PLoS ONE (Jan 2016)

Reduction in Hospital-Wide Clinical Laboratory Specimen Identification Errors following Process Interventions: A 10-Year Retrospective Observational Study.

  • Hsiao-Chen Ning,
  • Chia-Ni Lin,
  • Daniel Tsun-Yee Chiu,
  • Yung-Ta Chang,
  • Chiao-Ni Wen,
  • Shu-Yu Peng,
  • Tsung-Lan Chu,
  • Hsin-Ming Yu,
  • Tsu-Lan Wu

DOI
https://doi.org/10.1371/journal.pone.0160821
Journal volume & issue
Vol. 11, no. 8
p. e0160821

Abstract

Read online

Accurate patient identification and specimen labeling at the time of collection are crucial steps in the prevention of medical errors, thereby improving patient safety.All patient specimen identification errors that occurred in the outpatient department (OPD), emergency department (ED), and inpatient department (IPD) of a 3,800-bed academic medical center in Taiwan were documented and analyzed retrospectively from 2005 to 2014. To reduce such errors, the following series of strategies were implemented: a restrictive specimen acceptance policy for the ED and IPD in 2006; a computer-assisted barcode positive patient identification system for the ED and IPD in 2007 and 2010, and automated sample labeling combined with electronic identification systems introduced to the OPD in 2009.Of the 2000345 specimens collected in 2005, 1023 (0.0511%) were identified as having patient identification errors, compared with 58 errors (0.0015%) among 3761238 specimens collected in 2014, after serial interventions; this represents a 97% relative reduction. The total number (rate) of institutional identification errors contributed from the ED, IPD, and OPD over a 10-year period were 423 (0.1058%), 556 (0.0587%), and 44 (0.0067%) errors before the interventions, and 3 (0.0007%), 52 (0.0045%) and 3 (0.0001%) after interventions, representing relative 99%, 92% and 98% reductions, respectively.Accurate patient identification is a challenge of patient safety in different health settings. The data collected in our study indicate that a restrictive specimen acceptance policy, computer-generated positive identification systems, and interdisciplinary cooperation can significantly reduce patient identification errors.