Health professionals’ practices and perspectives of post-stroke coordinated discharge planning: a national survey

Introduction

Going home after a stroke is a significant milestone in the stroke recovery journey (Allen et al. 2002). Person-centred discharge planning is essential for a successful transition home, as it reduces length of stay, risk of rehospitalisation following discharge (Shepperd et al. 2010; Gonçalves-Bradley et al. 2022), and risk of secondary stroke (Johnston et al. 2010). Moreover, person-centred discharge planning improves compliance with community-based rehabilitation (Almborg et al. 2009a) and is associated with greater quality-of-life (Andrew et al. 2018). As such, stroke clinical guidelines from countries including the USA, Canada, the UK, Ireland, and Australia, recommend that appropriate, timely follow-up for the stroke survivor should be person-centred and arranged with specialist stroke services and primary care providers (Cameron et al. 2016; Adeoye et al. 2019; Mountain et al. 2020; Intercollegiate Stroke Working Party 2023; Stroke Foundation 2023).

As well as ensuring a person-centred approach to discharge planning, there is emerging evidence that a coordinated approach to discharge planning offers additional benefits. Coordinated discharge planning (CDP) is a multifaceted process that involves the person experiencing a particular health concern, their families, and all healthcare professionals involved in care (such as the multidisciplinary stroke team) and includes interorganisational coordination between inpatient and community care providers as well as a connection with a key contact in the community (such as a case manager or primary care provider). For example, CDP ensures timely follow-up of service provision (Carlsson et al. 2012) and effective communication between stakeholders whilst maintaining a person and family-centred approach (McDonald et al. 2007; Reeves et al. 2019). Although the Australian Clinical Guidelines for Stroke Management (Stroke Foundation 2023) recommend that the ‘discharge planning process may involve relevant community services…stroke support services, (and) any further rehabilitation or outpatient appointments’, the process of communicating and supporting an integrated transfer of care between providers is not explicitly addressed. While the provision of a written discharge plan is an important aspect to communicate information to all stakeholders (Carlsson et al. 2012), there is a distinction between the presence of a written discharge plan and the process of coordinating discharge planning with key stakeholders; the presence (or lack) of a written plan does not reflect whether CDP has occurred.

There is evidence suggesting breakdowns in many aspects of CDP practices that have the potential to impact the successful transition home. For example, studies have shown that the frequency of providing written discharge care plans, in addition to their quality and accuracy, is poor. A recent report from the Australian Stroke Clinical Registry identified that approximately one third of stroke survivors missed out on receiving a written care plan upon hospital discharge (Cadilhac et al. 2022). Omissions of critical components of discharge care plans have been reported (Kind et al. 2012), with stroke survivors identifying a lack of education and support when returning home (Jones et al. 2008; Ellis-Hill et al. 2009; Luker and Grimmer-Somers 2009). Furthermore, Andrew et al. (2018) found that only 18% of 200 participants received all aspects of ‘best practice’ discharge care planning, indicating significant evidence-practice gaps for aspects such as complete information about medications and support structures.

In addition to gaps in the provision of written discharge plans, there is evidence that stroke survivors, families, and healthcare professionals have unmet needs surrounding CDP processes. Stroke survivors have reported a lack of participation in discharge planning and have identified a need for a communicative ‘alliance’ when negotiating services with health professionals (Hedberg et al. 2008; Almborg et al. 2009b). The concept of an alliance was also reflected in a recent Swedish survey of 21 stroke survivors, their families, and healthcare professionals (Lindblom et al. 2020), where the need for an open and responsive discussion about all stakeholders’ perspectives was required to achieve success in coordinated transition from hospital. Primary care providers have also identified that communication from inpatient settings is often inadequate. A recent qualitative study conducted with 40 primary care providers found that discharge care plans often lacked content and clarity and were typically reported from a ‘medical perspective’ rather than focussing on stroke survivors’ recovery goals and community needs (Sheehan et al. 2022).

