Reproducibility of clinical data and decisions in the management of upper respiratory illnesses: a comparison of physicians and non-physician providers.

The ability of non-physician providers to collect the data required by an algorithm for upper respiratory illness management, and the appropriateness of resulting key management decisions, were studied by comparing non-physician data and management decisions on 426 patients with those of internists. The internists, blinded to Amosists' findings and plans, evaluated the same patients and indicated management without using the algorithm (AM-MD) study). To control for variability of internists' data collecting and illness management, 171 additional patients were evaluated and managed consecutively by two internists, each also kept unaware of the other's findings and plans (MD-MD study). Overall AM-MD agreement on history and physical findings (90 per cent and 81 per cent) and on the need for tests (84 per cent) and treatment (87 per cent) was as high as MD-MD aggrement (91 per cent, 80 per cent, 88 per cent, and 75 per cent, respectively). In both studies, there was significantly more agreement on history data than on physical findings, evaluation, and therapy.