Design

As recommended within the MRC framework [27], this process evaluation integrated qualitative and quantitative methods in the attempt to capture both the quality and quantity of the intervention. The study protocol is available elsewhere [28]. While no specific guidelines are available for the reporting of process evaluations [27], this study was reported in accordance with the Standards for Reporting Implementation Studies (STARI) statement [30] and the Criteria for Reporting the Development and Evaluation of Complex Interventions in Healthcare (revised guideline CReDECI 2) [31], where appropriate. A full STARI statement can be found in the S1 Table.

Context

The process evaluation was conducted in the ED of a hospital with a large catchment area in the Mid-West Region of Ireland, where the intervention was carried out; the ED provides acute care to over 60,000 adults and children a year. Older adults represent approximately 30% of presentations in this setting [32] and are more likely to present with a complexity of functional, medical and psychosocial issues. This cohort were identified as a target group who were most likely to benefit the most from this model of care in our extensive stakeholder engagement meetings prior to undertaking the RCT [17]. The intervention consisted of a dedicated team of HSCPs providing early assessment and intervention to older people aged ≥65 years presenting to the ED with complex care needs. The impact of the intervention when compared to usual care was tested on our two primary outcomes of ED length of stay and hospital admission rates. Secondary outcomes focused on a range of other patient, process, and clinical outcomes.

Participants

The study involved 26 participants (57% female) who took part in four focus groups (n = 20) and 1:1 interviews (n = 6). The sample included:

• the three HSCPs composing the team who carried out the intervention (one senior medical social worker, one senior occupational therapist, one senior physiotherapist);
• 10 ED doctors;
• 10 ED nursing staff members, including and the research nurse employed in the OPTI-MEND trial;
• three hospital staff members, including two Heads of Department and one Data Management officer, who contributed to the development and implementation of the intervention.

The four focus groups comprised the HSCP team, eight ED doctors, and nine nurses. The research nurse involved in the trial, two ED doctors, and the three hospital staff members completed 1:1 interviews.

The participants were recruited through purposive, convenience and snowball sampling, with the clinical and research team acting as gatekeepers; study leaflets were also distributed in the ED. All participants were provided with an information sheet outlining the aims and methods of the study and all signed a written informed consent. Ethics approval for the study was received from the HSE Mid-Western Regional Hospital Research Ethics Committee (REC 103/18) in September 2018.

Explored domains

In line with the study objectives and protocol [28], the domains of interest for this process evaluation included the implementation of the project, the mechanisms of delivery and impact, and the contextual influences on the intervention, as shown in Table 1. A detailed account of data sources and analysis types for each domain are available in the study protocol [28].

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Process evaluation domains.

https://doi.org/10.1371/journal.pone.0269117.t001

To avoid confusion, the following definitions were developed and agreed by the research team and are used hereafter:

• Stakeholder: Any individual who interacted with and/or facilitated the implementation of the HSCP team intervention in the ED, hospital staff; clinicians outside the hospital; researchers involved in the project.
• Participant: Any individual who took part in the process evaluation. This group may include stakeholders.

Data collection and analysis

Data collection for this study took place between June and July 2019. The main form of data collection for this study was qualitative, through focus groups and semi-structured interviews, to explore the direct experiences of individuals who had carried out the intervention or interacted with it in the ED. Using both focus groups and interviews enabled better accommodation of participants’ needs with regards to their availability and their preference for sharing their experience with others or individually. Qualitative data regarding the development and delivery of the intervention were integrated with quantitative data and notes from the recruitment and activity logs to develop a description of fidelity, dose, and reach.

A full interview schedule is included in the study protocol [28]; this was used as a guide to ensure that the three areas of interest of our investigation (implementation, mechanisms and context) were covered during the interview. The interviews and focus groups were audio recorded for verbatim transcription. One member of the research team (MC), who has experience in qualitative methods in the area of health, transcribed and analysed the qualitative data using the software NVivo version 11 Plus (QSR International Pty Ltd). Another researcher (RG) contributed to the data analysis and interpretation. The transcribed data were analysed iteratively and inductively, guided by the MRC framework, in accordance with the six steps of thematic analysis [34, 35]. In line with Saunders et al. [36], data saturation was agreed by the two researchers (MC and RG) during data collection when it was observed that new data repeated what had been already expressed in previous responses.

