SCOPE: Safer Care for Older Persons (in Residential) Environments: A Single Arm Pilot Study

Background: Nursing home residents require daily support. While care aides provide most of this support they are rarely empowered to lead quality improvement (QI) initiatives. A previous proof of principle study, called Safer Care for Older Persons in Residential Care Environments (SCOPE), demonstrated that care aide-led teams can successfully participate in QI interventions. In preparation for a large-scale study, this one-year pilot evaluated how well the bundle of SCOPE coaching strategies helped care-aide led teams to enact these interventions. A secondary aim was to determine if improvements in resident quality of care occurred. Methods: Using a modied IHI Breakthrough Collaborative Series model in a prospective single-arm study design, we randomly sampled 7 nursing homes in Winnipeg, Manitoba from the longitudinal Translating Research in Elder Care (TREC) cohort. Each SCOPE team had 5-7 front-line staff led by care aides. Teams received coaching to enact the intervention (i.e., to create actionable aim statements, implement QI interventions using plan-do-study-act [PDSA] cycles, use measurement to guide decision making) during three learning congresses, networked and shared learning experiences during these sessions, and received additional support from quality advisors between congresses. We used self-report data to code intervention enactment (‘poor’, ‘adequate’, ‘excellent’), and also measured improvement in team cohesion and communication. Secondarily, we observed changes in unit-level quality indicators using RAI-MDS 2.0 data. Results: Most teams successfully enacted SCOPE. Five of 7 teams created adequate-to-excellent aim statements throughout the pilot (e.g., statements were specic, measurable, time-bound). While 6 of 7 teams successfully implemented PDSAs, only 2 reported spreading their idea to involve more than a few residents and/or staff on their unit. Three of 7 teams explicitly stated how measurement was used to guide decisions. Team cohesion and communication scored high at baseline, and hence improved minimally. Resident quality indicators improved in 4 of the 7 nursing home units. Conclusions: Our bundled coaching strategies helped most care aide-led teams to enact SCOPE. Coaching modications are needed in follow-up studies to help teams more effectively use measurement, and to spread successful interventions within the unit. More detailed and robust approaches are also needed

Our coaching strategies successfully helped teams to enact SCOPE, and hence these strategies can be applied to larger follow-up studies. Coaching modi cations are needed to help teams more effectively use measurement, and to spread successful interventions within the unit. More detailed and robust approaches are also needed to monitor treatment enactment.

Background
Older adults are the fastest growing segment of the worldwide population [1]. As life expectancy increases so does the number of people with dementia and other co-morbid medical conditions [2][3][4][5][6]. Annually, 1.7 million North Americans reside in nursing homes [7] and at least half of these residents have some form of age-related dementia, often combined with additional impairments (e.g., di culties completing daily tasks, responsive behaviours, frequent incontinence) [8][9][10]. This vulnerable group requires complex health, personal, and social care, provided in ways that has meaning for residents [11] and that emphasize the importance of relational care and quality of life [12]. While media have highlighted the signi cant challenges with nursing home care during pandemic times [13][14][15], the quality of care provided in this sector has been recognized as suboptimal for decades, and many groups have called to improve nursing home structures and care processes [16][17][18][19][20].
Care aides (unregulated workers, also called personal support workers, orderlies or nursing assistants) provide 80% -90% of direct care to nursing home residents [21]. These staff are best situated to observe, interpret and respond to residents' daily needs [22,23], making them uniquely positioned to meaningfully participate in and, we contend, to lead quality improvement (QI) initiatives. However, research shows that care aides are routinely excluded from care planning processes, making them feel under-valued by other care staff and emphasizing the need to create more constructive collaborative care approaches [24]. Evidence shows that empowering care aides enhances their work performance and quality of work life [25][26][27], and that improving inter-professional collaboration can enhance nursing home care quality [28][29][30].
Given this knowledge, we had previously developed an intervention called Safer Care for Older Persons in Residential Care Environments (SCOPE) [31]. SCOPE is a multi-component intervention designed to empower care aides to lead, with coaching support, QI activities at the resident care unit level. SCOPE was designed to improve care aides' abilities to engage in QI initiatives, to facilitate the use of best evidence in their practice, and secondarily to improve their quality of work life and engagement. Enhancements in these areas should ultimately lead to improved quality of resident care and their associated health-related outcomes. In a proof-of-principle study, TREC researchers have shown that care aide-led teams can feasibly engage in SCOPE and learn about/apply QI interventions at the resident bedside, which in some instances resulted in unit-wide improvements in select clinical areas [32].
We undertook this pilot study in preparation for a larger trial. The aim of this pilot was to evaluate how well SCOPE coaching strategies helped care-aide led teams to enact their QI interventions, by speci cally (1) creating actionable QI aim statements, (2) implementing their QI plans using plan-do-study-act (PDSA) cycles, and (3) using measurement to guide decisions about the need to modify their intervention approaches. We also measured improvement in team cohesion and communication during SCOPE, and secondarily described changes in quality indicators (i.e., pain, mobility and responsive behaviours) at the resident care unit-level.

