A quality assurance program for the measurement of capillary blood cholesterol levels in private pediatric practices. The Children's Health Project.

Objective: To develop an easy to use quality assurance program for the measurement of capillary blood cholesterol levels in private pediatric practices. The program needed to comply with the guidelines laid down by the National Cholesterol Education Program.

Design: Intervention study.

Setting: Nine private pediatric practices in and around northern Philadelphia, Pa.

Participants: The analysts included clinic staff members with laboratory expertise ranging from none to some previous experience. None of the participants had previous experience with a quality assurance program.

Interventions: Progress was reported monthly to the Lipid Research Laboratory, Philadelphia, and action was taken to correct inaccuracies in bias or variance.

Main Outcome Measure: Compliance with the analytical guidelines laid down by the National Cholesterol Education Program in that the coefficient of variation was no greater than 5% and the bias was no greater than +/- 5% in the first year of the study.

Results: Within the first year of the study, there were 152 monthly quality assurance returns for each of two lyophilized control materials. On four occasions the coefficient of variation was greater than 5% while the overall bias was within the desired +/- 5% on 143 (94%) of 152 occasions. After the first 3 months of the study, as user confidence increased, intervention by the Lipid Research Laboratory became minimal. The internal quality assurance was further evaluated by a successful performance in a quarterly external quality assurance program.

Conclusions: It is possible to devise an easy to use quality assurance program for extra laboratory measurement of cholesterol levels in children, and, with minimal assistance, maintain acceptable standards of cholesterol analysis. The quality assurance improved following the first 3 months of training and education. Subsequent continuous quality improvement was maintained with minimal involvement of the specialist center. Should the controversial issue of private office measurement of blood cholesterol levels become universally acceptable, the implication from our study is that standards acceptable to the National Cholesterol Education Program and the Clinical Laboratory Improvement Amendments of 1988 are possible using a suitable quality assurance program.