Since HIV testing first became available testing has been a key component in national HIV prevention efforts1 because it provides an opportunity to counsel persons at risk for HIV infection about ways to reduce risk and identifies those already infected. Most importantly, evidence suggests that HIV-infected persons decrease their risk behaviors shortly after learning of their HIV infection.2
In addition, HIV testing early in the course of infection has become increasing important as advances in management of HIV infection have been shown to delay the progression to AIDS and death.3 Current recommendations call for initiating antiretroviral therapy before the development of AIDS.4 Cost estimates for caring for HIV-infected patients have shown that the cost of medical care for persons who are diagnosed with HIV infection when they meet the clinical criteria for AIDS is markedly higher than for persons who are diagnosed earlier.5 Despite these widely known benefits, HIV infections continue to be diagnosed late in the course of disease. In a study of 16 sites in the United States that did not include San Francisco, the Centers for Disease Control and Prevention (CDC) found that 45% of persons diagnosed were late testers.6 We wondered how prevalent this phenomenon was in a city with a high prevalence of AIDS7 and where a large proportion of the at-risk population has been tested.8 To that end, we used the San Francisco AIDS case registry to identify all persons in the city whose HIV diagnosis occurred within the 12 months before their AIDS diagnosis and examined temporal trends in the prevalence and the characteristics of these late testers.
Persons meeting the CDC's surveillance definition for AIDS9 are required by law to be reported to the San Francisco Department of Public Health. AIDS patients are reported primarily through active surveillance activities in which health department personnel review laboratory and pathology reports to identify persons with AIDS. Medical records are reviewed at the medical facility where the patient was initially diagnosed and at other healthcare facilities in San Francisco at which the patient was known to have received care. Reviews of the completeness of AIDS case reporting are conducted annually and have consistently found reporting to be highly complete10 (San Francisco Department of Public Health unpublished data).
The information collected on AIDS patients includes demographic and risk characteristics, health insurance status at the time of diagnosis, AIDS opportunistic illnesses, whether or not the patient was homeless at diagnosis, country of birth, the dates and results of the first and periodic subsequent CD4 and viral load tests, and the date of the patient's first positive HIV test. The date of patient's first positive HIV test was obtained either from a laboratory report or by physician documentation of HIV infection listed in the medical record. In 2001, the information collected on persons reported with AIDS was expanded to include the date of patient's self-report of the first known HIV positive date.
We included San Francisco residents (aged ≥13 yrs) diagnosed with AIDS between 2001 and 2005 and reported through March 27, 2006. Persons whose initial HIV diagnosis occurred 12 months or less before their AIDS diagnosis were defined as late testers. The date of HIV diagnosis was determined by identifying the earliest date of a documented positive HIV antibody test, self report of a positive antibody test, documented CD4 or viral load test, or use of antiretroviral therapies.
The χ2 test was used to measure associations between sociodemographic and risk characteristics and testing late. Independent predictors of late testing were measured using multiple logistic regression in which all variables that were significant at a P value of 0.1 or less from the χ2 test were entered into the model.
A total of 2243 San Francisco residents were diagnosed with AIDS between January 1, 2001 and December 31, 2005 and reported to the San Francisco Department of Public Health by March 27, 2006. Of these, 2139 (95.4%) were included in the analysis. We excluded 60 cases in which a medical chart review was not done, 42 cases that were missing information on health insurance, and 2 cases in which the mode of transmission was blood transfusion. The greatest proportion of patients was men, white, aged 30 to 49 years at the time of diagnosis, men who had sex with men, had private health insurance, and were born in the United States (Table 1). Eighty-three percent were diagnosed with AIDS because of low CD4 cell count. The number of patients diagnosed each year was relatively stable with the exception of 2005 in which fewer patients were identified.
Of the 2139 AIDS patients, 830 (38.8%) were late testers and this proportion was stable over time. Late testing occurred more frequently among persons of color than among whites, among persons <30 years old at the time of AIDS diagnosis, among persons whose risk for HIV infection was heterosexual contact or those without a reported risk, persons with private health insurance and without health insurance, persons born outside of the United States, and persons whose AIDS diagnosis included an opportunistic illness.
