Implementation of activities of daily living retraining for individuals in post-traumatic amnesia

Participants

Eighty OTs located across Australia completed workshop training in the PTA ADL intervention in 2017. Of these, 44 completed an anonymous survey at the time of the workshop and consented to research participation. Two years later, multi-stakeholder interviewee participation was purposively sought to span domains from the CFIR and garner varied perspectives regarding implementation. Seven OTs from the training workshop, two further OTs (one involved in the trial and one who received workplace training equivalent to the workshop) and two administrators (OT manager at a centre with TBI rehabilitation and a research funder involved in program translation) engaged in an interview (n = 11). Participant demographics are shown in Table 1. Participants who completed an interview were offered a AU$50 voucher.

Design and materials

A sequential explanatory mixed methods design was used to evaluate OTs’ perception of effectiveness of attending a training workshop in improving knowledge and skills in PTA ADL retraining quantitatively and then capture qualitative perspectives regarding implementation of the intervention 2 years following the workshop. The study has been reported according to the Good Reporting of a Mixed Methods Study (GRAMMS) checklist (O’Cathain et al. 2008; Supplementary File S1).

Procedures

Ethics approval was obtained from Monash University Human Research Ethics Committee (#11270). Written or electronic consent was provided by all participants. Participants who attended the PTA ADL workshop completed a survey and self-rating competency evaluation prior to and immediately following the workshop. At 2-year follow-up, participants were invited via email, to partake in an interview exploring their experiences utilising PTA ADL retraining. Following training, interviews were performed by KS with iterative feedback. Interviews were conducted via videoconference (n = 4; laptop or desktop) or telephone (n = 7). Interviewees participated in the interview at their workplace and the interviewer conducted the interview in a private meeting room or from home. Field notes were made and all interviews were recorded and professionally transcribed.

Data analysis

R Studio (2018) was used to analyse survey data. Descriptive statistics were calculated and repeated measure t-tests conducted to compare pre-training and post-training ratings.

An inductive approach to thematic analysis (Braun and Clarke 2006, 2021a) was employed to explore the use of PTA ADL retraining. Inductively derived themes were then mapped deductively to the CFIR domains to consider overarching barriers and facilitators of implementation. A critical realist ontology was adopted assuming access to participants’ perceived reality whilst acknowledging this is shaped by culture, language, context (Alvesson and Sköldberg 2017; Braun and Clarke 2021b), and interpretation by researchers (Braun and Clarke 2023). Braun and Clarke’s (2013) six phases of thematic analysis were followed, beginning with data familiarisation. Following transcript reviews, coding was performed using NVivo 12 software (QSR International) by KS with semantic and latent codes generated. Codes were refined with a codebook developed as per Mac-Queen et al. (1998). A second coder independently coded four randomly selected interviews (O’Connor and Joffe 2020). Intercoder reliability was calculated using a Cohen’s kappa statistic, ranging from 0 (no agreement) to 1 (perfect agreement) (Viera and Garrett 2005) and indicated moderate agreement (M = 0.56; range = 0.22–0.82). Discrepancies were discussed amongst the research team and reconciled to improve coding reliability and thematic validity. Themes were developed, compared to raw data, refined, and labelled through iterative team discussion (Braun and Clarke 2006, 2021a; Chenail 2011). Researcher subjectivity and their relationships with the investigated topic were considered throughout analysis (Braun and Clarke 2023). Alongside KS, the three authors engaged in thematic construction were also involved in the original RCT and training of clinicians, all with extensive backgrounds in TBI rehabilitation (JTP – female, clinical neuropsychologist and occupational therapist, BOT/DPsych; JP – female, clinical neuropsychologist, PhD; AM – male, clinical neuropsychologist, PhD). Actively involved in the research process, the researchers revisited the research question and interviews, fostering reflexive engagement with the data. Member checking was not feasible for the Honours student and, given theoretical inconsistencies with thematic analysis, this was not pursued by the authors to avoid potentially restricting theme construction (Morse et al. 2002; Varpio et al. 2017; Braun and Clarke 2019). Theoretical sufficiency (Dey 1999) as per deductive mapping to the CFIR was considered methodologically consistent for this study, as opposed to data saturation (Low 2019; Braun and Clarke 2021c). Data collection and analysis were informed by ‘meaning-richness’ to create a sufficient account of implementation of the intervention according to the CFIR (Dey 1999; Braun and Clarke 2022).

