San Francisco Department of Public Health HIV/AIDS Epidemiology Annual Report HIV Epidemiology Section

Transcription

1 San Francisco Department of Public Health HIV/AIDS Epidemiology Annual Report HIV Epidemiology Section 2012

2 San Francisco Department of Public Health HIV Epidemiology Annual Report 2012 HIV Epidemiology Section June 2013

3 The HIV Epidemiology Annual Report is published annually by the HIV Epidemiology Section. To obtain a free copy of the report, please contact: San Francisco Department of Public Health HIV Epidemiology Section 25 Van Ness Avenue, Suite 500 San Francisco, CA USA Phone (415) Fax (415) The HIV Epidemiology Annual Report 2012 is available on the internet at:

4 Acknowledgments This report was prepared by the HIV Epidemiology Section staff. We wish to thank the San Francisco Sexually Transmitted Disease Prevention and Control Services and the San Francisco STOP AIDS Project for providing data in this report. Data for opportunistic illnesses and associated mortality were analyzed and prepared by Kpandja Djawe, Centers for Disease Control and Prevention. In addition, we wish to acknowledge the contribution of persons with HIV, HIV health care providers, community groups, researchers, and members of the community. Publication of this report would not have been possible without their cooperation, dedication, and hard work.

5 San Francisco Department of Public Health Director of Health Health Officer and Director of Population Health and Prevention HIV Epidemiology Section Director HIV Surveillance Unit, Director Bio-Behavioral Surveillance Unit, Director Strategic Information, Director HIV Special Studies, Director Program Coordinators Epidemiologists/Data Managers Staff Researcher Barbara Garcia, MPA Tomás Aragón, MD, DrPH Susan Scheer, PhD, MPH Ling Hsu, MPH Henry Fisher Raymond, DrPH, MPH Willi McFarland, MD, PhD, MPH Sandra Schwarcz, MD, MPH Anthony Buckman, MS Viva Delgado, MPH Mike Grasso, MPH Theresa Ick Maree Kay Parisi Miao-Jung Mia Chen, PhD, MPH Yea-Hung Chen, MS Jennie CS Chin, MBA Priscilla Lee Chu, DrPH, MPH Anne Hirozawa, MPH Alison Hughes, MPH Tim Kellogg, MA Kara O Keefe, PhD, MPH Sharon Pipkin, MPH Tony Su, MPH Annie Vu, MPH Sean Arayasrikul Kristina Dhillon Aida Flandez Kelly Johnson Lauren Johnson Zachary Matheson Patrick Norton, PhD, MA Janella Parucha Gloria Posadas Rolando Ramirez John Rivie Rufina San Juan Nashanta Stanley Belinda Van Conrad Wenzel Maya Yoshida-Cervantes Adam Zeboski Erin Wilson, DrPH, MPH

6 i 2012 Annual Report HIV Epidemiology San Francisco Contents Contents.... i List of Figures and Tables.... ii Abbreviations....vii Executive Summary.... ix 1. Overview of HIV in San Francisco Trends in HIV Diagnoses Continuum of HIV Care Survival among Persons with HIV Infection Stage 3 (AIDS) Trends in HIV Mortality Opportunistic Illnesses and Associated Mortality Use of Antiretroviral Therapy among Persons with HIV Health Insurance Status at Time of HIV Diagnosis HIV among Men Who Have Sex with Men HIV among Injection Drug Users HIV among Heterosexuals HIV among Women HIV among Children, Adolescents and Young Adults HIV among the Aging Population HIV among Transgender Persons HIV among Homeless Persons Sexually Transmitted Diseases among Persons with HIV Geographic Distribution of HIV Technical Notes Data Tables...65

7 ii List of Figures and Tables 1. Overview of HIV in San Francisco Table 1.1 Characteristics of living HIV cases and newly diagnosed HIV cases in San Francisco, California and the United States...1 Figure 1.1 HIV infection stage 3 (AIDS) cases, deaths, and prevalence, , San Francisco....2 Figure 1.2 New HIV diagnoses, deaths, and prevalence, , San Francisco...3 Table 1.2 Table 1.3 Trends in persons diagnosed with HIV infection by demographic and risk characteristics, , San Francisco...4 Trends in persons living with HIV by demographic and risk characteristics, , San Francisco...5 Table 1.4 Characteristics of persons living with HIV as of December 2012, San Francisco...6 Figure 1.3 Estimated number of new HIV infections, , San Francisco...7 Table 1.5 Estimated rate of new HIV infections per 100,000 population by gender, race/ethnicity, age group, and exposure category in San Francisco, Trends in HIV Diagnoses Figure 2.1 Number of cases diagnosed with HIV infection by race/ethnicity, , San Francisco.. 9 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Table 2.1 Annual rates of male cases diagnosed with HIV infection per 100,000 population by race/ ethnicity, , San Francisco...10 Annual rates of female cases diagnosed with HIV infection per 100,000 population by race/ ethnicity, , San Francisco...10 Number of male cases diagnosed with HIV infection by exposure category, , San Francisco...11 Number of female cases diagnosed with HIV infection by exposure category, , San Francisco...11 Cases diagnosed with HIV infection by gender and age at diagnosis, , San Francisco Continuum of HIV Care Figure 3.1 Spectrum of engagement in care among persons diagnosed with HIV, , San Francisco...13 Table 3.1 Care and prevention indicators among new HIV diagnoses and living HIV cases, , San Francisco....14