Improved coordination of services across hospital and community settings is needed to support continuity of care after stroke; however, there is limited evidence about how to do so. Previous interventions have found that care transitions for stroke survivors improve when a community-based ‘key worker’ facilitates case management and support (e.g. Allen et al. 2002; Reeves et al. 2019) and when primary care teams are guided by integrated care pathways (Abdul Aziz et al. 2020), but these community-based interventions are unlikely to improve hospital processes. Given that the majority of stroke services are provided in inpatient hospital settings in many countries, including Australia (Lynch et al. 2019), it is important to understand how to improve CDP processes and interorganisational collaboration prior to discharge.

Research has identified a wide range of barriers and enablers that influence CDP practices in stroke care. One commonly reported barrier relates to organisational constraints, including a culture of deprioritising discharge planning (Cadilhac et al. 2017) and a lack of organisational resources to support the transition from hospital (Chen et al. 2020). These cultural and organisational barriers have been shown to complicate CDP processes, despite the potential for strategies such as information systems (e.g. e-records), specialist discharge roles, and group-based planning activities (e.g. ward rounds) to support CDP (Waring et al. 2019). Other reported hospital-based barriers include pressure to discharge patients (Magdon-Ismail et al. 2016; Kable et al. 2019) and decision-making dominated by medical issues (Waring et al. 2016). Post-discharge barriers, including inadequate, delayed, or inaccessible discharge care plans (Waring et al. 2016; Lynch et al. 2017; Kable et al. 2019; Sheehan et al. 2022), have also been reported to impact the continuity of care for stroke survivors discharged home. In contrast, having a shared understanding of stroke survivors’ and their families expectations, needs, and wishes (Allen et al. 2002; Sheehan et al. 2022; McDonald et al. 2007) has been found to facilitate CDP.

In order to develop strategies or innovations to enhance implementation of CDP within stroke services, it is necessary to understand the factors influencing health professionals’ practice using behaviour-change theory (Eccles et al. 2005). It is well recognised that implementation strategies developed and tailored to address context-specific barriers are more effective than non-tailored interventions or passive guideline dissemination (Baker et al. 2015), and this process can be facilitated by the use of theoretical frameworks. The Theoretical Domains Framework (TDF) (Michie et al. 2005; Cane et al. 2012) is one example of a framework that is useful in designing clinically meaningful and replicable implementation strategies (French et al. 2012; Phillips et al. 2015). The TDF is a synthesis of 33 behaviour-change theories that includes 14 domains: (1) knowledge; (2) skills; (3) social role and identity; (4) beliefs about capabilities; (5) optimism; (6) beliefs about consequences; (7) reinforcement; (8) intentions; (9) goals; (10) memory, attention, and decision processes; (11) environmental context and resources; (12) social influences; (13) emotion; and (14) behavioural regulation (Cane et al. 2012). The TDF has been used in many areas of healthcare research, including smoking cessation (Fulton et al. 2016), hand hygiene (Smith et al. 2019), and stroke rehabilitation (Lynch et al. 2017).

However, few studies have utilised behaviour-change theory to explore barriers and facilitators to CDP practice, and hence, there has been little progress towards tailored and theory-informed implementation. One exception is an Australian study (Cadilhac et al. 2017) that utilised the TDF (Cane et al. 2012) to investigate factors influencing adherence to stroke discharge care processes (including the provision of a discharge care plan) in two ‘top performing’ Australian hospitals. In their pilot study, Cadilhac et al. (2017) developed a multifaceted organisational intervention that improved and sustained discharge care in hospitals, which contributed to a 22% increase in the provision discharge care plans. The intervention included an educational meeting, performance feedback and action planning to agree on strategies, such as using the Stroke Foundation’s ‘My Stroke Journey’ education booklet and care planning guide (Stroke Foundation 2019). This preliminary study provides valuable evidence that theory-informed implementation can lead to improved discharge planning processes; however, the concept of ‘coordinated’ discharge planning was not specifically addressed. Moreover, there is a need to understand factors influencing discharge planning in a broader range of contexts (not only ‘top-performing’ hospitals) to inform the development of a contextually relevant implementation strategy to enhance CDP.