Training

Of the 44 OTs who completed a survey prior to training, 33 (76.74%) endorsed wanting to engage in more ADL retraining with individuals in PTA. Pre-training, on average they completed ADL retraining with approximately 12.83% (s.d. = 21.35%) of patients in PTA. OTs identified the following barriers in undertaking ADL retraining during PTA: time (65.91%), resources (36.36%), training (22.73%), confidence (22.73%), and it being beyond the scope of their role (11.36%). On a scale of 0 (very poor) to 10 (excellent), the quality of training within their completed Occupational Therapy degree relating to skills retraining following brain injury/severe cognitive impairment was rated as relatively poor (M = 3.32; s.d. = 2.71; range 0–9).

Forty-two of the 44 OTs completed the post-training survey. Overall, workshop training was rated to be helpful in improving knowledge, skills and confidence in PTA ADL retraining (M = 4.74; s.d. = 0.44; scale 1 (not at all useful)–5 (very useful)). Self-ratings of confidence in ADL retraining during PTA (0 (not at all confident)–10 (extremely confident), rated by 41 OTs) significantly improved from baseline (M = 4.48; s.d. = 2.4) to post-training (M = 7.70; s.d. = 1.24), t(40) = 9.78, P < 0.001. Competency self-assessment ratings in areas of knowledge and skills covered in the training workshop are summarised by domain in Fig. 2. Results of paired t-tests by domain indicated a positive change from baseline self-assessment of competencies compared to post-training (P < 0.001; Supplementary File S4).

Fig. 2.

Self-rated ADL retraining during PTA competencies pre- versus post-workshop. Competency ratings from 1 to 10 (1 = do not have, 5 = partly developed, 10 = advanced). All pre-post comparisons were statistically significant (P < 0.001).

Interviews

Interviewees (n = 11) discussed their experience or role in implementing PTA ADL retraining with interviews averaging 31.83 min (range: 8.37–62.00). Seven themes and 21 sub-themes were identified through thematic analysis. Fig. 3 depicts the relationship of themes. Concentric circles represent the influence on implementation of the organisation and the individuals within this (OTs and patients). Some themes overlap multiple domains, indicating dual influence. Themes were placed within the CFIR process of implementation (planning, engaging, executing, reflecting, and evaluating) to allow for evaluation of implementation efforts and to inform future endeavours (Damschroder et al. 2009).

Fig. 3.

Representation of interview themes in the context of the CFIR implementation process (Damschroder et al. 2009).

Facilitators of implementation

Key facilitators in implementing the PTA ADL retraining were reflected across the themes of OTs’ knowledge and skills, changing practice, and improving patient outcomes. Willingness to change, needing time and resources, working as a team, and tailoring it to their needs included some facilitatory elements. Reflecting the CFIR domain of characteristics of individuals, OTs’ knowledge, skills, and self-efficacy aided implementation and they indicated a change in their practice, using PTA ADL retraining more regularly and earlier with patients. Further, OTs advocated for use of the intervention and described being equipped to individualise the approach for the patient. Regarding the intervention characteristics, supportive evidence coupled with observed patient benefit were notable drivers. Patient benefits reflected the RCT findings including earlier independence in ADLs, briefer PTA duration and low levels of agitation (Trevena-Peters et al. 2018a, 2018b). Further advantages included fostering an earlier therapeutic alliance and enabling higher order ADLs to be addressed following PTA emergence, thereby maximising independence at discharge. At the inner setting, when time and resource needs were met, and there was managerial and team support, implementation was facilitated. However, when absent they served as barriers. Lastly, at the outer setting, whilst OTs prioritised ADL retraining for individuals in PTA, this behaviour change did not appear to have strong traction at an organisational level.

Barriers to implementation

Barriers to implementation of the intervention were captured within themes of changing practice, tailoring it to their needs, knowledge and skills (of the team), working as a team, willingness to change, and needing time and resources. In considering the inner setting, whilst OTs’ skill development facilitated implementation, relevant knowledge across the wider team was lacking, with a need for broader training identified. Clinicians’ knowledge and skills represent key features underpinning potential success of intervention implementation (Tavender et al. 2014; Pilli et al. 2023). Similarly, it is known that harmonising discipline-specific and team-wide approaches can be challenging, but facilitates implementation (Wright et al. 2016). OTs noted that working as a team could pose a challenge, for example, when required input from another staff member was not available or when seeking input for carry over of the intervention (i.e. skills retraining practice). Whilst OTs indicated willingness to implement the intervention, changing behaviour within their workplace was often challenging. There is a need for managerial support to underpin successful training, resourcing, and, ultimately, the implementation of PTA ADL retraining.

