Beginning in the early 1990s, the reporting of HIV infection case data, as an extension of AIDS case surveillance, began to be standardized. These data included CD4+ T-lymphocyte (CD4) results (especially the earliest result after HIV diagnosis), viral load test results, presence of AIDS-defining opportunistic illnesses, demographics, and information on transmission risk factors. By the end of 1993, all state, local, and territorial areas reporting HIV and AIDS case surveillance data to CDC were required to use a CDC-supported data collection practices and software system for data entry, record keeping, and case reporting. This software enabled the maintenance and reporting of as many as 20 CD4 and 9 HIV viral load test results per person.

State laws requiring laboratories to report CD4 and viral load results differ by state. By mid-2003, approximately 24 of the 33 states whose data are used in this report had some form of required reporting of CD4 test results by laboratories; and 29 of the 33 states had some form of required viral load reporting by laboratories (unpublished CDC data, 2004). In recognition of the contribution that laboratory reporting has made to HIV/AIDS surveillance, the Council of State and Territorial Epidemiologists recommended in 2004 that all states require laboratories to report all levels of CD4 results and both detectable and nondetectable results of viral load testing to state public health departments [1]. By January 2005, 29 of the 33 states had enacted some form of CD4 reporting (see Table 14). From 2003 through January 2005, no additional states included in this report added mandated viral load reporting.

Estimates of the proportion of persons without reported CD4 or viral load results should be interpreted with caution. The lack of laboratory test results reported to the national surveillance system does not mean that a person did not receive testing but means that for a variety of reasons, the results may not have been communicated to the local HIV/AIDS surveillance program or were not entered in the surveillance database that is transmitted to CDC. Because of the use of CD4 results in the AIDS definition, the historical surveillance prioritization of this stage of disease, and the likelihood that persons with AIDS will have symptoms that prompt clinical and laboratory evaluation for treatment, missing CD4 results are less common among persons with AIDS.

Persons without reported CD4 results after HIV diagnosis represent 2 distinct groups: (1) persons for whom this laboratory information is lacking because of surveillance practices and (2) persons for whom care and prevention services have failed. Surveillance practices differ among HIV reporting areas and can affect the completeness of case information. AIDS-defining CD4 results are prioritized over CD4 results obtained early in HIV infection, which may account for the missing CD4 results among persons who have HIV without AIDS. And, although the required level of CD4 reporting (e.g., fewer than 200 cells/µL compared with all levels of CD4 reporting) has not been found to affect timeliness or range of CD4 test results, multiple surveillance factors, including the burden of disease and health department staffing and resources, are probably involved in complete CD4 result reporting (unpublished CDC data, 2004). Persons for whom care and prevention services have failed include persons who have not been linked to health care, who have been lost to follow-up (e.g., moved out of the jurisdiction of the reporting HIV/AIDS surveillance program, died out of state), who don’t know they are HIV infected (e.g., never returned for HIV test result), who did not receive the recommended evaluation of CD4 status during follow-up, as well as those who may be in denial about their HIV-infected status.

Surveillance of CD4 Test Results

CD4 test results may include CD4 count or percentage or both. Although CD4 count is more widely used to monitor disease and recommend the initiation of antiretroviral treatment, CD4 percentage alone can be reported, and in some instances, especially after a case meets the immunologic criteria for AIDS, it is the only lab result that is reported after HIV diagnosis. In an effort to maximize the interpretability of CD4 results, the earliest CD4 percentages were interpreted within a category of CD4 counts or as discrete CD4 counts.

A random sample of 150,000 persons aged 13 years and older was obtained from CDC’s national surveillance database of all HIV/AIDS cases diagnosed through 2004. The earliest values for paired count and percentage (CD4 counts truncated at a maximum value of 1,500 cells/µL) were used to construct a nonlinear regression model that included additivity and variance stabilization (AVAS). The AVAS procedure transforms the independent and dependent variables to produce an additive model with constant residual variance. Percentage correlates for various count thresholds were obtained by using the AVAS regression results. From this regression, the following percentage cut-points were obtained and used in the analysis whenever the earliest CD4 test value was a percentage without a count. These imputations were used for Tables 1–7.

Of the paired CD4 count and percentage data used in this regression model, 78% were obtained from males. A gender difference was observed: for any given CD4 count, the resulting percentages were higher for females than for males. This phenomenon has been reported before [2]. Although the CD4 percentages used in the imputation for CD4 count categories were higher than those reported by others [3–5], the effect on this analysis is limited. Approximately 80 (less than 1%) of 30,000 annual cases had a CD4 percentage of 14 or higher and were affected by this imputation.

