Going North: Mexican migrants and their vulnerability to HIV.

BACKGROUND: Mexican migrants are at higher risk for HIV than Mexicans who do not migrate to the United States. Migration to the United States was the driving factor of the early Mexican HIV epidemic, and it is likely that it continues to strongly influence incidence. An overview of migration of Mexicans to the United States identifies many pervasive environmental and structural factors as well as risk behaviors that render migrants vulnerable to HIV infection. However, published studies sampling Mexicans while in the United States suggest a relatively low prevalence of HIV among the general migrant population. To better understand this apparent paradox, we sought to identify any demographic variables among Mexicans while in Mexico that may indicate that migrants have or acquire resources that have a protective effect from their vulnerability due to migration. METHODS: A California-Mexico binational collaboration project, with a respondent-driven sample with population-based quotas, was conducted in five Mexican states from December 2004 to January 2005, in areas with a high index of migration to the United States. We compared demographic and behavior variables of Mexicans with a history of migration to the United States in the past 12 months to nonmigrant Mexicans living in the same community. RESULTS: A total of 1539 migrants and 1236 nonmigrants were recruited from five Mexican states. Migrants (men and women) reported more HIV risk behavior than nonmigrants in the past 12 months. Migrants reported more sexual partners and noninjected drug use. Migrants reported higher condom use during vaginal sex and were more likely to have taken an HIV test. CONCLUSION: Though migrants reported higher HIV-related risk behaviors, they also reported higher condom use. Migrants were more likely to have accessed an HIV test indicating an opportunity for a prevention intervention. More binational collaborations are needed to research the different levels of vulnerability among Mexican migrants and actual acquisition of HIV infection. In addition, more research is needed to identify protective factors for HIV prevention interventions among Mexican migrant communities in Mexico and in the United States.

Knock-knock: a population-based survey of risk behavior, health care access, and Chlamydia trachomatis infection among low-income women in the San Francisco Bay area.

To estimate the prevalence of urogenital chlamydial infection among young, low-income women in northern California and to describe correlates of infection, a population-based door-to-door household cluster survey was conducted from 1996 through 1998. The participants included 1439 women 18-29 years of age, with a mean age of 24 years, most of whom were African American (43%) or Latina (23%) and had a median income of $500-$999 per month. Most (94%) had received health care in the past year, and approximately 50% was covered by state insurance programs. Although more than half (62%) had had a recent pelvic examination, only 42% had recently used a condom with a new partner. The prevalence of urogenital chlamydial infection was 3.2% (95% confidence interval, 2.2%-4.2%). Women with chlamydia were more likely to be younger (18-21 years of age) and nonwhite and to have lower socioeconomic status. These data demonstrated an approximately 2-3-fold greater burden of infection than routine surveillance data have suggested.

Prevalence of HIV infection, sexually transmitted diseases, and hepatitis and related risk behavior in young women living in low-income neighborhoods of northern California.

OBJECTIVE: To estimate the prevalence of human immunodeficiency virus (HIV) infection, sexually transmitted diseases, and hepatitis and the associated sexual and drug-using behavior among women residing in low-income neighborhoods in 5 northern California counties. METHODS: From April 4, 1996, to January 6, 1998, women aged 18 to 29 years were recruited door-to-door from randomly selected street blocks within 1990 census block groups below the 10th percentile for median household income for each county. RESULTS: Of 24,223 dwellings enumerated, contact was made with residents from 19,546 (80.7%). Within contacted dwellings, 3,560 eligible women were identified and 2,545 enrolled (71.5%). Weighted estimates for disease prevalence were HIV infection, 0.3% (95% confidence interval, 0.1%-0.4%); syphilis, 0.7% (0.3%-1.1%); gonorrhea, 0.8% (0.3%-1.3%); chlamydia, 3.3% (2.4%-4.8%); herpes simplex virus, type 1, 73.7% (71.6%-76.9%); herpes simplex virus, type 2, 34.4% (29.9%-39.0%); hepatitis A, 33.5% (28.3%-38.7%); chronic hepatitis B, 0.8% (0.3%-1.2%); and hepatitis C, 2.5% (1.4%-3. 6%). Condom use at last sexual intercourse with a new partner was reported by 44.0% (33.9%-54.1%). Injection drug use in the last 6 months was reported by 1.8% (1.0%-2.7%). CONCLUSIONS: The Young Women's Survey provided population-based estimates of the prevalence of 8 infectious diseases and related risk behavior within a population for whom data are often difficult to collect. Population-based data are needed for appropriate targeting and planning of primary and secondary disease prevention.

