Estimated HIV Prevalence in the United States at the end of 2003 ?????2003??HIV???

1
Estimated HIV Prevalence in the United States at
the end of 2003?????2003??HIV???

• M. Kathleen Glynn, DVM, MPVM
• Philip Rhodes, PhD
• Division of HIV/AIDS Prevention
• National Center for HIV, STD, TB Prevention
• 2005 National HIV Prevention Conference
• June 14, 2005
• 2005?6?14?2005???HIV????

2
Outline??

• Background??
• Methods??
• Results??
• Prevalence estimates 2003, 2003?????
• Comparison with estimates derived through
  previous methods ??????????
• Limitations???
• Conclusions/Recommendations??/??

3
The Spectrum of HIV InfectionHIV???

HIV??
???HIV
HIV??
???AIDS
?HIV/AIDS??
HIV??(???)
????
4
The Spectrum of HIV Infection HIV???

HIV??
HIV??
???AIDS
?HIV/AIDS??
???HIV
HIV??(???)
??
5
The Spectrum of HIV Infection HIV???

HIV??
???HIV
HIV??
???AIDS
?HIV/AIDS??
HIV??(???)
????
6
Background??

• National AIDS case surveillance has been
  conducted throughout the United States since the
  1980s
• 19??80?????????AIDS??
• Surveillance data can be used to directly
  generate an accurate picture of the AIDS epidemic
  
• ????????????????????
• National HIV (not AIDS) surveillance does not yet
  exist
• ?????????HIV(?AIDS)???
• At the end of 2003, only 35 states had
  implemented confidential name-based HIV infection
  surveillance
• ?2003??,?35????????HIV??
• We must estimate the extent of the national
  diagnosed HIV (not AIDS) epidemic???????????HIV(?A
  IDS)?????
• Surveillance data describe diagnosed population,
  estimation necessary for undiagnosed
  ???????????????,?????????????

7
Graphic of back calculation??????
8
Graphic of back calculation??????
9
Graphic of back calculation??????
10
Graphic of back calculation??????
11
Graphic of back calculation??????
12
Background??

• Previous HIV prevalence estimates
• ???HIV????
• End of 1992, 650,000 900,000 (Karon
  JAMA1996276126) 1992??
• End of 1998, 800,000 900,000 (CDC MMWR 1999
  RR19) 1998??
• End of 2000, 850,000 950,000 (Fleming, CROI
  2002) 2000??

13
Background??

• Back-calculation based on AIDS incidence stopped
  mid-1990s ??AIDS??????,??90????
• Effects of therapy on incubation distribution
  (ZDV, PCP prophylaxis, HAART)
• ???????????(ZDV, PCP??, HAART)
• Effect on demographics/risk group on incubation
  distribution????/?????????????
• Changes in AIDS case definition AIDS???????
• Back-calculation based on initial HIV detection
  (HIV with or without AIDS) shown to produced
  estimates of greater precision ?????HIV(??????)???
  ???,???????????

14
Objectives??

• To estimate HIV prevalence in U.S.
  adults/adolescents at the end of 2003 (50 states
  and DC) ???2003??????/?????HIV???
• To estimate the proportion of persons living with
  HIV who may be undiagnosed?????HIV????????????
• To characterize the estimated persons living with
  HIV by demographic and risk groups????????????????
  ??

15
Methods??

• Back calculation using HIV detection data (New
  Method)
• ?HIV??????????(??)
• Extrapolation of existing back calculation
  estimates (Previous Method)
• ???????????(??)

16
New Method Time to First Detection of HIV
Infection?? ?????HIV???

• Data source National HIV/AIDS Surveillance data
  reported to CDC through June 2004
• ???? 2004?6?????CDC???HIV/AIDS????
• HIV/AIDS Cases HIV/AIDS??
• Persons gt13 years old, diagnosed through 2003
• 2003?????,??13???????
• 30 states with established HIV surveillance
  systems
• 30?????HIV????
• AIDS Cases AIDS ??
• Persons gt13 years old, diagnosed through 2003
• 2003?????,??13???????
• 50 states and District of Columbia 50???DC
• Data elements Year and state of HIV/AIDS
  diagnosis, Disease status at initial HIV
  diagnosis, Sex, Race/Ethnicity, HIV Transmission
  Category
• ???? - HIV/AIDS?????????,????HIV?????,??,
  ??,HIV????

17
New Method Time to First Detection of HIV
Infection?? ?????HIV???

• Detection of HIV infection HIV?????
• Calendar year of first HIV positive test date
• ?????HIV?????
• Disease status at detection HIV only, or HIV
  with AIDS (diagnosed within same calendar year)
• ???????? ????????
• Types of Parameters Estimated ???????
• Number of HIV infections by time period,
• ??????HIV????
• Detection probabilities for HIV only status
• ?????????
• Estimated proportion in disease status and
  demographic and risk group
• ??????,???????????????

