The Body: The Complete HIV/AIDS Resource
Follow Us Follow Us on Facebook Follow Us on Twitter Download Our App 
Professionals >> Visit The Body PROThe Body en Espanol

HIV Transmission and Education >> Am I Infected?

Pages: 1

Roche claims to be 99.997% accurate???
      #21101 - 08/20/01 03:41 PM

I've seen several posters say that Roche says "PCR DNA to be 99.997% accurate". Really? I don't
think so...can you give me a number to call or a reference to this? If you think that calling
800-584-8183 is Roche, you're incorrect and if they told you Roche told them that, call Roche and
find out for yourself. I don't think that is correct. The PCR DNA test is around 92-98% sensitive in
HIV infected specimens...if anyone has any other information, PLEASE post it...I don't think
Specialized Testing should be telling people the the PCR DNA test is 99.997% accurate when it's not.
I think their counselors might be getting the RNA test mixed up with the DNA test. If I'm wrong I'd
love to know.


Post Extras: Remind Me!   Notify Moderator  

Re: Roche claims to be 99.997% accurate??? new
      #21118 - 08/20/01 06:35 PM

where exactly do you get your info from...i would belive the people who are giving the tests and not some lit from a doctor or the cdc who has an idiot reading info from 1987... i specifically called specialized testing and they assured me that this is their business and quest is thew best in the field and the results were 99.99% accurate...

Post Extras: Remind Me!   Notify Moderator  

Re: Roche claims to be 99.997% accurate??? new
      #21123 - 08/20/01 06:55 PM

The following is directly taken from the Johns Hopkins Medical Management of HIV under HIV Serologic Test section.

The link is

Viral Detection: Other methods to establish HIV infection include techniques to detect HIV antigen, DNA, or RNA (Table 2-1). HIV-1 DNA PCR is the most sensitive and can detect 1-10 copies of HIV proviral DNA. None of these tests is considered superior to routine serology in terms of accuracy, but some may be useful in patients with confusing serologic test results, when there is a need to clarify indeterminate test results, for virologic monitoring in therapeutic trials, and for HIV detection when routine serologic tests are likely to be misleading such as in patients with agammaglobulinemia, acute retroviral infection, neonatal HIV infection, and patients in the window following viral exposure. In most cases, confirmation of positive serology is accomplished simply by repeat serology. The sensitivity of tests for detection of HIV varies with the stage of disease and test technique, but is usually reported at >99% for DNA-PCR, 90% to 95% for quantitative HIV-RNA, 95% to 100% for viral culture of peripheral blood mononuclear cells (PBMC), and 8% to 32% for p24 antigen detection (J Clin Microbiol 1993;31:2557; N Engl J Med 1989;321:1621; J AIDS 1990;3:1059; J Infect Dis 1994;170:553; Ann Intern Med 1996;124:803). None of these tests should replace serology to circumvent the informed consent process.

Table 2-1: Tests for HIV-1
Assay Sensitivity Comments
serology 99.7% Readily available and inexpensive. Sensitivity and specificity >99.9% (MMWR 1990;39:380; N Engl J Med 1988;319:961; JAMA 1991;266:2861).
Rapid test
SUDS [Murex
Diagnostics, Norcross, Ga.] 99.9% Results are available in <10 minutes but test must be performed by a lab technician. Specificity is 99.6%; positive tests should be confirmed. Highly sensitive; negative tests do not usually require confirmation. Other rapid tests are available but are not FDA approved (Int J STD AIDS 1997;8:192; Vox Sang 1997;72:11).
Salivary test
(OraSure Test System) 99.9% Salivary collection device to collect lgG for EIA and WB. Advantage is avoidance of phlebotomy. Sensitivity and specificity are comparable to standard serology (JAMA 1997;227:254).
Urine test (Calypte 1) >99.9% Used for EIA test only, so positive results must be verified by serology. Must be administered by a physician. Cost is low - about $4/test.
PBMC culture 95% to 100% Viral isolation by co-cultivation of patient's PBMC with phytohemagglutinin (PHA)-stimulated donor PBMC with IL-2 over 28 days. Expensive and labor-intensive. May be qualitative or quantitative. Main use of qualitative technique is viral isolation for further analysis and for HIV detection in infants. Quantitative results correlate with stage: Mean titer is 20/106 cells in asymptomatic patients and 2,200/106 cells in patients with AIDS (N Engl J Med 1989;321:1621).
DNA PCR assay >99% Qualitative DNA PCR is used to detect cell- associated proviral DNA; primers are commercially available from Roche Laboratories. Sensitivity is >99% and and specificity is 98%. This is not considered sufficiently accurate for diagnosis without confimation and is not FDA-approved (Ann Intern Med 1996;124:803). Main use is for viral detection in the case of neonatal HIVand disputed or indeterminate serologic tests.