Factors that were independently associated with an increased likelihood of testing late were age younger than 30 years at the time of diagnosis, having acquired HIV infection through heterosexual contact, being reported without a risk factor, having private health insurance, or no health insurance at the time of AIDS diagnosis, being born outside of the United States, and having an opportunistic illness at the time of AIDS diagnosis (Table 1). Men who had sex with men and also injected drugs had a decreased likelihood of late testing.
Despite the visibility of HIV/AIDS in San Francisco, the availability of HIV testing, including free, confidential, and anonymous testing in the city, and the well-known benefits and availability of treatment for HIV infection, nearly 40% of persons diagnosed with AIDS in San Francisco between 2001 and 2005 were diagnosed with HIV infection only 12 months or less before their AIDS diagnosis and this proportion did not decline significantly during this time period. The extent of late testing in San Francisco is only slightly lower than what has been reported nationally.6 Although information on late testing is not available for San Francisco in the preHAART era, studies elsewhere found conflicting trends in late testing before and after HAART became available with the report of a decline in late testing postHAART,11 some showing an increase12,13 and one reporting relative stability.14 Thus, although we might expect the availability of effective therapy to decrease late testing, this does not seem to be consistently true.
Similar to national data, late testing in San Francisco occurred more frequently among persons who may not have been aware of their risk for HIV infection including younger persons, heterosexuals, and those without a reported risk. These factors were associated with late testing in other studies.12,15-18 In addition, health insurance status was a strong and independent predictor of late testing, suggesting that access to health care may affect testing. Late testing was also associated with being born outside of the United States. Studies conducted in countries other than the United States have shown that late testing was associated with being born outside of the country in which the study occurred.12,16,17 It is possible that these immigrants may not have access to care or testing or may not be aware of testing services. Alternatively, being born outside of the United States may be a marker for another factor that we were not able to measure that would decrease the likelihood of testing early, such as not being aware of ones risk or with a fear of learning of ones HIV infection, the reasons most frequently cited for not testing for HIV.19
Persons whose AIDS diagnosis included an opportunistic illness were more likely to be late testers. This finding is probably a reflection of persons seeking testing because of symptoms of AIDS. In 2 interview studies of newly diagnosed AIDS patients in the United States, the majority of patients cited illness as their reason for testing.6,18
Testing late occurred less frequently among men who have sex with men and inject drugs. This finding may be due to outreach programs aimed at promoting testing in this population or research studies that include testing in this risk group. An additional explanation is that men who have sex with men and also inject drugs are known to be the group in San Francisco that is at highest risk for HIV infection and markedly higher than heterosexual injection drug users in which new HIV infections are rare. Further, injection drug use is also associated with multiple health problems that may bring this group into healthcare settings where they are offered testing.
We did not find an independent association between race/ethnicity and late testing. Such an association was found in bivariate analysis in our study and in the United States based on 2 interview studies cited previously.6,18 We believe that the lack of independent association between race/ethnicity and late testing in our study is due to confounding by other independent predicators. Among African Americans, likelihood of late testing seems to be confounded by risk; when we removed risk group from the multivariate analysis, African Americans had a statistically significantly increased likelihood of testing late (OR: 1.44, 95% CI: 1.1, 1.9). For Latinos and persons of other race (primarily Asians and Pacific Islanders), confounding seems to be due to country of birth; when country of birth is excluded from the logistic regression model, the risk of late testing is statistically significantly increased among Latinos (OR: 1.28, 95% CI: 1.0, 1.6) and persons of other race (OR: 1.54, 95% CI: 1.1, 2.2).
There are several limitations to consider when interpreting the findings from this analysis. We estimated the date of initial HIV diagnosis and these may be inaccurate in some cases. The data used for this analysis come from mandatory AIDS case reporting. Although this has the advantage of providing very complete data on persons with AIDS, we were not able to include persons with advanced HIV disease who did not yet have AIDS. These persons may have also been diagnosed many years after acquiring HIV infection but had not yet progressed to AIDS and therefore were not included in our analysis. In addition, the data we used come exclusively from medical records. Thus we were not able to determine the reasons for late testing. We were not able to compare trends in late testing before and after HAART became widely available because information on the date of HIV diagnosis has only been collected since 2001. In addition, San Francisco differs from other areas in terms of HIV testing. A study of testing behaviors among men who have sex with men (the population most severely affected by HIV in San Francisco)7 found that 94% had been tested for HIV.8 In another recent study of young men who have sex with men, the proportion of HIV-infected men who were unaware of their infection was lower in San Francisco than in the other 4 cities studied.20 Given these differences, our findings might not generalize to other geographic areas and populations.
