Emerging Infectious Diseases: Consensus on Needed Laboratory Capacity
Could Strengthen Surveillance (Letter Report, 02/05/99, GAO/HEHS-99-26).

Pursuant to a congressional request, GAO reviewed the nation's
infectious diseases surveillance network, focusing on the: (1) extent to
which states conduct public health surveillance and laboratory testing
of selected emerging infectious diseases; (2) problems state public
health officials face in gathering and using laboratory-related data in
the surveillance of emerging infectious diseases; and (3) assistance
that the Department of Health and Human Services' Centers for Disease
Control and Prevention (CDC) provides to states for laboratory-related
surveillance and the value of this assistance to state officials.

GAO noted that: (1) surveillance and testing for important emerging
infectious diseases are not comprehensive in all states, leaving gaps in
the nation's infectious diseases surveillance network; (2) GAO's survey
found that most states conduct surveillance of five of the six emerging
infectious diseases GAO asked about, and state public health
laboratories conduct tests to support state surveillance of four of the
six; (3) over half of the state laboratories do not conduct tests for
surveillance of hepatitis C and penicillin-resistant S. pneumoniae; (4)
many state epidemiologists believe that their infectious diseases
surveillance programs should expand, and they cited a need to gather
more information on antibiotic-resistant diseases; (5) just over half of
the state public health laboratories have access to advanced molecular
technology, which could be valuable to all states' diseases surveillance
efforts; (6) few states require the routine submission of specimens or
isolated quantities of a pathogen from patients with certain diseases
for testing in state laboratories--a step CDC has urged them to adopt to
improve the quality of surveillance information; (7) many state
laboratory directors and epidemiologists reported that inadequate
staffing and information-sharing problems hinder their ability to
generate and use laboratory data to conduct infectious diseases
surveillance; (8) participants in the surveillance network often lack
basic computer hardware or integrated systems to allow them to rapidly
share information; (9) many state officials told GAO that they did not
have sufficient staffing and technology resources, and public health
officials have not agreed on a consensus definition of the minimum
capabilities that state and local health departments need to conduct
infectious diseases surveillance; (10) this lack of consensus makes it
difficult to assess resource needs; (11) most state laboratory directors
and epidemiologists placed high value on CDC's testing and consulting
services, training, and grant funding and said these services were
critical to their ability to use laboratory data to detect and monitor
emerging infections; (12) state officials said CDC needs to better
integrate its data systems and help states build systems that link them
to local and private surveillance partners; and (13) state officials
would like CDC to provide more hands-on training experience.

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

 REPORTNUM:  HEHS-99-26
     TITLE:  Emerging Infectious Diseases: Consensus on Needed 
             Laboratory Capacity Could Strengthen Surveillance
      DATE:  02/05/99
   SUBJECT:  Public health research
             Health care personnel
             Disease detection or diagnosis
             Infectious diseases
             Health research programs
             Laboratories
             Testing
             Federal/state relations
             Medical information systems
IDENTIFIER:  CDC National Immunization Program
             CDC Epidemic Intelligence Service Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** 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.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Chairman, Subcommittee on Public Health, Committee on
Health, Education, Labor, and Pensions, U.S.  Senate

February 1999

EMERGING INFECTIOUS DISEASES -
CONSENSUS ON NEEDED LABORATORY
CAPACITY COULD STRENGTHEN
SURVEILLANCE

GAO/HEHS-99-26

Surveillance of Emerging Infectious Diseases

(108351)


Abbreviations
=============================================================== ABBREV

  APHL - Association of Public Health Laboratories
  CDC - Centers for Disease Control and Prevention
  CSTE - Council of State and Territorial Epidemiologists
  EIP - Emerging Infections Program
  EIS - Epidemic Intelligence Service
  ELC - Epidemiology and Laboratory Capacity Program
  HHS - Department of Health and Human Services
  HISS - Health Information and Surveillance System
  HUS - hemolytic uremic syndrome
  INPHO - Information Network for Public Health Officials
  NCHSTP - National Center for HIV, Sexually Transmitted Diseases,
     and Tuberculosis Prevention
  NCID - National Center for Infectious Diseases
  NETSS - National Electronic Telecommunications System for
     Surveillance
  PFGE - pulsed field gel electrophoresis
  PHLIS - Public Health Laboratory Information System
  TTP - thrombotic thromobocytopenic purpura
  USDA - U.S.  Department of Agriculture

Letter
=============================================================== LETTER


B-280933

February 5, 1999

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

Dear Mr.  Chairman: 

The spread of infectious diseases is a public health problem once
thought to be largely under control.  However, outbreaks over the
last decade illustrate that infectious diseases remain a serious
public health threat.  For example, in 1993, more than 400,000 people
became ill from a city's drinking water contaminated with
Cryptosporidium parvum--a common parasite resistant to chlorination
and other water treatment measures.  Over 4,000 people were
hospitalized, and 55 died.  In 1996, drinking apple juice
contaminated with a virulent strain of E.  coli bacteria made more
than 60 people seriously ill and caused the death of one person.  And
in 1998, 26 children became ill from playing in a swimming pool
contaminated by a virulent strain of E.  coli.  Four of the children
developed a serious complication that affects the blood and kidneys. 

The resurgence of some infectious diseases is particularly alarming
because previously effective forms of control are breaking down.  For
example, some pathogens (disease-causing organisms) have become
resistant to antibiotics used to bring them under control or have
developed strains that no longer respond to the antibiotics. 

Monitoring infectious diseases--identifying diseases and their
sources--is critical for determining control and prevention efforts. 
Public health officials refer to this activity as surveillance--the
ongoing collection, analysis, and interpretation of disease-related
data to plan, implement, and evaluate public health actions.  Many
public health experts have raised concerns about the adequacy of the
nation's infectious diseases surveillance network, especially for
those diseases considered to be emerging--that is, ones more
prevalent now than 20 years ago or ones that show signs of becoming
more prevalent in the near future. 

In light of these concerns, you asked us to examine the nation's
surveillance network and to focus on the contribution of
laboratories, since new technology gives them an increasingly
important role in identifying pathogens and the sources of outbreaks. 
Specifically, you asked us to (1) determine the extent to which
states conduct public health surveillance and laboratory testing of
selected emerging infectious diseases, (2) identify the problems
state public health officials face in gathering and using
laboratory-related data in the surveillance of emerging infectious
diseases, and (3) describe the assistance that the Department of
Health and Human Services' (HHS) Centers for Disease Control and
Prevention (CDC) provides to states for laboratory-related
surveillance and the value of this assistance to state officials. 

To provide information on the contribution of laboratories to the
surveillance network, we surveyed the directors of all state public
health laboratories and infectious diseases epidemiology\1 programs
that report disease-related information directly to CDC, including
officials in all 50 states, 5 territories, the District of Columbia,
and New York City.\2 We also conducted case studies in Kentucky, New
York, and Oregon; spoke with additional state and local public health
officials around the country; and interviewed CDC officials.  We
focused our work on six specific emerging infectious diseases or
pathogens:  tuberculosis, Shiga-like toxin-producing E.  coli
(including E.  coli O157:H7),\3 pertussis, Cryptosporidium parvum,
hepatitis C virus, and penicillin-resistant Streptococcus pneumoniae. 
Our methodology is described in more detail in appendix I, the
results from our surveys are in appendixes II and III, and details on
the six diseases are in appendix IV.  Our work was conducted from
December 1997 through December 1998 in accordance with generally
accepted government auditing standards. 


--------------------
\1 Epidemiology is the study of the distribution and causes of
disease or injury in a population. 

\2 Throughout this report, we refer to this group collectively as
"states."

\3 Shiga-like toxin-producing E.  coli belong to a group of virulent
E.  coli that can produce severe intestinal bleeding.  Throughout
this report, we will refer to the group by the name of its most
well-known member, E.  coli O157:H7. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Surveillance and testing for important emerging infectious diseases
are not comprehensive in all states, leaving gaps in the nation's
infectious diseases surveillance network.  Our survey found that most
states conduct surveillance of five of the six emerging infectious
diseases we asked about, and state public health laboratories conduct
tests to support state surveillance of four of the six.  However,
over half of the state laboratories do not conduct tests for
surveillance of hepatitis C and penicillin-resistant S.  pneumoniae. 
Many state epidemiologists believe that their infectious diseases
surveillance programs should expand, and they frequently cited a need
to gather more information on antibiotic-resistant diseases.  Just
over half of the state public health laboratories have access to
advanced molecular technology, which many experts believe could be
valuable to all states' diseases surveillance efforts.  Furthermore,
few states require the routine submission of specimens or isolated
quantities of a pathogen from patients with certain diseases for
testing in state laboratories--a step CDC has urged them to adopt to
improve the quality of surveillance information. 

Many state laboratory directors and epidemiologists reported that
inadequate staffing and information-sharing problems hinder their
ability to generate and use laboratory data to conduct infectious
diseases surveillance.  For example, they believe that the number of
laboratory staff to perform tests and the number of epidemiology
staff who can analyze data and translate surveillance information
into disease prevention and control activities are insufficient. 
They also cited a need for training to ensure that their staffs have
the skills to take advantage of technological advances in laboratory
methods, information-sharing systems, or both.  Participants in the
surveillance network, particularly at the local level, often lack
basic computer hardware or integrated systems to allow them to
rapidly share information.  State officials also expressed concerns
about CDC's many separate data reporting systems, which result in
duplication of effort and drain scarce staff resources.  Although
many state officials told us that they did not have sufficient
staffing and technology resources, public health officials have not
agreed on a consensus definition of the minimum capabilities that
state and local health departments need to conduct infectious
diseases surveillance.  This lack of consensus makes it difficult to
assess resource needs.  We are recommending that the Director of CDC
lead an effort to help federal, state, and local public health
officials create consensus on the core capacities needed at each
level of government. 

CDC provides state and local health departments with a wide range of
technical, financial, and staff resources to help maintain or improve
their ability to detect and respond to emerging infectious disease
threats.  Most state laboratory directors and epidemiologists placed
high value on CDC's testing and consulting services, training, and
grant funding and said these services were critical to their ability
to use laboratory data to detect and monitor emerging infections. 
However, they identified a number of ways in which these services
could be improved.  Specifically, most state officials said CDC needs
to better integrate its data systems and help states build systems
that link them with local and private surveillance partners.  Many
state officials would also like CDC to provide more hands-on training
experience.  State officials also pointed out that obtaining
assistance with problems that cut across programmatic boundaries
could be improved if CDC's departments that focus on specific
diseases communicated better with one another. 


   BACKGROUND
------------------------------------------------------------ Letter :2

Emerging infectious diseases pose a growing health threat to people
in this country and around the world.  The causes of this increase
are complex and often difficult to anticipate.  For example,
increased development, deforestation, and other environmental changes
have brought people into contact with animals or insects that harbor
diseases only rarely encountered before.  Not all emerging infections
are unfamiliar diseases, however.  Some pathogens have developed
resistance to the antibiotics that brought them under control just a
generation ago.  Moreover, the threefold increase in international
travel during the past 20 years and greater importation of fresh
foods across national borders allow infectious diseases to spread
rapidly.  As these diseases travel, they interact with growing
numbers of people who have weakened immunity, such as transplant
recipients, elderly persons, patients treated with radiation, and
those infected with HIV/AIDS. 

With the introduction of antibiotics in the 1940s and the development
of vaccines for diseases like polio, there was widespread optimism
that infectious diseases could be eliminated completely.  As a
result, public health officials shifted some monitoring efforts to
other health problems, such as chronic diseases.  By 1986, CDC had
discontinued surveillance of drug-resistance trends in tuberculosis. 
The resurgence of tuberculosis and the appearance of HIV/AIDS thus
caught the nation's public health system off guard. 

