Attachment K Technical Guide

Attachment F. Supplemental Surveillance Activity 1: HIV Incidence Surveillance Data
Elements and Procedures
Status: new
Description: These data elements will be an electronic extension of the case report forms
(Attachment C); as such, there is no new form for this activity.

OMB No. 0920-XXXX Exp. Date XX/XX/XXXX

Standard HIV Incidence Surveillance Data Elements

Variable Description

Valid Value

Number of HIV tests in 2 years prior to the first
positive (For people who previously tested positive)

1 - 99
R=refused
D=don't know

Anonymous HIV test at first positive test

0=no
1=yes
7=refused
9=don't know

Antiretroviral medications within the last 6 months

0=no
1=yes
7=refused
9=don't know

ARV meds taken

22=Agenerase
30=Aptivus
24=Combivir
06 =Crixivan
11=Emtriva
03=Epivir
28=Epzicom
25=Fortovase
10=Fuzeon
19=Hepsera
02=Hivid
23=Hydroxyurea
18=Invirase
16=Kaletra
31=Lexiva
07=Norvir
88=Other
09=Rescriptor
26=Retrovir
15=Reyataz
08=Saquinavir

CLIA code for source lab where specimen originated

text

Currently taking antiretroviral medications

0=no
1=yes
7=refused

OMB No. 0920-XXXX Exp. Date XX/XX/XXXX

9=don't know
Date HAART use began

yyyymmdd

Date HAART use ended

yyyymmdd

Date information is extracted

yyyymmdd

Date of first HIV test ever

yyyymmdd

Date of first positive HIV test

yyyymmdd

Date of STARHS test

yyyymmdd

Date of the HIV test that resulted in a case report

yyyymmdd

Date specimen was obtained

yyyymmdd

Has specimen been approved for STARHS?

0=no
1=yes
2=pending

Laboratory ID

33D0654341=NYST
33D0654341=CDCSTARHS
33D0654341=NY
33D0654341=CDCSTAR
21D0649758=MARY01
50D0661430=WASH

Date of last HIV negative test before first positive

yyyymmdd

Date of last HIV negative test before first positive (or
before test that resulted in a case report if never
previously tested positive)

yyyymmdd

Name of site where first tested positive for HIV

text

Name of site where last tested negative for HIV

text

Ever tested for HIV

0=no
1=yes
7=refused
9=don't know

OMB No. 0920-XXXX Exp. Date XX/XX/XXXX

Ever tested negative

0=no
1=yes
7=refused
9=don't know

Number of HIV tests in last 2 years before first positive 1 - 99
(including first positive test)
R=refused
D=don't know
Number of HIV tests in last 2 years before first positive 1 - 99
(People who never had previous positive)
R=refused
D=don't know
Optical density

text

Ever tested positive

0=no
1=yes
7=refused
9=don't know

Reason for first positive HIV test*

text

Reason for testing at first positive - exposure to HIV
within the last 6 months*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - just checking to
make sure you are HIV negative*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - other reason*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - regular tester; time
for routine HIV test*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - required by either
insurance, military, court order, or for some other
required reason*

0=no
1=yes
7=refused
9=don't know

OMB No. 0920-XXXX Exp. Date XX/XX/XXXX

Reason for the test that led to the case report - exposure 0=no
to HIV within the last 6 months*
1=yes
7=refused
9=don't know
Reason for the test that led to the case report- just
checking to make sure you are HIV negative*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report- other
reason*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report- regular
tester; time for routine HIV test*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report - required
by either insurance, military, court order, or for some
other required reason*

0=no
1=yes
7=refused
9=don't know

Reason STARHS not performed

1=QNS
2=specimen never received at public lab
3=broken in transit
4=other

Results received

0=no
1=yes

Specify other reason for testing at fist positive*

text

Specify other reason for the test that led to the case
report*

text

Specimen ID number from source lab

text

STARHS ID

text

STARHS regional lab specimen ID number (same as
STARHS lab imported variable SPECIMEN ID)

text

OMB No. 0920-XXXX Exp. Date XX/XX/XXXX

STARHS test result

01=long term
02=recent
91=QNS
92=not rec’d by STARHS lab
93=broken
94=other

State lab CLIA code

text

State lab specimen ID number

text

State of site where first tested positive for HIV

text

State of site where last tested negative for HIV

text

Test assay

BED=BED
BVLS=BVLS (Vironostika LS)
OTLS=OTLS (Vironostika LS)
OTV=OTV (Vironostika LS)
AVID=AVID

Type of site where first tested positive for HIV

F01=inpatient facility
F01.01=inpatient facility/hospital
F01.04=inpatient facility/long term care
F01.50=inpatient facility/drug treatment
F01.OTH=inpatient facility/other
F01.UNK=inpatient facility unknown
F.OTH=facility/other
F.UNK=facility/unknown

Type of site where participant last tested negative for
HIV

F01=inpatient facility
F01.01=inpatient facility/hospital
F01.04=inpatient facility/long term care
F01.50=inpatient facility/drug treatment
F01.OTH=inpatient facility/other
F01.UNK=inpatient facility unknown
F.OTH=facility/other
F.UNK=facility/unknown

Type of test performed on specimen (LOINC)

5220-9=EIA / Elisa
21009-6=Western Blot
5472-6=CD4
25835-0=Viral Load (NASBA)
5017-9=Viral Load (bDAN)
25836-8=Viral Load (RT-PCR)

OMB No. 0920-XXXX Exp. Date XX/XX/XXXX

What type of specimen was obtained

1=blood finger stick
2=blood venipuncture
3=blood spot
4=oral mucosal transudate
5=urine
8=other
9=unknown

* The reason for testing for first positive test or for the test that led to the case report will no
longer be part of the standard HIV Incidence Surveillance data elements beginning
January 1, 2007.

Public reporting burden of this collection of information is estimated to average 10 minutes per response, including the time for reviewing instructions, searching existing
data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. An agency may not conduct or sponsor, and a
person is not required to respond to a collection of information unless it displays a currently valid OMB control number. Send comments regarding this burden estimate or
any other aspect of this collection of information, including suggestions for reducing this burden to CDC, Project Clearance Officer, 1600 Clifton Road, MS D-74, Atlanta,
GA 30333, ATTN: PRA (0920-XXXX). Do not send completed form to this address.

Page 1

Technical Guidance for HIV/AIDS Surveillance
Policies and Procedures for HIV Incidence Surveillance
DRAFT 10/31/2006
Although historically AIDS surveillance data have been of great value, current data do not represent the
entire population affected by the HIV epidemic, which in the United States consists primarily of HIV type 1
(HIV-1) infections. Unlike AIDS data, HIV data provide a window into the epidemic at an earlier stage of
disease, thereby allowing public health officials to monitor the epidemic more effectively and completely,
allocate resources, and plan and implement programs, particularly prevention programs. In the past,
however, biomedical technology did not discriminate between recent and chronic HIV infection and as a
result the incidence of HIV infection in the United States could not be measured directly. The serologic
testing algorithm for recent HIV seroconversion (STARHS) distinguishes between recent and long-standing
HIV-1 infection on a population level and should allow the estimation of local and national HIV incidence.
HIV Incidence Surveillance (HIS) is the aspect of the national HIV/AIDS surveillance system that uses
STARHS results as well as data on the history of testing and use of medications with antiretroviral (ARV)
properties for each case reported to HIV/AIDS surveillance programs to generate an HIV incidence
estimate. HIS will give a more representative picture of the HIV epidemic, its trends, and its impact on the
public’s health. CDC’s human subjects protection process has determined that the implementation of HIS
programs using STARHS, like other public health surveillance activities, is not research < National Center
for HIV, STD, and TB Prevention’s Non-research Determination for HIV Incidence Surveillance >.
The primary functions of HIS are to:
1. Incorporate STARHS into routine HIV surveillance activities by testing HIV seropositive specimens
obtained from newly diagnosed individuals using an assay approved for this purpose
2. Collect HIV testing and ARV use history information from individuals with newly reported HIV
infections as part of routine HIV surveillance
3. Apply a statistical model(s) to estimate HIV incidence locally and nationally using a combination of
STARHS results, information from surveillance case reports, including testing and ARV use history
information for all newly diagnosed individuals
4. Assist with local HIV prevention program planning and evaluation using incidence data.

The tasks to achieve the functions of HIS include:
1. Elicit collection and reporting of testing and ARV use history information by all providers reporting
HIV cases to the health department
2. Ensure that an aliquot of each confirmed seropositive specimen from a newly diagnosed case is
shipped to the CDC STARHS laboratory in New York
3. Determine the disposition of remnant HIV-positive specimens and coordinate communication
between the Incidence Surveillance Coordinator (ISC) and the public health laboratory or the CDC
STARHS laboratory regarding specimens shipped to and stored at the respective laboratories
4. Enter STARHS and testing and ARV use history information into the designated HIS database
5. Electronically transfer data to CDC
6. Ensure that HIS data handling procedures comply with all security and confidentiality guidelines as
described in the
7. Analyze and disseminate data locally
8. Train staff in the policies and procedures of the HIV/AIDS surveillance system as well as the HIS
System

Page 2
The prerequisites (structural requirements), best practices (process standards), and outcome standards for
HIS are described next.
Structural Requirements
All individuals with a new diagnosis of HIV infection tested confidentially should be reported to the HIV
surveillance system in accordance with the < Technical Guidance for HIV/AIDS Surveillance Programs,
Vol.I: Policies and Procedures >. In areas incorporating HIS into their HIV/AIDS surveillance systems, an
HIV testing history, information about ARV use, and the STARHS result of a remnant of the diagnostic
HIV positive specimen will complete information collection for a case. HIV/AIDS surveillance case report
data, in combination with these data elements will be used to make population-based HIV incidence
estimates.
Policies and Procedures – HIS is a fully integrated component of HIV/AIDS surveillance, therefore
documentation of HIS activities should be incorporated into locally tailored policy and procedures manuals
developed for HIV/AIDS surveillance < Technical Guidance for HIV/AIDS Surveillance Programs, Vol.I:
Policies and Procedures> to establish standardization, maintain continuity of meaning, document changes
over time, and develop training programs. All manuals describing policies and procedures of the local
surveillance program should address the needs of HIV/AIDS surveillance as well as the specific policies
related to HIS. In addition to the information listed in < Technical Guidance for HIV/AIDS Surveillance
Programs, Vol.I: Policies and Procedures> HIS specific policies and procedures should include information
related to:
• Training of testing providers in collection of additional data used for HIV incidence estimation
• Laboratory contacts
• Determining which specimens should be tested using STARHS (and which should be discarded)
• Specimen aliquoting
• Specimen shipping guidance
The BED Assay for STARHS – The assay STARHS currently uses is the BED HIV-1 Capture EIA (2)
manufactured by Calypte Biomedical Corporation®. The principle of the BED HIV-1 Capture EIA is based
on the observation that the ratio of HIV-specific IgG to total IgG increases with time after HIV infection (2).
The BED HIV-1 Capture EIA is applied to the diagnostic HIV-positive specimen and the assay is sensitive
to the length of time since the infection (i.e., antibody level present). The time from when a specimen
would first be reactive on the standard EIA to the time when the serum or plasma, if tested with the BED
HIV-1 Capture EIA, reaches an optical density (OD) level pre-determined to distinguish recent from nonrecent infections is defined as the STARHS mean window period. Although the mean STARHS window
period may vary slightly by HIV subtype, the mean window period for calculating population-based
incidence estimates in the United States is 156 days when using the BED HIV-1 Capture EIA.
The BED HIV-1 Capture EIA for STARHS is performed only on HIV-positive sera (2) and is not approved
as a diagnostic test. Because of the variability in antibody development in individuals, the predictive value
of an individual’s STARHS result is low; the results are only reliable as part of the population-based HIV
incidence estimate. The Food and Drug Administration (FDA) has ruled that the BED HIV-1 Capture EIA
be labeled “For Surveillance use. Not for diagnostic or clinical use.” Under FDA regulations, results of
STARHS performed for purposes of HIS cannot be returned to individuals or their health care providers or
used for clinical management. As with earlier assays used for STARHS, data show that the BED HIV-1
Capture EIA can produce a substantial number of false positive and false negative classifications on the

Page 3
individual level (4). At the population level, the number of false positives is approximately
equal to the number of false negatives thus effectively “canceling” each other out. However, the number of
misclassifications can be large, and each of the misclassified individual results would receive an incorrect
interpretation. STARHS results may also be misclassified due to the use of ARV therapies or late stage of
the disease. Evaluation of the BED HIV-1 Capture EIA has determined that the specimens of persons with
low HIV-1–specific antibodies resulting from ARV therapy or disease progression (i.e., AIDS) could lead to
the incorrect conclusion that these persons were recently infected. When a person has AIDS, this is thought
to be due to a loss of immune response as immune deficiency progresses. When a person is taking ARV
therapy, this result is thought to be due to suppression of the HIV viral load, which in turn reduces antigenic
stimulation and the quantity of circulating HIV-1–specific antibodies. The effect of ARVs taken for post
exposure prophylaxis or for concurrent hepatitis B infection, for example, is not known. Theoretically, it
could take longer to develop a full immune response to HIV infection. CDC data only reliably support
using STARHS for estimating incidence at the population level.
Testing and ARV Use History Data – Information on testing behavior is needed, such as recency of
testing and testing frequency. Additionally, history of ARV use (for example, pre- or post-exposure
prophylaxis or treatment for hepatitis) and immunological status (CD4 cell counts and viral loads) must be
included for all cases reported to the surveillance system.
Information needed for HIV incidence estimation is available as part of a standard case report and nearly all
testing and ARV use history information is gathered as part of a comprehensive HIV counseling session.
However, not all of the required HIS data elements have been collected uniformly, and many have not
previously been recorded. Therefore, a standard set of HIV testing and ARV use history data elements
needed for the HIV incidence estimate has been developed and providers of HIV testing should be trained in the appropriate collection of those data.
Staffing Needs – Implementation of the HIS system requires personnel with specific skills and dedicated
time to integrate HIS into the existing core HIV/AIDS surveillance system effectively. Generally, HIS staff
should have:
• An understanding of HIS and the characteristics of the HIV/AIDS surveillance in their area
• Good communication skills
• Strong leadership skills
• Enthusiasm about disease reporting for public health purposes.
• Dedication to the successful implementation of HIS
• Ability to work closely with CDC, other States, local sites, private providers, and laboratories
The recommended staffing plan including roles and responsibilities follows. In terms of personnel time,
CDC recommends that one full-time equivalent (FTE) be dedicated to the ISC position. Successful
implementation and integration of HIS requires a full time ISC dedicated to implementing and maintaining
the system. Other personnel assigned to HIS may vary depending on the implementation phase, prevalence
of HIV/AIDS, and available resources.
ISC
•
•

