Student Finds Flaws In Federal Flu Data

In its reports on flu and flu mortality, the Centers for Disease Control and Prevention (CDC) had inconsistencies in data and misrepresentations to the public, said Peter N. Doshi, a student in Harvard’s Graduate School of Arts and Sciences (GSAS) who is writing his dissertation in Japanese medical history. His criticism appeared in a Dec. 10 article published in the British Medical Journal.

Doshi’s article raised three main points of contention: an alleged misrepresentation on the CDC website, alleged inconsistencies in the CDC data, and a dubious causality between flu and deaths caused by pneumonia.

Doshi told The Crimson that his interest in the topic sparked after the CDC reported an increase in the average annual death rate, in 2003, from 20,000 to 36,000.

According to Doshi, the CDC made a strong connection between pneumonia-related deaths and the flu.

The CDC ascribed 8,000 of the 36,000 annual flu-attributed deaths to pneumonia stemming from flu patients’ weakened states.

Doshi disagreed with this connection, saying that common annoyances like “stress and colds” can leave a person vulnerable to pneumonia.

According to the CDC’s website, “Every year in the United States, on average: about 36,000 people die from flu.”

Doshi said this claim was misleading, as the people who died did not necessarily die of the flu but, rather, of a flu-related complication.

Director of Media Relations at the CDC, Glen Nowak, said that the organization “disagrees with [Doshi’s] position for various reasons.”

“His assessment fails to recognize that the data presents a range,” Nowak said.

He pointed out that flu deaths ranged from 17,000 to 51,200, and the number 36,000 is only an average.

Nowak also said that Doshi underscored the 12-percent increase in the 65-and-older population as too little to account for the 2003 spike in deaths. These people are much more susceptible to flu than the younger population. Doshi said that, while the number of people over the age of 65 may have been increasing, the number of flu deaths decreased by 30 percent, according to the CDC’s National Center for Health Statistics (NCHS).

Nowak, however, maintains that this increase in population accounts was one of the reasons for the 2003 increase.

John Abramson, a clinical instructor at the Harvard Medical School who provided some guidance throughout Doshi’s research, said the “public information about influenza is driving people towards remedies [like flu vaccines] that may not be very effective.”

David S. Rosenthal, director of University Health Services, said that it did not matter what the final cause of death was as long as flu, often a secondary cause of death, could be prevented.

“We want to get people used to getting vaccinated,” he said. “We’re due for a pandemic flu.”