Continuing high prevalence of HIV and risk behaviors among young men who have sex with men: the young men's survey in the San Francisco Bay Area in 1992 to 1993 and in 1994 to 1995.

Several recent studies have shown high rates of HIV infection and risk behavior among young men who have sex with men (MSM). To assess the direction of the epidemic in this population, we replicated a venue-based study performed in the San Francisco Bay Area during 1992 and 1993. From May 1994 to September 1995, we surveyed 675 MSM aged between 17 and 22. After statistical adjustment for age, ethnicity, residence, and site of recruitment, seroprevalence did not change significantly between the 1992 to 1993 (8.4%) and the 1994 to 1995 (6.7%) surveys. Similarly, no significant changes were found in the rates during the previous 6 months of unprotected receptive anal intercourse (23.4% versus 24.9%), injection drug use (8.0% versus 7.8%), or needle sharing among injection drug users (56.3% versus 64.5%) between the two surveys. Despite the increased attention that the problem of high risk behavior among young MSM has received, effective prevention interventions for MSM are needed as profoundly now as they had been several years ago.

Non-Hodgkin's lymphoma among people with AIDS: incidence, presentation and public health burden. AIDS/Cancer Study Group.

We describe the anatomic and histologic presentation and prognosis of non-Hodgkin's lymphoma (NHL) among people with AIDS (PWA) and determine their contribution to the NHL burden. We linked AIDS and cancer registries in selected areas of the United States and compared NHL sites and histologies in PWA and non-PWA, after adjusting for age, sex and ethnicity. Among 51,033 PWA, we found 2,156 cases of NHL (4.3%). Half of NHL cases occurring post-AIDS were not reported to AIDS registries. NHL was part of an AIDS-defining condition for 3.2% of all PWA; the relative risk of NHL with 3.5 years of another AIDS diagnosis was 165-fold compared to non-PWA within the cancer surveillance system. Of NHLs, 39% were high grade (vs. 12% among non-PWA), 60% were nodal (vs. 74% among non-PWA) and 15% had brain primaries (vs. 1% among non-PWA). Excluding brain sites, extranodal sites were still 20% more common than expected. Relative risk was elevated for all histologic types, with the risk ranging from 652-fold for high-grade diffuse immunoblastic tumors and 261-fold for Burkitt's lymphomas to 113 for intermediate-grade lymphoma to 14-fold for low-grade lymphoma. Survival among PWA with NHL was poor, and tumor grade had little impact. In high-risk AIDS areas, AIDS-related NHLs constitute a major share of the NHL burden. We conclude that NHL risk is considerably under-estimated in AIDS registry data. The major differences between PWA and non-PWA were the high frequency of brain lymphoma and the increase in high-grade lymphomas in PWA. However, the grade of NHL did not influence the prognosis among PWA with lymphoma. The increasing risk of NHL in PWA has contributed substantially to the general increase in NHL rates in the United States since 1981.

Projected incidence of AIDS in San Francisco: the peak and decline of the epidemic.

To predict the incidence of AIDS from 1978 through 1998 in San Francisco, we developed a model that combined annual HIV seroconversion rates for homosexual and bisexual men and for heterosexual injecting drug users with estimates of the incubation period distribution between HIV seroconversion and AIDS diagnosis and with estimates of the size of the at-risk populations. Our model assumed the availability of antiretroviral therapy at the efficacy level of zidovudine monotherapy. The annual number of new AIDS cases is estimated to have peaked at 3332 in 1992, and is projected to decline to 1196 annually by 1998. Although the projected number of cases decreased steadily during this period for homosexual and bisexual men, the projected number of cases for injection drug users, women, and persons with other risks increased between 1993 and 1998. The decline in the incidence of AIDS in San Francisco reflects the dramatic reductions in new HIV infections that occurred a decade previously and that were achieved as a result of significant changes in high-risk behaviors, primarily among homosexual and bisexual men. Changes in HIV seroincidence must be factored in before attributing the decrease in AIDS incidence to more effective combination antiretroviral treatment.

Prevention of Pneumocystis carinii pneumonia: who are we missing?

