to the editor: We would like to express our concern about several issues raised in the Cochrane for Clinicians department entitled “Vaccines for Preventing Influenza in Healthy Children” that appeared in American Family Physician.¹ Family physicians should actively encourage parents to follow the recommendations of the Advisory Committee on Immunization Practices (ACIP) and the American Academy of Family Physicians for annual influenza vaccination of children six to 59 months of age and of older children with certain underlying medical conditions such as asthma and diabetes.²

The article characterizes “prevention of influenza-like illness” as a clinically more important measure of vaccination effectiveness than prevention of influenza. But influenza-like illness is a surrogate marker of effectiveness, measured only to reduce expense and time associated with measuring the more specific laboratory-confirmed infection. Influenza vaccine protects against the influenza virus, not the many other viruses that co-circulate during winter months and cause “influenza-like” illnesses. It is unreasonable to expect any vaccine to protect against infections other than those for which it was designed.

Vaccinated children six months and older usually acquire protective antibody levels against the specific influenza strains.^2,3 Studies have repeatedly provided evidence of vaccine effectiveness, even in years when the match between circulating and vaccine strains was suboptimal. For example, a study conducted by the Centers for Disease Control and Prevention and a large health maintenance organization demonstrated effectiveness of 25 to 49 percent in children six to 23 months of age who were vaccinated for influenza compared with unvaccinated children.⁴ This study was performed during the 2003–2004 influenza season, which was characterized by a suboptimal match between the predominant circulating influenza strain and the influenza vaccine strain.

Variation in study design prevented a Cochrane meta-analysis of safety outcomes, but methodologic inconsistencies should not put the safety of influenza vaccine in doubt. The billions of doses administered without incident to persons of all ages and in varying degrees of underlying health is evidence of influenza vaccine safety.

Vaccination recommendations for healthy young children were made because of serious morbidity associated with influenza infection in this group.² Infants and toddlers are hospitalized with influenza complications at rates similar to or higher than elderly persons. And it appears children may be hospitalized because of influenza more often than we realize. A recent study reported that influenza diagnosis was missed in more than 70 percent of hospitalized children.⁵

Although vaccinating children will help limit the impact of influenza in those vaccinated, widespread pediatric immunization may provide an additional benefit in terms of reduced transmission to others at high risk of influenza complications (e.g., the elderly) because infected young children shed higher amounts of influenza virus for a longer time than adults, making them effective transmitters of this infection.

As with every vaccine and medication, the influenza vaccine is not perfect. However, influenza is often a serious and potentially deadly disease, and the vaccine is the best tool we have to prevent the spread and associated morbidity and mortality of this disease. Although we may all agree that additional research would be helpful, we cannot neglect our ongoing duty to protect children now.

in reply: Drs. Baker, Schaffner, and Zimmerman state that “influenza-like illness is a surrogate marker of effectiveness, measured only to reduce expense and time associated with measuring the more specific laboratory-confirmed infection.” Actually, influenza-like illness is a more patient-oriented outcome, whereas influenza infection is a disease-oriented outcome. In other words, patients and their parents are concerned about avoiding illness, missing school or work, and being hospitalized. They are much less concerned about which virus caused their illness. The data from the Centers for Disease Control and Prevention cited by the letter authors actually showed no statistically significant reduction in the rate of influenza-like illness in vaccinated children six to 23 months of age compared with unvaccinated children (hazard ratio = 0.75; 95% confidence interval [CI], 0.56 to 1.00).¹ There was a statistically significant reduction in the rate of “pneumonia and influenza” (hazard ratio = 0.51; 95% CI, 0.29 to 0.91). However, the letter authors did not report rates of pneumonia caused by any organism or all-cause hospitalization, both of which would be more patient-oriented outcomes than “pneumonia and influenza.” Since the publication of the Cochrane for Clinicans,² researchers have produced more limited quality evidence that the influenza vaccine is safe at these young ages,³ but evidence of its effectiveness is still limited.