In Reply: Dr Donta questions the methods, results, and conclusions of our study. The results of our earlier article1 are not inconsistent with those of our second study because the methods were different in several important ways. Contrary to Donta's assertion, the SF-36 was designed for and is valid with telephone interviews.2 One way of obtaining scientific information about the association between widely prevalent, nonspecific symptoms and prior disease is to compare the frequency of reports of such symptoms among patients who were diagnosed as having LD with reports of age-matched control subjects who do not have LD. Because the prevalence of LD is relatively low (even in Connecticut), it is very unlikely that enough control subjects were having problems from undiagnosed LD to distort the results; consequently, a control group from a nonendemic area is unnecessary. It is entirely appropriate to ask this sample of randomly-selected patients which symptoms they did and did not attribute to LD. Given the extraordinary publicity about LD, if there were bias, it should be in the direction of overreporting of symptoms among patients who had LD.
Seltzer EG, Shapiro ED, Gerber MA. Long-term Outcomes of Lyme Disease—Reply. JAMA. 2000;283(23):3069. doi:10.1001/jama.283.23.3068b