PCR specificity 6 November 2003
Previous Rapid Response Next Rapid Response Top
Christopher J Noble,
postdoctoral fellow
Bern Switzerland

Send response to journal:
Re: PCR specificity

Eleni Papadopulos-Eleopulos writes: "Would Christopher Noble also please explain the 0% - 100% specificity of the PCR test?"

I have already explained how Eleni is attempting to mislead the readers of this forum. She cites a meta-study of the use of PCR in the diagnosis of HIV infection. (1) A variation of between 40-100% was seen in the specifity of PCR between various studies. What Eleni does not tell you is that most of the studies showed close to 100% specificity (see figure 1). A SMALL number of studies found a poorer specificity.

It is not at all surprising that some studies had poor results. PCR is a complicated procedure and is prone to a number of technical problems. The metastudy however showed that PCR can be highly specific. Importantly the best studies in the meta-analysis (blinding, controls, etc.) showed high specificity.

"In studies in which the design was rated as either 3 or 4, sensitivity ranged from 83% to 100%, and specificity ranged from 95% to 100%"

The real question is how is this possible? How does this amazing coincidental correlation between the detection of HIV RNA and anti-HIV antibodies occur? Eleni would like people to believe that both of these measurements are non-specific yet they correlate so well. Eleni would like people to believe that her theory predicts all the experimental phenomena associated with AIDS. Does her theory predict this result?

I have previously asked Eleni to comment on this but she has remained silent.

(1) Owens DK, Holodniy M, Garber AM, Scott J, Sonnad S, Moses L, Kinosian B, Schwartz JS. Polymerase chain reaction for the diagnosis of HIV infection in adults. A meta-analysis with recommendations for clinical practice and study design. (1996) Annals of Internal Medicine 124:803-15.

Competing interests: None declared