by David Kopel
May 6, 2001, Rocky Mountain News
The award for the most biased "news" story to appear in a Denver daily this spring goes to The Denver Post's front page lead from April 24, "Colorado slips further in getting tots vaccinated." The story consisted of one-sided, fear-mongering promotion of a controversial bill in the legislature.
The "news" in the story was that Colorado had allegedly fallen from 39th to 42nd in the number of 2-year-olds who have all recommended vaccinations. The claim was made by an official from the Colorado Department of Public Health and the Environment.
Things aren't nearly as alarming as the state employee and the Post claimed. Immunization rates are based on polling conducted by the U.S. government's Centers for Disease Control and Prevention (CDCP) in the National Immunization Survey. Because polls only measure a tiny fraction of the total population, pollsters always report a "confidence interval." As the CDCP explains, "Where confidence intervals overlap, the observed difference might be due to chance."
Regarding Colorado's immunization rate for 24-month-olds for all vaccines, the confidence intervals overlap with the rates of every other state except Connecticut. So we're statistically tied with 49states. (See www.cdc.gov/rip/coverage/tables/99-00/24months-iap.xls .)You'll need a program that can read Excel spreadsheets.) A look at every subset of vaccines in the National Immunization Survey likewise shows Colorado to be statistically indistinguishable from most other states.
Colorado's estimated immunization rate was 70 percent; the national estimated rate was 71 percent. Again, the confidence intervals overlap, so Colorado's rate is essentially the same as the national rate.
The Post article blamed Colorado's supposedly low immunization rate on the fact that Colorado doesn't have a government-controlled "vaccine database." But for 2-year-olds (the subject of the Post article), the state already has a database - pursuant to the 1992 Infant Immunization Act. (See Colorado statute 24-5-1705.)
The Post article ran on the Monday of a week when the legislature was considering whether to expand the infant database so that it covers everyone in the state. On Wednesday, the Post ran a picture of two senators holding up the Post article as they promoted the expanded database.
In support of the database, the Post quoted, at length, the same Colorado official who made the frightening but dubious immunization rate claim, and also quoted the president of a group of Colorado pediatricians. Rather than interviewing a single critic of the database, the Post relied solely on the government official to describe the opponents' concerns, and to claim that they were groundless.
The government official said that opponents saw the database as "Big Brother-like." With no more detail supplied, the ordinary reader might wonder "Who cares if the government has a list of everyone who got a diphtheria vaccine?"
But if the reporter had reported the actual contents of the bill (Senate Bill 61), the privacy concerns might have been a lot more understandable. The bill would have expanded the database to cover vaccination information for everyone in Colorado, and this would include adults, since some vaccines, such as hepatitis A and flu, are intended for adults. (I should point out that the bill died in a House committee, but was revived last week within another bill; it could be amended or killed again by the time this column is published.)
Under current law, the infant vaccination database is allowed to collect "epidemiological information" without an individual's permission. In current usage, "epidemiological information" includes any piece of information that might be correlated with group or individual health - such as whether the child or anyone in his family owns guns, smokes, drinks alcohol (and how much), or has multiple sexual partners. Senate Bill 61 allowed the collection of "epidemiological information" from students, schools, and many other sources. The government would be allowed to retain the records permanently.
A balanced story would have included quotes from a database opponent explaining specifically what the perceived problem was, and a specific response from a database advocate - perhaps a statement that even though the database legally authorizes collection of wide-ranging personal information, the government doesn't currently intend to collect such information. Then readers could make up their own minds.
Contrary to what the Post claimed in an April 25 editorial, the legislation included no provision allowing an individual to remove his record from the state's database. The opt-out provision (proposed statute 25-4-1705(e)V) applied only for vaccine information itself, not for lifestyle-related "epidemiological information."
When my Independence Institute colleague Linda Gorman (who has written a study criticizing the database proposal) called the Post reporter, the reporter claimed that the focus of the article was vaccination rates, and readers could study other articles for the vaccine bill's details. Yet of the article's 18 paragraphs, 7 dealt with the vaccine database. The Post's coverage of the bill during the rest of the week consisted only of short legislative briefs, which continued to inaccurately describe the bill as pertaining only to vaccine data.
The Post editorial blasted Colorado State Rep. Mark Paschall for killing the database bill two days before witnesses were scheduled to testify, because Paschall "precluded public discourse on the bill." The same criticism might be leveled at the Post, because the paper failed to inform readers about both sides of the issue.
The Rocky Mountain News, meanwhile, covered in detail the legislative maneuvering on the bill, but reported scantily on the bill's contents.