therapeutic approach. Where no statistically significant difference was found, we chose not to include detailed statistical results in the text to maintain readability and focus on the main findings.

In the tables, means and standard deviations were reported to summarize the central tendency and variability of responses, which facilitated concise comparison across groups. In contrast, percentages for selected items were presented in the text to provide a more intuitive and accessible summary of key findings. While we acknowledged that Likert-type responses are ordinal, reporting means and SDs is a common practice in survey research and offers a compact way to convey group-level trends.

A total of 193 individuals responded and completed the questionnaire, of whom 51.8% were psychiatrists and 48.2% were psychologists. Most respondents were women (75.1%), and 77.2% were parents. The mean age of respondents was 47.5 years (SD = 11.4) and varied from 25 to 77 years. Detailed demographic information is seen in Table 2.

This section presents findings on how professionals typically identify parental status, communicate with patients about their children, and involve family members in the treatment process. These areas were assessed through five questions, comprising a total of seven items used in the analysis.

The results of two subscales derived from the exploratory factor analysis (Awareness of the negative impact of SMI and Workforce attitude regarding involving children) and an additional question on policy improvement are presented here.

In this study, almost all respondents reported that they regularly asked whether their patients have children. These findings are consistent with the results of our previous study based on medical records related to the identification of parental status in patients with SMI (3). The results from the Czech Republic indicate relatively good practice in identifying minors (95.9% of professionals routinely inquire), which is notably more frequent than that reported in international studies. The results of parental status identification vary in the international literature, for example, in a UK audit of 100 adult mental health cases, 62 were asked if they had dependent children, whereas, in more recent research, three-quarters of professionals were routinely asked if the service user had children (29, 18). Similarly, a Norwegian study reported that 56% of health professionals did not identify patients’ minor children, and a five-year follow-up study in the same clinic found that 28% of health professionals did not identify patients’ minor children (30).

However, asking about parental status does not imply that patients are asked about their needs and specifics in practice. For example, among multidisciplinary adult mental health practitioners in the UK, less than half of the professionals routinely ask patients about the quality of their relationship with their children (39.6%) or whether their children have emotional or behavioral difficulties (29.5%) (18). In the current study, more than half the respondents said that the topic of parenting was raised by patients during sessions, while 12.4% said that patients did not discuss it with them. It would be interesting to explore in more detail the barriers that patients face when not discussing parenting with professionals. If parents themselves were able to explain their illness to their children, it would help prevent negative effects on children (45). However, patients usually do not have the same awareness of the negative impact of the illness on their children, may lack communication skills, fear that the information will jeopardize their relationship with their child or take away their ability to care for their child (31), and sometimes have mixed feelings about receiving support with parenting (32). Therefore, it is the professionals who must educate patients and children about the implications and raise the issue with patients. In this study, three-quarters of respondents had advised parents on how to communicate with their children, felt fairly confident (83.4%) in advising on how to talk to patients’ children, and were comfortable (91.2%) in discussing parenting (parents were more confident than non-parents). Lack of time for parenting topics was reported by only 14.5% of respondents (psychiatrists more often than psychologists). They also do not think that the topic of parenting would disrupt the therapeutic alliance. We hypothesize that the greater certainty in parenting themes is related to the fact that professionals identify parenting conventionally, and perhaps see parenting topics as part of the patient’s treatment, but usually only when the topic is raised by the patients themselves (65% discuss parenting after patients bring it up) – same as reported by other studies (33, 34, 35).

The majority of respondents believed that COPMI are at increased risk of developing mental health problems and more than half believed that a parent’s SMI will always negatively affect the child. Three-quarters of respondents believed that involving children in care can be helpful, in contrast to a quarter who considered it inappropriate. This puts our findings in line with other studies where professionals have had generally positive attitudes toward the involvement of children in the treatment of adult patients (35, 36).

However, positive attitudes toward involving children in treatment are at odds with current practices. In this study, although respondents felt fairly confident about working on parenting issues with patients themselves and felt there was sufficient time in sessions to do so, they felt less confident about working directly with children (only 48.2% felt confident) and only a quarter of respondents had ever invited a child into a session. Yet only 14% of respondents felt that the potential inclusion of a child would disrupt the therapeutic relationship with the patient. This discrepancy may be related to structural and personal barriers, such as limited training in child-focused methods, uncertainty about appropriate ways to involve children, or a lack of institutional emphasis on family-focused practice. This trend is also evident abroad: research in Finland found that although nurses regularly collected information about clients’ children, most nurses did not regularly meet those children (37). In a Norwegian study, 5 out of 9 therapists from a psychiatric clinic invited the children, citing uncertainty, fear of increased workload, or more complex treatment as reasons for less involvement (38). In this study, the most common reasons for never having invited a child to a joint session were: outside area of expertise (35%), not thought of it (28.9%), patients do not want it (19%), lack of training (18.3%), and finding it inappropriate (15.5%). According to Haukø and Stamnes, it appears socially unacceptable for therapists treating adults to investigate children’s circumstances during treatment, as adult psychiatric care is primarily focused on the individual patient’s mental health (38). However, supportive interventions for families are limited when children are removed from adult care (37). In addition, we know that children want to be included in the care themselves and want to be informed about their parent’s treatment (39). Maybery and Reupert state that although professionals find the inclusion of children in care useful, these same professionals also assume limits in skill and knowledge (22). In this study, professionals perceived a lack of experience in working with children, a lack of appropriate procedures for working with families, a lack of time, inadequate financial compensation from health insurance companies, a lack of training, and taking on too much responsibility.

