Risk Factors for Childhood Pneumonia among the Urban Poor in Fortaleza, Brazil: A Case-Control Study
Fonseca, W., Kirkwood, B. R., Victoria, C. G., Fuchs, S. R., Flores, J. A., Misago, C., Bulletin of the World Health Organization
Acute respiratory infections (ARI) are one of the leading causes of childhood morbidity and mortality in developing countries, accounting for approximately a third of the 15 million deaths that occur annually among under-5-year-olds (1). The majority (75%) of these ARI deaths are due to pneumonia unassociated with measles.(a) Case management together with vaccination (measles, diphtheria, pertussis) is the main strategy for ARI control (2). Detailed understanding about the epidemiology of ARI is needed both to develop preventive programmes and to identify high-risk groups in order to target more effectively case-management interventions.
Among the factors that have been postulated to increase the risk of ARI among children in developing countries are the following: low birth weight, failure to breast-feed, malnutrition, indoor air pollution, and sociodemographic factors such as large family size, short birth interval, low income, low level of parental education, poor housing, and inappropriate child care practices (2, 3). However, in developing countries evidence on the association between these factors and pneumonia in children is scarce. In this article, we report the results of a case-control study carried out to elucidate the major risk factors for childhood pneumonia among the urban poor in Fortaleza, north-east Brazil.
Fortaleza, the capital of Ceara State, is located at latitude 3[degrees]S in a tropical zone; it includes the metropolitan area and greater Fortaleza and has a population of 2 million. The area's climate is determined by its proximity to the equator and its low altitude (sea level). The temperature is high throughout the year (25-38[degrees]C). The economy of the metropolitan area consists almost entirely of fishing, the textile industry, and commerce, with a heavy reliance on tourism. Patients with pneumonia were recruited for the study from the outpatient and inpatient departments of the Hospital Infantil Albert Sabin, the largest public paediatric hospital in the metropolitan area, catering mostly to low-income families. At the time of the study the infant mortality rate was approximately 90 per 1000 live births. According to official estimates, over half of the population live in favelas (shanty towns). A recent survey of 8000 families in Ceara State showed that respiratory infections are the second most important infectious cause of infant death (11%) referred to by mothers and the commonest reason for using health services (38%).(b)
Between 1 July 1989 and 30 June 1990 a total of 650 cases and 650 matched controls were recruited to the study. This gives a 90% power for detecting a relative risk of [less than or equal to]1.6 as significant at the 5% level if the prevalence of exposure among controls is 15-75%. (4). Such a sample size will detect a relative risk of [less than or equal to]2 as significant for a prevalence of exposure in the range 4-90%.
Selection of cases
Cases were children aged 0-23 months with radiologically confirmed pneumonia who were seen at the outpatient clinic of the paediatric hospital. It is routine practice to give all children with suspected pneumonia a chest X-ray. Only children whose chest X-ray showed evidence of pulmonary infiltration were included in the study. All the X-ray results were examined by a paediatric radiologist.(c) Children presenting with wheezing or a recent history of aspiration of a liquid or a foreign body were excluded, as were those with measles or underlying diseases such as symptomatic congenital heart disease, congenital malformation, cerebral palsy, cystic fibrosis or acquired immunodeficiency syndrome (AIDS). Approximately 800 children were recruited as potential cases and 4% were excluded using these criteria; also excluded were all children who died during the recruitment period (<2%).
The selection of cases took place over a full calendar year to cover seasonal variations in the incidence and etiology of pneumonia, with the same number of cases (12 or 13) being selected each week. …