Study design
This was a single arm prospective pilot study, lasting one year.

Setting & Sample
The SCOPE pilot was conducted with a random sample of 7 of 16 nursing homes located in Winnipeg, Manitoba that are enrolled in the Translating Research in Elder Care (TREC) program. TREC is a multilevel, longitudinal program of applied health services research designed to improve the quality of care and quality of life for nursing home residents, and also the quality of work life for their care staff [33]. TREC applies these constructs at the clinical microsystem (care units) where quality is created [34,35]. The overall TREC cohort was created using a strati ed (owner model, size, region) random sample [33].

The SCOPE Teaching and Coaching Strategies
The SCOPE intervention is based on a modi ed Institute for Healthcare Improvement (IHI) Breakthrough Collaborative Series model [36]. It was also informed by knowledge translation theory, speci cally focusing on the role that facilitation plays in implementation success [37,38]. Details of these coaching strategies are provided elsewhere [31,39] with each component shown in Figure 1. Components include (1) a 'Getting Started' evidence kit with clinical information and improvement strategies, speci c to one of three clinical areas (reducing pain, improving mobility, reducing dementia-related responsive behaviours) selected by teams; (2) three 2-day learning congresses designed to train SCOPE teams in basic QI approaches, and importantly, to provide them with peer to peer (from other units and sites) networking and learning opportunities; (3) a quality advisor who helped to design and implement the learning congresses, and who supported teams (in-person visits, telephone calls) regularly between these sessions; (4) a quality coordinator who provided oversight to the quality advisor, and who led virtual and in-person discussions to help unit and facility managers support front-line QI teams; and, (5) a celebratory congress held at the end of the pilot.
The quality advisor was the main liaison with each team. Duties included: 1) meeting with each team at the beginning of SCOPE to review the 'Getting Started' information kit; 2) working with the quality coordinator and research team to prepare and facilitate learning congresses; 3) conducting face-to-face meetings with each team at least monthly, to help them enact their PDSA plans and brainstorm solutions to challenges encountered; 4) being available for additional team consultation (phone, email) as needed; and, 5) keeping a diary of team interactions and progress.

