See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/338772522

EP Evaluator-CLIA (EE-CLIA) for Evaluating Clinical Laboratory Methods

(Consultant’s Version with Rapid Results Entry). Kennett Square, PA: David G.
Rhoads Associates, Inc., 2000,...

Article  in  Clinical Chemistry · November 2001

DOI: 10.1093/clinchem/47.11.2075

CITATIONS READS

0 21

2 authors:

Robert Rej Carol S Norton-Wenzel

Wadsworth Center, NYS Department of Health Wadsworth Center, NYS Department of Health
149 PUBLICATIONS   3,254 CITATIONS    7 PUBLICATIONS   74 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Aging Hudson River Residents Cohort View project

Hypothyroidism and PCB effects View project

All content following this page was uploaded by Robert Rej on 13 February 2020.

The user has requested enhancement of the downloaded file.

 Clinical Chemistry 47, No. 11, 2001 2075

ductory volume of this type, it is not CRC Handbook of Laboratory It also explains in detail the different
possible to provide an in-depth de- Safety, 5th ed. A. Keith Furr, ed. types of laboratory facilities required
tailed discussion of every type of Boca Raton, FL: CRC Press LCC, for microbiologic and biomedical
analysis. However, the discussion 2000, 774 pp., $149.99. ISBN 0-8493- studies, a list of very useful refer-
and literature references provided 2523-4. ences, and Internet addresses and
with each chapter are sufficient for links to websites to contact any of the
This handbook is a very useful and
an investigator to use this volume as resources directly. This book is an
practical guide designed to help any
the starting point to expand the par- invaluable resource for the safety
person working in a laboratory envi-
ticular area using primary literature and training of laboratory workers,
ronment. The handbook extensively
sources. The authors have included including students, researchers, and
covers topics from regulations to

Downloaded from https://academic.oup.com/clinchem/article-abstract/47/11/2075/5639292 by guest on 13 February 2020

two lengthy chapters, one on the their coworkers.
procedures. It includes detailed
analysis of a variety of small mole- questionnaires on different areas, in-
cules and the second on the analysis Rosa Sierra-Amor
cluding complete and detailed infor-
of biopolymers. These chapters pro- mation about chemicals, personal
Clinical Research Center
vide overviews of both the areas protective clothing, laboratory check-
Children’s Hospital
and the basis for development of lists, and the characteristics that
3333 Burnet Ave.
methods for a specific analyte. The members of the safety department
Cincinnati, OH 45229-3039
chapter on small molecules includes and its officers should have to teach
descriptions of the analysis of impor- and supervise their coworkers. It also
tant classes of compounds, such as includes how to design a laboratory
EP Evaluator-CLIA (EE-CLIA) for
polyaromatic hydrocarbons, and and how safe this should be as well
Evaluating Clinical Laboratory
pharmaceuticals, such as steroids as classification of laboratories,
Methods (Consultant’s Version
and tricyclic antidepressants. Meth- hoods, chemical storage, and animal
with Rapid Results Entry). Kennett
ods for the separation of biopoly- holding facilities.
Square, PA: David G. Rhoads Asso-
mers by CEC are still developing, but The author states that “if the scien-
ciates, Inc., 2000, 3 diskettes and
tific worker, for whom this hand-
descriptions of the analysis of oligo- manual (142 ⫹ 33 pp.), $695.00.
saccharides, peptides, proteins, and book is intended, follows the recom-
oligonucleotides are provided. For mendations in this handbook, the This software package is designed to
users who have an existing HPLC requirements to meet the desired assist the clinical laboratory in meet-
analytical procedure, the authors outcome of the standard should be ing minimum regulatory require-
met, but a written plan is required”. ments for method evaluation and
also provide a detailed chapter on
The handbook presents the require- validation. System requirements are
the transfer of HPLC methods to
ments to meet a key component of modest (PC with a 486 CPU, 8 MB
CEC. Analytical chemists or bio-
the Occupational Safety and Health RAM, and 10 MB hard drive space);
chemists already familiar with HPLC
Administration (OSHA) standard, the operating systems supported are
and capillary electrophoresis will the Chemical Hygiene Plan (CHP). Windows 95, 98, or NT. We installed
find this volume a valuable resource The information on Material Data and ran the program on several plat-
for bringing the techniques of CEC Safety Sheets (MSDSs) is very infor- forms (Athlon or Pentium II/III
and PEC into their laboratories. mative and will help the user to CPU), all using the Windows 98SE
In the final chapter, the authors better understand why MSDSs pro- operating system. After a straightfor-
attempt to predict the future of CEC vided by different companies vary ward installation, the software may
and PEC against the background of dramatically in quality. be used for 45 days, after which one
existing mature separation tech- Tables on chemical combinations must contact the vendor to obtain a
niques and emerging analytical capa- are very detailed, and the user of this serial number that is specific for that
bilities. The application of CEC and handbook should be able to readily computer.
PEC to the clinical laboratory will find this important information. Also Data may be entered into the pro-
require the production of commer- included are the sample hazard color gram either manually or by import of
cial instruments. Currently, the small codes for chemical containers used ASCII text files. Our version in-
number of commercially available by different manufacturers. The cluded the “Rapid Results Entry”
instruments limits these techniques chapter that covers the OSHA blood- feature, which speeds manual data
to largely research applications. borne pathogen standard provides entry when results for multiple ana-
details on basic provisions, exposure, lytes must be input for individual
HIV infection, hepatitis A and B, specimens. This module adds $100 to
Thomas J. Ryan Zoonotic diseases, reproductive haz- the cost of the program. We tested
ards, carcinogens, and neurologic the data import function, using sev-
Laboratory of Molecular Diagnostics hazards. Another chapter covers in- eral ASCII files, without success. We
Wadsworth Center fectious waste and alternative tech- contacted the software publisher,
Albany, NY 12201 nologies, storage, and transportation. who was unaware of such a problem
 2076 Book, Software, and Web Site Reviews

