Study Aim
This study delves into the developmental evaluation of the CHOICE Program, specifically designed for staff in Long-Term Care (LTC) facilities. Conducted at a single pilot site, the research aimed to ascertain two crucial aspects: firstly, whether the CHOICE Program could effectively modify the mealtime experience, encompassing physical, social, and Relationship-Centered Care (RCC) practices; and secondly, to identify necessary adaptations or additions to the program’s components for optimal implementation and impact.
Research Design
Employing a developmental evaluation approach, this study strategically assessed the CHOICE Program during its nascent stages. Developmental evaluation is particularly valuable for interventions in complex settings, allowing for real-time adjustments and learning. This method is crucial for understanding what aspects of a program are effective and which are not, especially in multifaceted programs within intricate environments [20]. The core principle of this evaluation is to foster program adaptation based on ongoing learning, adopting a ‘bottom-up’ strategy [20]. This approach is particularly relevant for complex interventions aimed at modifying social interactions and requiring coordination among diverse staff. Consequently, the CHOICE Program underwent revisions and expansions throughout this evaluation.
The study utilized an explanatory sequential mixed-methods design (QUANTITATIVE → qualitative) with a pre-test post-test time series. Initially, quantitative data was gathered using the Mealtime Scan for Long-Term Care (MTS) tool at four intervals over 24 weeks. These quantitative findings then guided subsequent qualitative data collection through semi-structured interviews [21]. The integration of both data sets was crucial for addressing the study’s objectives. Ethical approval was secured from the University of Waterloo Office of Research Ethics (#21413).
Intervention Participants and Setting
To enhance the breadth of learning, the study incorporated two distinct home areas—Parker and Wellesley (pseudonyms)—within a single LTC facility. This selection was strategic, anticipating that varying resident eating capacities would influence the team’s improvement strategies. The LTC home, part of a for-profit Ontario chain, was situated in Southern Ontario, Canada. The Schlegel-University of Waterloo Research Institute for Aging (RIA) played a key role in facilitating this innovation within a research-supportive environment. This collaboration aimed to translate the study’s insights into broader improvements across the LTC sector. The chosen facility, housing 192 beds across six home areas, was also selected due to its recognized need to enhance team member engagement during resident mealtimes and its limited prior exposure to mealtime-focused research initiatives.
Given that quantitative data collection was based on dining room observations, individual consent from team members and residents was deemed unnecessary by the University of Waterloo Office of Research Ethics, as the assessment focused on overall mealtime environment aspects. Prior to implementation, informational materials were distributed to residents, families, and staff to ensure awareness of the initiative and the observational data collection. Notices were also posted within the home areas to inform them of scheduled meal observations. Care team members and management pivotal to the CHOICE Program’s implementation were invited for qualitative interviews, for which informed written consent was obtained.
Implementation Overview
Prior to the pilot program launch, the research team, including the Principal Investigator (HK), Study Coordinator (HD), and doctoral student (SW), engaged with the LTC home’s administration. Discussions covered program components, the implementation process, and potential factors influencing care team behavior changes during mealtimes (Table 2). A consensus was reached regarding the study timeline (8 months of staff involvement), time commitments (staff huddles, coaching calls), resource allocation (informational materials, program prints), and achievable outcome goals for each home area.
The administrators appointed two Mealtime Champions (MT Champions) per home area. These champions were crucial in leading their teams through the implementation, acting as key contacts for feedback exchange with the research team. Selection criteria for MT Champions included: a) commitment to care enhancement, b) positive peer influence and respect, and c) full-time employment ensuring home area familiarity. Each home area focused on one CHOICE Principle at a time, allowing flexibility in the duration spent on each principle to accommodate specific needs. The Director of Food Services (DFS) served as an internal resource and CHOICE advocate, ensuring team accountability.
Early in the implementation, it became evident that MT Champions and the DFS required additional support to drive change effectively. The research team collectively addressed this need. The PI provided theoretical and technical guidance, while the study coordinator facilitated communication between home management, dining staff, and researchers. The doctoral student took on the role of ‘CHOICE Coach’, visiting the site bi-monthly for five months. During these visits, the Coach attended team huddles, supported huddle diary completion, observed mealtimes, gathered feedback, and collaborated with MT Champions on problem-solving. The Coach also met regularly with the DFS to review MTS-generated progress reports and devise strategies for improvement. In the final three months, the Coach reduced on-site presence to foster MT Champion and team independence, while bi-weekly coaching teleconferences with the research team continued.