8 iii 2012 Annual Report HIV Epidemiology San Francisco Figure 3.2 Comparison in proportions of cases receiving HIV care through January and April 2012 in San Francisco by case ownership...15 Figure 3.3 Living San Francisco HIV cases by care and current residence status, Survival among Persons with HIV Infection Stage 3 (AIDS) Figure 4.1 Kaplan-Meier survival curves for persons diagnosed with HIV infection stage 3 (AIDS) in four time periods, San Francisco...17 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Kaplan-Meier survival curves for persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 by race/ethnicity, San Francisco...18 Kaplan-Meier survival curves for persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 by exposure category, San Francisco...19 Kaplan-Meier survival curves for persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 by gender, San Francisco Five-year survival probability after HIV infection stage 3 (AIDS) for persons diagnosed between 2001 and 2012 by race/ethnicity, exposure category, and gender, San Francisco Trends in HIV Mortality Table 5.1 Deaths in persons with HIV infection, by demographic and risk characteristics, , San Francisco Figure 5.1 Table 5.2 Death rates due to due to HIV-related and non HIV-related causes among persons with HIV infection, , San Francisco...23 Underlying causes of death among persons with HIV infection, , San Francisco Table 5.3 Multiple causes of death among persons with HIV infection, , San Francisco Opportunistic Illnesses and Associated Mortality Table 6.1 Characteristics of persons diagnosed with HIV infection stage 3 (AIDS) and with OIs by calendar period of first OI diagnosis, , San Francisco...26 Table 6.2 Adults/adolescents diagnosed with HIV infection stage 3 (AIDS), by first OI and calendar period of OI diagnosis, , San Francisco...27 Figure 6.1 Survival time after first OI diagnosis for the 10 most frequently occurring OIs, , San Francisco Table 6.3 Five and ten-year survival probabilities after first OI diagnosis by calendar period, , San Francisco....30

9 iv List of Figures and Tables 7. Use of Antiretroviral Therapy among Persons with HIV Table 7.1 Estimate of ART use among persons living with HIV by demographic, risk, and socioeconomic characteristics, December 2012, San Francisco...31 Figure 7.1 Table 7.2 Figure 7.2 Table 7.3 Table 7.4 Estimate of ART use among living HIV cases with chart review by nadir CD4 level, December 2012, San Francisco...32 Trends in median CD4 count at time of diagnosis and at time of ART initiation among persons newly diagnosed with HIV, , San Francisco...33 Kaplan-Meier estimates of time from HIV diagnosis to ART initiation by year of HIV diagnosis and CD4 count at diagnosis among persons newly diagnosed with HIV, , San Francisco Trends in the number and percent of living HIV cases initiating ART each year among those who were ART naive at the beginning of each year, stratified by nadir CD4 count, , San Francisco...35 Characteristics of persons living with HIV between 2007 and 2011 who did not initiate ART, initiated ART at CD4 count 350 cells/µl or initiated ART at CD4 count above 350 cells/µl, San Francisco Health Insurance Status at Time of HIV Diagnosis Figure 8.1 Trends in health insurance status at time of HIV diagnosis by gender, , San Francisco...37 Figure 8.2 HIV cases by gender and health insurance status at time of HIV diagnosis, , San Francisco HIV among Men Who Have Sex with Men Figure 9.1 Cases diagnosed with HIV infection among MSM by race/ethnicity, , San Francisco...39 Figure 9.2 Figure 9.3 Percent of MSM reporting unprotected anal intercourse in the last six months by selfreported HIV status, the STOP AIDS Project, , San Francisco...40 Male rectal gonorrhea and male gonococcal proctitis among MSM by HIV serostatus, , San Francisco...41 Figure 9.4 Early syphilis among MSM by HIV serostatus, , San Francisco...42 Figure 9.5 Substance use among MSM, the STOP AIDS Project, , San Francisco HIV among Injection Drug Users Figure 10.1 Cases diagnosed with HIV infection among non-msm IDU by race/ethnicity, , San Francisco....44

