February 2005
Feature Story
Anne Paxton
When standards are allowed to slip, the human cost can
be incalculable. That’s a truism that was starkly brought home last year
when the Baltimore Sun published the story of Maryland General
Hospital’s laboratory.
Operating with faulty equipment despite repeated complaints from staff,
the laboratory had continued to release HIV and HCV test results—numbers
that turned out to be unreliable.
Results of tests on controls were edited to be brought within acceptable
ranges, while confirmatory tests that conflicted with the hospital’s test
results were ignored. At least 460 people were given potentially inaccurate
HIV or HCV test results, hospital officials estimate.
All of this came to light because of a staff member who alerted Maryland
state health officials.
"We heard about it right in the middle of a meeting on March 11, the
day the story first appeared," says Ronald B. Lepoff, MD, chair of the
CAP Commission on Laboratory Accreditation. There followed months of meetings,
conference calls, policy changes, Maryland Senate hearings, and even testimony
before Congress.
Now, CAP officials and other leading figures in the laboratory field
look back on what most agree was a preventable tragedy—one that
must serve as a wakeup call to laboratories and hospitals.
"Everybody has feared the cuts made in the infrastructure of some hospital
laboratories would start to show up, and that an MGH would occur," says
Pennell C. Painter, PhD, professor emeritus of pathology and director
of laboratory operations for LabCorp/Dynacare Tennessee Laboratory, Knoxville.
History explains much of it, he says.
Between 1966 when Medicare began and 1983 when DRGs came in, "that was
essentially the heyday of medicine from a financial perspective. Many
hospitals were built in that period, and actually if you look at laboratory
personnel and nurses, about 70 percent of all the people who have ever
been trained were trained in those years."
It was after 1983, he says, that the money gradually started going out
of health care. And in 1992 the government "cracked down severely on utilization
and used the judicial and enforcement branches to penalize for overuse
of laboratory tests. That further drew down the income for laboratories,"
Dr. Painter says.
"Now we’re over 20 years out since DRGs came into effect, and laboratories
have seen a gradual erosion of funds so that the vast majority of labs
have moved from being profit centers to cost centers for hospitals."
Some laboratories have faced progressive cuts in their resources, to
the point that "the whole infrastructure of quality started to fall out
from under them."
"What we’ve been existing off is essentially that ’bank account’ of trained
people," Dr. Painter says. "People are what keep the laboratories’ quality
systems from collapsing—not just bodies, but people who know what
they’re doing and have the time to oversee and foster all that goes into
making a quality-focused laboratory."
The hospital administrative leadership is the first tier in the hospital
charged with ensuring that the laboratory has what it needs to do quality
work, he says. "That’s something very few people in the laboratory are
aware of. It’s part of the hospital Medicare Conditions of Participation."
What tends to be forgotten, he says, is that the medical director and
owner of the laboratory are the only two names that appear on the CLIA
certificates, so they are held directly responsible for everything that
happens in that laboratory.
If they don’t know what’s going on, they can be in for a "big surprise,"
he says, when something goes awry. "Some of the top people in both lab
and hospital administration lost their jobs at Maryland General, and a
lot of this was because they did not follow through on some of their responsibilities."
Dr. Painter recalls a case reported several years ago involving erroneous
PT INR results because someone forgot to change the ISI factor used to
calculate the INR, resulting in patient deaths. "But that was literally
a single tech making a clerical mistake."
"The situation at MGH was very different. It was a quality systems failure,
with one or more instrumental analyzers turning out incorrect results
over a protracted period, the staff notifying management that something
was wrong with the testing equipment, and management not effectively correcting
the problem while permitting patient result reporting to continue."
If equipment is not working properly, "the remedy in lab sciences is
to cease testing immediately, stop testing until the testing equipment
begins working properly, and if there’s not enough money to fix it, then
it stays out of service. If management decides to continue testing even
though the testing systems are not working reliably, then that’s a serious
problem."