Participants and Study Procedures
Executive Directors from each facility received a written invitation to participate in the pilot followed by an in-person meeting to answer questions, to explain nursing home responsibilities, and to discuss available support. Sites were selected randomly with replacement; 1 site declined to participate, stating insu cient staff levels to engage in research. No sites were lost to follow-up during the research.
Following written consent to participate in the pilot, the Executive Director identi ed a senior sponsor (usually the Director of Care) to help promote SCOPE to other management staff, and to remove implementation barriers throughout the pilot as needed. This individual identi ed, at their discretion, one unit from their facility to participate in the pilot, and selected a unit-level team sponsor (usually a unitlevel clinical nurse manager) who was responsible to support day-to-day project activities. Senior and Team Sponsors collaborated to select a front-line team consisting of 5-7 members. At least 2 team members were care aides with one as team lead; other care staff (e.g., social workers) were selected as needed. Each team chose their intervention to focus on either reducing pain, improving mobility, or reducing dementia-related responsive behaviours. These three areas were selected based on a ranking exercise, completed by care aides before the pilot, using 4 criteria: (1) their perceived importance to care aides, (2) the ability to measure outcomes in these areas using the Resident Assessment Instrument-Minimum Data Set (RAI-MDS 2.0), (3) a distribution in the measures that demonstrated there was room for improvement, and (4) their modi ability [40,41].
Each congress occurred three months apart ( Figure 1); the agenda for each learning congress is provided in Appendix 1. In Learning congress 1, teams were coached to develop effective QI aim statements, while learning congresses 2 and 3 focused on measurement and strategies to spread effective QI strategies within each team's unit, respectively. Congresses also helped teams to problem solve and share solutions to challenges that they encountered (e.g., getting buy-in from peers), provided teams with knowledge sharing and socialization opportunities (e.g., through impromptu networking sessions and team presentations sharing their PDSA experiences), and provided dedicated planning time to integrate lessons learned into teams' upcoming daily care routines (action periods). During the celebratory conference teams celebrated their achievements, discussed lessons learned, and considered next steps. Examples of activities conducted during the congresses included storyboard sessions and team presentations (designed to help teams share their successes and opportunities to improve); technical training (creating aim statements, conducting PDSA cycles) using improv and simulation techniques, and interactive "games" designed to promote learning in speci c quality improvement areas; and time dedicated for team re ection and planning.

Ethics
Approval to conduct the research was provided by the University of Manitoba Health Research Ethics Committee (reference number H2015:045). Each nursing home received $3000 to help back ll participating team members who attended learning congresses. This study was funded by the TREC program (grant number PS 148582).

Measures Treatment Enactment
Enactment is an element of treatment delity that measures the extent to which people actually implement a speci c intervention skill, and differs from what is taught (treatment delivery), what is learned (treatment receipt), and the extent of its effect (treatment e cacy) [42]. Enactment is one of the most challenging aspects of treatment delity to measure [42]. Traditional approaches to measuring it include the use of questionnaires and self-reports, structured interviews, and activity logs [42].
Each team was asked to complete a self-assessment form every two months during the pilot (Appendix 2). Teams were asked to use this form to (1) create and re ne their QI aim statement; (2) report how well they were able to implement QI interventions using PDSA methods (e.g., starting with one or two residents, and involving other residents and/or staff depending on their success); and (3) document how they used measurement strategies and data to guide team decision making (e.g., to assess whether they were making progress towards their aims).

Measures -care aides
Workgroup cohesion is the "degree to which an individual believes that the members of his or her work group are attracted to each other, are willing to work together, and are committed to the completion of the tasks and goals of the work group" [44]. We measured work cohesion using 8 items adapted to align with the pilot (e.g., We have a lot of team spirit among team members; We know that we can depend on each other; We stand up for each other).
Workgroup communication is the "degree to which information is transmitted among the members of the work group" [44]. It was measured using 4 items adapted to align with the pilot (e.g., We frequently discuss resident care assignments with each other; We care share ideas and information).
Each of these measures was scored on a seven-point Likert scale ranging from 'strongly disagree' to 'strongly agree'; item scores were averaged to provide an overall score ranging from 1 to 7, with the latter representing strong agreement team cohesion/communication). Scores of '4' on these scales de ne groups with neutral agreement about group cohesion and communication.