HIV RNA PCR 95% to 98% False-positive tests in 2% to 9%; usually at low titer. Sensitivity depends on viral load (VL) threshold of assay and assumes no antiretroviral therapy.

Post Extras: Remind Me!   Notify Moderator  

The above info is for Shane new
      #21124 - 08/20/01 06:57 PM

I don't want to start a debate seeing as you have done quite a bit for this board but I thought it best that you have the correct statistics for the correct test.


Post Extras: Remind Me!   Notify Moderator  

Re: Roche claims to be 99.997% accurate??? new
      #21126 - 08/20/01 07:04 PM


I too got a test from Specialized Testing and it too was "not detected"...what I had a problem with is them
telling me that no further testing was necessary. I believed them...but as a matter of doing what I needed
to do to be as sure as I could, I called them back and asked for some references...I was passed to a
"manager" I forgot her name, but you can do the same...and she told me that "NO, Roche does not make
those claims, as the gentleman who worked for her had told me". Where they got their info was from the
same references you can get off the net. I asked her about Quest and she was NOT aware of any claims
made by them either. This did not satisfy me as I thought that maybe Quest (especially at Nichols
Institute where they do the test) or Roche could shed more light. I'm very good on the phone, and in my
career as an investment banker I've made my living getting through to people who typically can't be
gotten to. Anyway, I got to talk to a LAB supervisor, who was also an MD in Baltimore MD where they
also have the contract for the NIH to do PCR testing and he told me that they MAKE NO CLAIMS to
the sensitivity of the test...he did go on to say that does NOT MEAN that the test is invalid, but they
themselves cannot say for certain that the test is 99.9 anything. When I pressed him, he said just what Dr,
Holodniy says and that is, the PCR DNA test is highly sensitive and in "most" folks who are infected it
should show "detected"...but by NO MEANS is it fool proof...if you've really been exposed to HIV you
still should be tested at 3 months and 6 months via an ELISA. I was exposed to HIV through a freak
accident at a friends house who as AIDS.

I also called Roche and was told the same thing.

The problem I have with Specialized testing is that they are "lying" by telling people that they "don't need
to be re-tested and that it's "conclusive", when ALL the experts (and NOT some $10 hour paid phone
consultants) say that it isn't conclusive. Why do they say this...I think it's because they can get all the
"worried well" to pay for the test...if they told you the truth, MANY would not shell out the $$ for the test
if they knew they had to confirm it again...

I'm not saying that the PCR is NOT VALID...what I'm saying is that a NOT DETECTED result is
GREAT and it means that you're probably not infected...but it also means that you still need to get tested
via an ELISA at the correct times to be 99.9% sure. If it is only 99% sensitive...10 people out of 1000 will
have false negatives...meaning Specialized testing will send 10 people out who will believe them, only to
infect others, like their wives and lovers...I wouldn't want that on my conscious.

If you find any new info or if you can provide any legit references to the contrary, PLEASE POST
THEM. I'm waiting to go get my 6 month test (passed on the 3 month as I didn;t want the added stress)
and if you can provide any info to what they say is true, it would sure ease my mind...;)


Post Extras: Remind Me!   Notify Moderator  
Pages: 1

What's New at

Additional Information
0 registered and 1 anonymous users are browsing this forum.

Moderator:  TheBody, bogart, gigi, riverprincess 

      You cannot start new topics
      You cannot reply to topics
      HTML is enabled
      UBBCode is enabled

Thread views: 3132

Jump to

Contact Us | Privacy Statement The Body

UBB.threads™ 6.2.3