The positive predictive value of Borrelia serology in light of symptoms

Full article: http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=20461540&retmode=ref&cmd=prlinks

Excerpt:

OBJECTIVE: By using the published incidence of Lyme borreliosis
in endemic regions of the World, and the sensitivity and
specificity data of the best Lyme serological tests, we computed
the positive predictive value of Borrelia burgdorferi antibody
testing. METHODS: The calculation of predictive value was based
on Bayes’ theorem. We also analyzed the frequency distribution of
the specific and non-specific symptoms and complaints of 27,194
patients sent to the Centre for Tick-borne Diseases in Budapest
from 1986 to 2008. RESULTS: This evaluation demonstrated that
practitioners often use Lyme serology in a “trial and error” way,
without any reasonable ground. According to our calculation the
positive predictive value of the best Lyme antibody tests if
applied in this way is <9.1%. CONCLUSION: Our study suggests that
the present practice of applying Lyme serological tests may
result in more harm than benefit.