[Validating Latin-American and Caribbean Latin-American food security scale on pregnant adolescents]

Statistically validating the Latin-American and Caribbean household food security scale (ELCSA) which had been linguistically adapted for Colombia during 2008 by the University of Antioquia's School of Nutrition.

This was a descriptive study. The ELCSA scale (95 % confidence interval) was applied to a representative sample of 150 households containing pregnant adolescents from poor and vulnerable populations. The pregnant girls' families were covered by ESE Salud Pereira and had consulted between April and June 2009. Four professional nurses were trained as interviewers regarding the subject of each question. 32 adult-only households and households containing 118 adults, youngsters and children were surveyed. SPSS software was used for the statistical analysis; Cronbach's alpha, factorial analysis and multiple components were used.

ELCSA showed excellent reliability when applied to both adult-only households (Cronbach=0.927) and households having adults, teens and children (Cronbach=0.953). Factor analysis using the Kaiser-Meyer-Olkin sampling adequacy test presented high correlation amongst the items in both single adult households (p = 0.889) and households containing adults, teens and children (p=0.895). The scale's predictive ability was 75 % in the three components identified: quality and quantity of food access or availability of food and non-socially acceptable means of acquisition and distribution.

The results validated the ELCSA scale as being a reliable tool for measuring household food security in Colombian households.