Adult Malnutrition (Undernutrition) Screening: An Evidence Analysis Center Systematic Review.

Abstract Background Malnutrition screening relies on quick and easy-to-use tools that require minimal training of users. Valid and reliable tools should avoid under-referral of adults with a malnutrition diagnosis or over-referral of those without a malnutrition diagnosis to dietitians. Objective Our aim was to conduct a systematic review of adult malnutrition screening tools for validity, agreement, and reliability, and to determine the costs of the malnutrition screening procedure. Methods Following a structured process, a comprehensive search using PubMed, MEDLINE, EMBASE, and CINAHL was conducted for relevant research published between 1997 and 2017 that examined the validity, agreement, reliability, and costs of nutrition screening tools. Results Sixty-nine studies met the inclusion criteria. The Malnutrition Screening Tool exhibited moderate validity, agreement, and reliability based on Grade I (Good/Strong) evidence. The evidence supporting the conclusions for the remaining tools was Fair (Grade II). The Malnutrition Universal Screening Tool exhibited high validity and moderate agreement and reliability. The Mini Nutritional Assessment–Short Form exhibited moderate validity and reliability and low agreement. The Short Nutritional Assessment Questionnaire and the Nutrition Risk Screening 2002 exhibited moderate validity and reliability and the Mini Nutritional Assessment–Short Form Body Mass Index exhibited high validity and moderate agreement. However, these last three tools were missing reliability or agreement data. Limited data were available to evaluate malnutrition screening costs. Conclusions This review provides an analysis of validity, agreement, and reliability of tools to screen adults for malnutrition, regardless of their age, medical history, or location. No tools were found with high validity, reliability, and strong supportive evidence. Tools most often achieved moderate validity, agreement, and reliability, and had large variations in individual results. The minimum validity and reliability of tools to screen adults for malnutrition should be established to shape future research. Cost data for the screening process should be obtained and examined.