10 v 2012 Annual Report HIV Epidemiology San Francisco Figure 10.2 HIV prevalence, incidence, self-report and unrecognized infection among non-msm IDU, , San Francisco HIV among Heterosexuals Figure 11.1 Cases diagnosed with HIV infection among heterosexuals by race/ethnicity, , San Francisco Figure 11.2 Syphilis among heterosexual men, , San Francisco Figure 11.3 Syphilis among women, , San Francisco HIV among Women Figure 12.1 Female cases diagnosed with HIV infection by race/ethnicity, , San Francisco...48 Figure 12.2 Figure 12.3 Living female HIV cases diagnosed through December 2012 and female population by race/ethnicity, San Francisco...49 Females living with HIV diagnosed through December 2012 by exposure category, San Francisco HIV among Children, Adolescents and Young Adults Table 13.2 Cases diagnosed with HIV infection among adolescents and young adults, , San Francisco and the United States...50 Table 13.1 Living young adult HIV cases by exposure category, gender, and race/ethnicity, December 2012, San Francisco Figure 13.1 Pediatric HIV cases by time period of HIV diagnosis, , San Francisco HIV among the Aging Population Figure 14.1 Number and percent of persons diagnosed with HIV infection at age 50 years and older, , San Francisco...52 Table 14.1 Characteristics of persons diagnosed with HIV infection in by age at diagnosis, San Francisco Table 14.2 Characteristics of living HIV cases by age group, December 2012, San Francisco HIV among Transgender Persons Table 15.1 Characteristics of transgender HIV cases compared to all HIV cases diagnosed in , San Francisco Table 15.2 Characteristics of transgender persons living with HIV, December 2012, San Francisco HIV among Homeless Persons Figure 16.1 Number and percent of homeless cases diagnosed with HIV infection by year of diagnosis, , San Francisco...55

11 vi List of Figures and Tables Table 16.1 Characteristics of homeless HIV cases compared to all HIV cases diagnosed in , San Francisco Sexually Transmitted Diseases among Persons with HIV Figure 17.1 Number of HIV cases diagnosed with an STD by year of STD diagnosis, , San Francisco...57 Table 17.1 HIV cases diagnosed with an STD by demographic characteristics, , San Francisco Geographic Distribution of HIV Map 18.1 Geographic distribution of persons living with HIV, December 2012, San Francisco.. 59 Map 18.2 Geographic distribution of HIV rates for newly diagnosed in per 100,000 population, San Francisco...60

12 vii 2012 Annual Report HIV Epidemiology San Francisco Abbreviations ART Antiretroviral therapy CDC Centers for Disease Control and Prevention CI Confidence Interval HAART Highly active antiretroviral therapy HSF Healthy San Francisco IDU Injection drug users KS Kaposi s sarcoma NHAS National HIV/AIDS Strategy NHBS National HIV behavioral surveillance MAC Mycobacterium avium complex MMP Medical Monitoring Project MSM Men who have sex with men MSM IDU Men who have sex with men who also inject drugs OI Opportunistic illness OOJ Out-of-jurisdiction PCP Pneumocystis jirovecii pneumonia SFDPH San Francisco Department of Public Health STD Sexually transmitted diseases UAI Unprotected anal intercourse

13

14 ix 2012 Annual Report HIV Epidemiology San Francisco Executive Summary San Francisco has been a leader in HIV prevention, care, and treatment from the start of the epidemic in the United States. Providing high quality HIV care, including antiretroviral therapy (ART) to all regardless of ability to pay, has been a top priority for the San Francisco Department of Public Health (SFDPH) because of ART s well documented effect on reducing morbidity and mortality in HIV-infected persons. The optimal time to initiate ART has changed over time. ART was originally recommended only for persons who were severely immunocomproised (either a CD4 cell count under 200 cells/µl or the presence of an AIDS opportunistic illness). However, as additional information on the adverse effects of HIV even in the presence of moderate immunodeficiency became available and as ART-related toxicity and pill burden have declined, the CD4 count at which ART was recommended increased. In 2011, evidence of the effect of ART on reduced risk of sexual transmission of HIV was published, adding to the debate over the risks of delaying therapy. In 2012, the U.S. Department of Health and Human Services (DHHS) changed its policy and recommended that ART be offered to all HIV-infected persons regardless of their CD4 count. In keeping with its leadership role, the SFDPH adopted a policy to offer ART to all its HIV-infected patients in 2010, two years prior to the DHHS recommendations. This recommendation was based both upon the evidence of treatment benefits available at that time and our knowledge that patients receiving care in the private sector of San Francisco were already receiving ART at CD4 cell counts above the recommended threshold. We were aware of this trend from the data available from the SFDPH HIV surveillance case registry. This recommendation was coupled with expanded efforts to increase HIV testing, to facilitate entry into care, and to encourage remaining in or reengaging in care a policy which became known as Test and Treat and was adopted nationally. In our 2012 HIV Epidemiology Annual Report, we provide detailed information on the trends in the CD4 counts at the time of ART initiation and examine our data for evidence of disparities in early use of ART in order to assess the need for interventions to optimize treatment outcomes. We explored these trends in the years prior to and following the change in treatment recommendations ( ). Our findings demonstrated important trends. First, we found that the CD4 count at the time of diagnosis has increased among persons diagnosed from 2007 through 2011, suggesting that the test part of the Test and Treat strategy appears to be working. We also found that for patients newly diagnosed with HIV whose CD4 count at diagnosis was above 350 cells/µl, the CD4 count at the time of ART initiation increased. In addition, the time from HIV diagnosis to the start of treatment decreased for persons who were diagnosed with higher CD4 cell counts. Furthermore, among patients who had been diagnosed prior to 2007 but had not yet started treatment, there was also evidence of initiating treatment at higher CD4 counts over time. Taken together, these findings also support the success of the treat part of the Test and Treat approach. While we are encouraged by this evidence, there is still much that remains to be done. A higher proportion of African Americans, persons aged years and aged 50 years and older, and those with unknown health insurance at the time of diagnosis did not initiate ART and they were also less likely to initiate ART at higher CD4 counts, highlighting the need to address disparities in early treatment. Over time we expect the trends towards early initiation of ART to continue. Our surveillance data will allow us to measure this trend and its impact on new infections.