OBJECTIVE: To obtain population-based information on the characteristics of persons who were not receiving chemoprophylaxis against Pneumocystis carinii pneumonia (PCP) by examining the use of primary and secondary PCP prophylaxis among San Francisco residents whose AIDS-defining opportunistic illness was PCP in 1993. DESIGN: Retrospective medical record review. SETTING: Medical charts were obtained from San Francisco hospitals and outpatient facilities at which AIDS patients received their initial AIDS diagnosis. PARTICIPANTS: San Francisco residents whose AIDS-defining opportunistic illness was PCP in 1993. MAIN OUTCOME MEASURES: Use of primary and secondary PCP prophylaxis. RESULTS: Of the 326 eligible patients, 35% received primary PCP prophylaxis. Non-whites were significantly less likely to have received primary PCP prophylaxis than white patients [22 versus 40%, respectively; odds ratio (OR), 0.49; 95% confidence intervals (CI), 0.28-0.87]. Uninsured individuals-were also less likely to have received primary PCP prophylaxis than those with insurance (18 versus 41%; OR, 0.35; 95% CI, 0.17-0.73). The sociodemographic characteristics of patients who did and did not receive secondary PCP prophylaxis did not differ significantly. The most frequently cited reasons for not receiving primary PCP prophylaxis were that patients were unaware of their infection with HIV or were not receiving regular medical care. CONCLUSIONS: Barriers to receipt of PCP prophylaxis exist and are resulting in cases of preventable disease and unnecessary medical costs. Interventions to increase counseling, testing, and referral to medical care for persons at high risk for HIV infection are needed.

Cryptosporidiosis among patients infected with human immunodeficiency virus. Factors related to symptomatic infection and survival.

The authors reviewed the medical records of 194 human immunodeficiency virus (HIV)-positive patients newly diagnosed with cryptosporidiosis and all 3,564 patients with newly diagnosed acquired immunodeficiency syndrome (AIDS) at San Francisco General Hospital for the period 1986-1992. The study was designed to address three questions: 1) How do AIDS patients who present with cryptosporidiosis differ from other patients with AIDS? 2) What factors are associated with survival among AIDS patients with newly diagnosed cryptosporidiosis? 3) Does a diagnosis of cryptosporidiosis impact survival after AIDS diagnosis? A total of 194 cases of cryptosporidiosis among HIV-infected patients were identified during the study period. Of the 194 patients, 109 (56%) had no prior diagnosis of AIDS. These 109 patients represented 3.1% of the 3,564 newly diagnosed cases of AIDS in the same period. Among the 134 patients with CD4 T-lymphocyte counts performed within 3 months of Cryptosporidium diagnosis, 34 (25%) had CD4 counts greater than 209 cells/ml. In a multivariate conditional logistic regression model, the incidence of Cryptosporidium was related to ethnicity (for blacks vs. whites, matched odds ratio (OR) = 0.15, 95% confidence interval (CI) 0.03-0.73), CD4 count (for a CD4 count of < or = 53 cells/ml vs. > 53 cells/ml, matched OR = 12.60, 95% CI 4.01-39.61), and age (for a 10-year increase, matched OR = 0.51, 95% CI 0.27-0.98). Two factors measured at the time of Cryptosporidium diagnosis were identified as being independently associated with survival (p < 0.001) in the proportional hazards model: CD4 count < or = 53 cells/ml versus > 53 cells/ml (relative hazard = 6.18, 95% CI 2.99-12.76) and hematocrit < or 37% versus > 37% (relative hazard = 2.27, 95% CI 1.22-4.22). The median durations of survival in the four subgroups of Cryptosporidium-infected patients defined by these two variables differed significantly from each other (range, 204-1,119 days). Cryptosporidiosis as an initial AIDS-defining diagnosis was associated with an elevated relative hazard of death in comparison with other AIDS-defining diagnoses (relative hazard = 2.01, 95% CI 1.38-2.93). These data identify the groups of HIV-infected individuals at risk for presentation with symptomatic Cryptosporidium infection; the distinct survival patterns among subgroups of those patients already infected with this parasite; and the survival of AIDS patients with newly diagnosed cryptosporidiosis relative to patients with other AIDS-defining conditions. Such information is necessary for the design of prospective studies, the development of prophylactic strategies, the evaluation of candidate therapies, and the provision of prognostic information to patients.

The Young Men's Survey: methods for estimating HIV seroprevalence and risk factors among young men who have sex with men.

