Occupational Safety: Selected Cost and Benefit Implications of
Needlestick Prevention Devices for Hospitals (Correspondence, 11/17/2000,
GAO/GAO-01-60R).

Because of the serious concern for health care workers in the United
States, GAO examined the benefit and cost implications of purchasing
needlestick prevention devices for hospitals. GAO estimates about 69,000
needlesticks in hospitals can be prevented in 1 year through the use of
needles with safety features. Eliminating these needlesticks could
reduce the number of health care workers who become infected with the
hepatitis B virus (HBV), hepatitis C virus (HCV), or human
immunodeficiency virus (HIV) after sustaining a needlestick injury.
GAO's analysis of Centers for Disease Control and Prevention data shows
that reducing needlesticks may prevent at least 25 cases of HBV and at
least 16 cases of HCV infection per year. The reduction in the number of
HIV infections cannot be estimated. GAO estimates that the cost to
purchase needles with safety features would be between $70 million and
$352 million per year. The exact cost to adopt these needles is
difficult to determine because several factors must be considered,
including the cost to train workers to use the devices and the extent to
which the needles reduce injuries.

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

 REPORTNUM:  GAO-01-60R
     TITLE:  Occupational Safety: Selected Cost and Benefit
	     Implications of Needlestick Prevention Devices for
	     Hospitals
      DATE:  11/17/2000
   SUBJECT:  Health surveys
	     Infectious diseases
	     Occupational health standards
	     Hospitals
	     Health hazards
	     Health care personnel
	     Medical supplies
	     Occupational safety
	     Accident prevention
	     Proposed legislation

******************************************************************
** 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.                                         **
**                                                              **
******************************************************************

GAO-01-60R

Needlestick Prevention United States General Accounting Office

Washington, DC 20548

November 17, 2000 The Honorable Pete Stark House of Representatives

Subject: Occupational Safety: Selected Cost and Benefit Implications of
Needlestick Prevention Devices for Hospitals

Dear Mr. Stark: This letter responds to your request for an examination of
the potential benefits and costs of changes that would be mandated under the
proposed Health Care Worker Needlestick Prevention Act (HR 1899), which
would require the use of needles with safety features. 1 Percutaneous
injuries caused by needlesticks (puncturing of the skin by a needle or
similar sharp object) are a serious concern for the approximately 10 million
health care workers in the United States. These injuries pose a significant
risk of occupational transmission of bloodborne pathogens such as human
immunodeficiency virus (HIV), hepatitis B virus (HBV), and hepatitis C virus
(HCV) to health care workers. In addition, the emotional distress of a
needlestick injury can be severe and long lasting, even when a serious
disease is not transmitted. According to the Centers for Disease Control and
Prevention (CDC), approximately 384,000 percutaneous injuries occur annually
in U. S. hospitals, with about 236,000 of these resulting from needlesticks
involving hollow- bore needles. Although the proposed legislation applies to
health care workers in all settings, we focus only on hospital settings, as
there are no reliable data on percutaneous injuries sustained in other
settings. 2 It should be noted, however, that more than one- half of all
health care workers are in nonhospital settings.

This letter presents the number of needlestick injuries potentially
prevented by the use of needles with safety features and the estimated
ranges of the benefits and costs of using such needles in hospitals. Our
analysis is based on data provided by CDC, the International Healthcare
Worker Safety Center, the

1 Subsequent to the introduction of HR 1899, a similar bill, HR 5178, was
introduced and enacted into law. On Nov. 6, 2000, the President signed the
Needlestick Safety and Prevention Act (P. L. 106- 430), which mandates
changes in the bloodborne pathogens standard in effect under the
Occupational Safety and Health Act of 1970. The Act requires employers to
document the consideration and implementation of safer medical devices,
including safe needle devices, in their facilities.

2 CDC's National Institute of Occupational Safety and Health (NIOSH) is
beginning a study to determine the incidence of nonhospital percutaneous
injuries.

GAO- 01- 60R Needlestick Prevention 2 American Hospital Association, the
states of California and Maryland, the BectonDickinson

Corporation, articles published in peer- reviewed medical journals, and
other sources dated between 1995 and 1999. A detailed discussion of our
methods, results, and the potential limitations of our analysis is provided
in enclosure I. We conducted our analysis in August and September 2000 in
accordance with generally accepted government auditing standards.