In conducting the analysis, reflexivity was achieved by considering the researchers’ background and past experiences within the setting of the study (i.e., previous ED service user) in the design and conduct of the study. Further strategies to reduce response or analysis biases included the use of the MRC framework, as well as sharing the design and analysis of the study with the full research team.

In accordance with Data Protection policies, the study participants provided written informed consent to sharing anonymised extracts of interview transcripts; access to the full dataset is limited to the research team members who collected and analysed the data (MC and RG).

Findings

An overview of the implementation of the intervention together with the mechanisms and context of impact is presented in Fig 1. In this section, we firstly describe and quantify the implementation of the HSCP intervention, and then discuss key themes related to the mechanisms and contextual factors of impact.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Implementation framework of a HSCP team intervention for older adults in the ED [27].

(+) indicates an enabler, (-) refers to a barrier.

https://doi.org/10.1371/journal.pone.0269117.g001

Process–developing the HSCP team intervention in the ED

A detailed account of the implementation process is included in the S1 File. As shown in Table 2, the process of implementing the HSCP intervention was structured in three main phases. In the pre-implementation phase, research and communication activities were crucial to establish a clear evidence base to guide the intervention, to determine the structures and the supports needed for a successful implementation, to raise awareness and acceptability for the intervention among ED and hospital staff, and to identify the criteria for recruiting the most appropriate staff in the HSCP team.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Implementation process.

https://doi.org/10.1371/journal.pone.0269117.t002

The piloting phase, after recruitment of the three HSCPs (one Senior Medical Social Worker, one Senior Occupational Therapist and one Senior Physiotherapist) and space allocation in the ED, was crucial for the team to develop an interdisciplinary approach informed by existing models of care, get integrated into the ED environment and identify the most optimal care pathways and criteria. At the same time, liaison between the HSCP team, research team and support departments in the hospital enabled the intervention to work effectively within a trial.

During the intervention delivery (phase 3), the ongoing engagement between the HSCP team and other relevant stakeholders and parties was fundamental to ensure the appropriate operations both for the intervention and the trial.

Intervention delivery: Fidelity, dose, and reach

A detailed description of the assessment and interventions delivered by the HSCP team can be found in the trial report [29]. The target population for the intervention were older adults aged ≥65 years with a variety of index complaints; further details on patients’ inclusion criteria are described in detail in the trial report [29]. As part of the intervention, patients received holistic assessment by one or more members of the HSCP team with regards to their mobility, functional, cognitive, and psychosocial abilities. The team operated following an interdisciplinary approach whereby procedures were agreed as a team and decisions were made by a key case worker. Based on the results of the assessment, interventions and individualised discharge care plans were prescribed by the HSCP team in collaboration with the ED medical and nursing team.

The main findings related to fidelity, dose and reach are described in Table 3, and a detailed account can be found in S1 File.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Intervention delivery.

https://doi.org/10.1371/journal.pone.0269117.t003

In terms of fidelity, the HSCP team felt that that engagement with academic evidence and existing models of interdisciplinary care were critical to ensure that the content of the assessment and the target population were in line with international and national healthcare standards. However, adaptations to the assessment procedures and screening documentations were identified during the piloting phase, and communication with ED and hospital staff was pivotal to optimise care procedures (see S1 File for further details). These adaptations helped the team to share with the medical team a more comprehensive analysis and clearer recommendations for the patients, and thus enabled more timely care decisions.

