The Home Front

Flu Vaccine: Yes or No?

I’m a strong believer in vaccines, but I just can’t get a true sense on the flu vaccine and if it’s necessary for my kids every single year. One of the things that bothers me the most is that major media reports don’t do a very good job of reporting the actual facts on the studies behind the vaccine’s efficacy. For example, here’s CNN from January 3:

The exact number of flu-related adult deaths is hard to track and varies from year to year. The CDC has estimated that from 1976 through 2007, between 3,000 and 49,000 people died of flu-related causes.

“It depends on the season; it depends on the virus,” Jhung said.

Last year, 381,000 people were hospitalized and 171 children died in what’s being called a relatively severe season.

However, the CDC estimates that flu vaccination prevented 6.6 million illnesses last year, 3.2 million doctor visits and at least 79,000 hospitalizations.

Flu vaccines are recommended for everyone 6 months and older, especially pregnant women and those at high risk of complications, including the elderly, children younger than 5 years and those with underlying medical conditions such as asthma or diabetes.

Antiviral medications are a good treatment if you do get sick, Jhung said, particularly those at high risk for complications. Ideally, antivirals should be started within two days of when symptoms appear.

CNN never links to the actual report from the CDC, however. I did some digging, and you can read the CDC report for yourself here

CNN gives a good summary of the CDC’s findings, but doesn’t mention a single one of the CDC’s self-reported “limitations” with their own work. Here’s what the CDC wrote (emphasis mine):

The findings in this report are subject to at least six limitations. First, influenza vaccination coverage rates were derived from vaccination status reported by survey respondents, not vaccination records, and are subject to recall bias. Second, these rates are based on telephone surveys with relatively low response rates; therefore, selection bias might remain after weighting adjustments. Third, these surveys only cover the noninstitutionalized population. Fourth, estimates of the number of persons vaccinated based on these survey data exceeded the actual number of doses distributed, indicating coverage estimates used in this report overestimate averted illness resulting from vaccination (5). Fifth, the model only calculates outcomes directly averted by vaccination. If there is indirect protection from decreased exposure among unvaccinated persons in a partially vaccinated population (i.e., herd immunity), the model would underestimate the number of prevented illnesses. Also, although the impact of vaccination in preventing severe outcomes is most pronounced among persons aged ≥65 years, if vaccine effectiveness were lower among frail elderly persons, the model might have overestimated the effect in this group. Finally, adjustments for underreporting of influenza hospitalizations were based on studies conducted in 2009–10, as were the extrapolation of hospitalization rates to estimate rates of illness and medically attended illness. Because multipliers were calculated during a pandemic, if the ratio of hospitalizations to other outcomes or the underreporting of hospitalization rates were different in 2012–13 (e.g., through changes in health-seeking behaviors or testing practices), the model might have underestimated or overestimated the effect of vaccination.

These seem like pretty damning limitations to me. Why can’t we get better data on flu vaccines in general?