Today, infectious diseases account for considerable health care costs
and lost productivity.  In the United States, an estimated one-fourth
of all doctor visits are for infectious diseases.  Foodborne
illnesses, some of which were unrecognized 20 years ago, are
estimated to cause up to 33 million cases and 9,000 deaths annually
and to cost as much as $22 billion a year.  The number of pathogens
resistant to one or more previously effective antibiotics is
increasing rapidly, adding to health care costs and threatening to
return the nation to the pre-antibiotic era.  Antibiotic resistance
limits effective treatment options, with potentially fatal results. 
Resistant infections that people acquire during hospitalizations are
estimated to cost as much as $4 billion and cause 19,000 deaths a
year. 


      SURVEILLANCE IS THE PRIMARY
      PUBLIC HEALTH TOOL TO DETECT
      AND MONITOR INFECTIONS
---------------------------------------------------------- Letter :2.1

Surveillance is public health officials' most important tool for
detecting and monitoring both existing and emerging infectious
diseases.  Without an adequate surveillance system, local, state, and
federal officials cannot know the true scope of existing health
problems and may not recognize new diseases until many people have
been affected.  They rely on surveillance data to focus their staff
and dollar resources on preventing and controlling the diseases that
most threaten populations within their jurisdictions.  Health
officials also use surveillance data to monitor and evaluate the
effectiveness of prevention and control programs. 


         PASSIVE AND ACTIVE
         SURVEILLANCE
-------------------------------------------------------- Letter :2.1.1

Because known diseases can become emerging infections by changing in
unanticipated ways, the methods for detecting emerging infections are
the same ones used to monitor infectious diseases generally.  These
methods can be characterized as passive or active. 

When using passive surveillance methods, public health officials
notify laboratory and hospital staff, physicians, and other relevant
sources about disease data they should report.  These sources in turn
must take the initiative to provide data to the health department,
where officials analyze and interpret the information as it comes in. 

Under active surveillance, public health officials contact people
directly to gather data.  For example, state or local health
department staff could call commercial laboratories each week to ask
if any tests conducted for cryptosporidiosis yielded positive
results.  Active surveillance produces more complete information than
passive surveillance, but it takes more time and costs more. 

Infectious diseases surveillance in the United States depends largely
on passive methods of collecting disease reports and laboratory test
results.  Consequently, the surveillance network relies on the
participation of health care providers, private laboratories, and
state and local health departments across the nation. 


         SURVEILLANCE DEPENDS ON
         PARTICIPATION BY MANY
-------------------------------------------------------- Letter :2.1.2

States have principal responsibility for protecting the public's
health and, therefore, take the lead role in conducting surveillance. 
Each state decides for itself which diseases will be reported to its
health department, where reports should be submitted, and which
information it will then pass on to CDC. 

The surveillance process usually begins when a person with a
reportable disease seeks care.  To help determine the cause of the
patient's illness, a physician may rely on a laboratory test, which
could be performed in the physician's own office, a hospital, an
independent clinical laboratory, or a public health laboratory. 
State and local health departments that provide clinical services
also generate laboratory test results for infectious diseases
surveillance. 

Local health departments are often the first to receive the reports
of infectious diseases generated by physicians, hospitals, and
others.  Health department staff collect these reports, check them
for completeness, contact health care professionals to obtain missing
information or clarify unclear responses, and forward them to state
health agencies.  Staff resources devoted to disease reporting vary
with the overall size and mission of the health department.  Since
nearly half of local health agencies have jurisdiction over a
population of fewer than 25,000, many cannot support a large,
specialized staff to work on disease reporting. 

In state health departments, epidemiologists analyze data collected
through the disease reporting network, decide when and how to
supplement passive reporting with active surveillance methods,
conduct outbreak and other disease investigations, and design and
evaluate disease prevention and control efforts.  They also transmit
state data to CDC, providing routine reporting on selected diseases. 
Many state epidemiologists and laboratory directors provide the
medical community with information obtained through surveillance,
such as rates of disease incidence and prevailing patterns of
antimicrobial resistance. 

Federal participation in the infectious diseases surveillance network
focuses on CDC activities--particularly those of the National Center
for Infectious Diseases (NCID), which operates CDC's infectious
diseases laboratories.  CDC analyzes the data furnished by states to
(1) monitor national health trends, (2) formulate and implement
prevention strategies, and (3) evaluate state and federal disease
prevention efforts.  CDC routinely provides public health officials,
medical personnel, and others information on disease trends and
analyses of outbreaks.  Through NCID and other units--such as the
National Immunization Program and the National Center for HIV,
Sexually Transmitted Diseases, and Tuberculosis Prevention
(NCHSTP)--CDC offers an array of scientific and financial support for
state infectious diseases surveillance, prevention, and control
programs.  NCID officials said that most of their 1,100 staff and
$186 million budget in fiscal year 1998 were devoted to assisting
state infectious diseases efforts.  For example, CDC provides testing
services and consultation not available at the state level; training
on infectious diseases and laboratory topics, such as testing methods
and outbreak investigations; and grants to help states conduct
diseases surveillance.\4 The Epidemiology Program Office provides
training and technical assistance related to software for disease
reporting and oversees data integration efforts. 


--------------------
\4 The grants discussed in this report are cooperative agreements in
which CDC helps direct and monitor funded activities. 


      LABORATORIES PLAY AN
      ESSENTIAL ROLE IN
      SURVEILLANCE OF EMERGING
      INFECTIOUS DISEASES
---------------------------------------------------------- Letter :2.2

Public health and private laboratories are a vital part of the
surveillance network because only laboratory results can definitively
identify pathogens.  In addition, they often are an essential
complement to a physician's clinical impressions.  According to
public health officials, the nation's 158,000 laboratories are
consistent sources of passively reported information for infectious
diseases surveillance.\5 Independent commercial and hospital
laboratories may also share with public health agencies information
gathered through their private surveillance efforts, such as studies
of patterns of antibiotic resistance or the spread of diseases within
a hospital. 

Every state has at least one state public health laboratory to
support its infectious diseases surveillance activities and other
public health programs.  Some states operate one or more regional
laboratories to serve different parts of the state.  In five
states--Iowa, Nebraska, Nevada, Ohio, and Wisconsin--academic
institutions, such as university medical schools, provide public
health laboratory testing.  State laboratories conduct testing for
routine surveillance or as part of special clinical or epidemiologic
studies.  These laboratories provide diagnostic tests for rare or
unusual pathogens that are not always available in commercial
laboratories or tests for more common pathogens that use new
technology still needing controlled evaluation.  State public health
laboratories provide specialized testing for low-incidence, high-risk
diseases, such as tuberculosis and botulism.  Testing they provide
during an outbreak contributes greatly to tracing the spread of the
outbreak, identifying the source, and developing appropriate control
measures.  Epidemiologists rely on state public health laboratories
to document trends and identify events that may indicate an emerging
problem.  Many state laboratories also provide licensing and quality
assurance oversight of commercial laboratories. 

State public health laboratories are increasingly able to use new
advanced molecular technology to identify pathogens at the molecular
level.  Often, these tests provide information that is used not to
diagnose and treat individual patients but to tell epidemiologists
whether cases of illness are caused by the same strain of
pathogen--information that is not available from clinical records or
other conventional epidemiologic methods.  Public health officials
have already used this type of laboratory information to identify the
movement of diseases through a community in ways that would not have
been possible 5 years ago.  For example, staff in Minnesota's
laboratory use a molecular technology called pulsed field gel
electrophoresis (PFGE) to test "isolates" (isolated quantities of a
pathogen) of E.  coli O157:H7 that laboratories in the state must
submit.  From 1994 to 1995, the resulting DNA fingerprint patterns
identified 10 outbreaks--almost half of which would not have been
identified by traditional surveillance methods.  Using the laboratory
results, epidemiologists were able to find the sources of
contamination and eliminate them, thus preventing additional
infections. 

CDC laboratories provide highly specialized tests not always
available in state public health or commercial laboratories and
assist states with testing during outbreaks.  The staff at CDC's
laboratories also have a broad range of expertise identifying
pathogens.  These laboratories help diagnose life-threatening,
unusual, or exotic infectious diseases; provide information on cases
of infectious diseases for which satisfactory tests are not widely or
commercially available; and confirm public or private laboratory test
results that were atypical or difficult to interpret.  According to
NCID officials, CDC laboratories provide testing services and
consultations on conducting tests or interpreting results to every
state.  CDC also conducts research to develop improved diagnostic
methods and trains state laboratory staff to use them. 


--------------------
\5 U.S.  laboratories include about 90,000 laboratories in
physicians' offices; 5,800 independent clinical laboratories; 9,000
hospital laboratories; and 53,000 other laboratories, such as those
in state and local health departments, nursing homes, and other
health care facilities.  In 1993, about 60 percent of the nation's
approximately 3,000 local health departments provided at least some
laboratory services, often for a limited number of diseases. 


   NOT ALL STATES CONDUCT
   SURVEILLANCE AND TESTING FOR
   IMPORTANT EMERGING INFECTIONS
------------------------------------------------------------ Letter :3

While state surveillance and laboratory testing programs are
extensive, not all include every significant emerging infectious
disease, leaving gaps in the nation's surveillance network.  Each
state decides which diseases it includes in its surveillance program
and which diseases it routinely reports to CDC.  Many state
epidemiologists believe their surveillance programs need to add or
focus more attention on important infectious diseases, including
hepatitis C and antibiotic-resistant diseases.  Our survey found that
almost all states conduct surveillance of E.  coli O157:H7,
tuberculosis, pertussis, and hepatitis C, but fewer collect
information on cryptosporidiosis and penicillin-resistant S. 
pneumoniae.  State public health laboratories commonly perform tests
to support state surveillance programs for E.  coli O157:H7,
tuberculosis, pertussis, and cryptosporidiosis.  Most, however, do
not test for hepatitis C and penicillin-resistant S.  pneumoniae. 
Slightly more than half the state laboratories use PFGE, which state
and CDC officials believe could be valuable to most or all states'
diseases surveillance efforts.  Few states have followed CDC's
suggestion to improve surveillance by requiring medical providers and
laboratories to routinely submit specimens for testing in state
public health laboratories. 


      STATES DETERMINE WHICH
      DISEASES ARE UNDER NATIONAL
      AND STATE SURVEILLANCE
---------------------------------------------------------- Letter :3.1

Each year, the Council of State and Territorial Epidemiologists
(CSTE), in consultation with CDC, reviews the list of infectious
diseases that are "nationally notifiable"--that is, important enough
for the nation as a whole to merit routine reporting to CDC.  The
list currently includes 52 infectious diseases.\6 States are under no
obligation to adopt the nationally notifiable diseases for their own
surveillance programs, and state reporting to CDC is voluntary.  A
1997 CSTE survey of state health departments found that 87 percent of
states included at least 80 percent of the 52 nationally notifiable
diseases in their surveillance programs, and about one-third of
states included over 90 percent.\7 Lists of state reportable diseases
vary considerably, partly because of differences in the extent to
which diseases occur in different regions of the country.\8


--------------------
\6 State and CDC officials periodically revise the list of nationally
notifiable diseases. 

\7 Survey results did not include information from the District of
Columbia and three of the territories. 

\8 States also request or require reporting of diseases of local
importance that are not on the list of nationally notifiable
diseases.  The 1997 survey revealed that a total of 74 other
infectious diseases were each included in the surveillance program of
at least one state. 


      SURVEILLANCE OF SOME
      DISEASES IS NOT WIDESPREAD
---------------------------------------------------------- Letter :3.2

Of the six diseases covered by our survey, nearly all the states
include at least four in their diseases surveillance--most commonly
tuberculosis, E.  coli O157:H7, pertussis, and hepatitis C.  A
slightly smaller number of states include cryptosporidiosis in their
surveillance programs.  Penicillin-resistant S.  pneumoniae was
covered least often, with about two-thirds of the states including
it.  For all of the diseases except penicillin-resistant S. 
pneumoniae, most states require health care providers, laboratories,
and others to submit disease reports to public health officials. 
These reports contain information such as demographic characteristics
of the ill person, the date disease symptoms appeared, and the
suspected or confirmed diagnosis.  (See fig.  1.)

