Introduction
Moving from hospital to home for people with advanced cancer can be a time of heightened stress and uncertainty for both patients and their carers, where most of the care responsibilities shift onto family or friends (Adejoh et al. Reference Adejoh, Boele and Akeju2021). This transition of care is associated with uncontrolled symptoms for the patient and excessive burden on carers who are frequently unaware of what to expect in taking on the caring role for a person with advanced cancer (Lawson et al. Reference Lawson, Burge and Critchley2006). People with advanced cancer will experience hospital-to-home transitions across their illness trajectory, and these are particularly burdensome in the year after diagnosis (Whitney et al. Reference Whitney, Bell and Tancredi2017). Some authors have highlighted that symptom burden, disease progression, and carers reaching capacity can lead to increased hospitalization rates (Lawson et al. Reference Lawson, Burge and Critchley2006; Lage et al. Reference Lage, Nipp and D’Arpino2018).
Carers are critical to patients having a successful discharge from hospital to home at the end of life (Laugaland et al. Reference Laugaland, Aase and Barach2012; Higginson et al. Reference Higginson, Sarmento and Calanzani2013). First, they provide most of the care when a person returns home from hospital, which often incurs an emotional, physical, and financial cost (Wang et al. Reference Wang, Molassiotis and Chung2018; Adejoh et al. Reference Adejoh, Boele and Akeju2021). Secondly, their capacity to provide care remains one of the main factors that influences a person’s ability to remain at home and avoid readmission to hospital (Laugaland et al. Reference Laugaland, Aase and Barach2012; Higginson et al. Reference Higginson, Sarmento and Calanzani2013). Thirdly, they have a set of unique needs that are different from the patient and routinely remain unmet. These needs include detailed communication and information provision related to support and practical aspects of caring (Harding et al. Reference Harding, List and Epiphaniou2012; Wang et al. Reference Wang, Molassiotis and Chung2018).
International policy recommends that carers are included in discharge planning models and their support needs are addressed across the continuum of care (National Institute of Health and Clinical Excellence (NICE) 2004; World Health Organization (WHO) 2016; Naylor et al. Reference Naylor, Shaid and Carpenter2017; Carers Reference Carers2021). However, carers of people with advanced disease are under-represented in interventions and service initiatives aimed at improving hospital discharge. A recent systematic review of the reporting quality of hospital-to-home transition intervention studies found that only 22% of studies included carers in intervention design, delivery, and outcomes, and even less so for people with advanced cancer (Marston et al. Reference Marston, Morgan and Philip2022).
The Carer Support Needs Assessment Tool- Intervention (CSNAT-I) has shown promise as a potential intervention model that could be delivered in the hospital discharge setting (Aoun et al. Reference Aoun, Deas and Toye2015; Ewing et al. Reference Ewing G, Jones and Grande2018; Hall et al. Reference Hall, Ewing and Rowland2020). Originally developed for and validated with carers of people with advanced cancer in the community palliative care setting (Ewing et al. Reference Ewing, Brundle and Payne2013), the CSNAT-I is a systematic 5-stage process that creates opportunities for addressing carers’ needs and has been demonstrated to improve their ability to cope with care at home (Courvoisier et al. Reference Courvoisier, Aoun and Grande2015; Toye et al. Reference Toye, Parson and Slatyer2016; Grande et al. Reference Grande, Austin and Ewing2017). These improvements include a significant reduction in prolonged grief (Grande et al. Reference Grande, Austin and Ewing2017), carer strain (Courvoisier et al. Reference Courvoisier, Aoun and Grande2015), and sustained increased preparedness and quality of life (Toye et al. Reference Toye, Parson and Slatyer2016).