Surveillance of HIV Infection

In addition to the reporting of persons with AIDS, this report is focused on case reports from 33 states that had laws or regulations requiring confidential name-based HIV infection reporting of adults and adolescents with confirmed HIV infection that had not progressed to AIDS. After the removal of personal identifying information, these reports were submitted to CDC from Alabama, Alaska, Arizona, Arkansas, Colorado, Florida, Idaho, Indiana, Iowa, Kansas, Louisiana, Michigan, Minnesota, Mississippi, Missouri, Nebraska, Nevada, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, Virginia, West Virginia, Wisconsin, and Wyoming. The implementation of HIV reporting has differed from state to state. Before 1991, the surveillance of HIV infection that had not progressed to AIDS was not standardized, and the reporting of HIV infection was based primarily on passive surveillance. The information on many cases reported before 1991 is not complete.

Data on HIV infection should be interpreted with caution. HIV surveillance reports may not be representative of all persons infected with HIV because not all infected persons have been tested. Many states with confidential name-based HIV reporting offer anonymous HIV testing, and home-collection HIV test kits are widely available in the United States. The results of anonymous tests are not reported to the confidential name-based HIV registries of state and local health departments. Therefore, reports of confidential test results may not represent all persons who tested positive for HIV infection. Furthermore, many factors, including the extent to which testing is routinely offered to specific groups, may influence testing patterns and the availability of, and access to, medical care and testing services. These data provide a minimum estimate of the number of persons known to be HIV infected in states with confidential name-based HIV infection reporting.

For this report, cases in adults and adolescents were classified by using the 2000 revised HIV surveillance case definition, which incorporates positive test results or reports of a detectable quantity of HIV nucleic acid or plasma HIV RNA [6]. Additionally, cases were restricted to those in persons who resided in, and were reported from, states with confidential HIV reporting at the time of HIV diagnosis.

HIV diagnosis

The term HIV diagnosis refers to the earliest month and year of a documented HIV-positive lab result or a physician’s diagnosis. An earlier, self-reported date of an HIV-positive result, such as a date from anonymous testing, is not captured in the national HIV/AIDS surveillance system. As HIV-infected persons move from state to state, information about the earliest documented HIV test should be recorded in the HIV registry in the person’s current state of residence.

HIV without AIDS

Persons with HIV infection may be tested at any point on the clinical spectrum of disease; therefore, the time between the diagnosis of HIV infection and the diagnosis of AIDS may differ. For this report, a 12-month follow-up from HIV diagnosis was used. HIV without AIDS was defined as no evidence of progression from HIV infection to AIDS within 12 months after HIV diagnosis. In addition, because surveillance practices differ, the reporting and updating of persons’ clinical and vital status differ among states. Completeness of HIV case reporting is estimated at more than 85% [7]. By matching data in the national surveillance database, CDC estimates that approximately 2%–8% of the cases of HIV infection that had not progressed to AIDS are duplicates.

Surveillance of AIDS

Although all 50 states, the District of Columbia, U.S. dependencies, possessions, and associated nations report AIDS cases to CDC by using a uniform surveillance case definition and case report form, this report is focused on case reports from 33 states. Because some tables include HIV-infected persons in whom HIV infection progressed to AIDS within 12 months after HIV diagnosis, those analyses were limited to the 33 states also used for HIV infection reporting. Furthermore, AIDS cases were restricted to those in persons who resided in, and were reported from, states with confidential name-based HIV infection reporting at the time of AIDS diagnosis.

The original AIDS definition was modified in 1985 and 1987 [8, 9]. The case definition for adults and adolescents was modified again in 1993 [3, 10]. In the revisions, a broader range of AIDS-indicator diseases and conditions was incorporated, and the results of HIV diagnostic tests were used to improve the sensitivity and the specificity of the definition. Effective January 1, 2000, the surveillance case definition for HIV infection was revised to reflect advances in HIV virologic tests. The reporting criteria for HIV infection and AIDS were incorporated into a single case definition for adults and children [6].