Considering dose, the intervention was delivered during normal business hours, with 176 out of 214 eligible patients (82%) being included in the intervention. The team’s operating hours were described by many participants, including the HSCP team, as an important factor of dose and reach. For instance, anecdotally the HSCP team members reported that many eligible patients were GP referrals presenting in the afternoon, but these could not be picked up by the team due to the 5pm finish time. Participants identified important contextual barriers to this, including the limits of the scope of practice of allied health professionals, as well as the absence of community services outside normal business hours, which would hinder the ability of a HSCP team to discharge the patient safely.

Furthermore, while the patient age inclusion criterion (≥65 years) had been chosen to reach older adults who could benefit the most from the comprehensive assessment provided by a team of healthcare professionals [37, 38], it was felt that this may have limited reaching patients younger than 65 years old who would benefit from engaging with the HSCP team.

Lastly, an important factor of reach and dose was the delivery of the HSCP intervention within a randomised controlled trial, which is discussed in detail in section “Theme 3”.

Mechanisms and context of impact: Key themes

The participants’ views on the HSCP intervention were overall positive, identifying benefits of the ED-based HSCP team for patients in terms of safer discharges and for staff members as to reduced workload and increased confidence in decision-making.

Four main themes were identified based on the accounts provided by the participants in this study:

1. The team’s motivation and proactive communication promoted acceptability and integration in the ED;
2. The team’s specialised skills and interdisciplinary approach enhanced patient and staff’s ED experience;
3. The project stakeholders’ investment was a key enabler of implementation and acceptability;
4. Using a trial format promoted credibility but caused frustration among patients and staff.

A table of quotes for each theme is included in S2 File.

Summary of findings

This mixed-methods process evaluation aimed to describe the implementation and factors of impact of an ED-based HSCP team intervention providing early assessment and interventions to older adults with frailty and complex needs.

Considering the implementation process, the two phases preceding the intervention delivery (i.e., pre-implementation and piloting) were described as essential to the successful implementation of the intervention because enabling the gathering of inputs from relevant stakeholders, effective preparation, team building and buy-in from existing staff. The content and target population of the intervention aligned with existing evidence and practice [11, 12, 40], but some operational adaptations were necessary (e.g., assessment tool) to tailor the team’s procedures to the needs and dynamics of the ED. The team’s operating hours and patient inclusion criteria were perceived as limiting dose and reach, but at the same time they enabled the team to identify the patients most in need and dedicate enough time for an in-depth and comprehensive assessment.

In terms of mechanisms of impact, our participants felt that the HSCP team’s motivation and relational skills enabled them to get easily accepted and integrated in the ED team; also, their specialised skills and interdisciplinary approach contributed to enhance the ED experience of both patients and other staff members, promoting safer discharges, patient-centred care, reduced workload for the medical team and confidence in decision-making. Notably, our participants expressed positive views without knowing objective results on the intervention outcomes (i.e., trial results not available at the time of data collection), which highlights a positive impact of the intervention beyond clinical and cost effectiveness.

Participants expressed mixed feelings about introducing the HSCP intervention in the ED within a randomised controlled trial, appreciating the positive impact of the rigorous method on the profile of the intervention, but at the same time acknowledging some frustration among patient and staff due to the randomisation and trial procedures. At a contextual level, it was felt that the success of the implementation would not have been possible without the support and collaboration of invested and motivated stakeholders in the hospital and the broader healthcare setting, who facilitated the introduction of the HSCP team in the ED both at an operational and relational level.

Results in the context of existing evidence

Although stakeholders’ perspectives on HSCP services in the ED have been increasingly investigated [11–13, 17, 41], comparisons with previous literature are limited by the absence of a previous process evaluation of this model of care in the literature. Nonetheless, our findings on the implementation and mechanisms of impact of this intervention resonate with the available evidence in terms of positive views on HSCPs in the ED [17, 41].

With regards to the implementation process, the reported importance of having a pre-implementation and piloting period gathering inputs from multiple stakeholders, together with the perceived benefits of stakeholders’ involvement and collaboration (Theme 4), support previous studies [24, 40] that have reported benefits of consultations and building support among key stakeholders before implementing a new model of ED care.