   Figure 1:  State Surveillance
   of Selected Emerging
   Infections:  Diseases Included,
   Reporting Requirements, and
   State Public Health Laboratory
   Testing

   (See figure in printed
   edition.)

Note:  State surveillance and reporting requirement data include 55
states; state laboratory testing data include 54 states that provided
complete data. 

Over three-quarters (44) of the responding epidemiologists told us
that their surveillance programs either leave out or do not focus
sufficient attention on important infectious diseases. 
Antibiotic-resistant diseases, including penicillin-resistant S. 
pneumoniae, and hepatitis C were among the diseases they cited most
often as deserving greater attention.\9


--------------------
\9 The epidemiologist in one state reported taking steps to add
hepatitis C and penicillin-resistant S.  pneumoniae to the state's
list of reportable diseases.  Another state epidemiologist reported
adding hepatitis C to the list of reportable diseases, and a third
reported adding penicillin-resistant S.  pneumoniae. 


      STATE PUBLIC HEALTH
      LABORATORY TESTING DOES NOT
      ALWAYS ACCOMPANY
      SURVEILLANCE
---------------------------------------------------------- Letter :3.3

State laboratory testing to support state surveillance of the six
emerging infections in our survey varies across the nation.  Testing
is most common for four of the six:  tuberculosis, E.  coli O157:H7,
pertussis, and cryptosporidiosis (see fig.  1).  In 43 of the 54
state responses we analyzed,\10 the state public health laboratory
conducts testing for four or more of the diseases included in its
state's surveillance program.\11 Testing to support state
surveillance of hepatitis C and penicillin-resistant S.  pneumoniae
occurs in fewer than half of the states. 

State and CDC officials believe that most, and possibly all, states
should have PFGE technology, which can be used to study many diseases
and greatly improves the ability to detect outbreaks.  However, for
the diseases we asked about in our survey, state public health
laboratories are less likely to use advanced molecular technology
than more conventional techniques.  For example, slightly more than
half the state laboratories reported using PFGE technology to support
state surveillance efforts.  Twenty-nine of the 54 laboratory
directors responding to our survey reported using PFGE to support E. 
coli O157:H7 surveillance, and nine of these laboratories also use it
for pertussis surveillance. 

If a state laboratory provided testing in support of state-level
surveillance of a specific disease, we asked directors to assess the
adequacy of their testing equipment for that disease.  Laboratory
directors' views about the adequacy of the testing equipment they use
varied somewhat by disease but were generally positive.  Eighty
percent or more of the laboratory directors rated their equipment as
generally or very adequate for four diseases--tuberculosis, E.  coli
O157:H7, cryptosporidiosis, and hepatitis C.  Percentages were
slightly lower for pertussis (69 percent) and penicillin-resistant S. 
pneumoniae (68 percent).\12

State epidemiologists' views about the adequacy of the testing
information provided by state laboratories vary considerably by
disease.\13 More than 94 percent rated their state laboratory as very
or generally adequate to provide testing information for tuberculosis
and E.  coli O157:H7.  More than 70 percent said their state
laboratory is generally or very adequate for generating information
on pertussis and cryptosporidiosis.  In contrast, only about
one-third of epidemiologists said the information generated by their
state laboratory for hepatitis C (32 percent) and
penicillin-resistant S.  pneumoniae (37 percent) is generally or very
adequate. 

We also found that many states do not require other public and
private laboratories or medical providers to submit to the state
public health laboratory specimens or isolates from persons with
certain diseases.  CDC has urged states to consider developing such
laws because gathering specimens from across the state helps ensure
that the state's surveillance data include a diverse sample of the
state's population.  Such action by states also contributes to more
comprehensive national data.  In all, 29 states require specimens for
one or more of the six diseases in our survey:  5 states require
specimens for four diseases, 4 states require specimens for three
diseases, 9 states for two, and 11 for one disease.\14 Specimens of
tuberculosis and E.  coli O157:H7 are required most frequently. 


--------------------
\10 To study the full range of state surveillance, reporting, and
testing efforts for the conditions we asked about, we paired the
responses of laboratory directors and epidemiologists by state.  For
these data, we analyzed only the responses of the 54 pairs of
epidemiologists and laboratory directors who provided complete
information.  We excluded the two states where only the laboratory
director responded and the one state where the laboratory director
did not provide complete information on surveillance testing. 

\11 The 11 other state laboratories conduct tests for one or more of
these six infectious diseases.  In some cases, the laboratory tests
three or fewer of the diseases as part of the state's surveillance
efforts; in others, the laboratory tests on behalf of other public or
private laboratories. 

\12 These results exclude laboratories that do not provide
surveillance-related testing for the specific disease. 

\13 If testing by the state laboratory was part of state-level
surveillance of a specific disease, we asked state epidemiologists to
assess the state laboratory's adequacy in generating the data needed
for surveillance. 

\14 Twenty-two states require no specimens but ask for voluntary
submission of specimens for one or more of the six diseases, usually
for special studies rather than routinely.  Three states neither
require nor ask that specimens of these diseases be sent to their
state public health laboratory. 


   OFFICIALS REPORT THAT STAFFING
   CONSTRAINTS AND WEAK
   INFORMATION SHARING IMPEDE
   SURVEILLANCE OF EMERGING
   INFECTIONS
------------------------------------------------------------ Letter :4

As part of our survey and field interviews, we asked state officials
to identify the problems they considered most significant in
conducting surveillance of emerging infectious diseases.  The
problems they cited fall principally into two categories:  staffing
and information sharing.  State epidemiologists reported that
staffing constraints prevent them from undertaking surveillance of
diseases they consider important.  Laboratory directors told us they
do not always have enough staff to conduct tests needed for
surveillance; furthermore, their staff need training to remain
current with technological advances.  Epidemiologists and laboratory
officials both said that public health officials often lack either
basic computer equipment or integrated data systems that would allow
them to rapidly share surveillance-related information with public
and private partners. 


      STAFFING AND TRAINING
      LIMITATIONS AFFECT ABILITY
      TO EXPAND LABORATORIES' ROLE
---------------------------------------------------------- Letter :4.1

Public health officials reported that the nation's infectious
diseases surveillance system is basically sound but could improve its
ability to detect emerging threats.  Most state officials believe
they need to expand their infectious diseases surveillance programs. 
However, both state laboratory directors and epidemiologists said
that such expansion has been constrained by staffing and training
limitations.  Most of the 44 epidemiologists who reported that they
need to expand coverage of important infectious diseases said
insufficient staff and funding resources prevent them from taking
this action.  Some noted that they need more and better trained staff
just to do a better job on diseases already included in their
programs. 

We found considerable variability among states in laboratory and
epidemiology staffing per 1 million population.  In total, we found
that during fiscal year 1997, states devoted a median of 8 staff
years per 1 million population to laboratory testing of infectious
diseases.  Laboratory staff year medians for individual types of
testing ranged from 0.4 for foodborne pathogens to 2.4 for all other
infectious diseases not specifically listed in table 1.  The median
for total epidemiology staff years per 1 million population was 14;
the range was from 0.1 for foodborne pathogens to 5 for HIV/AIDS. 
(See table 1.)



                                     Table 1
                     
                        State Public Health Laboratory and
                      Epidemiology Staff Years Per 1 Million
                          Population in Fiscal Year 1997

                                                    Laboratory     Epidemiology
                                                   staff years     staff years
                                                  --------------  --------------
Disease or condition                              Median   Range  Median   Range
------------------------------------------------  ------  ------  ------  ------
Tuberculosis                                         1.2    0-21     1.8    0-45
HIV/AIDS\a                                             1    0.2-     5.1   0-193
                                                              33
Sexually transmitted diseases                        1.4    0.1-     3.3    0-72
                                                              50
Foodborne diseases                                   0.4    0-17     0.1    0-33
Other emerging infectious diseases                   0.7    0-14     0.2    0-33
All other infectious diseases                        2.4    0-16     3.2    0-50
Total staff years devoted to infectious diseases     8.1    1.3-      14    2.1-
 programs                                                     89             321
--------------------------------------------------------------------------------
\a HIV/AIDS was excluded from the "other emerging infectious
diseases" category. 

The majority of state laboratory directors indicated that their
staffing resources are generally adequate to generate test results
for the diseases in our study.\15 For each of the four diseases that
state laboratories most commonly support, more than 75 percent of
directors rated their staff as generally or very adequate to perform
the tests.\16 Among the smaller number of state laboratories that
conduct tests to support surveillance of hepatitis C and
penicillin-resistant S.  pneumoniae, a smaller percentage of
laboratory directors considered their staff resources at least
adequate (68 percent and 58 percent, respectively). 

Some state laboratory and epidemiology officials told us that
staffing constraints prevent them from making full use of testing
capacity.  For example, the laboratory director in a state that had
acquired PFGE technology cited lack of staff time as one reason for
not routinely using PFGE in surveillance of E.  coli O157:H7.  As a
result, he said, the incidence of E.  coli O157:H7 in his state is
probably understated.  If resources were available, he would also
like laboratory staff to test pertussis specimens collected during a
recent outbreak to determine whether the increase in reported cases
was a true outbreak or the result of increased awareness--and
reporting--of the disease following the death of a child.  Thirty-six
state laboratory directors reported having vacancies during the past
year and said the vacancies had negatively affected their
laboratory's ability to support their state's infectious diseases
surveillance activities.  Nine rated the impact as great or
significant.  Administrative and financial constraints, such as
hiring freezes or budget reductions, were most often responsible for
the vacancies. 

Laboratory officials noted that advances in scientific knowledge and
the proliferation of molecular testing methods have created a need
for training to update the skills of current staff.  They reported
that such training is often either unavailable or inaccessible
because of funding or administrative constraints.  For example,
several state officials said that in reducing costs, training budgets
are often cut first.  In other states, staff are subject to per
capita limits on training or travel expenses.  Therefore, if CDC or
another source provided additional funding, these funds could not be
used. 


--------------------
\15 This represents the views of laboratory directors whose staff
conduct tests to support surveillance of at least one of the diseases
we asked about. 

\16 The specific percentages for the four diseases are E.  coli
O157:H7 (82 percent), tuberculosis (75 percent), pertussis (78
percent), and cryptosporidiosis (77 percent). 


      LACK OF EQUIPMENT AND
      CUMBERSOME SYSTEMS HINDER
      INFORMATION SHARING
---------------------------------------------------------- Letter :4.2

For health crises that need an immediate response--as when a serious
and highly contagious disease appears in a school or among restaurant
staff--rapid sharing of surveillance information is critical.  Public
health officials told us, however, that many state and local health
departments do not have the basic equipment to efficiently share
information across the surveillance network.  Computers and other
equipment, such as answering or fax machines, that can shorten the
process of sharing surveillance information from weeks to a day or
less are not always available. 

Our survey responses indicate that state laboratory directors use
electronic communication systems much less often than state
epidemiologists use them.  Although about three-quarters of
responding state laboratory directors use electronic systems to
communicate within their laboratories, they do not frequently use
electronic systems to communicate with others.  Almost 40 percent of
laboratory directors reported using computerized systems to little or
no extent for receiving surveillance-related data, and 21 percent use
them very little for transmitting data.  While state epidemiologists
use electronic systems more than laboratory directors, they also use
them less commonly to receive information (42 percent) than to report
it (62 percent). 

One reason for the limited use of electronic systems may be the lack
of equipment.  A 1996 CDC survey found that, on average, about 20
percent of staff in most state health agencies did not have access to
desktop computers that were adequate for sharing information rapidly. 
Forty percent of local health officials responding to a 1996 survey
conducted by the National Association of City and County Health
Officials said they lacked such equipment.\17 State and local health
officials most often attributed the lack of computer equipment and
integrated data processing and management systems to insufficient
funding. 

The absence of equipment means some tasks that could be automated
must be done by hand--and in some cases must be done by hand even
after data have already been processed in electronic form.  For
example, representatives from two large, multistate private clinical
laboratories told us that data stored electronically in their
information systems had to be converted to paper so that it could be
reported to local health departments.  In one state we visited, a
local health department mails data stored on disk to the state health
agency because it lacks the equipment to transfer the data
electronically. 

