A Mixed-Method Study on the Palliative Care Capacity of Social Workers in Flanders (Belgium): A Call to Expand Their Role

,


Introduction
Social workers are well-positioned to make meaningful contributions to multidimensional palliative care practice.Teir contributions can be twofold.First, social workers can play an important role in addressing the nonclinical dimensions of palliative care.Tey can address psychosocial and spiritual dimensions of care, contributing to a more holistic approach to palliative care practice involving the social networks of persons with serious illnesses and their relatives [1,2].Second, since social workers are specifcally trained to identify social determinants of health, they can ensure equal access to professional palliative care delivery for their clients [3].
Even though social workers can make meaningful contributions, they are often inadequately involved in daily palliative care practice [4].Several explanations have been listed in the literature such as unclear role descriptions [5,6], lack of education related to end-of-life care in social work [7,8], or the medicalisation of palliative care [9].
Te palliative care capacity of social workers, which we defne as their "ability to perform general and disciplinespecifc tasks in palliative care delivery," therefore, remains underused.Tis capacity is largely defned by social workers' attitudes toward the performance of palliative care tasks as well as their competencies in performing these tasks.However, contextual factors such as the appreciation of social workers by other care professionals are also important as they infuence their ability to perform tasks.

Study Rationale.
To further examine the potential role and actual involvement of social workers in palliative care [10], this study aims to explore the palliative care capacity of social workers in Flanders (Belgium).First, the study examines the frequency with which social workers in Flanders are currently performing palliative care tasks.Second, it assesses their attitudes and competencies regarding the performance of these tasks.Tird, as contextual factors also infuence the performance of tasks, this study explores how social workers feel about being appreciated by other care professionals.

Study Design.
A mixed-methods explanatory sequential design with quantitative and qualitative methods was used.Based on the guidelines by Creswell and colleagues [11], this design is considered best as the qualitative data support better interpretations of the quantitative data by providing examples or by explaining and contextualising the results.As this design is related to Denzin's triangulation between methods [12], it guarantees an in-depth understanding of the situation while at the same time taking into account the faws and defciencies of each single-method.Tis study consisted of two phases.Te quantitative phase consisted of a selfconstructed cross-sectional survey questionnaire.Subsequently, focus groups were organised for the qualitative phase.Although both phases were integrated during the focus group discussions, priority was given to the quantitative phase in drawing the main conclusions, while the qualitative fndings were used to explain and deepen the fndings from the quantitative phase.

Sample
2.2.1.Eligibility Criteria.Social workers who (A) have a bachelor's (BSW) or master's (MSW) degree in social work or social nursing from an ofcial educational institution and (B) come into contact with palliative care during the exercise of their profession were eligible to participate in the survey round as well as in the focus group sessions.We approached and defned the term 'palliative care" broadly, not as a specialised care service but as a form of care that enhances wellbeing in all situations involving life-threatening illnesses and characterised by death, dying, and bereavement.

Sampling
Procedure.Due to the lack of an existing sampling frame, it was not possible to draw a representative sample of social workers in Flanders.We, therefore, had to construct a theoretical sampling frame: the survey questionnaire was distributed to social workers from fve types of care organisations in Flanders, which were likely to employ social workers who met the eligibility criteria.Tese included the (a) social services of the hospitals (n � 1268); (b) health insurance services (n � 393); (c) community health clinics (n � 42); (d) home care services (n � 566); and (e) nursing homes (n � 116).Two comments should be made on these numbers.First, for the frst four service types, survey questionnaires were sent to the total population of employed social workers.However, as the numbers mentioned above remain estimates of the actual population size, it was not possible to calculate an exact response rate.Second, for the nursing homes, survey questionnaires were sent to a potential number of respondents from a random sample of 80 nursing homes (roughly ten percent of the total number of nursing homes in Flanders).For privacy reasons, all survey questionnaires were distributed by contact persons in the organisations involved.
Focus group sessions were organised based on organisational afliation.Respondents could indicate in the survey questionnaire whether they wanted to be invited to participate in focus group sessions.If yes, they were asked to provide their contact information.Potential participants were then contacted via e-mail to ask whether and when they could participate.Furthermore, additional calls for participation were sent out by contacts in the organisations involved.

