Nutritional status and cognitive function in frail elderly subjects

Abstract

Longevity, frailty and chronic disease are often associated with protein energy-malnutrition (PEM). Subjects with dementia are at particular risk for PEM. The overall aims of this thesis were to evaluate relationships between nutrition, cognition and functional capacity and to assess interventional means of developing the nutritional care of elderly.

In study I and II, the nutritional status and the cognitive function were examined in 28 and 80 service flat (SF) residents. Furthermore, the effect of a nutrition education programme to SF care staff was evaluated. In study III, 22 and 14 subjects living in two group living units for demented (GLD) participated and the effects of nutritional supplementation and education to care personnel were evaluated. In study IV, possible relationships between weight, weight change and cognitive function were studied in 231 patients, admitted to a geriatric ward for examination of cognitive dysfunction. Seven-year mortality in relation to dementia diagnosis and nutritional status was also studied (IV).

The nutritional status was assessed by Body mass index (BMI=kg/M2), triceps skin fold (TSF), and arm muscle circumference (AMC). Study I and II included a modified Subjective Global Assessment (SGA). Muscle strength was measured with a Harpenden Grip Strength Dynamo-Meter@ (I, II). As biochemical nutritional markers in serum, albumin, transferrin, haemoglobin, C-reactive protein (CRP), vitamin B 12, folic acid and insulin-like growth factor-1 (Igf-1) (I-III) were analysed. Cognitive and functional status was assessed by the Mini Mental State Examination (MMSE) (0-30p, I-IV), Clinical Dementia Rating Scale (CDR) (0-3, 111) and Katz' ADL-index (I-III).

Care personnel at the SF (I-III) complexes attended 12-hour nutritional education program. The care personnel answered a questionnaire before and after the education that reflected their knowledge and attitudes about nutrition in elderly (I, II). Oral supplements (410 kcal/1720 kJ) were given daily during 5 months to residents in one of the two GLD units and the personnel attended a nutritional education programme (III). Mortality data (IV) were obtained from Swedish population records. Up to one third of the SF residents were assessed as malnourished or at risk for PEM according to SGA (I-II).

The education program tended to improve the knowledge of the care personnel in study I. In study II there was no significant difference in improvement between the staff at the two SF complexes. Objective measures of nutritional status and cognitive function in the residents was not changed at 5- and 6-months follow-ups, whereas SGA appeared to improve after the educational intervention (II). The combined intervention in study III resulted in a weight gain of 3.4 ±11.2 kg (p=0.00 1) 6 months later, that was not seen in the control group. The weight gain was not related to an improvement in ADL capacity or cognitive function (MMSE). A BMI <23 was noticed in more than half of 231 patients admitted for diagnostic evaluation of cognitive function (IV). The weight increased by 0.5 kg (p<0.001) during a 3-week hospital stay.

Concurrently, an increase in MMSE by 1 point was noticed (p=0.0001). The changes in weight and MMSE did not correlate. BMI <23 seemed to predict shorter survival (OR 3, 95% Cl 1.3-6.7), even after adjustment for age, gender and comorbidity, whereas type and degree of dementia were not related to 7-year mortality.

In conclusion, a limited nutrition education programme to care personnel did not affect the nutritional status of SF residents 5-6 months later. No overt improvements in the knowledge of the care staff were observed. Nutritional status was related to cognitive function. Care staff education and oral supplementation resulted in weight gain in GLD residents. A BMI <23 seemed to predict mortality in individuals with cognitive failure. More studies are needed to settle the relationships between nutrition and cognition in elderly individuals.