Provide overall management of the HIS system
o Serve as lead on the area specific implementation of the HIS guidance
Oversee data collection processes
o Determination of the disposition of HIV-positive specimens reported to the surveillance
system

Page 4

•

•
•
•

o Receipt of STARHS results
o Oversee collection of HIV testing and treatment history information from public HIV testing
sites and private providers
Collaborate with other HIS staff
o HIV incidence epidemiologist
ƒ Development of training materials and courses
ƒ Data collection procedures
o HIV incidence data manager
ƒ Data collection methods
ƒ Data entry and quality assessment
ƒ Data editing and file correction
ƒ Data transport procedures
ƒ Preparation of monthly reports
ƒ Security and confidentiality procedures
o HIV incidence laboratory liaison
ƒ Specimen transfer
ƒ Specimen tracking
Manage any employee or other service contracts related to HIS
Serve as the primary point of contact for CDC on HIS
Participate in CDC site visits, trainings, and workshops

Epidemiologist/Trainer
• Serve as lead on training HIV testing providers and laboratories on HIS, including
development/modification of surveillance area specific training materials
• Coordinate HIS and epidemiology activities with the ISC
• Participate in the development or modification of testing and ARV use history data elements
• Participate in data dissemination activities
o Collaborate with stakeholders to determine data needs and frequency of reporting
o Identify results and surveillance issues for review and dissemination
o Develop a data dissemination plan in collaboration with the ISC and CDC
• Participate in CDC site visits, trainings, and workshops as appropriate
Laboratory Liaison
• Act as the liaison between the public health, private, and community
laboratories and the ISC
• Oversee preparation and shipping of public health laboratory specimens to the CDC STARHS
laboratory
• Monitor quality control procedures outlined for preparing specimens for testing using STARHS
• Monitor security and confidentiality of specimens and STARHS results
• Track specimens identified for testing using STARHS (all laboratories)
• Participate in CDC site visits, trainings, and workshops as appropriate
Data Manager
• Assist ISC with daily management of HIS data
• Conduct data collection from areas
o Serve as subject matter expert on HIV incidence data elements and data management
programs

Page 5

•

•
•

o Receive data transfer from other health department entities and the CDC
STARHS lab and incorporate those data into the HIS database and datasets for transfer to
CDC
o Conduct data quality assessments
Conduct data management
o Modify CDC’s generic data management programs for use at the area level
o Develop and implement edit checks and conduct data cleaning
o Collaborate with the ISC, epidemiologist, and other area surveillance and prevention staff, as
needed, on data cleaning, data entry, and data set preparation
o Prepare datasets for local analysis
o Collaborate with CDC on dataset preparation for national incidence estimates
o Prepare HIV incidence data reports for local use in collaboration with the ISC,
epidemiologist, and CDC
Maintain security and confidentiality of HIV incidence data
Participate in CDC site visits, trainings, and workshops as appropriate

Process Standards
HIS involves the following:
• Obtaining testing and ARV use history data from providers
• Identifying laboratories that perform HIV related tests and obtaining remnant specimens for testing
using STARHS
• Determining specimen disposition as it relates to testing using STARHS
• Establishing a schedule for contact between the ISC and the Public Health Laboratory and the CDC
STARHS Laboratory to communicate regarding shipping, testing, and discarding remnant specimens
that are housed at the laboratory
• Entering data into the HIS database
• Electronically transferring data to CDC
• Ensuring that data handling procedures comply with Security and Confidentiality Guidelines
,
• Analyzing and disseminating data locally
• Training of staff in HIV/AIDS surveillance and HIS methods
Obtaining Testing and ARV Use History Data – The primary purpose of gathering HIV testing and ARV
use history is to calculate a statistical weight that will allow inference to the general population. The weight reflects the
probability that an individual will be tested during the STARHS window period and depends on the
following:
• HIV testing frequency
• Reason for the first positive test
• Information related to the most recent negative test
o Date
o Place
• Information related to the first positive test
o Date
o Place

•

Page 6
ARV use history
o Type
o Date started
o Date stopped

Testing and ARV use history data are collected by providers and surveillance staff using the area
standard reporting procedures or other procedures meeting the routine HIV/AIDS surveillance security
and confidentiality guidelines. These data:
• Should be included for all adult/adolescent (> 13 years at diagnosis) HIV/AIDS cases newly reported
to the HIV/AIDS surveillance system by all providers of HIV testing
• May be collected through client/patient interview or chart abstraction given that the provider has
been trained to obtain the data
• Can be collected when an individual presents for an HIV test or returns for the results
o Takes advantage of the individual’s ability to recall information that is more proximal to the
event
• Should be reported and recorded based on the patient’s self-report within three months of the HIV
diagnosis
o Longer intervals may increase the risk of recall bias, yet this consideration should not prevent
efforts to obtain the information even after three months if necessary
CDC has assisted in the development of materials for use in training providers to collect HIV testing and
ARV use history data. These materials are available at all sites, or upon request to CDC.
Obtaining Remnant HIV-Positive Specimens for Testing Using STARHS – To be most useful, testing
using STARHS should be performed on the HIV-positive diagnostic serum or plasma specimen. For the
purposes of this guidance, the HIV-positive diagnostic specimen is the HIV-positive specimen from the
diagnostic test that resulted or should have resulted in the case being reported to the HIV/AIDS surveillance
system. Remnant specimens for all confirmed HIV-positive diagnostic specimens should be tested using
STARHS.
• Laboratories performing routine diagnostic confirmatory HIV testing by Western Blot, indirect
fluorescent antibody (IFA) tests, immunological status tests such as CD4, or viral load counts should
report to the state/local health department surveillance program per existing requirements.
• In each surveillance area, all laboratories should be identified from a review of local HIV
surveillance data and laboratory licensing records and must be approached to request that remnants
of all diagnostic specimens be made available for testing using STARHS.
o Surveillance areas should maintain a directory of laboratory contacts at all reporting
laboratories to facilitate communication in the event that reporting or shipping of specimens
is disrupted or that changes in policy or procedures need to be communicated.
o Originating laboratories are those to which a specimen is first sent for testing
o Reference laboratories are those to which a specimen is sent for confirmatory testing when
the originating laboratory does not do confirmatory testing
• All remnant specimens from HIV-diagnostic Western Blot or IFA tests must be shipped to the CDC
STARHS laboratory in New York for testing using STARHS.
o A minimum of 0.5 ml HIV-positive serum or plasma specimen is necessary for testing using
STARHS
o Private or community laboratories performing HIV diagnostic testing should choose one of
two options for shipping the remnant HIV-positive serum or plasma specimen to the CDC
STARHS laboratory.

Page 7
ƒ
ƒ

Ship the specimen directly to the CDC STARHS laboratory.
Ship the specimen to the state public health laboratory affiliated with the health
department that receives the new HIV case report for processing or
• State public health laboratories can then batch and ship all specimens identified
for testing using STARHS to the CDC STARHS laboratory.
o State public health laboratories conducting HIV diagnostic testing should ship their own
HIV-positive specimens identified for testing using STARHS directly to the CDC STARHS
laboratory.
Specimen availability for testing using STARHS depends on the testing needs for the specimen. Uses of the
remnant of specimens should follow the CDC HIS recommended hierarchy (described below) for specimen
aliquoting to ensure adequate specimen volume for multiple diagnostic tests. When multiple tests must be
performed on a collected serum or plasma specimen, the aliquots must be made available with the following
hierarchy in mind:
1. HIV diagnostic testing
2. Testing with STARHS
3. HIV drug resistance genotyping (known as Variant, Atypical, and Resistant HIV Surveillance
[VARHS])
Aliquots made following this hierarchy will ensure that adequate specimen volume is available according to
the priority for data determined by CDC.
Determining the Disposition of Specimens and Communicating with the Public Health and CDC
STARHS Laboratories – A specimen will be held at the state public health laboratory or the CDC
STARHS laboratory (i.e., specimens shipped directly from private or commercial laboratories) until the area
ISC, using routine HIV/AIDS surveillance reporting procedures (i.e., HARS/eHARS), determines whether
the specimen represents the person’s first reported positive HIV test result in the HIS area. A specimen
should be tested using STARHS if:
1. the specimen represents the diagnostic specimen (the HIV-positive specimen that led, or should have
led a case to be reported to HARS/eHARS) or
2. the diagnostic specimen is unavailable and the specimen was drawn within 3 months of the
diagnostic specimen and
3. the specimen was drawn for an HIV-related test (viral load, qualitative PCR, CD4 level).
A specimen should not be tested using STARHS if:
1. the specimen is not the diagnostic specimen that led, or should have led the individual to be reported
to HARS/eHARS and
2. the individual had a previous specimen that was tested using STARHS or
3. the individual did not have a previous specimen tested using STARHS but the specimen was drawn
more than three months after the diagnostic specimen.
Because a remnant sample of every Western Blot positive blood specimen will be shipped by originating or
reference laboratories to either the state or local Public Health Laboratory or to the CDC STARHS
laboratory, the HIS program must inform the appropriate laboratory of the disposition of the specimen.
• Specimens for cases not previously reported to HARS/eHARS (or those drawn within three months
of a diagnostic specimen that is unavailable) will constitute the test list. A test list should:
o Be compiled for the state or local Public Health Laboratory and for the CDC STARHS
Laboratory

Page 8

•

o Include those specimens located at the individual laboratory and that should be
tested using STARHS.
o Be cumulative
Specimens that are neither diagnostic specimens nor drawn within three months of a diagnostic
specimen that is unavailable will constitute the toss list A toss list should:
o Be compiled for the state or local Public Health Laboratory and for the CDC STARHS
Laboratory
o Include those specimens held at the laboratory that should not be tested using STARHS.
o Be cumulative

Specimens will be handled, packaged, and shipped according to the CDC STARHS laboratory shipping
protocol . Specimens shipped as diagnostic specimens
and using dry ice for packing must follow the procedures for packing and shipping specimens using dry ice
.
The ISC will regularly, at an interval to be determined locally (e.g., monthly), inform area laboratory
designees (e.g., at the public health laboratory) and the CDC STARHS laboratory of all stored specimens, or
remnants specimens, with a positive HIV diagnostic test to be tested using STARHS (the test list) and those
to be discarded (the toss list).
• For those specimens on the test list:
o An aliquot of blood to be used for STARHS is drawn from the specimen
ƒ At the local public health laboratory prior to shipping
ƒ At the CDC STARHS laboratory if the specimen was shipped directly to the CDC STARHS
laboratory
o The aliquot is relabeled with a unique STARHS identification number (SID)
o The SID is paired with the corresponding specimen number and is sent to the ISC with no other
identifying data
o After STARHS the results are returned to the ISC with results identified by SID only
• Specimens on the toss list should be discarded according to routine laboratory protocols for HIVpositive serum or plasma specimens.
• The ISC should inform the CDC STARHS lab to discard all specimens that have been tested with
STARHS. If the ISC would like to have these specimens returned at the surveillance area’s expense
he/she should make arrangements with the CDC STARHS lab.
Entering Data into the Surveillance Database – Data elements needed for the calculation of statistical
weights used to make population-based HIV incidence estimates fall into one of four categories.
• Demographic data
o Age
o Sex
o Race/ethnicity
o Risks associated with HIV infection
• HIV testing and ARV use history data
• Clinical data
o CD4 count
o Viral load
• Laboratory Data
o Specimen collection date
o SID

Page 9
o STARHS results
All data elements needed for HIS are included in eHARS, but could not be added to HARS. As a result, an
HIS Access database was developed to store additional data related to HIS. Until a surveillance area begins
using eHARS, the jurisdiction should enter data into both HARS and the HIS Access database as
appropriate. Testing and ARV use history information and laboratory data related to the diagnostic HIV test
(including specimen collection date and SID) are variables entered into the separate HIS Access database
along with the unique state number (HARS/eHARS “stateno”) assigned to each case.
Creating the HIS Dataset and Transferring Data to CDC – HIS data entry and management take place
at state or local health departments using one of three data management systems:
• HIS Access database (until conversion to eHARS)
• eHARS
• Software that is compatible with CDC software
Information is merged into the data management systems from other sources:
• Excel spreadsheet containing STARHS results identified by SID only from the CDC STARHS
laboratory
• PEMS
• Other databases
For states that have not transitioned to eHARS, data are merged using unique identifiers reported with each
case for data transfer to CDC:
• the HARS/eHARS stateno links HARS/eHARS records to corresponding HIS Access database
records
When eHARS has the ability to import HIS data, merging datasets prior to transferring data to CDC will no
longer be necessary.

