[Methods of feeding and illness in infants in the first six months of life]

There is now a general agreement, that breast feeding prevents infantile gastrointestinal infections and non-gastrointestinal infections in all settings. In this study, we tried to pay attention to Kovar et al. who suggested that studies to detect the effects of breast feeding on health should be undertaken for other reasons and should include the collection of related information as part of the study.

Data collection. The population consisted of all infants born in the community of Krushevac in 1994. After at least 36 weeks of pregnancy their weight was 2500 g or more. The babies were selected by a modified cluster sampling. Data were collected by access to health cards where data of systematic medical examination of infants after the first, third and sixth month of life were recorded. Feeding method. Breast feeding was defined as the period of breast feeding from birth until the infant was totally weaned, even if other food was offered. The term bottle feeding referred to no breast feeding but to an infant formula or cow milk. Each infant was classified every month into either a breast feeding group or a bottle feeding group. Infectious diseases. We studied gastrointestinal and respiratory diseases, and all other illnesses except trauma were included in the third category. Statistical analysis. Every infant was classified each month as breast or bottle fed and thus was included in a 6-month observation. An episode of illness was associated either with breast feeding or bottle feeding according to the method of feeding in the month during which the episode occurred. The incidence density of diseases in breast fed infants was computed as follows: The rate numerator corresponded to the total number of episodes of diseases reported while infants were breast fed. The denominator was the number of month of breast feeding of all infants. The rate of diseases in bottle fed infants was obtained in the similar way. To compare the rates of illnesses in breast fed and bottle fed infants, we computed the incidence density ratio (IDR). So, when less than one, the IDR indicated a protective effect of breast feeding.

Figure 1 illustrates the distribution of infants by the method of feeding at each age in months. Sixty four percent of babies were breast fed at the end of the first month, 43 percent at the end of the third month and 37.8 percent at the end of the sixth month. A total of 305 episodes of diseases were observed in 448 infants. An approximate incidence density analysis revealed the significant protective effect of breast feeding in relation to the total number of illnesses during the first sixth months of life (Table 1). The incidence density for gastrointestinal diseases was 49 percent lower in breast fed infants than in bottle fed infants (IDR = 0.51). The protective effect of breast feeding on gastrointestinal diseases persisted after individual adjustment according to the infant's age (Table 2). The rough incidence density for respiratory diseases was 38 percent lower in breast fed infants than in bottle fed infants (IDR = 0.62)(Table 1). The protective effect of breast feeding on respiratory diseases persisted after individual adjustment according to the infant's age (Table 3).

The results of this retrospective cohort study revealed protective effects of breast feeding in relation to respiratory illnesses as well as gastrointestinal diseases during the first six months of life. The statistical analysis based on Person-time methods best emphasizes the common problems that arise in such studies, because it takes in account the fact that (1) the outcome may be recurrent in one individual; (2) the duration of breast feeding is associated with the infant's age, and (3) the incidence is also age related. We conclude that breast feeding was protective against respiratory illnesses including ear infection and against gastrointestinal diseases. The study thus reinforces the benefits of breast feeding for