Despite these limitations there are important conclusions that can be made from our analysis. Even with the current availability of HIV testing programs, expansion of testing services, particularly into places likely to test persons who do not consider themselves to be at high risk for HIV infection should be pursued. Routine testing of most persons has been recommended as cost effective and efforts to promote such testing should be developed.21,22 The availability of rapid tests, currently marketed for use in clinic settings, should be evaluated for personal use in which the test is self-administered and interpreted. Additionally, research should be undertaken to better understand the psychological and social underpinnings of late testing. Such information could help public health officials mount initiatives aimed at reducing this phenomenon.
1. CDC. Advancing HIV prevention: new strategies for a changing epidemic-United States, 2003. MMWR Morb Mortal Weekly Rep
2. CDC. Adoption of protective behaviors among persons with recent HIV infection and diagnosis-Alabama, New Jersey, and Tennessee, 1997-1998. MMWR Morb Mortal Weekly Rep
3. Hulgan T, Raffanti S, Kheshti A, et al. CD4 lymphocyte percentage predicts disease progression in HIV-infected patients initiating highly active antiretroviral therapy with CD4 lymphocyte counts >350 lymphocytes/mm3
. J Infect Dis
4. Yeni PG, Hammer SM, Hirsch MS, et al. Treatment for adult HIV infection: 2004 recommendations of the International AIDS Society-USA Panel. JAMA
5. Krentz HB, Auld MC, Gill MJ. The high cost of medical care for patients who present late (CD4 <200 cells/microL) with HIV infection. HIV Med
6. CDC. Late- versus early-testing of HIV-16 sites, United States, 2000-2003. MMWR Morb Mortal Weekly Rep
8. Schwarcz S, Stockman J, Delgado V, Scheer S. Does name-based HIV reporting deter high-risk persons from HIV testing? Results from San Francisco. J Acquir Immune Defic Syndr
9. CDC. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep
10. Schwarcz SK, Hsu LC, Parisi MK, et al. The impact of the 1993 AIDS case definition on the completeness and timeliness of AIDS surveillance. AIDS
11. Chadborn TR, Baster K, Delpech VC, et al. No time to wait: how many HIV-infected homosexual men are diagnosed late and consequently die? (England and Wales, 1993-2002). AIDS
12. Girardi E, Sampaolesi A, Gentile M, et al. Increasing proportion of late diagnosis of HIV infection among patients with AIDS in Italy following introduction of combination antiretroviral therapy. J Acquir Immune Defic Syndr
13. McDonald AM, Li Y, Dore GJ. Late HIV presentation among AIDS cases in Australia, 1992-2001. Aust N Z J Public Health
14. Neal JJ, Fleming PL. Frequency and predictors of late HIV diagnosis in the United States, 1994 through 1999. Ninth Conference on Retroviruses and Opportunistic Infections
. February 2002; Seattle WA; [abstract 474-M].
15. Castilla J, Sobrino P, de la Fuente L, et al. Late diagnosis of HIV infection in the era of highly active antiretroviral therapy: consequences for AIDS incidence. AIDS
16. Manavi K, McMillan A, Ogilvie M, et al. Heterosexual men and women with HIV test positive at a later stage of infection than homosexual or bisexual men. Intern J STD AIDS
17. Couturier E, Schwoebel V, Michon C, et al. Determinants of delayed diagnosis of HIV infection in France, 1993-1995. AIDS
18. Wortley PM, Chu SY, Diaz T, et al. HIV testing patterns: where, why, and when were persons with AIDS tested for HIV. AIDS
19. Kellerman SE, Lehman JS, Lansky A, et al. HIV testing within at-risk populations in the United States and the reasons for seeking or avoiding HIV testing. J Acquir Immune Defic Syndr
20. CDC. HIV prevalence, unrecognized HIV testing among men who have sex with men-five U.S. cities, June 2004-April 2005. MMWR Morbid Mortal Wkly Rep
21. Paltiel AD, Weinstein MC, Kimmel AD, et al. Expanded screening for HIV in the United States-an analysis of cost-effectiveness. N Engl J Med
22. Sanders GD, Bayoumi AM, Sundaram V, et al. Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy. N Engl J Med