Measures -residents
Quality indicators were assessed using RAI-MDS 2.0 [45], focusing on the percentage of people who showed improvements in mobility, the percentage of people whose responsive behavioral symptoms improved, and the percentage of people with pain. Resident mobility was assessed using the third generation [46] RAI-MDS 2.0 quality indicator "MOB1a" (the percentage of residents whose ability to locomote on the unit improved). This indicator excludes residents who are comatose, have six or fewer months to live, and/or who were independently mobile during their previous RAI-MDS 2.0 assessment [45]. The quality indicator entitled "BEHI4" was used to identify the percentage of residents on each unit whose behavioral symptoms (i.e., wandering, verbally abusive, physically abusive, socially inappropriate or disruptive behavior) improved from the previous RAI-MDS 2.0 assessment [45]. This indicator excludes residents who are comatose or who had missing behavioral scores in their previous assessment. Resident pain was measured using the RAI-MDS 2.0 pain scale [45]. This quality indicator assesses the percentage of residents with any amount of pain in the last seven days, excluding those with missing or con icting (no pain frequency but with some degree of intensity) item responses.

Treatment enactment
Each component of treatment enactment was scored using a 5-point scale ranging from poor (1) to excellent (5) ( Table 1). Aim statements were scored by the extent they met the SMART criteria of being Speci c, Measurable, Achievable, Relevant, and Timely [47]. Teams' ability to plan and implement their intervention using PDSA cycles were scored based on the degree to which their reported plans aligned with aim statements, and by the extent to which they reported spreading their improvement strategies to involve other residents and/or staff within their unit. Teams were also scored by the extent that they documented using measurement strategies and data to guide intervention revisions and related decisions. Two authors (MD, LG) independently rated teams' scores of treatment enactment, using bimonthly data reported during the pilot. Scoring discrepancies were resolved through iterative discussions.

Team cohesion and communication
Descriptive measures of workgroup cohesion and communication are shown for months 1, 7 and 12 of the pilot, after verifying that results are equivalent to bi-monthly scores. The team included speci c text documenting how measurement and data were used to guide improvement decisions in successive PDSA cycles.
The team made vague reference to measurement tools and/or strategies used to guide decision making in successive PDSA cycles.
The team did not report how measurement and data were used to guide decision making.
a Team aim statements had to include operational terms (e.g., de ne responsive behavior) (Speci c); contain a target goal (e.g., identify the degree of improvement sought) (Measurable); be realistic (e.g., initially focus on a smaller number of residents) and/or show progression throughout the pilot (Achievable); include information about how (e.g., by creating toolkits to support implementation) or when (e.g., during mealtime) the intervention would happen (Relevant), and; include a reference point/date by which intervention success would be measured (Timely).

Quality indicators
RAI-MDS 2.0 quality indicators were calculated at the unit-level using quarterly data collected during the pilot, using statistical process control (SPC) methods [48]. Data were not distributed normally and thus the following SPC zones were created using pre-SCOPE (January, 2013 to December 2016) data: a) zone -3=1 st -5 th percentile; b) zone -2=5 th -34 th percentile; c) zone -1=34 th -50 th percentile; d) zone + 1=50 th -66 th percentile; e) zone + 2=66 th -95 th percentile; f) zone + 3=95 th -99 th percentile. Following the SPC Western Electric rules [49], non-random variation was detected if (a) one or more data points during the SCOPE pilot were beyond zone 3 of pre-SCOPE results, (b) two of three successive data points were beyond zone 2, or (c) four of ve successive data points were beyond zone 1.

