CLIN.CHEM.36/12, 2027-2035 (1990)
Clinical Laboratory Regulation under the Clinical Laboratory Improvement Amendments of
1988: Can It Be Done?
K. MIchael Peddecord1
and Herbert
C. Hammond2
This report examines logical but not yet widely recognized
ramifications of the Clinical Laboratory Improvement Amendments of 1988 (CLIA’88), federal legislation that will require
certification of all laboratories examining human specimens.
Examination of the CLIA’88 committeereports and commit-
tee hearings suggests that more than the conventional
approach to laboratory standards will be needed to meet the
public’s expectations as articulated by our elected representatives. The conventional approach to clinical testing standards seeks to assure quality by regulating the laboratory
analytical process. However, little empirical evidence is available to support or refute this model,which has been used
during the past 25 years. One alternative paradigm for
laboratory standards is an approach that examines the total
laboratory testing process, including the selection, ordering,
and interpretation of the test as well as the laboratory
analysis per se. The history of controversy over laboratory
standards-especially personnel standards, the glacial federal regulatory rulemaking process, public expectations of
fail-safe technology, among other factors-suggests the implementation of CLIA’88 will be a lengthy and vigorously
debated contest. The risk of a test is seldom inherent in the
test itself, but rather is a function of the context in which the
test is being used to provide information for medical decision
making. Our premise is that diagnostic tests must be examined in the context of the laboratory testing situation. We
suggest that now is the appropriate time for laboratory
professionals, practicing physicians, and the public to abandon conventional thinking regarding clinical laboratory standards. We believe that CLIA’88 reflects a shift in public
expectations toward fail-safe laboratory testing and the need
for additional government oversight in laboratory test quality.
If these new expectations persist, CLIA’88 represents a
potential landmark in the course of federal authority and the
practice of medicine in the United States.
AdditIonal Keyphrases: physician’s office testing
of laboratory operation
.
quality control
.
economics
proficiency testing
‘Laboratory
Assurance
Program, Graduate School of Public
Health, College of Health and Human Services, San Diego State
University, San Diego, CA 92182-0405.
2Diion
of Public Health Policy, Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and
Human Services, Washington,
DC.
Opinions expressed are the authors’ and do not represent institutional policy.
Editor’s note: See this issue’s “Clinical Chemist” for the response
and comments of the AACC to the Health Care Financing Administration regarding CLIA’88.
Received June 4, 1990; accepted September 25, 1990.
To undertake an analysis of the Clinical Laboratory
Improvement Amendments of 1988 (CLIA’88) (1), we reviewed scientific and professional literature related to
laboratory quality assurance, quality control, and regulation-both
general and that specifically related to this
law.3 Committee reports from the House and the Senate,
prepared after the passage of the final version of the
legislation, were thoroughly
sional hearings held during
as were selected
examined. Records of congres1987 and 1988 were reviewed,
communications from professional organi-
to the Department of Health and
Services (DHHS) regarding CLIA’88. We also interviewed several congressional staff, DHHS employees,
and Washington
representatives
of several professional
organizations. Employees of DHHS involved in writing
regulations were not interviewed.
zations
who had written
Human
Background
As rationale
posed that
for enactment
of CLIA’88, Congress pro-
Because of the critical role played by laboratory testing in the
delivery of health services and in maintaining
good health, patients expect such tests to be done properly and rely heavily on
others to make sure that is the case. Patients assume, quite
reasonably, that their interests and the public health are being
protected by appropriate governmental agencies [2, p. 101.
We believe that protecting the interests of patients and the
public health necessitates a departure from the conventional view of laboratory regulation that focuses upon the
quality control of analytical activities within the laboratory. We suggest that an alternative paradigm, based on
promoting the quality of the total testing process, will be
needed to satisfy this goal. We believe that the dominant
objective of this law-the objective that reflects the public’s
expectations-is
the abstract goal of improving the contribution of the diagnostic testing process to the delivery of
health services and to the maintenance of health itself.
Furthermore, the patient should be protected from errors in
the testing process, regardless of where these errors occur
and regardless of who makes the error-the
laboratory or
the physician who orders and uses the test. If and when this
objective dominates laboratory standards policy, it will be
as important that these standards
(both professional and
governmental) improve the physician’s selection and use of
tests as it will be to assure the accuracy and precision of the
laboratory test itself.
This essay is an examination of the potential long-run
ramifications
of CLIA’88.
As of this writing,
neither
the
3Nonstandard abbreviations: CLIA, Clinical Laboratory Improvement Act/Amendments;
DHHS, Department of Health and
Human Services; and CAP, College of American Pathologists.
CLINICAL CHEMISTRY, Vol.36, No. 12, 1990
2027
Congress nor the regulatory
bureaucracy
has explicitly
stated that clinical laboratory
testing is anything
other
than the precision and accuracy of the analytical
phase. We
are not arguing that the total testing process approach that
we discuss in this paper is necessarily
the means which
Congress intended to be used to achieve the ends of protecting patients.
Congress has established
a policy goal
that the federal bureaucracy,
through the rulemaking
and
regulatory
enforcement
process,
is attempting
to implement. If the resulting
implementation
of CLIA’88 is too
expansive or too restrictive,
Congress will clarify its policy
intent through
hearings
and or additional
legislation.
When we examined the legislation (1) and various congressional reports (2, 3), we found a remarkable
congruence
between
the total testing process precepts of laboratory
performance,
the requirements
of the new law, and the
expectations
of its drafters.
Whether
the people who
drafted this legislation
necessarily
understood
that the
diagnostic value of a test is inextricably
linked
to its
sensitivity, specificity, and the prevalence of disease in the
tested population is not central to our thesis. Whether they
realized the limitations
of the conventional
wisdom in
laboratory
quality assurance is not critical
either. We
suggest that what will ultimately
matter is the ability of
the laboratory
industry within a regulated framework to
meet the public’s expectation that the diseases affecting the
populace
will be accurately
diagnosed
and effectively
treated.