15

16 Annual Report HIV Epidemiology San Francisco Overview of HIV in San Francisco Since the epidemic began, HIV/AIDS surveillance has been conducted in San Francisco through active and passive methods and evaluated on a regular basis. New in this report, the term AIDS is replaced by HIV infection stage 3 (AIDS) (see Technical Notes Stage of Disease at Diagnosis of HIV Infection ). As of December 31, 2012, there were 15,705 San Francisco residents diagnosed with HIV infection who were alive (Table 1.1). This accounts for 13% of California s HIV living cases and 2% of persons living with HIV reported nationally. Compared to cases reported in California and the United States, persons living with HIV in San Francisco were more likely to be male, white, men who have sex with men (MSM), and MSM who also inject drugs (MSM IDU). Compared to all persons living with HIV, newly diagnosed San Francisco HIV cases included more people of color and fewer injection drug users (IDU). Compared to newly diagnosed national HIV cases, newly diagnosed HIV cases in San Francisco were more likely to be male, white, and MSM. Table 1.1 Characteristics of living HIV cases and newly diagnosed HIV cases in San Francisco, California and the United States Living HIV Cases Newly Diagnosed HIV Cases San Francisco 1 California 2 United States 3 San Francisco 1, 2012 United States 3, 2011 (N = 15,705) (N = 117,213) (N = 898,529) (N = 413) (N = 42,842) % % % % % Gender Male 92% 87% 75% 94% 79% Female 6% 12% 25% 5% 21% Transgender 4 2% 1% -- <1% -- Race/Ethnicity White 62% 44% 33% 46% 28% African American 13% 18% 43% 11% 46% Latino 17% 33% 20% 26% 21% Asian/Pacific Islander 5% 4% 1% 12% 2% Native American 1% <1% <1% 2% <1% Other/Unknown 2% 1% 2% 3% 2% Exposure Category MSM 73% 66% 43% 76% 50% IDU 6% 7% 13% 3% 4% MSM IDU 15% 8% 5% 9% 2% Heterosexual 3% 9% 19% 5% 15% Other/Unidentified 2% 10% 20% 7% 29% 1 San Francisco data are reported through March 12, 2013 for cases diagnosed through December 31, California data are reported through December California data on newly diagnosed HIV cases are not available. 3 U.S. data are reported through June 30, 2012 and reflect cases diagnosed through December 31, U.S. data reflect unadjusted numbers for 50 states and 6 dependent areas and may be found in the CDC HIV Surveillance Report, 2011; vol Transgender data are not reported by the United States. See Technical Notes Transgender Status.

17 2 Overview of HIV in San Francisco The number of HIV infection stage 3 (AIDS) cases diagnosed each year among San Francisco residents reached a peak of 2,329 cases in 1992 and has declined since then (Figure 1.1). The sharpest decline in HIV infection stage 3 (AIDS) deaths occurred between 1995 and 1997, reflecting the impact of combination antiretroviral therapies (ART). Since 1999, slight declines have continued in both cases and deaths; however, delays in reporting affect the number of cases and deaths for recent years. Therefore, the numbers of cases and deaths for 2011 and 2012 may be revised upward in future reports. The number of San Franciscans living with HIV infection stage 3 (AIDS) has continued to rise every year since This is due to effective ART and a lower number of deaths than new cases each year. There were 9,607 San Francisco residents living with HIV infection stage 3 (AIDS) by the end of Figure 1.1 HIV infection stage 3 (AIDS) cases, deaths, and prevalence, , San Francisco 10,000 Number of Cases/Deaths 8,000 6,000 4,000 2, Year of Diagnosis/Death Persons living with HIV infection stage 3 HIV infection stage 3 cases HIV infection stage 3 deaths

18 Annual Report HIV Epidemiology San Francisco Figure 1.2 illustrates the number of persons newly diagnosed with HIV infection, number of deaths each year (lines), and number of persons living with HIV infection between 2006 and 2012 (bars). The date of HIV diagnosis for newly diagnosed cases was determined based on the earliest date of any of the following: positive HIV antibody test, viral load or CD4 test, initiation of ART, or patient self-report of a positive HIV test. The number of new HIV diagnoses shown by year includes persons who were diagnosed in that year with HIV, persons initially diagnosed with HIV infection stage 3 (AIDS), or persons initially diagnosed with HIV (stage 1, 2, unknown) and developed stage 3 in a later year. The number of new HIV diagnoses declined between 2007 and 2011 and leveled off in The number of deaths was steady from 2006 to 2007 and has declined since. For recent years, the number of cases diagnosed and deaths may be lower due to reporting delay. The number of living cases by year includes persons who were diagnosed with HIV during or prior to the year shown and not known to have died by the end of that year. Persons living with HIV increased from 14,425 in 2006 to 15,705 in The increasing number of living cases is a reflection of a steady addition of newly diagnosed cases over time coupled with a decline in deaths in each year. These data only include persons who have been diagnosed and reported to the health department. HIV-infected persons who are unaware of their infection and persons diagnosed with an anonymous HIV test are not included unless they also tested confidentially or entered care in the city. These figures therefore underestimate the true prevalence and incidence of HIV. Figure 1.2 Number of New HIV Diagnoses/Deaths New HIV diagnoses, deaths, and prevalence, , San Francisco 25, , ,425 14,636 14,885 15,089 15,273 15,469 15,705 15, , ,000 Number of Living HIV Cases Year Living HIV cases New HIV diagnoses Deaths 0

