Construct validity of the Dining Environment Audit Protocol: a secondary data analysis of the Making Most of Mealtimes (M3) study

BACKGROUND Research has demonstrated the importance of physical environments at mealtimes for residents in long term care (LTC). However, a lack of a standardized measurement to assess physical dining environments has resulted in inconsistent research with potentially invalid and unreliable conclusions. The development of a standardized, construct valid instrument that assesses dining rooms is imperative to systematically examine physical environments in LTC. The purpose of this study was to determine the construct validity of the new Dining Environment Audit Protocol (DEAP) tool. METHODS Secondary data collected from the Making Most of Mealtimes (M3) study was used for this analysis. Data were collected in 32 long term care homes, which included 82 dining rooms and 639 residents. A variety of resident and dining room level constructs were compared to the summative scales found on the DEAP using Spearman correlations and Student t-tests. A regression analysis identified individual characteristics assessed with DEAP that were associated with the summative scales of homelikeness and functionality. RESULTS Regression analysis (p < 0.05) identified that the DEAP homelikeness scale was positively associated with a view of the garden/green space, presence of a clock and a posted menu. The functionality scale was positively associated with number of chairs and lighting, while negatively associated with furniture with rounded edges and clutter. Additionally, the functionality scale was positively associated (p < 0.05) with the Mealtime Scan physical scale (ρ = 0.52), the dining room Mealtime-Relational Care Checklist (M-RCC) (ρ = 0.25), the DEAP total score (ρ = 0.56), and the Mini Nutritional Assessment- Short Form (ρ = 0.26). Homelikeness was positively associated (p < 0.05) with the DEAP total score (ρ = 0.53), staff Person Directed Care score (ρ = 0.49) and the resident Cognitive Performance Scale (t = 2.56), while negatively associated with energy (ρ = -0.26) and protein intake (ρ = -0.24). The homelikeness and functionality scales were also associated with one another (ρ = 0.26). CONCLUSION The construct validity of the DEAP was supported through significant correlations with a variety of measures that are theoretically related to the homelikeness and functionality of LTC dining rooms. This secondary analysis supports the use of the DEAP in future research to quantify the physical environment of LTC dining rooms. Protocol registered with ClinicalTrials.gov ID: NCT02800291; Registered retrospectively June 7, 2016.