RESUMEN
We have studied 219353 individual clinical chemistry results obtained in methods comparison studies. Each result was prospectively compared with its replicate, comparative, or repeat value to identify differences from expected values. Unacceptable results were defined as differing from the expected values by < or = 7 SDs or CVs. We believe these differences represent special-cause variation and should be expressed as unacceptable rates per million results (ppm). We observed 447 ppm unacceptables: 196 ppm in control samples and 251 ppm in patients' samples. Results judged likely to alter patient care occurred at a rate of 41 ppm. To better understand the magnitude of these rates, we compared these results with reports of error rates in HIV testing and the airline industry. The measurements reported were made for the purpose of quality improvement, not judgment or discovery. The significance of these findings for laboratorians, manufacturers, and regulators is discussed.
Asunto(s)
Química Clínica/normas , Laboratorios/normas , Errores Médicos , Química Clínica/estadística & datos numéricos , Humanos , Laboratorios/estadística & datos numéricos , Errores Médicos/estadística & datos numéricos , Planificación de Atención al Paciente , Control de Calidad , Reproducibilidad de los ResultadosRESUMEN
Forty-three consecutive cases from a community hospital with concomitant bone marrow iron stain, serum ferritin, and erythrocyte sedimentation rate (ESR) were reviewed. Cases were classified as iron present or absent by the bone marrow iron stain. A two-dimensional linear graphic relationship between ferritin and ESR correctly identified six of nine iron-deficient patients and 32 of 34 iron-present patients. Four cases yielded indeterminate results. One complex iron-deficient case was incorrectly classified. This graphic method developed with data from tertiary care patients was correct in 88.4% of cases, incorrect in 2.3%, and indeterminate in 9.3%. When absent iron stores were graphically predicted, the predictive value was 100%. When iron deficiency was graphically excluded, the predictive value was 97%. The authors conclude the graphic method is useful in a community hospital practice for the confirmation or exclusion of iron deficiency.
Asunto(s)
Anemia/metabolismo , Sedimentación Sanguínea , Médula Ósea/metabolismo , Ferritinas/sangre , Hierro/metabolismo , Anemia/sangre , Anemia/diagnóstico , Anemia Hipocrómica/diagnóstico , Diagnóstico Diferencial , Predicción , Hospitales Comunitarios , Humanos , Estudios RetrospectivosRESUMEN
Insurers, employers, and individuals create demands for laboratory testing in "wellness programs." Tests chosen to identify cases deserving intervention included routine automated chemical tests plus high-density lipoprotein cholesterol, ferritin, and thyroid tests. Participants' unwarranted concerns were addressed with a personalized reporting schema. We tested 1338 individuals, identified 224 (16.7%) with significant abnormalities, and made phone contact follow-up with 193 (86%) of these six to 14 months later. Cholesterol results suggesting increased risk of heart disease were frequent, and were not studied. Interventions were initiated in 55 of the 193 followup cases (49 by physician and six by participants), including prescription of iron or thyroid hormone, counseling on dietary or alcohol intake, and repeat testing. For 58, there was medical advice without intervention; abnormal results were ignored by 79. Noteworthy participant anxiety was manifested in two of the 193 cases, both of whom were treated with iron. We conclude that 4% of the original 1338 participants potentially benefitted from intervention. Ferritin and thyroid tests initiated 33 (61%) of these 55 specific therapeutic interventions.