19 4 Overview of HIV in San Francisco Table 1.2 shows the characteristics of persons diagnosed with HIV between 2008 and The majority were male, white, age years, and MSM. Race/ethnicity distributions were fairly similar during 2008 to 2012, but data for 2012 diagnoses suggest slight increases in proportions of Latinos and Asian/Pacific Islanders. With respect to age, data across the five years imply slightly increasing proportions of new diagnoses among persons aged years. There were no children (<13 years) diagnosed with HIV during these years. Proportions of HIV diagnoses among IDU (both non-msm IDU and MSM IDU) declined during this time period. Table 1.2 Trends in persons diagnosed with HIV infection by demographic and risk characteristics, , San Francisco Year of Initial HIV Diagnosis Total Number Gender Male 89% 91% 90% 87% 94% Female 7% 5% 8% 11% 5% Transfemale 2 3% 4% 3% 2% <1% Race/Ethnicity White 50% 52% 48% 52% 46% African American 16% 15% 14% 16% 11% Latino 23% 20% 24% 20% 26% Asian/Pacific Islander 8% 8% 10% 8% 12% Native American 1% 0% 1% 0% 2% Other/Unknown 3% 4% 4% 4% 3% Age at HIV Diagnosis (years) % 0% 0% 0% 0% % 0% <1% <1% 0% % 12% 12% 10% 14% % 13% 13% 15% 16% % 31% 31% 26% 31% % 27% 29% 32% 28% 50+ 9% 17% 14% 17% 12% Exposure Category MSM 74% 72% 65% 70% 76% IDU 5% 5% 8% 6% 3% MSM IDU 11% 15% 14% 11% 9% Heterosexual 6% 4% 8% 6% 5% Other/Unidentified 4% 4% 5% 6% 7% 1 Data include persons diagnosed with HIV infection in any stage and reported as of March 12, Percentages may not add to 100% due to rounding. 2 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

20 Annual Report HIV Epidemiology San Francisco The number of persons living with HIV continues to increase due to ongoing incidence of HIV combined with an increase in survival after diagnosis. Persons were counted as living in a year if their HIV diagnosis date was in or before that year and they were known to be alive at the end of the year. As of December 31, 2012, 15,705 San Francisco residents were living with HIV (Table 1.3). Demographic and risk characteristics of persons living with HIV remained mostly stable between 2009 and 2012; cases are predominately white, age 50 years and above, and MSM (including MSM IDU). This table demonstrates aging of persons living with HIV, with decreasing proportions in the and years age groups accompanied by persons aged 50 years or above rising from 42% to 51% between 2009 and Table 1.3 Trends in persons living with HIV by demographic and risk characteristics, , San Francisco Number (%) Number (%) Number (%) Number (%) Gender Male 13,872 ( 92 ) 14,044 ( 92 ) 14,215 ( 92 ) 14,446 ( 92 ) Female 863 ( 6 ) 874 ( 6 ) 899 ( 6 ) 909 ( 6 ) Transfemale ( 2 ) 355 ( 2 ) 355 ( 2 ) 350 ( 2 ) Race/Ethnicity White 9,484 ( 63 ) 9,542 ( 62 ) 9,644 ( 62 ) 9,726 ( 62 ) African American 2,005 ( 13 ) 2,019 ( 13 ) 2,033 ( 13 ) 2,046 ( 13 ) Latino 2,491 ( 17 ) 2,567 ( 17 ) 2,619 ( 17 ) 2,699 ( 17 ) Asian/Pacific Islander 736 ( 5 ) 771 ( 5 ) 796 ( 5 ) 842 ( 5 ) Native American 79 ( 1 ) 81 ( 1 ) 83 ( 1 ) 91 ( 1 ) Other/Unknown 294 ( 2 ) 293 ( 2 ) 294 ( 2 ) 301 ( 2 ) Age in Years (at end of each year) ( <1 ) 5 ( <1 ) 4 ( <1 ) 3 ( <1 ) ( <1 ) 12 ( <1 ) 11 ( <1 ) 8 ( <1 ) ( 1 ) 161 ( 1 ) 147 ( 1 ) 150 ( 1 ) ( 3 ) 440 ( 3 ) 435 ( 3 ) 444 ( 3 ) ,285 ( 15 ) 2,107 ( 14 ) 1,957 ( 13 ) 1,887 ( 12 ) ,771 ( 38 ) 5,660 ( 37 ) 5,463 ( 35 ) 5,150 ( 33 ) 50+ 6,385 ( 42 ) 6,888 ( 45 ) 7,452 ( 48 ) 8,063 ( 51 ) Exposure Category MSM 10,982 ( 73 ) 11,129 ( 73 ) 11,300 ( 73 ) 11,515 ( 73 ) IDU 1,027 ( 7 ) 1,026 ( 7 ) 1,017 ( 7 ) 998 ( 6 ) MSM IDU 2,347 ( 16 ) 2,344 ( 15 ) 2,336 ( 15 ) 2,338 ( 15 ) Heterosexual 438 ( 3 ) 465 ( 3 ) 487 ( 3 ) 503 ( 3 ) Transfusion/Hemophilia 28 ( <1 ) 26 ( <1 ) 26 ( <1 ) 26 ( <1 ) Other/Unidentified 267 ( 2 ) 283 ( 2 ) 303 ( 2 ) 325 ( 2 ) Total 15,089 15,273 15,469 15,705 1 Persons living with HIV at the end of each year. 2 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

