During the past two decades, one of the most important demographic changes seen has been the entry of women into the labor market. Currently, in some countries nearly 50% of mothers of infants work outside the home. Therefore, day care centers (DCC) are increasingly needed for providing care to the children of working mothers during working hours. Nevertheless, these facilities are somewhat heterogeneous, requirements to run are minimal and regulations are not uniform.

Prior to 1974, it was thought that the DCC had no influence on child health. Recently, however, reports have indicated the possible relationship between infectious diseases and the use of DCC. Overcrowding in the DCC and high child/caregiver ratio, in addition to host characteristics such as age and nutritional status, are thought to be critical in developing infectious diseases.

Acute respiratory infections (ARI) are particularly relevant in this context due to their high frequency and to their potential negative impact on the infant’s nutritional status. After 18 months of age, there is a continuous decline in the incidence rate. Some studies have suggested that children cared for at DCCs have a two-fold risk of infectious diseases when compared with children staying at home, the most frequent diseases being ARI episodes. Although this relationship has not been fully defined, it has been mentioned that DCCs are risk factors for ARI. Such risks may vary according to the geographical area and DCC characteristics.

At the international level, the type of childcare and frequency of use differs from country to country and is dependent on children’s ages. In the United States and Northern Europe, it is very common for children under 3 years of age to remain at home rather than to be cared for at DCCs. Social support programs that encourage parents to take care of their children during the first months of life facilitate this practice. Leave permits lasting for 3–6 months are granted for working parents and, in some cases, extend for up to 1 year. Only 20% of children attend public DCCs, which are often located near parents’ working places. DCC organizational structure is varied. For example, in Europe a DCC can provide care for up to 60 children, organized into groups ranging from 10–12 children each. There are up to two caregivers for groups of six infants. There are playgrounds at these facilities, and strict sanitary conditions are followed.

In Mexico, the Mexican Institute of Social Security (IMSS) provides health care for workers and their families including, among other services, day care for the children of employed mothers. The capacity of these DCCs varies, ranging from 40–400 children ranging in age from 43 days to 4 years. There are 20–40 children, divided according to ages, per room. In 1997, enrollment in the IMSS DCCs was 68,078 children. This number is increasing steadily, particularly for children under 3 months of age. This is due to the increasing number of mothers entering the labor market and to the short maternity-leave period permitted under the Social Security Law until 42 days after childbirth. This provides little time for the mother to provide care at home to the newborn.

Given the importance of the consequences of ARI on infant groups, this article assesses whether DCCs are risk factors for ARI in the context of the Mexican Institute of Social Security. The main objective is to determine the incidence, duration, and severity of ARI in infants attending DCCs compared with ARI incidence among infants cared for at home, and to assess the magnitude of the risk at institutional DCCs.

We conducted a cohort study in which two groups of infants, one cared for at DCCs and the other at home, were followed for 1 year. Infants aged 43 days to 4 months were included. Infants enrolling for the first time in one of seven DCCs were selected to participate in the study. The DCCs chosen for participation are located in Mexico City; they share the same sanitary conditions and the same city district. To assess the infants staying at home, we invited mothers of children entering for the first time into the well-child program at IMSS primary care facilities to take part in the study. Selection criteria included that they live in the same neighborhood as the DCC, but that they had not previously used DCC services. This method assured that both groups were exposed to the same outdoor environmental conditions. Infants having increased susceptibility to ARI, such as a history of perinatal morbidity, certain types of congenital malformations, metabolic diseases, and chronic infectious diseases were excluded. Those who changed DCCs during the follow-up period or who moved out of the neighborhood were eliminated because they were exposed to different environmental conditions.

We defined home care as the care provided by the family to the child in the house where the latter lived. Day care was defined as the care provided regularly to the child at the institutional DCC.

Acute respiratory infection was defined as the presence of a runny nose and/or a cough in addition to one or more signs or symptoms of infectious syndrome (fever, malaise, and hyporexia). The duration of illness should be defined as less than 21 days. We considered that ARI was resolved when the child had no symptoms and had returned to his/her normal activity. We defined a new episode of ARI as the reappearance of respiratory symptoms (as mentioned here) after at least 3 symptom-free days subsequent to the last episode. This included whether or not the child had a persistent runny nose and/or a cough and clinical data of infectious syndrome. During the follow-up, we registered whether or not they had ARI, the length of the illness, and the number of episodes. High morbidity of ARI was defined as having 7 or more episodes per year. Current criteria have defined that an infant may have 4 to 6 ARI episodes per year.

Every ARI episode was classified based on its severity. A severe episode was present when the infant had a fever for at least 3 days, and an illness lasting more than 10 days.

These include infant variables such as age, gender, ARI history, and the age of the child at the first ARI episode, breastfeeding, birth weight, and nutritional status. During follow-up, the interviewer recorded the age at weaning and the duration of breastfeeding. In addition, growth-related variables such as weight and height were registered. Infant nutritional status was evaluated by means of the National Center for Health Statistics references. The indices of weight for age, weight for height, and height for age were calculated by means of Z scores.

The variables of each family consisted of the following: the mother’s age and education level; whether she had a paid job outside the home; socioeconomic status; number of family members; number of older siblings at home; family history of chronic diseases (allergic, cardiovascular, diabetes mellitus, cancer, chronic respiratory); whether the child was exposed to tobacco smoke (more than five cigarettes daily at the home), and days of exposure to a relative having ARI.

We used a proxy26. M. Bronfman, H. Guiscafré, V. Castro, R. Castro and G. Gutiérrez, II. La medición de la desigualdad: una estrategia metodológica, análisis de las características socioeconómicas de la muestra. Arch Inv Med 19 (1988), p. 351. View Record in Scopus | Cited By in Scopus (67) to measure socioeconomic status including overcrowding, type of floor, piped water, and sewerage. In addition, the educational level of the head of household (either the mother or the father) was included in the proxy.