Another barrier to implementation was a mismatch between the needs of patients and the intervention and resources available in the inner setting, including environmental setup, equipment, and time. Such findings are not uncommon. The need for resources and time to spend with clients have previously been reported as implementation challenges in care and rehabilitation following TBI (Tavender et al. 2014; Wright et al. 2016). When considering the intervention characteristics, some OTs indicated the intervention did not suit all clients and they had to implement it in a less rigorous manner. This translation of knowledge-to-practice likely resulted in a ‘watered-down’ version of the intervention. It is also possible that the inclusion criteria of the original study (i.e. able to follow a single stage command, etc.) were not routinely applied.

Future implementation

To promote a holistic view of translation of evidence into practice, multiple domains such as that from the CFIR need to be considered (Grol and Grimshaw 2003; Damschroder et al. 2009; Damschroder and Hagedorn 2011; Damschroder 2020). To do so, this study employed a mixed-methods and longitudinal design to capture various outcomes and perspectives, as well as change over time regarding implementation of the intervention (Peters et al. 2013). To inform factors that influence implementation, data were gathered from the ‘real world’, specifically OTs and healthcare administrators (Peters et al. 2013).

Reflections on the further process stages of implementation are useful and can inform future endeavours. Within this, engagement warrants attention. Training efforts were focused at the level of upskilling OTs, with less consideration given to implementation leaders and champions who have influence, hold responsibility for overseeing implementation and thereby drive change (Scott et al. 2012; Damschroder 2020). In future, earlier liaison with management and the appointment of willing and engaged champions to train staff, facilitate, and monitor use of the intervention would be helpful (Bernhardsson et al. 2017).

Engagement also involves training in the intervention. Whilst workshop findings were suggestive of positive implementation momentum, the training lacked a longitudinal and holistic approach. Although important and commonly employed, education alone has a finite effect on implementation (Scott et al. 2012; Jones et al. 2015) and additional strategies such as reminders and feedback are recommended (Grol and Grimshaw 2003). Further allied health training opportunities and involvement in PTA ADL retraining (i.e. regular skills practice) warrant consideration to build a multi-faceted training package (Bernhardsson et al. 2017). For example, education regarding errorless and procedural learning would likely benefit other team members in delivering therapy during PTA, such as physiotherapy. There may also be potential for intervention co-delivery with nursing staff and/or allied health assistants to address staffing issues. Lastly, training in the intervention during university, clinical placements, and in work-commencement at rehabilitation centres is important to aid sustainability (Morris et al. 2020).

Beyond education, funding of OT time is required to enable the intervention. This is challenging in the context of hospital financial constraints. The study’s economic evaluation demonstrated that provision of PTA ADL retraining is cost saving through shorter length of stay (Mortimer et al. 2019). Therefore, the argument could be made that more patients could be treated in a given time-frame. Engagement of external change agents (e.g. funding bodies) to leverage the existing evidence and facilitate intervention decisions at rehabilitation centres may be helpful.

Execution of the implementation plan requires reflection. Changing existing and possibly, ingrained practices, is challenging. However, the fact that recent clinical practice guidelines have recommended the use of ADL retraining in PTA (Ponsford et al. 2023) will hopefully promote change and willingness of hospitals to invest resources in this practice. Inner and outer setting factors will need to be addressed along with provision of accessible training to promote behaviour change (Ritchie et al. 2017; Hyzak et al. 2023).

Limitations

The study had several limitations. Despite theoretical sufficiency, consenting interviewees may have been those with an interest in the intervention, conceivably impacting breadth of perspectives obtained. Further, only 46% of those interviewed worked in inpatient adult rehabilitation settings – the intended intervention setting. Capturing additional perspectives from those working in the primary inpatient TBI rehabilitation centres and further stakeholder involvement would have been valuable. Lastly, the study occurred prior to the update to the CFIR (Damschroder et al. 2022) and is, therefore, not reflective of these changes, which include benefits such as further delineation between implementation strategies and the ‘innovation’ being implemented. Despite this, given straightforward comparability of the original and updated CFIR domains, the study’s findings remain relevant (Damschroder et al. 2022).