Nutritional status of hospitalized elderly patients in Ethiopia: a cross-sectional study of an important yet neglected problem in clinical practice

Background Malnutrition is a common geriatric syndrome affecting approximately half of the older population with a more pronounced occurrence rate in those hospitalized. It affects the physiology, and results in poor humanistic and clinical outcomes. In Africa, particularly in Ethiopia, albeit multiple studies are available on malnutrition in non-hospitalized older population, similar studies in inpatient settings are scarce. Therefore, this study was conducted with the intention to quantify the prevalence of malnutrition in older patients on inpatient admission and determine its associated factors. Methods A total of 157 older inpatients aged 60 years and above were included in the present study. The data collection format was developed after an in-depth review of relevant literatures. The full Mini-Nutritional Assessment (MNA) tool was employed to assess the nutritional status on admission. Data completeness was checked thoroughly. Descriptive statistics and logistic regression analysis were conducted using STATA 15.0. The area under the receiver operating characteristic curve (ROC), Hosmer-Lemeshow test, and classification table were computed to evaluate the final model goodness-of-fit. Results Of the total study subjects, 81% were malnourished (MNA score <17) and 17% were at risk for malnutrition (MNA score of 17.5-23.5). However, upon review of the patients' medical charts, malnutrition diagnosis was recorded in only two patients. Rural residence (AOR = 2.823, 95%CI: 1.088, 7.324), self-reported financial dependence for expenses (AOR = 4.733, 95%CI: 1.011, 22.162), and partial dependence in functional autonomy on admission (AOR = 3.689, 95%CI: 1.190, 11.433) significantly increased the risk of malnutrition. The area under the ROC curve (0.754) and the Hosmer-Lemeshow test (p = 0.7564) indicated that the final model reasonably fits the data. The model`s sensitivity is 96.85%. Conclusion In the present study, an alarmingly high prevalence of malnutrition was identified older inpatients. The problem went undiagnosed in a similar percentage of patients. Several available literatures indicate the presence of an association between nutritional status and patient outcomes, thus strict nutritional screening at inpatient admission and intervention are recommended with special emphasis for those from rural areas, with financial dependence, and with functional impairment on admission.