Results
Nursing home characteristics, team composition and QI focus.
The characteristics of SCOPE nursing homes, units and team composition are found in Table 2. Five of the 7 nursing homes in the pilot were (public or voluntary) non-pro t, while 2 and 4 homes were medium (80-120 beds) and large (> 120 beds), respectively. Homes had between 1 and 6 units that ranged in size between 27 and 100 beds. A detailed analysis of the nursing home staff characteristics in TREC homes is available elsewhere [50][51][52].
Five of the seven SCOPE teams focused on reducing dementia-related responsive behaviors, 1 team focused on reducing pain, and 1 focused on improving resident mobility (Table 3). Team and senior sponsors were most often clinical nurse (unit) managers and Directors of Care, respectively. Team size (including the team and senior sponsor) ranged from 5 (n = 4 SCOPE sites) to 7 (n = 1 SCOPE site) individuals. With two exceptions (sites C and F), front-line SCOPE teams were comprised entirely of care aides.  We graded 3 of the 7 teams as creating excellent aim statements (rating of 5/5) during the course of the pilot (Table 4), 2 teams as creating adequate aim statements (rating = 3/5), and 2 teams as creating 'poor' (rating = 1/5) or 'poor-to adequate' (rating = 2/5) aim statements. To illustrate, Team D de ned responsive behavior in their aim statement ('hitting, screaming, arguing'; speci c), quanti ed their goals (reducing events by 60%; measurable), showed progression during the course of the pilot (reducing events by 60% at month 7, and 90% by Month 12; achievable and timely), and de ned when the intervention would occur (during activities of daily living; relevant) (data not shown). While Team B (graded as adequate) satis ed the 'speci c' (de ned responsive behavior), 'measurable' (included a target goal) and 'relevant' (reported when the intervention would occur) SMART criteria, this team did not show progression in its aim statement, and nor did it identify a timeline for achieving intervention success. We graded Team E as having a poor aim statement, as it met the 'relevant' SMART criteria only (de ned when the intervention would occur). We graded 5 teams as achieving adequate to excellent intervention progression (Table 4), however only Teams F and G reported scaling their intervention to involve other residents and/or staff on their unit (these teams received a score of 'excellent'). Team F reported using 'pain pocket card survival kits' to remind and help staff to implement the intervention, and re ected on how they engaged with non-SCOPE providers on their unit to enhance their care processes. We graded Teams B and E as achieving poor intervention progression; both teams reported a 'success story' for only one resident at the end of the pilot (data not shown).
Teams D, F and G speci cally reported how they used measurement tools (e.g., mobility tracking tools, use of RAI-MDS 2.0 data) to help make decisions throughout the pilot, and hence we graded these teams as excellent in this category (Table 4). Teams A, B and E vaguely referred to measurement (e.g., conducting baseline assessments) without providing details, and were graded as 'adequate'. Team C did not make any reference to using measurement to guide decisions.
Care aide outcomes.
Team cohesion and communication results are shown in Table 5. Most teams moderately (an average score of '6' across all scale questions) or strongly (an average score of '7' across all questions) agreed with statements about their cohesion and communication throughout the pilot. As the only exception, Team C provided a score of 3.8 (a neutral opinion) for team cohesion at month 12 of the pilot.  SPC charts for quality indicators are shown in Fig. 2. Patterns of quality care indicator data were nonrandom during the pilot for Sites D and E (responsive behaviors; one data point beyond zone + 3), Site G (mobility; one data point beyond zone + 3), and Site F (pain; 4 consecutive data points beyond zone − 1).
This non-random pattern for Site F commenced pre-SCOPE. During follow-up discussions Site F leaders disclosed that this was at least partly due to changes in their pain assessment approach (i.e., all residents receiving an analgesic were originally deemed as having pain).

