WEIGHT-GAIN IMPACTS MORBIDITY AND MORTALITY IN INSTITUTIONALIZED OLDER PERSONS

OBJECTIVE: To determine if a significant weight change in previously undernourished older persons is associated with mortality and morbidity. D DESIGN: Retrospective survey. SETTING: A chronic care hospital in Canada. SUBJECTS: Follow-up survey from an original data set (Time 1) on 88 previously diagnosed undernourished and 60 adequately nourished. Average age was 78 years; average length of residence was 40 months. Of the 148 subjects, 120 were male. MEASUREMENTS: Average time to follow-up (Time 2) was 319 days from Time 1. Data was collected on weight change, reported incidence of infection, use of antibiotics, number of falls, and admissions to acute centers, as well as incidence and cause of death. Three-day diet intake records were available on 15% of subjects within 1 month of Time 2. Five nutritional groups existed within this follow-up population: those previously undernourished who gained at least 5% of Time 1 weight (Group 1), those undernourished who lost at least 5% of Time 1 weight (Group 2), those undernourished who maintained their weight (Group 3), those adequately nourished who lost at least 5% of original weight (Group 4), and those who remained adequately nourished (Group 5). RESULTS: Forty-one of the 148 patients died; this was significantly associated with nutrition group (X(2)=11.23, P = 0.02), with fewer deaths in Groups 1, 3, and 5. As well, primary diagnosis (x(2)=32.6, P = 0.001), length of residence (X(2)=25.6, P = 0.01), and time to death (X(2)=26.4, P = 0.009) were significantly associated with nutrition group. Diet intake records indicated an improvement in energy and protein intake in Group 1 during the follow-up. Energy and protein change, whether increased or decreased from the original study, was significantly associated with nutrition group. Total number of feeding impairments and presence of dysphagia were significantly associated with nutrition group. Other morbidity events were not significantly associated with nutrition group, but trends suggest a positive effect of weight gain on incidence of falls, incidence of more than two infections, and incidence of recurring infections. CONCLUSIONS: A weight increase of at least 5% of body weight in previously undernourished patients is associated with a decreased incidence of death and may reduce morbidity events. Feeding impairments may affect ability of patients to gain weight, but dysphagia is associated with weight increase. Other nonnutritional variables of diagnosis, number of medications, and declining medical status also affect mortality and morbidity outcomes.