Carers and clinicians in a UK study recommended that CSNAT-I be implemented in hospital settings to expand the discharge planning focus to include carers. This could address “unrealistic expectations” of what care at home would entail and may reduce the likelihood of breakdown of care at home after discharge (Ewing et al. Reference Ewing G, Jones and Grande2018). Carers in this study also recommended implementing this tool as early as possible during hospital admission, with a review upon the initial transition home(Ewing et al. Reference Ewing G, Jones and Grande2018). Recent evaluation of the CSNAT-I implemented by palliative care nursing staff in a UK hospital revealed behavioral and organizational challenges impacting its delivery (Hall et al. Reference Hall, Ewing and Rowland2020). The structure and focus of the CSNAT-I were considered acceptable; however, there was uncertainty regarding the “fit” of carers’ needs in the health systems and which health professional should be responsible for delivery of the CSNAT-I (Hall et al. Reference Hall, Ewing and Rowland2020).
To investigate this further, we examined the CSNAT-I in an Australian inpatient acute cancer setting as part of routine care by occupational therapists (OTs; Marston et al. Reference Marston, Morgan and Philip2023). OTs play an important role in facilitating discharge home at end of life through interventions that optimize patients’ functional performance and carers’ capacity to care (Marston et al. Reference Marston, Agar and Brown2015; Eva and Morgan Reference Eva and Morgan2018) The CSNAT-I was accepted by patients, carers, and clinicians as an integrated part of existing discharge planning that should reach across hospital and home and target practical needs of caring (Marston et al. Reference Marston, Morgan and Philip2023). Study findings informed the development of a modified delivery model for the CSNAT-I – Caring for carers of people with advaNced cancer at hospiTal discharge (CARENET). The CARENET model was then used to proceed to the next phase of feasibility testing in a controlled trial. Feasibility refers to the practicality of an intervention – whether it is fit for purpose in a specific context (Bowen et al. Reference Bowen, Kreuter and Spring2009; Klaic et al. Reference Klaic, Kapp and Hudson2022). Three key areas of feasibility that were assessed were practicality, implementation, and limited efficacy based on Bowen et al.’s (Reference Bowen, Kreuter and Spring2009) framework. Practicality refers to the extent an intervention can be delivered within the given resources and circumstances. Implementation refers to the extent an intervention can be successfully delivered as intended within a defined context or according to protocol (i.e., fidelity). Limited efficacy refers to the intended effects of an intervention on relevant outcomes (Bowen et al. Reference Bowen, Kreuter and Spring2009). Thus, this study aimed to determine the feasibility of (1) delivering the CARENET intervention to carers of advanced cancer patients in an acute oncology setting as intended and (2) the study design to inform a phase 3 trial.
Methods
Design
A single-arm open-label study design was employed to determine the feasibility of the CARENET intervention and the study design. The selected study design aligns with the feasibility and piloting stage recommended in the Medical Research Council (MRC) framework for designing and evaluating complex interventions (Skivington et al. Reference Skivington, Matthews and Simpson2021). Feasibility was assessed using 3 key areas of focus recommended by Bowen’s framework for feasibility studies (Bowen et al. Reference Bowen, Kreuter and Spring2009) – practicality, implementation (fidelity), and limited efficacy, with links of areas of focus to trial aims and measures detailed in Table 1. Quantitative data were collected prospectively to address trial aims. CONSORT guidelines were used for reporting structure (Eldridge et al. Reference Eldridge, Chan and Campbell2016).
Table 1. Feasibility outcomes and alignment with research questions
AKPS = Australian Karnofsky Performance Scale; RUG-ADL = Resource Utilization Groups – Activities of Daily Living; EQ-5D-5L = EuroQol 5-Dimensions-5-Levels; SAS = Symptom Assessment Scale; CarerQoL-7D = Carer Quality of Life (QOL) Instrument.
a Supplementary File 2 for measurement properties.