In summary, CDP is a complex and multifaceted process, and there are evidence-practice gaps in current practice. In addition, there is a paucity of theory-informed research that has investigated and designed strategies to address the barriers faced by health professionals within the Australian healthcare context. To inform the development of potential solutions to improve discharge planning, further research is needed to better understand factors influencing post-stroke CDP, guided by a theoretical framework. Therefore, this study aimed to (1) describe the practices of stroke health professionals in undertaking stroke discharge planning and (2) identify the factors (i.e. barriers and facilitators) influencing CDP practices specific to the Australian healthcare context using the TDF.

Materials and methods

Results

Results are presented according to the structure of the survey, with demographic information reported first. The remaining results from Part 1 of the survey are presented under the heading ‘Post-stroke discharge planning practices’. Results from Part 2 of the survey relate to the TDF and are presented under the heading ‘Factors influencing post-stroke coordinated discharge planning’. The final section relates to the integration of qualitative and quantitative findings.

Factors influencing post-stroke CDP

Participant responses to statements regarding their role in CDP, corresponding to TDF domains (Michie et al. 2005; Cane et al. 2012), are shown in Fig. 1, where the percent ‘less than agreement’ (barrier) and percent in ‘agreement’ (facilitator) was calculated per domain. The main barriers included the socio-political context (81% less than agreement) and health professionals’ beliefs about capabilities (50% less than agreement). The domains with the highest percentage agreement included social/professional role and identity, knowledge, emotions, and intentions, indicating these were the main CDP facilitators. Percent agreement for individual TDF-aligned statements can be found in Supplementary Appendix S2.

Fig. 1.

Agreement with TDF statement regarding implementation of CDP (n = 42).

Participants’ responses to open-ended questions about the ‘most significant’ facilitators and barriers to engaging in CDP were analysed using deductive analysis with content units mapped to TDF domains; these are summarised in Tables 4 and 5. Overall, the ‘organisation’ domain was linked to the majority of reported barriers (n = 29) and facilitators (n = 25), described further below. Following this, the next most reported categories of barriers included the socio-political context (n = 8) and health professionals’ beliefs about capabilities (n = 6). Examples of these barriers included a ‘lack of community services/providers… [that] impacts wait lists, intensity of therapy and available discharge options’ (SLP2), and ‘competing demands’ (SW2) that had a negative impact on healthcare professionals’ perceived behavioural control. In addition to the ‘organisation’, the main facilitators to CDP were health professionals’ positive social influences (n = 7), such as a ‘supportive team environment’ (SW2) and social/professional role and identity (n = 6), including ‘expectations that discharge handovers and referrals will be completed’ (OT5).

Table 4.Barriers to CDP identified as most significant by participants mapped to the Theoretical Domains Framework (TDF).

Domain	Specific belief/subconstruct A	Example quotes	N reported
Organisation	The organisation in which I work, including its resources (time, money, staffing, general resources), its culture (of discharge summaries, staff access, discharge timing, focus, communication and external services), and my interaction with the environment, is a barrier to CDP.	‘Inadequate time available…not knowing when patients are discharging.’ (D1)	29
Socio-political context	The socio-political context, including service availability, the National Disability Insurance Scheme, external organisation demands, and funding, is a barrier to CDP.	‘Lack of community services/providers – this impacts wait lists, intensity of therapy and available discharge options (e.g. patients who are ‘too good’ for RACF [residential aged care faciity] but require some level of support/assistance/supervision that cannot be provided by family).’ (SLP2)	8
Beliefs about capabilities	Beliefs about my capabilities, including my caseload demands and perceived behavioural control, is a barrier to CDP.	‘…Competing demands…’ (SW2)	6
Patient factors	Patient characteristics and circumstances are barriers to CDP.	‘The biggest challenge in previous roles I had was patients having a GP to provide handover…’ (OT1)	5
		‘Complex family situations’ (SLP04)
		‘Client’s motivation/ability to take on a bigger role as they reach the end of the time at community rehabilitation (in relation to initiation).’ (SLP 5)
Knowledge	Knowledge, including procedural knowledge (of referring to available services, adapting to processes of different areas) and knowledge of impairments, are barriers to CDP.	‘…Knowledge of referral processes. ‘Who do I contact about that’?’ (NP9)	4
Innovation strategy	Innovation strategies, such as training, are barriers to CDP.	‘…Education/training’ (NP7)	1
Social influences	Social influences, including social support, are barriers to CDP.	‘Engagement of GPs…’ (SLP5)	1
Social/professional role and identity	My professional role, including the competing demands related to my role in CDP, is a barrier to CDP	‘We do not have a dedicated stroke OT on our acute wards.’ (OT2)	1