Prior to the 15th of each month, the complete dataset is transmitted to CDC over the Secure Data Network
(SDN). Data transmitted to CDC will include no personal identifiers and will be encrypted and password
protected according to the .
Security and Confidentiality – HIV testing is a medical procedure. Therefore, policies and procedures are
in place to protect the confidentiality of tested individuals and their medical records. STARHS will be
performed only on specimens that have tested positive for HIV. HIS data are considered part of routine
HIV surveillance data and should be held to the standards of security and confidentiality for HIV/AIDS
surveillance outlined in . Policies and procedures, based on these guidelines and local laws, are already
in place at state and local health departments and are used to secure hard copy and electronic information to
protect the confidentiality of persons reported as having HIV infection. These measures must be extended
to protect the STARHS information held locally. Access by HIS staff to information in HARS, HIV testing
and ARV use history, and STARHS data is governed by the same security and confidentiality requirements.
Data Analysis and Dissemination – Data from individuals who choose to have an HIV test and who test
positive will be used to estimate the incidence of HIV nationally and in participating areas, including the

Page 10
incidence of undiagnosed HIV infection. HIV incidence estimates will be used to assess
current HIV prevention programs locally, regionally, and nationally. HIS data will be stratified by selected
factors such as demographic or behavioral factors, thus creating subpopulation data at the national and local
levels. If the sampling procedure has sufficient statistical power, this stratification will allow comparisons
between different areas and among different risk groups. The methods used to generate the populationbased incidence estimate are described in which introduces the methods, statistical formulas, and different groups for which
incidence estimates will be made.
It is expected that analysis, interpretation, and dissemination of these data will be the primary responsibility
of CDC with appropriate contributions from surveillance areas. Results from the aggregate CDC database
will be analyzed regularly and feedback provided to areas. Aggregate results will also be published in
CDC’s HIV/AIDS Surveillance Reports. Area-specific analyses will be conducted at the discretion of
participating areas. As appropriate, results will be presented at conferences and published in peer reviewed
journals. The number of representative authors from areas and CDC will be determined for each
presentation or paper.
In addition to the variables needed to estimate HIV incidence, the following data elements will be used to
evaluate the performance of the BED HIV-1 Capture EIA for the determination of estimates of HIV
incidence:
• Whether AIDS has been diagnosed and if so, the date of diagnosis.
• At the time of HIV diagnosis,
o whether HIV ARV agents have been used for post exposure prophylaxis or for any other
medical condition (e.g., lamivudine for treatment of hepatitis B), and if so, the name(s) of the
agent, and dates and duration of use,
o where available, CD4 count, viral load, and HIV-1 subtype along with the type of test used to
determine the subtype.
As a result, all HIV-positive diagnostic specimens should be tested using STARHS irrespective of the time
to AIDS diagnosis for an individual, or evidence of previous ARV use.
Staff Training – Because HIS is a fully integrated component within the HIV/AIDS surveillance system, all
HIS staff should receive training in the local policies and procedures for core surveillance including:
• active and passive surveillance methods
• laboratory reporting mechanisms
• data management processes
In accordance with HIS staff must also receive training in security and confidentiality procedures,
and should sign a confidentiality statement upon being hired and annually thereafter.

Outcome Standards
Outcome standards described within the and
sections of
can be applied to HIS. These sections address issues of completeness of case ascertainment, timeliness of
reporting, passing standard data edits, and elimination of missing/unknown information.

Page 11
Meeting core surveillance standards for case ascertainment and timeliness are essential for
HIS to be successful given the time sensitive nature of HIS data elements including testing and ARV use
history data and STARHS. In addition, the quality of the HIV incidence estimate is dependent on the
quality of data included in the HIS system. As with core surveillance data elements, the minimum standard
for passing standard data edits related to HIS data is 97% with a target of 100%.
At least 85% of newly reported HIV/AIDS cases for a diagnosis year should have testing history and ARV
use data within 12 months of the date of the initial HIV/AIDS case report, measured at 12 months after the
close of the diagnosis year.
Related to shipment of specimens for testing using STARHS, the minimum standard is that 85% of newly
reported HIV/AIDS cases reported to the surveillance system in a jurisdiction should have a remnant of the
HIV-positive diagnostic blood sample (or a remnant of a sample drawn for an HIV related blood test drawn
within 3 months of an unavailable HIV-positive diagnostic specimen) transported to the CDC STARHS
laboratory within 12 months of diagnosis assessed for the most recent diagnosis year at 12 months after that
diagnosis year.

References
1. Janssen RS, Satten GA, Stramer SL, et al. New testing strategy to detect early HIV-1 infection
for use in incidence estimates and for clinical and prevention purposes. JAMA 1998; 280:42–48.
2. Parekh B, Kennedy MS, Dobbs T, et al. Quantitative detection of increasing HIV type 1
antibodies after seroconversion: a simple assay for detecting recent HIV infection and estimating
incidence. AIDS Research & Human Retroviruses 2002;18(4):295-307.
3. White E, Goldbaum G, Rao V. Application of the serologic testing algorithm to detect recent
HIV seroconversion (STARHS) in a clinical population. In: Program and abstract of the 9th
Annual Conference on Retroviruses and Opportunistic Infections; February 2002; Seattle.
Abstract 766W.
4. Linley L, Reed C. Applicability of Population-Based STARHS HIV Incidence Measure in
Determining Recency of Individual Infection among Patients Attending STD Clinics. In
Program and abstract of the 11th Annual Conference on Retroviruses and Opportunistic
Infections; February 2004; San Francisco. Abstract 854.
5. Coutinho RA, Duermeyer W, Van der Veen J. Epidemiology of hepatitis A in Amsterdam
October 1978–December 1979. J Virol Methods 1980;2(1-2):47–55.

National Center for HIV, STD, and TB Prevention’s Non-research Determination for HIV
Incidence Surveillance

Standard HIV Incidence Surveillance Data Elements

Variable Description

Valid Value

Number of HIV tests in 2 years prior to the first
positive (For people who previously tested positive)

1 - 99
R=refused
D=don't know

Anonymous HIV test at first positive test

0=no
1=yes
7=refused
9=don't know

Antiretroviral medications within the last 6 months

0=no
1=yes
7=refused
9=don't know

ARV meds taken

22=Agenerase
30=Aptivus
24=Combivir
06 =Crixivan
11=Emtriva
03=Epivir
28=Epzicom
25=Fortovase
10=Fuzeon
19=Hepsera
02=Hivid
23=Hydroxyurea
18=Invirase
16=Kaletra
31=Lexiva
07=Norvir
88=Other
09=Rescriptor
26=Retrovir
15=Reyataz
08=Saquinavir

CLIA code for source lab where specimen originated

text

Currently taking antiretroviral medications

0=no
1=yes
7=refused

9=don't know
Date HAART use began

yyyymmdd

Date HAART use ended

yyyymmdd

Date information is extracted

yyyymmdd

Date of first HIV test ever

yyyymmdd

Date of first positive HIV test

yyyymmdd

Date of STARHS test

yyyymmdd

Date of the HIV test that resulted in a case report

yyyymmdd

Date specimen was obtained

yyyymmdd

Has specimen been approved for STARHS?

0=no
1=yes
2=pending

Laboratory ID

33D0654341=NYST
33D0654341=CDCSTARHS
33D0654341=NY
33D0654341=CDCSTAR
21D0649758=MARY01
50D0661430=WASH

Date of last HIV negative test before first positive

yyyymmdd

Date of last HIV negative test before first positive (or
before test that resulted in a case report if never
previously tested positive)

yyyymmdd

Name of site where first tested positive for HIV

text

Name of site where last tested negative for HIV

text

Ever tested for HIV

0=no
1=yes
7=refused
9=don't know

Ever tested negative

0=no
1=yes
7=refused
9=don't know

Number of HIV tests in last 2 years before first positive 1 - 99
(including first positive test)
R=refused
D=don't know
Number of HIV tests in last 2 years before first positive 1 - 99
(People who never had previous positive)
R=refused
D=don't know
Optical density

text

Ever tested positive

0=no
1=yes
7=refused
9=don't know

Reason for first positive HIV test*

text

Reason for testing at first positive - exposure to HIV
within the last 6 months*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - just checking to
make sure you are HIV negative*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - other reason*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - regular tester; time
for routine HIV test*

0=no
1=yes
7=refused
9=don't know

Reason for testing at first positive - required by either
insurance, military, court order, or for some other
required reason*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report - exposure 0=no
to HIV within the last 6 months*
1=yes
7=refused
9=don't know
Reason for the test that led to the case report- just
checking to make sure you are HIV negative*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report- other
reason*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report- regular
tester; time for routine HIV test*

0=no
1=yes
7=refused
9=don't know

Reason for the test that led to the case report - required
by either insurance, military, court order, or for some
other required reason*

0=no
1=yes
7=refused
9=don't know

Reason STARHS not performed

1=QNS
2=specimen never received at public lab
3=broken in transit
4=other

Results received

0=no
1=yes

Specify other reason for testing at fist positive*

text

Specify other reason for the test that led to the case
report*

text

Specimen ID number from source lab

text

STARHS ID

text

STARHS regional lab specimen ID number (same as
STARHS lab imported variable SPECIMEN ID)

text

STARHS test result

01=long term
02=recent
91=QNS
92=not rec’d by STARHS lab
93=broken
94=other

State lab CLIA code

text

State lab specimen ID number

text

State of site where first tested positive for HIV

text

State of site where last tested negative for HIV

text

Test assay

BED=BED
BVLS=BVLS (Vironostika LS)
OTLS=OTLS (Vironostika LS)
OTV=OTV (Vironostika LS)
AVID=AVID

Type of site where first tested positive for HIV

F01=inpatient facility
F01.01=inpatient facility/hospital
F01.04=inpatient facility/long term care
F01.50=inpatient facility/drug treatment
F01.OTH=inpatient facility/other
F01.UNK=inpatient facility unknown
F.OTH=facility/other
F.UNK=facility/unknown

Type of site where participant last tested negative for
HIV

F01=inpatient facility
F01.01=inpatient facility/hospital
F01.04=inpatient facility/long term care
F01.50=inpatient facility/drug treatment
F01.OTH=inpatient facility/other
F01.UNK=inpatient facility unknown
F.OTH=facility/other
F.UNK=facility/unknown

Type of test performed on specimen (LOINC)

5220-9=EIA / Elisa
21009-6=Western Blot
5472-6=CD4
25835-0=Viral Load (NASBA)
5017-9=Viral Load (bDAN)
25836-8=Viral Load (RT-PCR)

What type of specimen was obtained

1=blood finger stick
2=blood venipuncture
3=blood spot
4=oral mucosal transudate
5=urine
8=other
9=unknown

* The reason for testing for first positive test or for the test that led to the case report will no
longer be part of the standard HIV Incidence Surveillance data elements beginning January 1,
2007.

Statistical Method for Generating Population-Based HIV Incidence Estimates
I. Background
This appendix contains a description of the procedures for estimating the total number of people infected
with HIV in a specific population of interest during a one-year period. These estimation procedures use data
collected from a population-based surveillance system for newly diagnosed HIV infections. The
information required to estimate the number of recent HIV infections falls into three general categories:
1) Data collected as part of HIV/AIDS surveillance
2) HIV testing history prior to the first positive HIV test and
3) Test results that indicate whether an infection is recent.
These procedures provide an estimate of the number of incident infections in the population covered by the
surveillance system and not merely the number of new infections among the persons tested for HIV.
Since it is not feasible to test everyone at risk for HIV infection in a given year to determine who has
seroconverted, the statistical approach that we will use to estimate population-based HIV incidence is
analogous to the sample survey approach (Karon et al., 2005). In this approach, all individuals who
seroconvert in the year of interest are considered to be the sampling frame, whereas those who have been
identified as recent seroconverters are regarded as a sample chosen from that sampling frame.
Based on the sample survey approach, the size of the sampling frame is estimated from the observed sample
by assigning a weight to each person in the sample. This weight is the inverse of the probability that a
person with similar characteristics is included in the sample. Thus, for estimating HIV incidence, the
weight assigned to each person identified as recently infected is the inverse of the probability that a
seroconverted person had a positive HIV test and was identified as having a recent infection based on the
serologic testing algorithm for recent HIV seroconversion (Janssen et al., 1998).
A person identified as recently infected is defined as a person who tests HIV positive by the standard
(sensitive) test, but negative by the BED test. The interval between the time an individual tests positive by
the sensitive HIV test and the time he/she tests positive by the BED HIV test is called the window period. It
depends both on the assay used and the critical SOD (standard optical density) value chosen to define a
positive reaction. The window period differs from person to person and is unknown for each individual.
However, the distribution of window period durations among all seroconverters can be estimated.