Traditional sampling methods are unsuitable for determining the levels of human immunodeficiency virus type I infection and related behavioral risk factors among young men who have sex with men. Most surveys of this hard-to-reach population have used nonprobability samples of young men in clinical or public settings. While these studies have revealed high rates of HIV infection and risk behaviors, their findings are not generalizable to broader populations of young men who have sex with men. To better understand the epidemiology of HIV within this population, the Centers for Disease Control and Prevention, in collaboration with state and local health departments, has developed a venue-based probability survey of young men who have sex with men. Conducted in seven metropolitan areas in the United States, the Young Men's Survey combines outreach techniques with standard methods of sample surveys to enumerate, sample, and estimate prevalence outcomes of a population of young men who frequent public venues and who have sex with other men. Venues where young men who have sex with men are sampled include dance clubs, bars, and street locations. At sampled venues, young men are enumerated, consecutively approached, and offered enrollment if they are determined eligible. Young men who agree to participate in the Young Men's Survey are interviewed, counseled, and tested for human immunodeficiency virus, hepatitis B, and syphilis in vans parked near sampled venues. The Young Men's Survey provides data on the locations and times at which demographic and behavioral subgroups of young men who have sex with men may be targeted for prevention activities. Behaviors and psychosocial factors associated with human immunodeficiency virus infection can be used to design culturally relevant and age-specific prevention activities for young men who have sex with men.

HIV seroprevalence and risk behaviors among lesbians and bisexual women in San Francisco and Berkeley, California.

Few data are available on human immunodeficiency virus (HIV) infection and risk behaviors among lesbians and bisexual women. A total of 498 lesbians and bisexual women was sampled from public venues in San Francisco and Berkeley, Calif, during 1993. The overall HIV seroprevalence was 1.2%. Ten percent of participants reported injecting drugs since 1978. Forty percent of the participants reported unprotected vaginal or anal sex with men during the past 3 years, including unprotected sex with gay and bisexual men and male injection drug users. The high rates of injection drug use and unsafe sexual behaviors suggest that lesbians and bisexual women frequenting public venues in San Francisco and Berkeley are at risk for HIV infection.

Temporal trends in human immunodeficiency virus seroprevalence and sexual behavior at the San Francisco municipal sexually transmitted disease clinic, 1989-1992.

The authors analyzed temporal trends in human immunodeficiency virus (HIV) infection among men and women who visited the San Francisco municipal sexually transmitted disease clinic between 1989 and 1992, using blinded HIV seroprevalence data. Temporal changes in sexual behavior were evaluated by abstracting self-reported information on sexual behaviors from a random sample of charts of men who visited the clinic between 1990 and 1992. From 1989 to 1992, HIV seropositivity declined from 2.0% to 1.0% among women (p = 0.06) and from 18.9% to 12.0% (p < 0.001) among men. The percentage of patients who reported having anal intercourse in the previous year did not change significantly during the study period. The percentage of male patients who reported having vaginal intercourse during the previous year decreased from 82.9% to 78.6% (p < 0.05), and the percentage of male patients who reported engaging in receptive oral sex during the previous year increased from 24.0% to 41.6% (p < 0.001). The percentage of male patients who reported that they always used condoms increased from 31.8% to 49.2% for anal sex, from 8.7% to 19.5% for vaginal sex, and from 1.4% to 6.3% for oral sex (p < 0.05). Among patients visiting the sexually transmitted disease clinic, there was a steady and significant decline in HIV seroprevalence. The decline in HIV seroprevalence was accompanied by a significant trend toward safer sexual practices. However, by the end of the study period, less than half of the patients reported using condoms all of the time, which suggests that there is a need to expand behavioral interventions to focus on high-risk persons.

Seroprevalence of HIV and risk behaviors among young homosexual and bisexual men. The San Francisco/Berkeley Young Men's Survey.

OBJECTIVE: To estimate the prevalence of human immunodeficiency virus (HIV) infection and risk behaviors among young homosexual and bisexual men sampled from public venues in San Francisco and Berkeley, Calif. DESIGN: A survey of 425 young homosexual and bisexual men sampled from 26 locations during 1992 and 1993. Participants were interviewed and blood specimens were drawn and tested for HIV, level of CD4+ T lymphocytes, and markers of hepatitis B and syphilis. SETTING: Public venues in San Francisco and Berkeley, including street corners and sidewalks, dance clubs, bars, and parks. POPULATION STUDIED: Homosexual and bisexual men aged 17 to 22 years. MAIN OUTCOME MEASURES: Prevalence of HIV infection and risk behaviors. RESULTS: The HIV seroprevalence was 9.4% (95% confidence interval, 6.8% to 12.6%). The prevalence of markers for hepatitis B was 19.8% (95% confidence interval, 16.1% to 23.9%), and that for syphilis was 1.0% (95% confidence interval, 0.3% to 2.4%). The HIV seroprevalence was significantly higher among African Americans (21.2%) than among other racial/ethnic groups (P = .002). Approximately one third (32.7%) of the participants reported unprotected anal intercourse, and 11.8% reported injecting drug use in the previous 6 months. At the time of interview, 70.0% of the HIV-infected men did not know that they were HIV seropositive, and only 22.5% were receiving medical care for HIV infection. CONCLUSIONS: The prevalence of HIV infection is high among this young population of homosexual and bisexual men, particularly among young African-American men. The high rates of HIV-related risk behaviors suggest a considerable risk for HIV transmission in this population. Prevention programs and health services need to be tailored to address the needs of a new generation of homosexual and bisexual men.

