Nutrition risk classification: a reproducible and valid tool for nurses.

BACKGROUND Incorporating the nursing staff to assist with the screening process on admission will allow patients who are at nutritional risk to be assessed by registered dietitians earlier in their hospital stay. The goal of this study was to develop an objective, valid, reproducible nutrition screen for use by registered nurses (RNs) to allow for nutrition classifications of hospitalized patients. METHODS The current nursing admission assessment form was modified to contain questions on weight loss history, percentage of ideal body weight, and alterations in dietary intake and gastrointestinal function. Assessments were completed within 48 hours of admission. On the basis of the answers to these questions, patients were classified as "at nutritional risk" or "low nutritional risk." In phase 1, to assess reproducibility of the form, a prospective study between staff RNs and a nutritionist was undertaken on 186 consecutive adult admissions. Nutrition screening and classification was done independently by both practitioners. In phase 2 of the study, prospective validation of the form contrasting prealbumin (PAB) levels with RN nutritional risk classification (n = 56) was investigated. RESULTS Interobserver agreement of nutrition classification between RN and nutritionist was 97.3% (p = .95). Twenty-nine patients were classified at low nutritional risk (27 normal PAB and 2 low PAB); 27 patients were classified as at nutritional risk (16 normal PAB and 11 low PAB) (chi 2 = 8.9, p < .003, power 0.8). The sensitivity of the tool was 84.6%. CONCLUSION To our knowledge, this is the first nutrition screening tool designed for staff RNs that has been tested for both validity and reproducibility.