In summary, we estimate that about 69,000 needlesticks in hospitals can be
prevented in 1 year through the use of needles with safety features.
However, the use of needles with safety features alone is insufficient to
prevent the majority of needlestick injuries. Our analysis indicates that
the use of needles with safety features may have financial benefits that
exceed the cost of these features because they can reduce needlesticks and
associated treatment costs for hospitals. The extent to which needles with
safety features are cost effective depends on their incremental costs, the
extent to which they reduce the risk of a needlestick injury, and the costs
of postexposure treatment of health care workers. Even though these factors
and their potential costs and benefits cannot be measured precisely,
eliminating 69,000 needlesticks could reduce the number of health care
workers who become infected with HBV, HCV, or HIV after sustaining a
needlestick injury. Our analysis of CDC data shows that reducing
needlesticks by this amount may prevent at least 25 cases of HBV infection
and at least 16 cases of HCV infection per year. The reduction in the number
of HIV infections cannot be validly estimated. In commenting on a draft of
this letter, CDC stated that it provides an objective presentation of the
information.

BACKGROUND Risk of exposure to bloodborne pathogens has always been a
problem for health care workers, but the emergence of acquired
immunodeficiency syndrome (AIDS) in 1981 brought the issue to the forefront
of public health policy. Percutaneous injuries, which include needlestick
injuries, expose health care workers to deadly bloodborne pathogens,
including HIV. At least 17 states have enacted safe needle laws. 3 These
laws usually either instruct health departments to require hospitals to use
needles with safety features or commission studies to evaluate the
feasibility of such features. It has been within the past 2 years that
states began enacting safe needle laws, beginning with California in
September 1998.

The Nature and Magnitude of the Needlestick Injury Issue The total number of
needlestick injuries sustained annually in the United States is unknown, and
the lack of data from nonhospital settings appears to be the

3 The following states have enacted safe needle laws: Alaska, California,
Connecticut, Georgia, Iowa, Maine, Maryland, Massachusetts, Minnesota, New
Hampshire, New Jersey, New York, Ohio, Oklahoma, Tennessee, Texas, and West
Virginia. In addition, Hawaii adopted a resolution supporting the
implementation of safer medical devices.

GAO- 01- 60R Needlestick Prevention 3 greatest obstacle in deriving a
national injury estimate. 4 Our analysis focuses on

health care workers in hospital settings, who account for approximately 40
percent of health care workers. According to CDC survey data, approximately
384,000 percutaneous injuries occur annually in hospitals, 5 with about 61
percent (236,000) resulting from hollow- bore needlestick injuries. CDC
adjusted their estimate of the number of percutaneous injuries for
underreporting 6 and other factors. However, these estimates exclude the
unknown number of needlestick injuries to health care workers in nonhospital
settings, where about 60 percent of health care workers are employed.

According to CDC survey data, most needlestick injuries occur after the
device has been usedï¿½and therefore exposed to potentially contaminated
bloodï¿½but before its disposal. Fifty percent of injuries occur between the
time the procedure is completed and disposal of the device; 20 percent are
associated with disposal of the device. Other injuries occur when the needle
pierces the syringe cap as the syringe is recapped after use, when a body
fluid is transferred from a syringe to a specimen container, and when used
needles are not disposed of in puncture- resistant containers. Devices
requiring disassembly (for example, prefilled cartridge injection syringes
that need to be detached from needles) are associated with higher rates of
injury.

However, many injuries to health care workers occur during use of a needle
in a patient, when the needle is inserted, manipulated, or withdrawn. Sudden
patient movement can also jar the needle loose and cause injury to a worker.
Among hospitals participating in CDC's survey, 26 percent of needlesticks
occur at this time. For the most part, these are not preventable with
current safer technology.

Risks Faced by Health Care Workers Needlestick injuries are a significant
risk for health care workers because these injuries expose workers to
diseases caused by bloodborne pathogens. The

4 Nonhospital facilities include nursing homes, physician and dental
offices, medical and dental laboratories, and residential care, hospice
care, home health care, and outpatient facilities. Other employees at risk
include personnel from funeral homes, schools, correctional facilities, and
waste removal, law enforcement, and fire and rescue services.

5 CDC based this estimate on data from 15 hospitals participating in its
National Surveillance System for Health Care Workers (NaSH) and on data from
45 hospitals participating in the Exposure Prevention Information Network
(EPINet) of the International Health Care Worker Safety Center. The 95
percent confidence interval for this estimate is 311, 000 to 464,000
percutaneous injuries per year in U. S. hospitals.

6 Rates of underreporting are difficult to ascertain. Hindrances to
reporting injuries include the perception that a low risk of infection is
associated with certain types of injuries, or patients, or both; lack of
knowledge of the appropriate procedures to follow after injury has incurred;
fear of punitive employer response; and time constraints.

GAO- 01- 60R Needlestick Prevention 4 primary diseases of concern in current
occupational settings are AIDS (from HIV),

hepatitis B (from HBV), and hepatitis C (from HCV). 7 HIV attacks part of
the body's immune system, and most health care workers who become infected
with HIV are likely to develop AIDS, which is characterized by severe
infections, other complications, and death. As of December 1999, CDC had
received reports of 56 documented and 136 possible cases of occupationally
acquired HIV infection in the United States. 8 Twenty- five of the 56
documented cases of HIV infection have progressed to AIDS. The average
transmission rate of HIV infection following a needlestick injury from an
infected patient is 0.3 percent. No vaccine currently exists to prevent HIV
infection, and there is no cure.

