January 15, 2010

Toast and marmalade

This popped up in the Google machine this morning;
A method for detecting whether an individual will progress to having active mycobacterial disease comprising determining whether the individual has a T cell response to one or more of the following mycobacterial antigens: CFP-10, Rv1989c, Rv3873, or Rv3878.

Initial reaction was along the lines of err arr right,  this looks nice - where's me marmalade.

Reading down the page I see that
The present inventors have shown that T cell responses to particular mycobacterial antigens act as prognostic markers that can be used to identify individuals at risk of progressing to active mycobacterial disease.
OK, so how do they do this invention thing?
Child contacts were clinically followed up every 6 months for 2 years at the clinic but asked to return immediately for further clinical assessment if they developed intercurrent symptoms. Diagnosis of active tuberculosis was made by the study paediatricians, taking into account symptoms, physical signs, and radiological and microbiological findings.
Right...so, overlooking the crumbs on the table, knife left in the jar and coffee getting cold just where the hell are we going with this?
Differences between the proportions of contacts responding to each particular antigen were compared using the McNemar's test for paired binary variables. The strength of T-cell responses to particular antigens amongst responders were compared using the Mann-Whitney test for continuous variables. Relative risk and 95% confidence intervals (95% CI), their significance being assessed by P-values calculated by the Chi squared test, were used to assess the prognostic value of the assay result for TST conversion, ESAT-6/CFP10 ELISpot conversion and progression to active TB disease. All analyses were undertaken in Graphpad Prism 4 for Windows (Version 4.03, GraphPad Software, Inc., CA, USA) and SPSS for Windows (Rel 13.0, SPSS Inc, Chicago, USA).
Sorry I asked - more toast anybody?

.