Analysis of Physician-Requested Manual Differentials in Peripheral Blood Samples: An Institutional Study for Quality Improvement. (21st September 2018)

• Record Type: Journal Article
• Authors: Angulo, Karen Arispe; Gannon, Victoria; Albright, TavaJo; Kroft, Steven; Cunningham, Ashley; Astle, John; Harrington, Alexandra
• Is Part Of: American journal of clinical pathology. Volume 150(2018)Supplement 1
• Page Start: S121
• Page End: S122
• Publication Date: 2018-09-21
• Subjects: Diagnosis, Laboratory -- Periodicals; Pathology -- Periodicals; 616.07
• DOI: 10.1093/ajcp/aqy100.292
• Languages: English
• ISSNs: 0002-9173

Abstract

Abstract: Introduction: In a recent audit of our orderables, we found a discrepancy between the Electronic Medical Record orderable "Manual Diff also order CBC" and the linked Laboratory Information System (LIS) translation "Physician requested manual smear review" (PRMSR). With this orderable, our providers believe they are ordering a manual differential (MD); however, our internal laboratory process has been to perform a smear review, assess the accuracy of the automated differential (AD), and evaluate instrument flags. Objectives: We evaluated the ordering practice for the MD and its impact on our laboratory workflow over a 4-month period. Methods: Institutional databases were queried for the number of PRMSRs received monthly in 2017. Detailed information was collected on the PRMSR requests over a 4-month period using our LIS and Sysmex's WAM. Results: A total of 1, 548 PRMSRs were ordered in 2017, ranging from 116 to 145 monthly requests (mean 129), including 629 PRMSRs during the study period. Patients' ages ranged from newborn to 93 years old (mean 49). The requests were received from inpatients (n = 385), cancer center outpatients (210), and outside clients (34). In total, 263 providers requested PRMSRs, including 44.5% residents/fellows, 25.5% physician faculty, and 18.3% APPs (28% from internal medicine). MDs were performed on 590/629 requests (93.8%) and ADs on 39. WAM OP Alerts occurred in 385/629 PRMSRs (61%), requiring manual review. In specimens without OP Alerts (244 PRMSRs), MDs were performed in 210, and the AD correlated with the MD in 205 cases (98%). Five showed significant disagreement between the AD and MD, requiring an MD report. Thirty-five PRMSRs met criteria for hematopathology review, 34 of which had OP alerts. Conclusion: In 2/3 of PRMSRs, instrument flags are necessitating manual review of the blood film. In the majority without flags, the MD correlates well with the AD results; therefore, this orderable should be discontinued at our institution.