II. Statistical procedure for estimating population-based incidence
The statistical procedure contains three major steps: stratification, weight calculation, and incidence
estimation. In the stratification step, all new HIV cases diagnosed in a specific time interval (e.g., a specific
calendar year) are stratified into incidence groups according to HIV transmission category or factors
associated with risk intensity and testing pattern. The number of new HIV diagnoses in each incidence
group should be sufficient large so that there will be at least 100 BED “recent” persons in each group.
Cases in each incidence group are then further divided into subgroups according to testing history,
motivation, and BED test information. In the weight calculation step, every newly diagnosed HIV case is
assigned an incidence weight. There are different types of weights determined by previous negative HIV
test, testing motivation, type of HIV diagnosis (HIV diagnosed with or without AIDS), result of the BED
test, and availability of needed information. The final step is to add weights in each incidence group to get
the incidence estimate for that group, and then aggregate the group estimates to get the incidence estimate
for the population.
A. Stratification
Since the accuracy of incidence estimates depends on homogeneity of risk level and testing pattern in the
group considered, we first stratify the population into groups with similar risk for HIV infection and testing
pattern in each group. However, the size of each group must be sufficiently large in order to have reliable
incidence weights assigned for cases in each group. Variables that could be used for stratification include
sex, race, transmission category, age at HIV diagnosis, state of diagnosis, or the type of testing facility. The
resulting strata are called incidence groups.
Within each incidence group, cases are divided into mutually exclusive subgroups. They are determined by
the availability of testing history information, the eligibility of BED testing, and the result of the BED test.
There are three major subgroups determined by previous negative HIV test. They are:
Gx – Cases without information on HIV testing before the first positive test for HIV
G0 – Cases without previous test before the first positive test for HIV
G1 – Cases not in groups Gx and G0, or tested before the first positive test for HIV
Each of the three groups is further divided into ten smaller groups based on (1) whether the first positive test
is motivated by a unique recent risk incident (2) whether AIDS is diagnosed at the same month of HIV
diagnosis (3) whether the case is eligible for BED test (no medication within six months prior the first
positive test) and BED tested with determinate result and (4) BED test result. These incidence subgroups are
shown in Tables 1-3. These subgroups are labeled as GABC with the following convention:
The first subscript is an indicator on previous test with A = 0 for cases with no test before the first positive
test, A = 1 for cases ever tested negative before the first positive test, and A = x for cases without
information on previous test.
The second subscript refers to test motivation with B = 0 for cases not motivated by a unique recent risk
incident, B = m for cases motivated by a unique recent risk incident, and B = x for cases without
information on test motivation.
The third subscript implies the status of BED test:
C = a for cases diagnosed with AIDS at HIV diagnosis

C = 0 for cases not AIDS at HIV diagnosis, no HIV related medication within 6 months before HIV
diagnosis, BED tested, and BED recent
C = b for cases not AIDS at HIV diagnosis, no HIV related medication within 6 months before HIV
diagnosis, BED tested, but not BED recent
C = x cases not in anyone of the above three subgroups.
B. Weight calculation
In this section, each newly diagnosed HIV case is assigned an incidence weight based on the incidence
subgroup associated with the case. There are four types of weights. They are (1) zero weight, (2) unit
weight, (3) calculated weight based on detection probability or testing frequency, and (4) derived weight
based on proportional method for cases with incomplete information.
Zero weight is assigned to AIDS cases (incidence subgroups GABC with C = a) and cases not BED recent
(incidence subgroups GABC with C = b and B = 0 or x):
W0 ma = W00 a = W0 xa = 0, W00b = W0 xb = 0
W1ma = W10 a = W1xa = 0, W10b = W1xb = 0

(1)

Wxma = Wx 0 a = Wxxa = 0, Wx 0b = Wxxb = 0
Unit weight is assigned to cases that were motivated by unique recent risk incident but not diagnosed with
AIDS at HIV diagnosis (incidence subgroups GABC with B = m but C ≠ a, they are G0mx, G1mx, and Gxmx):
W0 mx = W1mx = Wxmx = 1

(2)

Weights are calculated for cases that are known on previous test, not motivated by risk incident, eligible for
BED test (not AIDS and not on HIV medication within six months prior to the first positive test), and BED
recent. There are two types of calculated weights, one for those with no previous negative test (incidence
subgroup G000) and the other for those ever tested before the first positive test (incidence subgroup G100).
The weight for cases in G000 is given by
W000 =

1
∞

1+ ∫ e
0

−t ( q

−1 / α A

−1) / β A

S A (t )dSW (t )

(3)

where
q = N 00 a ( N 00 a + N 00 x + N 000 + N 00b )

(4)

is the proportion of individuals diagnosed with AIDS (CDC, 1993 definition) at the time of their HIV
diagnosis among those who had no previous HIV negative test and who’s first positive test was not
motivated by risk incident, SA(t) and SW(t) are the survival functions for the AIDS incubation time and the
duration of window period, respectively. The AIDS incubation time has a Gamma distribution (Longini et
al., 1991) with a shape parameter αA = 2 and a scale parameter βA = 48 months. Note that all of these
weights are equal. The duration of window period has a generalized log-logistic distribution with a survival
function given by (Robert H. Byers, Jr., personal communication)

(

SW (t ) = 1 + λe[ Ln ( t )−θ ] / σ

)

−1 / λ

(5)

With a SOD cut-off value equal to 1, the mean window period is about 184 days and the corresponding
parameters are λ = 1.044, θ = 5.048, and σ = 0.299. Using these parameters for the two survival functions,
weights for selected values of q from 0.01 to 0.45 are provided in Table 4.
The weights for cases in G100 and G1x0 are given by
T

W100 (T ) = W1x 0 (T ) =

∫

T

0

(6)

SW (t )dt

where T is the time from the last negative HIV test to the first positive HIV test. When T is greater than 24
months, the denominator is approximately equal to the mean window period μ = 184 days. Weights for
selected values of T from 0 to 24 months are provided in Table 5.
For all other cases (incidence subgroups GABC with A =x or B = x or C = x but B ≠ m), weights are derived
from other weights using the proportional method.
For cases in subgroup Gx00, the weight is

W x 00 = p0W000 + (1 − p 0 )W100

(7)

where
W100 =

1
N 100

N100

∑W
i =1

100

(Ti )

p0 = M 0 ( M 0 + M 1 )

(8)
(9)

and for A = 0,1,
M A = N A q A p A00
N A = N Ama + N Amx + N A0 a + N A0 x + N A00 + N A0b + N Axa + N Axx + N Ax 0 + N Axb
q A = ( N A0 x + N A00 + N A0b ) ( N Ama + N Amx + N A0 a + N A0 x + N A00 + N A0b )

(10)

p A00 = N A00 ( N A00 + N A0b )
For cases in subgroups G00x, G10x and Gx0x, the weights are

W00 x = p 000W000 , W10 x = p100W100 , W x 0 x = p x 00W x 00

(11)

where
p x 00 = N x 00 ( N x 00 + N x 0b )
For cases in subgroups G0x0, G1x0 and Gxx0, the weights are

(12)

W0 x 0 = p 0 mx + (1 − p 0 mx )W000
W1x 0 = p1mx + (1 − p1mx )W1x 0 (T )

(13)

W xx 0 = p xmx + (1 − p xmx )W x 00
where

p 0 mx = N 0 mx [ N 0 mx + N 000 + N 00 x ⋅ N 000 ( N 000 + N 00b )]
p1mx = N1mx [ N1mx + N100 + N10 x ⋅ N 100 ( N 100 + N 10b )]

(14)

p xmx = N xmx [ N xmx + N x 00 + N x 0 x ⋅ N x 00 ( N x 00 + N x 0b )]
W1x 0 (T ) is calculated using formula (6) and T is the time from the last negative HIV test to the first positive
HIV test.
Finally, for cases in subgroups G0xx, G1xx and Gxxx, the weights are
W0 xx = p 0 x 0W0 x 0 , W1xx = p1x 0 [ p1mx + (1 − p1mx )W1x 0 ], W xxx = p xx 0W xx 0
where

(15)

p0 x 0 = N 0 x 0 ( N 0 x 0 + N 0 xb )
p1x 0 = N1x 0 ( N1x 0 + N1xb )

(16)

p xx 0 = N xx 0 ( N xx 0 + N xxb )
W1x 0 =

1
N 1x 0

N1 x 0

∑W
i =1

1x 0

(Ti )

(17)

C. Incidence estimation

After each case has been assigned an incidence weight, the incidence for a specific population group can be
easily estimated by summing the weights of cases in the population considered:
I = ∑Wi

(18)

i

However, calculating the confidence interval for the incidence estimate is not straightforward because the
variance associated with the incidence estimate is complex. To estimate the variance, we need to know the
number of cases in each incidence group and each incidence subgroup.
Let r be the number of main incidence groups and Ii the incidence estimate for the i-th incidence group.
Then the total incidence in the population is the sum of incidences of all incidence groups:
r

I = ∑ Ii
i =1

The variance of this estimate is

(19)

r

V (I ) = ∑V (Ii )

(20)

i =1

Within each incidence group, e.g., group i, let kABC be the number of HIV/AIDS diagnoses in incidence
subgroups GABC within the specific population group of interest. Then the incidence estimate for the i-th
incidence group is
I i = I i 0 + I i1 + I ix
where

(21)

I i 0 = k 0 mx + k 000W000 + k 00 xW00 x + k 0 x 0W0 x 0 + k 0 xxW0 xx
= k 0 mx + (k 0 x 0 + k 0 xx p 0 x 0 ) p 0 mx

(22)

+ [k 000 + k 00 x p000 + (k 0 x 0 + k 0 xx p0 x 0 )(1 − p 0 mx )]W000
I i1 = k1mx + k100W100 + k10 xW10 x + k1x 0W1x 0 + k1xxW1xx
= k1mx + (k1x 0 + k1xx p1x 0 ) p1mx + (k100 + k10 x p100 )W100

(23)

+ (k1x 0 + k1xx p1x 0 )(1 − p1mx )W1x 0

and

I ix = k xmx + k x 00W x 00 + k x 0 xW x 0 x + k xx 0W xx 0 + k xxxW xxx
= k xmx + (k xx 0 + k xxx p xx 0 ) p xmx

(24)

+ [k x 00 + k x 0 x p x 00 + (k xx 0 + k xxx p xx 0 )(1 − p xmx )]W x 00
Let

v
v v v v v v
X = ( X 1 ,..., X 28 ) = ( K 0 , K1 , K x , P0 , P1 , Px , p0 ,W000 ,W100 ,W1x 0 )

where

v
K 0 = (k 0 mx , k 00 x , k 000 , k 0 xx , k 0 x 0 )
v
K 1 = (k1mx , k10 x , k100 , k1xx , k1x 0 )
v
K x = (k xmx , k x 0 x , k x 00 , k xxx , k xx 0 )
v
P0 = ( p0 mx , p 000 , p 0 x 0 )
v
P1 = ( p1mx , p100 , p1x 0 )
v
Px = ( p xmx , p x 00 , p xx 0 )

Then, we have

v
Ii = f (X )

Using the delta method, the variance of the incidence estimate can be approximated by
V (I i ) ≈ ∑
j ,l

The partial derivatives are

∂f ∂f
Cov( X j , X l )
∂X j ∂X l

(25)

∂f ∂ k 0 mx = 1
∂f ∂ k 00 x = p 000W000
∂f ∂ k 000 = W000
∂f ∂ k 0 xx = p0 x 0 [ p 0 mx + (1 − p 0 mx )W000 ]
∂f ∂ k 0 x 0 = p 0 mx + (1 − p 0 mx )W000
∂f ∂ k1mx = 1
∂f ∂ k10 x = p100W100
∂f ∂ k100 = W100
∂f ∂ k1xx = p1x 0 [ p1mx + (1 − p1mx )W1x 0 ]
∂f ∂ k1x 0 = p1mx + (1 − p1mx )W1x 0

∂f ∂ k xmx = 1
∂f ∂ k x 0 x = p x 00 [ p 0W000 + (1 − p0 )W100 ]
∂f ∂ k x 00 = [ p 0W000 + (1 − p 0 )W100 ]
∂f ∂ k xxx = p xx 0 { p xmx + (1 − p xmx )[ p0W000 + (1 − p 0 )W100 ]}
∂f ∂ k xx 0 = p xmx + (1 − p xmx )[ p 0W000 + (1 − p 0 )W100 ]
∂f ∂p 0 mx = (k 0 x 0 + k 0 xx p 0 x 0 )(1 − W000 )
∂f ∂p 000 = k 00 xW000
∂f ∂p 0 x 0 = k 0 xx p 0 mx + [k 0 xx (1 − p0 mx )]W000
∂f ∂p1mx = (k1x 0 + k1xx p1x 0 )(1 − W1x 0 )
∂f ∂p100 = k10 xW100
∂f ∂p1x 0 = k1xx p1mx + [k1xx (1 − p1mx )]W1x 0
∂f ∂p xmx = (k xx 0 + k xxx p xx 0 ){1 − [ p 0W000 + (1 − p 0 )W100 ]}
∂f ∂p x 00 = k x 0 x [ p 0W000 + (1 − p 0 )W100 ]
∂f ∂p xx 0 = k xxx p xmx + [k xxx (1 − p xmx )][ p 0W000 + (1 − p 0 )W100 ]
∂f ∂p 0 = [k x 00 + k x 0 x p x 00 + (k xx 0 + k xxx p xx 0 )(1 − p xmx )][W000 − W100 ]
∂f ∂W000 = [k 000 + k 00 x p 000 + (k 0 x 0 + k 0 xx p 0 x 0 )(1 − p 0 mx )]
+ [k x 00 + k x 0 x p x 00 + (k xx 0 + k xxx p xx 0 )(1 − p xmx )] p 0
∂f ∂W100 = [k100 + k10 x p100 ]
+ [k x 00 + k x 0 x p x 00 + (k xx 0 + k xxx p xx 0 )(1 − p xmx )](1 − p 0 )
∂f ∂W1x 0 = (k1x 0 + k1xx p1x 0 )(1 − p1mx )

We assume that each k is an independent proportion of the corresponding N and all N’s are independent
random variables each with a Poisson distribution. Therefore, the covariances between kz’s are zero and the
variances of kz’s can be estimated as
Vˆ (k z ) = k z
Note that the p’s are functions of N’s. Using the delta method, the variances of p’s can be estimated as (with
A = 0, 1, x):
Vˆ ( p A00 ) = p A00 (1 − p A00 ) ( N A00 + N A0b )
Vˆ ( p Ax 0 ) = p Ax 0 (1 − p Ax 0 ) ( N Ax 0 + N Axb )
Vˆ ( p ) ≈ p (1 − p ) [ N
+N +N
Amx

Amx

Amx

Amx

A 00

A0 x

N A00 ( N A00 + N A0b )]

Also, we have
Vˆ ( p 0 ) ≈ p 0 (1 − p 0 ) ( M 0 + M 1 )

Vˆ (W000 ) = [W00′ (q)]2 q(1 − q) ( N 00 a + N 00 x + N 000 + N 00b )
1
Vˆ (W100 ) = 2
N 100

∑W

1
Vˆ (W1x 0 ) = 2
N 1x 0

N1 x 0

N100
i =1

2
100

∑W
i =1

2
1x 0

(Ti )
(Ti )

