لخّصلي

"Follow-up losses" were considered when the mother-child pair was not located in two consecutive follow-up periods, when they moved to other cities, or when mothers no longer had a regular bond/coexistence with their children, as in cases of donation for adoption.

The data used for this study come from the birth cohort: “SAND — Health, Food, Nutrition, and Child Development: a cohort study”, which aimed to analyze aspects related to health, food, nutrition, and the development of infants from birth until the first year of life. The research, carried out in Rio Largo, the third largest municipality in the state of Alagoas, located 28 km away from the capital Maceió, was approved by the Research Ethics Committee, and all mothers signed an informed con- sent form. Methodological details are described in the inves- tigation by Melo et al.9
Between February 2017 and August 2018, data collections were carried out four times as follows:

1. Birth of the child (maximum seven days after birth); 2. At 3 months of age;


2. At 6 months of age; and


3. At 12 months of age.
   Teams of previously trained researchers applied structured and precoded questionnaires to obtain social, economic, demo- graphic, and health information about mothers and infants. Information on the child’s weight and length at birth was obtained through consultation of medical records.
   Of the total number of births that occurred in this period (n=394), 109 dyads did not meet the eligibility criteria, 41 refused to participate in the research, and 1 mother was iden- tified with a language disorder that compromised communica- tion, totaling 243 infants at baseline. “Follow-up losses” were considered when the mother-child pair was not located in two consecutive follow-up periods, when they moved to other cities, or when mothers no longer had a regular bond/coexistence with their children, as in cases of donation for adoption. Thus, at 3 months, 210 children participated in the research; at 6 months, 198; and at 12 months, 186 infants. Given the purpose of our study to investigate the association between sleep duration and iron deficiency anemia, our analyses were conducted with 123 infants, who had complete data on these variables.
   Sleep information was obtained at 3, 6, and 12 months of age, through the validated and translated version into Brazilian Portuguese of the semistructured Brief Infant Sleep Questionnaire (BISQ), which investigates the usual and last week’s sleep char- acteristics of the child, according to mothers’ reports.10
   The measures used in this study were as follows:


4. Nighttime sleep duration (from 7 p.m. to 7 a.m.) and 2. Daytime sleep duration (from 7 a.m. to 7 p.m.).
   Total sleep duration was represented by the sum of hours slept during the night and the day. To obtain the average daily sleep time in the first year of life, the arithmetic mean of sleep duration at 3, 6, and 12 months was calculated. The cutoff point for classifying short sleep duration was total sleep time