ENIGMA delivers good predictive ability in validation test

A novel malnutrition risk index, called the Elderly Nutritional Index for Geriatric Malnutrition Assessment (ENIGMA), retains its good predictive ability even upon external validation in a community-dwelling cohort of older adults, according to a new Singapore study.

“In the present study, the concurrent and predictive validity of the ENIGMA construct is replicated in an external evaluation study of community-dwelling older persons, supporting its accuracy and utility in assessing malnutrition risks among community-dwelling older adults,” the researchers said.

The external validation cohort consisted of 2,924 older adults who had participated in the second wave of the Singapore Longitudinal Ageing Study. Aside from ENIGMA. Participants were assessed using the Mini-Nutritional Assessment-Short Form (MNA-SF) and the Geriatric Nutritional Risk Index (GNRI); functional dependency, 10-year mortality, and known correlates of malnutrition were also included in the analysis.

Higher ENIGMA risk categories correlated significantly with lower MNA-SF scores, which indicated higher malnutrition risk. Such association between the two tools was also observed when looking at the individual MNA-SF components, including reduced food intake due to illness, weight loss, neuropsychological problems, and mobility, among others. [Br J Nutr 2021;doi:10.1017/S0007114521003433]

Similarly, ENIGMA could also capture malnutrition risk as assessed by GNRI, with high-risk grades in the former correlating significantly with low scores in the latter.

Nevertheless, ENIGMA’s concordance with either comparator tools was low. Its percentage overall agreement with MNA-SF and GNRI was 45.9 percent and 45.8 percent, respectively.

Despite its low agreement with other malnutrition risk tools, ENIGMA was significantly correlated with baseline functional dependency after controlling for confounders such as multimorbidity, mobility, and depression (odds ratio [OR], 1.64, 95 percent confidence interval [CI], 1.01–2.65).

Moreover, univariate analysis revealed that medium- (OR, 1.71, 95 percent CI, 1.01–2.89) and high-risk (OR, 2.68, 95 percent CI, 1.45–4.98) ENIGMA classifications could significantly predict functional dependency at follow-up. Neither MNA-SF nor GNRI correlated with functional dependency at baseline or at follow-up.

Similarly, only the moderate-to-high nutritional risk as determined by ENIGMA could significantly predict mortality outcomes (hazard ratio, 1.65, 95 percent CI, 1.04–2.62).

“As the ENIGMA was developed and validated among community-dwelling older adults, these results thus support its potential use in geriatric patients in primary care settings,” the researchers said. “It is brief and simple to use, without the need for comprehensive or brief geriatric assessment of physical or neuropsychiatric impairment.”

“More studies should also be conducted to evaluate whether the application of the ENIGMA in diverse settings result in measurable changes in nutritional status and tangible impacts on health outcomes and life expectancy,” they added.