Covariances between p’s and k’s that are not independent can be estimated as:
Coˆv( p A00 , k A00 ) = (∂p A00 ∂N A00 )k A00
Coˆv( p Ax 0 , k Ax 0 ) = (∂p Ax 0 ∂N Ax 0 )k Ax 0
Coˆv( p Amx , k Amx ) = (∂p Amx ∂N Amx )k Amx
s
Co v( p Amx , k A0 x ) = (∂p Amx ∂N A0 x )k A0 x
Coˆv( p Amx , k A00 ) = (∂p Amx ∂N A00 )k A00
Coˆv( p Amx , p A00 ) = [(∂p Amx ∂N A00 ) − (∂p Amx ∂N A0b )] p A00 (1 − p A00 )
for A = 0, 1, x
and

Coˆv( p 0 , k Amx ) = (∂p0 ∂N Amx )k Amx
Coˆv( p 0 , k A0 x ) = (∂p0 ∂N A0 x )k A0 x
Coˆv( p 0 , k A00 ) = (∂p0 ∂N A00 )k A00
Coˆv( p 0 , k Axx ) = (∂p0 ∂N Axx )k Axx
Coˆv( p 0 , k Ax 0 ) = (∂p0 ∂N Ax 0 )k Ax 0
∂p 0 ∂p Amx
∂p 0 ∂p Amx
Coˆv( p 0 , p Amx ) =
N Amx +
N A00
∂N Amx ∂N Amx
∂N A00 ∂N A00

+

∂p 0 ∂p Amx
∂p0 ∂p Amx
N A0 x +
N A0b
∂N A0 x ∂N A0 x
∂N A0b ∂N A0b

⎛ ∂p 0
∂p 0 ⎞
⎟⎟ p A00 (1 − p A00 )
Coˆv( p 0 , p A00 ) = ⎜⎜
−
∂
N
N
∂
A0b ⎠
⎝ A00
for A = 0, 1,

where

∂p A00 ∂N A00 = p A00 (1 − p A00 ) N A00
∂p Ax 0 ∂N Ax 0 = p Ax 0 (1 − p Ax 0 ) N Ax 0
∂p Amx ∂N Amx = p Amx (1 − p Amx ) N Amx
2
N Amx ) p A00
∂p Amx ∂N A0 x = −( p Amx
2
N Amx )[1 + N A0 x p A00 (1 − p A00 ) N A00 )]
∂p Amx ∂N A00 = −( p Amx
2
N Amx ) N A0 x p A00 (1 − p A00 ) N A0b
∂p Amx ∂N A0b = ( p Amx
for A = 0, 1, x

and
⎞
⎛ 1
∂p 0
qA
⎟⎟
= (−1) A p 0 (1 − p 0 )⎜⎜
+
∂N Amx
⎝ N A N A0 x + N A00 + N A0b ⎠
⎛ 1
∂p 0
(1 − q A ) 2
1
= (−1) A p 0 (1 − p 0 )⎜⎜
+
∂N A0 x
⎝ N A q A N Ama + N Amx + N A0 a

⎞
⎟⎟
⎠

⎛ 1
1 − p A00 ⎞
∂p 0
(1 − q A ) 2
1
A
⎟
= (−1) p 0 (1 − p 0 )⎜⎜
+
+
N A00 ⎟⎠
∂N A00
⎝ N A q A N Ama + N Amx + N A0 a
⎛ 1
1 − p A00 ⎞
∂p 0
(1 − q A ) 2
1
⎟
−
= (−1) A p 0 (1 − p 0 )⎜⎜
+
N A0b ⎟⎠
∂N A0b
⎝ N A q A N Ama + N Amx + N A0 a
∂p 0
p (1 − p 0 )
= (−1) A 0
∂N Axx
NA
p (1 − p0 )
∂p0
= (−1) A 0
NA
∂N Ax 0
∂p 0
1
= (−1) A p 0 (1 − p 0 )
NA
∂N Axb
for A = 0, 1.

We estimate the coveriances between W’s and other components that are correlated with W’s by
′ (q )(q 2 N 00 a )k 000
Coˆv(W000 , k 000 ) = −W000
′ (q )(q 2 N 00 a )k 00 x
Coˆv(W000 , k 00 x ) = −W000
′ (q ){q (1 − q ) (∂p 0 ∂N 00 a ) − (q 2 N 00 a )[(∂p 0 ∂N 00 x ) N 00 x
Coˆv(W000 , p 0 ) = W000
+ (∂p 0 ∂N 000 ) N 000 + (∂p 0 ∂N 00b ) N 00b ]}

where
⎛ 1
⎞
∂p 0
− q0
⎟⎟
= p 0 (1 − p 0 )⎜⎜
+
∂N 00 a
⎝ N 0 N 00 x + N 000 + N 00b ⎠

Finally, a 95% confidence interval for the total incidence is given by:

I ± 1.96 Vˆ ( I )

(26)

III. Data required for estimating HIV incidence

For each person reported with a newly diagnosed HIV infection:
1. Demographic information: sex, race, transmission category, age, and state of residence at first
positive HIV test.
2. Information about the first positive HIV test: testing date, type of testing site, reason for the positive
HIV test.
3. Testing history information: whether or not tested before the first positive HIV test. If tested before
the first positive HIV test:
a.
Date of the last negative HIV test,
b.
Date of the very first HIV test, and
c.
Testing frequency and pattern in the last two years.
Note: b and c are needed to evaluate the potential truncation effect on incidence estimation.
4. Information about eligibility for BED test: whether AIDS has been diagnosed, whether HIV
antiretroviral agents have been used in 6 months prior to the HIV diagnostic test, and if available, the
CD4 count and viral load at time of HIV diagnosis.
5. Information on BED test: date of blood drawn for BED test, (type of assay), BED result.
References

1.
2.
3.
4.

Karon, K.M., Song, R., Kaplan, E., and Brookmeyer, R. Estimating HIV incidence from
HIV/AIDS surveillance data and less-sensitive HIV test results. In preparation.
Janssen RS, Satten GA, Stramer SL, et al. New testing strategy to detect early HIV-1
infection for use in incidence estimates and for clinical and prevention purposes. JAMA
1998; 280:42–48.
Centers for Disease Control and Prevention. 1993 Revised classification system for HIV
infection and expanded surveillance case definition for AIDS among adolescents and adults.
MMWR 1992; 41(no.RR-17).
Longini IM, Clark WS, Gardner LI, et al. The dynamics of CD4+ T-lymphocyte decline in
HIV-infected individuals: A Markov modeling approach. JAIDS 1991; 4:1141–1147.

Table 1. Incidence subgroups and weights for those who were not HIV tested before their first positive test
(group G0)
Motivate AIDS
On medication
BE
N
Incidence
Su
at
unique re
BED or BED un
cas
weight
rec
gro
HIV
incide
W0 ma = 0
Yes
G0
N0
Yes
W0 mx = 1
No
G0
N0
G0

N0

W00 a = 0

G0

N0

W00 x = p 000W000

Ye

G0

N0

W000 , see equation (3)

N

G0

N0

W00b = 0

G0

N0

W0 xa = 0

G0

N0

W0 xx = p0 x 0W0 x 0

Ye

G0

N0

N

G0

N0

Yes
No

Yes
No

No

Yes
Unkno

Yes
No

No

W0 x 0 = p 0 mx + (1 − p0 mx )W
W0 xb = 0

Table 2. Incidence subgroups and weights for those who were HIV tested before their first positive test
(group G1)
Motivate AIDS
On medication
BE
N
Incidence
Su
at
unique re
BED or BED un
cas
weight
rec
gro
HIV
incide
W1ma = 0
Yes
G1
N1
Yes
W1mx = 1
No
G1
N1
G1

N1

W10 a = 0

G1

N1

W10 x = p100W100

Ye

G1

N1

W100 (T ) , use equation (6

N

G1

N1

W10b = 0

G1

N1

W1xa = 0

G1

N1

Yes
No

Yes
No

No

Yes
Yes
Unkno

W1xx = p1x 0 ( p1mx +

(1 − p1mx )W1x 0 )
W1x 0 = p1mx + (1 − p1mx )

No
No

Ye

G1

N1

N

G1

N1

W1x 0 (T )
use equation (6)
W1xb = 0

Table 3. Incidence subgroups and weights for those who had no information on previous HIV test (group
Gx)
Motivate AIDS
On medication
BE
N
Incidence
Su
at
unique re
BED or BED un
cas
weight
rec
gro
HIV
incide
W xma = 0
Yes
Gx
Nx
Yes
W xmx = 1
No
Gx
Nx
Gx

Nx

Wx0a = 0

Gx

Nx

W x 0 x = p x 00W x 00

Ye

Gx

Nx

N

Gx

Nx

W x 0b = 0

Gx

Nx

W xxa = 0

Gx

Nx

W xxx = p xx 0W xx 0

Ye

Gx

Nx

N

Gx

Nx

Yes
No

Yes
No

No

Yes
Unkno

Yes
No

No

W x 00 = p 0W000 + (1 − p0 )W

W xx 0 = p xmx + (1 − p xmx )W
W xxb = 0

Table 4. Weights and their first derivative values for selected values of q.
q
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.1
0.1

W0
1.6
2.0
2.3
2.6
2.9
3.2
3.5
3.7
4.0
4.2
4.5
4.7
5.0
5.2
5.5

W'0
47.4
37.0
32.5
29.9
28.3
27.2
26.4
25.9
25.5
25.2
25.0
24.9
24.9
24.9
25.0

q
0.1
0.1
0.1
0.1
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.3

W0
5.7
6.0
6.2
6.5
6.8
7.0
7.3
7.6
7.8
8.1
8.4
8.7
8.9
9.2
9.5

W'0
25.1
25.3
25.5
25.7
25.9
26.2
26.5
26.8
27.2
27.6
28.0
28.4
28.9
29.3
29.9

q
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.4
0.4
0.4
0.4
0.4
0.4

W0
9.8
10.1
10.5
10.8
11.1
11.4
11.8
12.1
12.5
12.8
13.2
13.6
14.0
14.4
14.8

W'0
30.4
30.9
31.5
32.1
32.8
33.5
34.2
34.9
35.7
36.5
37.3
38.2
39.1
40.1
41.1

Table 5. Weights associated with the time in months from the last negative HIV test to the first positive HIV
test (μ = 184 days is the mean window period).

T
0
1
2
3
4
5
6
7
8
9
10
11
12

W10
1.00
1.00
1.01
1.03
1.09
1.17
1.27
1.39
1.52
1.66
1.80
1.95
2.10

T
13
14
15
16
17
18
19
20
21
22
23
24
>24

W10
2.26
2.41
2.57
2.73
2.89
3.05
3.21
3.37
3.54
3.70
3.86
4.03
T/μ

Page 1
Guidance for the Transportation of Remnant HIV-Positive Specimens to the CDC
STARHS Laboratory from Private or Public Testing Laboratories
Purpose
This guidance describes two possible specimen transport models that originating laboratories
may use to ship remnant diagnostic serum specimens to the CDC STARHS laboratory for testing
for recent HIV-1 infection using the serologic testing algorithm for recent HIV seroconversion
(STARHS). Originating laboratories may choose to select either model, but must clearly
communicate their choice to and coordinate with the state / local HIV Incidence Surveillance
Coordinator (ISC) who will be responsible for managing the results.

Introduction
In December 2004, an expert consultation was convened by the Centers for Disease Control and
Prevention (CDC) and the Association of Public Health Laboratories (APHL). The purpose of
the 5th HIV Incidence Consultation on Laboratory and Specimen Transport was to discuss the
best approaches for transporting remnant HIV positive sera from private and public laboratories
to the CDC STARHS laboratory for testing using STARHS. The consultation participants
included HIV Incidence Surveillance staff from CDC and state / local areas, personnel from
commercial, private, university, and public health laboratories, APHL, and the American Clinical
Laboratory Association. The goal for the meeting was to gather input from stakeholders for the
development of an infrastructure for shipping specimens from private (including university
and/or medical center), commercial, and public health laboratories to the CDC STARHS
laboratory for STARHS.
The discussions concluded that two models were acceptable for shipping specimens from testing
laboratories to the CDC STARHS laboratory. This guidance describes both specimen
transportation models. The models differ in: 1) the extent of the testing laboratories involvement
in aliquoting / labeling samples for STARHS; 2) the physical storage location of the samples
until specimen disposition (to be tested using STARHS or to be discarded) is determined by the
ISC; and 3) the frequency of shipments to the CDC STARHS laboratory.
Each testing laboratory may choose either model, but this choice should be clearly
communicated to the ISC.

Laboratory Types
For the purposes of this guidance, there are three laboratory types. While each testing laboratory
may independently decide which specimen transport model will work best for that facility, CDC
has provided suggestions based on the type of laboratory and that facility’s relationship with the
state / local public health laboratory.

Page 2
Laboratory Types:
1) Commercial laboratories that process samples from many states and/or jurisdictions
(included in this category are: Quest Diagnostics Inc, Laboratory Corporation of America
[LabCorp], ARUP Laboratories, Specialty Laboratories, and Mayo Clinic)
2) Smaller private / university / hospital or medical center laboratories that provide
service primarily at the state or local level, but may also process samples for more than
one state and/or jurisdiction
3) Public Health Laboratories (PHL)
Specimen Information
Type of Specimens Shipped to CDC STARHS Laboratory - HIV positive serum from
Western Blot (WB) or Immunofluorescence Assay (IFA) confirmed diagnostic samples will
ultimately be shipped to the CDC STARHS laboratory, depending on the specimen transport
model chosen by the originating laboratory. Detailed information about which samples will
be shipped is included in the model descriptions of this guidance (Specimen Transport
Options).
Specimen Volume - The optimal quantity of serum required for STARHS testing is 0.5ml
per aliquot. However, if less than 0.5ml of the remnant sample is available for testing using
STARHS, the sample should still be sent to the CDC STARHS laboratory. The CDC
STARHS laboratory is the only laboratory that should determine if a sample is rejected due
to insufficient quantity.
Sample Storage - Short-term (less than one week) storage of samples in the refrigerator (2
– 8°C) is acceptable, but for long-term storage (more than one week), samples must be frozen
at negative (-)20°C or colder. This includes any period of time that the samples are kept at
the originating / testing laboratory or the “pass through” public health laboratory prior to
shipment to the CDC STARHS laboratory or the interim period while STARHS disposition
is being determined. Effort should be made to avoid repeated freezing and thawing of
samples, as this may give erroneous results.
•
•
•

It is recommended that, if not already in practice, a daily temperature log be kept to
ensure the freezer is operating properly.
The freezer should be housed in a location with proper ventilation to avoid
overheating and freezer failure.
The freezer must contain adequate space to store specimens.