SLP, speech language pathologist; OT, occupational therapist; D, dietitian; NP, nursing professional; SW, social worker.

^A Each specific belief/subconstruct has been created based upon subconstructs of the TDF in combination with categories derived from participant responses.

Table 5.Facilitators to CDP identified as most significant by participants mapped to the Theoretical Domains Framework (TDF).

Domain	Specific belief/subconstruct A	Exemplar quotes	N reported
Organisation	The organisation in which I work, including its associated resources (time, staffing, documentation/handovers), culture (staff access, expectations/views, communication and collaboration), and my interaction with the environment, is a facilitator to CDP.	‘Weekly multidisciplinary case conference and accessibility of the allied health team – both of these provide the opportunity to discuss the patient’s goals, recommendations for follow-up and essentially bounce ideas around regarding appropriate discharge options and plan.’ (SLP2)	25
Social influences	Social influences, including social support and modelling, are facilitators to CDP.	‘Supportive team environment’ (SW2)	7
Social/professional role and identity	My professional role, including conducting ward rounds, the expectation of completing documentation and handovers, and showing leadership, is a facilitator to CDP.	‘Expectations that discharge handovers and referrals will be completed.’ (OT5)	6
Knowledge	Procedural knowledge, including knowledge of available services and referral processes, is a facilitator to CDP.	‘Awareness of post-discharge services and processes to refer them to.’ (AHP – unspecified)	2
Intentions	Intentions, including my motivation, is a facilitator to CDP.	‘My own motivation’ (SLP5)	1
Beliefs about consequences	Beliefs about my consequences, including my outcome expectancies, are facilitators to CDP.	‘Seen as ‘part of the process’ for stroke patients.’ (SLP7)	1
Skills	Skills, including practice, are facilitators to CDP.	‘Previous experience working for the stroke foundation.’ (SLP5)	1
Patient factors	Patient characteristics and circumstances are facilitators to CDP.	‘When patients and their families are able to advocate for themselves.’ (SLP8)	1
Behavioural regulation	Behavioural regulation, including self-reflection, is a facilitator to CDP.	‘…Self-reflection on pros/cons of providing coordinated discharge planning in the past…’ (SLP5)	1
Beliefs about capabilities	Beliefs about my capabilities, including my perceived behavioural control, are facilitators to CDP.	‘…Writing good chart entries and discharge summaries…’ (PT1)	1
Socio-political context	The socio-political context, including good community services, is a facilitator to CDP.	‘…Good follow-up services in the community, however, these may be stretched at times.’ (OT7)	1

SLP, speech language pathologist; OT, occupational therapist; PT, physiotherapist; SW, social worker; AHP – unspecified, allied health professional who did not specify their discipline.

^A Each specific belief/subconstruct has been created based upon subconstructs of the TDF in combination with categories derived from participant responses.

Discussion

This study explored Australian stroke healthcare professionals’ discharge planning practices and identified factors influencing CDP. Findings suggested that although most participants reported high levels of involvement from key stakeholders, including the stroke survivor, family and hospital-based clinicians in discharge plan development, there was inconsistent coordination of this process between care providers across different settings, such as community. Reported practice and health professionals’ perceived needs for improvement are discussed below, in addition to the key factors influencing practice and barriers that should be considered in future implementation efforts.