Participants and setting
Participants were adult inpatients with advanced cancer and their carers (patient–carer dyad). The setting was a 30-bed medical oncology ward situated in a 140-bed specialist cancer center in metropolitan Melbourne. For the purposes of this study, carers were defined according to the NICE definition: “Lay people in a close supportive role who share in the illness experience of the patient and who undertake vital care work and emotion management” (National Institute of Health and Clinical Excellence (NICE) 2004). Advanced cancer was defined as cancer that had spread or returned (recurred) after treatment, and included locally advanced and metastatic cancer. All people with advanced cancer (regardless of prognosis) and identified carers were potentially eligible and screened for the study. Eligibility criteria included: English speaking, identified as a carer of a person with a diagnosis of advanced cancer with a planned discharge home and able to provide informed consent. Planned discharge home was defined as a decision for future discharge supported by an assessment using locally developed criteria to identify a relevant pathway home. Participants were excluded if patients had a planned discharge to residential care or subacute settings. Patients who were part of the dyad and were able to provide informed consent, or had proxy written consent, were eligible and invited to participate if interested.
Procedure
The study protocol was prospectively registered with the Australian New Zealand Clinical Trials Registry Registration No: ACTRN12622001325796. Ethical approval was obtained by the Peter MacCallum Cancer Centre Ethics Committee (HREC No. 83438 and the UTS Ethics Committee (ETH22-7440). From November 2022 to July 2023, potential participants were identified and referred for eligibility screening. Potentially eligible participants were identified by their own clinicians (medical, nursing, or allied health) and referred to the research assistant (RA) after discussions with potential participants. Eligible participants were invited to partake in the study followed by a formal consent process. Carers were consented first, then if they consented, the patient consent process would occur. The proposed sample size was 20–30 patient–carer dyad participants as per “general rule of thumb” for feasibility studies (Totton et al. Reference Totton, Lin and Julious2023) and based on pragmatic findings from the preliminary study.
Once consent was obtained, the study involved the collection of measures across 3 time points and the delivery of CARENET across hospital and home settings (before and after hospital discharge). The 3 time points for study measures (T1, T2, and T3) are defined and represented in Figure 1 with details related to the intervention as follows.
Figure 1. Intervention components (Needs Assessment, Responding to Needs, and Review) and data collection timepoints (T1 – within 1–3 days before CARENET, T2 – within 1–2 days before discharge, and T3 – after CARENET and within 1 week after discharge).
Intervention
The features of the CARENET intervention are described in Supplementary File 1 and represented in Figure 1. CARENET consisted of the CSNAT-I, and a modified intervention delivery mode whereby the CSNAT-I was delivered by OTs as an integrated part of usual discharge planning practices across hospital and home settings. During the hospital stay, carers completed 3 components: (1) an assessment and prioritization of their needs using the paper-based CSNAT-I tool (Needs assessment), (2) an action plan to address their needs (Responding to needs) via support directly from treating OTs or signposting other avenues of support for the carer to initiative themselves, and (3) within 72 h after returning home from hospital, the carers received a review of their needs via a phone call (Review). The assessment part of the CSNAT-I that was embedded in the first component consisted of 15 domains of support need for the carer to choose from. They related to the practical aspects of caring as well as their own personal needs as a carer (Ewing et al. Reference Ewing, Brundle and Payne2013). The time it took to deliver the CARENET was tailored to the individual and the period the patient spent moving from hospital to home.
Implementation process
OTs completed training and participated in a planned implementation strategy to support the delivery of the intervention. This included a 90-min individual online learning module on the CSNAT-I (2022) and 2 one-hourly group education sessions on the CARENET intervention protocol and trial processes, before recruitment commenced. During recruitment, clinician support and monitoring of adherence occurred via (1) Assignment of OT clinical champions; (2) Expert working group of OTs, researchers and consumers; and (3) File audit and informal assessment of clinician compliance (conducted by RA [M.L.T.] and primary investigator [C.M.]; Supplementary File 3).
Outcomes
The primary outcome was intervention and trial feasibility (recruitment and adherence). Secondary outcomes were eligibility/attrition (practicality) and intended effect of intervention on patient and carer outcomes (limited efficacy). Alignment of these outcomes with research questions, areas of feasibility, measures, and endpoints are presented in Table 1.