Specimen Numbering - The specimen number on the samples shipped to the CDC
STARHS laboratory will either be the original laboratory-assigned specimen accession
number or the STARHS identification number, depending on the transport model selected by
the originating laboratory. Detailed information about specimen numbering is included in
the model descriptions of this guidance (Specimen Transport Options).

Page 3
Specimen Retention - The ISC must coordinate with the laboratory storing HIV positive
remnant sera (the CDC STARHS laboratory and/or their state / local PHL) to identify
samples that should be tested using STARHS. However, not all stored samples will be tested
using STARHS, and those that will not be tested will have to be identified for disposal. The
ISC should regularly notify the storage laboratory of which samples should be tested using
STARHS and which should be disposed of by submitting a list of laboratory-assigned
specimen accession numbers with “test” or “toss” for each specimen according to the
decision reached. The state / local ISC and the storing laboratory should communicate
regularly (every 1-3 months) to discuss any specimens for which no disposition has been
communicated to determine if the sample can be disposed of or if further investigation is
needed. Samples should not be destroyed or disposed of until the disposition is definitively
determined.

Packaging and Shipping Procedures
Shipping Guidance - Shipping procedures are described in detail in the CDC document
.
Specimens may be shipped from originating laboratories to the state PHL as a “pass-through”
facility or to the CDC STARHS laboratory as ‘Diagnostic Specimens. However, due to the
requirement for dry ice, all laboratories shipping HIV positive samples must be certified to
ship dangerous goods.
Frequency of Shipments - The frequency of specimen shipments to the CDC STARHS
laboratory or the pass-through facility will be determined by the shipping laboratory,
considering factors such as specimen retention policies and freezer / storage space, and in
consultation with the ISC and the receiving laboratory.
Shipping Couriers - Specimens must be shipped on dry ice by same-day or overnight
delivery service to ensure that specimens do not thaw in transit. The shipping laboratory may
decide which courier service to use for specimen transport.
Additional Information for Commercial Laboratories Only - APHL has set up a Federal
Express billing account for the large commercial laboratories to defer costs of shipping
samples to the CDC STARHS laboratory.
Additional Information for All Other Private Laboratories - Intra-state shipments from
private laboratories to the PHL may be shipped by Federal Express (or a similar
commercial courier) or an established local courier service. Funding permitting, states
may elect to set-up a billing account with Federal Express (or a similar commercial
courier) to pay for shipping costs incurred by the private laboratory to either the state /
local PHL or the CDC STARHS laboratory (see Funding for Specimen Handling).
Tracking Shipments - The shipping laboratory should notify the receiving laboratory (state
PHL or CDC STARHS laboratory) by fax or email when specimens are shipped, including
the name of the courier and the tracking number of the shipment. The receiving laboratory
will be responsible for tracking the shipments and will notify the originating laboratory if the
specimens are not received

Page 4
Additional Information Related to Commercial Laboratories Only - As part of the
contract between APHL and the large commercial laboratories, APHL will track
shipments from the commercial laboratories to the CDC STARHS laboratory.
Commercial laboratories must provide APHL with a list of sample numbers sent to the
CDC STARHS laboratory. The CDC STARHS laboratory must notify APHL of any
shipments sent from a commercial laboratory to the CDC STARHS laboratory that are
received. This notification is needed for billing reconciliation purposes at APHL.
Sample Rejection Criteria - Sample rejection due to thawing, breakage, insufficient
quantity, or lost-in-transit will be determined and recorded by the CDC STARHS laboratory.
Confidentiality and HIPAA Regulations
STARHS must ensure that confidentiality is protected and maintained to meet standards for HIV
surveillance. The Privacy Rule of the Health Insurance Portability and Accountability Act
(HIPAA) regulations permit protected health information to be shared for the purposes of public
health surveillance activities (1). This protection allows the originating laboratories to send
specimens labeled with their laboratory-assigned accession number to either the state PHL or the
CDC STARHS laboratory, where samples to be tested using STARHS will be re-assigned a
unique STARHS identification number prior to testing. This process will minimize re-labeling
errors and simplify the shipment procedures for private laboratories. However, the state / local
public health department must have the laboratory accession number to link the test result to the
patient information in the surveillance record. Therefore, the laboratory accession number must
be included on the HIV laboratory report sent to the state / local public health department by the
originating laboratory.
Funding for Specimen Handling
As a rule, surveillance is not a remunerated activity. However, through a Cooperative
Agreement with APHL (# U60/CCU303019-17), a pre-determined fee to offset personnel,
administrative, and handling costs incurred by the high volume, multi-jurisdictional commercial
laboratories (Section 3, Laboratory Type 1) will be reimbursed for an initial start-up period. The
APHL Cooperative Agreement is effective through June 2006, but may be renewable.
Reimbursement for other private laboratories is not covered by the APHL Cooperative
Agreement; funding for specimen handling costs may be made available through the state’s
Cooperative Agreement with CDC for HIV/AIDS Incidence Surveillance (Program
Announcement 04017), but handling fee reimbursement is not recommended.
Specimen Transport Options
Option A: Specimen Originated at Private Laboratory and is Sent Directly to the
CDC STARHS Laboratory

Page 5
Transportation Overview - In this transportation model, the originating private laboratory
performing the confirmatory testing will send all confirmed HIV positive diagnostic
specimens directly to the CDC STARHS laboratory (bypassing the state / local PHL) by
overnight shipping in accordance with the procedures described in the document . The CDC
STARHS laboratory will store specimens until specimen disposition is determined by the
state / local ISC, at which point, the CDC STARHS laboratory will pull samples to be tested
using STARHS, aliquot, re-label them with a STARHS identification number, and perform
STARHS. All samples that should not be tested using STARHS (i.e., samples that are not
the diagnostic specimen) will be discarded.
In this model, the originating laboratory would continue to submit laboratory report
information in the current manner to the appropriate jurisdiction, but must also include the
laboratory-assigned specimen accession number and the collection date on the report.
Figure 1 graphically depicts the flow of specimens and reports when samples originate at a
private laboratory and are then shipped directly to the CDC STARHS laboratory.
Procedures for Specimens Sent Directly from a Private Laboratory to the CDC STARHS
Laboratory
•

The ISC notifies the CDC HIV Incidence Surveillance (HIS) Coordinator about
each private laboratory that plans to send specimens directly to the CDC
STARHS laboratory. The ISC should send the following information to the CDC
HIS Coordinator for each laboratory:
o Name of laboratory and laboratory point of contact,
o Full contact information, including mailing address, phone number, fax, and
email
o Estimated number of positive samples expected per year.

•

The CDC HIS Coordinator will contact the CDC STARHS laboratory with this
information from private laboratories. This information is important for the CDC
STARHS laboratory to plan for storage capacity.

•

The state/local ISC should provide the private laboratory with a copy of the
. The form
should be pre-filled with the laboratory contact information, the name and address
of the person who will receive the STARHS results (ISC), and the appropriate
check box marked for Incidence Surveillance:

INCIDENCE SURVEILLANCE (HICSB)
•

The submitting private laboratory must include a copy of the with a list of all laboratoryassigned accession numbers included in the shipment. If more than one specimen
box is included in the shipping container, then each box should contain its own
STARHS Specimen Submission Form. If possible, an encrypted electronic
version of the list of specimen accession numbers should also be included in the

Page 6
shipment; this version will help the CDC STARHS laboratory log the samples
with minimal chance of entry errors. Note: the manifest should not contain any
patient identifiers other than specimen accession numbers. At the time of
shipment, the submitting laboratory should also mail a copy of the shipping
manifest to the ISC, notifying him/her of the shipment. This notification is critical
for the ISC to be able to track specimens. As previously noted, identifying
information for specimens should not be faxed or e-mailed, even if encrypted.
•

The CDC STARHS Laboratory will not provide the private laboratories with any
shipping materials, labels or cryovials, but will return the shipping container if a
prepaid return air bill is included in the shipment. Surveillance sites may provide
the private laboratories with prepaid shipping labels or shipping account numbers
(i.e., Federal Express) to cover shipping expenses.

•

Specimens shipped from private laboratories directly to the CDC STARHS
laboratory will be stored frozen, indicated only by their original laboratoryassigned accession number. The ISC will send to the CDC STARHS laboratory a
written list of specimens (identified by the original laboratory-assigned accession
number, and, if known, the name of the originating lab) that are to be tested using
STARHS (test list) or to be disposed of (toss list). The CDC STARHS laboratory
will not assign a STARHS identification number unless the ISC notifies the
laboratory that the sample is to be tested using STARHS. The CDC STARHS
laboratory will continue to hold specimens that are not on one of these two lists.

•

The CDC STARHS laboratory will test the specimens and send the STARHS
results back to the designated ISC listed on the . Samples will be tested and reported by
the newly assigned STARHS identification number.

•

Periodically the CDC STARHS laboratory will review the stored specimens to
reconcile the status of any samples that have been stored for a lengthy period of
time. This will reveal any specimens that the ISC never ordered to be tested or
discarded. The length of time specimens must be held will vary widely by
surveillance site depending on such factors as reporting delays, etc.

Roles of Parties Involved
Role of Private Laboratories - The private laboratories are responsible for forwarding two
items for HIV incidence testing: (1) a laboratory report to the public health surveillance
department per local requirements, including the collection date, the laboratory-assigned
specimen accession number, and identification information about the testing facility; (2)
remnant HIV positive serum from WB or IFA confirmed diagnostic samples labeled with
the laboratory-assigned specimen accession number.
The private laboratories may elect to aliquot 0.5 ml of the remnant sera for shipment to
the CDC STARHS laboratory so that any additional portion of the remnant sera may be
stored at their facility, or they may send the entirety of their remnant sera, without any
further manipulation, to the CDC STARHS laboratory.

Page 7
Prior to sending shipments to the CDC STARHS laboratory, private laboratories should
carefully review the to ensure proper shipping and handling of specimens.
Role of CDC STARHS Laboratory - The CDC STARHS laboratory must store all
remnant HIV positive serum samples received until specimen disposition has been
determined by the appropriate jurisdiction’s ISC. The ISC will provide the CDC
STARHS laboratory with a list of all samples to be tested using STARHS (test list) and a
list of all samples to be discarded (toss list), listed by specimen accession number. The
samples on the toss list should be discarded according to established laboratory methods.
The CDC STARHS laboratory will pull all samples on the test list and aliquot them into
the designated cryogenic vial for testing. The CDC STARHS laboratory will
simultaneously re-label the samples to be tested using STARHS with a STARHS
identification number. The CDC STARHS laboratory must provide the appropriate ISC
with a link between the STARHS identification number and the original specimen
accession number. After the ISC has been provided with the linkage information, the
CDC STARHS laboratory will destroy the laboratory copy of the specimen accession
information.
The CDC STARHS laboratory will test all samples on the test list by the STARHS
identification number and send results to the ISC from the appropriate jurisdiction. The
STARHS results are for surveillance purposes only therefore results will not be reported
back to the originating laboratory, provider, or client.
Role of the State / Local HIV Incidence Surveillance Coordinator - The ISC from the
jurisdiction where the specimen originated will determine the disposition of the specimen
and coordinate with the CDC STARHS laboratory to ensure that the specimen is either
tested or discarded as appropriate. The ISC will also maintain the link between the
original specimen accession number and the STARHS number, and will manage the
STARHS results.
Specimen Numbering - Specimens will be stored at the CDC STARHS laboratory by the
original laboratory-assigned specimen accession number. Once a specimen appears on the
test list the sample will be assigned a unique STARHS identification number and will be
tested using STARHS. All subsequent procedures use only the STARHS identification
number.
Theoretical Laboratory Types for this Transportation Model - The laboratories that
would best use this model are high volume, multi-jurisdictional commercial laboratories.
However, other private laboratories may also choose to elect this specimen transport model.
Note: The testing laboratory may choose either of the two transport models. With the
exception of public health laboratories, the examples listed in this section are merely
suggestions, not requirements, for the types of laboratories that may choose to elect this
model.

Page 8

Option B: Specimen Originated at or Sent Via State / Local Public Health
Laboratory
Transportation Overview - In this transportation model, confirmatory testing will have
been performed at either the PHL or a private laboratory. For samples originating at a
private laboratory, that laboratory will send all confirmed HIV positive diagnostic
specimens to the state / local PHL. The state / local PHL will store all specimens
received from the private laboratories until sample disposition is determined by the ISC.
All specimens on the test list (those to be tested using STARHS) will be pulled, aliquoted
into the designated cryogenic vials provided by the CDC STARHS laboratory, relabeled
with a STARHS identification number, and shipped to the CDC STARHS laboratory by
overnight shipping in accordance with the procedures described in the document
. All
specimens that are on the toss list (those that are not to be tested using STARHS) will be
pulled and discarded according to existing laboratory procedures.
In this model, the originating laboratory would continue to submit laboratory report
information in the current manner, but must also include the laboratory-assigned
specimen accession number, other relevant specimen identifiers, and testing laboratory
identification on the report.
Many laboratories send enzyme immunoassay (EIA) positive specimens to a reference
laboratory for confirmatory WB or IFA, which usually results in different laboratory
accession numbers. In this case, care must be taken to ensure that the appropriate
specimen accession numbers are associated with the correct surveillance report.

