Single-Use Medical Devices: Little Available Evidence of Harm From Reuse,
but Oversight Warranted (Letter Report, 06/20/2000, GAO/HEHS-00-123).

Pursuant to a congressional request, GAO reviewed the practice of
single-use device (SUD) reprocessing in the United States, focusing on
the: (1) extent of SUD reprocessing; (2) health risks associated SUD
with reprocessing; (3) cost savings from reprocessing; and (4) Food and
Drug Administration's (FDA) oversight of SUD reprocessing.

GAO noted that: (1) while some health care institutions have chosen to
reprocess and reuse some kinds of SUDs, accurate and comprehensive
information about the number of facilities that use reprocessed SUDs and
the types of SUDs that are reprocessed is not available; (2) surveys by
professional associations and other groups have found that approximately
20 to 30 percent of American hospitals reported that they reuse at least
one type of SUD and that at least one-third of the hospitals that do so
contract with third-party reprocessing companies; (3) most hospitals
using reprocessed SUDs reuse only a few types of devices; (4) it is
likely that some hospitals do not report their use of reprocessed SUDs,
and the estimates do not fully include ambulatory surgery centers or
physicians' practices that also may reuse SUDs; (5) although SUD
reprocessing does pose theoretical health risks, clinical evidence shows
that certain devices can be reprocessed safely; (6) some infection
control experts told GAO that the careful reprocessing of appropriate
SUDs has not been demonstrated to be a public health risk; (7) GAO found
that several reports of patient adverse events allegedly due to SUD
reprocessing that GAO identified were inaccurate or not relevant to the
debate; (8) however, this does not mean that SUD reprocessing is always
safe; (9) current surveillance systems almost certainly do not detect
all infections and injuries resulting from the use of reprocessed SUDs;
(10) FDA, device manufacturers, and third-party reprocessors generally
agree that many types of SUDs cannot be safely cleaned and sterilized,
and even for devices that usually can be reprocessed, some models are
impossible to clean and sterilize effectively; (11) substantial cost
savings can be achieved by reprocessing SUDs; (12) independent
reprocessing firms charge hospitals approximately one-half the price of
a new device for a reprocessed device, while the in-house cost of
reprocessing some devices can be less than 10 percent of the price of a
new device; (13) under FDA's revised framework for SUD reprocessing,
independent reprocessing firms and hospitals will have to obtain FDA's
approval before they can reprocess many devices labelled for single-use;
and (14) the revised regulatory framework will give FDA more information
about SUD reprocessing and strengthen its oversight of reprocessing.

--------------------------- Indexing Terms -----------------------------

     TITLE:  Single-Use Medical Devices: Little Available Evidence of
	     Harm From Reuse, but Oversight Warranted
      DATE:  06/20/2000
   SUBJECT:  Medical equipment
	     Consumer protection
	     Safety standards
	     Health care cost control
	     Product safety
	     Safety regulation
	     Health resources utilization

** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **

Appendix I: Scope and Methodology


Appendix II: Comparison of New ESRD Patients by Age and Primary Diagnosis,
1989, 1993, and 1997


Appendix III: Medicare Conditions of Participation for Dialysis


Appendix IV: Comments From the Health Care Financing


Table 1: Number and Percentage of Dialysis Facilities Resurveyed,
1993-99 10

Table 2: Top Five Conditions of Participation Identified as Deficient
and Their Potential Adverse Effects, 1993-99 12

Table 3: Overview of Enforcement Tools Available to HCFA 17

ESRD end-stage renal disease

HCFA Health Care Financing Administration

NIH National Institutes of Health

OSCAR On-Line Survey, Certification, and Reporting

USRDS United States Renal Data System

Health, Education, and
Human Services Division


June 23, 2000

The Honorable Charles E. Grassley
The Honorable John B. Breaux
Ranking Minority Member
Special Committee on Aging
United States Senate

More than 288,000 people suffering from kidney failure depend on Medicare to
cover the cost of the life-sustaining kidney dialysis treatments they
receive several times each week. These end-stage renal disease (ESRD)
beneficiaries are among Medicare's sickest and most vulnerable patients,
costing Medicare about $10 billion in 1998. Dialysis is a technically
complicated process, and mistakes or poor procedures can cause patients
serious injury or even death. The quality of care that these Medicare
beneficiaries receive at some of the nation's 3,817 dialysis facilities is
in dispute. On the positive side, death and hospitalization rates related to
dialysis appear to have declined over time. But at the same time, concerns
have been raised about reduced staffing levels at ESRD facilities and the
greater use of potentially less skilled technicians rather than nursing
personnel to administer dialysis treatments.

The Health Care Financing Administration (HCFA), the agency that administers
Medicare, is responsible for overseeing adherence to its quality-of-care
standards and promoting quality improvement among ESRD facilities. HCFA pays
state agencies to perform on-site inspections of these facilities and
contracts with 18 organizations, called ESRD networks, to gather data about
dialysis treatments and conduct activities to improve the quality of care
patients receive. You asked us to evaluate HCFA's processes to ensure that
ESRD facilities meet quality-of-care standards. We focused our work on
determining (1) the extent to which on-site inspections of dialysis
facilities are performed and problems are identified, (2) whether an
effective process exists to ensure that dialysis facilities correct
problems, and (3) what steps are being taken to use available monitoring
resources as effectively as possible.

Our report is based in part on analysis of information from national
databases compiled by HCFA, state survey agencies, and ESRD networks. For a
more in-depth review of actual monitoring and enforcement activities, we
focused on work being done by state agencies in California, New Jersey,
Texas, Oregon, and Washington, and at the four HCFA regional offices and the
four ESRD networks that oversee dialysis facilities in those states. We
conducted our work between November 1999 and May 2000 in accordance with
generally accepted government auditing standards. Appendix I contains a more
detailed explanation of our scope and methodology.

Over the past 7 years, the number of HCFA-funded inspections of dialysis
facilities has declined significantly. These unannounced inspections,
commonly called surveys, which are HCFA's primary tool for ensuring that
facilities meet standards protecting patients' health and safety, were
conducted at only 11 percent of the dialysis facilities eligible for
recertification in 1999, compared with 52 percent in 1993. When such surveys
were conducted, they showed that noncompliance is a problem. For example, in
1999, 15 percent of the surveyed facilities had deficiencies severe enough,
if uncorrected, to warrant terminating their participation in Medicare. To
enable more frequent surveys, HCFA has requested a threefold increase in
funding for on-site inspections in its budget request for fiscal year 2001.
This funding level would support a survey of all dialysis facilities every 3

While increasing on-site surveys will likely encourage more facilities to
improve conditions, the enforcement system provides little assurance that
corrections will be sustained. Essentially, HCFA's only current enforcement
tool is to terminate a facility from the Medicare program if it does not
correct its deficiencies. The threat of termination brings nearly all
facilities into compliance for a while, but they do not necessarily stay
that way. In every state we visited, we found instances in which facilities
that had corrected their problems were found to have serious problems
shortly afterward. The Congress has authorized HCFA to use other enforcement
tools, such as the denial of payment for Medicare services, but HCFA
maintains that this authority would have limited effectiveness and
applicability. For example, HCFA has not taken steps to use denial of
payments because, like termination from the program, this sanction could be
applied only if the facility failed to return to compliance.

HCFA is planning to use clinical and outcome data (such as patient death
rates) more extensively in deciding which facilities to survey and monitor
more closely. Although the information HCFA intends to use may help in that
regard, it has limitations as well. These data are designed to give a
picture of the care being provided to ESRD patients generally, but they are
often not current, detailed, or reliable enough to detect specific
facilities that are providing substandard services. For example, we found
instances in which facilities had above-average clinical outcome scores but
were found to have serious deficiencies during on-site surveys. HCFA's ESRD
networks already collect considerable facility-specific information, such as
patient complaints, that is more timely, but they do not necessarily share
it with state survey agencies. One state where such sharing had occurred
showed positive results.

To give facilities a greater incentive to remain in compliance, we suggest
that the Congress consider strengthening HCFA's authority to impose monetary
penalties on dialysis facilities that have the most severe or repeated
serious deficiencies. We are also recommending that HCFA strengthen its
systems for targeting on-site surveys and make use of additional available
enforcement tools.

The Medicare program covers dialysis services for patients suffering from
ESRD, the stage of kidney impairment that is considered irreversible and
requires either regular dialysis treatments or a kidney transplant to
maintain life. Kidney failure can result not only directly from kidney
disease but also indirectly from other diseases, such as diabetes and
hypertension. Dialysis is a technically complicated process that is
individualized to accommodate each patient's needs. There are two general
modes of dialysis treatment: hemodialysis and peritoneal dialysis, both of
which can be performed at a dialysis facility or at home. During
hemodialysis, the patient's blood is filtered through a dialysis machine
that withdraws fluid and toxic materials before returning cleansed blood to
the patient. In peritoneal dialysis, the removal of fluid and toxic
materials takes place within the abdominal cavity by means of cleansing
fluid and drainage. The vast majority of ESRD patients (86 percent) receive
hemodialysis. Generally, an ESRD patient has three dialysis sessions per
week, lasting 3 to 4 hours each, usually provided on an outpatient basis.

Almost all dialysis patients, regardless of their age, are
Medicare-eligible, making Medicare the main payer of dialysis services.
Total expenditures for the Medicare ESRD program, authorized in 1972, have
grown steadily from $229 million in 1974 to over $11.4 billion in 1998. A
major reason for the increase in program costs is the dramatic rise in
enrollment: total enrollment for those beneficiaries requiring dialysis or
transplants has risen from approximately 16,000 in 1974 to over 360,000 in
1998. The increase in enrollment has been fueled by expansion of the
criteria that determine who is an acceptable candidate for dialysis. For
example, physicians are recommending dialysis for older patients--the number
of patients in the ESRD program who are 65 or older increased from 5 percent
in 1973 to 50 percent in 1997. In addition, the program is admitting more
patients with hypertension and severe diabetes (see app. II for additional
information on the changing demographics of dialysis patients). The number
of dialysis facilities has grown in step with the growth in the number of
dialysis patients. Since 1993, the number of facilities has increased at an
average rate of 6 percent annually, reaching 3,817 participating facilities
in 1999.

Medicare payments, which are based primarily on a fixed rate per treatment,
have essentially remained unchanged since program inception. For facilities
that aim to maximize profits, such fixed payment rates can create incentives
for efficiencies, but they can also be an incentive for underservice. This
movement toward greater efficiencies has spurred considerable industry
consolidation into for-profit facilities and chain providers. The Medicare
Payment Advisory Commission reported that in 1997, 68 percent of the
non-hospital-based facilities were for-profit. And three-quarters of all
for-profit dialysis facilities were affiliated with a chain. In 1998,
dialysis facilities used about 12 percent fewer staff to administer dialysis
than in 1993. Furthermore, they increasingly rely on lower-cost technicians
rather than nursing personnel to monitor dialysis treatments.

HCFA has established a set of quality-of-care standards, called "conditions
of participation," that dialysis facilities are required to meet before they
can receive Medicare payments. The conditions of participation are
regulatory standards, first established in 1976, designed to ensure that
dialysis facilities are capable of furnishing quality care in a safe
environment. There are 11 conditions of participation covering areas such as
the physical environment of the facility, the adequacy of patient care
plans, and the management of the facility (see app. III for a more detailed
description of the 11 conditions of participation).

Oversight of the program falls primarily on state survey agencies and ESRD
networks working under contract with HCFA. Each plays a separate oversight
role. State survey agencies--generally state departments of health--are
responsible for verifying that dialysis facilities comply with conditions of
participation. They do so primarily through unannounced site surveys of
dialysis facilities. These agencies, which have expertise in health and
safety issues, are frequently responsible for surveying other types of
health care facilities that require certification for participation in the
Medicare program, including nursing homes and home health agencies. No
statutory requirements exist for the frequency of state surveys of dialysis
facilities; rather, the frequency is determined mainly by the funding
available. For fiscal year 2000, state agencies are expected to receive
about $2 million for survey and certification of dialysis facilities.

State agencies, with HCFA's concurrence, determine whether problems
identified during a survey are serious enough to warrant finding a facility
out of compliance with a condition of participation. If a facility is found
to be out of compliance and the deficiencies are not corrected--generally
within 90 days--the facility is subject to termination from the Medicare
program. If deficiencies are so severe that they put patients' health and
safety in immediate jeopardy, the facility has only 23 days to make
corrections (this is called the "fast track" for termination). To determine
whether deficiencies have been adequately addressed, the agency conducts
another on-site survey. If the facility is still out of compliance, the
state agency refers the facility to HCFA, which is responsible for
prescribing and reviewing additional corrective actions and, if these
additional steps are insufficient, proceeding with the termination process.
If deficiencies are corrected or plans for correction are developed at any
time during this process, the process to terminate is stopped.

ESRD networks are organizations that contract with HCFA to help ensure
effective and efficient administration of the ESRD program and improve
program performance. The 18 networks are funded through a fifty-cent charge
on each Medicare dialysis treatment, which for fiscal year 2001 is expected
to total about $18 million. ESRD networks have medical staff with experience
in dialysis, and their boards of directors and medical review boards are
composed of dialysis facility representatives, physicians, and dialysis
patients. As a result, they tend to have more clinical expertise
specifically on dialysis than do state survey agencies.