Even with adequate computer equipment, the difficulty of creating
integrated information systems can be formidable.  Not only does
technology change rapidly, but public health data are currently
stored in thousands of places, including the record and information
systems of public health agencies and health care institutions,
individual case files, and data files of surveys and surveillance
systems.  These data are in isolated locations that have differing
hardware and software structures and considerable variation in how
the data are coded, particularly for laboratory test results. 

CDC operates over 100 data systems to monitor over 200 health events,
such as specific infectious diseases.  Many of these systems collect
data from state surveillance programs.  This patchwork of data
systems arose, in part, to meet CDC and state needs for more detailed
information for particular diseases than was usually reported.  For
example, while information collected to determine incidence rates of
many nationally notifiable diseases consists of minimal geographic
and demographic data, the information collected to determine
incidence rates of tuberculosis includes information on personal
behavior, the presence of other diseases, and stays in institutional
settings, as well as geographic and demographic data.  The additional
information collected on tuberculosis also helps guide prevention and
control strategies. 

Public health officials told us that the multitude of databases and
data systems, software, and reporting mechanisms burdens staff at
state and local health agencies and leads to duplication of effort
when staff must enter the same data into multiple systems that do not
communicate with one another.  Furthermore, the lack of integrated
data management systems can hinder laboratory and epidemiologic
efforts to control outbreaks.  For example, in 1993 the lack of
integrated systems impeded efforts to control the hantavirus outbreak
in the Southwest.  Data were locked into separate databases that
could not be analyzed or merged with others, requiring public health
investigators to analyze individual paper printouts. 


--------------------
\17 Questionnaires were mailed to a random sample of 800 health
officials in local health agencies; 384 responded. 


      OTHER CONCERNS MAY ALSO
      AFFECT USE OF LABORATORY
      DATA
---------------------------------------------------------- Letter :4.3

State officials also raised concerns about a lack of complete data
for surveillance and the increased reliance on fees to fund state
laboratories, which they believe undermine their infectious diseases
surveillance efforts. 


         COMPLETENESS OF DATA
-------------------------------------------------------- Letter :4.3.1

Public health officials and experts acknowledge that, even when
states require reporting, the completeness of data reported varies by
disease and type of provider.  As might be expected, reporting of
severe and life-threatening diseases is more complete than reporting
of mild diseases.  However, when mild diseases are not reported,
outbreaks affecting a large number of people may go unnoticed until
deaths occur among people at higher than normal risk.  In addition,
reporting by practitioners in frequent contact with infectious
diseases, such as family practitioners, is more complete than
reporting by those who are not, such as surgeons.  Although
surveillance need not be complete to be useful, underreporting can
adversely affect public health efforts by leading to erroneous
conclusions about trends in incidence, risk factors for contracting a
disease, appropriate prevention and control measures, and treatment
effectiveness. 

Completeness of reporting is a concern for the surveillance of
illnesses that can produce mild symptoms, such as diarrheal
illnesses, which include many foodborne and waterborne conditions. 
Reported cases of some illnesses represent the tip of the iceberg, at
best.  A recent CDC-sponsored study estimated that 340 million annual
episodes of acute diarrheal illness occurred in the United States,
but only 7 percent of people who were ill sought treatment.  The
study further estimated that physicians requested laboratory testing
of a stool culture for 22 percent of those patients who sought
treatment, which produced about 6 million test results that could be
reported.\18 In cases of mild diarrheal illness, physicians may not
request laboratory tests to identify the pathogen because patients
with these diseases can get better without treatment or effective
treatments do not exist. 

Public health officials expressed varying views about how managed
care growth and the consolidation of the laboratory industry might
affect the completeness of surveillance data.  Some public health
officials and physicians believe that managed care--with its emphasis
on controlling costs--could lead doctors to order fewer diagnostic
tests, particularly those not needed for treatment decisions.  Also,
to the extent that managed care organizations less frequently use
specialists, results from specialized tests they employ would not be
generated.  Concerns about laboratory consolidation--particularly
when specimens are shipped to central testing facilities in other
states--stem from fears that out-of-state testing centers will not
report test results needed for surveillance, possibly because they
might not be aware of state reporting requirements regarding what
information should be reported and where to direct it.  In two states
we visited, representatives of large multistate independent
laboratories said their policy is to report test results in
accordance with state requirements.  One representative provided us
with documentation showing the various reporting requirements of
states in one region served by the laboratory.  Each of these
laboratories is participating in electronic laboratory reporting
pilot programs in different states. 

Other CDC and state public health officials believe that managed care
organizations and concentrated ownership of laboratories could
provide information that is potentially more consistent, complete,
and reliable than what public health officials now routinely obtain
through passive reporting.  They argue that because information on a
large number of patients is concentrated in a small number of
organizations, the number of contacts for active surveillance
projects is smaller and more manageable and information can be
analyzed from large databases.  Moreover, they add, these
organizations are likely to collect and store laboratory data
electronically, which could speed disease reporting. 

Our survey asked epidemiologists whether they or other agencies in
their states had evaluated the impacts of managed care and laboratory
consolidation on surveillance data; we could identify no systematic
evaluations on this issue.  Similarly, researchers who conducted a
survey for HHS did not find data that address concerns about the
impact of managed care.\19


--------------------
\18 H.  Herikstad and others,"Population-Based Estimates of the
Burden of Diarrheal Illness:  FoodNet 1996-1997" (Atlanta, Ga.: 
International Conference on Emerging Infectious Diseases, 1998). 

\19 Public Health Laboratories and Health System Change, The Lewin
Group, Oct.  6, 1997. 


         INCREASED RELIANCE ON
         FEES TO FUND STATE
         LABORATORIES
-------------------------------------------------------- Letter :4.3.2

Another concern state officials frequently mentioned is an increasing
reliance on fees to fund the operations of state public health
laboratories.  Over 30 laboratory directors responding to our survey
said their budgets were partly supported by fees for genetic
screening and tests for regulatory and licensure programs.  State
officials told us that an imbalance of fees in relation to
appropriated funding shifts the focus of laboratory operations away
from testing services beneficial to the entire community and toward
services that can be successfully marketed--a shift that they believe
could jeopardize fulfilling their public health mission.  One state
laboratory director said that over the past 15 years, state funding
has declined by more than half and fees are expected to cover the
difference.  He believes that if the laboratory loses contracts for
genetic or blood lead-level testing, he will have to reduce other
testing, such as for sexually transmitted diseases or CDC's influenza
surveillance. 


      NO PUBLIC HEALTH CONSENSUS
      DEFINES CORE CAPACITIES
      NEEDED FOR SURVEILLANCE
      SYSTEM
---------------------------------------------------------- Letter :4.4

Although many state officials are concerned about their staffing and
technology resources, public health officials have not developed a
consensus definition of the minimum capabilities that state and local
health departments need to conduct infectious diseases surveillance. 
For example, according to CDC and state health officials, there are
no standards for the types of tests state public health laboratories
should be able to perform; nor are there widely accepted standards
for the epidemiological capabilities state public health departments
need.  Public health officials have identified a number of elements
that might be included in a consensus definition, such as the number
and qualifications of laboratory and epidemiology staff; the
pathogens that each state laboratory should be able to identify and,
where relevant, test for antibiotic resistance; specialized
laboratory and epidemiology capability that should be available
regionally; laboratory and information-sharing technology each state
should have; and support services that CDC should provide. 

Recognizing this lack of guidance, CSTE, the Association of Public
Health Laboratories (APHL), and CDC have begun collaborating to
define the staff and equipment components of a national surveillance
system for infectious diseases and other conditions.  Their work is
to include agreements about the laboratory and epidemiology resources
needed to conduct surveillance, diseases that should be under
surveillance, and the information systems needed to share
surveillance data.  One goal of reaching this consensus would be to
give state and local health agencies the basis for setting priorities
for their surveillance efforts and determining the resources needed
to implement them. 


   CDC SERVICES ARE WIDE-RANGING
   AND GENERALLY PERCEIVED AS
   VALUABLE
------------------------------------------------------------ Letter :5

CDC provides state and local health departments with a wide range of
technical, financial, and staff resources to help maintain or improve
their ability to detect and respond to disease threats.  Many state
laboratory directors and epidemiologists said this assistance has
been essential to their ability to conduct infectious diseases
surveillance and to take advantage of new laboratory technology. 
However, a small number of laboratory directors and epidemiologists
believe CDC's assistance has not added much to their ability to
conduct surveillance of emerging infections, and many state officials
indicated that further improvements are needed, particularly in the
area of information-sharing systems. 


      SERVICES INCLUDE BOTH
      TECHNICAL AND FINANCIAL
      ASSISTANCE
---------------------------------------------------------- Letter :5.1

CDC's various units, particularly NCID, provide an array of technical
and financial support for state infectious diseases surveillance
programs.  In general, this support falls into the following six
areas:  testing and consulting, training, grant assistance, funding
for regional laboratories, staffing assistance, and
information-sharing systems. 

  -- Laboratory testing and consultation.  CDC staff and laboratories
     support state infectious diseases surveillance efforts with
     technical assistance and testing services that may not be
     available at the state level.  CDC staff provide consultation
     services on such matters as epidemiological methods and
     analysis, laboratory techniques, and interpretation of
     laboratory results.  Almost all of the state laboratory
     directors and epidemiologists responding to our survey said they
     use CDC's laboratory testing services and frequently consult
     with CDC staff. 

  -- Training.  CDC provides public health and medical personnel with
     training on a wide range of topics.  The training is offered
     through such means as interactive audio- or video-conferences,
     computer-assisted instruction, seminars, and hands-on workshops. 
     Since 1989, CDC has offered laboratory training through a
     collaboration with APHL.  An APHL and CDC assessment identified
     the need for training on current advances in food microbiology,
     fungal and viral infections, rabies, tuberculosis, and new and
     emerging pathogens.  To meet these needs, CDC developed a series
     of courses incorporating hands-on experience, offered in various
     locations around the country.  State laboratory directors and
     epidemiologists indicated they use CDC training extensively, and
     most said they participated in CDC-sponsored training in 1997. 

  -- Grant programs.  CDC's various grant and staffing assistance
     programs provide at least some support to the infectious
     diseases surveillance programs of all states.  In fiscal year
     1998, NCID distributed $31.2 million of its $185.7 million
     budget to state and local health agencies for infectious
     diseases programs.  NCID supports three major grant programs
     that aid state surveillance programs for emerging infectious
     diseases (see table 2).\20 Together these three grant programs
     provided about $20 million to state and local health departments
     in fiscal year 1997. 



                                         Table 2
                         
                            NCID's Grant Assistance for State
                           Infectious Diseases Programs During
                                     Fiscal Year 1997

Program description                     FY 1997 funding          1997 recipients
--------------------------------------  -----------------------  ------------------------
Tuberculosis grants
-----------------------------------------------------------------------------------------
Helps state laboratories improve their  $9.6 million, with       All 50 states, Los
testing ability to support state        awards ranging from      Angeles, and New York
tuberculosis surveillance and           $8,000 to $1.3 million   City
elimination efforts.


Emerging Infections Program (EIP) grants
-----------------------------------------------------------------------------------------
Helps states improve their              $5.8 million, with       California, Connecticut,
surveillance of emerging infections     annual awards ranging    Georgia, Maryland,
and produce information of national     from $645,000 to $1.2    Minnesota, New York,
significance. States have used funds    million                  Oregon
for active surveillance of drug-
resistant infections, foodborne and                              (CDC intends to add 3
waterborne diseases, and vaccine-                                states by 2000, bringing
preventable conditions; to conduct                               the total to 10.)
applied research on epidemiologic and
laboratory methods; and to implement
prevention projects. CDC began EIP
with funding for programs in four
states.