About one- third to one- half of those with acute HBV infection develop
symptoms of hepatitis such as jaundice, fever, nausea, and abdominal pain.
Most acute infections resolve, but 5 to 10 percent of patients develop
chronic infection and become carriers of the disease. Chronic carriers of
the infection have an estimated 20- percent lifetime risk of dying from
cirrhosis and a 6- percent risk of dying from liver cancer. The average
transmission rate of HBV infection following a needlestick injury from an
infected patient is estimated to range from 6 to 30 percent. Hepatitis B
vaccines have been available since 1982. Currently, the genetically
engineered hepatitis B vaccine is recommended for all health care personnel
who are occupationally exposed to blood and has a 96- percent vaccine
efficacy rate. According to CDC, about 71 percent of workers who are at risk
for occupational exposure to blood had been vaccinated by 1995. The
Department of Health and Human Services (HHS) has established a goal of
increasing hepatitis B vaccine coverage within this group to 98 percent by
2010.

HCV infection often occurs initially with only mild symptoms or none at all.
However, approximately 75 to 85 percent of persons with HCV infection
subsequently develop chronic infection, and 70 percent develop active liver
disease. Of the patients with active liver disease, 10 to 20 percent develop
cirrhosis, and 1 to 5 percent develop liver cancer. The average transmission
rate of HCV following a needlestick injury from an infected person is 1.8
percent. Currently, no vaccine is available to prevent HCV infection.

Postexposure treatment is recommended for health care workers following a
needlestick exposure from an HIV- infected patient or an HBV- infected
patient. Many drugs used for HIV postexposure treatment are expensive and
have unpleasant side effects. There are no medications for postexposure
treatment for health care workers following a needlestick exposure to an
HCV- infected patient.

7 Other diseases from bloodborne pathogens include HTLV- I- associated
myelopathy (from the human T- lymphotrophic virus Type I [HTLV- I]),
syphilis, malaria, dengue, babesiosis, brucellosis, leptospirosis,
Creutzfeldt- Jakob disease, arboviral infections, relapsing fever, viral
hemorrhagic fever, and Colorado tick fever.

8 Documented cases include those health care workers who were HIV negative
before the injury and were HIV positive after the injury. Possible cases
include health care workers who acquired the infection without a documented
occupational exposure and without identifiable behavioral or transfusion
risks.

GAO- 01- 60R Needlestick Prevention 5 Even when a serious disease is not
transmitted, the emotional distress of a

needlestick injury can be severe and long lasting, often requiring
counseling. This is especially true if the injury involves exposure to HIV.
Not knowing the infection status of the source patient can also create
distress. Emotional distress from needlestick injuries may also extend to
colleagues and family members.

Safer Medical Devices Technology Hospitals and other facilities can use many
types of safer medical devices to reduce the number of percutaneous
injuries. Examples of needles with safety features include protected needle
intraveneous (IV) connectors; needles that retract into a syringe or vacuum
tube holder; hinged or sliding shields attached to phlebotomy needles
(needles for drawing blood), winged- steel needles, and blood gas needles;
protective encasements to receive an IV stylet as it is withdrawn from the
catheter; sliding needle shields attached to disposable syringes and vacuum
tube holders; self- blunting phlebotomy and winged- steel needles; and safer
IV catheters that encase the needle after use.

According to OSHA, facilities using needles with safety features are
reducing the number of needlestick and other types of percutaneous injuries.
9 Training and education are necessary for health care workers to learn how
and when to use these safer medical devices properly. Other changes in work
practices such as not allowing disposal containers to overfill can also
reduce the risk of needlestick injury.

These devices have limitations. Many new devices have been developed to
reduce the risk of needlestick injuries, but those that have been assessed
vary considerably in their clinical efficacy and in their effectiveness in
reducing rates of injuries. Also, needles with safety features may not be
available or may not be a practical alternative to conventional devices in
certain situations. 10 In some cases, these devices have caused needlesticks
while in use. Besides these limitations, there are obstacles to the use of
needles with safety features, which include their increased purchase price
compared with conventional devices, possible staff resistance to changes in
the devices used, and the time required to train staff in the use of new
devices.

9 On Sept. 9, 1998, OSHA published a Request for Information in the Federal
Register. This request asked for information on engineering and controls
used to eliminate or minimize the risk of occupational exposure to
bloodborne pathogens due to percutaneous injuries from contaminated needles
and other sharp instruments. OSHA received 396 responses from nursing homes,
clinics, and acute care, tertiary care, rehabilitation, and pediatric
hospitals. See Occupational Safety and Health Administration, "Record
Summary of the Request for Information on Occupational Exposure to
Bloodborne Pathogens Due to Percutaneous Injury” (Washington, D. C.:
Occupational Safety and Health Administration, May 1999) http:// www. osha-
slc. gov/ html/ ndlreport052099. html (downloaded Sept. 5, 2000).