In contrast to state agencies, which check for adherence to conditions of
participation, the networks are responsible for quality improvement, which
focuses on improving the clinical outcomes of dialysis facilities. Network
activities include identifying and collecting data on key clinical
indicators and furnishing individual facilities with regional performance
data on clinical indicators so a facility can compare its performance with
that of other facilities. The networks also provide technical support to
help facilities improve their performance on the key indicators. In the
aggregate, these indicators show that the quality of dialysis care
nationwide has been improving. As evidence, HCFA's 1999 data report cited
first-year patient death rates, which, after adjustments for some patient
conditions, declined from more than 30 per 100 patient years in 1986 to
slightly more than 21 in 1996.1 The data also showed that in 1997, 72
percent of the sampled patients received adequate dialysis as measured by
urea reduction, an increase from 59 percent in 1995. The use of clinical
outcome data has evolved from a tool to assess the overall quality of
dialysis services at the patient level to being considered by HCFA as a
method to assess the quality of services at individual facilities.

In addition, networks conduct specific quality improvement projects with
dialysis facilities, handle grievances regarding patient care, and assist
patients in finding dialysis providers. Networks also conduct on-site
inspections at facilities to assess procedures and assist facilities in
improving the quality of care they provide. To participate in Medicare,
facilities must cooperate with network data collection efforts and quality
improvement projects.

Oversight of state survey agencies is coordinated by HCFA's Center for
Medicaid and State Operations in its central office and its 10 regional
offices. Oversight of the 18 ESRD networks and their activities is
coordinated by HCFA's Office of Clinical Standards and Quality and regional
offices in Boston, Dallas, Kansas City, and Seattle.


On-site inspections by state survey agencies are HCFA's primary oversight
tool to ensure that ESRD facilities meet Medicare conditions of
participation. An effective monitoring program should ensure that
deficiencies are identified and corrected at surveyed facilities and that
facilities are surveyed often and with enough randomness to give facilities
an incentive to remain in compliance with standards. However, the number of
recertification surveys performed each year is decreasing and has reached
the point that only a small fraction of the facilities are surveyed. This is
a matter for concern because we found ample evidence that serious health and
safety problems exist in a number of dialysis facilities. Recognizing that
dwindling surveys presents a serious risk to effective monitoring, HCFA has
requested a nearly threefold increase in funding for ESRD surveys in its
2001 budget.

Inspections are required (1) when a facility begins to participate in
Medicare, (2) when a facility changes or expands services, such as starting
a dialyzer reuse program,2 and (3) when a facility relocates. Aside from
these requirements, there is no provision in law or regulation that sets a
maximum period between surveys. Rather, the interval between a facility's
initial survey and subsequent recertification surveys depends on HCFA's
survey goals; indications that additional surveys are needed because of a
complaint or a grievance; and the extent of the survey resources made
available through HCFA's contract payments to the states and through other
funding sources, such as state appropriations. Generally, states determine
which facilities to survey with only limited input from HCFA or ESRD
networks. State agency officials told us that they use criteria such as the
date of the last survey and the volume and type of complaints received to
set their survey agendas.

Since 1993, the number of HCFA-funded dialysis facility surveys has declined
substantially. At the same time, the number of new facilities entering the
program annually has increased. These new facilities--each requiring a
survey--along with a decrease in funding from HCFA, have led to a
substantial drop in the percentage of existing facilities surveyed (see
table 1). In 1993, 52 percent of facilities in the program prior to 1993
received a recertification survey. By 1999, only 11 percent of the
facilities subject to a recertification survey were resurveyed. At the
current survey rate, once a dialysis facility receives its initial
certification survey, it is not likely to be resurveyed for about 9 years.
Currently, 772 active dialysis facilities have not been resurveyed in the
last 5 years.

                            Total number of
         Total number of    facilities that
 Year of facilities         could be resurveyed Total number  Percentage
 survey  participating in   (existing           of facilities resurveyed
         Medicare           facilities          resurveyed
                            only--excludes new
 1993    2,559              2,334               1,216         52
 1994    2,741              2,517               727           29
 1995    3,000              2,697               389           14
 1996    3,209              2,942               476           16
 1997    3,448              3,148               469           15
 1998    3,659              3,370               398           12
 1999    3,817              3,589               409           11

Note: Our analysis starts with 1993 because it represents the point where
the downward trend in resurvey activity starts. In addition, data from prior
years are less complete and likely understate the true level of survey
activity. Nevertheless, the prior-year data show that the number of existing
facilities resurveyed in prior years was comparable to 1993 levels.

Source: GAO analysis based on data from HCFA.

Deficiencies Is Rising

The infrequency of surveys makes it impossible to determine the exact extent
to which dialysis facilities are currently in compliance with the conditions
of participation. However, data indicate that the percentage of inspected
facilities found to be out of compliance has increased significantly during
the 1990s. In 1993, 6 percent of facilities surveyed were cited for a
condition-of-participation deficiency; that number rose to 15 percent in
1999.3 In two states we visited, state survey officials have conducted more
frequent on-site inspections. They were able to do this either by
reallocating survey resources from other types of health care facilities,
like rural health clinics, to dialysis facilities or by using additional
funding from their state governments to fulfill their role in state dialysis
facility licensing laws. In these states, inspectors found facilities out of
compliance at high rates.

� Oregon . During a 20-month period from June 1998 to March 2000, Oregon's
state agency conducted 41 surveys spread across the state's 39 dialysis
facilities.4 Eleven facilities (26 percent) were found to be out of
compliance with the Medicare conditions of participation. Had the state not
stepped up its efforts, it would have taken 4 to 10 years to identify these
seriously deficient facilities.5

� Texas. The passage of a state dialysis licensing requirement in 1996 led
to a dramatic increase in the number of dialysis facility surveys in Texas.
In 1996, in order to license the facilities, the agency surveyed all 244 in
the state and found that 33 (14 percent) were out of compliance with
Medicare conditions of participation, compared with a national average at
the time of about 9 percent.

The five conditions of participation most commonly cited as deficient
accounted for 75 percent of all deficiencies reported during 1993 through
1999. Table 2 lists these conditions of participation as well as describes
examples of the potential for harm resulting from these deficiencies.

 Percentage of total                             Example of potential
 deficiencies,       Condition of participation  adverse effects of
 1993-99                                         noncompliance
                                                 Certain procedures are
                                                 associated with dialysis
                                                 for which failure to
                     The facility's governing    follow established
                     body should adopt and       protocols could result in
                     enforce written rules and   serious injury. For
 23                  regulations, including      instance, inadequate
                     operational rules and       medication delivery system
                     patient care policies, to   policies and procedures
                     safeguard the health and    can lead to medication
                     safety of patients.         errors and adverse drug
                                                 events that increase a
                                                 patient's risk of
                                                 complications or death.
                                                 Deficient equipment could
                                                 lead to life-threatening
                     The facility's physical     complications. For
                     environment should be       instance, if a dialysis
                                                 pump is not inspected and
 19                  functional, sanitary, safe, calibrated properly, the
                     and comfortable for
                     patients, staff, and the    patient may experience
                     public.                     blood loss, receive an air
                                                 bubble, or sustain other
                                                 serious injury during
                                                 Deficient reuse practices
                                                 can expose patients to
                                                 chemical or infectious
                     The reuse of hemodialyzers  hazards by means of direct
                     and supplies should occur   introduction into their
 13                  only in facilities that     circulatory systems. ESRD
                     meet certification          patients are more
                     standards.                  susceptible to infection,
                                                 and close attention to
                                                 infection control is a
                                                 critical prevention
                                                 Deficient patient care
                                                 planning can result in
                                                 ineffective treatment. For
                     The long-term program and   instance, an inadequate
                     patient care plans should   patient care plan could
                     show that a professional,   fail to identify and refer
                     multidisciplinary health    a patient who is eligible
 12                  care team developed a       for kidney transplant. Or
                     written long-term-care plan the care plan could fail
                     to ensure each patient      to include monitoring
                     receives individualized     alerts for patients with
                     care and the appropriate    cardiac conditions such as
                     type of dialysis treatment. arrhythmia, which can be a
                                                 complication during
                     The director of the renal
                     dialysis facility should be If dialysis staffs are not
                     a Board-certified physician properly trained, they
                     and trained in the care of  cannot be expected to
 9                   ESRD patients. The          respond quickly and
                     director, among other       effectively to the range
                     things, is also responsible of complications that can
                     for ensuring the proper     arise during dialysis
                     training of staff.          treatment.

Source: GAO analysis of HCFA data.

In its 2001 budget submission to the Congress, HCFA requested a nearly
threefold increase in the funding for dialysis facility surveys--from $2.2
million in fiscal year 2000 to $6.3 million in 2001. This increase,
according to HCFA, will ensure that ESRD facilities are surveyed at least
every 3 years. HCFA is seeking this additional funding in response to the
declining survey frequency and the rising number of deficiencies identified,
as well as information from states regarding complaints about dialysis
facilities. Nationwide, complaints to state survey agencies rose 22 percent
from 1998 to 1999. As a case in point, the Oregon Department of Health
received just 2 complaints in 1997, 6 in 1998, and 19 in 1999.


Even if the frequency of state on-site inspections increases, HCFA's
enforcement actions against noncomplying facilities provide little incentive
for facilities to make more than temporary improvements. The effectiveness
of HCFA's enforcement of condition-of-participation requirements is limited
because HCFA relies on termination from Medicare--or, in reality, the threat
of termination--as its sole enforcement tool. To escape termination from the
program, facilities almost always bring themselves back into compliance, but
they face minimal consequences if they again slip out of compliance. For a
variety of reasons, HCFA has not developed or used other sanctions that
would give facilities more of an incentive to maintain compliance with
conditions of participation.6 In combination with the decreasing frequency
of state surveys, these factors severely limit HCFA's ability to promote
long-term compliance.

Necessarily Keep Them There

HCFA uses the threat of termination as its primary enforcement tool. When
state agencies identify problems that are sufficiently serious to put the
facility out of compliance with a condition of participation, they begin a
process, through HCFA, by which the facility either corrects its
deficiencies or is terminated from the Medicare program. Before a facility
can be terminated, it has an opportunity to correct its deficiencies or
develop an acceptable plan of correction. Actions and plans may include
establishing new procedures and policies, documenting and clarifying roles
and responsibilities of facility staff and managers, recruiting qualified
staff, and conducting in-service training of personnel. Once the state
agency determines, normally by a revisit, that the deficiency has been
corrected and has reasonable assurance that it will not recur, the
termination process is stopped.

In practice, facilities nearly always correct such deficiencies and are
rarely terminated. For example, 481 of the surveys conducted since 1993
resulted in at least one condition-of-participation deficiency,7 but only
three facilities have been terminated for not correcting a deficiency.8
According to HCFA officials, the goal of the monitoring and enforcement
program is to bring problem facilities back into compliance with conditions
of participation, not to punish them. They stated that the threat of
termination from Medicare is an effective method to bring about compliance.

Although the threat of termination is effective in bringing a facility into
compliance, it provides little assurance that a facility, once recertified,
will not immediately slip out of compliance again. For one thing, while
facilities are correcting their deficiencies, they are allowed to continue
to receive full Medicare payments, and they do not have to reimburse
Medicare for payments they received when the services and care they provided
were not at the level required for payment. Moreover, if they slip out of
compliance again and face termination, they can avoid it by returning to
compliance during the grace period.

The length of time between surveys makes it difficult to determine how
quickly and how often facilities fall out of compliance. However, analysis
of the survey deficiency database suggests a pattern of repeated
deficiencies. For example, of facilities with four or more surveys,9 38
percent of those that had deficiencies on their most recent survey were also
deficient on at least one of the same requirements on their last prior
survey. More than half of them had two or more such repeat deficiencies.

In some situations, termination is not used even when a facility fails to
take appropriate corrective action after the termination process has begun.
State, network, and HCFA officials told us that termination is not always an
option because it could create serious access problems for patients using
that particular facility. In fact, to avoid such access problems, throughout
the termination and corrective action process--which can last 90 days or
more--noncomplying facilities continue to receive Medicare payments and may
continue to accept new Medicare patients.

During our state visits, we also identified cases in which facilities
returned to compliance only to be found out of compliance again a short time
later. Three examples follow.

� Washington. On March 24, 1999, a facility was cited for noncompliance with
such requirements as following physician orders, following anemia management
protocols, and following up on adverse incidents at the facility. The state
accepted a corrective action plan on July 21. However, on October 13, a
lengthy complaint was filed alleging that the same types of deficiencies
found during the survey were still occurring and that the facility's
management was not correcting the problems. The complaint also included a
long list of incidents that allegedly occurred over a 6-month period,
including the months the facility was reported to be taking corrective
actions. Many of the allegations and incidents in the complaint were
substantiated during the state investigation, including problems that were
also cited on the prior survey: for example, not writing reports for serious
incidents, such as medication errors, in which patients did not receive
prescribed medication and in which other patients received medications that
had not been prescribed for them. During this same investigation, the state
found poor patient care practices, such as leaving a patient on a bedpan
throughout the 3-hour dialysis treatment, causing blisters. Overall, the
deficiencies found were so severe that they posed immediate jeopardy to
patient health and safety, and the facility was placed on a fast track to
termination. The facility again took corrective actions that were acceptable
to the state and HCFA, and at the time of our work, continued to dialyze
Medicare patients.

� New Jersey. A facility's initial certification survey on February 26,
1996, found numerous deficiencies, including having untrained personnel
responsible for water treatment, not testing chloramine levels of water
daily, not having a quality assurance plan, and poor patient care planning.
After developing an acceptable plan of correction, the facility was
certified to operate six dialysis stations, treating 35 patients. Over the
next 18 months, the ESRD network conducted several on-site visits at the
facility and each time found serious and continuing problems. For example,
patients were placed at serious risk because dialysate (the fluid used to
extract toxins from the blood) was prepared using untreated water.
Furthermore, the facility's treated water, dialysate, and dialysis machines
had bacterial contamination that exceeded acceptable levels.10 In 1998, the
state agency resurveyed the facility and found the problems identified by
the network as well as the same deficiencies found earlier by the state. In
response, the facility again developed an acceptable plan for corrections.
Since then, the facility has continued to treat Medicare patients and has
not been resurveyed in more than 2 years.

