Prevalence and Determinants of Poor Food Intake of Residents Living in Long-Term Care

Abstract

Poor food intake is known to lead to malnutrition in long-term care homes (LTCH), yet multilevel determinants of food intake are not fully understood, hampering development of interventions that can maintain the nutritional status of residents. This study measures energy and protein intake of LTCH residents, describes prevalence of diverse covariates, and the association of covariates with food intake. Multisite cross-sectional study. Thirty-two nursing homes from 4 provinces in Canada. From a sample of 639 residents (20 randomly selected per home), 628 with complete data were included in analyses. Three days of weighed food intake (main plate, estimated beverages and side dishes, snacks) were completed to measure energy and protein intake. Health records were reviewed for diagnoses, medications, and diet prescription. Mini-Nutritional Assessment-SF was used to determine nutritional risk. Oral health and dysphagia risk were assessed with standardized protocols. The Edinburgh-Feeding Questionnaire (Ed-FED) was used to identify eating challenges; mealtime interactions with staff were assessed with the Mealtime Relational Care Checklist. Mealtime observations recorded duration of meals and assistance received. Dining environments were assessed for physical features using the Dining Environment Audit Protocol, and the Mealtime Scan was used to record mealtime experience and ambiance. Staff completed the Person Directed Care questionnaire, and managers completed a survey describing features of the home and food services. Hierarchical multivariate regression determined predictors of energy and protein intake adjusted for other covariates. Average age of participants was 86.3 ± 7.8 years and 69% were female. Median energy intake was 1571.9 ± 411.93 kcal and protein 58.4 ± 18.02 g/d. There was a significant interaction between being prescribed a pureed/liquidized diet and eating challenges for energy intake. Age, number of eating challenges, pureed/liquidized diet, and sometimes requiring eating assistance were negatively associated with energy and protein intake. Being male, a higher Mini-Nutritional Assessment–Short Form score, often requiring eating assistance, and being on a dementia care unit were positively associated with energy and protein intake. Energy intake alone was negatively associated with homelikeness scores but positively associated with person-centered care practices, whereas protein intake was positively associated with more dietitian time. This is the first study to consider resident, unit, staff, and home variables that are associated with food intake. Findings indicate that interventions focused on pureed food, restorative dining, eating assistance, and person-centered care practices may support improved food intake and should be the target for further research.