Development of the Survey Questionnaire and Focus
Group Questions.Te main guideline for the development of the survey questionnaire was the "core competencies framework for palliative care in Europe" by Hughes and colleagues [13].Te tasks included in this list were adapted to the Flemish context and presented to feld experts in Flanders, including social workers, representatives of care organisations, social work education managers, and policy makers.Based on their feedback, the survey questionnaire was further refned.
Te fnal survey questionnaire was divided into three parts.Te frst part consisted of items measuring respondents' personal, educational, and professional backgrounds.Te second part measured -for a total number of 49 tasks in palliative care divided into ten work packageson a 5-point Likert scale the following: (a) the frequency with which respondents are currently performing a certain task, (b) the attitude of respondents regarding the performance of this task, and (c) the competency of respondents regarding the performance of this task.Te third part measured on a 5point Likert scale how often (a) respondents feel appreciated by other care professionals when they share their views on end-of-life matters and (b) respondents think that their views on end-of-life matters are considered by other care professionals.
Semistructured focus group questions were formulated after an initial analysis of the quantitative data for each organisational group of social workers.We chose this approach in order to explain the quantitative data while leaving room for spontaneous discussions.

Data Collection Period. Data were collected between
April and September 2021.Te electronic survey questionnaires were sent by means of an anonymous link on QUALTRICS© in April and May 2021.Reminders were sent according to the total design method [14].Subsequently, online focus group sessions, which were audio and video recorded and anonymously transcribed, took place between June and September 2021.
2.2.5.Quantitative and Qualitative Data Analysis.Te quantitative data were analysed using the statistical software program SPSS©.Univariate and bivariate descriptive analyses were used to discern and describe the main tendencies in the data.To compare the three dimensions of the tasks list (frequency, attitude, and competency), we used TYPE I and TYPE II disconfrmation scores.TYPE I scores were used to identify which tasks were considered important (scores 4-5 on the 5-point Likert scale) but not frequently performed in daily practice (scores 1-2 on the 5-point Likert scale).Furthermore, TYPE II scores were used to identify which tasks were considered to be unimportant (scores 1-2 on the 5-point Likert scale) but frequently performed in daily practice (scores 4-5 on the 5-point Likert scale).Te same strategy was used to identify the tasks for which respondents considered themselves competent to perform (scores 4-5 on the 5-point Likert scale), but which they did not frequently perform in practice (scores 1-2 on the 5-point Likert scale), or vice versa.Missing data were not considered for analysis.Response patterns in which no data were collected for the task list items or which contained missing data for more than 50% of all survey items were not retained for analysis.
Subsequently, the qualitative data were analysed using Nvivo12©.Te qualitative data analysis was guided by the framework method set out by Ritchie and colleagues [15].Coding was done both deductively and inductively.Te deductive coding was related to the research questions, while the inductive coding was related to focus group participants' comments on the quantitative results in spontaneous discussions.Codes were grouped into categories and applied to each transcript fle until no additional codes emerged.Tendencies and diferences in the data were then identifed.Both audio and video recordings were destroyed after the identifcation of the main tendencies and diferences.