For persons with laboratory-confirmed HIV infection, the 1987 revision incorporated encephalopathy, wasting syndrome, and other indicator diseases that are diagnosed presumptively (i.e., without confirmatory laboratory evidence of opportunistic illness). In addition to the 23 clinical conditions in the 1987 definition, the 1993 case definition for adults and adolescents included HIV infection in persons with CD4 counts of fewer than 200 cells/µL or a CD4 percentage of less than 14 and a diagnosis of pulmonary tuberculosis, recurrent pneumonia, or invasive cervical cancer. For adults and adolescents, the 2000 revised HIV surveillance case definition incorporated positive test results or reports of a detectable quantity of HIV nucleic acid or plasma HIV RNA.

Although the completeness of reporting of AIDS cases to state and local health departments differs by geographic region and patient population, studies conducted by state and local health departments indicate that the reporting of AIDS cases in most areas of the United States is more than 85% complete [11–14]. In addition, multiple routes of exposure, opportunistic illnesses diagnosed after the initial AIDS case report was submitted to CDC, and vital status may not be determined or reported for all cases. However, among persons reported as having AIDS, the reporting of deaths is estimated to be more than 90% complete [15]. By matching data in the national surveillance database, CDC estimates that approximately 3%–4.5% of AIDS cases are duplicates.

Since January 1, 1994, CDC has not accepted AIDS case reports that meet only the laboratory-based immunologic criteria of the 1993 expanded surveillance case definition [3] if information on sex or race/ethnicity is missing. A small number of cases previously reported to CDC without those variables have been returned to the health departments for follow-up and have been deleted from the totals. Included in this report are persons known to be infected with HIV type 2 [16].

HIV with AIDS and HIV to AIDS

Over time, HIV infection may progress to AIDS and be reported to surveillance. Persons with HIV infection who are later reported as having AIDS are categorized under AIDS to reflect the most current information. For this report, because of the 12-month follow-up after HIV diagnosis, which was used to assign persons to a HIV disease category, AIDS cases comprise (1) HIV with AIDS (i.e., diagnoses of HIV infection and AIDS were made during the same calendar month) and (2) HIV to AIDS (i.e., diagnosis of AIDS was made 1–12 months after HIV diagnosis).

Tabulation and Presentation of Data

In all sections of this report, data have been statistically adjusted to correct for delays in the reporting of cases; unreported risk factors, or transmission categories, have been statistically redistributed to better present the trends in the epidemic and the distribution of risk factors among affected populations. To assess trends in diagnoses or prevalence, it is preferable to use adjusted data, presented by year of diagnosis instead of year of report. In section 1 (Tables 1–3), the earliest CD4 counts within 12 months after diagnosis are reported by year of HIV diagnosis and demographic characteristics of persons with HIV/AIDS. In section 2 (Tables 4–6), the earliest CD4 count is presented by year of diagnosis and HIV disease category. In section 3 (Tables 7 and 8), the lowest prevalent CD4 test result for persons who were alive through December 2004 and a longitudinal view of the first (incident) CD4 test result for cases of HIV infection diagnosed in 2001 are reported. In section 4 (Tables 9a, 9b, and 10), viral load and other findings in addition to CD4 results are reported by year of HIV diagnosis and HIV disease category. In section 5 (Tables 11–14), descriptive demographic information, by year of diagnosis and HIV disease category, are presented in addition to information about CD4 and viral load reporting in the United States.

CD4 or viral load date

Specimens for lab tests may be drawn or collected on one day (collection date) and tested on another day. For this report, the collection date (month and year) was used to establish the time from HIV diagnosis to lab test (in most instances, within 12 months after HIV diagnosis). Because of the lag between specimen collection and testing, it is possible for a specimen to be collected within 12 months after HIV diagnosis, but for the specimen to be tested more than 12 months after HIV diagnosis. Phrases such as “count within 12 months after HIV diagnosis” and “test performed within 12 months after HIV diagnosis” reflect the fact that the specimen was collected within 12 months after HIV diagnosis.

Year of diagnosis and year of lab test were required for these analyses. When month was missing, a midyear point was used to complete the assessment of a CD4 test result obtained within 12 months after HIV diagnosis; it was also used (Table 8) to examine the incident CD4 test result after HIV diagnosis.

Age groups

In most tables, age groups of persons with a diagnosis of HIV infection, with or without AIDS, are based on the person’s age at the time of the earliest documented positive HIV test result or a physician’s diagnosis. The age category for adults and adolescents comprises persons aged 13 years and older. For Table 7, the age groups of persons living with HIV infection or with AIDS are based on the person’s age as of December 31, 2004. Also in Table 7, the lowest CD4 count was included only if the CD4 test was performed when the person was at least 13 years old.

