Near-patient testing. Quality of laboratory test results obtained by non-technical personnel in a decentralized setting.

The authors evaluated the quality and reliability of four desktop analyzers in the outpatient clinic. Twenty-seven nontechnologists (NTs) participated in the study. These included nurses, physicians, and medical students. The instruments and tests evaluated were as follows: Reflotron (glucose, cholesterol, triglycerides, gamma-glutamyltransferase and urea); Seralyzer (creatinine, glucose, potassium, aspartate aminotransferase, and hemoglobin); Vision (glucose, urea, cholesterol, triglycerides, alkaline phosphatase, and uric acid); and DT60 (sodium, potassium, glucose, amylase, uric acid, bilirubin, and creatinine). For precision studies, low and high control material was used, and method comparison was done with methods in routine use in the laboratory. The range of coefficients of variation (CVs) for the analyzers with NTs was as follows: Reflotron: CV, 2.4-7.9%; Seralyzer CV, 1.4-18.7%; Vision: CV, 1.5-2.7%; DT60: CV, 2.5-46.8. The percentage results that is different by greater than 10% between the NTs and trained technologists was related to the complexicity of procedure for each analyzer and was the lowest for the Vision analyzer and greatest for the Seralyzer.