Mayberry and Reupert, in their hierarchical model of barriers, rank organizational support, such as policies focused on family and child screening, assessment guidelines, and protocols, at the very base (22). More than half of the respondents in this study would welcome a change in legislation to allow greater involvement of children, which would set the foundation for systemic change in our country. Gregg et al. conclude that perceptions of practitioners, personal attitudes, perceived organizational support, and skill and knowledge are more influential than service-based factors (23). From the available data, we cannot draw conclusions about which barriers are more influential; what we ultimately find important anyway is the need to influence both personal (skill and knowledge, etc.) and organizational (policy settings, legislation, reimbursement from insurance companies) barriers at the same time, because in both cases professionals perceived a limitation to providing more care to COPMI.

Concerning experience in working with children, the training of psychiatrists and clinical psychologists in the Czech Republic is differentiated (child certification is a superstructure after clinical psychology and specialization in psychiatry), which results in a very demanding training and contributes to a shortage of professionals in this area. A shortage of professionals was perceived as a problem by 80% of respondents, and 71% recommended greater coherence between child-oriented and adult-oriented services. Similar systemic barriers have been described in other countries as well. For example, studies from Norway and Australia have reported that the lack of formal procedures, insufficient collaboration between adult and child services, and time constraints are common obstacles in addressing the needs of COPMI (22, 24). In the Czech Republic, the reported lack of referral options and perceived shortage of specialists may reflect the demanding qualification requirements for professionals working with children and families, which can limit workforce capacity. These international comparisons support the relevance of our findings and highlight areas where organizational improvements and resource allocation may be needed. In this context, the fact that only half of the respondents said they felt inadequately experienced to work with children still appears to be a fairly promising result. Psychiatrists and non-parents rated themselves as significantly less competent in involving children to patients’ treatment compared to psychologists and parent respondents. Some studies confirm that professionals are not taught about family-focused themes in their curriculum (37, 40). In this study, three-quarters of professionals expressed a desire to see the topic included in professional training and would welcome endorsement for themselves. Previous research shows that family/child training has a positive impact on family-focused practice; professionals who have received family training are more likely to access parents’ awareness of their children’s well-being and more likely to gather information about patients’ children (23). Although COPMI care is still at a nascent stage in our country, some programs are already available, such as Child Talks + (41) or the translated manuals by Solantaus (42, 43). A structured interview guide entitled “Let’s talk about children” can help professionals with more precise and systematic identification of children’s needs (44).

Our study has certain limitations. First, relying on diagnostic criteria, such as those from the ICD-10, can be limiting, as there are alternative definitions of SMI that may consider factors beyond symptomatology, such as functional impairment or other criteria. In this study, we define SMI based on the ICD-10 diagnosis rather than functional criteria, as this diagnostic system is more commonly used in mental health settings in the Czech Republic. One limitation of the study is the inability to determine the exact percentage of respondents who consented to participate. This is due to the data collection method, which involved approaching both individuals and entire facilities, where the total number of individuals approached was beyond our control. Another study limitation was the use of exploratory factor analysis (EFA), which inherently involves a degree of subjectivity—particularly in decisions about the number of factors to retain, where the risk of under- or over-extraction cannot be entirely ruled out. Another of the constraints of this study was the relatively small sample size, which did not allow for more detailed statistical analyses (e.g., between different types of workplaces). As such, the findings should be interpreted as exploratory and indicative, rather than broadly generalizable. Selection bias may have occured due to the self-selection of participants, as professionals with more positive attitudes toward working with families or a greater interest in the topic may have been more likely to participate. In addition, as with all self-report data, socially desirable responding cannot be ruled out, although the fully anonymous nature of the questionnaire may have helped reduce this risk. Despite efforts to reach professionals across various settings and roles, access to the survey may have varied. Individuals contacted directly might have had a higher motivation to respond than those invited through institutional channels. These factors could have contributed to an overrepresentation of more engaged professionals, which may further limit the generalizability of the findings.

The use of a non-standardized questionnaire may also be a limitation, but in our setting, there was no standardized instrument available to measure the attitudes of professionals. Therefore, the results cannot be quantitatively compared with other studies. One limitation of the present study is that it did not explore potential interconnections between the identified barriers. Future research could benefit from examining how these factors may influence each other.

This research provides the first step for follow-up studies focused on mapping the needs of families with a member with SMI in detail and from the perspectives of families themselves. Based on our findings, we offer suggestions that could help improve COPMI care. First, it is important to spread awareness among professionals not only about the risk factors for COPMI but also about the possibilities of support for the families. This could be done through a comprehensive publication or brochure providing an overview of available support options, not just educational materials for families, but also a summary of organizations providing intervention and support programs. It would be appropriate to encourage and increase awareness of existing intervention programs and develop comprehensive recommendations and working practices for professionals wishing to increase the involvement of their patients’ children in their practice. A change in policy and the inclusion of the issue in the curriculum of professional training could be adopted as a long-term goal for the future.