10 Respondents to OSHA's Request for Information indicated that this is true
in dentistry and pediatric applications. Currently, safer needles are only
available in limited sizes.

GAO- 01- 60R Needlestick Prevention 6 In addition to these obstacles, some
needles with safety features have been

reported to affect patients adversely. Adverse effects reported in response
to OSHA's Request for Information include additional venipunctures with some
blood draw devices and safety catheters, increased pain, and hematomas.
However, the majority of respondents indicated that delivery of patient care
has not been affected by the use of needles with safety features.

CDC recommends that use of needles with safety features be combined with
comprehensive programs that include reducing the unnecessary use of needles,
modifying procedures and work practices, training health care workers in
safer work practices involving the use of needles, promoting safety
awareness in the work environment, and evaluating the effectiveness of these
measures.

BENEFITS OF NEEDLES WITH SAFETY FEATURES EXCEED THEIR COSTS IN SOME
CIRCUMSTANCES

Adoption of needles with safety features would prevent about 69,000
needlesticks each year. Many HBV, HCV, and HIV infections would be prevented
as well. Needles with safety features are currently more expensive than
conventional needles. However, our analysis of available data on the costs
and preventability of needlestick injuries shows that the adoption of
needles with safety features may be justifiable based solely on decreased
initial treatment costs. The greatest dollar savings resulting from a
needlestick reduction program would be the reduced cost of treating health
care workers who have sustained needlesticks. Other costs also would be
reduced, but these cost reductions are difficult to quantify as they are
highly dependent on specific situations. These costs include medical
treatment costs for health care workers who become infected after sustaining
a needlestick; wages and time lost by these workers; emotional distress
suffered by injured workers, their colleagues, and family members; reduced
quality of life; and while rare, lives lost. Needles with safety features
may also reduce liability and worker's compensation costs to hospitals when
health care workers acquire diseases after a needlestick injury. These exact
cost reductions cannot be determined from the available data, and we have
not included them in our analysis.

Reduction of Needlestick Injuries in Hospital Settings Using needles with
safety features could prevent a sizable number of needlestick injuries in
hospitals. According to our analysis, about 69,000 of these injuries are
preventable by the use of needles with safety features (see table 1).
Additionally, 109,000 needlesticks are preventable by eliminating the
unnecessary use of needles 11 and by using safer work practices. The 69,000
needlestick injuries represent about two- fifths of the estimated 177,000
preventable needlesticks reported by hospitals participating in the National
Surveillance System for Health

11 Unnecessary use of needles is partially dependent on available
technology. In particular, improved technologies that eliminate the use of
needles as connectors in IV lines have proven useful.

GAO- 01- 60R Needlestick Prevention 7 Care Workers (NaSH), 12 which is
managed by CDC's Hospital Infections Program.

Hospitals use this system to report percutaneous injuries, health care
worker exposure to blood and other body fluids, and other information
related to preventing occupational exposures and infections among health
care workers. Participation in CDC's NaSH system is voluntary, and the
number of preventable needlesticks and means of preventing them in these
hospitals may not be representative of all hospitals in the United States.
The approximately 60 hospitals that currently volunteer to participate in
NaSH tend to be large and are concentrated in the northeastern United
States.

Table 1: Projection of the Percentage and Number of Preventable Needlesticks
in Hospitals in 1 Year

Percentage a Number b Projected number of annual needlesticks 236,000 Not
currently preventable needlesticks c 25 59,000 Preventable needlesticks 75
177,000

Preventable by eliminating unnecessary use 25 58,000 Preventable by using
needles with safety features 29 69,000 Preventable by using safer work
practices 21 51,000 a Percentage is based on the number of annual
needlesticks.

b Totals may not add due to rounding. c Needlesticks that are not currently
preventable often occur while the needle is in use in the patient.

Source: GAO projection of CDC NaSH data. For our analysis, we assumed that
the percentage of preventable needlesticks and the means of their prevention
as shown in the NaSH data were reasonable models for all hospitals in the
United States, regardless of their size or location. The NaSH estimates
appear to be consistent with other published reports that show actual
reductions in the percentage of needlestick injuries sustained in hospitals
after needles with safety features were adopted. 13 These estimates are
general in nature.

12 The percentage of preventable needlesticks varied in the 31 hospitals
participating in NaSH (the preventability within specific hospitals ranged
from 48 to 85 percent). The percentage of needlesticks preventable by method
also varied. See S. Campbell, L. Chiarello, P. Srivastava, D. Cardo, and The
NaSH Surveillance Group, “Preventability of Needlestick Injuries to
Health Care Workers in the National Surveillance System for Healthcare
Workers,” Abstracts-- 4th Decennial International Conference on
Nosocomial & Healthcare- Associated Infections( Atlanta, Ga.: Centers for
Disease Control and Prevention, July 2000), http:// www. cdc. gov/ ncidod/
hip/ NASH/ 4thabstracts. htm - 7 (downloaded Sept. 5, 2000).