Use of Bayes theorem to estimate the impact of the proposed CD4-based expansion of the AIDS case definition.

To assess the immediate impact of the proposed CD4-based expansion of the AIDS case definition, we determined two key proportions from a subsample of men from the San Francisco City Clinic Cohort (SFCCC). We then used Bayes theorem to project the number of persons fitting the proposed definition in the entire SFCCC and in the city of San Francisco. Among those men meeting the 1987 AIDS case definition, the proportion with a CD4 cell count < 200 cells (within 6 months of their AIDS diagnosis) is 0.70 (16 of 23). Among those with a CD4 count ever < 200 cells, the proportion with AIDS is 0.40 (29 of 73). Our estimates show that 446 persons in the SFCCC and 3,603 persons in San Francisco would fit only the expanded definition. Thus, the proposed definition would likely more than double the number of persons who could be diagnosed with AIDS. Bayes theorem offers a simple method for estimating the immediate impact of the proposed CD4-based expansion of the AIDS case definition.

Comparison of voluntary and blinded human immunodeficiency virus type 1 (HIV-1) seroprevalence surveys in a high prevalence sexually transmitted disease clinic population.

To compare the seroprevalence of and risk factors for human immunodeficiency virus infection (HIV) among patients attending a public sexually transmitted disease clinic, the authors conducted both voluntary and blinded seroprevalence surveys between June 1989 and August 1990. For the voluntary survey, every twenty-fifth patient attending the clinic for a new problem was invited to receive anonymous testing for HIV antibody. For the blinded survey, sera obtained for syphilis serologies from 2,297 (86%) of the 2,682 patients attending the clinic for a new problem were tested for HIV antibody after all personal identifiers were removed. Of the 946 eligible patients, 631 (66.7%) agreed to participate in the voluntary survey. Black men were significantly less likely to participate than other men and women (p < 0.0001). The prevalence of HIV antibody was 25% greater in the blinded survey than in the voluntary survey (15.2% and 11.4%, respectively, p < 0.05). This difference was due primarily to black homosexual/bisexual men, who had a 12.7-fold greater risk of HIV infection in the blinded survey than in the voluntary survey. These results suggest that blinded seroprevalence surveys may provide a better prevalence estimate of HIV infection than voluntary surveys. The greater risk for HIV infection observed among homosexual and bisexual black men, who were tested only in the blinded serosurvey, suggests that efforts to increase voluntary testing for HIV infection in this group should be developed.

HIV-1 seroprevalence among women attending sexually transmitted disease clinics in California. California Family of Surveys and Sentinel Surveillance Consortia.

To evaluate the distribution of the human immunodeficiency virus type 1 (HIV-1) epidemic among California women, we analyzed HIV-1 seroprevalence and risk factors among women attending sexually transmitted disease (STD) clinics in 21 local health jurisdictions. Using standardized protocols developed by the Centers for Disease Control, we tested unlinked serum specimens from women attending participating STD clinics in 1989. We analyzed demographic characteristics, HIV risk exposure groups, and results of HIV-1 antibody testing on 17,210 specimens with an overall HIV-1 seroprevalence of 0.57%. Seroprevalence rates were highest for African-American women, women 25 to 29 years of age, injection drug users, and women attending STD clinics in San Francisco. After multiple logistic regression analysis, HIV-1 seropositivity remained highest for these four groups. The rate of HIV-1 infection among women attending STD clinics in California underscores the continued need to make HIV counseling and testing an integral component of routine services for women being evaluated for, or presenting with, sexually transmitted diseases.

The spectrum of acquired immunodeficiency syndrome (AIDS)-associated malignancies in San Francisco, 1980-1987.