Figure 2a graphically depicts the flow of specimens and reports when samples originate
at the PHL and are then shipped to the CDC STARHS laboratory. Figure 2b describes
the flow of specimens and reports when samples originate at a private laboratory and are
sent to the PHL for storage prior to shipment to the CDC STARHS laboratory.
Procedures for Specimens Sent from a Private Laboratory through a State Public Health
Laboratory to the CDC STARHS Laboratory
•

The ISC works with each private laboratory to set up procedures for shipping
specimens to the state PHL.

•

The ISC should provide the private laboratory with a copy of the . The form can be
pre-filled with the state PHL contact information, or a new form may be
developed in agreement with the PHL. Some private laboratories already have
existing mechanisms for transporting specimens to the PHL. These procedures
may also be used.

•

The submitting private laboratory must include a list of all laboratory-assigned
specimen accession numbers included in the shipment to the PHL. At the time of
shipment, the private laboratory will mail a copy of the list to the ISC, notifying
him/her of the shipment. The private laboratory should also notify the PHL of the

Page 9
shipment by calling or emailing the PHL with the shipment tracking number, if
applicable, and the number of samples sent. This information is critical for both
parties to be able to track specimens.
•

The PHL will store the specimens from the private laboratory, holding them until
specimen disposition is determined by the ISC. Specimens should be stored
frozen by the original laboratory-assigned specimen accession number. On a
regular basis, the ISC will notify the PHL which specimens they are storing
should be pulled for STARHS testing (test list) and those that can be discarded
(toss list).

•

The PHL will discard all specimens on the toss list and will prepare all specimens
on the test list for testing using STARHS.
o All specimens to be tested using STARHS will be pulled, thawed, and
aliquoted into the designated cryovials provided by the CDC STARHS
laboratory.
o The PHL will re-label the samples to be tested using STARHS with a unique
STARHS identification number using labels provided to the PHL by the CDC
STARHS laboratory.
o The PHL will send the ISC the linkage information between the original
laboratory-assigned specimen accession number and the new unique STARHS
identification number.
o The PHL will ship all re-labeled specimens to the CDC STARHS laboratory
according to the procedures described in the document .
o The PHL should provide the CDC STARHS laboratory with a completed
, listing
all samples in the shipment by the newly-assigned STARHS identification
number. The PHL should also include an encrypted electronic version of the
specimen list in the shipment; this version will help minimize data entry errors
at the CDC STARHS laboratory. The list of STARHS identification numbers
on the
serves as verification from the ISC that all samples in the shipment are to be
tested using STARHS.
o A copy of the completed should also be mailed to the ISC as notification of the
shipment.
o The PHL should also notify the CDC STARHS laboratory of the shipment by
calling or emailing the laboratory to provide the shipment tracking number
and number of samples sent.

•

The CDC STARHS laboratory will test all samples received from the PHL with a
pre-assigned and labeled STARHS identification number and send STARHS
results back to the designated ISC listed on the .

Page 10
•

Periodically the PHL should review their stored specimens to reconcile the status
of any samples that have been stored for a lengthy period of time. This review
will reveal any specimens that the ISC did not order to be tested or discarded. The
length of time specimens must be held will vary widely by surveillance site
depending on such factors as reporting delays, etc.

Roles of Parties Involved
Role of Private Laboratories - The private laboratories are responsible for forwarding two
items for HIV incidence testing: (1) a laboratory report to the public health surveillance
department per local requirements with specimen identifiers, the laboratory-assigned
specimen accession number, and identification information about the testing facility; (2)
remnant HIV positive serum from WB or IFA confirmed diagnostic samples labeled with
the laboratory-assigned specimen accession number and testing laboratory identification
information.
The private laboratories may elect to aliquot 0.5 ml of the remnant sera to send to the
PHL so that any additional portion of the remnant sera may be stored at their facility, or
they may send the entirety of their remnant sera, without any further manipulation, to the
PHL.
Prior to sending shipments to the CDC STARHS laboratory, private laboratories should
carefully review the to ensure proper shipping and handling of specimens.
Role of Public Health Laboratories - The PHL must store all remnant HIV positive serum
samples (tested in their own facility or shipped from a private laboratory) until specimen
disposition has been determined by the ISC. The ISC will provide the PHL with a list of
all samples to be tested using STARHS (test list) and a list of all samples to be discarded
(toss list) listed by specimen accession number.
The PHL will pull all samples that will not be tested using STARHS and discard them
according to existing laboratory procedures.
The PHL will pull all samples to be tested using STARHS and aliquot them into the
designated cryogenic vial provided by the CDC STARHS laboratory. The PHL will
simultaneously re-label the samples with a STARHS identification number using labels
provided to the PHL by the CDC STARHS laboratory. The PHL must also provide the
ISC with a link between the STARHS identification number and the original specimen
accession number. The PHL will ship all samples to be tested using STARHS, labeled
only with the STARHS identification number, to the CDC STARHS laboratory according
to the procedures described in the document .
Role of State / Local HIV Incidence Surveillance Coordinator - The state / local ISC from
the jurisdiction where samples originated will determine the disposition of all samples
stored at the PHL and coordinate with the PHL to ensure that only specimens to be tested

Page 11
using STARHS are shipped to the CDC STARHS laboratory and all specimens that will
not be tested using STARHS are discarded. The ISC will also maintain the link between
the original specimen accession number and the STARHS number, and will manage the
STARHS results.
Role of CDC STARHS Laboratory - The CDC STARHS laboratory will test all samples
received from a PHL using the STARHS identification number. Once testing is
complete, the CDC STARHS laboratory will return results to the appropriate
jurisdiction’s ISC.
Specimen Numbering - Specimens will be stored at the PHL labeled with the original
laboratory-assigned specimen accession number. Once specimen disposition is determined,
each sample to be tested using STARHS will be assigned a unique STARHS identification
number by the PHL prior to shipment to the CDC STARHS laboratory. All subsequent
procedures use only the STARHS identification number.
Theoretical Laboratory Types for this Transportation Model - The laboratory types that
would best use this model are single-jurisdiction laboratories such as hospital, medical
center, university, small independent reference laboratories, or local branches of large
commercial laboratories. In many cases, these laboratories already have working
relationships and established procedures for submitting samples to their state and/or local
PHL and would prefer not to change their existing practices. All state / local PHL
performing confirmatory testing for HIV also fall into this category, except they would
simply hold samples until the ISC determines specimen disposition.
Note: The testing laboratory may choose either of the two transport models. With the
exception of public health laboratories, the examples listed in this section are merely
suggestions, not requirements, for the types of laboratories that may choose this model.
Responsibilities
Private Laboratories
Select a Model Type - Each private laboratory performing confirmatory testing of HIV
diagnostic specimens must select one of the two specimen transport model types and
inform the ISC which model was chosen. The private laboratory must send remnant sera
from confirmed HIV-seropositive samples to either the state / local PHL or to the CDC
STARHS laboratory.
Additional Laboratory Report Information - The private laboratory must include the
laboratory-assigned specimen accession number on the laboratory report form sent to the
HIV surveillance department, per state / local disease reporting requirements.
Many laboratories send EIA positive specimens to a reference laboratory for
confirmatory WB or IFA. In this case, care must be taken to ensure that the appropriate
specimen accession numbers are associated with the correct surveillance report.

Page 12
Public Health Laboratories
Sample Storage and Retention - The PHL will often serve a dual function as a testing
laboratory or a “pass through” facility for private laboratories. The PHL will store
(Section 4.3) all WB or IFA confirmed positive samples and/or all samples received from
private laboratories until sample disposition is determined by the ISC.
Aliquoting and Sample Shipment - Once sample disposition has been determined by the
ISC, the PHL will be responsible for pulling the identified samples, aliquoting samples
into the appropriate tubes, and re-labeling the samples with a STARHS identification
number for testing. The PHL will send the ISC the linkage information between the
laboratory-assigned specimen accession number and the STARHS identification number.
The PHL will ship all samples to be tested using STARHS (test list) to the CDC
STARHS laboratory and discard all samples on the toss list according to existing
laboratory procedures.
State / Local HIV Incidence Surveillance Coordinator
Sample Disposition - The ISC will determine sample disposition for all HIV seropositive
diagnostic samples tested in the jurisdiction. The ISC will coordinate with the PHL
and/or the CDC STARHS laboratory to ensure the proper samples are tested.
Data Management - The ISC will retain the linkage information between the laboratoryassigned specimen accession number and the STARHS identification number to ensure
that STARHS results can be matched to surveillance data. The ISC will send incidence
data to CDC on a monthly basis on or before the 15th of each month. If the ISC is in a
local jurisdiction, then results should be sent to the state ISC for matching purposes
before submitting data to CDC.

Page 13

CDC STARHS Laboratory
Sample Rejection Criteria - Sample rejection due to thawing, breakage, insufficient
quantity, or lost-in-transit will be determined and recorded by the CDC STARHS
laboratory.
Sample Storage and Retention - All samples that are shipped directly from a private
laboratory and not a state PHL must be stored (Section 4.3) at the CDC STARHS
laboratory until sample disposition is determined by the ISC. Storage time may vary
from state-to-state depending on the state’s surveillance practices. Once sample
disposition has been determined by the ISC, the CDC STARHS laboratory will be
responsible for pulling all samples on the test and toss lists from the ISC. The samples on
the toss list will be discarded. The samples on the test list will be aliquoted into the
appropriate tubes and re-labeled with a STARHS identification number for testing. The
CDC STARHS laboratory will apply a label to the sample tube and then to a line listing
of specimen accession numbers received from the ISC or PHL for those samples to be
tested using STARHS. The CDC STARHS laboratory will send the ISC the linkage
information and then destroy the linkage information held at the CDC STARHS
laboratory. All subsequent testing and results will only refer to the STARHS
identification number and will no longer include the original specimen accession number.
Result Reporting - The CDC STARHS laboratory will report STARHS results by the
STARHS identification number only to the ISC with jurisdiction over the sample as
designated by the .
The results reporting mechanism will adhere to the methods agreed upon between CDC
and the CDC STARHS laboratory.

References
1. CDC. HIPAA Privacy Rule and Public Health: Guidance from CDC and the U.S. Department
of Health and Human Services. MMWR, April 11, 2003; 52, 1-12. Available at
http://www.cdc.gov/mmwr/preview/mmwrhtml/m2e411a1.htm.

Page 14

Figure 1. Specimen Originates at National Commercial Laboratory or Private
Laboratory and is Sent Directly to the CDC STARHS Laboratory

NATIONAL
COMMERCIAL or
PRIVATE LAB
1. HIV Laboratory Report
with Specimen Accession #
(1234), Collection Date,
Other Identifiers, &
Laboratory Identifying
Information

(e.g., Specimen Accession # 1234)
2. Sends aliquots of all HIV+ samples
with Specimen Accession # (1234),
Collection Date, and Laboratory
Identifier

3. Contacts CDC-STARHS lab with lists of all
specimens to be tested using STARHS or to be
discarded by Specimen Accession #

STATE PUBLIC
HEALTH
SURVEILLANCE
PROGRAM

5. Links STARHS # & Accession # and then destroys
linkage at STARHS lab.

CDC
STARHS
LAB

(ISC)
6. Sends Results by STARHS # to ISC
4. STARHS Lab pulls and re-labels
specimens to be tested using
STARHS. STARHS # assigned to
Specimen Accession # at CDCSTARHS Lab. Sample tested by
STARHS #. STARHS Lab pulls and
discards specimens that will not be
tested using STARHS.

KEY:
= Specimens
= Specimen disposition information
= Laboratory report
= Linkage information
= Results

Page 15

Figure 2a. Specimen Originates at a Public Health Laboratory (PHL Performed the
Confirmatory HIV Testing)

PHL
with assigned Specimen #
(e.g. Specimen # 1234)
1. HIV Laboratory
Report with Specimen
# (1234) and Other
Laboratory Identifiers
2. Notifies lab
of which
specimens
should be
tested using
STARHS and
which should
be discarded by
4. Links Specimen # with
Specimen #
STARHS # for all specimens
shipped to CDC-STARHS lab.

STATE PUBLIC
HEALTH
SURVEILLANCE
PROGRAM

3a. Aliquots all samples to be tested using
STARHS into correct tubes.
3b. Assigns STARHS # and attaches labels
provided by CDC-STARHS lab.
3c. Ships only samples to be tested using
STARHS to CDC-STARHS Lab.
3d. Discards all samples that will not be tested
using STARHS according to existing laboratory
procedures.

CDC
STARHS
LAB
5. Sends Results by STARHS # to ISC

(ISC)

KEY:
= Specimens
= Specimen disposition inform
= Laboratory report
= Linkage inform
= Results

Page 16

Figure 2b. Specimen Originates at a Private Laboratory (for example, a University
Hospital Laboratory, Regional or Local Independent Commercial Laboratory) and Sends
Sample to State Public Health Laboratory (serves as a pass-through facility)
UNIVERSITY OR
PRIVATE LAB
Assigns Specimen Accession #

2. Ships aliquots of all HIV+ samples to
State PHL with testing lab information &
Specimen Accession # (1234)

PHL
(Repository for all HIV+ Sera)
Stores with Specimen
Accession #

(e.g., Specimen Accession # 1234)

(e.g., Specimen Accession # 1234)
3. Notifies PHL of
which specimens
should be tested using
STARHS and which
should be discarded
by Specimen
Accession #

1. HIV Laboratory
Report with Testing
Lab information,
Specimen Accession #
(1234), and Other
Laboratory
Identifiers

4a. Aliquots all samples to be tested
using STARHS into correct tubes.
4b. Assigns STARHS # and attaches
labels provided by CDC-STARHS lab.
4c. Ships only samples to be tested
using STARHS to CDC-STARHS Lab.
4d. Discards all samples that should not
be tested using STARHS according to
existing lab procedures.
5. Links Specimen
Accession # with
STARHS # for all
specimens shipped to
CDC-STARHS lab.