13 See Centers for Disease Control and Prevention, “Evaluation of
Safety Devices for Preventing Percutaneous Injuries Among Health- Care
Workers During Phlebotomy Procedures- MinneapolisSt. Paul, New York City,
and San Francisco, 1993- 1995,” Morbidity and Mortality Weekly Report,
Vol. 46, No. 2 (1997), pp. 21- 25; and F. Roudot- Thorval, O. Montagne, A.
Schaeffer, et al., "Cost and Benefits of Measures to Prevent Needlestick
Injuries in a University Hospital," Infection Control and Hospital
Epidemiology, Vol. 20, No. 9 (1999), pp. 614- 17.

GAO- 01- 60R Needlestick Prevention 8 The magnitude and methods for
preventing needlesticks may not precisely match

these estimates in every hospital. Reducing the number of needlestick
injuries may also reduce the number of health care workers who become
infected with HBV and HCV. The specific number of infections avoided is
difficult to determine, as the risk depends on the type of virus and the
nature of the exposure. The number of HIV infections that would be avoided
cannot be validly estimated. Projections for the approximate number of HBV
and HCV infections avoided, based on NaSH data, are shown in table 2.
According to CDC, the prevalence of these viruses may be higher for patients
in hospitals than in the general population. This would increase the risk of
infection for health care workers, as the percentage of infected persons
they may be exposed to may be greater than the percentage of infected
persons in the general population. Therefore, these projected reductions in
infections may be underestimates.

Table 2: Projection of the Number of HBV and HCV Infections Avoided From
Needlesticks in Hospitals in 1 Year

Method of prevention HBV Infections

HCV Infections Infections avoided by eliminating unnecessary use of needles
21 14 Infections avoided by using needles with safety features

25 16 Infections avoided by using safer work practices 19 12

Source: GAO projection of CDC NaSH data. While our analysis is focused on
reducing needlesticks through the use of safety features, we also found that
using safer work practices could prevent about 51,000 needlesticks. Safer
work practices include such measures as not recapping needles unless no
alternative exists; properly disposing of used needles in puncture-
resistant sharps containers; and consolidating specimen collection from
patients. We did not find any valid estimates of the costs of using safer
work practices, so we have not estimated the potential costs and benefits of
adopting them. However, as with the adoption of needles with safety
features, the benefits of adopting safer work practices are likely to be
significant due to the savings resulting from decreased postexposure
treatment costs.

Costs of Needles with Safety Features and Number of Needles Used The
increased purchase costs of using needles with safety features in hospitals
would be between $70 million and $352 million per year. These do not include
the costs associated with training or changing work practices; however,
eliminating the unnecessary use of needles would also produce savings. The
exact cost of adopting needles with safety features is difficult to
determine. Needles with safety

GAO- 01- 60R Needlestick Prevention 9 features generally cost more than
those without, but the cost varies with the type

of feature, the number of times the feature is used, the cost of training
workers in its correct use, and other factors. For example, data reported to
OSHA as part of its Request for Information indicated that the added cost of
a needle with a safety feature ranges from $. 07 to $. 15 for a syringe/
needle combination, from $. 15 to $. 30 for a blood collection needle or
set, and to about $. 70 for an intravenous catheter. Other OSHA respondents
reported that a hypodermic syringe/ needle without a safety feature would
cost $. 05, whereas a similar syringe/ needle with safety features would
cost about $. 25. For the purposes of our analysis, we estimated costs at
three possible levels, assuming that the cost of a needle with a safety
feature would be 1.5, 2.0, or 3.5 times the cost of a similar needle without
a safety feature. These cost estimates fall within the general range of
other published cost estimates.

The number of hollow- bore needles used in hospitals is difficult to
determine. Our estimate is based on data for a hospital with 250 to 300
beds. The devices included in this projection are the ones most commonly
used to penetrate tissues or to enter arteries or veins. Table 3 shows our
estimate of the number of needles by type per hospital bed.

Table 3: Estimate of the Number of Needles Used in Hospitals Per Year Needle
type Number used per

hospital bed Number used per year

Vacuum tube blood collection 217 217,000,000 Winged- steel needle 56
56,000,000 IV catheter 111 111,000,000 Hypodermic syringe/ needle 367
367,000,000

Source: GAO estimate based on data from OSHA, the International Healthcare
Worker Safety Center, and other sources.

