Abstract (eng)
Rationale
In the near future, the demand for residential facilities such as nursing homes (NHs) will enormously increase, due to the drastic rise in old, dependent people. Accordingly, the demands for quality of nursing, including nutritional care influencing the well-being and health of the elderly, become increasingly important. This thesis, therefore, focuses on nutritional management in European nursing homes and the nutritional status of their residents in order to identify nutrition-related factors that may predict outcomes such as weight loss or mortality.
Methods
The data for this thesis are derived from the survey "nutritionDay in nursing homes" 2007-2010, which is a voluntarily, annually repeated one-day cross-sectional audit with an outcome evaluation after six months. Data were collected with the help of questionnaires, requesting unit characteristics, residents´ general characteristics, residents´ actual nutritional intake and outcome. In the first section of this thesis, data from 470 units in 234 nursing homes in 14 European countries were quantitatively described and summarised (descriptive analysis). For the analyses in the second section, Hungarian units were excluded due to missing outcome data. Univariate generalised estimating equations were performed with binary logistic regression to describe the relationship of each baseline independent variable to the outcomes "weight loss of at least 10% or death", respectively "weight loss of at least 10%" and "death". Significant predictor variables (P<.05) were included in multivariable regression models.
Results
19 167 residents with a mean age of 76.8±16.8 years, 71% female, took part in the survey.
One third of the population had a BMI below 22 kg/m2, showing a potential risk for malnutrition. 31% of the residents had lost weight and 19% had gained weight in the previous year. On the day of evaluation, 62% of the residents ate everything at lunch, 27% ate half a portion, 6% a quarter of the meal and 2% ate nothing. Only half of the "non-eaters" received nutritional support. After six months, the outcomes of 54.4% of the residents were collected. Of these, 85% were still in the NH, 1.1% had been transferred, 1.2% had been discharged home and 12.1% had died.
For the outcome analyses, data from 9 946 residents, aged 85.1±9.9 years and 78% female, were included. The main finding was that higher age, immobility, a short time since unit admission, high drug intake, pressure sores, impaired cognitive status, antibiotics use, and reduced food intake before and on nutritionDay (p<0.0001-0.018) were independent predictors for the outcome "weight loss of at least 10% or death within 6 months". Furthermore, a high BMI, weight loss before, impaired cognitive status, and reduced food intake before and on nutritionDay were significant influencing factors for the outcome "weight loss of at least 10% within 6 months". Lastly, reduced food intake before and on nutritionDay, higher age, immobility, pressure sores, antibiotics use, and male gender were independent risk factors for "death within 6 months". Thereby, food intake before and at lunch on nutritionDay were the only independent influencing variables on all three outcome categories.
Conclusion
This analyses revealed that very simple individual resident and nutrition related data, like past and actual nutritional intake, showed the highest significance for predicting a bad outcome. Therefore, besides detailed anamnesis for risk factors, simple regular questions on actual intake (with the help of plate symbols), regular weighing and inquiring about appetite should be included routinely in the residents’ charts in NHs.