Race and ethnicity

In the Federal Register for October 30, 1997 [17], the Office of Management and Budget announced the Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity, also known as Statistical Policy Directive 15. These standards, which superseded the 1977 standards, reflect a change in federal policy regarding the collection of race and ethnicity data; implementation by January 1, 2003, was mandated. At a minimum, information on the following race categories should be collected: American Indian or Alaska Native; Asian; black or African American; Native Hawaiian or Other Pacific Islander; and white. Additionally, systems must have the capacity to retain information when multiple race categories are reported. Two ethnicity categories should be collected regardless of race: Hispanic and not Hispanic.

Because data for this report were compiled from reports to CDC through June 2005, race and ethnicity information may have been collected under 2 systems. The race and ethnicity categories in the system used through December 2002 have been maintained in this report because most were submitted under that system. Persons who reported multiple racial categories or whose race was unknown have been included in the total numbers in Tables 1–3, 7, and 11–13. Also, persons reported as non-Hispanic may include persons whose ethnicity was not reported.

Tabulation of persons living with HIV/AIDS

Tabulations of persons living with HIV infection or AIDS at the end of 2004 (Table 7) include persons who were reported as alive or whose vital status was missing or unknown as of the last update of the data. Persons in whom HIV infection progressed to AIDS remained categorized as having AIDS even if their CD4 count increased to more than 200 cells/µL at the end of 2004. Table 7 thus represents the lowest documented CD4 count among persons living with HIV/ AIDS. For persons living with HIV without AIDS (Table 8), data on the time from diagnosis to first CD4 test were tabulated through June 2005. Table 8, like Table 7, includes data on persons who were reported as alive or whose vital status was missing or unknown as of the last update of the data. Caution should be used in interpreting these data because states differ in how they review the vital status of persons reported with HIV infection or with AIDS. In addition, some persons may be lost to follow-up.

Transmission categories

For surveillance purposes, cases of HIV/AIDS are counted only once in a hierarchy of transmission categories. Persons with more than 1 reported mode of exposure to HIV are classified in the transmission category listed first in the hierarchy. The exception is male-to-male sexual contact and injection drug use, which makes up a separate transmission category.

Persons whose transmission category is classified as male-to-male sexual contact include men who report sexual contact with other men (i.e., homosexual contact) and men who report sexual contact with both men and women (i.e., bisexual contact). Persons whose transmission category is classified as heterosexual contact are persons who report specific heterosexual contact with a person with, or at increased risk for, HIV infection (e.g., an injection drug user).

Adults and adolescents who were born in, or who had sex with someone born in, a country where heterosexual transmission was believed to be the predominant mode of HIV transmission (formerly classified as Pattern II countries by the World Health Organization) are no longer classified as having heterosexually acquired AIDS unless they meet the criteria stated in the preceding paragraph. Similar to other cases among persons who were reported without behavioral or transfusion risks for HIV infection, these cases are now classified (in the absence of other risk factor information that would classify them in another transmission category) as “no risk reported or identified” [18]. Cases in children whose mother was born in, or whose mother had sex with someone born in, Pattern II countries are now classified (in the absence of other risk information that would classify them in another transmission category) as “Mother with, or at risk for, HIV infection: has HIV infection, risk not specified.”

Cases in persons with no reported exposure to HIV through any of the routes listed in the hierarchy of transmission categories are classified as “no risk reported or identified.” No identified risk factor (NIR) cases are cases that have been followed up by local health department officials; cases in persons whose exposure history is missing because they died, declined to be interviewed, or were lost to follow-up; and cases in persons who were interviewed or for whom other follow-up information was available and no mode of exposure was identified.

As of September 2000, the procedures for investigating cases reported without risk factors changed from ascertaining risk factors for all reported cases to estimating risk distributions from statistical models and population-based samples. States continue to investigate reports of unusual exposure to HIV and report these cases to CDC. CDC will continue to tabulate the number of documented unusual exposures to HIV reported by the states.

Because recently reported HIV or AIDS cases are more likely to be reported as NIR, recent AIDS incidence in some transmission categories will be underestimated unless an adjustment is made. For tables showing the estimated diagnoses of HIV infection and diagnoses of AIDS, the adjustment of NIR cases among adults and adolescents is based on the redistribution of transmission category, by sex, race, and region, of cases that were diagnosed during 1992–1999 and initially assigned to the NIR category but that were later reclassified.