The majority of respondents reported that stroke survivors in their service frequently received a discharge plan (90%) and that stroke survivors and families were frequently involved in this plan’s development (88%). These results highlight a discrepancy between clinicians’ perceptions and national data, indicating that 69% of stroke survivors receive a discharge care plan and 31% do not (Cadilhac et al. 2022). This higher self-rating aligns with previous literature, indicating that clinicians tend to overestimate their adherence to guidelines (Sadeh-Sharvit et al. 2022). Similarly, although our findings suggest higher levels of stroke survivor and family involvement than previously reported (Almborg et al. 2009a, 2009b; Perry and Middleton 2011), these differences between health professionals, stroke survivors, and families’ perspectives may also represent an overestimation of stakeholder involvement. Several participants in our study reported a need for more collaborative and communicatively accessible discharge planning processes that include the stroke survivor and family, which aligns with previous findings (Hedberg et al. 2008; Almborg et al. 2009a, 2009b; Perry and Middleton 2011) and suggests ongoing work to optimise stroke survivor and family participation is needed.

Discrepancies were also identified between the level of CDP involvement of inpatient staff and external care providers, with inpatient staff involved more frequently. Specifically, results indicated high levels of contribution to discharge plans from multidisciplinary hospital staff (90%), whereas over one third of participants reported that coordination with external care providers (e.g. general physicians, community services) occurred only ‘sometimes’ or ‘occasionally’. These findings indicate inconsistent coordination of care between inpatient and external care providers, supporting previous reports of poor communication with external care providers in stroke and other populations (Røsstad et al. 2013; Waring et al. 2016).

The impacts of inconsistent coordination between care providers were demonstrated in reports of uncertainty around enacting discharge plan recommendations. More than half of respondents indicated they were unsure if recommendations had been actioned from care plans. These findings align with previous literature that health professionals perceive poor follow-up care as a threat to safe discharge, with concerns relating to perceived limitations in community and primary care providers’ involvement in post-discharge care (Waring et al. 2016). This uncertainty reinforces the existing evidence that highlights the need for enhanced communication between care providers across the care continuum (Lindblom et al. 2020) and further supports the argument that giving information in the form of a written discharge plan does not equate to care coordination. Consequently, further research into improving CDP processes is a priority.

Despite the generally positive perceptions of discharge planning processes overall, few participants reported CDP was managed well in their practice setting. These results suggest that CDP practices are inconsistent between hospitals and that some stroke survivors may receive better quality care depending on the hospital to which they are admitted, highlighting the need for a standardised discharge protocol (McCann et al. 2009; Cadilhac et al. 2011; Cadilhac et al. 2017). These results may also indicate that participants’ perceptions of discharge planning differed and that some may not have been aware of or valued the importance of coordination with external care providers. To explore the underlying reason for these results further, more detailed observational or in-depth exploratory studies are required.

Explicit exploration of the factors influencing CDP identified that the organisation domain was perceived as both a barrier and a facilitator, reflecting complex and multifaceted organisational structures. Almost half (46%) of our respondents rated this domain as a barrier, with 29 qualitative responses relating to how the organisation hindered CDP. The finding that organisational culture can prevent prioritisation of discharge planning is similar to previous research, where it has been reported that stroke survivors are often discharged from hospital survivors before CDP can occur, due to low bed capacity (Magdon-Ismail et al. 2016; Waring et al. 2016; Cadilhac et al. 2017). Several other organisational barriers were common to the study by Hesselink et al. (2014) in a general population, including workload and time pressure, work shift structures, and a lack of communication. As mentioned above, poor collaboration with external care providers was also common to previous research (Røsstad et al. 2013; Waring et al. 2016), suggesting that further guidance may be needed regarding expectations of reciprocal communication between hospital and primary/community care providers.