� Texas. A facility cycled in and out of compliance over a 9-year period
while developing numerous plans of correction at the direction of both the
state and the ESRD network. On many occasions, the deficiencies were so
severe they put the health and safety of the facility's 227 patients in
immediate jeopardy. For example, the facility had repeated problems
regarding providing adequate levels of dialysis, managing patient anemia,
and planning patient care. In 1999 HCFA put the facility on a fast track to
termination, citing such deficiencies as not providing care necessary to
address patients' medical needs, not complying with physicians' orders, lack
of physician planning of and supervision over patient care, and not
following up on adverse incidents. It took more than 4 months and two
revisits from the state before the facility came back into compliance.
However, when the state conducted a survey 4 months later, the facility was
again out of compliance. At the time of our review, state agency officials
were exploring enforcement options under state licensing authority.

Termination is one of several enforcement tools available to HCFA, but it is
the only one in use (see table 3). HCFA maintains that the other tools have
varying limitations that have prevented them from being used as effective
alternatives. The following sections discuss each enforcement tool for
dialysis facilities and the limitations that might be affecting its use.

 Type of             Enforcement                     Concerns or
 noncompliance       tool            Extent used     limitations

 Failure to comply                   Invoked when a  Successful in bringing
 with Medicare       Termination     facility is not facilities back into
 conditions of       from the        in compliance   compliance, but not
 participation for   Medicare        with a          necessarily at keeping
 dialysis facilities program         condition of    them in compliance

                     Denial of                       Like termination,
                                     Not implemented facilities can avoid
                     payment for new into regulation this sanction by
                     patients        by HCFA         returning to
 Failure to follow   denial of
 industry standards  payments for    Not implemented HCFA maintains that
 and practices for   services        into HCFA       applying this sanction
 reusing             provided when   procedures      would be cumbersome
 hemodialyzers       the facility
                     was out of
 Failure to
 participate in ESRD
 network             Termination
 quality-of-care     from the        Never levied    Only option available
 initiatives, or to  Medicare        against a       if the deficiency is
 pursue              program         facility        serious
                     Denial of
                     payment for new                 Limited
                     patients        Never levied    applicability--can be
                     admitted after  against a       used only for
                     the effective   facility        nonserious
                     date of the                     deficiencies
                     Reduction of a
                     payment rate by
                     20 percent for
                     each 30-day                     Limited
                     period that the Never levied    applicability--can be
                     facility        against a       used only for
                     continues to    facility        nonserious
                     not participate                 deficiencies
                     or pursue goals
                     after being
                     directed to do
                     Withholding all
                     payments,                       Limited
                                     Never levied    applicability--can be
                     without         against a       used only for
                     interest, for
                     all ESRD        facility        nonserious
                     services                        deficiencies

Denial of Payment for New Medicare Patients

In 1987 the Congress gave HCFA the authority to develop regulations allowing
the agency to deny Medicare payments for new patients at facilities that are
not in compliance with the conditions of participation. At that time, the
Congress noted that HCFA may be reluctant to use termination, even in cases
of serious deficiencies, but that persuasion or technical assistance alone
may not be sufficient to bring facilities into compliance. However, HCFA has
not promulgated regulations for denying payments. HCFA officials told us
that denying payments would offer no advantages over termination because,
under the law, facilities can avoid the penalty by returning to compliance
within the grace period. In that sense, denial of payments would operate the
same as termination--it would occur only if the facility did not comply.

Retroactive Denial of Payment for Improper Dialyzer Reuse

The Congress provided HCFA additional and broader authority to address
facilities not complying with standards and requirements for reprocessing
and reusing dialyzers. Compliance with accepted standards is important to
prevent the weakened immune systems of dialysis patients from being exposed
to microbial contamination and dangerous levels of the germicide used to
clean the dialyzers. HCFA was authorized to impose sanctions retroactively
when a facility failed to follow industry guidelines on appropriate reuse
procedures, even if the facility had corrected its deficient practices.
Unlike termination, this tool also can be used for deficiencies that are not
considered severe enough to constitute a violation of the applicable
condition of participation. HCFA has not incorporated this authority into
its procedures, believing that it would be too cumbersome to do so. HCFA
officials explained that it is administratively difficult to use this
sanction because it is hard to identify which specific dialysis treatments
are actually affected by a facility's deficient process for reusing

We disagree that this authority would necessarily be cumbersome to
implement--at least not in all instances. Many of the important reuse
standards relate to processes and procedures that affect almost all patients
in a facility. As a result, if a deficiency is cited that affects all or
most of a facility's patients, determining which payments should be denied
may not be as difficult as HCFA assumes. Our state-level reviews showed
instances in which such conditions applied. That is, many of the
deficiencies affected all patients that were dialyzed during the period
examined, and surveyors were able to identify specific days of
noncompliance. Payments made for services provided during the period of the
deficiency would thus be subject to recoupment under current regulations,
requiring relatively little effort on the part of claims processing
contractors to establish the appropriate amounts.

Penalties for Noncompliance With ESRD Network Activities or Initiatives

HCFA has several financial sanctions at its disposal if facilities do not
cooperate with ESRD network activities or pursue the network's quality goals
and initiatives. After providing notice to chronically deficient facilities,
HCFA can deny payment for new patients, reduce payments for services
provided, or withhold payments altogether. However, the law only authorizes
use of these financial sanctions if the deficiency does not "jeopardize
patient health and safety." This, in practice, creates an enforcement
paradox. Networks are inclined to refer only facilities with serious
deficiencies to HCFA for sanction, but only the nonserious deficiencies
would be subject to the financial sanction. For serious deficiencies,
termination is the only sanction available.

In practice, the networks try to educate, provide technical assistance,
require corrective action plans and progress reports, and generally use more
collegial means to change the behavior of noncomplying facilities. Since
1993, only two facilities nationwide have been recommended for alternative
sanctions by ESRD networks.11 Each involved a situation in which the network
determined that patient health and safety were being jeopardized because of
a lack of fundamental processes and systems, but the facility did not
respond to the network's efforts to address the problems. In both cases,
HCFA did not proceed with sanctions but instead relied on surveys to
document problems and on the threat of termination to bring about needed

Than Those Available for Nursing Homes

HCFA does not have the same tools to create strong incentives for ESRD
facilities to maintain compliance that it does for nursing homes. In 1987,
largely in response to studies showing that many nursing homes tended to
cycle in and out of compliance with standards, the Congress authorized HCFA
to levy civil monetary penalties of up to $10,000 per day on homes that do
not meet Medicare requirements of participation. The Congress intended these
penalties to create a strong incentive to maintain compliance. In July 1995
HCFA established in regulation that nursing homes are subject to these
financial sanctions on the basis of the severity of their deficiencies and
can also face financial sanctions if they have repeated serious
deficiencies. These latter penalties can be levied without allowing a grace
period to correct the deficiencies, and they can be applied even if a
nursing home corrects the deficiencies. In our previous review of
enforcement of nursing home standards, we reported that while administrative
problems with appeals had not yet been resolved, civil monetary penalties
may provide a strong deterrence to severe or sustained noncompliance.12

HCFA has been working on a pilot project that will use available
facility-specific data to help state surveyors select facilities for review.
While this idea has merit, for such a screening process to be effective, the
data must be more timely and reliable than what HCFA currently has at its
disposal. Moreover, the extent to which outcome measures, which would be
included, would accurately predict the presence of serious health and safety
deficiencies that would be identified through on-site inspections is
unclear. In contrast, opportunities exist to better target resources through
improved communication between the ESRD networks and state survey agencies.
Thus far, HCFA's efforts to facilitate the exchange of information between
networks and survey agencies have been inconsistent.

Target Surveys

In May 2000, as part of a pilot project, HCFA sent individual dialysis
facility profiles created using available facility-specific data to the
seven state survey agencies participating in the pilot. These profiles are
designed to help state agencies determine which facilities to select for
on-site inspections. The information focuses on the adequacy of dialysis
provided, the frequency of some dialysis-associated complications and
diseases, and the types of practices used by the facilities in administering
dialysis and reusing dialyzers. This information comes from a number of
sources. Part of it is data currently used to prepare annual reports on
renal care, such as standardized mortality and hospitalization rates. HCFA
obtains other data through claims for payment that facilities file with
intermediaries. These claims include information on the adequacy of dialysis
treatments (the urea-reduction ratio) and an assessment of anemia in
patients (hematocrit). HCFA is also using data on patient infections
collected by the Centers for Disease Control and Prevention.

HCFA plans to collect feedback from the seven pilot states in the fall of
2000 and to begin training surveyors in the use of the profiles in early
2001. The evaluation of the pilot project is scheduled to be completed in
early 2001, but at the time of our review, the evaluation plan and criteria
had not been set.

Individual Facilities

Because the facility profile project is now being tested, we did not
comprehensively evaluate it. However, we did identify several issues that
need to be considered before the data are used to significantly influence
the survey selection process. The major concern is whether the data are a
strong predictor of noncompliance with Medicare standards. In the states we
visited, we found cases in which facilities had good clinical outcome scores
but were identified in on-site surveys as seriously out of compliance with
Medicare standards. For instance, during a complaint investigation, state
surveyors and network quality assurance staff found serious,
life-threatening deficiencies, such as a lack of knowledge of basic medical
and dialysis practices like anemia management, infection control, and water
purity. However, when network officials reviewed the facility's clinical
outcomes, the facility had better-than-average scores.

Available Data Are Neither Timely nor Necessarily Reliable

Whether the data come from Medicare claims or through collection by ESRD
networks, the process by which HCFA collects and aggregates data on ESRD
patients and services takes time. Much of the data for the facility-specific
profiles is at least 2 years old. For example, the facility profiles for the
year 2000 report hospitalization and mortality data from 1996 through 1998.
The Centers for Disease Control and Prevention surveillance data included in
these profiles were collected through a 1997 survey. The screening tool
proposed in the HCFA pilot would thus reflect conditions at the facility
that were at least 2 years old. It is reasonable to assume that, given the
dynamic nature of the industry, such a screen would not reflect current

Although clinical outcome measures, such as hematocrit levels and the
urea-reduction ratio, are generally accepted as good measures of dialysis
service quality, the assessment of the reliability of the measures reported
to fiscal intermediaries yielded mixed results. For example, an initial
internal study found differences between the clinical measures facilities
reported to fiscal intermediaries and the information collected by ESRD
networks. Preliminary results of a later HCFA study found the two data sets
to be more closely correlated. A primary concern that remains is the lack of
assurance that a single set of procedures to collect, store, assay, and
report laboratory values is being followed consistently.

Predictive Power of Outcome Measures Is Unclear

Another significant issue involved in using clinical outcome data in
conjunction with the facility selection process is whether outcome measures
are a reasonable predictor of a facility's level of compliance with Medicare
standards. Although a limited analysis found outcome measures can have a
predictive power,13 there is disagreement on the extent to which outcome
measures currently available to HCFA are strong predictors of compliance
with Medicare standards. Moreover, concerns exist that using outcome
measures to inform the survey selection process may complicate the process
of collecting accurate data.

For example, clinical outcome measures like urea-reduction ratios were
designed to estimate the extent to which health care providers conformed
with clinical practice guidelines, and not necessarily to reflect the extent
to which facilities complied with important condition-of-participation
standards. As a result, ESRD network and state agency staff told us that
dialysis providers could have clinical outcome scores within the average
range for the region and still have serious deficiencies, often in such
critical areas as water purity, staff competence, and infection control.

The experience of the Texas network shows the difficulty of using outcome
measures as the key tool to predict which facilities do not comply with
Medicare conditions of participation. The network compared clinical outcome
data with the results of state surveys for 179 facilities for 1996.14 An
analysis of the data found that using outcome measures would have been an
improvement over the random chance that selected facilities would have
condition-of-participation deficiencies. However, network officials cited
methodological difficulties that, in their view, would have limited the
usefulness of these results for targeting surveys. For example, clinical
outcome data are not current enough and would not have been available in the
same year as the surveys. Network officials also pointed out that the data
did not account for the severity of the deficiencies, in that some
facilities with the most severe noncompliance problems had acceptable
outcome measures. As a result of these and other concerns, the network's
medical review board reported that its analysis was inconclusive about the
merits of using clinical outcome data as a controlling factor in targeting
state survey resources.

Over time, the process of using clinical performance measures to score
facilities and then conduct surveys on the basis of these scores could, in
itself, complicate efforts to improve the accuracy of the facility-reported
data. In the long term, the use of facility-specific data to inform the
regulatory oversight process creates an incentive for facilities to report
data that indicate acceptable performance whether they are providing an
acceptable level of service or not. HCFA quality assurance specialists
reported in 1999 that clinical performance data were to be used primarily
for population-based quality improvement rather than for evaluating
facilities' care of specific patients or compliance with quality assurance
standards. The report noted considerable concern that, if inappropriately
used (particularly by regulators), the clinical performance measures could
potentially have a deleterious effect on the care of dialysis patients,
presumably by creating incentives for facilities to "game the reporting
system."15 Such incentives are particularly problematic with the ESRD
program because currently most of the data are self-reported. Verification
of the data is limited to a review for transcription errors.

By building stronger cooperation between ESRD networks and state survey
agencies, HCFA has an opportunity to improve the quality of
facility-specific performance data used in selecting facilities to survey.
ESRD networks collect a variety of data from individual dialysis facilities
and in some cases have facility performance information that is available on
a real-time basis, rather than after a lag of several years. However, HCFA
has not consistently encouraged this coordination, and, in some cases,
through conflicting policy interpretations, has actually impeded it. As a
result, the level of coordination and information sharing varies
dramatically across the nation, and in most cases little of it takes place.