Population-based disease registries for acquired immunodeficiency syndrome (AIDS) and cancer were linked for San Francisco residents to study the pattern of AIDS-associated malignancies diagnosed during the time period 1980-1987. A total of 1,756 newly diagnosed malignancies were identified during these years among members of the AIDS cohort. Of these, 1,752 (99.7%) occurred in males, 1,454 (83%) were Kaposi's sarcoma, 235 (13%) were non-Hodgkin's lymphoma, and 16 (1%) were Hodgkin's disease. The distributions of AIDS patients with cancer differed significantly from those without cancer by race and by risk group. Malignancies known to be human immunodeficiency virus (HIV)-associated, and now diagnostic of AIDS (Kaposi's sarcoma, non-Hodgkin's lymphoma), were, as would be expected, dramatically in excess among AIDS patients. Some malignancies not traditionally thought to be HIV-associated appear to have occurred more often than expected in the study cohort. These include Hodgkin's disease, rare non-melanoma skin cancers, and cancers of the rectum, anus, and nasal cavity. Malignancies known to be HIV-associated were more likely to be diagnosed concurrent with or subsequent to first AIDS diagnosis. Conversely, malignancies not known to be HIV-associated were more likely to be diagnosed before AIDS diagnosis. Compared with the concurrent reference population of the San Francisco Bay Area, there was little or no increase in Kaposi's sarcoma over the time interval of this study. For non-Hodgkin's lymphoma, and suggestively for Hodgkin's disease, however, the temporal increase has been quite dramatic.

Survival for women and men with AIDS.

To compare trends in the length of survival for women and men after diagnosis of AIDS, data were analyzed for 139 women and 7045 men who were reported with AIDS in San Francisco between July 1981 and 31 December 1990. Patients were followed prospectively through 15 May 1991. The median survival for women (11.1 months) was significantly shorter than that for men (14.6 months). When data were stratified by year of diagnosis, significantly improved survival was observed in recent years for both women and men, although survival for women remained significantly shorter than that for men. Among those who received either zidovudine or 2',3'-dideoxyinosine, survival did not differ by gender. However, among those not receiving therapy, survival was significantly shorter for women. These results suggest that the shorter survival of women may be a result of factors other than gender, possibly including less use of antiretroviral therapy.

A comparison of younger and older gay men's HIV risk-taking behaviors: the Communication Technologies 1989 Cross-Sectional Survey.

Data from the 1989 Communication Technologies cross-sectional survey of gay men in San Francisco indicate that both levels and correlates of sexual risk are different between younger and older gay men. Gay men under the age of 30 report higher risk behavior for human immunodeficiency virus (HIV) infection than do gay men who are 30 years of age or older. Further, the set of correlates of unprotected anal intercourse are different between younger and older gay men. Both young and old gay men report that having a primary partner and a lower perceived impact of the AIDS epidemic on their sexual behavior are associated with risk. However, among young gay men, reporting a lower attributed risk for HIV infection to unprotected anal intercourse, higher concern about AIDS risks, and shorter length of residence in San Francisco are positively correlated with risk-taking behavior. These associations were not statistically significant among gay men 30 years of age or older. It appears that the circumstances and/or reasons for taking sexual risk are different between older and younger gay men. Prevention programs must be designed so that they are sensitive to the needs of each generation of gay men.

Progressive multifocal leukoencephalopathy in persons infected with human immunodeficiency virus, San Francisco, 1981-1989.

Progressive multifocal leukoencephalopathy (PML), a rare neurological disease, has been sporadically reported in persons infected with human immunodeficiency virus (HIV), the causative agent of acquired immune deficiency syndrome (AIDS). From January 1981 through February 1989, in San Francisco, we identified 94 HIV-infected persons with PML, of whom 48 (51%) were pathologically confirmed (as required for AIDS case reporting). These 48 patients were significantly older when diagnosed with AIDS (20% older than 50 years) than patients with AIDS without PML. The remaining 46 (49%) patients, diagnosed clinically and by neuroimaging, did not differ significantly from definitive patients in demographic or survival characteristics after PML diagnosis. We detected antibodies to JC virus, the causative agent of PML, in 9 of 14 (64%) AIDS-related patients with PML, and in 9 of 14 (64%) matched control subjects, suggesting that determination of JC virus antibody status before AIDS diagnosis does not reliably indicate which patients will contract PML. Our study shows that the proportion of patients with AIDS who contracted PML remained stable between 1981 and 1988, but increased in the first 2 months of 1989. Our findings further indicate that PML in HIV-infected patients may be underestimated by as much as 50%.