Dr. Painter is not comfortable with the term "whistleblower" in this
context. "The techs doing the testing have always been our first-line
guardians of laboratory quality. You cannot reasonably expect to uncover
the kinds of things we’re talking about here during a typical laboratory
inspection, so our safeguard for quality goes back to the personnel doing
the testing who, under CLIA, are required to report such problems to management."
But that big "savings account" of formally trained medical technologists
is drying up now as technologists retire, he says, and that looming laboratory
testing personnel shortage sets the stage for potential quality lapses.
"Some places bring people in off the street and train them, which you
can do under CLIA for anything moderately complex." That includes almost
any automated analyzer, "so the most complicated system on the market
that can do hundreds of chemistry tests at high speed is considered moderate
complexity."
"The people trained to be operators often don’t understand what the results
mean, and don’t know when they don’t make sense," he says. That doesn’t
mean they can’t turn out reliable lab work, but it does put an additional
burden on the lab’s quality system.
Autoverification systems are starting to gain the power to make some
of the same informed decisions a technologist would, however, and just
like a good technologist, Dr. Painter says, the better systems are able
to look across all other test results in a panel and determine whether
they appear reasonable before they are reported. "So as those techs who
’understand’ laboratory tests and laboratory science leave the profession,"
he says, "even better quality surveillance and more intelligent instrumental
systems will be needed."
In the Maryland General case, there were signs that some employees were
not aware that the CAP is open to hearing complaints from employees, Dr.
Lepoff says.
Since 1992, when the government granted deeming authority to the CAP’s
Laboratory Accreditation Program, the CAP has had in place a person dedicated
just to complaints. "Complainants are a very important source of information,
since no accrediting organization can be there all the time and know what
happens every day to every sample," Dr. Lepoff says.
To re-emphasize that message, beginning Oct. 1 all 6,000 CAP-accredited
laboratories were required to post a sign asking employees with concerns
about the quality of patient testing or laboratory employee safety to
call a toll-free number, and promising that their identity will be kept
confidential.
Since then, the College has received about 20 calls a month, with perhaps
10 percent of them indicating potentially serious problems.
Gerald Hoeltge, MD, is the CAP’s special commissioner for non-routine
processes, also known as the "complaints commissioner." The lessons that
emerged from the Maryland General situation apply to all laboratories,
he says.
"No. 1 is that leadership is essential to laboratory quality. The most
senior people in the organization have to be committed to quality, and
every laboratory has to have indicators that are meaningful and monitored
by those at the top."
No. 2 is that technical staff need to take pride in excellence. "They
do need clear direction regarding quality, but they also need to be involved
in discussions about quality, and they need feedback on their own performance
and that of the laboratory as a whole."
The third lesson is that it’s essential to pay attention to the details
of every step in the analytic process, he says.
The CAP’s move to make it easier to lodge complaints has eliminated the
"threshold of inconvenience," Dr. Hoeltge says. The College employs trained
technologists who have a structured interview form to follow when receiving
these calls.
"Sometimes I’ll call back, or it will be one of the senior people in
CAP accreditation, if we need more information, because we want our followup
to the laboratory to be as efficient and helpful as possible.
"The whole thing is about laboratory improvement. We’re not out to get
on people’s cases; we’re here to help people find opportunities to improve."
In Dr. Lepoff’s view, the biggest lesson of the Maryland General situation
is that the responsibility of the laboratory director for quality cannot
be abrogated.
"Of course the situation was not simple, and I don’t pretend we’ll ever
fully understand what happened. But this was a typical, inner-city urban
safety-net hospital under financial pressure, as a lot of hospitals are.
It has a lot of indigent patients, a lot of homeless patients."
"There were cutbacks, and at one point apparently they placed an administrative
director into the laboratory who basically took administrative control
from the laboratory director."
"So the laboratory director, who is a pathologist, either ceded or was
forced to cede administrative control to a person who reported, as nearly
as we can tell, directly to the hospital."
"The hospital was very cost-conscious, and clearly in the process of
cutting back staff they went too far, and employees were hurt and patients
were exponentially hurt because of poor decisions made in terms of quality."