HCFA has not been clear on the type of relationship and coordination it
expects between networks and states. HCFA's current policy is that networks
may readily share facility-specific information with state survey agencies
to aid in the certification process. This stance reinforces HCFA contract
requirements with networks from prior years, in which networks were
instructed to achieve a working relationship with state agencies and HCFA
regional offices that would assist each in improving the quality of care
provided to ESRD patients. Activities the networks are to undertake with
state agencies include sharing information and data reports, communicating
on patient quality-of-care issues, providing facility-specific data to the
state agency, and working to support their survey activities.

HCFA regional offices that oversee network and survey agency activities have
not applied this policy consistently. In fact, most HCFA regional offices
restrict networks from sharing facility-specific information and support
ESRD networks when they deny requests by state survey agencies for such
information, saying that federal confidentiality restrictions prohibit this
sort of exchange. In contrast, with the knowledge of the HCFA regional
office, the ESRD network in Texas began providing facility-specific
information to the Texas Department of Health after the state passed a
licensure law for dialysis facilities in 1996. More recently, in early 2000,
some HCFA regional offices have begun efforts to facilitate the
communication and exchange of information, including facility-specific
performance information, between ESRD networks and state agencies.

By sharing information and knowledge, ESRD networks and state agencies can
effect a more complete picture of ESRD facilities. Each has different
information and knowledge about a facility that together provide a more
accurate overall assessment of the quality of care a facility provides. ESRD
networks work solely with ESRD facilities; have information on the clinical
aspects of the care in facilities; and also may be more aware of staffing
and management changes, patient complaints, and the results of network
quality improvement initiatives, which can have a major impact on the
quality of care provided. In contrast, networks do not have detailed
information about facilities' systems and processes that are key to quality
of care, such as the quality of water used, infection control procedures,
reprocessing of dialyzers, and care planning. This type of information can
be provided by state survey agencies.

Oversight of ESRD facilities needs improvement. While many facilities may be
conscientiously and consistently providing quality care, some do not, and
current oversight efforts are not enough to find and correct the problems in
a timely manner. HCFA's request for a threefold budget increase for
inspecting ESRD facilities is a sign that the agency realizes additional
oversight is necessary.

While increasing the number of inspections should help improve oversight,
other things can be done as well. One is to put some teeth into the
enforcement process. Currently, when condition-of-participation violations
are found, even on a recurring basis, ESRD facilities essentially face no
actual penalty as long as they correct any problems identified. Part of the
reason is that HCFA has chosen not to exercise its authority to levy certain
sanctions. HCFA has not instituted procedures to deny Medicare payments for
dialysis if a facility does not meet dialyzer reuse standards. However, in
practice, other sanctions now available to HCFA have little application
because either they are restricted to less serious deficiencies or, in the
case of more serious deficiencies, facilities can take corrective action,
even temporarily, and avoid them altogether.

One way to give facilities more of an incentive to stay in compliance is to
have available the kinds of monetary penalties that can be used when nursing
homes are found to have severe or repeated serious deficiencies. For
example, HCFA can fine nursing homes, and the fines are not forgiven when
the facility corrects its problems. We have previously reported that such
penalties can give nursing homes a strong incentive to remain in compliance
with Medicare standards. Making such financial penalties more applicable to
ESRD facilities would require action by the Congress.

Another way to strengthen oversight is for state agencies and the ESRD
networks to share information on complaints and known quality-of-care
problems at specific facilities. Doing so would help target inspection
resources where they are most needed. HCFA's efforts to use available
outcome data for targeting its survey efforts may also eventually help in
this regard, but more testing and evaluation are needed to ensure that the
data used are sufficient to predict noncompliance with Medicare quality

We recommend that the Administrator of HCFA take the following actions to
strengthen oversight of ESRD facilities:

� Develop procedures on how and when to use HCFA's existing authority to
impose partial or complete payment reductions for ESRD facilities that do
not meet Medicare quality standards for dialyzer reuse.

� Establish procedures to facilitate better and more routine cooperation and
information sharing between ESRD networks and state survey agencies,
particularly in targeting facilities for on-site surveys.

� Evaluate the results of HCFA's project for using clinical outcome data to
select facilities for on-site review before it recommends that states use
such data as a key factor in the selection process. A central component of
the evaluation should be determining the extent to which the data are
sufficient to predict which facilities have a higher likelihood of not
complying with Medicare's conditions of participation.

To improve ESRD facilities' incentives to maintain compliance with
Medicare's conditions of participation, the Congress should consider
authorizing HCFA to assess monetary penalties on ESRD facilities like those
it is authorized to assess on nursing homes that have severe or repeated
serious deficiencies.

In commenting on the report, HCFA agreed with the report's findings and
expressed overall agreement with its recommendations. HCFA cited a number of
steps it intends to take or that are already under way to address our
recommendations. HCFA also pointed to a variety of patient outcome measures
over the last several years as evidence of improved overall quality of ESRD
treatment. While these data are encouraging about nationwide quality, they
do not mean that particular facilities are not problematic. This is
evidenced by the fact that the number of facilities found to be out of
compliance with Medicare conditions of participation increased from 6
percent in 1993 to 15 percent in 1999.

Regarding the recommendation about sanctions for inappropriate dialyzer
reuse, HCFA stated that it would develop necessary regulations and
procedures to implement such sanctions. In response to our recommendation to
facilitate cooperation among state agencies and ESRD networks, HCFA stated
that it is now taking steps to clearly delineate responsibilities of state
survey agencies and ESRD networks that would encourage cooperative
information-sharing to help identify poor-performing facilities.

Regarding our recommendation to evaluate whether outcome data are an
appropriate means of selecting facilities for on-site surveys, HCFA stated
that this process is already under way. HCFA cited an analysis of recent
data on facilities in Texas that indicated a strong relationship between
state survey results and outcome measures. We have included information in
the report about this analysis. However, we believe additional testing and
evaluation are needed before outcome measures are used as a significant
factor in selecting ESRD facilities for survey. HCFA stated its intention to
continue studying this issue.

HCFA did not specifically comment on our suggestion that the Congress
consider authorizing it to assess monetary penalties on ESRD facilities
similar to those authorized for nursing homes. However, HCFA did state that
it was pursuing a legislative strategy to consolidate and clarify current
alternative or intermediate sanctions and possibly establish new authorities
across all provider types.

HCFA also provided detailed technical comments, which we incorporated in the
report where appropriate. HCFA's comments are in appendix IV.

As agreed with your offices, we will make no further distribution of this
report until 4 days after its issue date. At that time, we will send copies
to the appropriate authorizing committees; the Honorable Nancy-Ann Min
DeParle, Administrator of HCFA; and interested congressional committees. We
will also make copies available to other interested parties.

Please contact me at (202) 512-7119 if you have any questions about this
report. Major contributors included Margaret Buddeke, Timothy Bushfield, and
Mark Ulanowicz, under the direction of Frank Pasquier.

Janet Heinrich
Associate Director, Health Financing and
Public Health Issues

Scope and Methodology

In order to evaluate the procedures and processes employed by HCFA, state
survey agencies, and ESRD networks to monitor dialysis facilities, we
interviewed (1) HCFA officials at its central office and four regional
offices; (2) state survey officials in California, New Jersey, Texas,
Oregon, and Washington; (3) ESRD network officials in five networks; and (4)
officials from the Network Forum, which is the organization that represents
all of the ESRD networks. We also collected data on the policies and
procedures used by HCFA, state survey agencies, and ESRD networks to monitor
dialysis facilities. We judgmentally selected these five states because they
appeared to be typical based on available data on clinical outcome measures
for each ESRD network and HCFA data on the number of
condition-of-participation deficiencies. We also considered other factors,
such as networks with larger states and more surveys, networks in which
innovative monitoring practices were being employed, and networks with a mix
of geographic oversight responsibility (networks with small geographic
areas, large geographic areas, and multistate coverage). Within each network
we selected and visited state survey agencies in the largest states. We
reviewed and obtained documentation on facility surveys from HCFA and state
agencies and clinical performance data collected by ESRD networks. We also
analyzed data on the results of state surveys and the clinical outcomes of
dialysis treatments from national databases.

To determine the extent to which on-site inspections of dialysis facilities
are done to ensure compliance with Medicare quality standards, we analyzed
HCFA's nationwide database of health care facility inspection results--the
On-Line Survey, Certification, and Reporting (OSCAR) system. This data
system records state survey results in a standard format. We analyzed data
to identify the level of survey activity over time and to determine the
extent that survey resources are spent on recertification surveys or initial
surveys. We analyzed the frequency of citation of condition-of-participation
deficiencies, which, unless corrected, are severe enough to warrant a
facility's termination from the Medicare program. Determinations of such
deficiencies are made by state agencies and receive HCFA's concurrence.
Although we did not thoroughly assess the reliability of the database for
the purpose of analyzing the frequency of recertification surveys, HCFA
officials generally recognize it to be reliable for this purpose. However,
the extent to which the data provide a consistent measure of quality of care
across states is unknown. To make such a determination would require a
review of the consistency of state survey processes nationally, which was
beyond the scope of our work.

To determine the effectiveness of the processes used to ensure that
facilities correct identified deficiencies, we reviewed the procedures used
by state agencies and networks to require corrective actions and to evaluate
whether facilities return to compliance. To gain more insight into the
effectiveness of HCFA's procedures to ensure sustained compliance with
quality-of-care standards, we looked particularly at the cases in which
state agencies and/or ESRD networks knew about facilities that had serious
and recurring problems. We reviewed the enforcement tools HCFA has available
to address noncompliant facilities and assessed the extent to which these
tools are utilized. We also analyzed HCFA data to identify the number of
facilities that were terminated from the program.

In assessing HCFA's efforts to improve the targeting of facilities to
inspect and monitor, we focused on HCFA's ongoing pilot project to profile
facilities using a variety of facility-specific data. Because this project
is in process and no strong indicators currently exist that identify
facilities with quality-of-care problems, it is difficult to assess the
overall effectiveness of this approach as a tool to identify noncompliant
facilities. Instead, we assessed the limitations of the data that HCFA is
planning to use to target facilities for on-site inspections. To this end,
we reviewed the data HCFA plans to use and discussed data reliability issues
with ESRD networks, HCFA researchers, noted renal care researchers, and the
peer review organization that has contracted with HCFA to develop the pilot
program. In addition, we discussed with state survey agency, ESRD network,
and HCFA officials the extent to which state agencies and ESRD networks
share information and coordinate their oversight activities.

Comparison of New ESRD Patients by Age and Primary Diagnosis, 1989, 1993,
and 1997

                   1989                  1993                 1997
                            Percentage           Percentage            Percentage
                   Patients              Patients             Patients
                            of total             of total              of total
 Under 15          430      1.0          475     0.8          583      0.7
 15-24             1,309    3.0          1,337   2.3          1,373    1.7
 25-34             3,435    7.8          3,652   6.2          3,833    4.8
 35-44             4,649    10.6         5,840   10.0         7,080    9.0
 45-54             5,850    13.3         7,846   13.4         10,936   13.8
 55-64             9,100    20.8         11,383  19.4         15,317   19.4
 65-74             11,978   27.3         16,964  28.9         22,056   27.9
 75 or older       7,090    16.2         11,127  19.0         17,924   22.7
 Total             43,841   100          58,624  100          79,102   100
 Primary diagnosis
 Diabetes          14,404   32.9         21,319  36.4         33,096   41.8
 Hypertension      12,786   29.2         17,333  29.6         20,066   25.4
 Glomerulonephritis5,863    13.4         6,439   11.0         7,390    9.3
 Cystic kidney     1,307    3.0          1,624   2.8          1,772    2.2
 Other urologic    772      1.8          888     1.5          1,388    1.8
 Other cause       4,453    10.2         5,400   9.2          8,284    10.5
 Unknown cause     2,209    5.0          2,621   4.5          2,920    3.7
 Missing cause     2,047    4.7          3,000   5.1          4,186    5.3
 Total             43,841   100.0        58,624  100.0        79,102   100.0

Source: National Institutes of Health (NIH), United States Renal Data
System, USRDS 1999 Annual Data Report (Bethesda, Md.: NIH, Apr. 1999); and

Medicare Conditions of Participation for Dialysis Facilities

 Condition of              Number of
 participation             standards and       Description
                                               The facility and personnel
                                               employed by the facility
 Compliance with federal,                      must be licensed as required
 state, and local laws and 4                   by federal, state, or local
 regulations                                   laws. This includes
                                               compliance with all public
                                               safety laws and
                                               The facility must be under
                                               the control of an
                                               identifiable body that
 Governing body and                            adopts and enforces rules
 management                70                  and regulations, including
                                               operational rules and
                                               patient care policies to
                                               safeguard the health and
                                               safety of individuals.
                                               A professional,
                                               multidisciplinary health
                                               care team and the patient
                                               must develop a written
                                               long-term-care plan to
                                               ensure each patient receives
                                               the appropriate type of
 Patient long-term-care                        dialysis and care. Patient
 program and patient care  20                  care plans, which have
 plan                                          shorter time lines, must be
                                               personalized for each
                                               patient to address their
                                               specific medical,
                                               psychological, social, and
                                               functional needs. Both plans
                                               are to be regularly reviewed
                                               and updated to respond to
                                               changing patient needs.
                                               Dialysis facilities must
                                               have written policies
                                               describing the rights of the
                                               patients in order to ensure
                                               patients are fully informed
 Patients' rights and                          about the services
 responsibilities          12                  available, their medical
                                               condition, whether the
                                               facility reuses dialysis
                                               supplies, and whether the
                                               patient is a candidate for
                                               transplantation and home
                                               Patient medical records must
                                               be maintained to document
 Medical records           21                  patient assessments,
                                               diagnosis, and treatment,
                                               and medical and nursing
                                               Dialysis services are to be
                                               provided in a setting that
 Physical environment      29                  is functional, sanitary,
                                               safe, and comfortable for
                                               patients, staff, and the
                                               Facilities that reuse
 Reuse of hemodialyzers                        hemodialyzers and other
 and other dialysis        92                  dialysis supplies must
 supplies                                      follow established protocols
                                               and standards to ensure
                                               patient and staff safety.
                                               Agreements between dialysis
                                               facilities and inpatient
                                               dialysis centers must be in
 Affiliation agreement or                      writing to ensure inpatient
 arrangement               4                   care and other hospital
                                               services are promptly
                                               available to dialysis
                                               Dialysis treatments must be
                                               under the general
 Director of renal                             supervision of a qualified
 dialysis facility         6                   director, who is responsible
                                               for planning, organizing,
                                               conducting, and directing
                                               professional services.
                                               Properly trained and
                                               qualified personnel must be
 Staff of a renal dialysis                     present in adequate numbers
 facility or center        6                   to meet the needs of
                                               patients, including needs
                                               arising in emergencies
                                               Dialysis facilities must
                                               provide dialysis services as
 Minimal service                               well as laboratory, social,
 requirements              27                  and dietetic services needed
                                               to address ESRD patient

Comments From the Health Care Financing Administration


Table 1: Number and Percentage of Dialysis Facilities Resurveyed,
1993-99 10

Table 2: Top Five Conditions of Participation Identified as Deficient
and Their Potential Adverse Effects, 1993-99 12

Table 3: Overview of Enforcement Tools Available to HCFA 17

1. National Institutes of Health (NIH), United States Renal Data System
(USRDS), USRDS 1999 Annual Data Report (Bethesda, Md.: NIH, Apr. 1999), p.

