Background: Some clinicians link chronic disease in certain patients to 'food intolerance'. This is currently diagnosed by exclusion dieting, a time-consuming and tedious technique. It has been claimed that IgE/IgG4 antibody testing is a rapid and valid method of determining food intolerance.
Objective: To determine the test/retest reliability of IgE/IgG4 antibody testing as a diagnostic tool.
Methods: Blinded testing of duplicate blood samples from nine patients with suspected food intolerance was undertaken by tertiary referral centre using the services of a commercial laboratory. The proportions of consistent and inconsistent results for tests of 95 different foods were analysed.
Results: Test/retest reliability was low. Even though the study method systematically overestimated kappa, this value never exceeded 0.51, regardless of the statistical model used. All but one patient had a greater number of inconsistent results than had been prespecified as an unacceptable level of disagreement. In one case, 50 out of 95 test results were inconsistent on retest.
Conclusions: We found no evidence that IgE/IgG4 antibody testing as performed by this laboratory is a reliable diagnostic tool.