but verified that our conforming file cate analysis of the zero and a non- provided; some are likely unlocat-
could not be successfully imported. zero calibrator. This is the sole mod- able, and none is current. Each mod-
In response to our communications ule without a graphical output, a ule has a help screen; sometimes this
with the publisher, a revised version feature that should be included. information is at odds with the man-
of the program was available at their The Verification of Reference Inter- ual. We found neither program help
download site in a short time and val module provides an estimate of nor the manual to be particularly
was functional. Import is partially the reference interval as the central enlightening or grammatic.
menu-driven, but the process is not 95% of a given population using Guidance for the data to be col-
straightforward. For example, in the from 20 up to 65 000 results. A histo- lected is of limited help and usually
method comparison module, two gram and statistics, including per- unreferenced; for example, the fol-

Downloaded from https://academic.oup.com/clinchem/article-abstract/47/11/2075/5639292 by guest on 13 February 2020

different imports are required, even centage of results outside a proposed lowing is stated for the number of
if data are in the same file. After reference interval, mean, SD, me- individual specimens to be used in
import, sample numbers are dis- dian, and range, are calculated. The verification of a reference interval:
played in order by the first character, “Pass/Fail” evaluation is based on
so that a specimen numbered “2” ⬎10% of the results falling outside “For verifying a proposed refer-
appears after a specimen numbered the reference interval to be verified; ence interval: 20 points
“199”. Export of data from EE-CLIA inappropriately broad reference in-
For creating or changing a refer-
in various ASCII file formats, includ- tervals are certain to be issued a
ence interval: Minimum of 120
ing those usually parsed by spread- “Pass”.
points. In some cases, 300 to 400
sheet and database programs and by Both “regular” and Deming re-
are recommended“
other versions of EP Evaluator, can gression analysis calculations are
be accomplished by options within provided in the Alternate Method
each module. Comparison module. One should At nearly $700, we believe that this
EE-CLIA operates from a main note that t-test statistics are included package should offer more than five
menu that allows selection from five as a default in this module, although statistical modules (several of which
statistical modules: Simple Precision, this function may be disabled if the could be set up in short order with a
Sensitivity, Verification of Reference user considers this statistical param- spreadsheet program). A baroque
Interval, Alternate Method Compar- eter inappropriate for use in compar- pricing structure exists for this prod-
ison, and Linearity. An irrelevant ison studies and would like to pre- uct [the current web page (http://
cartoon appears on the first screen, vent this calculation from being www.dgrhoads.com/lab_statistics_
but this is mercifully a defeatable performed. In addition to the results pricelst.sht#ecl) lists 21 different
“feature”. The five analysis modules summary page, up to three different prices, depending on use], so your
are similarly structured with menu graphs [scatter, bias (i.e., Bland–Alt- cost may be higher or lower. As an
options provided for selection of ex- man), and percent bias] may be aid for interested potential custom-
periment (analyte, instrument), spec- viewed and printed by this module. ers, a fully functioning demonstra-
ification of parameters (e.g., lot num- The Linearity module performs tion copy can be downloaded at
bers and allowable error limits), several functions, including assess- www.dgrhoads.com.
input of test data, data analysis, and ment of accuracy, linearity, report- In summary, this package pro-
file utilities (import and export of able range, calibration verification, vides the modules and functions as
data). Each module can generate a and probability of proficiency test advertised. We liked the convenient,
printed summary of each data anal- failure. A nice feature contained in well-organized, and professional
ysis and corresponding plot(s); this module is the “History” option, summaries of all graphs and statis-
screen previews are also an option. which allows the user to plot up to tics in a printable format. The results
The Simple Precision module pro- three sets of linearity data (e.g., cur- are particularly handsome when
vides an estimate of mean, SD, and rent data and data saved from two printed in color. For the small to
CV for replicate results. The analysis previous comparisons) on the same medium-sized laboratory seeking to
assumes a gaussian distribution, and graph. meet minimum validation standards,
no estimates of kurtosis or skewness The manual provides the user both this package definitely warrants in-
are provided. The graphical analysis basic instruction on the operation of vestigation.
displays SDI (⫺3 to ⫹3) vs sample each program module and some
number. A very convenient feature guidance for the interpretation of the
allows exclusion of data points by statistical parameters used. Errors Robert Rej
selecting them directly from the abound in the manual, particularly Carol S. Norton-Wenzel
graph. with misidentification of figure and
EE-CLIA incorrectly defines “sen- table numbers within the text and in Wadsworth Center for
sitivity” (see Clin Chem 1997;43: figure legends. At least two of the Laboratories and Research
1831–7) but does spell out how this figures that were referenced in the New York State
parameter—actually the detection text were not included in the manual. Department of Health
limit—is calculated based on repli- A few bibliographic references are Albany, NY 12201-0509

View publication stats