The Total TestIng
Process Approach to Laboratory
QualityManagement
Mounting evidence argues that the goal of patient protection cannot be realized by merely promoting accuracy in
the analytical
phase of the testing process (4,5). Students
of the total testing process are well aware that three key
factors in the performance
of the testing process are (a) the
formulation
of the clinical question, (b) the interpretation
of the laboratory results in the context of that question, and
(c) the use of those results in subsequent decisions affecting
patient care (6). Most of the evidence is theoretical
or
anecdotal;
however, in general, a false-positive
test obtained from a patient in a very low risk group can have
serious
consequences,
even if only follow-up diagnostic
procedures
are performed
to conuirm the initial positive
test. Concern over inappropriate
use of such apparently
harmless
tests such as dipstick urinalysis
have recently
been documented
(7, 8). Although
the problem of low
predictive value of testing for human immunodeficiency
virus antibody in low-prevalence
populations has received
considerable
attention (9), a recent study found that potential problems in the communication
and interpretation
of
such test results were more prevalent than were analytical
errors (10). These findings provide evidence of the need to
evaluate
all steps in the total testing process (11). A
complete review of the problems associated
with the predictive value of diagnostic tests is beyond the scope of this
policy analysis; several other publications
provide useful
discussions (12, 13).
The Regulatory ImplementatIon Process
During May 1990, the DHHS promulgated
the first set of
regulations
for the implementation
of CLIA’88
(14). These proposed regulations, known as Notice of Proposed Rule Making, are the second major step in implementing this potentially
historical
legislation.
The first
proposed
2028
CLINICALCHEMISTRY,Vol.36, No. 12, 1990
step was the establishment
of the so-called self-implementing provisions of CLIA’88 as part of the March 14, 1990,
Final Rule governing Medicare
Laboratory Standards
(15,
p. 9539). Although
the current
efforts to implement
CLIA’88 do not clearly articulate
a total testing model for
laboratory regulation,
we believe they do not foreclose by
any means the ultimate
development
of a total testing
process paradigm. In the government’s words, “We [DHHS]
will urge commenters
to assist us in the formulation
of a
regulatory
scheme that is test-complexity
based, rather
than locationltype
of laboratory based” (15, p. 9568). Unless
the long-term policy process proves to be totally impervious
to scientific reasoning, sooner or later protecting the patient will be translated
into sets of regulatory
incentives
designed to maximize the efficiency of the laboratory testing process, the use of the information
gained, and the
analytical
accuracy of the laboratory
results. Unfortunately, whereas the initial regulations
required 18 months
to develop and publish, only a 120-day comment period was
provided on these initial regulations,
definitely not enough
time to develop data or a scientific consensus based on a
participative
model.
In this essay we seek to provoke analysis and discussion
of the ramifications
of the goal of patient protection and a
federal requirement
for universal
laboratory
regulation,
including physician’s office laboratories.
The fact that the
total testing process approach is not a central focus at this
stage in the bureaucratic
rulemaking
process
does not
necessarily constrain the long-term evolution of this alternative laboratory policy. Despite the conscientious
efforts
of officials in the agencies charged with implementing
CLIA’88 to establish a framework in the recent Notice of
Proposed Rule Making (14), we predict that this notice will
be but a single step in the extended implementation
process
for CLIA’88.
A Brief History of Laboratory
Regulation
Proposed regulations
are subject to change. In 1979 a
proposed rule (16) was published
to establish a single,
comprehensive,
and integrated
set of technical
and scientific laboratory
standards
(17, p..’ 111-24). A “storm” of
controversy
ensued and the proposed regulations
were
withdrawn in 1980. An official postmortem
concluded that
“...the
1979 [Notice of Proposed Rule Making] failed for
want of empirical evidence
.
(17, p. 111-25).
Notwithstanding
that Section 4 of CLIA’88 requires
empirical studies (1), at this writing empirical verification
for the recent notice is also lacking (14). Furthermore,
we
suggest that this initial notice represents
a conventional
wisdom about implementing
the law, an approach and
philosophy subject to change and evolution through public
dialogue and scientific research. We believe this law has
far-reaching
impact and potential ramifications
that cannot be appreciated
within the traditional
framework
for
laboratory
standards.
Whatever
quality assurance
paradigm ultimately prevails-the
conventional approach, the
total testing process approach,
or another alternativethat paradigm
will be a result of the scientific and public
policy debate of which this article is an element.
Even if the implementation
of CUA’88 were to be restricted to only the precision and accuracy of the analytical
phase of testing in the physician’s office, this law nevertheless means the regulation of physicians’ decisions: e.g., who
can be hired to run tests, or how much training
must they
have before they can run tests. For this law to result in the
.
.“
regulation of laboratory testing on the basis of its impact on
medical care is a difference of degree, not one of philosophy.
Thus we consider CLIA’88 a landmark
in government
oversight of medicine, the full implications of which need to
be thoroughly examined and discussed.
independent laboratories (24). Both the Senate and House
versions of this bill (25) proposed the establishment
of a
single set of national standards
for clinical laboratories.
Although CLIA’77 proposed to exempt physician’s office
laboratories,
this exemption would have required
participation in a federally approved proficiency testing program.
Regulating the Practice of Laboratory
Medicine
Heretofore,
the “long arm” of federal laboratory
standards enforcement
has stopped
at the physician’s
door.
Both Medicare (18) and the Clinical Laboratory
Improvement Act of 1967 (CLIA’67) (19) exempted physicians from
regulation with respect to testing done in their own facilities for their own patients
(20, 21). This exemption
is
consistent with a free market, i.e., that government
should
regulate only where there is a market failure, if we define
physicians
as consumers.
The consequences
of market
failure in medical services is discussed by Pauly (22). If the
producer of a good or service has knowledge of an inferior
quality attribute that can be concealed from the consumer,
the government
may have a role in “protecting” the consumer. if we were to imagine that laboratory services were
being provided in an idealized market, then physicians
would have perfect knowledge of the performance
characteristics and prices of various laboratories.
They would also
know and be able to specify the level of precision, accuracy,
timeliness,
etc. desired. Each physician would then order
tests accordingly. However, in an imperfect world, physicians would have to incur very high costs to determine the
performance characteristics of laboratories; thus, arguably,
mandatory
government
performance
standards are a more
efficient way of correcting
for asymmetry
of information
between providers and consumers. This concept is applicable to the services of a nonhospital laboratory or a hospital
laboratory with respect to a physician (although we know
of no empirical evidence that proves the benefit vs cost
question one way or the other). That is, the physician
cannot readily learn about the quality assurance,
competency of personnel, maintenance
of records, etc. of a laboratory performing tests on specimens from his or her
patients. Although gross errors that are obviously inconsistent with clinical findings may be readily apparent, less
dramatic errors compromising
test quality would be opaque
to the practicing physician.
CUA’88 extends this philosophy of regulation. CLIA’88
dramatically
redefines the consumer
being protected, who
is now the patient. As a result, the regulatory
scope is
expanded
to universal
coverage of all laboratory
testing,
including that done in doctors’ offices, direct-to-consumer
testing such as is done in shopping malls, and even testing
in government-owned
laboratories;
tests done by the consumer at home are not within the scope of this regulation.