In line with existing evidence [11, 12, 40], this process evaluation supports the appropriateness of focusing allied health services for older adults in the ED on providing comprehensive assessment and intervention to support safe discharges. Our participants’ concerns about the limitations of the HSCP team’s operating hours and the need for extended services have been expressed by both ED service users and providers in several studies [11, 12, 24, 41], and it was observed in a qualitative investigation that was conducted to inform the design of the present HSCP intervention [17]; however, the evidence on the impact of ED operating hours on patient outcomes is unclear [42, 43], and changing service hours requires a considerable revision of the scope of practice of both acute allied health and community services.

Our findings on the HSCP team’s relational, professional, and interdisciplinary skills as key enablers of the successful implementation of the intervention, as well as a positive ED experiences for patients and staff, resonate with extensive evidence on the positive impact of teamwork, trustworthiness and proactivity within the ED on acceptability, buy-in and effectiveness of new models of care, particularly when changes in the culture of care are required [11–14, 41, 44, 45]. These findings point at the benefits of having a team of professionals with high motivation, a strong interdisciplinary skillset, and a patient-centred approach for the implementation of a collaborative intervention in a complex and dynamic environment such as the ED.

Despite these benefits, our study resonates with previous studies suggesting that interdisciplinary allied health interventions for older adults in the ED would be further optimised if providing care integration and community follow-ups together with a more advanced scope of practice for HSCPs and the support of geriatric specialist services [4, 46–49], although previous investigations have noted potential issues related to funding and sustainability [49]. More broadly, the logistic and operational issues associated with introducing a new model of care to a complex healthcare setting within randomised controlled trial have been acknowledged in previous research [50, 51], pointing at the need to consider more pragmatic methods [42].

Strengths and limitations

To the best of our knowledge, this is the first structured process evaluation of a dedicated HSCP team providing early assessment and intervention to older adults in the ED. In this process evaluation, we adopted a standardised framework [23] and sought inputs from an interdisciplinary steering group of experts in allied health and emergency care to carry out a rigorous and comprehensive investigation of the multiple factors influencing the development and delivery of a new model of care in a complex healthcare environment. Using a mixed-methods approach enabled us to capture information both on the quality and quantity of the intervention; also, conducting both group and individual interviews helped our participants’ recruitment as well as enriching our data collection.

The findings of the study should be considered in light of some limitations. Firstly, the process evaluation did not involve ED patients or carers, as their perceptions of the intervention have been investigated via a survey in the clinical effectiveness study [29] and it was felt that involving patients also in the process evaluation may cause an additional burden on potentially vulnerable individuals; nonetheless, we acknowledge that using accounts from staff members may have affected the accuracy of our findings on patients’ experiences. Furthermore, exploring the views of community stakeholders (not involved in this study given the focus on the ED experience) could have enriched our understanding on the contribution of the intervention to patients’ safe discharges. Our considerations on the implementation fidelity of the intervention could not be quantified due to the lack of an appropriate tool for this kind of intervention. Also, the generalisability of our findings may be limited given the nature of the intervention and the setting; contextual influences within and outside the ED may impact the implementation of this type of intervention elsewhere. Nonetheless, we found that the characteristics of intervention, as well as the mediators of impact, were in line with existing practice and evidence.

Implications for practice

The insights gathered in this study suggest that a HSCP team intervention aimed at older adults in the ED is a viable strategy to improve the experiences of patients and staff members, provided adequate time for preparation and the recruitment of skilled and motivated professionals who are capable of building positive relationships and foster an interdisciplinary and patient-centred ethos in their practice. Furthermore, the active involvement of relevant stakeholders in the pre-implementation phase is essential to facilitate acceptance and the intervention operations. Our findings support the added value of introducing an interdisciplinary team of HSCPs over individual professionals for enhancing the of older adults with complex needs, because promoting collaborative decision-making and integrated care. Although funding and financial factors were not the focus of this study, understanding the economic costs and benefits associated with an interdisciplinary team-based model of care for this population is key to guide future implementation. A cost effectiveness evaluation is currently in preparation which will provide an account of this.