18
New Method - Structure of Extended Back
Calculation Model?? ??????????Stepwise
Structure Example ??????
19
Previous Method - Extrapolation of existing
estimates?? ????????

• Overall HIV prevalence extension of 2000 CDC
  estimates
• ????HIV??? 2000?CDC??????
• Diagnosed AIDS cases from national AIDS
  surveillance data through Dec. 2003
• ???AIDS??? ????2003?????AIDS????
• Diagnosed HIV cases - Poisson regression models
  based on data from states with mature, stable
  systems for HIV reporting
• ???HIV??? ??????????HIV??????????????poisson
  ????
• Undiagnosed HIV prevalence Overall prevalence
  - Diagnosed HIV prevalence AIDS Prevalence
• ???HIV??? ????HIV??? ???HIV??? AIDS
  ???

20
Results Estimated HIV Prevalence?? HIV?????
21
Results - Estimated HIV Prevalence?? HIV?????
22
Results - Estimated HIV Prevalence, United
States, 2003?? - 2003???HIV?????
Total HIV Prevalence???? 1,039,000 1,185,000
23
Estimated HIV prevalence by Sex, United States,
20032003???HIV????? ????
24
Estimated HIV Prevalence by Race/Ethnicity,
United States, 2003 2003???HIV????? ????
25
Estimated HIV Prevalence by Transmission
Category, United States, 2003 2003???HIV?????
????
26
Estimated HIV Prevalence by Disease Status and
Selected Characteristics, United States, 2003
2003???HIV????? ???????
27
Distribution of Diagnosed and Undiagnosed HIV
Infections among Estimated Persons Living with
HIV, by Sex, United States, 2003
2003???????????????????
28
Distribution of Diagnosed and Undiagnosed HIV
Infections among Estimated Persons Living with
HIV, by Race/Ethnicity, United States, 2003
2003???????????????????
29
Distribution of Diagnosed and Undiagnosed HIV
Infections among Estimated Persons Living with
HIV, by Transmission Category, United States,
20032003???????????????????
30
Comparisons of Estimated HIV Prevalence, United
States, 2003New Method and Previous Method
2003????????????
31
Comparisons of Estimated HIV Prevalence, United
States, 2003New Method with Previous Method
2003????????????
32
Limitations of 2003 HIV Prevalence
estimates2003?HIV??????????

• Estimates and ranges are dependent on underlying
  surveillance data
• ????????????????
• Surveillance data subject to local variations in
  reporting practices, completeness, timeliness
• ??????????????????????
• Persons who may have tested anonymously, and
  therefore are not reported to surveillance as
  diagnosed, may represent a portion of those
  estimated to be undiagnosed
• ??????????,??????????????,??????????(??)
• Important Assumptions ????
• Detections dates for simultaneous HIV/AIDS are
  complete
• HIV/AIDS?????????
• Standardization procedures for demographic and
  risk groups consistent across geographic areas
• ????????????????????????????

33
Conclusions??

• Approximately one million adults/adolescents were
  living with HIV/AIDS in the U.S. at the end of
  2003
• ?2003??,????1????/????HIV/AIDS
• About one-fourth of infected persons have not
  been diagnosed
• ?1/4????????
• Growing population in need of prevention and
  treatment services
• ?????????????

34
Conclusions??

• Black non-Hispanics represent almost half of all
  persons living with HIV infection
• ???????????????
• MSM remain most affected risk group, high risk
  heterosexuals also affected
• MSM??????????,????????????
• Undiagnosed population has higher proportion of
  Black non-Hispanics, high risk heterosexuals,
  than diagnosed population
• ????HIV???????????????????????????

35
Recommendations??

• Increased HIV testing early in the course of
  disease and programs to link diagnosed persons to
  on-going care and prevention services are needed
• ?????????,?????????????????????
• Comprehensive national HIV surveillance data are
  critical ????????????
• Would improve estimates of HIV prevalence
• ???HIV?????
• Would allow better monitoring the epidemic
• ????????
• Continue efforts to fully implement national
  incidence surveillance to provide annual
  incidence estimates
• ??????????????,??????????

36
Acknowledgements??

• HIV/AIDS Surveillance staff in state and local
  health departments ??????????HIV/AIDS?????

• Previous Methods for Estimating HIV Prevalence
• ?????????
• Patricia Fleming
• Robert Byers
• Patricia Sweeney
• Danni Daniels
• John Karon
• Robert Janssen

• Applying Previous Methods to Update
  Estimates???????????
• Robert Byers
• Tebitha Kajese