Qualitative responses from 25 participants in this study also provided evidence of the organisation facilitating CDP. Communication and collaboration across all professionals, with access to multidisciplinary staff who shared a person-centred approach to discharge planning maximised the opportunity to provide CDP. Although resources and processes such as documentation and handover were reported to facilitate CDP, no participants specifically identified the Stroke Foundation’s ‘My Stroke Journey’ resource (Stroke Foundation 2019). This resource, which aims to support care planning in the transition from hospital to home (Stroke Foundation 2019), has previously been used to improve provision of discharge care plans as part of an intervention promoting collaboration in discharge planning (Cadilhac et al. 2017). The role of this and other resources in facilitating CDP and their perceived value by health professionals requires further exploration.

Other key barriers and facilitators were identified that could be addressed in a future implementation intervention. One specific barrier identified within the socio-political context domain was a lack of community services available to refer stroke survivors to upon discharge, a finding common to several other studies (Hickey et al. 2012; Waring et al. 2016; Lynch et al. 2017). Although this barrier may not be easily modifiable for individual clinicians, as it typically requires system-level change (Halladay and Bero 2000), facilitating factors, such as procedural knowledge of referral processes and available services (knowledge domain) and a positive team culture (social/professional role and identity domain), could be leveraged to decrease barriers. For example, Cadilhac et al. (2017) addressed inconsistent procedural knowledge through regular multidisciplinary team meetings to foster social support within their intervention that improved provision of discharge care plans.

The social/professional role and identity domain was identified as the main factor facilitating post-stroke CDP, with the highest percent agreement (93.15%) and several qualitative responses. This reflects findings from Lynch et al. (2017), who found that adherence to a rehabilitation assessment tool in acute stroke units was strongly influenced by whether participants perceived it as part of their role. Methods for optimising this domain should be considered, such as role clarification within procedural and governance documents, which could contribute to a standardised discharge protocol. As positive intentions were also reported to facilitate CDP, a social learning environment with well-intentioned health professionals may be an effective way to improve practice. This strategy is supported by a theory-informed overview of systematic reviews, which showed that implementation strategies requiring health professionals to learn in a social environment to enact new practices had the greatest impact on behaviour change (Johnson and May 2015). A novel finding from our study was that positive emotions also facilitated CDP, suggesting that leveraging this domain through strategies such as promoting positive patient stories may be important to successful implementation.

Implications and future directions

This study expanded on the work by Cadilhac et al. (2017) by exploring health professionals’ perspectives more broadly across Australia to reinforce the presence of an evidence-practice gap in post-stroke discharge planning processes. Our study highlights ongoing perceptions of inconsistent coordination of discharge planning in post-stroke care, suggesting that clearer guidance and support is needed to improve communication between hospital-based clinicians and primary and community care providers. These inconsistencies in CDP practices have the potential to negatively impact the discharge experience and associated outcomes for stroke survivors and their families. Although the Australian and New Zealand Living Clinical Guidelines for Stroke Management (Stroke Foundation 2017, 2023) state that ‘discharge planning relies on effective communication between team members, stroke survivors, families/carers, and community service providers including general practitioners’, health professionals are in a difficult position to achieve CDP without further support from the healthcare systems in which they work. Improved integration of systems and processes are likely required to support reciprocal communication and collaboration between inpatient and community care providers. In addition, future research involving the development and evaluation of a tailored implementation strategy to address the key barriers and leverage the key facilitators identified in our study is required.

Conclusion

There are inconsistencies and evidence-based practice gaps in post-stroke CDP processes. Despite hospital-based health professionals perceiving CDP as part of their social/professional role and identity, and generally having the knowledge and intentions to engage in CDP processes, there are numerous barriers to practice. In particular, post-stroke CDP is limited by the socio-political context, health professionals’ beliefs about capabilities, and aspects of the organisation. ‘Best practice’ in post-stroke discharge planning processes may be achieved by addressing the identified barriers and facilitators through future development and evaluation of a tailored implementation strategy.

Supplementary material

Supplementary material is available online.

Data availability

The data that supports the findings of this study are available in the tables and supplementary material of this article.

Conflicts of interest

The authors declare no conflicts of interest.

Declaration of funding

This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Ethics standard

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

Author contributions

All authors read and approved the final paper.