Frederick A Meier1, Rhona J Souers, Peter J Howanitz, Joseph A Tworek, Peter L Perrotta, Raouf E Nakhleh, Donald S Karcher, Christine Bashleben, Teresa P Darcy, Ron B Schifman, Bruce A Jones. 1. From the Department of Pathology and Laboratory Medicine, Henry Ford Health System, Detroit, Michigan (Drs Meier and Jones); the Departments of Biostatistics (Ms Souers) and Surveys (Ms Bashleben), College of American Pathologists, Northfield, Illinois; the Department of Pathology, State University of New York, Brooklyn (Dr Howanitz); the Department of Pathology, St Joseph Mercy Hospital, Ypsilanti, Michigan (Dr Tworek); the Department of Pathology, West Virginia University Health Sciences Center, Morgantown (Dr Perrotta); the Department of Pathology, Mayo Clinic, Jacksonville, Florida (Dr Nakhleh); George Washington University Medical Center, Washington, DC (Dr Karcher); Clinical Laboratories, University of Wisconsin Hospitals and Clinics, Madison (Dr Darcy); and Diagnostic Service Line, Southern Arizona Veterans Administration Health Care Systems, Tucson (Dr Schifman).
Abstract
CONTEXT: Many production systems employ standardized statistical monitors that measure defect rates and cycle times, as indices of performance quality. Clinical laboratory testing, a system that produces test results, is amenable to such monitoring. OBJECTIVE: To demonstrate patterns in clinical laboratory testing defect rates and cycle time using 7 College of American Pathologists Q-Tracks program monitors. DESIGN: Subscribers measured monthly rates of outpatient order-entry errors, identification band defects, and specimen rejections; median troponin order-to-report cycle times and rates of STAT test receipt-to-report turnaround time outliers; and critical values reporting event defects, and corrected reports. From these submissions Q-Tracks program staff produced quarterly and annual reports. These charted each subscriber's performance relative to other participating laboratories and aggregate and subgroup performance over time, dividing participants into best and median performers and performers with the most room to improve. Each monitor's patterns of change present percentile distributions of subscribers' performance in relation to monitoring durations and numbers of participating subscribers. Changes over time in defect frequencies and the cycle duration quantify effects on performance of monitor participation. RESULTS: All monitors showed significant decreases in defect rates as the 7 monitors ran variously for 6, 6, 7, 11, 12, 13, and 13 years. The most striking decreases occurred among performers who initially had the most room to improve and among subscribers who participated the longest. All 7 monitors registered significant improvement. Participation effects improved between 0.85% and 5.1% per quarter of participation. CONCLUSIONS: Using statistical quality measures, collecting data monthly, and receiving reports quarterly and yearly, subscribers to a comparative monitoring program documented significant decreases in defect rates and shortening of a cycle time for 6 to 13 years in all 7 ongoing clinical laboratory quality monitors.
CONTEXT: Many production systems employ standardized statistical monitors that measure defect rates and cycle times, as indices of performance quality. Clinical laboratory testing, a system that produces test results, is amenable to such monitoring. OBJECTIVE: To demonstrate patterns in clinical laboratory testing defect rates and cycle time using 7 College of American Pathologists Q-Tracks program monitors. DESIGN: Subscribers measured monthly rates of outpatient order-entry errors, identification band defects, and specimen rejections; median troponin order-to-report cycle times and rates of STAT test receipt-to-report turnaround time outliers; and critical values reporting event defects, and corrected reports. From these submissions Q-Tracks program staff produced quarterly and annual reports. These charted each subscriber's performance relative to other participating laboratories and aggregate and subgroup performance over time, dividing participants into best and median performers and performers with the most room to improve. Each monitor's patterns of change present percentile distributions of subscribers' performance in relation to monitoring durations and numbers of participating subscribers. Changes over time in defect frequencies and the cycle duration quantify effects on performance of monitor participation. RESULTS: All monitors showed significant decreases in defect rates as the 7 monitors ran variously for 6, 6, 7, 11, 12, 13, and 13 years. The most striking decreases occurred among performers who initially had the most room to improve and among subscribers who participated the longest. All 7 monitors registered significant improvement. Participation effects improved between 0.85% and 5.1% per quarter of participation. CONCLUSIONS: Using statistical quality measures, collecting data monthly, and receiving reports quarterly and yearly, subscribers to a comparative monitoring program documented significant decreases in defect rates and shortening of a cycle time for 6 to 13 years in all 7 ongoing clinical laboratory quality monitors.