Costs for Treatment of Health Care Workers Injured by Needlesticks Costs of
initial postexposure treatment vary widely and depend on the situations
faced by injured workers. Published estimates run from $500 to $3,000 per
injury sustained. Depending upon the situation, an injured worker may need
treatment for exposure to HIV, HBV, or other bloodborne pathogens. In
addition, the patient involved may need to be tested for diseases. For the
purposes of our analysis, we assumed postexposure treatment costs of $500,
$1,500, and $2,500. These cost estimates fall within the general range of
other published cost estimates. 14 We estimate that eliminating 69,000
needlesticks per year would reduce postexposure treatment costs for injured
health care workers in hospitals by between $37 million and $173 million per
year.

14 Occupational Safety and Health Administration, “Record Summary of
the Request for Information on Occupational Exposure to Bloodborne Pathogens
Due To Percutaneous Injury” (Washington, D. C.: Occupational Safety
and Health Administration, May 1999), http:// www. oshaslc. gov/ html/
ndlreport052099. html (downloaded Sept. 5, 2000).

GAO- 01- 60R Needlestick Prevention 10 While only a subset of health care
workers who suffer needlestick injuries

subsequently become infected, adoption of needles with safety features also
may reduce costs associated with longer term treatment for those workers.
However, we did not estimate these reductions as they are highly dependent
on the worker's situation. These situational factors include the worker's
age and health status at the time of infection, the type of infection
acquired, and the severity of diseases resulting from the infection. While
we did not estimate these costs, they are potentially significant. For
example, the average annual cost of treating a person with HIV has been
estimated at between $20,000 and $24,700 in 1996. 15

By reducing the risk of needlestick injuries, the use of needles with safety
features may also reduce the potential liability costs to hospitals when
health care workers become infected after a needlestick injury. Fears of HIV
and other infections have led many health care workers to pursue legal
action for compensation for a disease acquired at work. Even in cases where
diseases have not been transmitted, health care workers are suing for
compensation for the emotional distress experienced while waiting for test
results. We were unable to identify data concerning the dollar amounts
awarded for compensation.

Costs Avoided by Adopting Needles With Safety Features Using the assumptions
above, we estimated the potential costs of adopting a national requirement
to use safe needle technologies. This analysis shows that the use of needles
with safety features is cost efficient when the costs of postexposure
treatment are moderate or high and the added costs per feature are low (see
table 4).

15 F. Hellinger and J. Fleishman, “Estimating the National Cost of
Treating People With HIV Disease: Patient, Payer, and Provider Data, ”
Journal of Acquired Immune Deficiency Syndromes, Vol. 24 (2000), pp. 182-
88.

GAO- 01- 60R Needlestick Prevention 11 Table 4: Estimates of Benefits Over
Costs of Needles With Safety Features in

Hospitals for 1 Year Cost scenarios for postexposure treatment Low ($ 500
per injury) Medium ($ 1,500

per injury) High ($ 2,500 per injury) Low cost (1.5 times more costly)

-$ 47 million $ 21 million $ 90 million Medium cost (2.0 times more costly)

-$ 129 million -$ 60 million $ 9 million Cost for

needles with safety features compared with conventional needles High cost
(3.5

times more costly)

-$ 374 million -$ 306 million -$ 237 million Shaded figures indicate
benefits that exceed costs. Unshaded figures indicate costs that exceed
benefits.

Source: GAO analysis. The scope of this analysis is limited to the selected
financial costs that hospitals might incur that are associated with using
needles with safety features, but it omits the effects of several relevant
factors. For example, we did not factor in (1) decreases in subsequent
medical treatment costs for health care workers who become infected, (2)
reductions in health care workers' risks to life and health, (3) reductions
in time lost from work, and (4) the emotional distress suffered by injured
and infected workers. While it is not easy to quantify the additional
benefits of using needles with safety features, they are real and likely to
be substantial. If we were able to incorporate these additional factors, the
estimated net benefits of needles with safety features would have been
greater than the estimates reported above.

AGENCY COMMENTS We provided a draft of this report to CDC for review and
comment. In written comments, CDC stated that the agency generally agreed
with our results and methodology, recognizing the limitations of the data
(see encl. II). CDC officials also provided technical comments, which we
incorporated where appropriate.

-----

GAO- 01- 60R Needlestick Prevention 12 As arranged with your office, unless
you publicly announce the contents of this

letter earlier, we plan no further distribution of it until 30 days from the
date of this letter. At that time we will send copies to the Honorable
Jeffrey P. Koplan, Director of CDC, and other interested parties. We will
also make copies available on request. If you have any questions regarding
this letter, please contact me at (202) 512- 7119. Marcia Crosse, Timothy
Clouse, David Goodman, and Deborah Miller made major contributions to this
work.

