Validation of the Mini Nutritional Assessment-Short Form in a population of frail elders without disability. Analysis of the Toulouse Frailty Platform population in 2013

**Objective** To assess the validity of the Mini Nutritional Assessment-Short Form (MNA-SF) in elderly patients from the Toulouse Frailty Platform. **Participants** Overall, 267 patients aged 65 and over, without severe cognitive impairment (i.e. Mini Mental Status Examination > 20 and CDR<1), no physical disability (i.e. Activities of Daily Living ≥ 5) and no active cancer history (over the past 12 months) were included in 2013. **Measurements** Receiver operating characteristic (ROC) analyses were used to assess the predictive validity of the French version of the MNA-SF for good nutritional status (defined as a full MNA score≥24/30). Analyses were conducted in the overall sample and then in subgroups of frail and pre-frail subjects according to the frailty phenotype. Optimal cut-off points were determined to obtain the best sensitivity/specificity ratio and the highest number of correctly classified subjects. **Results** Among 267 patients, mean age=81.5±5.8; women=67.0%; 138 (51.7%) were frail, 98 (36.7%) were pre-frail and 31 (11.6%) were robust. Given their MNA-SF scores, 201 (75.3%) had a good nutritional status, 61 (22.8%) were at risk of malnutrition and 5 (1.9%) were malnourished. In the overall sample, but also in subgroups of pre-frail or frail elders, the areas under ROC curves were 0.954, 0.948 and 0.958 respectively. The 11 points cut-off provided the best correct classification ratio (91.4%); sensitivity=94.0%, specificity=83.3%. **Conclusion** The MNA-SF appeared to be a validated and effective tool for malnutrition screening in frail elders. Implementing this tool in clinical routine should contribute to improving the screening of malnourished frail individuals.