Characteristics
3.1.1.Quantitative Phase.A total of 499 respondents completed the survey, out of which 111 respondents could not access the full survey questionnaire, either because they did not give informed consent to process their data (n � 5) or because they did not meet the eligibility criteria (n � 106).Tese response patterns were not retained for analysis.Furthermore, additional 36 response patterns were not retained for analysis because no data were collected for the task list items (n � 19) or because missing data accounted for more than 50% of all survey items (n � 17).Te response patterns of 352 respondents were retained for data analysis.
Table 1 presents the main characteristics of the survey respondents.Respondents were mostly female (88.6%), with a BSW or MSW degree (81%) and had moderate professional experience (48.4%).Furthermore, most respondents worked in hospitals (42.3%) and health insurance services (33.2%), followed by home care services (13.3%) and nursing homes (9.7%).Te majority of respondents (54.2%) did not receive information on palliative care during their education, and a minority of respondents (29.9%) followed continuing education courses on palliative care.Despite the lack of specifc education, 81.5% of respondents indicated coming into contact with palliative care at least monthly in their professional lives.Tese percentages difer, however, between organisations: respondents from hospitals and health Health & Social Care in the Community insurance services indicated coming into contact with palliative care on a more frequent basis than respondents from nursing homes and home care services (65.1% and 61.5% against 52.9% and 42.5%, respectively).
3.1.2.Qualitative Phase.Twenty-four social workers participated in a total of fve focus groups, out of which nine participants worked in a hospital social service, eight worked in a nursing home, four worked in a health insurance service, and three worked in home care services.No focus group session was organised for the community health clinics since no willing participants could be found, while two focus group sessions were organised for the hospital social workers.Twenty-three participants were female and one participant was male.
3.2.Palliative Care Tasks.Table 2 presents survey respondents' performance of palliative care tasks, i.e., the frequency with which they perform tasks, their attitudes regarding the performance of these tasks, and their competencies in performing these tasks.

Te Frequency with which Social Workers Perform
Palliative Care Tasks.Respondents reported frequently talking with clients (64.3%) about the social consequences of a serious illness and with relatives about the social (64.3%) and fnancial (64.7%) consequences of a serious illness.Furthermore, they reported frequently assessing clients' needs (81.6%), especially the applicability of fnancial reimbursements (78.9%).Based on this assessment, they frequently guided clients to nonprofessional (67.3%) as well as professional (78.9%) forms of support.Consequently, they frequently cooperated with other care professionals (76%) and referred clients to them (78.3%).
Respondents stated that they did not frequently perform tasks related to dealing with clients and relatives from diferent cultural backgrounds.Respondents did not frequently assess the needs of these clients and their relatives (43.7% and 44.6%, respectively), while a majority of respondents did not frequently start a conversation about endof-life with these clients and their relatives (66.5% and 65.3%, respectively).Furthermore, respondents do not frequently talk about religion or spirituality with clients and relatives (65.6% and 63.9%, respectively) or perform politicising tasks such as recognising and addressing (40.4% and 45%, respectively) structural problems in the guidance and support of clients.Two other separate tasks that respondents indicated they did not frequently perform are advocating for relatives' needs after the death of a client (41.5%) and mediating conficts between clients and relatives (40.9%).

