In the past few years, there has been an increase in morbidity from influenza in infants and young children. Because of this increase, in 2003 the Advisory Committee on Immunization Practices changed its recommendation for annual trivalent influenza vaccine to include all children six to 23 months of age. Despite this recommendation, there have been no population-based studies that evaluated the safety of the vaccine. In the published controlled studies using the vaccine, it is evaluated in terms of outcomes, such as the prevention of otitis media, or as a control to compare with the live attenuated intranasal influenza vaccine. There also are no postmarketing studies assessing the trivalent influenza vaccine, though in the past, studies of this type have been used to identify or rule out adverse events associated with other vaccines. France and associates evaluated the records of a large cohort of children who had received the trivalent influenza vaccine to determine whether there was an increased risk for medically attended events within two weeks of the vaccination.

The study design was a self-controlled screening analysis based on the case-crossover method, spanning from January 1, 1993, through December 31, 1999. The participants were children younger than 18 years who were enrolled in one of five managed care organizations for at least 28 days before and after immunization with the trivalent influenza vaccine. These organizations had extensive computerized records for their patients. Participants were divided at random into two equal groups: group 1 was evaluated for the risk of outpatient, emergency department, and inpatient visits within 14 days of receiving trivalent influenza vaccine compared with the risk for visits in two other control periods. Group 2 was evaluated for medically attended events that were considered significant and plausible in group 1, with the same control periods. Further analysis using medical records was carried out on medically attended events that were significant in both groups within 14 days of vaccination compared with the control periods.

There were 251,600 children involved in the study, distributed evenly by sex, with an average age of 10 years. Over the course of the study, participants received 438,167 trivalent influenza vaccinations. There were 1,165 distinct medically attended events in the outpatient setting, 230 in the emergency department, and 489 in the inpatient setting. Analysis of the data determined four diagnoses that could be associated with the vaccine: impetigo, dermatitis, uncomplicated diabetes, and unspecified ureteral disorders. After medical review, impetigo was the only medical event associated significantly with the vaccine, and this occurred in nine children.

The authors conclude that there is no evidence of important medically attended events resulting from the trivalent influenza vaccine in the study population during the two weeks after vaccination. They add that this study is useful for national policy groups, who will be looking at the efficacy, safety, cost-effectiveness, and feasibility of annual immunization in all children before expanding current recommendations. Further studies are needed to evaluate long-term effects of the vaccine.