RESUMEN
OBJECTIVE: This study aimed to develop and validate protein energy malnutrition (PEM) screening tool for older adults in public residential homes, and to test its practicality. DESIGN: This cross-sectional study consisted of two phases: tool development/validation and tool practicality evaluation. In Phase 1, the questionnaire was developed based on literature review and tested for content validity. Older residents were interviewed using this questionnaire to identify potential PEM risk factors. A 24-h recall was used to collect dietary data, and body composition and serum albumin were measured. In Phase 2, practicality of new PEM screening tool was evaluated by intended users. Data were analysed by χ2 test, Fisher's exact test, t-test, Mann-Whitney U test and multiple logistic regression. Akaike Information Criterion (AIC) was used to estimate the best fit model. SETTING: Four public residential homes in central region, Thailand. PARTICIPANTS: 249 older residents residing in public residential homes and eight intended users. RESULTS: 26·9 % had PEM (serum albumin <3·5 g/dl). According to multiple logistic regression and AIC values, PEM predictors were having pressure ulcer, experiencing significant weight loss and taking ≥ 9 types of medicine daily. These predictors were included in PEM screening tool. Regarding the tool performance test, area under the ROC curve was 0·8 (P < 0·001) with sensitivity and specificity of 83·9 and 45·5 %, respectively. For its practicality, eight intended users reported that it was useful and easy to use. CONCLUSIONS: New screening tool may be capable of identifying PEM in older residents, and further testing is required before being recommended for use.