Social Workers' Attitudes regarding the Performance of
Palliative Care Tasks.Te results show that respondents generally have a positive attitude regarding the performance of palliative care tasks since more than 50% of respondents considered 46 out of 49 tasks to be important for their social work practices.Tree tasks were considered important by less than 50% of respondents: talking with clients or relatives about religion/spirituality (38.4% and 31.7%,respectively) and assessing clients' needs using instruments (49%).
To compare respondents' attitudes regarding the performance of palliative care tasks and the frequency with which they perform them, TYPE I and TYPE II disconfrmation scores were calculated.TYPE II disconfrmation scores equalled 0%.Tus, there are almost no respondents who indicated frequently performing palliative care tasks while having a negative attitude towards them.TYPE I disconfrmation, or tasks that respondents did not frequently perform while having a positive attitude to perform them, was found in 18% to 29% of respondents as shown in Table 3.
Based on the focus group discussions of these results, fve tendencies can be described.First, dealing with clients and relatives from diferent cultural backgrounds was considered important by respondents for their social work practices but was not frequently performed.Focus group participants referred to the lack of diversity in client populations despite the fact that the current Flemish society is characterised by superdiversity: "I think it is more difcult to reach those people who do not speak our language (Dutch).Working with interpreters creates gaps in understanding and the diferent culture creates barriers as well since talking about grief difers between cultures." Second, high disconfrmation scores were found for tasks relating to politicising work such as addressing (24.6%) and advocating (22.9%) structural problems in the guidance and support of clients.In the focus groups, nursing home social workers often referred to the fact that care managers did not expect social workers to perform a politicising role.As one participant commented, "Sometimes I have the feeling that they have difculties with our critical point of view and the questions we ask."Tird, 20.7% of respondents think it is important to advocate for relatives' needs after the death of clients even though they do not frequently perform this task in practice.In the focus groups, hospital social workers pointed to time constraints in organising aftercare for clients' relatives.According to these participants, home care social workers should be responsible for tasks related to bereavement support and other forms of aftercare for relatives.However, home care social workers pointed to organisational and fnancial constraints to perform these tasks: "It is strange when you can make just one phone call to relatives after the death of a client or that you have to take a day of to go to a client's funeral.Especially when you have been guiding and supporting those people for a long period of time, sometimes even for years."Fourth, 21.3% of respondents consider mediating conficts between clients and relatives to be important although they do not frequently perform this task in practice.Focus group participants confrmed that social workers should perform this task on a more regular basis.As one nursing home social worker argued: "Colleagues (doctors or nurses) often avoid relatives who want to enter into a confict with clients or care professionals, while these are the people who 4 Health & Social Care in the Community 6 Health & Social Care in the Community you should talk to and get to know better to understand their situation!"Fifth, tasks related to evaluating care plans are considered important but not frequently performed.High disconfrmation scores were found for the tasks of recognising (17.9%) and reporting (18.6%) gaps in clients' care plans.Focus group participants explained this by stating that social workers are not structurally involved in the development or evaluation of care plans, which was especially emphasised by the nursing home social workers.As one participant put it, "I never considered social work as being a purely administrative job.We have to be able to coordinate the care planning and adjust it when necessary.We need to be the spokespersons of clients and relatives and continue to point out their wishes to other care professionals."

Social Workers'
Competencies regarding the Performance of Palliative Care Tasks.More than 50% of respondents felt competent to perform the majority of palliative care tasks listed in this study (28 out of 49 tasks).Respondents felt competent to assess (84.8%) and advocate for (77%) clients' needs, especially the applicability of fnancial reimbursements (75.2%).Respondents also felt competent to guide clients to professional forms of support (82.9%), to cooperate with other care professionals (81.4%), and to refer clients to them (82.8%).
Respondents felt that they were not competent to perform tasks in two areas.First, more than a quarter of respondents felt that they were not competent to start a conversation with clients about end-of-life (27.1%) or answer questions on end-of-life matters (28.4%).Tis was especially the case with clients or relatives from diferent cultural backgrounds.A majority of respondents felt that they were not competent to start a conversation about endof-life issues with clients (57.8%) or relatives (54.3%) from diferent cultural backgrounds.A majority of respondents also felt they were not competent to talk about religion or spirituality with clients (52.8%) or relatives (54.2%), which could be important in this group.Second, more than a quarter of respondents felt that they were not competent to recognise (25.9%) or report (27.9%) gaps in care plans for individual clients, as well as to address structural problems observed in care plans (29.2%).
To compare respondents' competencies regarding the performance of palliative care tasks and the frequency with which they perform them, TYPE I and TYPE II disconfrmation scores were calculated.All scores were between 0% and 10%, which implies that respondents gave similar scores on the variables measuring their competency regarding the performance of tasks and the variables measuring the frequency with which they perform these tasks.Consequently, there are no tasks for which a substantial number of respondents indicated that they do not feel competent to perform them, while frequently performing them, or vice versa.
Focus group participants explained the low disconfrmation scores as follows: when social workers frequently perform certain palliative care tasks, they will also feel more competent to perform these tasks and vice versa.According to the participants, this is a direct consequence of the lack of content on end-of-life themes in educational curricula and the necessity to gain work experience to develop competencies.As one hospital social worker explained, "I am less competent to evaluate care plans just because I just do it less."A colleague responded that "We don't learn how to do this during our education time so it all depends on the specifc service you are working in.[. ..]You gradually learn to perform these tasks as you perform them."