Data collection
Measures used for data collection are presented in Table 1. Patient- and carer-related data were collected via standardized measures and questionnaires. Service, process data were collected via medical records and project-specific case report forms. As illustrated in Figure 1, data collection at the 3 time points (T1, T2, and T3) for study measures was as follows: T1 1–3 days before intervention commenced; T2 1–2 days before discharge, and T3 after CARENET and within 1 week after discharge. Patient and carer outcome measures were the only measures collected across all 3 time points. Patient-related measures were the Australian Karnofsky Performance Scale (AKPS) (Abernethy et al. Reference Abernethy, Shelby-James and Fazekas2005), Resource Utilization Groups – Activities of Daily Living (RUG-ADL) (Fries et al. Reference Fries, Schneider and Foley1994), Symptom Assessment Scale (SAS) (Daveson et al. Reference Daveson, Allingham and Clapham2021), and EuroQol 5-Dimensions-5-Levels (EQ-5D-5L) (Herdman et al. Reference Herdman, Gudex and Lloyd2011). Carer-related measures were Preparedness for Caregiving Scale (Archbold et al. Reference Archbold, Stewart and Greenlick1990) and Carer Quality of Life (QOL) Instrument (CarerQOL-7D) (Hoefman et al. Reference Hoefman, van Exel and Brouwer2017). The main carer-related outcome was carer-preparedness measured via Preparedness for Caregiving Scale (Archbold et al. Reference Archbold, Stewart and Greenlick1990), where a high score reflects improved preparedness, which was the only measure collected across all 3 time points. Features of these valid and reliable measures are summarized elsewhere (Supplementary File 2). Demographic data were collected at baseline (T1), while adherence, process, and service data were collected at T3.
Data analysis
Data analysis was conducted using Excel (Microsoft Office V23.08) and IBM SPSS v28 software. Mean and standard deviation were reported for continuous normally distributed data, and interquartile range and median for data deviating from normal distribution. Results of the categorical variables (eligibility, recruitment, attrition, and intervention adherence) were presented as percentages and frequencies. All questionnaires were scored according to their respective scoring guidelines. To evaluate changes in patient- and carer-related measures (i.e., functional performance, quality of life, and carer preparedness) between T1 and T3, we calculated difference scores and 95% confidence interval (CI). Additionally, we conducted paired sample t-tests and Wilcoxon signed-rank tests to assess the statistical significance of these changes.
Results
Participant recruitment and eligibility
Twenty-five dyads were recruited across the 8-month accrual period (Fig. 2). A high number (n = 336; 87%) of identified eligible participants were not invited to participate. Rapid and unpredicted hospital discharge was the primary reason (54%) for not inviting identified eligible participants, followed by the RA not being available (35%). There was a 54% consent rate for the remaining dyads. Figure 2 shows lower completion rates of patient- and carer-related measures at T2 (n = 4, 16%) compared to T1 (n = 24, 96%) and T3 (n = 13, 52%). Reasons included RA unable to contact the carer at time of discharge.
Figure 2. Recruitment and participation. T1 = timepoint 1 – 1 day before CARENET, T2 = timepoint 2 – 1–2 days before discharge, T3 = timepoint 3 – after CARENET and within 1 week after discharge.
Participant characteristics
Key characteristics of patient–carer dyads are shown in Tables 2 and 3. Carer participants had a mean age of 56.79 (SD = 15.54), were more likely to be female (56%, n = 14), were born in Australia (76%, n = 19), were English speaking at home (80%, n = 20), and were a spouse or partner to the person they were caring for (72%, n = 18). Seventy-two per cent (n = 18) lived with the person they cared for and 48% (n = 12) were still employed. Patient participants had a mean age of 59.72 (SD = 15.89), almost two-thirds (64%, n = 16) were female. Most patients spoke English at home (84%, n = 21), more than half were born in Australia (64%, n = 16), and 40% (n = 10) were still employed. There was high variance (heterogeneity) in the participant diagnosis, a median (IQR) length of stay in hospital of 17.0 (IQR = [14.5, 34.5]) days, and home was the discharge destination for 76% (n = 19) of patient participants.