Did the drafters of the legislation anticipate the far-reaching potential of this consumer-protection
legislation, especially for the practice of medicine? Perhaps not; in which
case, the CLIA’88 may be explained as technical naivete.
On the other hand, perhaps they did, which makes CLIA’88
a true landmark.
Legislative History and CLIA’88
The congressional
actions in CLIA’88, as far-reaching
as
they may appear, are not without precedent. During 1976
and 1977 there was considerable
momentum
toward passage of similar federal legislation.
The bill known as
Clinical Laboratory Improvement
Act of 1977 (CLIA’77)
(23) proposed to extend federal licensure to all hospital and
CLIA’77 envisioned an essential enforcement role for states
and private
accreditation
agencies,
to be coordinated
through a federal Office of Clinical Laboratories.
Unlike CLIA’77, CLIA’88 does not create a specific
agency to coordinate the activities. This decision could
prove to be the Achilles heel of the law, which will ultimately thwart its successful implementation.
As is common knowledge, DHHS has not changed the organizational
strategy for implementing
CLIA’88 from the strategy used
for the past decade to implement the joint MedicareCLIA’67 programs. In this strategy, the primary enforcement responsibility
is vested in the Health Care Financing
Administration
arm of DHHS, with technical support to be
provided by the Public Health Service (per interagency
agreement of March 17, 1979). Even if Congress believes
that this bifurcated
approach is effective for the 12 000
hospital and nonhospital
interstate
laboratories being regulated under the aforementioned
programs, CLIA’88 presents enormous technical challenges that could easily exceed the inherent limitation of this strategy.
Like CUA’88, the earlier proposal would have required
the Secretary of DHHS to conduct studies and report to
Congress regarding the relationship
of laboratory
personnel and quality. While quite progressive in extending the
federal role, the 1977 bill’s explicit exemption of physician’s office laboratories
would have assured only a limited
role for the government
in that arena. CLIA’77 passed the
Senate but died without a vote in the House of Representatives (24).
For the ensuing decade, congressional
discourse on clinical laboratory testing was diverted from quality issues to
financial questions.
Such issues as physician
markups,
laboratory
fee schedules,
and prospective
payment took
center stage. In 1987, Congress once again addressed the
perceived problems of laboratory quality in Section 4064(e)
of the Budget Reconciliation
Act (26) that was to have
taken effect January
1, 1990, and would have extended
Medicare laboratory
regulations
to any physician’s office
laboratory completing more than 5000 tests per year for
Medicare patients. Recognizing that CLIA’88 would eventually regulate these laboratories,
Congress
provision in the 1989 Budget Reconciliation
Raising Laboratory
repealed
Act.
this
Quality on the Congressional Agenda
Media stories highlighting
laboratory quality problems
in the areas of cervical cytology screening, AIDS testing,
and drug screening served to arouse Congressional
concern
and initiated the movement toward enactment
(27-31). Hearings in both the House
cluded representatives
of professional
of CLIA’88
and the Senate inand trade associa-
tions expressing their interest in protecting “quality” (not
to mention their turf), particularly
in cervical cytology
screening and physician’s office laboratories (32-34). In
addition, individuals who were the victims of poor laboratory testing provided vivid anecdotes of laboratory testing
problems. There was little empirical
evidence to guide
Congress’s policymaking,
especially
with regard to the
qualifications
of technical personnel, supervisors,
and laboratory
directors.
Laboratory representatives
also laCLINICAL CHEMISTRY, Vol. 36, No. 12, 1990 2029
the growth of physician’s office laboratory technologies and increasing test volumes, claiming that laboratories were at a competitive
disadvantage
because
of
regulation. Typical of the comments by laboratory industry
professionals
was that of the spokesperson
for the American Clinical Laboratory Association, who requested a level
playing field, asking Congress to
fix the system to
enable that [sic] all participants
perform at the same high
levels.” (35). The legislation was signed by President Reamented
...
gan on October 31, 1988.
Points of CLIA’SB Relevant to the Total Testing Process
EmpiricalEvidence
Necessitated
by the relative dearth of empirically based
studies to guide policymaking,
research
studies are required under CLIA’88. The need to study the connection of
laboratory
testing and health outcome is made explicit
because the studies were to address the “extent to which
deviation from accurate and reliable results affect diagnosis, treatment,
and ultimately patient health.” Other studies mandated include (a) validity, reliabililty,
and accuracy
of proficiency testing; (b) the correlation between standards
for personnel employed and the accuracy and reliability of
results of the tests performed by laboratories
subject to
standards; (c) the relationship
of internal quality assurance
and quality control on results; and (d) the relationship of
problems in diagnosis and treatment of patients caused by
inaccurate laboratory test results (1, Sec. 4). This last study
area would seek to understand
laboratory
errors in each
component of the clinical testing process, including com-
munication
between the attending physician and the labo-
ratory, selection of tests, and limits (normal ranges) of
tests. In addition, specimen collection and transport, analysis, and reporting of results would be subject to study. Our
conclusion from the requirement
for empirical studies is
that the intent of Congress appears to go well beyond the
traditional bounds of laboratory regulation to extend over
the total testing process-further
supporting our alternative view of how the regulatory
system should evolve.
The Secretary
of DHHS is to “act through” the Public
Health Service. Congress made note of the past involvement in laboratory improvement
of the Centers for Disease
Control and the potential
role of the “health outcomes
research” of the Agency for Health Policy and Research, as
well as the expertise of the National Institutes of Health in
diagnosis and treatment
of patients. The committee report
states that the
Committee intends that the Secretary
make use of the expertise of such agencies in conducting
the studies.” Although these studies were to have been
completed by May of 1990, no funds for their performance
were authorized in CLIA’88 (1, p. 41). Because CLIA’88 is
to be self-supporting,
we can infer that funding for studies
could be generated
from certification
fees, enabling the
government
to proceed with these essential studies.
“.
..
Universal Coverage
CLIA’88 applies federal
outside
the legitimate
standards
to any testing site
application
of home test kits. All
clinical laboratories are required to have a certificate
issued by the federal government [1, Sec. 353(b)]. Laboratories are defined to be any facility for examining materials
from the human body for the purpose of providing information for diagnosis, prevention,
or treatment
of any disease
or for assessing the health of human beings [1, Sec.353(a)].
