Iron deficiency anemia remains a public health problem in children under the age of five,
women of childbearing age and pregnant women, whose consequences can reverberate over
the years. Thus, this dissertation aimed to carry out a narrative review of the algorithms and
qualitative models available in the literature to estimate the availability of dietary iron, that is,
the amount of absorbable iron, and to assess the availability of iron in school meals at the
Municipal Centers of Infant Education (CMEI), from Maceió-AL. This work consisted of
three stages, namely: 1) narrative review of the algorithms and qualitative models available in
the literature to estimate available iron; 2) selection of algorithms to be used in the second
stage of the dissertation; 3) evaluation of the availability of iron in the school meals of the
CMEI in Maceió-AL. Eighteen algorithms and 2 qualitative models were found to estimate
iron availability, over 39 years, between 1978 and 2017. The selected algorithms were:
Monsen and Balintfy (1982), Singer et al. (1982), Du et al. (2000) and Rickard et al. (2009).
Estimates of the availability of iron in school meals, obtained in each of the four algorithms,
were expressed as mean and mean standard error. Correlation analysis was performed
between the estimates, followed by the analysis test with repeated measures (ANOVA) to
compare the estimates, and the Bland-Altman test to verify the agreement of the results.
Correlation analysis between diet constituents was also performed to evaluate the proposed
menu. Absorbable iron estimates ranged from 0.23 to 0.44 mg/day, and the estimated
absorption efficiency ranged from 4.8% to 9.7%. When comparing the estimates with the
nutritional recommendations, it was found that, although the amount of total iron apparently
met the iron recommendations of the children assisted by the CMEI, on the vast majority of
the days analyzed, the amount of absorbable iron did not meet the nutritional
recommendations adopted by the PNAE. Based on the Bland-Altman test, it was found that
the pair of algorithms by Monsen and Balinfty (1982) and Rickard et al. (2009) showed the
greatest agreement with each other. The results of the present study, associated with the ease
of its application, allow us to recommend the algorithm of Monsen and Balinfty (1982) as the
one of choice, in the context investigated here. The correlation analysis between the
constituents of the meals showed that animal tissue (AT) and non-heme iron (FNH) showed a
moderate positive correlation (r = 0.42; p <0.05), while TA and calcium (r = -0.54) and
calcium and phytic acid (r = -0.46) showed a moderate negative correlation (p<0.05). It is
concluded that the amount of iron available in school meals estimated in this study does not
meet, on most days, the nutritional recommendations adopted by the PNAE. In addition, the
analysis of the menu's constituents confirmed the need to adopt strategies to increase the
availability of iron in school meals, such as associating the consumption of FNH and ascorbic
acid and reducing the presence of inhibitory factors, especially in meals with higher amounts
of iron.