Sincerely yours, Janet Heinrich Director, Health Care - Public Health Issues

Enclosures-2

ENCLOSURE I ENCLOSURE I

GAO- 01- 60R Needlestick Prevention 13 METHODOLOGY

Our analysis is based on data provided by the Centers for Disease Control
and Prevention (CDC), the International Healthcare Worker Safety Center, the
American Hospital Association, the states of California and Maryland, 16 the
Becton- Dickinson Corporation, articles published in peer- reviewed medical
journals, and other sources. Two surveillance systems provide most of the
data relating to needlestick injuries: CDC's National Surveillance System
for Health Care Workers (NaSH) and the International Healthcare Worker
Safety Center's Exposure Prevention Information Network. While data from
these two surveillance systems may not be representative of all hospitals,
the data appear to be similar in terms of the types of devices that cause
injuries and the approximate frequency of percutaneous injuries within
participating hospitals. Given this similarity, we believe that, while data
from these systems are not representative of hospitals generally, they do
show injury patterns that could reasonably be expected to occur in
hospitals.

CDC's Hospital Infections Program manages NaSH, which is a voluntary system
that hospitals use to report percutaneous injuries, health care worker
exposure to blood and other body fluids, and other information related to
the prevention of occupational exposures and infections among health care
workers. Hospitals that have volunteered to participate in NaSH tend to be
large and are concentrated in the northeastern United States. As NaSH
participation is voluntary, data from NaSH may not be representative of
hospitals across the nation. Participation in NaSH has varied over time;
currently about 60 hospitals are participating.

Our estimates for the number of hollow- bore needles used and the cost of
these features are based on information reported by the American Hospital
Association, the states of California and Maryland, and the Becton-
Dickinson Corporation. The cost of needles with safety features depends on
the specific feature, usage patterns, and related factors such as the
training needed to use the device properly. Because of these potential cost
variations, we analyzed the costs and benefits of needles with safety
features using the assumption that the unit cost for hollow- bore needles
with safety features would be 1.5, 2.0, and 3.5 times more than similar
needles without safety features (see table 5).

16 California and Maryland are the only states with published estimates on
the costs and benefits of requiring the use of devices with safety features.

ENCLOSURE I ENCLOSURE I

GAO- 01- 60R Needlestick Prevention 14 Table 5: Cost of Using Needles With
Safety Features in 1 Year

Current With use of safety features

Additional costs Needles with safety features 1.5 times more expensive than
conventional needles

$164,000,000 $245,000,000 $81,000,000 Needles with safety features 2.0 times
more expensive than conventional needles

$164,000,000 $327,000,000 $163,000,000 Needles with safety features 3.5
times more expensive than conventional needles

$164,000,000 $572,000,000 $408,000,000 Source: GAO projections are based on
data from the American Hospital Association, the states of California and
Maryland, and the Becton- Dickinson Corporation.

These cost estimates are consistent with those published elsewhere. For
example, the Becton- Dickinson Corporation estimates that a typical
hypodermic syringe without a safety feature costs about $. 09. A similar
hypodermic syringe with a safety feature costs about $. 30 or about 3.3
times more. For our analysis, we assumed that baseline costs for
conventional devices were $0.10 for a vacuum tube blood collection needle,
$0.65 for a winged- steel needle, $0.65 for an IV catheter, and $0.09 for a
hypodermic needle/ syringe.

We used a range of estimated costs for the treatment required after a
needlestick to reflect the range of costs reported. Cost estimates given to
the Occupational Safety and Health Administration and CDC and those
published in peer- reviewed medical literature range from $500 to more than
$3,000 per injury. These variations are due to the types of treatments
needed, facility procedures for treating and accounting for the cost of
needlesticks, and the extent to which other costs not directly related to
the injury (such as administrative reporting requirements and training
costs) are included. For our analysis, we used a range of costsï¿½$ 500,
$1,500, and $2,500 per injury (see table 6). These cost estimates are within
the range of cost data we reviewed.

ENCLOSURE I ENCLOSURE I

GAO- 01- 60R Needlestick Prevention 15 Table 6: Estimates for Postexposure
Treatment Costs in 1 Year

Current With use of safety features

Avoided treatment costs $500 per injury assumed $118,000,000 $ 84,000,000
$34,000,000 $1,500 per injury assumed

$354,000,000 $ 251,000,000 $103,000,000 $2,500 per injury assumed

$591,000,000 $ 418,000,000 $173,000,000 Source: GAO analysis. Our estimates
do not include lifetime medical treatment costs, lost wages, or workmen's
compensation costs for health care workers who acquire infections after
sustaining needlestick injuries. These costs vary with the specific
circumstances surrounding the injury.