Thus the statute defines clinical laboratories in the broad2030 CLINICAL CHEMISTRY, Vol. 36, No. 12, 1990
est possible manner. Not only are the traditional
hospital
and independent laboratories covered, but so are rural
health clinics, physicians’ offices, nursing homes, ambulatory surgical centers, renal dialysis centers, the everpopular mobile cholesterol-screening
facilities, and arguably all military
and Department
of Veterans
Affairs
facilities as well as state and local public health laboratories. Testing specimens for insurability and employment
purposes is covered; testing for forensic purposes is not.
Nevertheless,
the statute empowers the Secretary of DHHS
to oversee the testing in any place that a patient’s specimen
is tested for health purposes.
Site Neutrality
Under CLIA’88, two laboratories
performing
the same
tests would be subject to the same standards,
the Energy
and Commerce Committee
wrote, “notwithstanding
that
one was located in a physician’s office and the other was
located in a different setting” (2, p. 27). A laboratory test
would be subject to the same personnel standards whether
the test was performed in a hospital or at an independent
laboratory or a physician’s office. The Senate expressed
a
similar view:
While the Committee does not intend to discourage physicians
from performing tests in their own laboratories, it does intend that
there be assurances of high quality in these laboratories.
Thus,
the committee proposes that physician-office laboratories be regulated no differently than other laboratories performing the same
categories of tests [3, p.21].
. . .
We observe that this concept of site neutrality
is both
logical from the perspective of the official rhetoric, which we assert is patient protection, and, at the same
eminently
time, exquisitely difficult from the perspective of the actual
practice of regulatory implementation. Because risk to the
patient is not merely a function of the analyte, e.g., glucose,
but also of the analytical
method used, the instrument,
the
clinical question/interpretation
(sensitivity, specificity, and
prevalence), and very importantly,
the subsequent medical
intervention (e.g., further testing vs treatment with a
powerful
drug), a “test” will need to be carefully defined.
An example
of careful definition
able automated
glucose methods
laboratories
would be a list of acceptfor diabetes screening in
with minimum personnel criteria.
Variable Standards
Some may view CLIA’88 as a political triumph of certain
laboratory guilds over other providers of laboratory
services, such as physicians in their offices. We suggest that an
alternative conclusion is reasonable. The concept of consumer/patient protection means that the consumer’s pocketbook, convenience, and access to laboratory tests in remote
areas-what
testing-are
might be termed the total economic utility of
to be protected. CLIA’88 supports this total-
utility concept by opting for the administratively
more
difficult strategy of variable standards. The drafters
of the
legislation recognized that a variety of laboratory providers
were desirable, observing that one set of standards
would
be
unnecessarily
onerous for some and inadequately
rigorous for others” (2, p. 27). Requiring the highest possible precision and accuracy in all situations would have the
perverse
effect of limiting
access to testing outright
or
creating avoidable delays in providing laboratory information to physicians. Either case adversely affects the information content of medical decisions. The strategy of variable
standards addresses this problem. As the Senate observed:
“...
At the same tune the Committee does not wish to impose
unnecessarily rigorous regulatory burdens on those laboratories,
whether they are in doctors office or other sites, which perform
only certain categories of tests [3, p.21].
Standards
are to be based on performance
considerations,
not administrative
convenience, e.g., the location of the
laboratory.
CLIA’88 not only addresses
the problem of
variable standards
at the level of official rhetoric but also
provides logical guidance-”considerations”
upon which
the law is to be implemented.
There are three sets of these
considerations:
one for “waiver” tests, one for personnel
qualifications, and another for other laboratory standards.
Exempted or waivered tests. In lieu of a “certificate,” a
laboratory (by the CLIA’88 universal definition) may operate under a “certificate of waiver,” provided it performs
only simple laboratory examinations
that “have an insignificant risk of an erroneous
result” [1, Sec. 353(a)(3)],
including:
tests approved for home use, simple methods
with negligible likelthood of error, and tests that pose no
“reasonable
risk of harm” [1, Sec.353(a)(3)(A
thru C)]. A
laboratory operating under a “certificate of waiver” would
be exempt from proficiency
testing,
personnel
and quality
assurance standards,
and inspection.
Clearly, routine tests exist where mandatory
federal
oversight would be illogical; for example, a stool guaiac
test performed by a physician as part of a routine physical.
Nevertheless, who is prepared to say that the stool guaiac
test is in and of itself a “foolproof” test or that there is no
result? On the one hand,
a false-negative
result may delay diagnosis and treatment
of gastrointestinal
bleeding; on the other, a false-positive
result would expose the patient to additional testing costs
and perhaps
radiographic
or endoscopic
examinations.
Thus the concept of an exempted
or waivered test, although logical from the perspective
of the policy goal
(which we suggest is to maximize
the benefit to the
patient), is incompatible
with the current regulatory
apparatus. We predict that ultimately
the list of waivered
tests will need to be defined not as tests, e.g., stool guaiac,
but rather as medical-practice
laboratory procedures, e.g.,
a stool guaiac test performed by a physician as part of a
routine physical.
Viewed from a total testing perspective, CUA’88 is, we
suggest, essentially
a tool for risk management.
The risk of
a test is seldom inherent in the test itself, but rather is a
function of the context in which the test is being used to
provide information for medical decision making. A logical
consequence
of this line of reasoning
is that defining
procedures
for waiver must be a process of identifying
patient indications,
clinical questions,
test methods and
instruments,
personnel,
and consequences of results for
which the risk is so low that further regulatory oversight
would have an inconsequential
effect.
Personnel
qualifications.
CLIA’88 requires
that
only
qualified personnel be used to perform or supervise testing.
The law states that a laboratory
may
risk associated
with
an erroneous
use only personnel meeting such qualifications as the Secretary may establish for the direction, supervision, and performance
of examinations and procedures within the laboratory, which
qualifications
shall take into consideration competency, training,
experience, job performance and education and which qualifications shall, as appropriate, be different on the basis of the type of
examinations
and procedures being performed by the laboratory
and the risks and consequences of erroneous results associated
with such examinations and procedures. [1, Sec. 353U)(1)(C)].
Education, measured by academic credentials of personnel, has been widely used by the laboratory community
and
regulators as a proxy for a predictor of good performance;
however, the empirical
basis for this practice is mixed
(36-38). The economic incentive for an emphasis on aca-
demic credentials is as old as economics itself. Limiting the
available labor pool to individuals with academic degrees,
given the law of supply and demand, results in higher
wages for laboratory specialists than would be the case if
the supply of eligible workers included qualified individuals whose formal laboratory training was not obtained in
an academic
degree-granting
program.