Appreciation of Social Workers by
Other Care Professionals in Palliative Care. Figure 1 illustrates the level of appreciation respondents indicated they had experienced from other professionals in palliative care.Te frst graph shows how often respondents feel appreciated for their views by other care professionals, while the second graph shows how often respondents experience that these views are considered.
Although 76% of respondents indicated that they frequently cooperate with other care professionals, they do not always feel appreciated for their views on end-of-life matters.
Table 3: Highest type I * disconfrmation scores between respondents' attitudes regarding the performance of palliative care tasks and the frequency with which these are performed (N � 352).

Palliative care tasks
Type I disconfrmation score (1) Starting a conversation about the end-of-life with clients from a diferent cultural background 29% (2) Starting a conversation about the end-of-life with clients' relatives from a diferent cultural background 28.9% (3) Assessing the needs of clients from a diferent cultural background 26% (4) Addressing structural problems in the guidance and support of clients 24.6% (5) Recognising structural problems in the guidance and support of clients 22.9% (6) Mediating conficts between clients and their relatives 21.3% (7) Advocating for relatives' needs after the death of a client 20.7% (8) Assessing the needs of relatives from a diferent cultural background 20.6% (9) Reporting gaps in clients' care plans 18.6% (10) Recognising gaps in clients' care plans 17.9% Two fndings deserve further explanation.First, hospital social workers are the only group in which a majority of respondents (66.7%) frequently feel appreciated (compared to 38.2%, 41.1%, and 45.5% in nursing homes, health insurance, and home care services, respectively).Focus group participants, however, nuanced this result by pointing out the diference between acute and chronic hospital services.
As one participant put it, "Appreciation depends on colleagues and multidisciplinary cooperation.In the geriatric service, I am just as important a member as the occupational therapist or the physiotherapist.In acute services, however, this is not the case.Te role of social workers is really limited there."Second, nursing home social workers feel least appreciated for their views, as a ffth of the respondents in this group (20.6%) indicated that they seldom or never feel appreciated by other care professionals (compared to 9.8%, 4.3%, and 9% in the health insurance services, hospitals, and home care services, respectively).One participant in the focus group explained this result by pointing to the limitation of the social worker's role to administrative tasks, while another participant described the general underappreciation of nursing home staf.Furthermore, a majority of respondents in each organisational group indicated that their views on endof-life matters are not always considered by other professionals.However, there was a substantial diference between hospital respondents and the other organisational groups.In the former group, 48.5% of respondents indicated that their views are frequently considered, compared to 16.8%, 24.5%, and 15.6% in nursing homes, health insurance, and home care services, respectively.In the focus groups, hospital social workers explained this result by pointing to their responsibilities in the hospital discharge process.As one participant commented, "Tey (other care professionals) need us to discharge patients from the hospital.[. ..]Our views are frequently considered because we have a practical function in the hospital.Te sooner we act, the faster patients can leave.[. ..]Because of this we can exert pressure but there are of course also many discussions."However, participants from all groups did not describe such discussions as disturbing, but rather as a precondition to critically examine the care provided.As one nursing home social worker commented, "I feel that other professionals do not always take my view into account.I do not have a monopoly on the truth so it is not even necessary.Yet, I always try to expand the views of others.You just have to be critical as a social worker.So, even if they do not agree, we should at least be able to have a discussion!"