Discussion
This pilot demonstrated that the bundle of SCOPE teaching and coaching strategies − a "getting started" information kit, learning congresses, quality advisor guidance, and discussions to help senior and team sponsors support front-line teams − effectively supported most care aide-led teams' enactment of their QI strategies. Five of the seven teams provided adequate or excellent aim statements during the course of the pilot, 5 self-reported achieving at least adequate intervention progression (i.e., showed learning and re nement through PDSA cycles), and 3 teams speci cally discussed how they used measurement to guide intervention decision making processes during the course of the pilot. While measures of team cohesion and communication did not change appreciably during the pilot, opportunities for positive change were limited by high baseline scores. While this pilot was not powered to detect statistically signi cant differences in measures of treatment e cacy, some trends for improvement in quality indicators were noted at the resident care-unit level.
These ndings align with and are complemented by an earlier qualitative study conducted by Ginsburg et al. (2018) who analyzed data from 6 focus groups conducted during our nal SCOPE celebratory congress [39]. The previous ndings suggested that front-line providers saw great value in the bundle of SCOPE facilitated coaching strategies − and in particular the learning congresses, senior and team sponsor support, and interactions with quality advisors − for helping them to enact their interventions.
While care aides in that qualitative study emphasized the many advantages of SCOPE, they also acknowledged the challenges with implementing QI interventions and offered solutions to them [39].
These include having (1) regular time dedicated for teams to plan and enact their PDSA cycles, (2) clearer team member and leadership roles, and (3) stronger support (e.g., from leaders) to help spread successful intervention strategies to other residents and/or staff on their units. Also complementing the ndings reported here, Ginsburg et al. (2018) reported that front-line team members experienced considerable challenges with measurement, and recommended less didactic teaching approaches coupled with more measurement tools included in the Getting Started information kit [39]. Together, these two studies provide insights for adapting the SCOPE intervention in future trials.
Our pilot study ndings contribute to the existing nursing home improvement and implementation research [53,54] in three ways. These contributions apply to future SCOPE interventions, and can also help to transform QI interventions and care aide engagement in NH improvement work more broadly.
First, our results contribute to the growing body of literature showing that care aides can successfully lead QI initiatives, with the proper support. This is important, given care aides' essential role in providing day-to-day nursing home support coupled with their high degree of knowledge about the wants and needs of residents [21][22][23]. Actively engaging with care aides is important to enhance nursing home quality of care, particularly given the need to balance effective medical care with relational and social approaches [11,12]. Several researchers have demonstrated the bene ts of meaningfully engaging both care staff [25][26][27] and residents [55,56] during care processes.
Second, these ndings contribute to our understanding of the role of facilitated coaching to support quality improvement. Rycroft-Malone and colleagues propose that both task-oriented and enabling facilitation approaches are needed to support quality improvement, pending the improvement target and learning environment [38,57]. Our SCOPE pilot results, combined with the ndings of Ginsburg et al [39], suggest that a combination of technical (e.g., teaching teams to create and use measurement tools), education (e.g., enabling teams to apply these tools within the context of their care environment) and supportive facilitation (e.g., enabling teams to problem solve using simulation and allowing informal network time) approaches are likely needed. Additional research is required to understand how various combinations of facilitation techniques can best support improvement efforts both within the SCOPE intervention and more broadly.
Third, these study ndings highlight the value of and need for detailed process evaluation work, including approaches to measure intervention delity that allows us to better understand both how and why interventions succeed or fail [58]. While intervention delity is traditionally measured using self-report strategies [42], these type of data are prone to information bias [59], and techniques are required to differentiate between what an intervention has taught ( delity delivery), what is learned ( delity receipt), and what is enacted ( delity receipt). A compendium of data collection methods (e.g., self-report and researcher observations) are required to more effectively differentiate between these delity subcomponents, and to determine their importance [58].
Limitations SCOPE teams were recruited from a single Canadian health region, and hence lessons learned should be applied cautiously to other jurisdictions and countries. Teams also provided an overall description of treatment enactment without explaining how these scores were decided (e.g., by team consensus, by one person on behalf of the team). As suggested above, more detailed approaches to assessing delity enactment will help to provide more robust data on this important construct. Similarly, data on team cohesion and communication were captured at the team level and hence social desirability bias may explain the high scores on these measures. In future research, individual team-member responses may provide a more accurate data on measures of enactment, team cohesion, and communication.

Conclusion
This pilot provided guidance for a larger intervention trial. Results demonstrate that our facilitated coaching strategies supported most care aide-led teams to enact their QI interventions. To further enhance SCOPE, we recommend modifying certain coaching strategies and improving select research/intervention evaluation tools.

Consent for publication
Not applicable.

Availability of data and materials
The data that support the ndings of this study are available from Translating Research in Elder Care, c/o Dr. Carole Estabrooks, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of Dr. Carole Estabrooks.