Epidemiology and Laboratory Capacity (ELC) grants
-----------------------------------------------------------------------------------------
Helps states and large local health     $4.3 million, with       California, Colorado,
departments strengthen and enhance      awards ranging from      Florida, Georgia,
their basic capacity for surveillance   $128,000 to $379,000     Hawaii, Illinois,
of and response to infectious                                    Kansas, Louisiana,
diseases. Funds allow states to                                  Maine, Massachusetts,
implement new technology, upgrade                                Michigan, New Jersey,
information systems, hire and train                              New York, New York City,
staff, and purchase office and                                   Ohio, Pennsylvania,
laboratory equipment. Projects include                           Tennessee, Utah,
building electronic reporting systems;                           Vermont, Washington,
using molecular laboratory methods in                            West Virginia,
outbreak investigations; and enhancing                           Wisconsin
surveillance of hepatitis C, diarrheal
illnesses, and other conditions. CDC                             (CDC added 8 states in
awarded ELC grants initially to 10                               1998 and plans to
states.                                                          involve all 50 state
                                                                 health departments as
                                                                 well as many territorial
                                                                 and large local health
                                                                 departments by 2002.)
-----------------------------------------------------------------------------------------
EIP and ELC grants, designed to strengthen and enhance state
surveillance abilities, are components of CDC's overall plan to
address emerging infectious diseases.\21

  -- Funding for regional laboratory networks.  To help with both
     state-specific and nationwide control and prevention efforts,
     CDC has sponsored development of regional laboratory networks
     that give states access to molecular testing services that may
     not be available in their own state laboratory.  The two main
     laboratory networks are PulseNet, which currently focuses on E. 
     coli O157:H7, and the Tuberculosis Genotyping Network (see table
     3). 



                                     Table 3
                     
                        CDC-Sponsored Regional Laboratory
                                     Networks

                           Laboratories
Description                participating              Notable results
-------------------------  -------------------------  --------------------------
PulseNet
--------------------------------------------------------------------------------
EIP and ELC funding        In 1998, the network       In 1997, Colorado's public
helped build PulseNet, a   included 4 state public    health laboratory, using
laboratory network that    health laboratories that   PFGE to develop DNA
uses PFGE to study E.   provide testing to nearby  fingerprints of E. coli
coli O157:H7.              states, 20 public health   O157:H7 isolates submitted
Salmonella typhimurium  laboratories that test     from laboratories in the
and other non-typhoidal    specimens from within      state, found identical
Salmonella were         their borders, and U.S.    fingerprint patterns in
recently added to the      Department of Agriculture  samples from 13 different
organisms under study;     (USDA) and Food and Drug   patients. Subsequent
more will be added in the  Administration             testing at a USDA
future. Participating      laboratories that test     laboratory matched the
laboratories are           food products.             fingerprints with those of
electronically linked to                              E. coli O157:H7
rapidly share PFGE                                    isolates recovered from
patterns of foodborne                                 ground beef taken from
pathogens for comparison.                             packages used by two of
                                                      the patients. State
                                                      officials concluded the
                                                      cases were linked to the
                                                      meat, which had been
                                                      distributed nationally.
                                                      The manufacturer, in
                                                      cooperation with federal
                                                      officials, removed 25
                                                      million pounds of
                                                      potentially contaminated
                                                      ground beef from U.S.
                                                      markets.


Tuberculosis Genotyping Network
--------------------------------------------------------------------------------
This network of enhanced   CDC selected seven         CDC and participating
tuberculosis surveillance  regional laboratories in   laboratories established a
uses restriction fragment  April 1996. In 1997, CDC   national database of
length polymorphism--a     gave these laboratories a  tuberculosis fingerprints.
DNA fingerprint            total of over $900,000.    Patterns in the database
technology--to trace the                              showed that drug-
spread of specific                                    resistant strains first
strains of the disease.                               found in New York City
The laboratories also                                 have spread to other parts
help states investigate                               of the country. The
outbreaks and identify                                fingerprints also showed
instances of laboratory                               that tuberculosis can be
contamination that                                    transmitted during brief
resulted in false                                     contact among people who
diagnoses.                                            do not work or live
                                                      together, an important
                                                      discovery that led to
                                                      improved treatment and
                                                      control programs.
--------------------------------------------------------------------------------
  -- Staffing assistance.  CDC provides a small number of staff
     resources to assist state infectious diseases programs through
     2-year Epidemic Intelligence Service (EIS) placements and
     fellowships in state or local health departments or
     laboratories.  About one-fourth of the 60 to 80 EIS participants
     selected each year work in state and local health departments. 
     Additionally, by February 1998, CDC had trained 18 laboratory
     fellows to work in state, local, and federal public health
     laboratories through its Emerging Infectious Diseases Laboratory
     Fellow Program, a collaborative effort with APHL; CDC plans to
     make 9 emerging diseases laboratory fellowships available
     through APHL and the CDC Foundation.\22 One goal of the
     fellowships is to strengthen the relationship of public health
     laboratories to infectious diseases and drug-resistance
     surveillance, prevention, and control efforts. 

  -- Information sharing.  Over the past several decades, CDC has
     developed and made available to states several general and
     disease-specific information management and reporting programs. 
     Virtually all states use two of these programs to report data on
     some infectious diseases to CDC--the Public Health Laboratory
     Information System (PHLIS) and the National Electronic
     Telecommunications System for Surveillance (NETSS).  PHLIS is
     used primarily by laboratories; NETSS is used primarily by
     epidemiology programs. 


--------------------
\20 CDC's NCHSTP provides grants that aid state surveillance of
HIV/AIDS, sexually transmitted diseases, and tuberculosis. 

\21 Centers for Disease Control and Prevention, Preventing Emerging
Infectious Diseases:  A Strategy for the 21st Century (Atlanta, Ga.: 
Department of Health and Human Services, Sept.  1998). 

\22 The CDC Foundation is a nonprofit corporation established under
the authority of the Preventive Health Amendments of 1992 (42 U.S.C. 
section 280 d-11) to support CDC's mission. 


      STATE OFFICIALS VALUE CDC'S
      ASSISTANCE BUT SEE A NEED
      FOR IMPROVEMENT IN
      INFORMATION-SHARING SYSTEMS
---------------------------------------------------------- Letter :5.2

Our surveys showed that overall state laboratory directors and
epidemiologists highly value the support CDC provides for their
surveillance efforts.  Usage and satisfaction levels were highest in
the areas of testing and consultation, training, and grant support. 
The area most often identified as needing improvement was the
development of information-sharing systems. 


         LABORATORY TESTING,
         CONSULTATION, AND
         TRAINING ASSISTANCE ARE
         VIEWED AS CRITICAL
-------------------------------------------------------- Letter :5.2.1

Many state laboratory directors and epidemiologists told us that
CDC's testing, consultation, and training services are critical to
their surveillance efforts.  In all three areas of assistance, more
than half of those responding to our survey indicated that the
services greatly or significantly improved their state's ability to
conduct surveillance (see fig.  2).  According to officials who spoke
with us, CDC's testing for unusual or exotic pathogens and the
ability to consult with experienced CDC staff are important,
particularly for investigating cases of unusual diseases.  However,
about 15 percent of survey respondents said CDC's testing services
made only modest improvements in their state's surveillance capacity. 

   Figure 2:  Views of State
   Laboratory Directors and
   Epidemiologists on Extent CDC's
   Testing, Consultation, and
   Training Improved Emerging
   Infectious Diseases
   Surveillance Ability

   (See figure in printed
   edition.)

Over 70 percent of epidemiologists responding to our survey said that
knowledgeable staff at CDC are easy to locate when they need
assistance, but many noted that help with matters involving more than
one CDC unit is very difficult to obtain.  Many state officials who
spoke with us thought that this problem arose because staff in
different units do not seem to communicate well with each other.  One
official described CDC's units as separate towers that do not
interact. 

A number of state officials commented that CDC provides tests and
consultation very promptly when people are at risk--for example
during outbreaks of life-threatening diseases--but less quickly in
other circumstances.  To provide more timely consultation, CDC has
developed an on-line image-sharing ability that allows CDC staff and
health professionals in remote locations to view an organism under a
microscope at the same time.  In one state, staff at CDC and a
surgeon in another state used this capacity during an operation to
identify a parasite as the cause of the patient's eye problem,
allowing the surgeon to rule out cancer as a diagnosis and
eliminating the need to remove the patient's eye. 

Some state officials and survey respondents said that in less urgent
circumstances, CDC's test results were often not returned quickly
enough to be useful to physicians or, in some cases, to
epidemiologists.  For example, state officials have waited up to a
year for CDC to return test results on unusual organisms, making it
difficult--if not impossible--to recognize any subsequent encounters
with these organisms.  Some of these officials suggested that
competing priorities at CDC often prevented the timely return of test
results in the absence of immediate need. 

Training is another CDC service that state officials believe is
important.  As figure 2 shows, the percentage of respondents
indicating that training greatly or significantly improved their
ability to conduct surveillance of emerging infections was even
higher than for testing and consultation.  Participant evaluations of
recent courses offered in collaboration with APHL were generally
consistent with our survey results.  These evaluations indicated that
the courses provided information the participants needed on the most
current technologies available.  However, about 11 percent of our
survey respondents did not believe that the training they received
appreciably improved their surveillance ability.\23

Although state officials generally valued the training CDC provides,
they also said more training is needed, especially hands-on,
skill-based training in new laboratory techniques.  Laboratory
officials in particular said that the use of distance learning
through audio- or video-conferences--as opposed to hands-on workshops
in CDC laboratories--diminished opportunities to develop close
collaboration between state and CDC laboratory staff.  According to
CDC officials, the use of distance learning became desirable when
downsizing of staff in state public health laboratories and the costs
of sending staff to Atlanta led to declining attendance at courses at
CDC headquarters.  State officials also cited a need for training and
technical assistance in information-sharing systems. 


--------------------
\23 The officials who considered the benefits from training moderate
consisted of 29 percent of responding laboratory directors and 19
percent of epidemiologists.  The 11 percent who felt the benefits
were minimal represented 15 percent of responding epidemiologists and
6 percent of laboratory directors. 


         MOST RESPONDENTS SEE
         SUBSTANTIAL VALUE IN
         GRANT ASSISTANCE PROGRAMS
-------------------------------------------------------- Letter :5.2.2

Most state officials responding to our survey reported that funding
through CDC's disease-specific grants and epidemiology and laboratory
capacity grants had made great or significant improvements in their
ability to conduct surveillance for emerging infectious diseases (see
fig.  3).\24 Over 70 percent of responding laboratory directors and
80 percent of responding epidemiologists--comprising more than
three-quarters of all survey respondents--said disease-specific
funding had greatly or significantly enhanced their state's capacity
to conduct infectious diseases surveillance.  With one exception,
epidemiology, laboratory, and combined capacity grants were similarly
valued, with at least 68 percent of recipients saying the enhancement
was great or significant.  Laboratory directors reported benefitting
more from grants specifically directed to laboratory or combined
laboratory and epidemiology capacity than from grants specifically
designed to enhance epidemiology capacity. 

   Figure 3:  Views of State
   Laboratory Directors and
   Epidemiologists on Extent CDC's
   Funding Assistance Improved
   Efforts to Use Laboratory Data
   in Emerging Infectious Diseases
   Surveillance

   (See figure in printed
   edition.)

Note:  Sixty-five officials provided views on disease-specific
grants, 28 provided views on epidemiology capacity grants, 34 on
laboratory capacity, and 33 on combined capacity. 

Officials cited several examples in which CDC assistance was
instrumental in helping states improve their surveillance and
laboratory testing efforts for high-priority conditions, such as
antibiotic-resistant diseases. 

  -- After state laboratories began receiving funds from CDC's
     tuberculosis grant program, they markedly improved their ability
     to rapidly identify the disease and indicate which, if any,
     antibiotics could be used effectively in treatment.  State
     laboratory officials attributed this improvement to the funding
     and training they received from CDC. 

