N R Sahyoun1, P F Jacques, G E Dallal, R M Russell. 1. National Center for Health Statistics, Office of Analysis Epidemiology and Health Promotion, Hyattsville, Md. 20782, USA.
Abstract
OBJECTIVE: To evaluate the Nutrition Screening Initiative (NSI) checklist as a screening and an awareness/educational tool in an elderly population. DESIGN: Epidemiologic follow-up study. Information similar to the questions of the NSI checklist was collected by the Nutrition Status Survey of Boston elders between 1981 and 1984. Vital status of volunteers was obtained during 8 to 12 years of follow-up. SUBJECTS/ SETTING: Community-dwelling men (n = 200) and women (n = 381) aged 60 years and older who participated in the survey. STATISTICAL ANALYSES PERFORMED: Multivariate analysis was used to assess the association between mortality and each of the NSI-similar questions and the cumulative score, which is the sum of the values assigned to each question. Attributable risk percent, a measure of association, was calculated to measure the percentage of deaths that could potentially be prevented if the risk factors or their consequences were eliminated. RESULTS: Eating meals alone, problems biting or chewing, difficulties with shopping or cooking, and taking more than three medications per day were positively associated with mortality (P < .05). The cumulative score, although significant, was a weaker predictor of mortality. Attributable risk percent of mortality was 19.9% and 51.2% for men and women, respectively. APPLICATIONS/ CONCLUSIONS: Some but not all of the individual questions of the NSI checklist equivalent were significantly associated with mortality and identify specific problems that may have a long-term negative effect yet may be missed if the cumulative score were the sole criterion for screening people. The attributable risk percent suggests that the checklist may be best used as an awareness/educational tool as intended originally and could have an important public health effect on early death of community-dwelling elderly people.
OBJECTIVE: To evaluate the Nutrition Screening Initiative (NSI) checklist as a screening and an awareness/educational tool in an elderly population. DESIGN: Epidemiologic follow-up study. Information similar to the questions of the NSI checklist was collected by the Nutrition Status Survey of Boston elders between 1981 and 1984. Vital status of volunteers was obtained during 8 to 12 years of follow-up. SUBJECTS/ SETTING: Community-dwelling men (n = 200) and women (n = 381) aged 60 years and older who participated in the survey. STATISTICAL ANALYSES PERFORMED: Multivariate analysis was used to assess the association between mortality and each of the NSI-similar questions and the cumulative score, which is the sum of the values assigned to each question. Attributable risk percent, a measure of association, was calculated to measure the percentage of deaths that could potentially be prevented if the risk factors or their consequences were eliminated. RESULTS: Eating meals alone, problems biting or chewing, difficulties with shopping or cooking, and taking more than three medications per day were positively associated with mortality (P < .05). The cumulative score, although significant, was a weaker predictor of mortality. Attributable risk percent of mortality was 19.9% and 51.2% for men and women, respectively. APPLICATIONS/ CONCLUSIONS: Some but not all of the individual questions of the NSI checklist equivalent were significantly associated with mortality and identify specific problems that may have a long-term negative effect yet may be missed if the cumulative score were the sole criterion for screening people. The attributable risk percent suggests that the checklist may be best used as an awareness/educational tool as intended originally and could have an important public health effect on early death of community-dwelling elderly people.