2. A dialyzer is a filter that is used to clean waste material from the
patient's blood. Dialyzers can be used multiple times on the same patient if
dialysis facilities establish procedures--that comply with Medicare
standards--to clean and disinfect dialyzers after each use.

3. These data are based on our analysis of recertification surveys only.

4. These inspections included initial surveys, recertification surveys, and
surveys required before the facility can initiate a dialyzer reuse program.

5. Both the minimum and maximum estimates assume that the state would survey
10 percent of its facilities each year (the HCFA goal at the time). The
minimum estimate assumes that the 11 out-of-compliance facilities were
surveyed first, and the maximum estimate assumes they were surveyed last.

6. We use the term "sanctions" in this report to refer to all of the
penalties available for noncompliance, including denial of Medicare payments
and termination from the Medicare program.

7. This figure includes both recertification surveys and complaint surveys.

8. One additional facility voluntarily withdrew from Medicare because of the
threat of termination. While HCFA's deficiency data identified 12 facilities
involuntarily terminated, we excluded those terminations that were not
linked with a facility's failure to correct condition-of-participation

9. Only a facility's four most recent surveys are included in HCFA's survey

10. Federal surveyors from the HCFA regional office accompanied the network
surveyors on one of the facility visits and observed many of these problems.

11. Two other facilities voluntarily withdrew from Medicare before HCFA
could consider network recommendations. In one case, a facility failed to
improve and sustain improvement in removing an adequate amount of
contaminants from patients' blood. After considerable monitoring and various
approaches to improving the facility's performance over an 8-month period,
the network recommended that HCFA impose an alternative sanction. Although
the facility withdrew from Medicare before the recommendation could be
considered, the HCFA project officer stated that because the issues involved
patient health and safety, HCFA could not pursue alternative sanctions.

12. Nursing Homes: Additional Steps Needed to Strengthen Enforcement of
Federal Quality Standards (GAO/HEHS-99-46 , Mar. 18, 1999).

13. Robert Wolfe, Facility Statistics, Patient Care and Science: A
Re-evaluation of Network 14 State Surveyor Data, a presentation to the HCFA
Dialysis Facility-Specific Reporting Workgroup, Aug./Sept. 1999.

14. ESRD Network 14 Medical Review Board, Position Paper on the Use of
Outcomes Data for Survey Selection Purposes (Dallas, Tx.: June 2, 2000).

15. PRO-West, Developing Clinical Performance Measures for the Care of
Patients With End Stage Renal Disease, final report to HCFA (Seattle, Wash.:
PRO-West, Jan. 1999).

Appendix I: FDA's List of Frequently Reprocessed SUDs


Appendix II: Comments From the Department of Health and Human Services


Table 1: Surveys of SUD Reprocessing 9

CDC Centers for Disease Control and Prevention

EP electrophysiology

FDA Food and Drug Administration

GI gastrointestinal

HCFA Health Care Financing Administration

HHS Department of Health and Human Services

HIV human immunodeficiency virus

JCAHO Joint Commission on Accreditation of Healthcare Organizations

MDR Medical Device Reporting

PMA premarket approval

SUD single-use device

VA Department of Veterans Affairs

Health, Education, and
Human Services Division


June 20, 2000

The Honorable James M. Jeffords
The Honorable Edward M. Kennedy
Ranking Minority Member
Committee on Health, Education, Labor, and Pensions
United States Senate

The Honorable Bill Frist
Subcommittee on Public Health
Committee on Health, Education, Labor, and Pensions
United States Senate

The Honorable Mike DeWine
The Honorable Christopher Dodd
The Honorable Richard Durbin
The Honorable Mike Enzi
The Honorable Patty Murray
The Honorable Jack Reed
United States Senate

Medical devices approved for sale in the United States as single-use devices
(SUD) sometimes are reprocessed and used again on other patients.1
Reprocessing involves cleaning and sterilizing a device and verifying that
it functions properly. Some reprocessed devices are relatively simple items
for external use, such as sequential compression devices (inflatable sleeves
to improve blood circulation), while others are complex and invasive, such
as catheters that are inserted into the heart to monitor cardiac
functioning. Some devices, both SUDs and those marketed as reusable, have
been reprocessed in-house by hospitals and other treatment facilities for
decades. An industry of third-party medical device reprocessing companies
has developed within the last 10 years. The practice of SUD reprocessing
raises public health concerns, primarily regarding the potential risks of
infection and device malfunction, and has led to complaints by the original
device manufacturers that the Food and Drug Administration (FDA), the
federal agency within the Department of Health and Human Services (HHS) that
approves medical devices for marketing, has not maintained consistent
regulatory standards for different types of medical device companies.

When manufacturers seek approval to market a device as single-use, FDA
cannot require them to show that reusing the device would be inappropriate
or hazardous. Manufacturers that intend to market a device as reusable must
give FDA supporting data demonstrating to the agency's satisfaction that a
device can be cleaned and sterilized without impairing its function. All
other devices are intended for single-use. Thus, a device may be marketed
for single-use because the manufacturer believes that it cannot be safely
and reliably used more than once or because the manufacturer chooses not to
conduct the studies needed to prove that the device should be labeled as
reusable. In effect, because FDA can only evaluate a device relative to the
use intended for it by its manufacturer, its approval of a device as
single-use means that a device can be used safely and reliably once, not
necessarily that it cannot be used safely and reliably more than once if it
is appropriately reprocessed.

You asked us to review the practice of SUD reprocessing in the United
States. We focused our work on (1) the extent of SUD reprocessing, (2) the
health risks associated with SUD reprocessing, (3) the cost savings from SUD
reprocessing, and (4) FDA's oversight of SUD reprocessing. We looked only at
the practice of reprocessing SUDs for use on another patient; we did not
examine devices approved for marketing as reusable, the resterilization of
opened but unused devices, or devices reprocessed for additional use on the
same patient. To conduct our work, we reviewed the relevant scientific
literature, met with FDA officials, reviewed FDA documents and documents
submitted to FDA by interested parties; interviewed officials at the Centers
for Disease Control and Prevention (CDC) and the Health Care Financing
Administration (HCFA); gathered information from other experts in government
and industry; contacted third-party reprocessing companies; and interviewed
physicians, hospital administrators, and other health care providers. We
conducted our work between November 1999 and May 2000 in accordance with
generally accepted government auditing standards.

While it is clear that some health care institutions have chosen to
reprocess and reuse some kinds of SUDs, accurate and comprehensive
information about the number of facilities that use reprocessed SUDs and the
types of SUDs that are reprocessed is not available. Surveys by professional
associations and other groups have found that approximately 20 to 30 percent
of American hospitals reported that they reuse at least one type of SUD and
that at least one-third of the hospitals that do so contract with
third-party reprocessing companies. Most hospitals using reprocessed SUDs
reuse only a few types of devices. It is likely that some hospitals do not
report their use of reprocessed SUDs, and the estimates do not fully include
ambulatory surgery centers or physicians' practices that also may reuse

Although SUD reprocessing does pose theoretical health risks, clinical
evidence shows that certain devices can be reprocessed safely. In addition,
some infection control experts told us that the careful reprocessing of
appropriate SUDs has not been demonstrated to be a public health risk.
Almost all of the professional associations we contacted believe that
selected devices can be reprocessed safely if appropriate procedures are
followed and closely monitored. We found that several reports of patient
adverse events allegedly due to SUD reprocessing that we identified were
inaccurate or not relevant to the debate. However, this does not mean that
SUD reprocessing is always safe. Current surveillance systems almost
certainly do not detect all infections and injuries resulting from the use
of reprocessed SUDs (or from the use of medical devices in general).
Furthermore, FDA, device manufacturers, and third-party reprocessors
generally agree that many types of SUDs cannot be safely cleaned and
sterilized, and even for devices that usually can be reprocessed, some
models are impossible to clean and sterilize effectively.

Substantial cost savings can be achieved by reprocessing SUDs. Independent
reprocessing firms charge hospitals approximately one-half the price of a
new device for a reprocessed device, while the in-house cost of reprocessing
some devices can be less than 10 percent of the price of a new device. The
competition created by SUD reprocessing appears to have caused some original
device manufacturers to reduce their prices to certain purchasers.

FDA's regulation of SUD reprocessing for different types of device
manufacturers has been inconsistent, but the agency is about to institute a
new regulatory framework intended to address this concern. Currently,
although third-party reprocessing firms are considered by FDA to be
manufacturers of reusable medical devices, they are not required to seek
premarket approval to reprocess SUDs, and FDA until now has chosen not to
exercise its jurisdiction over hospitals and other health care institutions
that reprocess SUDs in-house. Under the revised framework, independent
reprocessing firms and hospitals will have to obtain FDA's approval before
they can reprocess many devices labeled for single-use. The revised
regulatory framework will give FDA more information about SUD reprocessing
and strengthen its oversight of reprocessing. However, there are significant
barriers to the framework's successful implementation. FDA told us that the
additional work involved in reviewing applications for SUD reprocessing may
interfere with the agency's ability to complete timely reviews of premarket
applications for new medical devices. Also, the framework will involve the
agency in regulating SUD reprocessing practices in hospitals. To get help
with the monitoring of reprocessing in these facilities, FDA has asked HCFA
and the Joint Commission on Accreditation of Healthcare Organizations
(JCAHO) to include SUD reprocessing in their hospital quality-of-care
standards. Neither HCFA nor JCAHO has committed to do this in the near term.

FDA is responsible for ensuring the safety and effectiveness of medical
devices sold in the United States, ranging from bandages and thermometers to
cardiac catheters and artificial hearts. Approximately 80,000 to 100,000
models of medical devices are currently in use in the United States, and the
domestic market for medical devices totaled roughly $56 billion in 1999. FDA
regulates the safety and effectiveness of medical devices, the packaging and
labeling that describes how they should be used, and the facilities that
manufacture them. FDA's requirements for approving devices for marketing
depend on the device's potential for harming patients. Class I, or low-risk,
devices include such things as elastic bandages and orthopedic saw blades.
Medium-risk devices, class II, include items like urethral catheters and
blood pressure cuffs. Class III devices, such as heart valves and balloon
angioplasty catheters, support or sustain human life and present significant
risk of patient injury. Most class I devices can be marketed without
obtaining prior approval from FDA. FDA requires the manufacturers of most
class II and class III devices to submit either a premarket notification
application (510(k)) to show that the device is substantially equivalent to
one already on the market or an application for premarket approval (PMA),
which provides evidence, often including clinical data, demonstrating that
the device is safe and effective. FDA

requires a satisfactory inspection of a manufacturer's facilities before a
PMA is approved.2 FDA received 4,458 510(k) applications and 72 PMAs for
medical devices in fiscal year 1999. Manufacturers are required to register
with FDA, list the devices they produce, follow good manufacturing
practices, and inform FDA about problems with their marketed devices.
Manufacturers also are subject to inspection by FDA. In fiscal year 1999,
FDA inspected 30 percent of the nearly 3,500 domestic manufacturers of class
II and class III medical devices.

Generally, FDA can evaluate applications to market new devices only in terms
of a device's intended use as described on its label.3 Thus, manufacturers
that wish to market a device as single-use need to convince FDA only that it
can be used safely and effectively once--they do not need to demonstrate
that the device cannot be used more than once. Conversely, manufacturers
that wish to market a device as reusable must either provide data
demonstrating that the device will be safe and effective for a specified
number of uses or provide a measure to determine whether or not a
reprocessed device still meets performance specifications. They must show
that the device can be cleaned and sterilized and that its function will not
degrade with repeated uses. Devices that are not marketed as reusable are
intended for single-use.4

Single-use device reprocessing is a small part of the large and varied
medical device industry. FDA has confirmed the existence of only a small
number of third-party reprocessing firms, although it suspects there are
more. Surveys by professional associations and other groups have
consistently found that roughly 20 to 30 percent of American hospitals
reported reusing at least one type of SUD. Many health care professionals
told us that they believe manufacturers market devices with a single-use
label because of the economic benefits of doing so and that they therefore
believe many SUDs can be reused.

