Results: Without additional staffing, the combination of a robust
sampling and data review protocol and the electronic accessioning
process eliminated error and reduced sample processing time. The
processing efficiencies allow what would have taken up to a week to
be achieved in hours. The permanently attached cryogenic barcoded
sample label made processing activities more efficient and the
location code ensured the correct sample storage location for future
retrieval.
Conclusion: The developed sample management system proved to
be an effective and accurate tool for processing large number of
samples for population study.
Presenter: Collin Riker, New Jersey Department of Health, Ewing,
NJ, [email protected]
When Does a Public Health Laboratory Reject Specimens?
A Look at Specimen Rejection by Facility and Error Type
During a Measles Virus Outbreak in Brooklyn, NY, 2018
A. DeVito 1 , C. Mahle 1 , U. Siemetzki-Kapoor 1 , M. Iwamoto 2 , J. Rosen 2 ,
J. Rakeman 1 ; 1 New York City Public Health Laboratory, 2 New York City
Department of Health and Mental Hygiene
PHL received specimens for measles testing from 47 submitters,
five of which were considered to be high volume. Forty-seven
specimens (15.8%), representing 55 test orders, were rejected for
testing. Thirty one of these specimens (66.0%) were rejected due to
specimen hemolysis. The remaining specimens were rejected due
to various specimen integrity issues such as insufficient specimen
quantity, sample received without proper transport medium,
improper storage of the specimen, old collection date, or requests
to cancel the order. Of 16 rejected specimens received from high
volume submitters, the most common rejection error was specimen
hemolysis (n=11, 64.7%). Hemolysis was the reason for rejection of
20 specimens submitted from low volume submitters (n=31, 64.5
%). No specimen was rejected due to incomplete submission forms
with errors that could not be fixed.
The most common reason for specimen rejection was specimen
hemolysis, and there was no difference in the most common reason
for rejection between high and low volume submitters. Submissions
with errors in paperwork or specimen integrity cause slowdowns
in the testing process due to time spent correcting errors, while
specimen rejection can cause additional delays that may involve the
patient returning to the provider for a specimen to be recollected.
These issues impact the speed at which results are reported to the
provider and Health Department for follow up. Improving practices in
82
LAB MATTERS Summer 2019
Presenter: Andrea DeVito, New York City Department of Health and
Mental Hygiene, New York, NY, [email protected]
A User-friendly Shiny Web Application for Choosing Pool
Sizes When Testing Pooled Specimens
C. Bilder 1 , B. Hitt 1 , J. Tebbs 2 , C. McMahan 3 ; 1 University of Nebraska-
Lincoln, 2 University of South Carolina, 3 Clemson University
Background: High volume screening of clinical specimens for
infectious diseases is often made possible by a process known as
pooled testing. This algorithmic process involves testing portions
of specimens from separate individuals together as one unit (or
“pool”) to detect infection. Follow-up retesting is performed on
members of positive-testing pools to decode the positive members
from the negative ones. An important decision needed prior to
implementation of pooling is what pool size to use. Choosing too
large of a pool size can lead to a large number of retests, perhaps
even resulting in a total number of tests larger than what would
occur by individually testing specimens. Choosing too small of a
pool size can also lead to an overall larger number of tests than
necessary.
Methods: To help laboratories choose a pool size, we developed a
Shiny web application that leverages the power of the R statistical
software package. This application uses analytical derivations and
computer simulation methods to emulate the pooling process. A
user-friendly web interface is provided so that users do not need
experience with R to perform calculations.
Results and conclusions: The application calculates the expected
number of tests and the expected accuracy for commonly used
pooling algorithms. This application can also determine the
“optimal” pool size(s) based on minimizing the expected number of
tests. Access to the application is available through our
www.chrisbilder.com/shiny website.
This research is supported by Grant R01 AI121351 from the
National Institutes of Health.
Presenter: Christopher Bilder, University of Nebraska-Lincoln,
[email protected]
Analytical Validation of a Sample-to-Sequence Pipeline for
Non-Targeted Pathogen Detection in Clinically Relevant
Matrices
K. Parker 1 , B. Knight 1 , H. Wood 1 , D. Yarmosh 1 , J. Russell 1 , J.R.
Aspinwall 1 , K. Werking 1 , P. Chain 2 , P.E. Li 2 , R. Winegar 1 , 1 MRIGlobal,
2
Los Alamos National Laboratory
The ability to identify an unknown infectious agent in a clinical
sample is often limited by the tools available to the clinician.
Current microbial and molecular methods are complicated by
factors such as fastidious growth conditions, the need to perform a
series of differential growth tests, and the challenges of designing
large panels of molecular assays that are both sensitive and specific
over a broad range of organisms. Next generation sequencing (NGS)
provides a means for unbiased detection of pathogens from a
variety of clinical matrices. With a non-targeted approach, NGS has
PublicHealthLabs
@APHL
APHL.org
In October 2018, the NYC Public Health Laboratory (PHL) began
responding to a measles outbreak in Brooklyn. Measles virus testing
includes real-time PCR testing of nasopharyngeal (NP) swabs and
IgM and IgG testing of serum or blood samples. Requirements for
testing clinical specimens include completing a paper submission
form, proper labeling of the specimen, and proper collection,
storage, and transport of the specimen. DOHMH works to ensure
that any correctable error (i.e., missing or illegible information) is
resolved through follow up with the submitter. Between 10/1 and
12/14, PHL received 419 requests for measles virus testing on
298 specimens. We examined specimen rejection rates by type
of submission error and type of facility (high volume submitters
were those that submitted >100 specimens of any type in 2018; all
others were considered low volume submitters).
the pre-analytical phase especially during an outbreak is of critical
importance for timely initiation of testing, follow up, and public
health action.