Table 2. Demographic characteristics of consented carer participants (n = 25)
Table 3. Demographic characteristics of consented patient participants (n = 25)
IQR: interquartile range.
a Other = Breast, head and neck, urological, cancer of unknown primary, neurological.
Intervention adherence
Partial adherence was reached with 60% of enrolled participants completing all 3 intervention components (Fig. 2). Adherence to components occurring before discharge was higher than the review component occurring after discharge, with 84% of carer participants completing the needs assessment and 68% completing the needs assessment and creating an action plan to respond to needs. Reasons for non-adherence included patient death prior to completing the intervention (n = 5), patient discharge too early (n = 1), transfer to palliative care ward (n = 1), unable to contact carer (n = 2), and unknown (n = 1).
Limited efficacy
Table 4 presents patient and carer outcomes (n = 12) assessed before (T1) and after (T3) the CARENET intervention. Upon returning home (T3), patients demonstrated a trend toward increased care needs compared to their hospital stay (T1), although they remained able to manage their own needs with limited physical assistance (AKPS mean [SD] = 68.89 [12.96] at T1 vs. 61.11 [24.21] at T3). Symptom burden also showed an upward trend upon returning home (SAS mean [SD] = 6.44 [4.56] at T1 vs. 11.11 [12.23] at T3).
Table 4. Patient and carer outcome measures taken before (T1) and after CARENET (T3) (n = 12)
AKPS = Australian Karnofsky Performance Status, RUG-ADL = Resource Utilization Group – Activities of Daily Living, EQ-5D-5L = EuroQol 5-Level-5-Dimensions.
a 0–4 scale, response options 0 = not at all to 4 = very well prepared, MCID = 0.25 point difference per item or more than 2 point difference overall (REF).
Carers exhibited mixed outcomes, reporting a decreased sense of preparedness for addressing patients’ emotional needs (mean [SD] = 2.50 [1.01] at T1 vs. 2.25 [0.87] at T3) but improved preparedness for practical caregiving tasks, including obtaining system support (mean [SD] = 2.75 [1.06] at T1 vs. 3.17 [0.72] at T3), managing emergencies (mean [SD] = 2.64 [1.03] at T1 vs. 3.09 [0.83] at T3), finding services (mean [SD] = 2.75 [0.87] at T1 vs. 2.92 [0.79] at T3), and providing physical care (mean [SD] = 2.50 [0.91] at T1 vs. 2.83 [0.84] at T3). The Preparedness for Caregiving Scale revealed increases in mean scores across practical domains, although these changes were not statistically significant. Overall preparedness improved marginally from 2.75 (SD = 0.86) at T1 to 3.25 (SD = 0.62) at T3 (p = 0.05).
Notably, carers’ quality of life showed a significant improvement, increasing from 78.95 (SD = 14.09) before CARENET (T1) to 79.45 (SD = 12.45) after CARENET T3 (p = 0.01). While trends toward improvement were observed across several domains, including physical care, support from the health system, and overall preparedness, statistical significance was limited to carers’ quality of life and marginally to overall preparedness.
Discussion
This was the first phase 2 trial to test the feasibility of delivering the CSNAT- I as an integrated part of discharge planning practices to assist with discharge from hospital to home for carers of people with advanced cancer. Our findings have shown that while the recruitment target was reached, there were several challenges to the feasibility of the intervention and study conduct.
First, a limited group of carers received the intervention, while a vast number of eligible carers did not participate. This is almost certainly due to a mismatch between the hospital discharge context and the trial’s eligibility procedures. Rapid discharge home for people with palliative care needs is commonplace in acute hospitals and can happen quickly, often within 24 h (Department of Health Victoria 2021; Australian Institute of Health and Welfare 2024). We found that for people planning to return home in 24 h there was insufficient time to complete all aspects of the intervention and study measures according to the protocol. Furthermore, the limited grant funding restricted the RA’s availability to 2 days per week and hindered the ability to reach all eligible participants. Another reason for smaller recruitment numbers could be clinician “gatekeeping” that excluded patient carers with perceived high levels of distress. Another CSNAT-I trial reported that clinicians who seek to avoid overwhelming carers may inadvertently limit support for those with greater needs (Lund et al. Reference Lund, Ross and Petersen2020).