STATE PUBLIC
HEALTH
SURVEILLANCE
PROGRAM

CDC
STARHS
LAB

6. Sends Results by STARHS # to ISC

(ISC)

KEY:
= Specimens
= Specimen disposition inform
= Laboratory report
= Linkage inform
= Results

HICSB Incidence Surveillance
STARHS SPECIMEN SUBMISSION FORM
Please complete this form and send it with each shipment. Specimens should be sent to:
NYSDOH Wadsworth Center
Axelrod Institute
Diagnostic HIV Testing Lab: STARHS
120 New Scotland Avenue
Albany, NY 12208
SHIPPING FACILITY INFORMATION:

RESULTS SENT TO:

Name:____________________________________

Name________________________________________

Address:__________________________________

Address______________________________________

__________________________________

______________________________________

Phone Number:_____________________________

Phone Number:________________________________

Fax:______________________________________

Fax_________________________________________

Email:____________________________________

Email:_______________________________________

Contact Person:_____________________________

INCIDENCE SURVEILLANCE (HICSB) - List of eligible specimens sent separately

INCIDENCE TESTING FOR HIV DRUG RESISTANCE SURVEILLANCE (HRS)
( A S)
BEHAVIORAL SURVEILLANCE (BSCB)
EVALUATION OF DRIED FLUID SPOT SURVEILLANCE (DFS)
RANGE OF SPECIMEN NUMBERS SENT (OR ATTACH LIST):
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
Please identify any specimens above that are not collected under the Standard Procedures and should be tested
using the Vironostika HIV-1 Less Sensitive Assay.

Operational Public Health Laboratory Flow Chart for HIV Incidence Surveillance
Specimens handled and testing for HIV infection initiated
per standard laboratory protocol

EIA reactive

Remove one 1 ml aliquot;
freeze at -70ºC
(see VARHS guidance)*

EIA non-reactive

Specimens handled per
standard laboratory protocol

Proceed with HIV testing

Confirmatory HIV test (+)

Confirmatory HIV test (-)

Notify Incidence Surveillance Coordinator (ISC) and await
further guidance

Receive test list from ISC

Remnant specimens from
new cases in HARS to be
tested using STARHS

Receive toss list from ISC

Remnant specimens from cases
previously reported to HARS not to
be tested using STARHS

Remove one, 0.5 ml aliquot
per specimen

Batch and send to STARHS
laboratory for testing

*Applies only to laboratories participating in VARHS

Standard
laboratory
HIV testing
procedures.
Unrelated to
HIV Incidence
Surveillance

Epidemiologic Flow Chart for HIV Incidence Surveillance

All persons with a confirmed positive
HIV-diagnostic blood test result

Was the test from a
previously reported case
in HARS?

NO
Test using STARHS

YES
Incidence Surveillance Coordinator (ISC)
adds specimen to test list by lab-assigned
accession number, depending on specimen
storage location

YES

Was the
specimen collected
within 3 months of an
unavailable diagnostic
specimen?

NO
Where is the
specimen
Public Health located? CDC STARHS
Laboratory
Laboratory
(PHL)
(CSL)
ISC sends test
list to PHL

ISC sends test
list to CSL

Lab aliquots
specimen and
assigns a STARHS
ID (SID)

Lab aliquots
specimen and
assigns a STARHS
ID (SID)

PHL sends SID to
health department,
batches and sends
specimens to
STARHS Lab

STARHS
performed;
results sent to
health department

Lab sends SID to
health department

STARHS
performed;
results sent to
health department

Do not test using STARHS

Incidence Surveillance Coordinator (ISC)
adds specimen to toss list by lab-assigned
accession number, depending on specimen
storage location

Public Health
Laboratory
(PHL)

Where is the
specimen
located?

CDC STARHS
Laboratory
(CSL)

ISC sends toss
list to PHL

ISC sends toss
list to CSL

Specimens
discarded;
not shipped
to STARHS
laboratory

Specimens
discarded

Training and Certification for Shipping Infectious Substances

FedEx 800-GO-FEDEX 3 day IATA based training Covers all hazardous materials Cost is $550

Saf-T-Pak 800-814-7484 Specifically for infectious and diagnostic substances, and dry ice 3 options--One day seminar, On-site programs, or Interactive CD For interactive CD: for one sitting, can be
done in 3-5 hours Certificate good for 2 years OR until regulations change Cost is ~$250

Viking Packaging (Oklahoma) 800-788-8525—David Weilert Seminars monthly in Tulsa/ $300 per
person Covers all nine classes of hazardous materials Covers shipping under IATA Certificate
good for 2 years Will do group classes in local area---$3,000 plus travel costs

These are some companies that provide training for dangerous goods shipping. The Centers for
Disease Control and Prevention does not endorse any particular company.

Page 1
Guidance for Processing, Storage, and Shipping of Specimens
to the CDC STARHS Laboratory
(Revised: 8/2006)
Purpose
This standard operating procedure describes methods for the handling, storage, and shipping of serum
specimens that will be tested for recent HIV-1 infection using STARHS. Results from these tests will
help estimate HIV incidence.
Introduction
Remnant serum from positive HIV diagnostic specimens is to be collected and frozen using vials and
labels specified or supplied by the CDC STARHS laboratory. Ideally, 0.5 ml should be collected for
each aliquot. Frozen serum will be shipped to the CDC STARHS laboratory for testing.
CDC STARHS Laboratory
The CDC STARHS laboratory is the Wadsworth Center Retroviral Immunology Diagnostic HIV
Testing Laboratory which is part of the State of New York Department of Health. Frozen aliquots will
be shipped to:
NYSDOH Wadsworth Center
Axelrod Institute
Diagnostic HIV Testing Lab: STARHS
120 New Scotland Avenue
Albany, New York 12208
Attn: Brian Granger
Setting and personnel for specimen processing
•

Centrifugation, aliquoting, and shipping should be performed at or under the auspices of a
laboratory that is CLIA-certified for handling HIV+ specimens.

•

All personnel handling specimens should receive blood borne pathogens training. See the
Occupational Safety and Health Association (OSHA) Occupational Exposure to
Bloodborne Pathogens Standard:
http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=10051.

•

Personnel handling or processing specimens should have appropriate laboratory training in the
relevant laboratory techniques for handling HIV+ specimens and for performing the specific
tasks required.

•

The setting in which centrifugation, aliquoting, and shipping occurs should meet Biosafety level
2 specifications required by the U.S. Department of Health and Human Services for handling of

Page 2
specimens containing HIV:
o [http://www.cdc.gov/od/ohs/biosfty/bmbl4/bmbl4toc.htm, Biosafety in Microbiological and
th
Biomedical Laboratories (BMBL) 4 Edition (4th%20BMBL.pdf), pages 20-27, 171-175].

Materials
•

Cryogenic vials - Supplied by CDC STARHS Laboratory

•

Specimen labels - Supplied by CDC STARHS Laboratory – Label will identify sample (barcode,
number, etc) by STARHS identification number

•

Cardboard storage boxes for cryogenic vials – Can be supplied by CDC STARHS Lab if
requested

•

Freezer—STARHS samples can be refrigerated at 2-8oC, but for long term storage and shipping,
samples should be frozen at -20oC
o It is recommended that, if not already in practice, a daily temperature log be kept to ensure
the freezer is operating properly
o The freezer should be housed in a location with proper ventilation to avoid overheating and
freezer failure.
o Staff must be certain there is adequate space in freezer to store specimens.

•

A supply of dry ice in pellet form

•

Insulated shipping containers certified to ship frozen diagnostic specimens (HIV+ serum and dry
ice)

•

Shipping courier air bills

•

Materials for shipper packing—See Packing procedures for shipping to the CDC STARHS
laboratory later in this document

•

Specimen Collection and Processing
All processing of specimens should be done by personnel qualified to handle HIV+ specimens under the
auspices of a laboratory equipped for the handling of HIV + specimens
[http://www.cdc.gov/od/ohs/biosfty/bmbl4/bmbl4toc.htm, Biosafety in Microbiological and Biomedical
th
Laboratories (BMBL) 4 Edition (4th%20BMBL.pdf), pages 20-27, 171-175].
•

Aliquot the serum (0.5 ml per cryogenic vial). Use labels to identify the specimen and record
this information in the proper setting. (Specimen log for eventual transfer to HIV Incidence
Surveillance database.)

•

Store aliquots in refrigerator or freezer until specimen disposition has been determined and
scheduled shipping date has arrived.

Page 3
Shipping
•

Specimens for STARHS should be sent to CDC STARHS laboratory using the address written
on the and in the “CDC
STARHS Laboratory” section in this document. All specimens will be shipped as diagnostic
specimens using IATA (International Air Transport Association) Packing Instructions 650. Dry
ice will be included with each shipment using IATA Packing Instructions 904.
o Because samples will be shipped using dry ice, shipping personnel must be trained and
certified to ship dangerous goods. See Appendix 7.1.B for a list of companies that provide
training.
o Establish contact with Lea N’ko Ali-Napo ([email protected]) at the CDC STARHS
laboratory.
o Ensure that adequate STP320 or equivalent shipping containers are available. The CDC
STARHS Laboratory will return them to the submitting laboratory if a return air bill is
included in the shipment. The shippers are expensive and need to be re-used.
o Ensure that you have an adequate supply of shipping courier air bills which can be obtained
free of charge from most couriers.

•

Packing procedures for shipping to the CDC STARHS laboratory - It is recommended that all of
these steps are read and understood before starting the preparation of the actual shipment
o Bring the STP320 shipper or equivalent that is to be used for the shipment and materials
needed for packing the specimens into the area in which the shipment will be prepared.
o If the shipper is new and being used for the first time, check to be sure that it includes the
following items:
ƒ

Two (2) sheets of bubble wrap

ƒ

Two (2) STP 710 or equivalent certified secondary containers

ƒ

Two (2) 250 ml absorbent strips

ƒ

Class 9 label and dry ice quantity label

ƒ

Other hazard and handling labels

ƒ

One (1) instruction sheet

o For a diagram of the above contents, refer to the SaftPak catalog.
o Use only what is needed of the above contents for each individual shipment. Save left over
supplies for future shipments.
o If the shipper is being re-used, the proper labels will already be in place on the outer
cardboard container. Ensure that adequate supplies of the other materials listed above are on
hand.
o Put on personal protective equipment.
o Remove cryogenic vials from freezer and accurately record the specimen accession or
STARHS identification numbers. The specimen numbers can either be written directly onto

Page 4
the STARHS Specimen Submission Form or on a separate list that will be attached to the
form. Return them to the freezer. Repeat the process until all specimen numbers have been
recorded for each vial that is going to be shipped.
o These specimens should remain frozen at all times and therefore should not be removed
from a frozen environment for more than a few minutes.
o Prepare 3 copies of the STARHS Specimen Submission Form (Appendix 7.1.A) — listing or
attaching the specimen number on each vial to be shipped.
ƒ

Copy 1 (original) will be sent with the specimens in the shipment.

ƒ

Copy 2 should be mailed to ISC as notification of shipment.

ƒ

Copy 3 will be retained by the submitting laboratory for their records.

o If possible, on a floppy disk or CD, also include an encrypted electronic version of the list of
specimen numbers in the shipment. This will help the CDC STARHS laboratory minimize
the amount of data entry they have to do when logging in the samples, thereby minimizing
errors.
o Prepare the Shipping courier air bill that the CDC STARHS laboratory will use to return the
shipper back to the submitting laboratory for re-use. The air bill MUST be completely filled
in with the return address, the CDC STARHS laboratory address, and the proper billing
number.
o If dry ice is in another location which requires leaving the area in which the shipment is
prepared, use a separate container to bring the dry ice that is needed for shipping back into
the shipping area at this time.
o Bring the specimens to the area in which the shipment is prepared. Work quickly, keeping in
mind that these specimens should remain frozen at all times and therefore should not be
removed from freezing temperature environment for more than a few minutes.
o Re-check the screw-cap lids on the specimen vials—tighten if necessary.
o Place the specimens into the secondary leak-proof container and make sure samples are
surrounded by bubble wrap and absorbent strips. The vials should not move around or rattle
inside the vessel.
o Place the secondary vessel into the inner box and place the inner box into the polystyrene
cooler.
o Pack dry ice pellets in the shipper and around the inner box. The STP320 shipper will hold
~8 kg of dry ice (~10lbs) and, if packed completely, will keep the contents frozen for greater
than 80 hours.
o DO NOT PUT DRY ICE INSIDE THE INNER BOX
o Place the lid on the polystyrene cooler
o Place one copy of the completed on top of the shipping box return form with the completed return FedEx
air bill stapled to it. Fold in half and place on top of the polystyrene lid.
o Fold over the top flaps and seal the shipping container with clear shipping tape.

Page 5
o The outer box must have a mark in the form of a square set at an angle of 45o (diamond
shaped). The mark must be at least 2 inches by 2 inches and include the UN 3373
designation. The proper shipping name “Diagnostic specimens” must be marked on the outer
package adjacent to the diamond shaped mark. Labels can be purchased to place on the outer
box that fulfill this requirement.
o Apply the Class 9 Hazard Label over the lower diamond shaped outline on the box.
o Apply the net quantity dry ice label to the outlined area adjacent to the Class 9 Hazard Label.
Write the approximate amount (in kg) of dry ice you used to pack the container.
o Prepare the shipping courier paper work addressed to the CDC STARHS laboratory. Select
the overnight shipping option.
o Call or email the CDC STARHS laboratory to notify them of the shipment. Provide the CDC
STARHS laboratory with the shipment tracking information and the total number of samples
in the shipment. Note: Do not fax or email laboratory-assigned specimen accession
numbers or STARHS identification numbers.