Table 2 CHOICE Intervention Components and Functions
Table 2 detailing the components and functions of the CHOICE Intervention, outlining strategies for enhancing mealtime experiences in long-term care.
Recognizing the need for structured change management support, the Theoretical Domains Framework was adopted as a coaching foundation [22, 23]. This framework, simplifying fourteen domains into capability, opportunity, and motivation (COM-B model [24]), was chosen for its emphasis on context (opportunity), crucial for healthcare interventions. Coaching, without formal team member education on the framework, focused on supporting program component adoption and execution. Positive reinforcement, such as celebrating successes, was used to motivate team participation.
The intervention adapted throughout the pilot, aligning with developmental evaluation principles [20]. The final iteration included eight components (Table 2), tailored by team members and researchers to meet specific home area needs. For example, the in-person CHOICE Principles education session was converted into an online module for broader accessibility, particularly for new hires and those unable to attend the initial sessions.
Quantitative Phase
Data Collection
To assess the CHOICE Program’s impact on mealtime dynamics and interactions, standardized mealtime observations were conducted bi-monthly using the Mealtime Scan [MTS] [25]. The MTS is a validated tool for measuring the psychosocial and physical aspects of dining environments in LTC that influence mealtime experiences [25, 26]. It provides an overall dining room level assessment rather than individual resident assessments. A modified MTS version, incorporating recommendations from extensive prior use [11, 25], was tested for its responsiveness to the intervention. Modifications included enhanced tracking of social interactions and refined item scaling (dichotomous to 0–4 scale) to better capture changes over time. The physical environment meter was removed due to potential mealtime disruption [25]. A new summative scale assessing overall dining environment quality was added, showing good inter-rater reliability (ICC = 0.76; unpublished). The MTS includes four summative scales, each with a maximum score of 8. Detailed MTS information is available in Keller et al., 2018 [25].
Observations were conducted by two trained assessors in their respective dining rooms, starting before resident entry and continuing until meal completion. Originally planned for five time points (baseline, 8, 16, 24, and 32 weeks), data collection was limited to four due to external factors. At each time point, MTS was completed across five meals (1 breakfast, 2 lunches, 2 dinners) over two days, consistent with prior mealtime environment assessments [25]. The same assessor conducted all 20 assessments within each home area to ensure consistency.
Resident-level data, including age, sex, Cognitive Performance Score (CPS), and Activities of Daily Life – Long Form (ADL-LF), were collected from inter-RAI Minimum Data Set (MDS) to characterize residents. CPS, ranging from 0 (intact cognition) to 6 (severe impairment) [27], was dichotomized into CPS ≤2 (mild impairment) and CPS >2 (moderate to severe impairment) for analysis [28].
Data Analysis
Descriptive comparisons using chi-square or t-tests were performed to contextualize home area differences in resident data and stable mealtime characteristics. Normality tests were conducted on MTS summative scales and subscales, with descriptive statistics (mean, SD) presented by time point and dining room. Linear mixed models (Proc MIXED) with repeated measures analyzed MTS summative scales as outcomes, with dining room, time point, and their interaction as fixed effects. As both areas received the CHOICE program, no control group was present. Time effects indicated MTS scale changes over time, while time-home area interactions showed area-specific differences. Statistical significance was set at p < 0.05, and SAS® Studio Statistical Software was used for analysis.
Qualitative Phase
Data Collection
Qualitative semi-structured interviews were conducted to provide explanatory insights into the quantitative findings, particularly focusing on effective program components and implementation processes. Key informants, including MT Champions, management, and actively involved team members, were recruited post-study. Interview guides were informed by progress reports based on MTS data trends for each home area, focusing on:
- Team members: Perceptions of intervention impact on resident mealtime experiences, experiences with program components, suggestions for improvement and sustainability, and observed mealtime changes.
- Home Management: Perceptions of program components’ reception, implementation experiences, strategies for team support and program sustainability, and observed mealtime changes.
Additionally, brief qualitative comments from MTS audit forms, contextualizing meal-specific scores, were reviewed to support quantitative-qualitative data linkage. The CHOICE Coach (SW) conducted interviews over one month following the intervention’s conclusion in November 2016.
Data Analysis
Data analysis, conducted concurrently with data collection by SW, continued until informational redundancy was achieved [29]. Verbatim interview transcripts were analyzed using a generalized inductive approach [30], involving coding data segments, collapsing codes into categories, and interpreting themes. Emergent codes, categories, and themes were reviewed with co-authors (HK, VV, SI). The analysis focused on intervention implementation, components, and mealtime changes in each home area. MTS text data was deductively coded and integrated into interview themes to contextualize dining area findings [31].
