Patient Safety Tip of the Week
May 26, 2015 How Safe is the Lab
You Use?
The Milwaukee Journal Sentinel has done an excellent series
on patient safety issues related to laboratories (Gabler
2015). Gabler does a great job using real
people stories to illustrate the devastating impact a lab error can have on the
lives of patients and their families. But she goes much further and delves into
the lack of transparency about laboratory errors and the inability of the
general public or even you, the healthcare professional, to know how well the
laboratory to which you refer patients is actually performing.
Our regular readers know we are fond of “stories, not
statistics” to foster patient safety. Gabler shows
how inaccurate HIV diagnoses and false paternity tests destroyed families and
incorrect pregnancy testing and blood compatibility testing put neonates in
jeopardy. And how failure to heed a lab technician’s warnings about a faulty
analyzer resulted in that lab tech being exposed to HIV and HCV and more than
400 patients possibly receiving inaccurate HIV and HCV results.
But, while the stories in this series are indeed compelling,
the statistics are even more bothersome. Gabler
uncovered not only numerous specific errors and deviations occurring in various
labs but also found significant flaws in the regulatory oversight of the labs. Labs
appear to have the ability to choose some quality parameters as they see fit
and they can choose their oversight organization in some cases. One example
given was a lab in which staff failed proficiency testing in pregnancy testing
one quarter. Yet rather than do repeat proficiency testing for pregnancy
testing that lab “simply didn’t participate in an outside check of its
pregnancy testing the following quarter”. Gabler also
discusses labs cutting corners to save money (eg.
using expired reagents, putting off fixing faulty equipment, etc.).
There are 35,000+ labs in the US. While they must meet the
federal standards of the Centers for Medicare and Medicaid Services (CMS), the
surveys are usually outsourced to other organizations. State inspectors handle
about 50% of labs, with accrediting organizations like The Joint Commission
(TJC) or the College of American Pathologists (CAP) reviewing 46%, including
most hospitals and large clinics. New York and Washington run their own
programs, accounting for the remaining 4%. CMS sends teams to survey a small
percentage of labs (about 2%) and to audit the surveys being done by those
other organizations.
When CMS does those audits they allow up to a very generous
20% disparity rate (between the CMS findings and the other organization’s
findings) before CMS reviews the work being done by those organizations. In
2013 The Joint Commission exceeded that 20% threshold, with 9 of 43 audited
inspections not meeting the CMS expectations.
But the public might become aware of faulty laboratory
practices only if a regulatory body sanctions that lab. And such sanctions are
relatively rare. Findings of surveys are not made public if the lab submits a
plan of corrective action that is accepted by the oversight agency.
We all appreciate it when organizations like The Joint
Commission take on an “educational” role and an approach to facilitate
improvement. But at the same time we don’t want to be sending our patients to a
lab that has substandard processes.
So how do you know if your favorite lab is safe or not? You
won’t find the answer on any website. Your best bet is to be tied into your
hospital’s quality improvement system and make sure that your medical staff
representation on all the quality committees is actively participating. But
that will only help you get a feel for how safe the hospital lab is. For those
using proprietary and/or commercial labs, good luck!
So why aren’t there publicly available measures of lab
quality and safety? There is certainly considerable public reporting of
hospital quality and safety measures (albeit some good and some bad measures).
But a patient is actually much more likely to use a lab than a hospital. The
vast majority of quality indicators monitored in laboratories are either too
technical or of little interest to the general public. But some potential candidates
for publicly reported measures might be:
- % of mislabeled specimens
- Specimen loss rate
- TAT (turnaround time) for
some key tests (eg. cervical cancer tests)
Of course, you’d have to have standardized ways of
determining these measures so the system cannot be “gamed”.
We’ve done multiple columns on errors related to laboratory
studies (see list below), many or most of which actually occur before the
specimen arrives at the lab (as in our Patient Safety Tips of the Week for October
9, 2007 “Errors
in the Laboratory“ and March 6, 2012 ““Lab”
Error”). Lost lab specimens
can leave patients without a diagnosis (see our November 16, 2010 Patient
Safety Tip of the Week “Lost
Lab Specimens”). One particularly serious error highlighted in
several of our columns is mislabeling of
specimens. A CAP study in 2011 had shown a specimen labeling error rate of
1.1 per 1000 cases (Nakhleh
2011). Because that has such potentially devastating consequences
for patients, the College of American Pathologists (CAP) has just issued guidelines for uniform labeling of blocks and
slides in surgical pathology (Brown
2015). CAP put together a panel of experts to develop that guideline. A
systematic literature review found the overall evidence inadequate to inform
the guideline so the panel had to rely on expert consensus opinion for 10 of
their 12 recommendations.
In several of our columns on specimen mislabeling errors we’ve
mentioned that you, as a clinician, should be suspicious when a “surprise”
diagnosis (or lack of an expected diagnosis) comes back from the lab. There are
DNA-based tools that labs can use to look for switched or cross-contaminated lab
specimens.
There are other checks that can be done for some highly
sensitive tests. For example, in the Gabler article
one lab has two samples of each person's DNA tested by two different lab
technicians. Another safeguard requires that whenever a man is excluded as
father of a child, the company double-checks to make sure the child's swab
wasn't accidentally switched with the mother's swab, since the two often have
their cheeks swabbed at the same time.
So, how safe is the lab you use? What quality indicators
would you like to see for those labs?
Some of our other
columns on errors related to laboratory studies:
- October
9, 2007 “Errors
in the Laboratory“
- November
16, 2010 “Lost
Lab Specimens”
- October
11, 2011 “LEAN
in the Lab”
- March 6, 2012 ““Lab”
Error”
- April
2012 “Specimen
Labeling Errors”
- January 22, 2013 “You
Don’t Know What You Don’t Know”
- April 15, 2014 “Specimen
Identification Mixups”
- November 25, 2014 “Misdiagnosis
Due to Lab Error”
- March 24, 2015 “Specimen
Issues in Prostate Cancer”
References:
Gabler E. Hidden Errors. A
Watchdog Report. Weak oversight allows lab failures to put patients
at risk. Milwaukee Journal Sentinel 2015; May 16, 2015
Nakhleh RE, Idowu
MO, Souers RJ, et al. Mislabeling
of cases, specimens, blocks, and slides: a College of American Pathologists
study of 136 institutions. Arch Pathol Lab Med 2011; 135(8):
969-974
http://www.archivesofpathology.org/doi/pdf/10.5858/2010-0726-CPR
Brown RW, Speranza VD, Alvarez JO,
et al. Uniform Labeling of Blocks and Slides in Surgical Pathology. Guideline From the College of American Pathologists Pathology and
Laboratory Quality Center and the National Society for Histotechnology.
Arch Pathol Lab Med 2015; Early Online Release April
21, 2015
http://www.archivesofpathology.org/doi/pdf/10.5858/arpa.2014-0340-SA
Print “PDF
version”
http://www.patientsafetysolutions.com/