21 6 Overview of HIV in San Francisco Among living male HIV cases diagnosed in San Francisco, the majority were white and MSM (Table 1.4). White and African American male HIV cases had similar age distributions at the end of 2012, while Latino, Asian/Pacific Islander, and Native American males were younger. Among living female HIV cases, injection drug use was the predominant exposure category for white and African American women while heterosexual sex was the predominant exposure category for Latina, Asian/ Pacific Islander, and Native American women. Age 50 years and older is the largest age category for both men and women living with HIV (52% and 47%, respectively). Table 1.4 Characteristics of persons living with HIV as of December 2012, San Francisco African Asian/Pacific Islander White American Latino & Native American Number (%) Number (%) Number (%) Number (%) Male Exposure category MSM 7,625 ( 81) 866 ( 56) 1,966 ( 81) ) IDU 205 ( 2) 245 ( 16) 61 ( 3) 21 3) MSM IDU 1439 ( 15) 315 ( 20) 283 ( 12) 91 11) Heterosexual 30 ( <1) 70 ( 5) 47 ( 2) 9 1) Total Number 1 ( 11,330 ( 540 ( 2,181 ( 160 Transfusion/Hemophilia 6 ( <1) 3 ( <1) 2 ( <1) 4 ( <1) 15 Other/Unidentified 84 ( 1) 50 ( 3) 54 ( 2) 16 ( 2) 220 Age in Years (at end of 2012) ( 0) 0 ( 0) 0 ( 0) 0 ( 0) ( 0) 0 ( 0) 1 ( <1) 1 ( <1) ( <1) 30 ( 2) 41 ( 2) 12 ( 1) ( 2) 47 ( 3) 111 ( 5) 48 ( 6) ( 9) 134 ( 9) 460 ( 19) 186 ( 23) ,937 ( 31) 466 ( 30) 922 ( 38) 314 ( 38) 4, ,420 ( 58) 872 ( 56) 878 ( 36) 265 ( 32) 7,515 Male Subtotal 9,389 1,549 2, ,446 Female Exposure category IDU 156 ( 60) 212 ( 56) 60 ( 34) 16 ( 23) 456 Heterosexual 75 ( 29) 131 ( 35) 85 ( 48) 42 ( 59) 337 Transfusion/Hemophilia 5 ( 2) 2 ( 1) 2 ( 1) 2 ( 3) 11 Other/Unidentified 26 ( 10) 31 ( 8) 30 ( 17) 11 ( 15) 105 Age in Years (at end of 2012) ( 0) 1 ( <1) 1 ( 1) 0 ( 0) ( 0) 0 ( 0) 5 ( 3) 0 ( 0) ( 2) 6 ( 2) 7 ( 4) 2 ( 3) ( 3) 9 ( 2) 14 ( 8) 0 ( 0) ( 12) 43 ( 11) 31 ( 18) 16 ( 23) ( 39) 108 ( 29) 47 ( 27) 27 ( 38) ( 44) 209 ( 56) 72 ( 41) 26 ( 37) 429 Female Subtotal Transfemale Total 9,726 2,046 2, ,705 1 Includes persons with multiple race or whose race/ethnicity information is not available. 2 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population. See Technical Notes Transgender Status.

22 Annual Report HIV Epidemiology San Francisco HIV incidence estimates The SFDPH serves as one of the 25 national HIV incidence surveillance sentinel sites monitoring the number and rates of new HIV infections. Estimates of new infections track the leading edge of the HIV epidemic and are critical for allocating resources and evaluating effectiveness of prevention programs. To identify incident HIV cases, blood from newly diagnosed HIV individuals is retested using a laboratory assay (called BED) that classifies individuals as having either a recently acquired HIV infection or a longstanding infection. These results are used with a statistical adjustment for HIV testing history to estimate HIV incidence. We applied this method, developed by the CDC, to data. Overall, the estimated number of new HIV infections has remained relatively stable since 2007 (Figure 1.3). While there are fluctuations in the estimates, the confidence intervals overlap from year to year indicating that the changes in incidence are not statistically significant. Figure 1.3 Estimated number of new HIV infections, , San Francisco 800 Number of Estimated Incident Cases (CI: ) 477 (CI: ) 372 (CI: ) 365 (CI: ) 437 (CI: ) Year CI: Confidence Interval.

