Hiding an epidemic: Our broken Lyme reporting system

lyme-cases-82-08-small4.jpg One thing I’ve always found puzzling is why the Centers for Disease Control (CDC) continues to disseminate misleading data from a Lyme disease tracking system that undercounts human cases by a factor of ten. (Click here to view or download PDF version of above graph showing the magnitude of this miscount.) For example, in the article, “Bad Lyme Bug Spreading,” CDC medical epidemiologist Kevin Griffith, MD, MPH said:

“Although the 20,000 cases reported to the CDC in 2006 were fewer than the 23,000 cases reported in 2005, Griffith says the true number of cases is probably larger.”

While this statement is technically factual, it cites two-year old data and doesn’t present the information that is relevant to health care providers, state health departments and people at risk for Lyme disease. The CDC needs to communicate the true risks of getting Lyme disease in each state and how many people are suffering from the disease, not some internal tracking numbers used by a few CDC epidemiologists. A more accurate, honest statement would be:

“In 2007 there were 27,444 reported cases of Lyme disease, a 37% increase over the previous year. It’s important to note that the actual number of new Lyme cases in the U.S. could possibly exceed 300,000 per year due to inherent flaws in the Centers for Disease Control’s passive case reporting system.”

The U.S.’s Lyme reporting system is complex, labor-intensive, and overly reliant on an expensive, two-tiered testing protocol that misses about 75% of the positive cases. (In contrast, the AIDS screening test only misses about 5% of the positive cases.) Many physicians are afraid to report Lyme cases, after the highly publicized medical board hearings of Lyme specialists like Dr. Charles Ray Jones and Dr. Joseph Jemsek. What’s more, the double-digit growth of the Lyme epidemic and cutbacks in state health department services is driving this reporting system to its breaking point. The overworked health department in New Jersey quantified the absurdity of the Lyme surveillance process in their report, “Burden of Lyme Reporting” (2001-2006). In this study, they found that: --Each Lyme case took an average of 2 months to close --24% of Lyme cases were never closed or counted --The Lyme reporting process in New Jersey alone required 5 full-time employees to handle Unfortunately, when the CDC and the Council of State and Territorial Epidemiologists (CSTE) were presented with this report in 2007, they revised the surveillance process in a way that will further hide the true size of the Lyme epidemic. Rather than fixing the root cause of the reporting inaccuracy -- bad tests -- they added complexity to the process, creating three distinct categories of Lyme cases ("confirmed", "probable", and "suspected") for health surveillance workers to sort out. I suspect that these busy employees will throw time-consuming cases into the “probable” and “suspected” buckets, and based on past CDC behavior of sweeping bad news under the rug, those cases will never see the light of day, creating the illusion of a drop in total US Lyme cases. The tragedy of this broken reporting system is that it is ruining lives. Physicians make clinical decisions based on the CDC case numbers, and if they think that Lyme is rare in their area, they may not consider it as a differential diagnosis. Congress allocates research funding based on case numbers, too, and by underestimating these numbers by a factor of ten, grants will go to other diseases. And the band will play on. It’s time for the CDC and CSTE to stop downplaying the magnitude of the Lyme epidemic and to put the truth back into the reporting system. Click here to view/download PDF version of graph "Lyme Cases 1982-2008"