We estimated the infection risk that health care workers face from
needlestick injuries by using CDC reports on the risk of human
immunodeficiency virus (HIV), hepatitis B virus (HBV), and hepatitis C virus
(HCV) infections per needlestick and combined them with published reports on
the prevalence rates of these diseases nationwide. For example, we estimated
the probability that a needlestick injury could result in HCV infection by
multiplying the probability that a needlestick involving a person with HCV
infection would result in infection in the health care worker (about 1.8
percent) by the probability that the needlestick involved a person infected
with HCV (about 1.3 percent). Multiplying these probabilities together
indicates that the risk of HCV infection from a random needlestick involving
a randomly selected person is about 0.02 percent or about 1 in 4,000. The
results of our analysis are shown in table 7.

ENCLOSURE I ENCLOSURE I

GAO- 01- 60R Needlestick Prevention 16 Table 7: Estimated Risk Faced by
Health Care Workers From Needlesticks

Characteristic Percentage Percentage of U. S. population with chronic HBV
infection 0.42 Percentage of U. S. population with chronic HCV infection
1.30 Percentage of U. S. population with HIV infection 0.32

Probability of contracting HBV, from an infected patient, per needlestick

30.00 Probability of contracting HCV, from an infected patient, per
needlestick 1.80 Probability of contracting HIV, from an infected patient,
per needlestick

0.30 Probability of contracting HBV, from a patient, per needlestick a 0.126
Probability of contracting HCV, from a patient, per needlestick 0.024
Probability of contracting HIV, from a patient, per needlestick 0.001 a This
probability is for a person who is not immune to HBV infection. In 1995,
about 71 percent of workers at risk for HBV infection had been immunized and
would not face this specific risk.

Source: GAO projection of CDC data. According to CDC, this table may
underestimate the risk to health care workers because it uses the population
prevalence for the U. S. population as an estimate of the prevalence of
these viruses in persons seeking medical care. Prevalence rates for persons
seeking care will be different, and for some populations, the prevalence
rates will be much higher. 17 Studies show prevalence rates as high as 76.9
percent for HCV and 65.7 percent for HBV for some inner city injecting drug
user populations, 18 and HCV rates as high as 10.4 percent for patients in
dialysis units. 19

17 R. S. Janssen, M. E. St. Louis, G. A. Satten, et al., “HIV
Infection Among Patients in U. S. Acute Care Hospitals,” New England
Journal of Medicine, Vol. 327 (1992), pp. 445- 52; G. D. Kelen, S. Fritz, B.
Qaqish, et al., “Unrecognized Human Immunodeficiency Virus Infection
in Emergency Department Patients,” New England Journal of Medicine,
Vol. 318 (1988), pp. 1645- 50; P. Charache, J. L. Cameron, A. S. Maters, E.
I. Frantz, “Prevalence of Infection With Human Immunodeficiency Virus
in Elective Surgery Patients,” Annals of Surgery, Vol. 214 (1991), pp.
562- 68; M. A. Montecalvao, M. Sung Lee, H. DePalma, et al.,
“Seroprevalence of Human Immunodeficiency Virus- 1, Hepatitis B Virus,
and Hepatitis C Virus in Patients Having Major Surgery,” Infection
Control and Hospital Epidemiology, Vol. 16 (1995), pp. 627- 32.

18 R. S. Garfein, et al.,“ Viral Infections in Short- Term Injection
Drug Users: The Prevalence of the Hepatitis C, Hepatitis B, Human
Immunodeficiency, and Human T- Lymphotrophic Viruses,” American
Journal of Public Health, Vol. 86 (1995), pp. 655- 61.

19 J. I. Tokars, E. R. Miller, M. J. Alter, M. J. Ardunio, “National
Surveillance of Dialysis Associated Diseases in the United States,
1995,” ASAIO Journal, Vol. 44 (1998), pp. 98- 107.

ENCLOSURE I ENCLOSURE I

GAO- 01- 60R Needlestick Prevention 17 LIMITATIONS OF THE DATA AND ANALYSIS

Our analysis of the costs and benefits of using needles with safety features
excludes other factors that could affect the results shown above. We
excluded these factors to focus on a broad range of costs and benefits
associated with these features. Additional factors that could be considered
include the relative costs of specific features, the costs and benefits
associated with the reduced use of needles, the costs and benefits
associated with improved training in the use of needles, the avoided
treatment costs resulting from reduced numbers of needlestick injuries, and
the additional time needed to perform a procedure without using a needle.

The total costs of needlestick prevention devices may be affected by the
relative cost of specific features and by how many of those features are
used. Our model assumes that the costs of all needlestick prevention devices
will increase by roughly the same amount. However, if the increased cost
varies by feature type, the overall benefits also may vary. For example,
safety features for hypodermic syringes may cost three times as much as
conventional needles, while other types of safety features may cost twice as
much. In the latter case, the total cost of adopting these features will be
between two and three times the cost of conventional needles.

ENCLOSURE II ENCLOSURE II

GAO- 01- 60R Needlestick Prevention 18 COMMENTS FROM THE CENTERS FOR DISEASE
CONTROL AND PREVENTION

(201098)
*** End of document. ***