23 8 Overview of HIV in San Francisco Table 1.5 presents the estimated rate per 100,000 of new infections by gender, race/ethnicity, age group, and exposure category. The rate of infection among MSM are disproportionately high; the rate of infection per 100,000 MSM in 2011 was 782 compared to an overall rate of 62 in San Francisco. With caution given to the large margin of error, the data suggest higher incidence among African Americans and Latinos compared to whites for the years we are able to provide an estimate. Table 1.5 Estimated rate of new HIV infections per 100,000 population by gender, race/ ethnicity, age group, and exposure category in San Francisco, Rate per 100,000 Rate per 100,000 Rate per 100,000 Rate per 100,000 Rate per 100,000 (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) Overall (47-94) (50-86) (37-69) (31-72) (41-83) Sex at Birth Male (85-175) (94-163) (70-132) (52-182) (75-155) Female ** ** ** ** ** Race/Ethnicity White (49-127) (56-110) (39-89) (25-84) (43-108) African American ** (33-192) (33-198) ** ** Latino (30-179) (71-197) (19-108) (27-193) (24-163) Other 33 (10-57) Age (years) ** ** ** ** (69-204) (64-161) (53-149) (30-160) (50-188) (44-132) (56-126) (36-100) (18-95) (26-106) (23-128) (55-144) (23-95) (31-142) (35-136) 50+ ** ** ** ** ** Exposure Category 1 MSM (545-1,138) (601-1,061) ( ) ( ) (505-1,058) Non-MSM 10 (2-17) ** ** ** ** ** Incidence estimate not calculated due to incomplete data. 1 MSM includes MSM IDU; Non-MSM includes heterosexuals, non-msm IDU, and other.

24 Annual Report HIV Epidemiology San Francisco Trends in HIV Diagnoses Race/ethnicity Trends by race/ethnicity category for cases diagnosed with HIV infection show that whites accounted for the majority of cases diagnosed between 2006 and 2012 (Figure 2.1). The number of white HIV cases declined for most of this time period and leveled off between 2010 and Numbers of HIV cases for other race/ethnicity groups remained fairly stable. Figure 2.1 Number of cases diagnosed with HIV infection by race/ethnicity, , San Francisco Number of Cases Year of HIV Diagnosis White African American Latino Other 1 1 Cases in the Other race/ethnicity category include 68% Asian/Pacific Islanders, 5% Native Americans, 22% Multi-race, and 5% Unknown race.

25 10 Trends in HIV Diagnoses Among men, the annual rates of HIV diagnosis are highest in African Americans, although rates have declined from 2006 to 2012 (Figure 2.2). There was also a declining trend in rates of HIV diagnosis for white men during 2006 to 2010, and the rates have leveled since HIV rates for Latino men have decreased between 2006 and 2011 but show an increase in the most recent year Rates among men in other race/ethnicity group remained fairly level in this time period but also show an increase in the most recent year. In 2012, the rate of HIV diagnosis per 100,000 population was 169 among African American men, 142 among Latino men, and 103 among white men. In San Francisco, rates of HIV diagnosis are much lower in women compared to men. Among women, the rates of HIV diagnosis from 2006 to 2012 were highest for African Americans with rates peaking in 2008 and 2011 (Figure 2.3). Rates for Latino women show signs of increase in recent years of 2011 and Rates for white women were relatively level throughout the time period. In 2012, the rate of HIV diagnosis per 100,000 population was 27 for African American women, 15 for Latino women, 3 for white women, and 2 for women of other race/ethnicity group. Figure 2.2 Rate per 100,000 Figure 2.3 Rate per 100, Annual rates 1 of male cases diagnosed with HIV infection per 100,000 population by race/ethnicity, , San Francisco Year of HIV Diagnosis White African American Latino Other 2 1 See Technical Notes HIV Case Rates. 2 Cases in the Other race/ethnicity category include 69% Asian/Pacific Islanders, 5% Native Americans, 21% Multi-race, and 5% unknown. Annual rates 1 of female cases diagnosed with HIV infection per 100,000 population by race/ethnicity, , San Francisco Year of HIV Diagnosis White African American Latina Other 2 1 See Technical Notes HIV Case Rates. 2 Cases in the Other race/ethnicity category include 59% Asian/Pacific Islanders, 4% Native Americans, 30% Multi-race, and 7% unknown.

26 Annual Report HIV Epidemiology San Francisco Exposure category Most males diagnosed with HIV infection in San Francisco are MSM. In recent years, trends in the number of male cases diagnosed with HIV infection for exposure categories other than MSM non-idu were relatively stable (Figure 2.4). The annual number of MSM non-idu HIV cases declined from 2006 through 2010 and increased since then. In 2012, 81% of male HIV cases were MSM non-idu, 10% were MSM IDU, and 3% were heterosexual IDU. Figure 2.4 Number of male cases diagnosed with HIV infection 1 by exposure category, , San Francisco Number of Cases Year of HIV Diagnosis MSM IDU MSM IDU Other 1 Excludes transfemale cases diagnosed with HIV infection. Includes persons with HIV by year of their initial HIV diagnosis. Beginning in 2007, the number of annual female cases diagnosed with HIV infection due to heterosexual contact showed a similar trend as the number of IDU cases, with heterosexual cases slightly exceeding IDU cases each year (Figure 2.5). In 2012, 55% of female HIV cases were infected due to heterosexual contact and 14% were due to IDU. Figure 2.5 Number of female cases diagnosed with HIV infection 1 by exposure category, , San Francisco Number of Cases Year of HIV Diagnosis IDU Heterosexual Other 1 Excludes transmale cases diagnosed with HIV infection. Includes persons with HIV by year of their initial HIV diagnosis.