Higher
wages
aside, tight labor markets also produce undesirable
consequences for clinical laboratory workers and managers, e.g.,
longer hours and higher stress. This may be a particular
problem in public health-care systems, which may not have
funds to raise salaries or attract qualified personnel in a
tight labor market.
From the perspective of protecting
the patient/consumer,
a tight labor market can mean high prices and no improvement in service. Consistent
with the patient/consumerprotection
paradigm,
CLIA’88 eschews the administratively convenient proxy (reliance on formal education) and
requires comprehensive
personnel standards based on consideration of competency, experience, job performance,
and
training, with formal education being but one component.
In the words of Congress:
The Committee’s main concern is that the person in charge of
the laboratory be knowledgeable about laboratory science, commensurate with the range and sophistication of the testing being
performed, and be capable of maintaining quality service. The
Secretary would need to determine what specific qualifications are
both necessary and sufficient to satisfy these objectives. The
Committee does not believe it is essential that every laboratory be
under the direction a pathologist or Ph.D. scientist nor does it
believe that an M.D. degree alone is necessarily sufficient to assure
competence to manage a laboratory. The Committee also urges the
Secretary to explore various means for directors to establish [sic]
the requisite level of competence, in addition to formal degree
training, such as passing a specific test or participating in continuing education or other similar alternatives.
The Committee is also concerned that supervisory and other
personnel have appropriate qualifications to assure accurate and
reliable tests. The bill would not, however, specifically direct the
Secretary to establish particular personnel standards to technicians and technologists. 1f however, the Secretary did decide to
adopt such standards, the bill would direct him not to rely
exclusively
on academic
degrees
to establish
competence,
but to
take into account as well such other factors as experience, job
performance, other avenues of training, and qualification by examination [2, p.28].
In addition, personnel qualifications
shall appropriately
differ on the basis of the difficulty of the testing being
performed and the risk and consequences
of erroneous
results [1, Sec. 353(f)(1)(C)].
Other considerations
for laboratory standards. CLIA’88
directs that laboratory standards should be developed with
consideration
of the methodologies
used, the judgement
needed, the interpretation
required, the difficulty of calculations, quality control requirements
of the instruments,
and
“such other factors” as are relevant.
Another obvious factor
would be risk to the patient from an erroneous result [1, Sec.
353(f)(2)(A thru G)]. These “considerations”
form what is
now commonly referred to as the “complexity model” of
laboratory regulation. The three-tier complexity model provides the flexibly for CLIA’88 to resemble
the multi-level
CLINICAL CHEMISTRY, Vol.36, No. 12, 1990 2031
models now used by certain states to regulate laboratories
(39).
How long it will take to develop and reach scientific
consensus on meaningful levels of complexity is a matter of
speculation. The goal-to provide both highly accurate and
widely available testing, the public’s expectation-cannot
be achieved, we believe, without a workable and technically defensible model of complexity. Therefore, the admin-
istrative process will necessarily require the input of experts from both clinical practice and the analytical disciplines of laboratory medicine. Notwithstanding
the fact
that the government
has proposed a rule (14) generally
regarded as a complexity
model, the technical advisory
process referred to in the regulation is only proposed; it is
not in place. Therefore, the technically defensible complexity model remains to be developed. We believe a substantial
window of opportunity
exists for the industry to develop a
complexity model in the context of test use and patient
protection.
Special Requirements for Cytology
The major quality problem identified in the Congressional
Hearings
[Papanicolaou
this discipline
of 1988 was cervical cytology screening
smears]. The added attention
given
(Pap)
is consistent
with the patient/consumer-
protection paradigm
we are advancing
in this analysis.
There are few better examples of where the patient is more
vulnerable to the consequences
of a falsely negative result
than in the Pap test for cervical cancer. Furthermore,
little,
if any, information
redundancy
is provided by the early
clinical indications of cervical cancer that would enable the
physician to compensate for poor laboratory performance.
Cytology is unique within CLIA’88 in that it is the only
laboratory
specialty for which Congress wrote specialtyspecific standards
[1, Sec. 353(0(4)(B)1. The law requires
the Secretary to establish guidelines for personnel, working conditions, quality control, and the maximum workload
for cytotechnologists doing screening, including the maximum number of slides to be screened by a cytotechnologist
in a 24-h period, criteria for rescreening cytological preparations, record-keeping to document compliance, a requirement that screening be performed on the premises, periodic
evaluation of each individual’s proficiency (not just the
laboratory’s), and unannounced
on-site proficiency testing.
The law also provides for standards for the adequacy of
cytology slides and requires notification of the physician
when an inadequate slide is submitted [1, Sec. 352(0(4)1.
Proficiency Testing
The testing
of quality-control
specimens submitted via
by an outside agency is a long-accepted
part of the laboratory
quality assurance.
Although this
process provides only some indication of the best level of
work that a laboratory
is capable of producing, no “improvement program” would be complete without it. Within
mail to a laboratory
technical limitations, this form of external quality validation is required by all governmental and voluntary agencies concerned
tion. Programs
with laboratory
regulation
and accreditaare also available
for physician’s
office
laboratories, and participation in these programs is required by some states as the major component of physician’s office laboratory licensure (40, 41).
Proficiency testing can be viewed as a learning
mechanism
for laboratories
to learn about their
mance, their instruments,
and their methods.
2032
CLINICAL CHEMISTRY, Vol. 36, No. 12, 1990
tool, a
perforWhile
CLIA’88 does not explicitly diminish the education function of proficiency testing, the legislation demands much
more from proficiency testing than discretionary laboratory
learning. The House conference
committee wrote that it
believed that”..
proficiency testing should be the central
element in determining
a laboratory’s competence since is
purports
to measure
actual test outcomes rather than
merely gauging the potential for accurate outcomes” (2, p.
28). Recognizing
the shortcomings
of this assessment
method, the law
is designed to strengthen
and improve
proficiency testing” (2, p. 28). Among these improvements
are the requirements
that laboratories
must agree to treat
proficiency testing specimens as routine samples [1, Sec.
353(d)(1)(E)1 and the authority
for the government
to
conduct on-site proficiency testing to assure that samples
are handled without special treatment. Proficiency test
results
are to be available
to the public [1, Sec.
353(f)(3)(D)], a requirement that is significant in the context of the implicit philosophy of the regulation we ascribe
to CLIA’88. If consumers of laboratory tests-physicians,
patients,
health maintenance
operations,
or whoeverwant better precision
than the minimum
standards,
CLIA’88 guarantees them access to information that would
assist them in making this choice. Although we looked for
language suggesting that drafters of CLIA’88 appreciated
.
“.
.