That became very clear when the government, the Joint Commission on Accreditation
of Healthcare Organizations, the state, and the CAP went in to look at
the laboratory, he says. "As long as the laboratory director’s name is
on the CLIA certificate, he or she has ultimate responsibility for everything
that happens in the laboratory. The argument that the administration has
control just doesn’t hold water."
It may sound trite, Dr. Lepoff adds, but a second lesson is that quality
improvement is important. "It is so easy to pay lip service to quality
assurance and not actually close the loop and fix things. And that is
not going to fly. The quality assurance practices of the laboratory are
the key to maintaining a quality product."
He acknowledges there’s no way to prevent hospitals from making poor
management decisions, but says, "What we can do is make sure the laboratories
we accredit function at a higher level than is required by law by doing
a thorough and a fair inspection that represents the state of the art
in how laboratories should practice laboratory medicine."
"The way to fix medical errors is not to have them happen—and the
way to do that is to make sure the system works right."
Because people in the laboratory and the hospital are busy, sometimes
oversight is not what it should be and followups don’t occur, says Kent
Lewandrowski, MD, associate chief of pathology at Massachusetts General
Hospital and associate professor at Harvard Medical School. "It reflects
more of an organizational disease whereby bad practices or complaints
from employees don’t necessarily get the attention they should."
The CAP and JCAHO inspections are equally vulnerable to missing these
lapses, in his view. "No inspection is going to ensure uniformly across
the U.S. that every laboratory is functioning optimally in all situations."
"The regulations are not built around a system designed to detect and
eliminate fraud, and I think that’s a key idea here," Dr. Lewandrowski
contends. "People can be engaged in good, bad, or indifferent medicine,
but those engaged in fraud are entering into a potentially more criminal
realm."
Is a new regulatory approach needed? One Maryland congressman thinks
so: He has introduced a bill to require unannounced inspections for laboratories
under Medicare and federal whistleblower protections for laboratory personnel.
But Dr. Lewandrowski says the reflex response should not be to install
a raft of new regulations, conduct a public witch-hunt, or panic and overhaul
the inspection process.
"You have to put it in perspective. You recognize acetaminophen toxicity
kills people every day. Should it be pulled off the market because it’s
a hazard to public health? Do we have to change the basic system of laboratory
oversight? In my opinion, the answer is probably no.
"No amount of system and regulation, short of one that consumes the entire
gross domestic product of the U.S., can prevent a similar incident happening
somewhere sometime," Dr. Lewandrowski says. As long as patients are sleeping
on the street outside the hospital who may have HIV and don’t know it,
and tens of millions of patients are uninsured, he says, he thinks "health
care dollars could be better spent."
"Was this a local problem with people clearly not doing the job they
were supposed to? Or is this a widespread problem with the system coming
apart and breaking down? My feeling is it’s the former. Every debacle
should not result in billions of dollars in new regulations that rob productive
funding from the system for little yield."
Nevertheless, says Dr. Hoeltge, who is head of the section of transfusion
medicine at the Cleveland Clinic Foundation, Maryland General is not unique.
"This is a hospital just like the one in your own neighborhood. It happens
to be urban, but it’s a community-based primary-care hospital. It has
family practitioners. People come for treatment for diabetes and high
blood pressure. It’s as typical as can be. The only difference is it happens
to be located in an old part of Baltimore."
It’s critical to remember that bad things can happen and the probability
is they will happen, Dr. Painter says, if a lab does not have a good quality
system that is fully supported. "Laboratory instrumental and noninstrumental
testing systems are inherently unstable, meaning laboratory testing systems
will go wrong. Not ’might,’ but ’will,’" he says.
"Laboratory testing is very complex and some institutions don’t give
adequate credence to just how likely a bad result is to occur," he adds.
But the quality systems on which high-quality labs operate have long been
in place, and they work. "There is no magic bullet that can allow you
to skip steps and still maintain consistent quality."
That’s why Dr. Painter says the Maryland General experience should perhaps
not be thought of in terms of the lessons learned there but in terms of
the "lessons we should have remembered."
Anne Paxton is a writer in Seattle. |