  -- In addition to supporting such core activities as active
     surveillance of antibiotic-resistant conditions, four states use
     EIP funds to conduct active surveillance of unexplained deaths
     and severe illnesses in previously healthy people under age
     50--a potentially critical source of information to detect new
     or newly emerging diseases.  This project will also provide
     information on known infectious diseases that health care
     professionals are not recognizing in their patients.  The
     epidemiologist in one of these states said that although
     reporting of such cases had been required for a long time,
     efforts to improve the completeness of the reporting and analyze
     the data began only after the state received CDC funds. 

Our survey provided one other possible indication of the effect of
CDC's assistance on state surveillance and testing for
antibiotic-resistant conditions.  In comparison to its funding for
tuberculosis, which goes to programs in all states and selected
localities, CDC funds active surveillance and testing for
penicillin-resistant S.  pneumoniae in only eight states.  This
pattern of funding parallels the pattern of testing reported by our
survey respondents.  Of the 54 states that reported conducting
surveillance for tuberculosis, 49 have laboratories that test for
antibiotic-resistance.  In contrast, of the 37 states that reported
conducting surveillance for penicillin-resistant S.  pneumoniae, only
about half have laboratories that provide testing support.  Moreover,
while all but one of the states require health care providers to
submit tuberculosis reports to public health officials, fewer than
half require reporting of penicillin-resistant S.  pneumoniae.\25


--------------------
\24 More states received disease-specific grant funding than
epidemiology, laboratory, or combined capacity building grants. 

\25 In the 55 states where epidemiologists responded to our survey
regarding their surveillance programs for the conditions we asked
about, only 54 laboratory directors provided complete information on
their testing to support state surveillance. 


         REGIONAL LABORATORY
         NETWORKS SEEN AS
         VALUABLE, BUT FEWER THAN
         HALF OF STATES USE THEM
-------------------------------------------------------- Letter :5.2.3

Although CDC-sponsored regional laboratory networks are intended to
expand states' access to advanced testing services, our survey
responses indicate that only about half of the states have used these
laboratories during the past 3 years.  Among those state officials
who did use the networks, views on their usefulness are generally
favorable, although networks were not valued as highly as other types
of assistance (see fig.  4).  Of the 19 laboratory directors who used
the services of regional laboratories, 10 reported great improvement
in their surveillance capacity as a result, 6 reported moderate
improvement, and the remaining 3 said improvement was minimal.  Of
the 21 epidemiologists who used regional laboratory services, 11
reported the services made great improvement, 5 said the improvement
was moderate, and 5 said the improvement was slight. 

   Figure 4:  Views of State
   Laboratory Directors and
   Epidemiologists on Extent CDC's
   Regional Laboratories Improved
   Emerging Infectious Diseases
   Surveillance Ability

   (See figure in printed
   edition.)

Note:  Officials are from 29 states. 


         MIXED VIEWS ON STAFFING
         ASSISTANCE
-------------------------------------------------------- Letter :5.2.4

Almost two-thirds of the 33 epidemiologists and about half of the 13
laboratory directors who had hosted CDC field placements reported
that their staff had greatly or significantly improved their
program's capacity to conduct surveillance.  State officials we spoke
with generally highly praised field placement programs because
participants--who might continue their careers in federal or state
government--gained hands-on experience working in state programs.  An
epidemiologist commented that these placements, which spanned most of
the past 20 years, had been invaluable as they provided staff to
supplement his state's surveillance program.  One state official,
however, said that the benefits of such placements are limited
because it takes almost 2 years of training for new staff to
effectively assist in state programs. 


         INFORMATION-SHARING
         SYSTEMS SEEN AS AREA
         NEEDING CONSIDERABLE
         IMPROVEMENT
-------------------------------------------------------- Letter :5.2.5

According to officials who spoke with us, CDC's information-sharing
systems have limited flexibility for adapting to state program
needs--one reason many states have developed their own information
management systems to capture more or different data, they said. 
State and federal officials told us that NETSS and PHLIS often cannot
share data for reporting or analysis with each other or with state-
or other CDC-developed systems.  CDC officials responsible for these
programs said that the most recent versions can share data more
readily with other systems but that the lack of training in how to
use the programs and high staff turnover at state agencies may limit
the number of state staff and officials able to use the full range of
program capabilities. 

NETSS supports the collection and management of information such as
patient demographics and residence, the suspected or confirmed
diagnosis, and the date of disease onset.  PHLIS contains more
definitive information on the pathogen provided by the laboratory
test.  Both programs also offer optional disease-specific reporting
modules states may use to gather additional data.  When
epidemiologists cannot electronically merge data from different
sources, they must manually match the records to analyze disease
trends and determine the relevant risk factors needed for effective
prevention and control efforts.  Sharing data between systems also
identifies multiple records on the same case and can help
epidemiologists take steps to improve reporting. 

Epidemiologists responding to our survey rated NETSS more highly for
flexibility and overall helpfulness than laboratory directors rated
PHLIS.  About half (48 percent) of responding epidemiologists said
NETSS was highly flexible for meeting their needs while only
one-quarter (27 percent) of laboratory directors said the same for
PHLIS.\26

Fifty-eight percent of epidemiologists said NETSS greatly helped them
conduct surveillance, while 22 percent said it was moderately helpful
and the remaining 20 percent said it was minimally helpful.  In
contrast, 76 percent of laboratory directors said PHLIS was of little
help, 13 percent said it was very helpful, and 11 percent said it was
moderately helpful. 

Many epidemiologists and laboratory directors thought the system they
use does not share data well with other systems.  About two-thirds of
the laboratory directors who use PHLIS and one-quarter of the
epidemiologists who use NETSS said the systems have little to no
ability to share data.  Many officials we spoke with complained about
a substantial drain on scarce staff time to enter and reconcile data
into multiple systems, such as their own system plus one or more
CDC-developed systems.  One large local health department has one
person working full time to enter and reconcile data for a single
disease. 

As some of CDC's disease-specific electronic reporting and
information management systems become outdated and need to be
replaced, CDC has responded to state and local requests for greater
integration of reporting systems and for flexibility in the use of
grant funds to build information systems.  In late 1995, CDC
established the Health Information and Surveillance System (HISS)
Board to formulate and enact policy for integrating public health
information and surveillance systems.  Subcommittees of the HISS
Board bring together federal and state public health officials to
focus on issues such as data standards and coding schemes,
legislation for data security, assessing hardware and software used
by states, and identifying gaps in CDC databases.\27 As of August
1998, the HISS Board or its subcommittees had identified barriers to
implementing effective laboratory reporting standards and some
solutions, established mechanisms to assess information needs and
gaps in state and local data systems, and begun to assess ways to
integrate NETSS and PHLIS. 

CDC provides some training and technical assistance related to NETSS
and PHLIS, although state officials we interviewed said such training
and assistance are in short supply.  Responses to our survey suggest
that CDC's training for these two systems was less widely used and
less highly valued than its technical assistance.  Nearly all
respondents used CDC's technical assistance for these two programs,
while two-thirds of laboratory directors and 82 percent of
epidemiologists used the training.  Almost half of the
epidemiologists and 40 percent of the laboratory directors found the
technical assistance highly valuable, but less than 30 percent of
either group found the training highly valuable.  Staff at two local
health departments told us that no training was offered to them by
state or CDC staff and the wait for technical assistance could last a
month or more.  State and local officials appreciated the help CDC
offered but said CDC had few staff or other resources devoted to
helping them use these reporting systems. 

CDC and the states have made progress in developing more efficient
information-sharing systems through one of CDC's grant programs.  The
Information Network for Public Health Officials (INPHO) is designed
to foster communication between public and private partners, make
information more accessible, and allow for rapid and secure exchange
of data.  By 1997, 14 states had begun INPHO projects.  Some had
combined these funds with other CDC grant moneys to build statewide
networks linking state and local health departments and, in some
cases, private laboratories.  In New York, state officials developed
a network that will link all local health agencies with the state
health department and over 4,500 health care facilities and
diagnostic laboratories.  The network provides electronic mail
service and access to surveillance data collected by the state.  In
Washington, systems for submitting information electronically reduced
passive reporting time from 35 days to 1 day and gave local
authorities access to health data for analysis.\28

In addition to funding specific projects through INPHO grants, in
April 1998 CDC adopted a policy that allows states to submit
proposals to use disease grant funds to build integrated information
systems.  As of November, no states had submitted proposals, although
several indicated they planned to do so.  This initiative involves no
new funding but allows states to use money from existing grants in
more flexible ways. 

While state officials were supportive of additional CDC efforts in
this area, they also recognized that progress in developing effective
networks could be affected by the actions--or lack of action--of
others in the surveillance network.  For example, officials in some
states said autonomous local health departments may elect not to
adopt or link with state-developed systems, thereby continuing some
level of fragmentation among data systems regardless of efforts
undertaken by CDC or others. 


--------------------
\26 Twenty-four percent of epidemiologists said NETSS was not very
flexible, while 57 percent of laboratory directors said PHLIS had
little flexibility.  The remaining 28 percent of epidemiologists and
16 percent of laboratory directors said the programs were moderately
flexible. 

\27 Integrating data systems also requires agreement on policy
issues, such as access, sharing, and confidentiality of data.  Our
work did not address these issues.  A recent federal mandate
requiring the use of uniform standards when medical records are
shared electronically has begun to intensify efforts to reach these
types of agreements.  Section 262 of the Health Insurance Portability
and Accountability Act of 1996 (P.L.  104-191) provided for
electronic data exchange standardization for certain administrative
and financial transactions, while protecting the security and
confidentiality of transmitted data.  HHS, through its Data Council,
is responsible for establishing data and privacy standards.  CDC is a
member of the Data Council. 

\28 J.  Davies and D.  B.  Jernigan, "Development and Evaluation of
Electronic Laboratory-Based Reporting for Infectious Diseases
Surveillance" (Atlanta, Ga.:  International Conference on Emerging
Infectious Diseases, 1998). 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

Public health officials agree that the importance of infectious
diseases surveillance cannot be overemphasized.  The nation's
surveillance network is considered the first line of defense in
detecting and identifying emerging infectious diseases and providing
essential information for developing and assessing prevention and
control efforts.  Laboratories play an increasingly vital role in
infectious diseases surveillance, as advances in technology
continually enhance the specificity of laboratory data and give
public health officials new techniques for monitoring emerging
infections. 

Public health officials who spoke with us said that the nation's
surveillance system is essentially sound but in need of improvement. 
They point to outbreaks rapidly identified and contained as visible
indications of the system's strength.  Our survey results tend to
support this view:  surveillance of five of the six emerging
infectious diseases we asked about is widespread among states, and
surveillance of four of the six is supported by testing in state
public health laboratories.  Officials also view CDC's support as
essential and are generally very satisfied with both the types and
levels of assistance CDC provides. 

However, our survey also revealed gaps in the infectious diseases
surveillance network.  Just over half of the state public health
laboratories have access to molecular technology that many experts
believe all states could use, and few states require the routine
submission of specimens to their state laboratories for testing--a
step urged by CDC.  In addition, many state epidemiologists believe
their surveillance programs do not sufficiently study all infectious
diseases they consider important, including antibiotic-resistant
conditions and hepatitis C. 

Both laboratory directors and epidemiologists expressed concerns
about the staffing and technology resources they have for
surveillance and information sharing.  They were particularly
frustrated by the lack of integrated information systems within CDC
and the lack of integrated systems linking them with other public and
private surveillance partners.  CDC's continued commitment to
integrating its own data systems and to helping states and localities
build integrated electronic data and communication systems could give
state and local public health agencies vital assistance in carrying
out their infectious diseases surveillance and reporting
responsibilities. 

The lack of a consensus definition of what constitutes an adequate
infectious diseases surveillance system may contribute to some of the
shortcomings in the surveillance network.  For example, state public
health officials assert that they lack sufficient trained
epidemiologic and laboratory staff to adequately study infectious
diseases, as well as sufficient resources to take full advantage of
advances in laboratory and information-sharing technology.  Without
agreement on the basic surveillance capabilities state and local
health departments should have, however, it is difficult for
policymakers to assess the adequacy of existing resources or to
identify what new resources are needed to carry out state and local
surveillance responsibilities.  Moreover, public health officials
make decisions about how to spend federal dollars to enhance state
surveillance activities without such criteria to evaluate where
investments are needed most. 


