Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the 'malnutrition universal screening tool' ('MUST') for adults.

Abstract

The 'malnutrition universal screening tool' ('MUST') for adults has been developed for all health care settings and patient groups, but ease of use and agreement with other published tools when screening to identify malnutrition requires investigation. The present study assessed the agreement and the prevalence of malnutrition risk between 'MUST' and a variety of other tools in the same patients and compared the ease of using these tools. Groups of patients were consecutively screened using 'MUST' and: (1) MEREC Bulletin (MEREC) and Hickson and Hill (HH) tools (fifty gastroenterology outpatients); (2) nutrition risk score (NRS) and malnutrition screening tool (MST; seventy-five medical inpatients); (3) short-form mini nutritional assessment (MNA-tool; eighty-six elderly and eighty-five surgical inpatients); (4) subjective global assessment (SGA; fifty medical inpatients); (5) Doyle undernutrition risk score (URS; fifty-two surgical inpatients). Using 'MUST', the prevalence of malnutrition risk ranged from 19-60% in inpatients and 30% in outpatients. 'MUST' had 'excellent' agreement (kappa 0.775-0.893) with MEREC, NRS and SGA tools, 'fair-good' agreement (kappa 0.551-0.711) with HH, MST and MNA-tool tools and 'poor' agreement with the URS tool (kappa 0.255). When categorisation of malnutrition risk differed between tools, it did not do so systematically, except between 'MUST' and MNA-tool (P=0.0005) and URS (P=0.039). 'MUST' and MST were the easiest, quickest tools to complete (3-5 min). The present investigation suggested a high prevalence of malnutrition in hospital inpatients and outpatients (19-60% with 'MUST') and 'fair-good' to 'excellent' agreement beyond chance between 'MUST' and most other tools studied. 'MUST' was quick and easy to use in these patient groups.