tspot.interp: Interpret the results of the Oxford Immunotec TSPOT.TB assay...

In mmparker/tbdiag: Functions for Tuberculosis Diagnostics Research

Description

Given vectors of nil, TB antigen (panels A and B), and mitogen results in spots, this function computes TSPOT qualitative interpretations. The function uses the Oxford Immunotec North America criterion by default; alternative criteria sets can be created as methods for the tspots.criteria function

Usage

tspot.interp(nil, panel.a, panel.b, mito, criteria = "oxford.usa",
  verbosity = "terse", ...)

Arguments

nil	A vector of nil results (in spots)
panel.a	A vector of Panel A TB antigen (ESAT-6) results (in spots)
panel.b	A vector of Panel B TB antigen (CFP10) results (in spots)
mito	A vector of mitogen results (in spots)
criteria	The name of the desired result criteria (defaults to the Oxford Immunotec criteria for North America).
verbosity	The level of verbosity ("onechar", "terse", "verbose") of the output.
...	Other arguments passed to the crtieria evaluation function chosen by the "criteria" argument.

Details

All spot counts greater than 20 are automatically censored to 20 for the purposes of calculating qualitative results, following Oxford's interpretation instructions.

Value

The function returns a vector of qualitative results. The verbosity of results depends on the argument passed to "verbosity":

onechar	Returns a single character indicating the result (N for Negative, B for Borderline, P for Positive, I for Indeterminate).
terse	Returns a single word indicating the result (Negative, Borderline, Positive, Indeterminate).
verbose	Returns the same results as "terse", with the addition of a short comment indicating the reason for an "Indeterminate" result.

Multiple criteria sets are available. The function defaults to the standard Oxford North American criteria (criteria = "oxford.usa"), but other currently available options include:

criteria = "oxford.global"	The Oxford global criteria, for which the criterion for positivity is lowered from an 8-spot difference between the antigen and nil panels and which does not include the borderline qualitative result;
criteria = "10spot"	A criteria set in which the borderline result is extended to include differences of 5 to 9 spots and only differences of 10 or more spots indicate a positive result.

Note

This function is provided purely as a convenience and is not a replacement for manual interpretation, manufacturer-provided software, or common sense. Absolutely not for clinical use.

References

Oxford Immunotec <http://www.oxfordimmunotec.com/>

See Also

qft.interp for Quantiferon interpretation.

Examples

# Calculate results
test.tspots$result.check <- with(test.tspots,
                                 tspot.interp(nil = nil,
                                              panel.a = panel.a,
                                              panel.b = panel.b,
                                              mito = mito))

# Compare lab and calculated results
with(test.tspots, table(lab.result, result.check, exclude = NULL))

# Compare different levels of verbosity
test.tspots$verbose.check <- with(test.tspots,
                                  tspot.interp(nil = nil,
                                               panel.a = panel.a,
                                               panel.b = panel.b,
                                               mito = mito,
                                               verbosity = "verbose"))

test.tspots$onechar.check <- with(test.tspots,
                                  tspot.interp(nil = nil,
                                               panel.a = panel.a,
                                               panel.b = panel.b,
                                               mito = mito,
                                               verbosity = "onechar"))

unique(test.tspots[ , c("lab.result", "result.check",
                      "verbose.check", "onechar.check")])

mmparker/tbdiag documentation built on May 23, 2019, 5:05 a.m.