Third-party reprocessing firms and some hospitals and other health care
institutions reprocess SUDs. While the exact size of the reprocessing
industry is unknown, it is clearly only a small part of the medical device
industry. For example, FDA has identified only 13 third-party reprocessing
companies, although it suspects that more are in operation. Last year, a
trade association representing major third-party firms said that its members
collectively received about $20 million annually for their services.
Although there are many hospitals (more than 6,000) and ambulatory surgery
centers (about 2,700) in the United States, evidence indicates that only a
minority of them reprocess SUDs in-house. Furthermore, reprocessors (both
third-party firms and hospitals) each typically reprocess only a few types
of SUDs.

While it is clear that some health care institutions have chosen to
reprocess and reuse some kinds of SUDs, neither FDA nor any other
organization has accurate and comprehensive information about the number of
facilities that use reprocessed SUDs or the types of SUDs that are
reprocessed. Table 1 presents the results of six surveys about SUD
reprocessing by professional associations and other groups. The surveys
typically asked members of selected professional groups to describe the SUD
reprocessing practices at the institution with which they were affiliated.
Most of the surveys found that approximately 20 to 30 percent of American
hospitals reused at least one type of SUD and that at least one-third of the
surveyed hospitals that reused SUDs contracted with independent reprocessing
companies. While the results of the various surveys are fairly consistent,
it is difficult to assess the validity of the findings because the response
rates for the surveys are low. The only survey with a response rate greater
than 50 percent, from the Metropolitan Chicago Healthcare Council,
unfortunately does not have separate results for the reprocessing of SUDs
used on patients--its findings combine SUD reprocessing and the
resterilization of opened but unused devices.

                                                Percentage of
                                                institutions Percentage of
 Source         Sample            Date Response that         that have
                                       rate     reprocess
                                                SUDs used on third-parties
                                                patients     reprocess SUDs
 ECRIa          subscribers to    1996 (N=more  31           7
                ECRI publications      than
                Random sample of
                membership of the
 R.J. Cheung andAmerican Society       40%
 othersb        for               1997          29c          d
                Gastrointestinal       (N=294)
 Society for                           22%
 Healthcare     ASHCSP members    1998          21           7
 Central Service                       (N=214)
                Random sample of       44%
 OR Manager f   readers of        1999          Between 18   24g
                publication            (N=132)  and 31
 Chicago                               72%
 Healthcare     Member hospitals  1999          20g          17g
 Councilh                              (N=71)
                Membership of the
 J.W. Birk and  Society of             46%
 othersi        Gastroenterology  1999          16c          d
                Nurses and             (N=223)

aECRI, Special Report: Reuse of Single-Use Medical Devices: Making Informed
Decisions (Plymouth Meeting, Pa.: ECRI, 1997), pp. 85-86.

bR.J. Cheung and others, "GI Endoscopic Reprocessing Practices in the United
States," Gastrointestinal Endoscopy , Vol. 50, No. 3 (1999), pp. 362-68.

cGI endoscopic instruments only.

dQuestion not included in survey.

eASHCSP, presentation at conference, "The Re-Use of Single-Use Devices:
Practice, Patient Safety and Regulation," May 5-6, 1999.

fOR Manager , Vol. 15, No. 9 (1999), pp. 1, 11, 14.

gIncludes opened but unused SUDs in addition to SUDs that have been used on
a patient.

hMetropolitan Chicago Healthcare Council survey.

iJ.W. Birk and others, "A National Survey on the Re-Use Patterns of `Single
Use Only' ERCP Supplies," Gastrointestinal Endoscopy , Vol. 49, No. 4
(1999), p. AB139.

The estimate of 20 to 30 percent may be low because some hospitals and other
health care facilities do not report this practice. For example,
representatives of manufacturers told us that their examination of SUDs that
malfunctioned indicated that devices from some hospitals had clear evidence
of reprocessing, even though the institutions denied reusing SUDs.
Similarly, an official of the Department of Veterans Affairs (VA) told us
that some VA hospitals reuse single-use cardiac catheters, even though SUD
reprocessing is contrary to VA policy.5 The surveys also may not completely
capture the use of reprocessed SUDs in ambulatory surgery centers,
physicians' practices, or other nonhospital institutions.6

The frequency of reprocessing varies widely among different devices, and
most hospitals that reuse SUDs reuse only a few types of devices. For
example, electrophysiology (EP) catheters (devices inserted into the heart
to measure and correct cardiac rhythm disorders) have been reprocessed for
20 years, even though all models of them were approved for single-use only.
Some types of EP catheters are relatively easy to clean (because they do not
have long, hollow tubes), sterilize, and test. They also are expensive
(ranging roughly from $100 to $1,500), and a typical EP procedure could
involve the use of several catheters. In the course of our work, we
contacted representatives of nine EP centers; seven of them acknowledged
using reprocessed EP catheters. Several hospitals told us that EP catheters
were among the very few SUDs they reused. Conversely, gastrointestinal (GI)
biopsy forceps are more difficult to reprocess. The forceps are long and
have hollow tubes and delicate mechanisms that make them harder to clean and
sterilize. We contacted physicians from 17 gastroenterology centers, and
none of these physicians said that their facilities reused GI biopsy
forceps. (A list compiled by FDA of frequently reprocessed SUDs is in app.

Many health care personnel believe that some SUDs can be reused. They told
us they distrust the single-use label for some devices because (1) FDA
cannot require manufacturers to support the designation of a device as
single-use, (2) they perceive that manufacturers have an economic incentive
to market devices as single-use that could just as well be sold as reusable,
and (3) FDA's approval requirements for SUDs are less extensive than those
for reusable devices. In addition, the application of the single-use label
to noncritical medical devices erodes its meaning for some health care

On occasion, manufacturers have contributed to the sense that compliance
with the single-use label is not always necessary. We found three examples
of this. First, a major manufacturer of pulse oximeter sensors (devices that
measure blood oxygen levels) has a program to "recycle" the sensors,
essentially offering to sell "remanufactured" sensors for reduced prices to
health care institutions that return their used single-use sensors to the
company. These SUDs are also reprocessed by third-party reprocessing firms,
which cite the manufacturer's recycling program as evidence that the
single-use label on these devices is not meaningful. Second, in a 1998 U.S.
District Court case, the judge found that the manufacturer's only purposes
in labeling a device for single-use were to comply with FDA's requirements
and to limit its own liability from reuse, not to prevent a hospital from
using it more than once.8 Third, manufacturers have written letters to
hospitals containing detailed instructions for the sterilization of SUDs.
The letters typically caution against resterilizing the SUD and then give
detailed sterilization instructions. Some of the letters note that the
sterilization procedures may be used for open but unused devices, but others
do not include that restriction. One letter volunteered that the
manufacturer had verified that the device could be resterilized three times,
although the manufacturer had not tested devices that had been used on
patients. One hospital we talked with used the instructions in the letters
as guidelines for its in-house reprocessing program.


While SUD reprocessing does pose some theoretical health risks, the
available evidence indicates that some SUDs can be safely reprocessed and
reused on other patients. The safety of reprocessing some types of devices,
such as some types of EP catheters, is supported by a well-developed
clinical literature. The infection control and patient safety experts we
consulted told us that the reprocessing of certain SUDs is not a
demonstrated public health risk, and SUD reprocessing is seen as safe by
many associations representing health care professionals. Several reports of
patient adverse events allegedly related to SUD reprocessing that we
investigated were inaccurate, not relevant to the debate, or difficult to
interpret. However, this does not mean that SUD reprocessing is always safe.
For example, some reports of nonsterile reprocessed SUDs merit further
investigation, and current surveillance systems are unlikely to detect all
infections or injuries caused by reprocessed SUDs.

To successfully reprocess a device that has been used on a patient,
institutions must be able to clean it thoroughly, sterilize it to acceptable
standards, and ensure that reprocessing and reuse will not degrade its
functioning. Cleanliness is important because even measurably sterile
devices can harbor biological material from previous uses that may prove a
health risk for subsequent patients. Potentially, this biological residue by
itself can prove toxic to new patients, and it also can form a crust to
shield harmful bacteria from sterilization procedures. For these reasons,
reprocessors assert that they choose devices to reprocess carefully,
rejecting those that cannot be cleaned thoroughly or that are damaged by
sterilization or reprocessing procedures.

The reprocessors we contacted, both third-party firms and hospitals,
followed similar reprocessing procedures. Devices to be reprocessed are
collected following established procedures and are frequently rinsed or
otherwise cleaned soon after use, before they are sent to the reprocessing
facility. There the devices are cleaned, refurbished, inspected, and
sterilized. The third-party reprocessors told us that they check the
function of every device before it is sterilized and returned to the client.
These firms also told us that they do not mix devices from different
hospitals--each hospital receives only devices from the batch it sent to the
reprocessor. According to the reprocessors, many devices are rejected during
reprocessing because they have been damaged, even among device models that
are especially amenable to reprocessing. The reprocessors also told us that
they or their client health care facilities set limits on the maximum number
of times an individual device can be reused. They also said that they keep
track of the number of uses for each device and discard the devices when the
limit is reached.

Reprocessors also resterilize open but unused SUDs. These are devices that
were opened in preparation for a surgical procedure--and are therefore no
longer sterile--but were not used on a patient. FDA, device manufacturers,
and hospitals all agree that there is less risk in reprocessing these
devices. Manufacturers told us that hospitals frequently ask them for
sterilization instructions for opened but unused SUDs and that the
manufacturers either provide instructions or advise the hospitals that the
devices cannot be resterilized.


To assess the health risks of reprocessed SUDs relative to the risks from
new devices, it would be best to compare the rates of patient injuries and
illnesses caused by each. Unfortunately, neither comprehensive data about
the numbers of adverse events caused by medical devices nor data about the
numbers of patients exposed to particular devices exist today. Furthermore,
new medical devices are not always perfect, and some patient injuries or
infections are caused by SUDs at their first use. This means that individual
cases of adverse events associated with reprocessed SUDs are not informative
because we do not know how often these events occur with new SUDs.
Therefore, to assess the safety of reprocessed SUDs, we evaluated
information from a variety of complementary, but less than ideal, data

Four types of information indicate that some devices labeled for single-use
can be reprocessed safely and used again on other patients. First, the
safety of reprocessing some types of devices has been established by
well-developed clinical studies. Studies have shown both that reprocessing
procedures can be safely accomplished and that patient outcomes are not
adversely affected by the use of reprocessed SUDs. For example, several
studies have documented the safe reprocessing and reuse of EP catheters. One
study of more than 14,000 EP procedures found that the overall rate of
patient infections was very low and did not differ between clinical centers

that reused EP catheters and centers that used each catheter only once.9 A
later study of 69 EP catheters used in 336 procedures concluded that
carefully reprocessing one model of single-use catheter up to 5 times posed
no increase in health risks.10 Similarly, some evaluations of the
reprocessing of single-use endoscopic instruments published in peer-reviewed
scientific journals found that those SUDs could be reused at least several
times without increasing patient risk.11

Second, the hospital infection control practitioners, risk management
executives, and patient safety experts we interviewed all told us that
careful reprocessing of the types of SUDs that can be properly cleaned and
sterilized does not pose a risk to patient health. For example, the hospital
infection control practitioners we contacted told us that all types of
infectious bacteria and some key viruses (including human immunodeficiency
virus (HIV) and hepatitis C) can be destroyed if devices are properly
cleaned and sterilized and that they were not aware of any infections
resulting from the reuse of SUDs in their hospitals. Hospital infection
experts at CDC told us that the evidence showed that SUD reprocessing poses
minimal, if any, public health risk. The CDC experts said that they were not
aware of patient illnesses caused by SUD reuse in the last decade.12 The
head epidemiologist of CDC's Hospital Infection Program told us that
although CDC does not specifically monitor SUD reuse, he was confident
hospital infection surveillance systems would have uncovered infections
resulting from SUD reuse if they had occurred. Risk management professionals
told us that the hospitals they worked with had not received any claims of
patient injury caused by the use of reprocessed SUDs. An official of a
health quality consulting organization told us that his firm could find no
evidence in its databases that treating patients with reprocessed SUDs was
more dangerous than using new devices.

Partly because of the clinical literature and expert opinion just described,
with the exception of groups representing device manufacturers, all of the
professional organizations with positions on SUD reuse that we contacted or
that submitted comments to FDA on the agency's regulatory proposal expressed
at least qualified support for this practice. None sought to ban SUD
reprocessing, although some supported FDA's plan to more closely regulate
SUD reprocessing. These organizations included groups representing
physicians, nurses, in-hospital sterilization professionals, infection
control practitioners, and health care facilities.13 Some of these
associations and other organizations have issued guidelines to help
hospitals develop SUD reprocessing programs. The guidelines usually give
advice about assessing the costs and benefits of reprocessing SUDs, choosing
devices that can be reprocessed safely, and evaluating the services offered
by third-party reprocessing companies.