   RECOMMENDATION TO THE DIRECTOR
   OF CDC
------------------------------------------------------------ Letter :7

To improve the nation's public health surveillance of infectious
diseases and help ensure adequate public protection, we recommend
that the Director of CDC lead an effort to help federal, state, and
local public health officials create consensus on the core capacities
needed at each level of government.  The consensus should address
such matters as the number and qualifications of laboratory and
epidemiologic staff, laboratory and information technology, and CDC's
support of the nation's infectious diseases surveillance system. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

CDC officials reviewed a draft of this report.  They generally
concurred with our findings and recommendation and provided technical
or clarifying comments, which we incorporated as appropriate. 
Specifically, CDC agreed that a clearer definition of the needed core
epidemiologic and laboratory capacities at the federal, state, and
local levels would be useful and that integrated surveillance systems
are important to comprehensive prevention programs.  CDC noted that
it is working with other HHS agencies to address these critical
areas. 

We also provided the draft report to APHL and CSTE.  APHL officials
said the report was comprehensive and articulated the gaps in the
current diseases surveillance system well.  They also provided
technical comments, which we incorporated as appropriate.  CSTE
officials did not provide comments. 

As agreed with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days
from the date of this letter.  At that time, we will send copies to
the Secretary of HHS, the Director of CDC, the directors of the state
epidemiology programs and public health laboratories included in our
survey, and other interested parties.  We will make copies available
to others upon request. 

If you or your staff have any questions, please contact me or Helene
Toiv, Assistant Director, at (202) 512-7119.  Other major
contributors are included in appendix V. 

Sincerely yours,

Bernice Steinhardt
Director
Health Services Quality and
 Public Health Issues


OBJECTIVES, SCOPE, AND METHODOLOGY
=========================================================== Appendix I

The Chairman of the Subcommittee on Public Health of the Senate
Committee on Health, Education, Labor, and Pensions asked us to study
the nation's public health surveillance of emerging infectious
diseases, focusing on the contribution of laboratories.  This report
discusses (1) the extent to which states conduct public health
surveillance and laboratory testing of selected emerging infectious
diseases, (2) the problems state public health officials face in
gathering and using laboratory-related data in the surveillance of
emerging infectious diseases, and (3) the assistance CDC provides to
states for laboratory-related surveillance and the extent to which
state officials consider it valuable. 


      SCOPE OF OUR STUDY
------------------------------------------------------- Appendix I:0.1

Although laboratories are only one part of the surveillance network,
they merit attention because newly developed laboratory technology is
an increasingly important means to more quickly identify pathogens
and the source of outbreaks.  We could describe laboratories'
contributions in more detail only by focusing on a small sample of
diseases because the specific contribution of laboratory testing to
surveillance varies with each disease.  Due to the lack of a
consensus definition of the types of public health laboratory testing
that should occur and the lack of explicit, widely accepted standards
to assess epidemiologic capacity, we were not able to assess the
overall adequacy of the nation's emerging infectious diseases
surveillance efforts. 

We selected--with the assistance of officials from CDC, APHL, CSTE,
and the American Society for Microbiology--a sample of six bacterial,
viral, and parasitic pathogens that can be identified using
laboratory tests and pose nationwide health threats (see table I.1). 
Our sample includes diseases transmitted by food and water as well as
ones that had previously been controlled by the use of antibiotics
and vaccines.  These diseases affected up to 1.5 million people in
the United States in 1996 and caused an unknown number of deaths. 



                                    Table I.1
                     
                     Emerging Infectious Diseases Covered in
                                    Our Review

Disease or pathogen        Public health threat
-------------------------  -----------------------------------------------------
Tuberculosis               The appearance of strains resistant to one or more
                           commonly used antibiotics threatens U.S. efforts to
                           control the spread of tuberculosis.

Shiga-like toxin-          This deadly--often foodborne--group of E. coli
producing E. coli,      first appeared in 1982. No effective treatment exists
including E. coli       and infection can result in death or long-term
O157:H7                    disability.

Pertussis (whooping        Pertussis is one of the nation's most commonly
cough)                     reported vaccine-preventable childhood diseases.
                           Incidence is increasing despite high rates of
                           immunization.

Cryptosporidium         This parasite is frequently found in the nation's
parvum                  surface and treated water supplies and the risks of
(Cryptosporidiosis)        low-level exposure from its presence are unknown. The
                           disease it causes has no effective treatment.

Hepatitis C virus          Identified only in 1988, hepatitis C is a leading
                           cause of chronic liver disease and is the nation's
                           most common bloodborne infection. Chronic liver
                           disease related to hepatitis C is also the most
                           frequent indication for liver transplantation.

Penicillin-resistant       S. pneumoniae, a leading cause of death and
Streptococcus           illness, is rapidly becoming resistant to penicillin,
pneumoniae                 with resistance rates as high as 30 percent of cases
                           in some areas.
--------------------------------------------------------------------------------
These six emerging infectious diseases or pathogens are described in
more detail in appendix IV. 


      SURVEY DEVELOPMENT AND
      DISTRIBUTION
------------------------------------------------------- Appendix I:0.2

To gather nationwide data on state public health surveillance efforts
for the sample of six emerging infections, we surveyed the directors
of all state public health laboratories and infectious diseases
epidemiology programs that report disease-related information
directly to CDC.  These include programs in each of the 50 states,
the District of Columbia, New York City, and 5 U.S.  territories
(American Samoa, the Commonwealth of the Northern Mariana Islands,
Guam, Puerto Rico, and the Virgin Islands). 

To develop questions used in our surveys, we reviewed documentation
on surveillance and emerging infectious diseases prepared by CDC,
professional organizations representing state public health
laboratorians and epidemiologists, professional laboratorians, and
public health experts.  We also spoke with officials and
representatives from each of these groups.  We worked with officials
from professional organizations of public and private laboratories
and CDC to judgmentally select a sample of six emerging infections
with nationwide significance and to identify appropriate laboratory
tests used to generate data for state public health surveillance
efforts. 

We pretested our surveys in person with both laboratory directors and
epidemiologists in each of four states and asked knowledgeable people
at CDC and in the laboratory and public health fields to review the
instruments.  We refined the questionnaire in response to their
comments to help ensure that potential respondents could provide the
information requested and that our questions were fair, relevant,
answerable with readily available information, and relatively free of
design flaws that could introduce bias or error into our study
results. 

We mailed 57 questionnaires to laboratory directors in April 1998 and
57 questionnaires to epidemiologists in May 1998.  We sent at least
one follow-up mailing and conducted telephone follow-ups to
nonrespondents.  We ended data collection in July 1998.  At that
time, we had received responses from all 57 laboratory directors and
from 55 epidemiologists, for response rates of 100 percent and 97
percent, respectively. 


      SURVEY ANALYSIS
------------------------------------------------------- Appendix I:0.3

In preparing for our analysis, we reviewed and edited the completed
questionnaires and checked the data for consistency.  We tested the
validity of the respondents' answers and comments by comparing them
with data we gathered through interviews with public health experts
and other public health officials in a total of 30 states and with
documentation obtained at CDC and in case study states. 

We combined responses from epidemiologists and laboratory directors,
by state, to analyze for each of our six specific diseases the extent
to which state public health laboratories supported state
surveillance efforts and the views of epidemiologists and laboratory
directors on the adequacy of testing equipment, staff, and the
resulting surveillance information.  To analyze the extent to which
state public health laboratories supported state surveillance
efforts, we selected only those states that met the following
conditions:  for each disease, (1) the state public health laboratory
director indicated the laboratory performed tests that generated
results used in state surveillance and (2) state epidemiologists
indicated that the state conducted surveillance.  Using these
criteria, we analyzed responses from 54 states. 


      CASE STUDY WORK
------------------------------------------------------- Appendix I:0.4

We also conducted on-site work at CDC and in three states--New York,
Kentucky, and Oregon.  These three states were selected as a
nonrandom judgmental sample representing diverse geographic areas and
public health surveillance programs.  In the three states, we
interviewed state and local public health officials as well as other
interested groups, including representatives from hospitals, large
private clinical laboratories, managed care organizations, and
medical associations.  At CDC, we interviewed officials responsible
for infectious diseases surveillance and laboratories, information
systems development, and support services for states.  We interviewed
officials and obtained documentation to determine how these various
programs were organized and how they interacted with other public
health and private parties to obtain, analyze, and share
disease-related data for surveillance.  In addition, we reviewed the
general literature on public health surveillance and emerging
infectious diseases and interviewed officials from organizations
representing state public health laboratory directors, state
epidemiologists, state and local public health officials, laboratory
professionals, and public health experts. 

Our work was conducted from December 1997 through December 1998 in
accordance with generally accepted government auditing standards. 




(See figure in printed edition.)Appendix II
LABORATORY DIRECTORS' SURVEY
RESULTS
=========================================================== Appendix I



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)




(See figure in printed edition.)Appendix III
EPIDEMIOLOGISTS' SURVEY RESULTS
=========================================================== Appendix I



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


SIX EMERGING INFECTIOUS DISEASES
========================================================== Appendix IV

Given the multitude of infectious diseases and varying state
surveillance programs, we consulted experts to select a sample of
emerging disease threats of nationwide significance.  These six
conditions are described in greater detail below. 


   SHIGA-LIKE TOXIN-PRODUCING E. 
   COLI, INCLUDING E.  COLI
   O157:H7
-------------------------------------------------------- Appendix IV:1


      THE PATHOGEN AND DISEASE
------------------------------------------------------ Appendix IV:1.1

E.  coli are normal bacterial inhabitants of the intestines of most
animals, including humans, where they suppress the growth of harmful
bacteria and synthesize vitamins.  For reasons not completely
understood, a minority of strains cause illness in humans. 
Shiga-like toxin-producing E.  coli are one of five recognized
classes of E.  coli that cause gastroenteritis in humans.  The group
derives its name from producing potent toxins, closely related to
those produced by Shigella dysenteriae, which cause severe damage to
the lining of the intestine.  E.  coli O157:H7, first identified as a
human pathogen in 1982, causes severe abdominal cramping and diarrhea
that can become heavily bloody.  Although people usually get well
without treatment, the illness can be fatal. 

E.  coli O157:H7 is easily killed by heat used in pasteurization and
cooking.  However, it can live in acid environments.  The amount of
bacteria needed to cause illness is thought to be low. 


      COMPLICATIONS
------------------------------------------------------ Appendix IV:1.2

Three to 5 percent of victims develop hemolytic uremic syndrome
(HUS), which is characterized by kidney failure and anemia.  Some
elderly victims develop thrombotic thromobocytopenic purpura (TTP),
consisting of HUS plus fever and neurologic symptoms.  Approximately
1 percent of HUS victims die, though many more develop long-term
complications.  Death rates from TTP can be as high as 50 percent. 


      TRANSMISSION
------------------------------------------------------ Appendix IV:1.3

The disease is often associated with consumption of undercooked
ground beef, but sources of contamination are diverse.  Recent
outbreaks of E.  coli O157:H7 have been linked to consumption of
contaminated apple juice and cider, raw vegetables such as lettuce,
raw milk, and processed foods such as salami.  Illness can also be
caused by ingesting contaminated water at recreational sites such as
swimming pools or spread from child to child in day care settings. 


      COSTS AND PREVALENCE
------------------------------------------------------ Appendix IV:1.4

For E.  coli O157:H7, the estimated annual cost in the United States
from the acute and long-term effects of illness and from lost
productivity is $302 to $726 million, most of which is due to lost
productivity.  The number of reported cases fluctuates seasonally,
peaking in June though September.  Northern states report more cases
than southern states.  In the Pacific Northwest, E.  coli O157:H7 may
be second only to Salmonella as a cause of bacterial diarrhea.  The
true prevalence is unknown and the disease has only recently been
added to the list of nationally notifiable diseases.  CDC received
reports of over 2,741 cases from 47 states in 1996. 