Secondly, in this trial, delivery of all intervention components as intended was difficult, particularly the review after discharge. Recorded reasons for not completing all components (i.e., availability of carers and patient death) were like other trials delivering adapted CSNAT-I models (Toye et al. Reference Toye, Parson and Slatyer2016; Patchwood et al. Reference Patchwood, Woodward-Nutt and Rhodes2021). However, it is possible that problems with delivering the CARENET as planned may be because of the shortfalls in the implementation process and subsequent impact on clinicians’ proficiency. The recommended implementation and training strategy was used for the CSNAT-I component (The University of Manchester/ University of Cambridge 2022). However, this was impacted by high staff turnover, as well as reduced access to clinical champions, and the research staff for support and fidelity monitoring. Previous CSNAT-I studies recommend that these features are important for successful implementation in both trial and usual practice settings (Diffin et al. Reference Diffin, Ewing and Harvey2018; Hall et al. Reference Hall, Ewing and Rowland2020).
It is also possible that difficulty with delivering CARENET could be attributed to using a protocolized, multi-staged delivery model that is not congruent with the fast-paced and dynamic nature of the hospital discharge context. Discharge home from hospital can occur along different pathways and timelines, compared to more stable and predictable community and home-based settings where most trials testing the CSNAT-I have previously occurred (Aoun et al. Reference Aoun, Deas and Toye2015; Grande et al. Reference Grande, Austin and Ewing2017; Lund et al. Reference Lund, Ross and Petersen2020).
Thirdly, the study showed that there were difficulties with the completion of the selected outcome measures. Obtaining data from patients and carers was difficult to achieve at discharge (T2), compared to before and after intervention (T1 and T3). The positive trends observed for carer quality of life and overall levels of preparedness after intervention align with findings of other CSNAT-I trials (Toye et al. Reference Toye, Parson and Slatyer2016; Grande et al. Reference Grande, Austin and Ewing2017), supporting the use of these outcome measures to test the impact of carer-targeted interventions. However, results must be interpreted with caution, given the small sample size and the large group of carers deemed eligible but discharged prior to recruitment
Implications and future directions
Our trial has highlighted the opportunities and the challenges of conducting carer-focused interventions in a compressed and unpredictable hospital discharge context. Foremost is the need to consider a study design that will increase access to a more diverse group of patient–carer dyads (i.e., age, ethnicity, and higher care needs) and ultimately improve the clinical translation of this type of intervention. This could involve using a more flexible, multi-modal eligibility process that allows recruitment to take place at various times and in settings that align with carers’ needs and their different pathways home (Gustavson et al. Reference Gustavson, Horstman and Cogswell2024). An example of this could be an opt-out “early” approach, where carers are automatically recruited to the study upon patient admission and have the choice to opt out. This opt-out “early” approach may also overcome barriers from limited resources and staff availability (Treweek et al. Reference Treweek, Lockhart and Pitkethly2013)
A modifiable intervention model could also improve access for all carers who need support at discharge. This could entail intervening with carers at different levels with tailored delivery modes to accommodate their needs and circumstances. Research into carers of older hospitalized people supports flexible and adaptable models that offer single and multistage approaches and are considered feasible and effective (Levoy et al. Reference Levoy, Rivera and McHugh2022). For carers of hospitalized people with advanced cancer, further research is required to understand the degree of modifications that can occur without compromising the efficacy of the intervention itself (Levoy et al. Reference Levoy, Rivera and McHugh2022). This could include a review of what aspects of CARENET are considered core and those where flexibility of inclusion may apply. Tools to assist with adapting existing interventions are recommended to systematize this process and achieve a balance between access and impact (Kirk et al. Reference Kirk, Moore and Wiltsey Stirman2020).