.
the desirability of flexibility in the application
ciency testing to various levels of test complexity
of profi-
and risk
to patients, we did not find any explicit recognition
of the
fact that proficiency testing deals only with the analytical
phase of the testing process. However, the Section 4 studies
do require an assessment
of the “validity”
of proficiency
testing [1, Sec. 4(a)(1)]. The proposed regulations,
perhaps
in recognition of the difficulty in administering
proficiency
testing to all laboratories who might benefit from it, do not
require proficiency testing for exempted tests, even though
such procedures are available for some of these tests (e.g.,
Gram stain, screening for anti-streptolysin
0, urine pregnancy testing, microhematocrit)
(14, p. 20918). This is
quite ironic, if in fact the lowest levels of quality control are
found in physician’s office laboratories.
laboratories
that would benefit
proficiency testing.
It may be these
the most from mandated
Accreditation by Nonfederal Organizations
Nonprofit accreditation.
Congress provides an opportunity for professional organizations
to assist in enforcement
of CLIA’88 [1, Sec. 353(o)]. The policy of exempting laboratories from licensure if they meet professional accreditation
standards equal to federal standards
was established
for
interstate laboratories
under CLIA’67. The Inspection and
Accreditation
Program of the College of American Pathologists (CAP) was recognized in CLIA’67, allowing CAPaccredited
interstate
laboratories
to be exempted
from
interstate licensure. Medicare’s policy of providing deemed
status to hospitals accredited by the Joint Commission on
Accreditation
of Healthcare
Organizations
or to the American Osteopathic Association provides an analogous example for purposes of Medicare payment. CLIA’88 provides
that the Secretary of DHIIS shall set standards for accreditation organizations.
Those laboratories
accredited by approved bodies will be exempt from federal certification
[1,
Sec. 353(o)]. The recently established Commission on Office
Laboratory Assessment-a
joint venture of the CAP, the
American Medical Association,
the American Academy of
Family Practice, and the American
Society of Internal
Medicine-was
established with the intent of setting standards and accrediting all sizes of physician’s office laboratories (42). The accreditation
route may prove more acceptable, especially to physicians’ offices, but it will not exempt
these laboratories
from oversight, because the Commission
on Office Laboratory
on-site verification
iting body [1, Sec.
not been the norm.
deemed status for
Assessment
surveys
will have to conduct
to become an approved
accred-
(e)(2)(i)]. Wide use of accreditation
has
For example, CAP has never been given
hospital laboratories
under Medicare,
even though it is recognized by the Joint Commission on
Accreditation
of Healthcare
Organizations
and enjoys that
status under the CLIA’67 statute.
The role of private organizations
in laboratory oversight
could be to pioneer laboratory
performance
assurance
based on the total testing process. However, accreditation
by private organizations
will be hampered by skepticism
over whether private operations can be trusted to regulate
as stringently as the government. One criticism of accreditation has been the lack of information
on performance
and inspection deficiencies of accredited laboratories. Federal oversight of deemed organizations has also been criticized for this lack of detailed
information
on the perforproviders
(43). CLIA’88 answers
this problem by requiring that a
laboratory
must authorize
records of accreditation
inspection as well as”..
other information as the Secretary may
require
to be submitted
to the appropriate
federal
agency [1, Sec. 353(e)(1)(B)]. CLIA’88 requires that laboratories participating
in accreditation
programs do not slip
below federal minimum
performance
levels; however, no
limits are placed on how restrictive
or how expensive
mance and inspection deficiencies of accredited
.
.
.
.“
private accreditation
can become.
The role of states under CLIA’88. States are treated
similarly to the professional
accrediting
bodies: laboratories licensed in states deemed to have licensure programs
equal to or more stringent than federal standards will be
exempt from federal certification. CLIA’88 gives the Secretary of DHHS authority
to exempt laboratories
in states
with requirements
equal to or more stringent
than
CLIA’88, and also expressly allows states to enforce their
own laboratory regulations.
Any state wanting its laboratory industry to exceed federal minimum standards is free
to do so [1, Sec. 353(p)(l)]. Thus a state or group of states
could foster total testing process concepts ahead of the
federal bureaucracy.
Currently,
only New York’s laboratory standards have been accepted by federal authorities as
equivalent
to the CLLA’67 regulations.
In a study of physician’s office laboratories,
the DHHS Office of the Inspector
General identified fewer than a dozen states with relatively
complete programs for such laboratories
(41). Pennsylvania
is often viewed as a prototype because of its multi-level
system for classifying laboratories
(39). Unlike CLIA’67,
which emphasized
the role of state-federal
partnerships,
CLIA’88 is silent on this thorny issue. No funds or other
incentives for improving state laboratory improvement
353(m)]. There is no history on which to judge whether a
pay-as-you-go
program will be more or less constrained
than would a program that taps the U.S. Treasury directly.
Development
of alternative
regulatory
strategies
and research will be an overhead expense to the program. If the
user fee schedule raises more funds than would otherwise
be the case, the development
of a new paradigm
could
happen sooner-but
this is speculation
(44).
Regulation, Complexity, and the Total Testing
Approach
Regulating based on complexity. While conceptually
simple, regulating based upon the complexity of testing will
prove to be the biggest challenge to policy makers, physicians, and the laboratory industry.
Specific considerations
in CLIA’88 [1, Sec. 353(0(2)] and statements
in committee
reports make clear that laboratory location can not be used
as a proxy for text complexity (2, p. 27). However, developing a scientifically valid consensus on what should constitute noncomplex tests suitable for exemption will prove
difficult. Professional organizations
have provided the Secretary of DHHS with lists of tests proposed for exemption
(45,46) but these lists appear to be common tests performed
in physician’s office laboratories.
Example of Using the Total Testing Approach. If, as we
suggest, CLIA’88 reflects the expectation
that utility of
laboratory testing and patient care management
should be
optimized,
clinical testing performance
standards
must
either address the manner in which tests are used by the
physician
or fail to respond to that expectation.
Applying
the total testing process philosophy requires that the type
of medical practice and the purpose of a test must be taken
into account when evaluating what level of quality assurance is needed. A family practitioner ordering glucose for
screening a population with a low prevalence of diabetes
may not be able to use the same glucose methodology as an
endocrinologist who is monitoring insulin therapy in a
practice involving young active diabetics. The information
needs of these two situations
are different and must be
taken into consideration
if laboratory data are to be used in
a medically sound way.
