Five-year follow-up of routine outpatient test turnaround time: A College of American Pathologists Q-Probes study

Archives of Pathology & Laboratory Medicine, Nov 2003 by Valenstein, Paul, Walsh, Molly

Context.-Timely reporting of outpatient tests can increase efficiency of care and improve customer satisfaction.

Objectives.-We conducted a survey in 2002 to determine how quickly hospital-based laboratories turned around routine requests for 3 common assays and compared the results with a similar survey conducted in 1997.

Design.-One hundred eighteen laboratories prospectively recorded the collection-to-verification turnaround time for 9252 complete blood cell counts (CBCs), 8832 thyroid tests, and 9193 basic metabolic panels.

Results.-The median facility reported all test results by 7:00 AM of the weekday immediately after the date of specimen collection. The bottom 10% of institutions reported 99% of CBCs and basic metabolic panels within 1 day and 60% of thyroid tests within 1 day. The 65 institutions that participated in both the 1997 and 2002 surveys showed significant overall improvement in turnaround time for all 3 types of tests (P

Conclusion.-The turnaround time of routine outpatient tests appears to have improved between 1997 and 2002.

The prompt and predictable reporting of routine outpatient tests can increase efficiency of outpatient care and improve customer satisfaction, even when it does not directly affect medical outcome.1 Furthermore, the ability to monitor outpatient test turnaround times and meet certain performance expectations is important to many managed care organizations contracting for laboratory services.2

In 1997, the College of American Pathologists (CAP) surveyed 614 laboratories that collectively reported on the collection-to-verification turnaround time of 30 240 outpatient complete blood cell counts (CBCs), 25 683 biochemical profiles, and 14 801 thyrotropin assays.3

A similar survey was conducted in 2002 to assess whether industry performance had changed during the past 5 years. This brief report summarizes the results of the 2002 CAP survey of outpatient turnaround time and compares its performance to 1997 survey results.

METHODS

Study Format and Data Collection

The study was conducted according to a format previously described.4 After pilot testing and refinement of the data collection instrument, CAP Q-Probes subscribers were mailed data collection instructions during the first quarter of 2002. Participating institutions were first asked 7 questions regarding the source and size of the participant's outpatient activity and the practices related to test ordering and specimen transportation as well as whether the institution was involved with managed care plans. Participants were then asked to record the turnaround time of 3 types of tests: automated CBCs, thyroid tests, and basic metabolic panels. Participants prospectively identified 4 tests of each type collected, beginning on a Monday, and determined whether the result was verified within 1 day of specimen collection. The accumulation of tests continued for 4 weeks (weekdays only) until a total of 80 specimens of each test type had been examined. The data acquisition process used in 2002 differed slightly from the process used in 1997. Whereas participants in 1997 were asked to examine the turnaround times of tests received on different laboratory shifts, 2002 survey participants were not required to select specimens from different shifts, since this variable did not prove to have a large impact on outpatient turnaround time in 1997.

Definitions

To ensure the comparability of participant responses, common definitions were used by all participants. Eligible outpatient specimens were nonstat specimens obtained on a weekday from hospital outpatients or were outreach specimens collected by an office-based phlebotomist, by a home-draw phlebotomist, in an outpatient phlebotomy service center, or from a nursing home resident. Weekend specimens, stat specimens, and specimens from inpatients, the emergency department, employee or business health services, health screening fairs, and veterinary testing were excluded. CBCs with a manual differential count were excluded. Participants were directed to evaluate the basic metabolic panel and thyroid test(s) most commonly ordered in their institution. A test was considered to have been completed within 1 day if results for all of the ordered analytes were available to clinicians by 7:00 AM on the first nonholiday weekday after the date of specimen collection. For example, all components of a basic metabolic profile collected on a Thursday at 4:00 PM had to be verified by the next day (Friday) before 7:00 AM in order for the specimen to have been considered turned around within 1 day. The time of paper report delivery was not measured.

Statistical Analysis

Data were analyzed with SAS software (SAS Inc, Cary, NC). Some laboratories did not report data for all 80 specimens. Specifically, 5 of 117 laboratories submitted incomplete data for CBCs, 7 of 117 laboratories for metabolic profiles, and 11 of 115 laboratories for thyroid testing. Laboratories with missing data were excluded from analyses that required the data. To test for associations between practice variables and turnaround time, the Kruskal-Wallis test was used for discrete-valued independent variables, and regression analysis was used for continuous-valued independent variables. Those with significant associations (P