Third, only a very small percentage of the reports FDA has received through
its Medical Device Reporting (MDR) program concerned patient adverse
outcomes associated with reused SUDs, although this program probably
underestimates the number of injuries with reprocessed SUDs.14 For a roughly
3-year period ending in December 1999, FDA's Manufacturer and User Facility
Device Experience database received nearly 125,000 reports of patient
injuries, device malfunctions, or other potential problems associated with
SUDs. FDA told us that 1,131 of those reports involved SUDs that had been
reprocessed but that nearly 700 of them concerned dialysis equipment that
was reprocessed for use on the same patient. Only 49 of the reports were for
SUDs that are included on FDA's list of frequently reprocessed devices (36
for malfunctions, 9 for injuries, and 4 for other

reasons), and it is not known whether those injuries were caused by
reprocessing, by device failure unrelated to reprocessing, or by some other
aspect of the medical procedure.15

Fourth, several of the reports we identified of patient adverse events
allegedly related to SUD reprocessing were inaccurate, not relevant to the
debate, or difficult to interpret. For example, it was reported that a
recent patient death occurred in a Colorado hospital as a result of an
infection transmitted by a reprocessed cardiac catheter. CDC officials told
us that this incident occurred some time ago in a hospital that did not use
reprocessed cardiac catheters. The infection ultimately was traced to
improperly sterilized glass medicine cups. Similarly, it was alleged that
SUD reuse caused increased rates of pneumonia in one group of children.16
This was supported by a study of home use of tracheostomy tubes in children
with breathing difficulties.17 This is not relevant to the current debate
because the reused tubes were cleaned at home with hydrogen peroxide,
vinegar, or soap and water for use on the same child, not reprocessed by
hospitals or third-party companies for use on other patients. Likewise, FDA
received a report that the tip of a reused EP catheter broke off and lodged
in a patient's heart. However, FDA also received two reports of similar
injuries resulting from procedures with new, not reprocessed, EP catheters.
In addition, a published report of blindness caused by the broken tip of a
reused EP catheter migrating to a retinal artery is difficult to interpret
because of the year and location of the case. It occurred overseas in 1984,
and the sterilization procedures used on the catheter were different from
the ones used today in the United States.

While the evidence shows that carefully controlled reprocessing of some SUDs
is safe, it is also clear that some SUDs cannot be safely reprocessed,
procedures for safe reprocessing are not always followed, and the
limitations of the information available about SUD reprocessing argue for
monitoring of the practice. FDA researchers, original device manufacturers,
and third-party reprocessors all agree that many types of SUDs cannot be
reprocessed safely. For example, the largest third-party reprocessing firm
told us that it reprocesses only 15 "families" of devices and that many of
these involve only the resterilization of opened but unused devices. There
is also agreement that, even for some categories of SUDs that can be
reprocessed, some models can be thoroughly cleaned and sterilized, while
others cannot. For instance, two third-party reprocessing firms told us that
they identify for clients particular device models that can be successfully
reprocessed. Thus, the hospitals and ambulatory surgery centers that
contract with these reprocessing firms can, in purchasing new devices that
they intend to reuse, avoid ones that the firms will not reprocess.

For devices that can be reprocessed safely, cleaning and sterilization
procedures are not always followed correctly. For example, a 1997 survey of
gastrointestinal endoscopy physicians found that about one-quarter of
endoscopic facilities failed to follow all of a professional association's
guidelines for cleaning and sterilizing endoscopic instruments.18 Four
percent of the respondents to that survey reported patient infections
associated with endoscopic procedures, although there was no indication that
the infections resulted from the use of reprocessed SUDs. Also, underlining
the potential risks of SUD reprocessing, infection outbreaks occur
occasionally that are due to sterilization failures for devices approved for
marketing as reusable. The outbreaks are detected in hospitals when
unusually high numbers of patients become ill with the same infectious
agent. For example, CDC reported that the failure of automatic cleaning
machines to properly clean bronchoscopes and endoscopes led to at least five
infectious outbreaks.19

For reprocessed SUDs, device manufacturers have forwarded to FDA reports of
allegedly damaged, unclean, or nonsterile devices taken from hospital stocks
that had been reprocessed by third-party reprocessing firms. FDA found that
at least one of these claims had merit. In March 1999, a manufacturer told
FDA that six of its reprocessed GI biopsy forceps it retrieved from a
Florida hospital were not sterile. The devices were labeled for single-use
only and had been reprocessed by a third-party reprocessing company. These
biopsy forceps are nearly 8 feet long, and the sterility testing procedure
used by the manufacturer involved cutting the devices into segments to allow
better access to the center portions of the hollow tubing. Using established
test procedures that did not segment the biopsy forceps, both FDA and the
reprocessing firm subsequently tested devices from the same lot and found
them to be sterile. FDA now believes that the sterility test protocol it
used was not the best one for these devices, and it is preparing a new
protocol. Although there is no evidence that these reprocessed devices have
harmed patients, this case demonstrates the possibility that some
reprocessed SUDs sterilized according to current protocols may not be free
of bacterial contamination.

Current surveillance systems for medical errors and adverse events almost
certainly do not detect all infections and injuries resulting from the use
of reprocessed SUDs (or from the use of medical devices in general). It is
well known that surveillance systems based on spontaneous reports by health
care providers and manufacturers are plagued by underreporting, incomplete
reports, and other problems.20 As we have previously reported, FDA's
surveillance system for medical devices--the MDR program--is no exception.21
For example, when FDA conducted a pilot test of a sentinel system for
gathering adverse event reports for medical devices (working closely with
selected health care institutions to increase reporting rather than trying
to maximize reporting among all institutions), it received adverse event
reports at a rate 10 times greater than the MDR program, even though MDR
regulations mandate the reporting of the same types of events. In addition,
FDA officials and infection control experts told us that it is often
difficult to identify the source of infections in individual patients, and
it is particularly difficult to trace infections back to the use of specific
medical devices.

Reprocessed SUDs cost less than new devices. Independent reprocessing firms
charge hospitals and ambulatory surgery centers approximately one-half of
the price of a new device for each reprocessed SUD, while three hospitals
that reprocess EP catheters in-house told us that their reprocessing costs
were less than 10 percent of the price of a new device. Although there is
some debate about how to calculate the true costs of reprocessing
(including, for example, the staff time needed to study the safety and
cost-effectiveness of reprocessing particular devices and to perform
quality-control functions), hospitals that use reprocessed SUDs told us that
they save money by doing so. For example, one group of hospitals that uses
reprocessed SUDs estimated that their annual savings per hospital were
$44,000 for sequential compression devices, $17,000 for pulse oximeter
sensors, and $115,000 for EP catheters. Other hospitals with active
cardiology services gave us higher estimates for their savings from reusing
EP catheters, ranging from $200,000 to $1 million annually.

The exact prices paid for new SUDs result from negotiations between
individual manufacturers and individual purchasers. The competitive
alternative offered by SUD reprocessing has affected negotiations between
manufacturers and purchasers and may have caused some manufacturers to lower
their prices to some purchasers. For example, we found evidence that
manufacturers sometimes offer lower prices to facilities that agree not to
reprocess. We obtained copies of marketing materials from a manufacturer of
single-use sequential compression devices offering to reduce prices if the
purchasing hospital signed a contract stipulating that it would not
reprocess the devices. For two hospitals we contacted, manufacturers offered
to reduce the price of new EP catheters by as much as one-half, matching the
price of third-party reprocessing, if the facilities would agree to not
reprocess the devices. A major third-party reprocessing firm told us that
some hospitals stopped using its services when offered this arrangement by
manufacturers. We were not able to determine how often manufacturers offer
these price breaks.

The overall prices of some SUDs that are reprocessed appear to have
decreased in recent years, even for health care institutions that do not
reuse them, although we were unable to attribute the price drops to
reprocessing. For instance, one third-party reprocessing firm showed us
evidence that the prices its clients paid for new GI biopsy forceps, pulse
oximeter sensors, and sequential compression devices had declined 20 percent
or more over the last few years. A manufacturer of one of these devices
confirmed to us that its prices had declined, although the company cited
reasons other than reprocessing. Similarly, a majority of the
gastroenterologists we contacted told us that the prices their facilities
paid for new single-use GI biopsy forceps had decreased, and none of them
reprocessed these devices.

by Manufacturers to Most SUD Reprocessors

FDA's current regulatory scheme results in uneven requirements for FDA
review. Its proposed revisions would treat most SUD reprocessors as
manufacturers. The proposed framework also would give FDA more information
about SUD reprocessing and strengthen its oversight of reprocessing.
However, there are significant barriers to the framework's successful

FDA's current regulation of SUD reprocessing represents a balance between
its regulatory obligations, its judgment that SUD reprocessing has not posed
a significant risk to the public health, and its limited resources. FDA
categorizes all entities that reprocess SUDs, including third-party
reprocessing firms, hospitals, and ambulatory surgery centers, as device
manufacturers,22 and therefore they are technically required to comply with
good manufacturing practices, FDA inspection, and manufacturers' adverse
events reporting regulations. FDA has enforced these provisions for
third-party reprocessing firms but not for hospitals and other health care
institutions that reprocess SUDs. In addition, because FDA has not required
reprocessors to seek premarket approval for reprocessing SUDs, technically
all reprocessors are engaged in the practice of selling adulterated medical
devices. However, because FDA has judged that SUD reprocessing has not posed
a significant risk to the public health, it has exercised "regulatory
discretion" to allow SUD reprocessing to continue and to focus its limited
enforcement resources in other areas. FDA has the authority to immediately
halt any practice that threatens the public's health and has not done so
with SUD reprocessing because it believes there has been no evidence that
the public health has been threatened.

Some manufacturers have complained that FDA's inconsistent enforcement of
premarket requirements has disadvantaged them relative to reprocessors.
Manufacturers that want to market a reusable device must submit data to FDA,
through a 510(k) or PMA, that convinces the agency that a device can be
safely reprocessed for a set number of times without compromising its
function. Currently, while third-party firms must register with FDA and meet
FDA's standards for good manufacturing practices, they can reprocess SUDs
without seeking premarket approval from FDA. FDA has not regulated hospitals
and other health care institutions that reprocess SUDs in-house.

Another difficulty with the current policy has been FDA's inability to
inspect all third-party reprocessors because it has been unable to identify
them. In early June 2000, FDA officials told us that FDA had identified 14
reprocessing facilities operated by 13 different reprocessing firms and that
the agency had inspected all but two of those facilities. FDA discovered two
of the third-party reprocessors only when the firms identified themselves to
the agency by submitting comments on FDA's proposed regulatory framework. In
the course of our work, we found one third-party reprocessing firm that was
not known to FDA, and we forwarded information about it to the agency. FDA
suspects that there are more third-party reprocessors that have not
registered with the agency.

Requirements for Reprocessors

FDA's proposed regulatory framework will make major changes to the oversight
of SUD reprocessing. The framework will extend enforcement of all of FDA's
requirements for device manufacturers to hospitals that reprocess SUDs and
third-party reprocessing firms.23 There will be three major changes. First,
hospitals will be expected to satisfy all the requirements now faced by
third-party reprocessing firms, such as registering with FDA, telling FDA
which devices they reprocess, fulfilling the MDR reporting requirements for
manufacturers (in addition to current user facilities' MDR requirements),24
using reprocessing procedures that meet the standards for good manufacturing
practices, and facing inspection by FDA. Second, both hospitals that
reprocess SUDs and third-party reprocessing firms will be required to meet
all applicable premarket requirements. That is, they will have to conduct
the appropriate studies and submit relevant documentation to FDA as if they
were seeking to market a new device. Thus, for many devices, SUD
reprocessors will be required to submit a 510(k) demonstrating that the
reprocessed device is substantially equivalent to a device already on the
market, presumably the same device before it has been reprocessed. For other
devices with higher risk, SUD reprocessors will be required to conduct
clinical trials and gather other information to submit a PMA to show that
the device is safe and effective. For example, among reprocessed devices, GI
biopsy forceps will require a 510(k), and cardiac ablation catheters (a type
of EP catheter) will require a PMA. Finally, all reprocessors will be
required to follow general requirements for labeling SUDs, including
providing adequate instructions for use. Neither the hospitals nor the
third-party reprocessors we contacted now include instructions for use on
their labels because reprocessed devices ordinarily are returned to
facilities that already have instructions from the manufacturer's original
labeling of the device.25

FDA's proposed regulatory framework for SUD reprocessing specifically
exempts opened but unused SUDs. The proposed framework also does not apply
to health care facilities other than hospitals that reprocess SUDs in-house.
By at least temporarily excluding ambulatory surgery centers, physicians'
practices, and other nonhospital health care institutions from regulation,
the proposal maintains the inconsistency of the current policy by exempting
some categories of reprocessors from FDA oversight.

FDA plans to issue a final guidance document in July 2000, with the new
requirements taking effect over an 18-month period starting then. Hospitals
that reprocess SUDs will be subject to FDA's manufacturer facilities
requirements (registration, listing, inspection, and MDR reporting)
6 months after the final guidance is issued. Premarket approval requirements
for frequently reprocessed SUDs will also begin to take effect 6 months
after the final instructions are issued for devices FDA considers high risk,
in 12 months for moderate-risk devices, and in 18 months for low-risk
devices. FDA is phasing in the enforcement of these requirements for several
reasons: (1) it believes that its regulatory activities should be
implemented in accordance with the potential health risk associated with the
reprocessed device; (2) the potential exists for unintended and
unpredictable consequences if FDA enforces all requirements immediately,
such as shortages of reprocessed SUDs in certain hospitals; (3) hospitals
need time to learn about and comply with FDA regulations; and (4) FDA lacks
the resources to immediately enforce all regulatory requirements.


FDA's proposed regulatory framework imposes a structure designed to oversee
the manufacture of new medical devices onto the different enterprise of SUD
reprocessing. Implementation of this new framework will face a number of
barriers, including SUD reprocessors' inexperience with FDA's regulations
for medical device manufacturers.

Hospitals that reprocess SUDs have no experience with FDA's regulation of
medical devices and device manufacturers, even though FDA technically
considers them to be device manufacturers now. And, while third-party
reprocessing firms already collect some of the data FDA will require for
premarket approval of reprocessed SUDs, their ability to adjust to the new
requirements is not assured. For example, third-party reprocessors may find
it difficult to conduct the clinical studies needed to submit PMAs. FDA and
some third-party reprocessors are working together to develop prototype
premarket applications for reprocessing.

Important details about the operation of the new framework that will affect
its implementation have yet to be finalized by FDA. For instance, the extent
to which FDA will accept premarket applications for groups of similar
devices, rather than for each model of a device, has yet to be determined.
Similarly, some manufacturers have told FDA that some class I devices that
are exempt from premarket approval would pose risks if reprocessed. The
number of SUDs requiring FDA approval to be reprocessed could increase if
FDA agrees with that assessment.

