Bad Science on Lyme in the NYT (Warning: Semi-Personal Post)

Print
Monday, 12 August 2013 12:40

Sorry folks, I usually restrict this blog to economic issues, but I am going to stray a little bit here to beat up the NYT over its Room for Debate on Lyme disease. (My wife has Lyme disease.)

Three of the participants in the debate assert that the research shows long-term antibiotic treatment is ineffective for treating people who supposedly suffer from chronic Lyme. The argument is that Lyme is an acute illness that can be effectively treated with 2-3 weeks of antibiotics. In this view, people who continue to experience symptoms after treatment either suffer from some other ailment or are hypochondriacs.

However the claim that the research shows long-term treatment is ineffective is not accurate. Allison DeLong, a statistician at Brown University, reviewed the three most often cited studies that claim to find long-term antibiotic treatment is ineffective. Her analysis showed that one of the studies had insufficient power to reach any conclusion about the effectiveness of treatment.

A second study showed that, while they were being treated, patients were significantly healthier than patients in the control group. This result has been ignored because the study also found that patients relapsed after the end of treatment. In other words, the study concluded that because 3 months of treatment did not cure patients (some of whom had already had years of antibiotic treatment), that treatment was ineffective.

The third study in fact did find that treatment led to a statistically significant improvement in patients' health according to the main measure the researchers had chosen (a measure of fatigue). However they opted to ignore this finding because the measure was subjective. The researchers also were confused about their own findings, wrongly believing that the double-blind nature of the study had been compromised even though the treatment and control group gave nearly identical answers when asked whether they thought they were being treated.

Given the importance of DeLong's findings to the Lyme debate it would have been appropriate to include her views in this exchange or to at least find an expert who was familiar with her research. It is a serious disservice to have an exchange on Lyme that does not include any mention of the latest research on the topic.