Discussion
Tis mixed-method study explores the palliative care capacity of social workers in Flanders, defned as their "ability to perform generic and discipline-specifc tasks in palliative care."Te results show that respondents have a positive attitude regarding the performance of palliative care tasks and the competencies to perform the majority of them.Nevertheless, the results also illustrate that respondents in various care settings in Flanders do not always feel appreciated by other care professionals for their views on end-of-life matters and that these views are not always considered.
Although the palliative care capacity of social workers in Flanders is limited to tasks related to the assessment and referral of clients, their potential role extends beyond these tasks.Tere are, for example, tasks for which respondents in this study have a positive attitude but which they do not  frequently perform.However, these are tasks that we may expect to be performed by social workers for two reasons.First, social workers are well-positioned to address the social dimension of palliative care, for example, by involving clients' social contexts [2] and by acting as a bridge between clients, relatives, and professionals in care institutions [16].
Although respondents in this study do not frequently mediate conficts between clients and relatives or advocate for relatives' needs after the death of clients, they have a positive attitude regarding these tasks.Tis also applies to tasks related to evaluating care plans, through which they could ensure that the wishes of clients and relatives are respected by all professionals involved.Second, tasks related to politicising work are discipline-specifc tasks through which social workers can put typical social work values, such as social justice, into practice.For example, by performing a politicising function, social workers can address unequal access to professional palliative care for vulnerable and underserved populations [17,18] at the organisational or policy level.Although they do not frequently perform policitising tasks in palliative care, respondents in this study have a positive attitude regarding the performance of these tasks.Future research should examine the most acceptable and feasible strategies to expand the role of social workers in Flanders so they can fully address the social dimension of palliative care and engage in politicising work.However, this study points to two important preconditions for implementing this role.First, although respondents indicated having a positive attitude toward the tasks associated with this role, they also indicated feeling less competent or less capable of performing them.Te competencies of social workers to perform these tasks must therefore be increased.Second, in line with previous research [16,19], this study illustrates that competencies in palliative care are a product of education and work experience.For social workers to gain further experience, it is important to increase their actual involvement in palliative care by addressing contextual factors that prevent them from performing tasks related to an expanded role.All Flemish care organisations should therefore improve the appreciation of social work in palliative care among other care professionals, which can, in turn, increase the job satisfaction of social workers [20].Furthermore, as focus group discussions also pointed to a lack of time and resources or the nonstructural involvement of social workers in multidisciplinary cooperation, these factors must also be considered.
Te contribution of this study is twofold.First, the study contributes to contemporary social work research by addressing the need to further examine the role and involvement of social workers in the broader feld of palliative care (i.e., the feld of death, dying, and bereavement).Additionally, it may inspire other researchers to consider our approach to palliative care capacity, namely the idea that social workers' ability to perform tasks not only depends on competencies.It is equally important to consider their attitudes toward palliative care tasks [21,22] and contextual factors that facilitate or hinder the performance of tasks [10].Second, this study contributes to overall palliative care practice in Flanders.As medical professionals cannot address all realms of death, dying and bereavement, holistic, and multidimensional palliative care practice requires the actual involvement of other professionals such as social workers.Tis study shows that social workers in Flanders can contribute to a holistic and multidimensional approach to palliative care by specifcally addressing the social dimension of care.Terefore, palliative care practice can be improved overall if there is more actual social work involvement and more opportunity for social workers to gain experience in this feld.
Tis study has three limitations.First, the results of this study are not representative of the whole population of social workers in Flanders due to the lack of an existing sampling frame.To account for this limitation, we constructed a theoretical sampling frame aiming to reach as many social workers as possible in relevant care settings.Moreover, by adopting the principles of triangulation between methods in the sequential explanatory design, we ensured that we got as clear a picture of the situation as possible.Second, for the sake of clarity, the diferences between the various groups based on organisational afliation were not fully addressed in this article.However, our data remain available for future research or practice development that aims to examine and address the diferences between the various groups.Tird, the opinions of other professionals on the social work role and the involvement of social workers in palliative care were not covered in this study.Tis could also be addressed in further research.

Conclusion
Te palliative care capacity of responding social workers in Flanders, defned as their "ability to perform generic and discipline-specifc tasks in palliative care," is limited to tasks related to client assessment and referral.However, as their potential role extends beyond these tasks, future research should examine the most feasible and acceptable strategies to expand the role of social workers so they can fully address the social dimension of palliative care and engage in politicising work.An expanded social work role and increased involvement of social workers would contribute to a holistic and multidimensional approach to palliative care practice.Nevertheless, this study points to a need to increase the competencies of social workers for tasks associated with this role and to address relevant contextual factors that prevent social workers from performing these tasks.

Data Availability
Te quantitative data are currently available upon request from the corresponding author and they will be made available in a depository.Access to the qualitative data is restricted for privacy reasons as focus group participants could potentially be recognised and traced.