Another implementation hurdle is that FDA will probably not be able to
identify all of the reprocessors that will be subject to the new regulatory
framework, at least in the short term. FDA has not yet located all of the
third-party reprocessing firms that it suspects operate today. In addition,
although it is engaged in an outreach effort to educate hospitals that
reprocess SUDs in-house about the new requirements, we believe FDA will find
it difficult to identify reprocessing hospitals unless they voluntarily
register with the agency.

Furthermore, the potentially large number of additional premarket
applications and manufacturing facilities to inspect could overburden FDA's
already stretched resources. FDA officials told us that the agency has about
500 staff involved in reviewing premarket applications for medical devices
and inspecting device manufacturers. These staff reviewed about 5,000
premarket applications and completed approximately 1,050 inspections of
medical device manufacturing facilities in fiscal year 1999. FDA's fiscal
year 2001 budget request includes an additional 5 staff years for work on
reprocessing issues. The number and complexity of 510(k)s and PMAs that will
be submitted for reprocessing are unknown, as is the number of hospitals
that will register with the agency. But FDA could receive many premarket
applications because applications are required from each entity for each
device that it wishes to reprocess. A large number of applications may
impede FDA's ability to oversee reprocessing and may compromise its work in
other areas. For example, premarket submissions for reprocessing will be
placed in the same queue as 510(k)s and PMAs for new medical devices. An FDA
official told us that this additional work may decrease the percentage of
marketing applications for new devices that are reviewed in a timely manner.

Recognizing its resource constraints, FDA has asked HCFA and JCAHO for
assistance in monitoring SUD reprocessing in hospitals.26 We found that
neither HCFA nor JCAHO plans to make a substantial contribution to this
effort in the near term. HCFA could potentially affect SUD reuse in two
ways--by altering its coverage policies or by changing the terms of
participation for hospitals to participate in Medicare. Regarding coverage
policies, Medicare generally does not cover medical devices that are not
approved by FDA, and HCFA has agreed that it will not pay for SUDs
reprocessed without FDA's approval. In commenting on a draft of this report,
HHS stated that HCFA will review Medicare coverage and payment rules once
FDA's new framework takes effect. However, HCFA currently lacks the means to
determine whether it is paying for a new or a reprocessed device because it
pays for the treatment of particular conditions, not for individual pieces
of equipment that may be used in treatment. For the terms of participation,
a HCFA official told us that it has no plans to include requirements about
the reuse of SUDs in any of its standards for Medicare conditions of
participation for health care facilities.

FDA has consulted with JCAHO with the hope that it could eventually perform
inspections to FDA's standards for SUD reprocessing in hospitals.27 This
would include providing hospital identifying information to FDA so that FDA
could take enforcement actions. As of mid-May 2000, JCAHO had agreed to
inform hospitals about FDA's policy on SUD reuse and to ask three questions
about in-house SUD reprocessing on FDA's behalf during its hospital surveys
for a 6-month period: (1) Does the hospital reprocess and reuse devices
labeled for single-use, and if so, which devices does it reprocess or reuse?
(2) Is the hospital aware of FDA's requirements for registration and listing
of the devices it chooses to reprocess and reuse? (3) Does the hospital
intend to continue to reprocess and reuse such devices? JCAHO will provide
the answers to FDA on an aggregate basis, without identifying individual
hospitals. FDA is paying JCAHO a small fee for this service. In the long
term, JCAHO's suitability for conducting inspections on FDA's behalf has not
been established. For example, JCAHO's survey practices and policies for
protecting confidential hospital information may conflict with FDA's need to
take public enforcement actions.28 In addition, if their collaboration
proceeds, FDA may need to pay JCAHO for these inspections.

FDA's proposed framework imposes significant new requirements on
institutions that reprocess SUDs, but it also grants specific FDA approval
for SUD reprocessing. Because of these conflicting consequences, the net
effect on SUD reprocessing is uncertain. It may lead to an overall decrease
in the number of SUDs that are reprocessed, at least until the new
regulatory system is functioning well. If this happens, there is a chance
that the price of new devices will increase as the competitive alternative
of SUD reprocessing becomes less viable. Also, an FDA official expressed
concern that temporary shortages of reprocessed SUDs may occur in some
hospitals, causing the hospitals to seek devices from other sources.

FDA officials, hospital administrators, physicians, and device manufacturers
all told us that hospitals will be much less likely to maintain in-house SUD
reprocessing operations under the new framework. Some hospitals that
currently reprocess in-house are likely to contract with third-party
reprocessing firms for that work. This shift may ease FDA's task of
inspecting hospitals that reprocess SUDs. It also could increase the costs
of reprocessing to hospitals, because, according to the hospitals we
contacted, third-party firms are more expensive than their internal
reprocessing operations. At least some third-party firms anticipate an
increase in business, both because of the expected shift in reprocessing
from hospitals and because they expect that FDA's formal approval of the
reprocessing of particular SUDs will improve their marketing success.

The evidence suggests that some SUDs can be safely reprocessed if
appropriate cleaning, testing, and sterilization procedures are carefully
followed. However, SUD reprocessing is not invariably safe, and relatively
little is known about the practice of SUD reprocessing in health care
institutions. For this reason, FDA has taken steps to increase its oversight
of SUD reprocessing. Nonetheless, the new framework does not treat all types
of reprocessors consistently and will be difficult to implement.
Furthermore, because the demonstrated health risks from SUD reprocessing are
small, it may have only a limited impact on public health.

In its comments on a draft of this report, HHS said that the report
accurately describes current reuse practices, the potential health risks of
SUD reprocessing, and current and planned regulatory approaches to SUD
reprocessing. HHS also emphasized that FDA is the lead agency for the
regulation of medical devices and that HCFA will review its policies when
FDA's new regulatory framework takes effect. HHS also provided technical
comments, which we incorporated where appropriate. (HHS' comments are in
app. II.)

As arranged with your offices, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days after
its issue date. We will then send copies to the Honorable Donna E. Shalala,
Secretary of Health and Human Services, and the Honorable Jane E. Henney,
Commissioner of FDA. We will also make copies available to others who
request them.

The major contributors to this report were Elizabeth A. Bradley, Marcia
Crosse, Martin T. Gahart, Janina R. Johnson, and Stefanie Weldon. If you or
your staffs have any questions, please contact me at (202) 512-7119.

Janet Heinrich
Associate Director, Health Financing and
Public Health Issues

FDA's List of Frequently Reprocessed SUDs

Angiography catheter
Blood pressure cuff
Cardiac ablation catheter
Cardiac guidewire
Compressible limb sleeve
Electrophysiology recording catheter
Intra aortic balloon catheter
Percutaneous transluminal coronary angioplasty (PTCA) catheter
Percutaneous transluminal angioplasty (PTA) catheter

Breathing mouthpiece
Endotracheal tubes
Oral and nasal catheters
Respiratory therapy and anesthesia breathing circuits
Tracheobronchial suction catheter

Biliary sphincterotomes
Biopsy needles
Endoscopic guidewires
Endoscopic staplers
Extraction balloons/baskets
Non-electric biopsy forceps
Urethral catheters

Hemodialysis blood tubing

Laparoscopic dissectors
Laparoscopic graspers
Laparoscopic scissors

Arthroscopy instruments
Carpal tunnel blade
Drill bits
External fixation device
Flexible reamers/drills
Saw blades
Surgical drills

Biopsy forceps
Biopsy needles
Electrosurgical electrodes/handles/pencils
Endoscopic blades
Endoscopic guidewires
Endoscopic staplers
Fascia holders
Laser fiber delivery systems
Scissor tips, removable inserts
Surgical cutting accessories

Glucometer lancets
Keratome blade
OR drapes
Phacoemulsification needle
OR gowns
Sharps containers
Syringes, piston
Infusion pump, implanted
Syringe, irrigating

Source: FDA, Reprocessing and Reuse of Single-Use Devices: Review
Prioritization Scheme (Feb. 8, 2000).

Comments From the Department of Health and Human Services

The following is GAO's comment on the Department of Health and Human
Services' letter dated June 12, 2000.

1. We have added language to the report to indicate that HCFA will review
Medicare coverage and payment policies once FDA's new regulatory framework
takes effect.


Table 1: Surveys of SUD Reprocessing 9


1. SUDs are also referred to as disposable devices because they are intended
to be discarded after one use.

2. Manufacturers submitting 510(k)s for class III devices must have been
inspected by FDA in the 2 years preceding the 510(k) application. Facility
inspection is not required for other 510(k) applications.

3. Food and Drug Administration Modernization Act of 1997 (P.L. 105-115),
205(b), adding 513(i)(1)(E)(i) to the Federal Food, Drug, and Cosmetic Act.

4. FDA did not require manufacturers of reusable devices to include cleaning
and sterilization instructions with them until 1995, following the release
of reprocessing standards for reusable devices by an industry association.
FDA also has the authority to require a manufacturer to change the label of
a device that it markets for an intended use other than that on the label
and that poses a health risk.

5. The same VA official told us that there had been no reported problems
with reused catheters.

6. This is because some of the surveys did not include nonhospital

7. In addition, we repeatedly heard two claims that we were unable to
verify. Health care personnel told us that they believed that some SUDs were
identical to reusable devices. Similarly, FDA officials and health care
personnel told us that they recalled that the labels of some devices were
changed from reusable to single-use in years past without significant design

8. United States Surgical Corp. v. Orris, Inc. , 5 F. Supp.2d 1201 at 1207.

9. S. O'Donoghue and E.V. Plata, "Reuse of Pacing Catheters: A Survey of
Safety and Efficacy," Pacing and Clinical Electrophysiology , Vol. 11, No. 9
(1988), pp. 1279-80.

10. B. Avitall and others, "Repeated Use of Ablation Catheters: A
Prospective Study," Journal of the American College of Cardiology , Vol. 22,
No. 5 (1993), pp. 1367-72.

11. J. Cohen and others, "A Prospective Study of the Repeated Use of
Sterilized Papillotomes and Retrieval Baskets for ERCP: Quality and Cost
Analysis," Gastrointestinal Endoscopy , Vol. 45, No. 2 (1997), pp. 122-27;
and R.A. Kozarek and others, "Reuse of Disposable Sphincterotomes for
Diagnostic and Therapeutic ERCP: A One-Year Prospective Study,"
Gastrointestinal Endoscopy , Vol. 49, No. 1 (1999), pp. 39-42.

12. However, CDC is aware of infections caused by hemodialyzers that were
reprocessed for reuse on the same patient.

13. The organizations included the American College of Cardiology, the North
American Society for Pacing and Electrophysiology, the American College of
Surgeons, the American Society for Gastrointestinal Endoscopy, the
Association of periOperative Registered Nurses, the Society of
Gastroenterology Nurses and Associates, the American Society for Healthcare
Central Service Professionals, the International Association of Healthcare
Central Service Material Management, the Association for Professionals in
Infection Control and Epidemiology, and the American Hospital Association.

14. This is because the information on MDR reports that identifies SUDs as
reused is inconsistent and probably incomplete. For example, an FDA official
told us that FDA had received only six MDR reports that mentioned a
third-party reprocessing firm by name and that three of them were for the
same incident.

15. The remaining reports were for devices other than those on FDA's list of
frequently reprocessed devices or for devices that were reused on the same

16. Statement of Robert H. O'Holla, before the Oversight and Investigations
Subcommittee, House Committee on Commerce (Feb. 10, 2000).

17. S.C. Bahng and others, "Parental Report of Pediatric Tracheostomy Care,"
Archives of Physical Medicine and Rehabilitation , Vol. 79, No. 11 (1998),
pp. 1367-69.

18. R.J. Cheung and others, "GI Endoscopic Reprocessing Practices in the
United States."

19. CDC, "Nosocomial Infection and Pseudoinfection from Contaminated
Endoscopes and Bronchoscopes--Wisconsin and Missouri," Morbidity and
Mortality Weekly Report , Vol. 40, No. 39 (1991), pp. 675-78; and CDC,
"Bronchoscopy-Related Infections and Pseudoinfections--New York, 1996 and
1998," Morbidity and Mortality Weekly Report , Vol. 48, No. 26 (1999), pp.

20. See Adverse Events: Surveillance Systems for Adverse Events and Medical
Errors (GAO/T-HEHS-00-61 , Feb. 9, 2000).

21. See Medical Device Reporting: Improvements Needed in FDA's System for
Monitoring Problems With Approved Devices (GAO/HEHS-97-21 , Jan. 29, 1997).

22. 21 C.F.R. 820.3 (o). All reprocessors of SUDs are considered

23. FDA, Enforcement Priorities for Single-Use Devices Reprocessed by Third
Parties and Hospitals (Rockville, Md.: FDA, Feb. 2000).

24. MDR reporting requirements for manufacturers include reporting deaths,
serious injuries, and malfunctions to FDA within 30 days; reporting events
that require immediate remedial action to FDA within 5 days; and filing
baseline reports to communicate basic data about each device that is the
subject of a report.

25. To the extent that these instructions infringe on the copyrighted
instructions of the original manufacturers, it may be very difficult for
reprocessors to meet this requirement.

26. HCFA administers Medicare and its related facility survey and
certification programs. JCAHO is a private organization that inspects and
accredits hospitals for participation in Medicare. JCAHO surveys hospitals
every 3 years.

27. In lieu of inspections by FDA, manufacturers can now pay to be inspected
by independent organizations that are acceptable to FDA.

28. HHS, Office of Inspector General, The External Review of Hospital
Quality: Holding the Reviewers Accountable , OEI-01-97-00053 (Washington,
D.C.: HHS, July 1999).
*** End of document. ***