Despite the high visibility of E.  coli O157:H7 due to recent
outbreaks, clinicians often do not consider it when diagnosing
patients or collect appropriate specimens.  Although laboratory
testing to detect E.  coli O157:H7 is relatively straightforward and
inexpensive, a recent study showed that at the end of 1994 only about
half of the clinical laboratories in the United States were screening
stool samples for it. 


   TUBERCULOSIS
-------------------------------------------------------- Appendix IV:2


      THE PATHOGEN AND DISEASE
------------------------------------------------------ Appendix IV:2.1

Tuberculosis, caused by Mycobacterium tuberculosis, was the leading
cause of death from infectious diseases in the United States at the
turn of the century; it remained the second leading cause of death
until the development of antibiotics in the 1950s.  Worldwide, about
one-third of all people are infected.  Tuberculosis kills over 2.9
million people a year--making it a leading cause of death. 
Tuberculosis of the lungs destroys lung tissue and, if left
untreated, half of victims die within 2 years.  The risk of
contracting the disease is highest in the first year after infection
and then drops sharply, although reactivation can occur years later. 
Only about 10 percent of healthy people infected with the pathogen
develop clinical disease.  Tuberculosis is difficult to treat,
requiring a 6-month regimen of multiple antibiotics to effect a cure
and prevent the emergence of antibiotic-resistant strains.  When
health care is adequate and compliance with treatment is maintained,
cure rates should exceed 90 percent, even in those whose immune
systems have been compromised by HIV/AIDS. 


      COMPLICATIONS
------------------------------------------------------ Appendix IV:2.2

The emergence of strains resistant to one or more antibiotics puts
not only tuberculosis patients at risk, but also health care workers,
social workers, and any other people in frequent contact with them. 
For cases of multidrug-resistant tuberculosis, fatality rates can
exceed 80 percent for immuno-compromised and 50 percent for
previously healthy individuals.  Multidrug-resistant cases are
extraordinarily difficult to treat, and most patients do not respond
to therapy. 


      TRANSMISSION
------------------------------------------------------ Appendix IV:2.3

Tuberculosis is spread primarily by the respiratory route from
patients with active disease.  Shouting, sneezing, and coughing can
easily spread the pathogens in the environment.  The risk of
transmission varies with the length of exposure, degree of crowding
and ventilation, virulence of the strain, and health of the person
exposed. 


      COSTS AND PREVALENCE
------------------------------------------------------ Appendix IV:2.4

From the 1950s through the early 1980s, the incidence of tuberculosis
declined in the United States, then began to increase in 1988,
reaching a peak in 1992.  The HIV/AIDS epidemic, immigration from
countries with high rates of tuberculosis, and outbreaks in
facilities such as correctional institutions and nursing homes have
contributed to the resurgence.  Treatment costs for an individual
with multidrug-resistant tuberculosis can be as much as $150,000, 10
times the cost of treating a nonresistant case.  In 1996, 54 states
reported 21,337 cases to CDC. 


   PERTUSSIS
-------------------------------------------------------- Appendix IV:3


      THE PATHOGEN AND DISEASE
------------------------------------------------------ Appendix IV:3.1

Pertussis, caused by the bacterium Bordetella pertussis, is
characterized by uncontrollable spells of coughing in which one cough
follows another too quickly to allow a breath in between.  An intake
of breath that produces a high-pitched "whooping" sound follows each
coughing spell, hence the name whooping cough.  The illness lasts
about 2 weeks and responds to antibiotic therapy.  In the early to
mid-1900s, pertussis was a common childhood disease and a leading
cause of death among children in the United States.  Today, pertussis
is one of the nation's most commonly reported childhood
vaccine-preventable diseases. 


      COMPLICATIONS
------------------------------------------------------ Appendix IV:3.2

Complications associated with pertussis may be severe, especially
among infants.  Secondary bacterial pneumonia causes most
pertussis-related deaths.  Other complications include seizures,
encephalopathy, and ear infections.  About 1 percent of affected
infants died in 1993.  The risk of complications is highest among
infants and under-vaccinated preschool aged children.  In 1994, a
strain resistant to the antibiotic preferred for treatment appeared
in the United States. 


      TRANSMISSION
------------------------------------------------------ Appendix IV:3.3

Immunity to pertussis can decrease with age.  Consequently, young
adults and adolescents who contract the disease can be an important
source in transmitting it to unimmunized infants.  Pertussis among
adults and adolescents is often not diagnosed by physicians--despite
the presence of a persistent cough--because they do not expect to see
the disease in this age group.  Pertussis is endemic in the United
States. 


      INCIDENCE
------------------------------------------------------ Appendix IV:3.4

Pertussis incidence is cyclical, with peaks every 3 to 4 years. 
Incidence has decreased from 150 cases per 100,000 population prior
to 1940 to about 1.2 cases per 100,000 by 1991.  In 1996, 7,796 cases
were reported to CDC, an estimated 10 percent of the true number. 
Although the total number of reported cases remains well below the
annual number reported during the pre-vaccine era, the total number
of cases has increased steadily in each peak year since 1977.  The
reasons for the increase in reported cases are unclear but appear
unrelated to decreased vaccination rates or reduced vaccine efficacy. 
Because few pertussis specimens are tested for resistance, the
prevalence of antibiotic-resistant strains is unknown. 


   PENICILLIN-RESISTANT
   STREPTOCOCCUS PNEUMONIAE
-------------------------------------------------------- Appendix IV:4


      THE PATHOGEN AND DISEASE
------------------------------------------------------ Appendix IV:4.1

Worldwide, S.  pneumoniae infections are among the leading causes of
illness and death for young children, individuals with underlying
medical conditions, and elderly people.  S.  pneumoniae is the most
common cause of bacterial pneumonia and is implicated in infections
of the ears, sinuses, lungs, abdominal cavity, bloodstream, and
tissues that envelop the brain and spinal column.  A vaccine that
controls the 23 most common strains has been available since the
1980s, but it is largely underutilized.  In the past, S.  pneumoniae
uniformly responded to treatment with penicillin, allowing physicians
to treat even severely ill patients without testing for antibiotic
resistance.  During the 1990s, however, resistance to penicillin
spread rapidly in the United States, and strains resistant to
multiple antibiotics account for a small, but growing, proportion of
cases. 


      COMPLICATIONS
------------------------------------------------------ Appendix IV:4.2

Case fatality rates--which vary by age, type of infection, and
underlying medical condition--can be as high as 40 percent among some
high-risk patients, despite appropriate antibiotic therapy. 


      TRANSMISSION
------------------------------------------------------ Appendix IV:4.3

Transmission occurs through contact with infected saliva. 


      PREVALENCE
------------------------------------------------------ Appendix IV:4.4

In the United States, S.  pneumoniae causes up to 3,000 cases of
meningitis, 135,000 cases of hospitalized pneumonia, and as many as 7
million ear infections each year.  Resistance to penicillin varies
widely by region and age group but accounts for 30 percent of cases
in some communities.  The prevalence of resistance for most areas of
the United States is unknown, possibly because the condition was not
nationally reportable until 1996.  Limited knowledge of local
patterns of resistance and the lack of a rapid diagnostic test often
result in therapy that uses either unnecessary or overly broad
antibiotics, thereby contributing to the development of resistant
strains. 


   CRYPTOSPORIDIOSIS
-------------------------------------------------------- Appendix IV:5


      THE PATHOGEN AND DISEASE
------------------------------------------------------ Appendix IV:5.1

Cryptosporidiosis, caused by the parasite Cryptosporidium parvum, can
affect human intestinal and, rarely, respiratory tracts.  The disease
has long been known to veterinarians but was first recognized as a
human pathogen in 1976.  The intestinal disease is generally
characterized by severe watery diarrhea and can include abdominal
cramps, nausea, vomiting, and low-grade fever.  Most healthy
individuals recover after 7 to 10 days.  Infection of the respiratory
tract is associated with coughing and a low-grade fever, often
accompanied by severe intestinal distress.  Unlike many bacterial
infections, the infective dose of cryptosporidiosis is thought to be
small, perhaps as few as 10 organisms, each about half the size of a
red blood cell.  An infected person or animal can shed millions of
organisms per milliliter of feces.  Once in the environment, the
organisms can remain infective for many months.  No safe and
effective treatment for cryptosporidiosis has been identified. 


      COMPLICATIONS
------------------------------------------------------ Appendix IV:5.2

Among persons with weakened immune systems, the disease can lead to
dehydration and death. 


      TRANSMISSION
------------------------------------------------------ Appendix IV:5.3

The infectious stage of the parasite is passed in the feces of
infected humans and animals.  Infection can be transmitted from
person to person, from animal to person, through ingesting
contaminated food or water, or through contact with fecally
contaminated environmental surfaces. 


      PREVALENCE
------------------------------------------------------ Appendix IV:5.4

The parasite is common among herd animals and is present in virtually
all the surface--and much of the treated--waters of the United
States.  The parasite, small enough to slip through most water
filters, is resistant to chlorine treatment.  The public health risk
of contracting the disease from tap water is unknown.  Tests on body
fluids indicate as many as 80 percent of the United States population
have had cryptosporidiosis.  Throughout the world, the organism has
been found wherever it was sought.  In 1996, 42 states reported 2,426
cases to CDC. 


   HEPATITIS C VIRUS
-------------------------------------------------------- Appendix IV:6


      THE PATHOGEN AND DISEASE
------------------------------------------------------ Appendix IV:6.1

The virus that causes hepatitis C was discovered in 1988 and is the
major cause of chronic liver disease worldwide.  Since 1990,
molecular-based laboratory tests have allowed detection of specific
antibodies in the blood of infected people.  Prior to 1990, diagnosis
of hepatitis C was made by excluding both hepatitis A and hepatitis
B.  The incubation period for acute hepatitis C averages 6 to 7
weeks.  Typically, adults and children with acute hepatitis C are
either asymptomatic or have a mild clinical illness.  More severe
symptoms of hepatitis C are similar to those of other types of viral
hepatitis and include anorexia, nausea, vomiting, and jaundice.  Most
patients do not achieve a sustained response to treatment. 


      COMPLICATIONS
------------------------------------------------------ Appendix IV:6.2

At least 85 percent of persons infected with hepatitis C develop
persistent infection.  Chronic disease develops in 60 to 70 percent
of infected individuals, and up to 20 percent may develop cirrhosis
over a 30-year period.  Hepatitis C is a leading cause of chronic
liver disease in the United State and a major reason for liver
transplants.  An estimated 8,000 to 10,000 people die annually from
hepatitis C and its related chronic disease. 


      TRANSMISSION
------------------------------------------------------ Appendix IV:6.3

Hepatitis C is most efficiently transmitted through large or repeated
contact through the skin with infected blood.  Intravenous drug use
is the most common risk factor for acquiring hepatitis C.  Currently,
transfusion-associated hepatitis rarely occurs due to donor screening
policies instituted at blood banks and to routine testing of blood
donors for evidence of infection. 


      COSTS AND PREVALENCE
------------------------------------------------------ Appendix IV:6.4

In the United States, the annual number of newly acquired acute
hepatitis C infections has ranged from an estimated 180,000 cases in
1984 to an estimated 28,000 in 1995.  The prevalence of hepatitis C
in the general population is about 1.8 percent, which corresponds to
approximately 3.9 million people with chronic infection.  Hepatitis C
and related chronic diseases cost about $600 million annually (in
1991 dollars). 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

GAO CONTACTS

Helene Toiv, Assistant Director, (202) 512-7162
Cheryl Williams, Evaluator-in-Charge, (503) 235-8451

STAFF ACKNOWLEDGMENTS

In addition to those named above, the following individuals made
important contributions to this report:  Linda Bade, Senior Health
Policy Analyst; Nila Garces-Osorio, Health Policy Analyst; Julian
Klazkin, Attorney; Susan Lawes, Senior Social Science Analyst; and
Stan Stenersen, Reports Analyst. 


*** End of document. ***