What is the appropriate level of quality for determining
an analyte? We suggest that the question is unanswerable
except in the context of the clinical question and the
follow-up medical interventions, because the clinical question and its ramifications (not the laboratory test itself) are
what determine
the risk to the patient. Is it possible for a
physician to perform a test in his or her office, with the
sensitivity and specificity of an office-based analytical system (operated by trained office personnel) for certain clinical questions, and yet need to use a “reference” laboratory
to achieve the same level of confidence for another clinical
question
involving
the same analyte?
If the answer is yes,
Pay As You Go
as we suggest, the next question is whether the laboratory
regulatory process can manage an issue of such subtlety. In
our opinion, the answer to this question is currently “no”
but in the long term must also be “yes.” If the regulatory
system categorically exempts a long list of analytes, it will
be guilty of a sham as far as assuring the reliability of
Unlike the current regulatory
systems, Congress provided that CLIA’88 would be totally supported from “user”
fees. Certified laboratories would pay a fee that would cover
not only inspection
costs but also an appropriate
share of
the total administrative
costs. Certificate
of waiver (exempt) laboratories
would pay only small fees [1, Sec.
office-based testing. On the other hand, if the process
restricts too many common analytes from office testing,
patient convenience
and the quality of care provided by
physicians will be compromised and costs will be increased
(47). We believe either extreme will be a misinterpretation
of the legislation.
capabilities
are included.
CLINICAL CHEMISTRY, Vol. 36, No. 12, 1990
2033
ConcludingRemarks
2. US House of Representatives.
Our premise is that diagnostic tests must be examined in
the context of the total laboratory testing process. Improvement of the quality of testing is dependent not simply on
the analytical component of testing, but also on the use of
appropriate
decision rules regarding which test to select
and how to use the results in the context of patient care.
Many of the real gains in laboratory “quality” and laboratory “effectiveness”
must come in the examining
room or at
the patient’s bedside, as well as at the laboratory
bench.
Identifying analytes and situations where the effectiveness
of laboratory
testing can be improved is a fundamental
challenge for clinicians, researchers,
laboratory professionals, and policymakers.
We believe that this is the real
challenge of the empirical studies required by CLIA’88. We
have discussed the congruence of the major provisions of
this legislation in the context of protecting the patient and
maximizing
the utility of the total testing process. Our
policy analysis has attempted
not only to inform readers of
the salient features of CLIA’88 but also to raise the level of
debate on this legislation
and its implementing
regulations. Instead of arguing over the lists of exempted tests for
physicians’ offices, personnel requirements,
proficiency
testing, and the other issues of detailed regulation that
characterized
previous discussions
of CLIA’88 and
other federal laboratory regulation programs
(15, 17, 38),
what is needed is serious discussion of the more fundamenhave
tal issues of quality patient care. We will have achieved a
worthwhile
objective if we have prompted
commentary,
pro
or con, on the desirability of the total testing process approach to
performed
laboratory
shift from
laboratory
standards.
However, we will have
a true service if we have alerted both clinical and
medicine
to the implications of the government’s
attempting to protect physicians from low-quality
laboratory testing to attempting
to protect patients from
low-quality outcomes associated with testing errors.
We have suggested
an alternative
of how laboratory
testing,
not just laboratories,
should be evaluated.
Our
analysis of CLIA’88 and its accompanying
committee reports have lead us to the conclusion that the fundamental
intent of this law is patient
protection
and improved
effectiveness of testing. This goal can be realized only by
in-depth
examination
of the total laboratory testing proset of activities at the very core of the practice of
medicine. From this perspective,
it is equally important
cess-a
that
the
regulatory
program
now
being
implemented
be
concerned with improving physicians’ selection and use of
tests as well as the analytical accuracy and precision of the
laboratory test itself. Ultimately,
to better serve patients,
we must improve the effectiveness of the physician’s use of
the laboratory, both in the office and in the hospital setting.
Failure to recognize this challenge is likely to lead to yet
another regulatory program that does little to improve the
quality and effectiveness of medical care.
Research for this analysis was initiated during sabbatical leave
(K.M.P.) while a health policy consultant
with the Assistant
Secretary for Planning and Evaluation (ASPE), DHHS. H.C.H. is
currently on leave from ASPE serving as Science Policy Fellow
with the U.S. Congress. We thank Dr. Ronald Cada, Dr. Stanley
Edinger, Ms. Karen Ferran, Ms. Diane Francis, and several other
reviewers for their helpful suggestions.
References
1. Public Law 100-578. Clinical Laboratory Improvement Amendments of 1988. Stat 42 USC 201. H.R. 5471. Oct. 31, 1988.
2034 CLINICAL CHEMISTRY, Vol. 36, No. 12, 1990
Committee
on Energy
and Com-
merce. Clinical Laboratory Improvement Amendments of 1988.
Report to accompany H.R 5150. Report 100-899.
3. US Senate. Committee on Labor and Human Resources. Medical Laboratory Improvement
Act of 1988. Report to accompany
S.2477. Report 100-561. September 26, 1988.
4. Inhorn SL, Addison By, eds. Proc, 1986 Institute on Critical
Issues in Health Laboratory Practice. Managing the quality of
laboratory test results in a changing health care environment.
Atlanta, GA: Centers for Disease Control, 1987.
5. Dorsey DB. Evolving concepts of quality in laboratory practice:
a historical overview of quality assurance in clinical laboratories.
Arch Pathol Lab Med 1989;113:1329-34.
6. Watts NB. Medical relevance of laboratory tests: a clinical
perspective. Arch Pathol Lab Med 1988;112:379-82.
7. Woolhandler S, Pels RJ, Bor DH, Himmelstein
DU, Lawrence
RS. Dipstick urinalysis screening of asymptomatic adults for
urinary tract disorders. I. Hematuria and proteinuria. J Am Med
Assoc 1989;262:1215-9.
8. US Preventive
Services Task Force. Recommendations on
screening for hematuria and proteinuria
by dipstick urinalysis. J
Am Med Assoc 1989;262:1214.
9. Meyer KB, Pauker SA. Screening for HIV: can we afford the
false positive rate? N Engl J Med 1987;317:238-41.
10. Benenson AS, Peddecord KM, Hofkerr LK, Ascher MS, Taylor
RN, Hearn TL. Reporting the results of human immunodeficiency
virus testing. J Am Med Assoc 1989;262:3435-8.
11. Drotman DP, Valdeserri RO. Are human immunodeficiency
virus test reports clear to clinicians? Editorial]. J Am Med Assoc
1989;262:3465-6.
12. Galen RS, Gambino SR. Beyond
normality:
the predictive
value and efficiency of medical diagnosis. New York: John Wiley &
Sons, 1975.
