Nutritional risk in a university hospital. Challenges and consequences in clinical practice

Introduction: Malnutrition is present in 20-50% of hospitalised patients, leading to increased risk for adverse clinical outcomes and even mortality. Nutritional status is often ignored during hospitalisation. The Bergen Nutritional Strategy was a multi-pronged effort introduced to increase focus on improving patients’ food and mealtime routines, and the routines used by staff to evaluate nutritional risk. Another element of this strategy was to ensure proper nutritional care during patients’ hospitalisation. Guidelines were implemented and hospital staff was educated. Repeated point-prevalence surveys were introduced in 2008 to increase awareness about patients’ nutritional status and to improve nutritional care. Aims: The main objective of this dissertation research was to objectively evaluate the Bergen Nutritional Strategy. This was accomplished by conducting three studies, each addressing different aspects of the strategy. The aim of the first study was to evaluate whether the Bergen Nutritional Strategy had positive effect overall on nutritional care of patients at Haukeland University Hospital (Paper I). The aim of the second study comprised two aspects. The first was to study in detail the components of the Nutritional Risk Screening (NRS 2002) tool to determine the minimum number of components necessary to clearly classify a patient as being ‘at nutritional risk’. The intention was to simplify the screening procedure, if possible. The second aspect was to assess whether being ‘at nutritional risk’ is associated with increased morbidity, mortality, and health-care costs. This was assessed during a oneyear follow-up (Paper II). The aim of the third study was to determine the prevalence of nutritional risk as a function of patients’ age, disease category, and the hospital department in which they were treated in order to better understand in which departments and patients groups nutritional care is most crucial to monitor (Paper III). Methods: Nutritional registrations performed as point-prevalence surveys were conducted every three months during 2008 and 2009. Any changes in clinical practice at hospital units were assessed by repeated surveys. The first survey was conducted at 14 hospital units and the next seven at 51 units. NRS 2002 was used to classify patients as ‘at nutritional risk’ or ‘not at risk’, according to their nutritional status and severity of illness (See Appendix 2, section 11.2). Data on length of hospital stay, new hospital admissions, and mortality were obtained from the patient administrative system. Patients: For the eight point-prevalence surveys in 2008 and 2009, 5849 adult hospitalised patients were subject for inclusion; 3604 patients were included in study I, and 3279 patients were included in studies II and III. Results: In study I, 1230 (34%) of 3604 patients were at nutritional risk. Among these, 53% received nutritional treatment, and dieticians were involved in the treatment of only 5%. The proportion of patients who were screened increased significantly from the first to the last survey (p=0.012). However, the proportion of patients who received nutrition treatment did not increase during the study period (p=0.66). In study II, 3279 patients were followed for one year. Of these, 29% were at nutritional risk, as assessed by NRS 2002. Being at nutritional risk was strongly associated with increased morbidity and mortality. Even the initial screening robustly identified adverse outcomes. Every single item of the screening tool was found to be a significant independent risk predictor. A positive response to one or more of the initial four questions in NRS 2002 was associated with increased risk of morbidity and mortality, and positive answers to all four questions were associated with a 13 times greater risk of dying during the following year (OR 13.0, 95% CI 4.52 to 37.6). In study III, compared to well-nourished patients, those at nutritional risk were more often female (53% vs.50%); underweight (mean Body Mass Index [BMI] 21.4 vs. 25.3 kg/m²); and older (mean age: 67.8 vs. 63.0 years). The prevalence of nutritional risk increased with age, being 40% for patients ≥80 years and 21% for those 25 kg/m2), and patients with fewer than four diagnoses were frequently found to be at nutritional risk. A high prevalence of nutritional risk was found in nearly all patient groups and hospital units. However, it was most common among patients with infections, cancer, or pulmonary diseases. The greatest numbers of patients at nutritional risk were in the departments of general medicine or surgery. Nearly half (40%) of the patients who were discharged from hospital to nursing homes, and 25% of the patients who were discharged to their own home were at nutritional risk. Conclusions This comprehensive study of a university hospital patient population revealed that a high proportion of the patients in this university hospital were at nutritional risk during the study period. Far from being simply an academic finding, this risk was strongly associated with adverse outcomes, sometimes even death. Nutritional depletion is a significant risk factor for morbidity, increased use of hospital services, and premature death. Our findings support the elevated need for nutritional screening in hospitals. Patients at nutritional risk were identified in all disease categories and all ages. A screening tool is immensely valuable for categorising patients at nutritional risk, and NRS 2002 was found to be suitable for identifying high-risk patients. The initial four questions of NRS 2002 were strong predictors of hospitalisation, morbidity, and most importantly, mortality, among hospitalised patients. Thus the combined use of just these four questions would be appropriate and effective to use as an initial screening of hospitalised patients. Implementation of the Bergen Nutritional Strategy improved the screening performance among the hospital staff, but did not improve the patients’ nutritional treatment. Therefore, more intense efforts are necessary to improve nutritional practice and staff knowledge in hospitals