Influence of Psychosocial Factors on the Energy and Protein Intake of Older People on Dialysis

Objective To explore the relationship between nutritional parameters and psychosocial factors in older people on dialysis. Design A cross-sectional observational study in prevalent older people on hemodialysis (HD) and peritoneal dialysis (PD). Setting A secondary analysis from a quality of life study in older people (Broadening Options for Long-term Dialysis in the Elderly). Subjects One-hundred and six patients 65 years of age or older and on dialysis for at least 90 days were purposively recruited (HD patients matched to PD patients by age, sex, dialysis vintage, ethnicity and Index of Deprivation). Half were on HD, the mean age was 72.7 years, 72% were male, 92% were from a White ethnic background, and 26% had diabetes. Intervention The patients attended one visit at which they completed nutritional assessments (3-day food diary, subjective global assessment, handgrip strength, and body mass index) and questionnaires: Short Form-12 (SF-12), the Hospital Anxiety and Depression Scale (HADS), the Mini Mental State Exam, and social networks. Main outcome measures The differences in nutritional parameters between patients on PD and HD were determined by univariate analyses, and the relationships between nutritional intake and demographic, clinical, and psychosocial variables were determined by multivariate analyses. Results There was no difference in the energy and protein intake and nutritional status between older people on HD and PD. For the whole sample, multivariate analyses found that lower energy intake was related to fewer social networks ( P = .002) and lower SF-12 Physical Component Scale (PCS) scores ( P = .021). A lower protein intake was related to worsening Index of Deprivation scores ( P = .028) and an interaction between SF-12 PCS and presence of possible depression ( P = .015). Conclusion Energy and protein intake in older people (regardless of modality) appears to be independently associated with psychosocial variables.