13. Sox HC, Blatt MA, Higgins MC, Marton KI. Medical decision
making. Boston: Butterworths, 1988.
14. US Department of Health and Human Services. Health Care
Financing Administration.
Medicare, Medicaid and CLIA programs; regulations
implementing
the Clinical Laboratory Improvement Amendments of 1988; proposed rule. Fed Reg 1990;55:
20896-959.
15. US Department
of Health and Human Services. Final rule
with comment period. Medicare, Medicaid and CLIA programs;
revision of the laboratory regulations for the Medicare, Medicaid
and Clincal Laboratories Improvement Act of 1967 programs. Fed
Reg 1990;55:9534-610.
16. Fed Reg 1979;44:58912-9111.
17. Macro Systems. Final report on assessment of clinical laboratory regulations.
Office of Assistant Secretary for Planning and
Evaluation.
NTIS No.PB86210441AS(793)87450,
Washington,
DC: DHHS, 1986.
18. US Code of Federal Regulations, Title 20, Chapter III, Subpart
M, Part 405. Medicare Independent Laboratories. Conditions for
participation. 1968 et seq.
19. Public Law 90-174. Clinical Laboratories Improvement Act of
1967. Code of Federal Regulations, Title 42, Part 72, 1967.
20. Edinger SE. The Medicare and Clinical Laboratories
Act of
1967; proficiency testing requirements and its relationship
to the
private sector. Arch Pathol Lab Med 1988;112:357-62.
21. Hammond HC. Federal regulation of clinical laboratories.
Arch Pathol Lab Med 1989;112:363-7.
22. Pauly MV. Is medical care different? In: Luke RI), Bauer JC,
eds. Issues in health economics. Rockville, MD: Aspen Systems
Corp., 1982:3-23.
23. US Congress. Senate Committee on Human Resources, Subcommittee on Health and Scientific Research. Hearing for S.705,
Clinical Laboratory Improvement Act of 1977. March 29-30, 1977:
765 pp.
24. Schaeffer M. Perspectives
and prospectives. In: Schaeffer M,
ed. Federal legislation
Hall, 1981.
and the clinical laboratory. Boston: GK
25. US Congress. House of Representatives. Committee on Interstate and Foreign Commerce, Subcommittee on Health and the
Environment.
Hearing for H.R. 6221, Clinical Laboratory
Improvement Act of 1977. June 14-15, 1977. Serial No. 95-17. 642
pp.
26. Public Law 100-203. Section 4064 Omnibus Budget Reconcil-
iation Act of 1987. December 22, 1987.
27. Bogdanich W. False negative: medical labs, trusted as largely
error-free are far from infallible-haste,
misuse of equipment,
specimen mix-ups afflict even best labs at times-regulation:
weak
and spotty. Wall St J, Feb. 2, 1987.
28. Bogdanich W. Risk factor: inaccuracy in testing cholesterol
hampers war on heart disease-some
diagnoses are skewed by
glitches such as use of ill-calibrated lab gear-missing the mark by
100 percent. Wall St J, Feb. 3, 1987.
29. Bogdarnch W. Lax laboratories: the Pap test misses much
cervical cancer through labs’ errors-cut-rate
“Pap mills” process
slides using screeners with incentives to rush-misplaced
sense of
security? Wall St J, November 2, 1987.
30. US Congress. House of Representatives.
Committee on Small
Business, Subcommittee on Regulation and Business Opportunities. Quality AIDS testing. Hearings, Oct. 19, 1987. Serial No.
100-32. 118 pp.
31. Ashton P. Testimony of the American Society for Cytotechnology before the U.S. Congress. Senate Committee on Human
Resources, Cancer detection in women. Hearing January 18, 1988.
No. 100-529, pp 167-224.
32. US Congress. House of Representatives.
Committee on Small
Business, Subcommittee
on Regulation and Business Opportunities. Deadly mistakes: are laboratory results reliable? Hearings,
March 2 and May 3, 1988. No. 100-43. 242 pp.
33. US Congress. Senate Committee on Government Affairs. Subcommittee
on Oversight
of Government
Management.
Health
Care Financing Administration’s
management
of medical laboratories. Hearings, March 23-24, 1988. No. 100-765. 796 pp.
34. US Congress. House of Representatives.
Committee on Energy
and Commerce, Subcommittee on Oversight and Investigations.
Clinical laboratories. Hearings, March 10 and July 27, 1988. No.
100-146. 552 pp.
35. Groome H. Testimony of the American Clinical Laboratory
Association before the US Congress. Op. cit. (ref. 33):119-120.
36. Peddecord KM, Cads RL. Clinical laboratory proficiency test
performance:
its relationship to environmental,
structural
and
process variables. Am J Clin Pathol 1980;73:380-5.
37. Kenney ML. Laboratory performance and director qualifications. Clin Chem 1987;33:725-38.
38. Peddecord KM. The regulatory model for clinical laboratories:
an empirical evaluation.
Clin Chem 1989;35:691-700.
39. Bartola J. Testimony before the US Congress. Op. cit. (ref.
33):248-78.
40. Crawley R, Belsey R, Brock D, Baer DM. Regulation of
physicians’
offices: the Idaho experience.
J Am Med Assoc 1988;
255:374-82.
and Inspections. Office of Inspector General,
US Department of Health and Human Services. Quality assurance
in physician office labs [sic]. March 1989. OAI-05-88-00330.
21 pp.
42. Goldsmith MF. Federal proficiency testing requirements
set to
start for physician office laboratories
[News]. J Am Med Assoc
41. Office of Analysis
1989;262:2355-6.
43. Hays LB. Testimony
before the US Congress.
Op cit. (ref.
33):85-102.
44. US Department
of Health and Human Services. Health Care
Financing Administration.
Proposed rule, laboratory certification
fees under CLIA’88. Fed Reg 1990;55:31758-73.
45. Aukerman GD. Chairman, Board of Directors of the American
Academy of Family Physicians. Letter to Louis W. Sullivan, M.D.,
Secretary, US Department of Health and Human Services, August
29, 1989.
46. Keffer JH. President, American Society of Clinical Pathologists. Letter to T.G. Morford, Director of the Health Standards and
Quality Bureau, Health Care Financing Administration, US Department of Health and Human Services, December 15, 1988.
47. Nelson AR. President-elect, American Medical Association.
Summary Statement to the Committee on Health and the Environment, Committee
on Energy and Commerce, US House of
Representatives,
July 6, 1988.
CLINICALCHEMISTRY, Vol. 36, No. 12, 1990 2035