Objectives: The aim of this study was to determine the: (1) prevalence of undernutrition as determined by the 'DETERMINE Your Nutritional Health Checklist' (NHC) and (2) factors independently associated with undernutrition among the older residents of these publicly funded shelter homes in Peninsular Malaysia. Design: A total of 1081 elderly people (59%M) over the age of 60 y were surveyed using questionnaires determining baseline demographics, nutritional and cognitive status, physical function and psychological well-being. Setting: Shelter homes, Peninsular Malaysia. Results: In all, 41.4% (n ¼ 447) were nourished (score o3), 32.1% (n ¼ 347) at moderate risk (score between 3 and 5) and 26.6% (n ¼ 287) were at high risk of undernutrition (score45) according to the NHC. A large proportion of subjects were underweight with 14.3% of subjects recording a low body mass index (BMI) o18.5 kg/m 2 and a further 18.2% recording a BMI between 18.5 and 20 kg/m 2 . The residential geriatric depression score (GDS-12R) (relative risk (RR) ¼ 1.03 (95% confidence interval (CI) 1.01-1.05); P ¼ 0.002) and the number of illnesses (RR ¼ 1.14 (95% CI 1.07-1.21); Po 0.001) were found to be independently associated with nutritional risk (NHC score Z 3). Using a BMIo18.5 kg/m 2 as an objective marker for nutritional risk, the NHC was shown to have a sensitivity of 66.4% (95% CI 58.0-74.2%), specificity of 42.7% (95% CI 39.3-46.1%), positive predictive value of 16.2% (95% CI 13.3-19.5%) and a negative predictive value of 88.4% (95% CI 84.9-91.4%). Conclusions: Many elderly people residing in publicly funded shelter homes in Malaysia may be at-risk of undernutrition, and were underweight. The NHC is better used as an awareness tool rather than as a screening tool.