Stepped-care intervention models could offer a stratified approach where the level of care (or carer support) is matched to the level of complexity of the condition or individual need (General Practice Mental Health Standards Collaboration (GPMHS) 2017). Stepped care models are well established in mental health care and are used to optimize access and efficiencies within given resources (Berger et al. Reference Berger, Fernando and Churchill2022; Mughal et al. Reference Mughal, Salmon and Churchill2023). Similarly, cancer prehabilitation models use a stepped care approach (MacMillan Cancer Support 2019) where all people having surgery are offered prehabilitation, but scope and intensity will vary. CARENET could similarly be implemented in this stepped-care approach, whereby intensity of support may differ according to need. For example, carers of people experiencing escalating carer needs would need a greater intensity of intervention to assist with discharge planning compared to carers of people whose care needs remain unchanged. If this approach were applied, additional feasibility studies would be needed to ensure intervention models are fit for purpose before proceeding to a larger scale study.
It can also be argued that understanding the impact of a person-centered intervention like CARENET should be informed by the carers’ engagement with the intervention, not just adherence to the intervention protocol. Assessing fidelity of enactment is considered a measure of engagement, specifically whether the intervention content leads to a change in behavior that is then enacted in daily life (Walton Reference Walton2024). Even though consensus has not been reached in the literature regarding the best way to measure enactment, there is a shared opinion that high levels of enactment equate to good outcomes. Given that CARENET relies on carers responding to the needs assessment and the support offered for it to work, this aspect of fidelity should be considered in future feasibility study design alongside fidelity of delivery and receipt (Walton Reference Walton2024).
Several limitations of this study should be noted. First, while the trial design testing feasibility is aligned with recommendations of the MRC framework (Skivington et al. Reference Skivington, Matthews and Simpson2021), the single-arm design with no comparable control group limited our scope to assess against usual care outcomes. Secondly, the generalizability of the findings to other carer groups and different hospital settings is limited. Thirdly, the small sample size limits the ability to comment upon the efficacy of CARENET on patient–carer outcomes. However, this was a feasibility trial where a pragmatic sample size was selected to focus on “limited efficacy” (Bowen et al. Reference Bowen, Kreuter and Spring2009) and help determine where the conduct of chosen outcome measures was practical and fit for scaling up for large efficacy trials.
Conclusion
In conclusion, our trial supports the understanding that carers of people with advanced cancer or palliative care needs will continue to need significant support during the transition from hospital to home, no matter what the context. However, delivering a protocolized multi-staged intervention for carers in this discharge setting may have limited feasibility and compromise access to more support for all carers. Further exploration of what intervention content and delivery is needed in order to benefit all carers and the people they are caring for, before proceeding to larger-scale trials.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1478951525100710.
Acknowledgments
The authors extend their appreciation to all patients and families who participated in this study. Thanks to the carer consumers (Catherine Nguyen and Ros Alcot), Professor Brian Le, Catherine Ludbrook, Occupational Therapy Department at Peter MacCallum Cancer Centre for their inputs into the study. Special acknowledgment to the original developers of the Care Support Needs Assessment Tool – Intervention (CSNAT-I), Dr Gail Ewing and Professor Gunn Grande who provided support and mentorship throughout the study.
Author contributions
All authors meet the criteria for authorship as defined by the International Committee of Medical Journal Editors (ICMJE). C.M., M.R.A., J.P., and D.M. were responsible for study concept and design, and drafting of article. C.M. and M.L.T. were responsible for the acquisition of the data. All authors contributed to the analysis and interpretation of the data and critical revision of the article. All authors approved the final version of this article to be published.
Funding
The conduct of the study was supported by a Peter MacCallum Cancer Foundation Grant 2022/2023, ID#2149, awarded to C.M.
Competing interests
There are no conflicts of interest to declare.
Open access