Reporting Delays

Reporting delays (time between diagnosis of HIV infection or AIDS and report to CDC) may differ among transmission, geographic, racial/ethnic, age, sex, and vital status categories; for some AIDS cases, delays have been as long as several years. About 52% of AIDS cases were reported to CDC within 3 months of diagnosis, and about 88% were reported within 1 year. Approximately 88% of deaths of persons with AIDS are reported within 1 year. For cases of HIV infection diagnosed since the implementation of uniform reporting through the HIV/AIDS Reporting System on January 1, 1994, about 66% of all cases were reported to CDC within 3 months of diagnosis, and about 93% were reported within 1 year.

Adjustments of the estimated data on HIV infection and on AIDS to account for reporting delays are calculated by a maximum likelihood statistical procedure, taking into account differences in reporting delays among transmission, geographic, racial/ethnic, age, sex, and vital status categories and assuming that reporting delays in these groups have not changed over time [19, 20].

References

1. Council of State and Territorial Epidemiologists. Laboratory reporting of clinical test results indicative of HIV infection: new standards for a new era of surveillance and prevention [position statement]. Available at: http://www.cste.org/ps/2004pdf/04-ID-07-final.pdf. Accessed September 14, 2005.
2. Munoz A, He D, Cohn S, et al. Factors influencing correspondence between CD4% and CD4 count in HIV-infected homosexual men and intravenous drug users. In: Final Program and Oral Abstracts of the Eighth International Conference on AIDS; July 19– 24, 1992; Amsterdam, the Netherlands. Abstract MoC0086.
3. CDC. 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR 1992;41(No. RR-17):1–19.
4. Mellors JW, Munoz A, Giorgi JV, et al. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med 1997;126:946–954.
5. Yu LM, Easterbrook PJ, Marshall T. Relationship between CD4 count and CD4 percent in HIV-infected people. Int J Epidemiol 1997;26:1367–1372.
6. CDC. Guidelines for national human immunodeficiency virus case surveillance, including monitoring for human immunodeficiency virus infection and acquired immunodeficiency syndrome. MMWR 1999;48(No. RR-13):29–31.
7. CDC. Diagnosis and reporting of HIV and AIDS in states with integrated HIV and AIDS surveillance— United States, January 1994–June 1997. MMWR 1998;47:309–314.
8. CDC. Current trends: revision of the case definition of acquired immunodeficiency syndrome for national reporting— United States. MMWR 1985;34:373–375.
9. CDC. Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. MMWR 1987;36(No. SS-1):1S–15S.
10. CDC. Current trends update: acquired immunodeficiency syndrome— United States, 1994. MMWR 1995;44:64–67.
11. Buehler JW, Berkelman RL, Stehr-Green JK. The completeness of AIDS surveillance. J Acquir Immune Defic Syndr 1992;5:257–264.
12. Rosenblum L, Buehler JW, Morgan MW, et al. The completeness of AIDS case reporting, 1988: a multi-site collaborative surveillance project. Am J Public Health 1992;82:1495–1499.
13. Schwarcz SK, Hsu LC, Parisi MK, Katz MH. The impact of the 1993 AIDS case definition on the completeness and timeliness of AIDS surveillance. AIDS 1999;13:1109–1114.
14. Klevens RM, Fleming PL, Li J, et al. The completeness, validity, and timeliness of AIDS surveillance data. Ann Epidemiol 2001;11:443–449.
15. Karon JM, Rosenberg PS, McQuillan G, Khare M, Gwinn M, Petersen LR. Prevalence of HIV infection in the United States, 1984 to 1992. JAMA 1996;276:126–131.
16. CDC. Update: HIV-2 infection among blood and plasma donors— United States, June 1992–June 1995. MMWR 1995;44:603–606.
17. National Archives and Records Administration. Revisions to the standards for the classification of federal data on race and ethnicity. Federal Register 1997;62:58781–58790. Also available at: http://www.whitehouse.gov/omb/fedreg/ombdir15.html. Accessed January 5, 2006.
18. CDC. Current trends: heterosexually acquired AIDS—United States, 1993. MMWR 1994; 43:155–160.
19. Green TA. Using surveillance data to monitor trends in the AIDS epidemic. Stat Med 1998;17:143–154.
20. Karon JM, Devine OJ, Morgan WM. Predicting AIDS incidence by extrapolating from recent trends. In: Castillo-Chavez C, ed. Mathematical and Statistical Approaches to AIDS Epidemiology. Berlin: Springer-Verlag; 1989:83:58–88. Lecture Notes in Biomathematics.