APHL 2019 POSTER ABSTRACTS
Results: This intermediate-level multimedia online course discusses
a stepwise plan for setting up and validating a new LC-MS/MS
based assay for application in a biochemical genetics laboratory.
A specific example of quantifying methylmalonic acid in human
plasma is used to describe each of the steps required to develop
and validate the method. An outline of the validation report is
provided that includes how to review, interpret and document
validation tests, quality control results, and proficiency tests.
Continuing education credits are available from this course free
of charge, including 1.5 hours of the ASCLS P.A.C.E. credit. The
course is expected to be publicly available on CDC TRAIN website at
https://www.train.org/cdctrain/home in Spring 2019. 23 submitters accounted for 50.1% of errors. More than half (52%)
of specimens with submission errors were associated with 13/152
NYC PHL submitters. The most common error types were: specimen
identifiers not matching those on the submission form (CDC 56%,
NYC PHL 26%), no submission form (CDC 16%, NYC PHL 39%),
specimen labeling errors (CDC 2%, NYC PHL 30%). Additionally,
CDC had a specific error of >1 specimen per submission form
(9%). Existing QA procedures resolved most submission errors,
often through correcting CDC ELIMS and NYC LIMS entries.
CDC ELIMS data showed 35 CDC lab units corrected identifier-
related submission errors in ELIMS (682 corrections per 100,000
specimens with errors).
Conclusions: This online training module is intended to help
laboratory professionals understand the regulatory environment
of a biochemical genetics laboratory, define the steps required to
develop and validate an assay based on LC-MS/MS, and describe
the contents of the validation report. Course evaluation results and
feedback from the participants will be closely monitored to assess
the utility and learning outcomes. Conclusions: Overall, the percentage of specimens submitted to
CDC and NYC PHL with errors was low. This analysis shows the
power of LIMS to inform evidence-based improvement of pre-
analytic processes. Accessioning QA data revealed that most
errors associated with specimen submissions were related
to incomplete or inaccurate submission forms. This analysis
identifies specific areas for quality improvement and provides a
reference to implement future mitigations (e.g., job aids and LIMS-
integrated analytic tools) that would allow labs to conduct internal
assessments.
Presenter: Bertina Su, Association of Public Health Laboratories,
Silver Spring, MD, [email protected]
Improving Specimen Submission Through Data Analysis
of Mislabeled Specimens Received by the Centers for
Disease Control and Prevention and the New York City
Public Health Laboratory
R. Fowler 1 , D. Lowe 1 , C. Kretz 1 , R. Stinnett 1 , M. Lawrence 1 , A.
Marinova-Petkova 1 , M. Hardy 1 , M. Petway 1 , E. Wilson 2 , P. Mandel 1 ,
S. Soroka 1 , J. Rakeman 3 , A. Muehlenbachs 1 ; 1 Centers for Disease
Control and Prevention, 2 New York City Department of Health and
Mental Hygiene, 3 New York City Public Health Laboratory
Background: Accurate specimen identification and test
requisitioning are cornerstones of quality assurance (QA) for
laboratory testing. Specialized groups that receive and accession
specimens, such as CDC’s Specimen Triage and Transportation
Team (STATT) and the New York City Public Health Laboratory (NYC
PHL) requisition team, provide QA measures to identify specimen
submission errors. Laboratory information management systems
(LIMS) data can be used to evaluate the source of these errors
and to identify opportunities for improvement. Our objective was to
identify the frequency and types of submission errors at CDC and
NYC PHL through analysis of LIMS data and QA records.
Methods: CDC’s Enterprise LIMS (ELIMS) was queried to identify
submission errors among the 856, 120 specimens received by CDC
from various PHLs during October 2014–January 2018. Data for
submitter, patient, and CDC lab unit identifiers were blinded. CDC
lab units (n=108) were queried for changes to patient identifiers.
CDC STATT QA records for 265, 397 specimens received by CDC
during July 2014–June 2018 were reviewed to identify additional
specimen submission errors that were resolved before ELIMS
accessioning. NYC PHL LIMS was queried for submission errors
that could be corrected before testing and errors that resulted in
specimen rejection during May 2017–May 2018. Proportion of
specimens with submission errors was estimated by type, submitter
and laboratory unit.
Results: NYC PHL, CDC ELIMS, and CDC STATT identified errors in
1.5%, 0.68% and 0.1% of specimen submissions, respectively. CDC
received specimens with submission errors from 144 submitters;
PublicHealthLabs
@APHL
APHL.org
Presenter: Marlon Lawrence, Centers for Disease Control and
Prevention, Atlanta, GA, [email protected]
Analyzing the Performance of Different Types of Coolers
and Coolants to Improve Cold Chain Transportation
D. Lowe , G. Pellegrini, A. Carter, E. LeMasters and A. Weiner,
Centers for Disease Control and Prevention
Background: Clinical and research specimens often must remain
cold during transportation to the laboratory. Long distances, poor
infrastructure, and limited coolants frequently lead to samples
arriving too warm and possibly unusable. Yet, there is little empirical
evidence addressing how different coolers or coolants perform. Our
objective was to analyze different types of coolers and coolants,
and to develop a model to estimate the amount of coolant needed
based on variable transit times.
Methods: We compared 16 different coolers and 10 different
coolants by the time needed to reach 6°C. Two categories of coolers
were evaluated: currently utilized polystyrene foam and injection
molded coolers, and newer rotation-molded coolers. A standard
amount of coolant filled each cooler and a data logger recorded
the temperature. The coolant categories evaluated were cold packs
and frozen filled water bottles. We measured how long coolants
kept temperatures below a target temperature of 6°C versus cost.
Finally, we varied the amount of coolant to generate a predictive
model for transit times.
Results: Polystyrene foam and rotation-molded coolers remained
below the target temperature longer than injection-molded coolers
(average: 61.4 hrs vs 39.7 hrs). This duration varied with rotation-
molded coolers (range: 36.2-109.5 hrs). The coolant sources also
varied in their time below 6°C (range: 11.1–27.4 hrs) and their
cost (range: $0.11-$12.95/pound). Frozen water bottles remained
below 6°C for a similar time as other cold packs, but were the most
economical. Finally, we varied the amount of water bottles to build
a predictive model that predicted the time to 6°C to within 3hr
(R-squared = 0.994) for rotation-molded coolers.
Summer 2019 LAB MATTERS
79