27 12 Trends in HIV Diagnoses Age Table 2.1 shows the annual number of HIV diagnosis between 2009 and 2012 by gender and age at HIV diagnosis. Among males, proportions of each age group were relatively stable from year to year, with the leading number of cases in the years age group followed by the years age group. Among females, the proportion of HIV diagnoses in females 40 years and older increased from 59% in 2009 to 77% in Table 2.1 Cases diagnosed with HIV infection by gender* and age at diagnosis, , San Francisco Year of Initial HIV Diagnosis Number (%) Number (%) Number (%) Number (%) Male (Age in years) ( 0 ) 0 ( 0 ) 0 ( 0 ) 0 ( 0 ) ( 0 ) 2 ( 1 ) 1 ( 0 ) 0 ( 0 ) ( 12 ) 44 ( 11 ) 35 ( 10 ) 55 ( 14 ) ( 13 ) 49 ( 13 ) 55 ( 15 ) 63 ( 16 ) ( 31 ) 127 ( 33 ) 97 ( 27 ) 125 ( 32 ) ( 29 ) 111 ( 29 ) 111 ( 31 ) 104 ( 27 ) ( 15 ) 50 ( 13 ) 56 ( 16 ) 43 ( 11 ) Male Subtotal 413 ( 100 ) 383 ( 100 ) 355 ( 100 ) 390 ( 100 ) Female (Age in years) ( 0 ) 0 ( 0 ) 0 ( 0 ) ( 0 ) 0 ( 0 ) 0 ( 0 ) 0 ( 0 ) ( 13 ) 3 ( 9 ) 5 ( 11 ) 2 ( 9 ) ( 4 ) 3 ( 9 ) 4 ( 9 ) 1 ( 5 ) ( 25 ) 5 ( 15 ) 6 ( 14 ) 2 ( 9 ) ( 13 ) 11 ( 33 ) 17 ( 39 ) 11 ( 50 ) ( 46 ) 11 ( 33 ) 12 ( 27 ) 6 ( 27 ) Female Subtotal 24 ( 100 ) 33 ( 100 ) 44 ( 100 ) 22 ( 100 ) * Transgender data by age are not presented in the table due to small numbers and potential small population.

28 Annual Report HIV Epidemiology San Francisco Continuum of HIV Care Spectrum of engagement in care To improve health outcomes for HIV-infected persons, rapid entry into care, ongoing engagement in care, and use of ART to achieve viral suppression are required. The SFDPH monitors these outcomes using reports of CD4 and viral load tests. The number of persons newly diagnosed with HIV decreased from 479 in 2009 to 423 in 2011 (Figure 3.1). Over these years, the proportion of newly diagnosed persons who entered care within six months of diagnosis remained stable (~88%) (1). However, not all persons who entered care, continued to receive care; 72%-75% of persons who entered care remained in care three to six months after their first medical visit (2). For those who remained in care through a second medical visit, retention was higher (83% in 2009, 78% in 2010, and 77% in 2011) (3). Among persons who are retained in care for three visits, a very high proportion achieve viral suppression within 12 months of diagnosis and this proportion has increased from 77% among persons diagnosed in 2009 to 100% in 2011 (4). Unfortunately, because not all newly diagnosed persons enter or remain in care and others move outside of San Francisco, which impedes our ability to measure these outcomes, the overall proportion of newly diagnosed persons who achieved viral suppression is lower (47% -57%) (5). The increase in the proportion virally suppressed in more recent years is encouraging. Figure 3.1 Spectrum of engagement in care among persons diagnosed with HIV, , San Francisco Diagnoses 2010 Diagnoses 2011 Diagnoses 350 Number of Cases New diagnoses* Linked to care within 6 months of diagnosis (1) Retained in care for 3-6 months after linkage (2) Retained in care for 6-12 months after linkage (3) Viral suppression^ within 12 months among persons retained in care for 3 visits (4) Viral suppression^ within 12 months among all new diagnoses (5) * Number of new diagnoses shown each year is based on the evidence of a confirmed HIV test and does not take into account patient self-report of HIV infection. ^ Defined as the latest viral load test during the specified period 200 copies/ml.

29 14 Continuum of HIV Care In July 2010, the White House Office of National AIDS Policy released its National HIV/AIDS Strategy (NHAS) setting goals for increasing access to care for people living with HIV, optimizing their health outcomes and reducing HIV-related health disparities. To monitor the impact of the NHAS, HIV surveillance data are used for core outcome measures and indicators. We report on key HIV care and prevention indicators using routinely collected surveillance data from 2010 to 2011 in San Francisco. Between 2010 and 2011, the proportions of new cases who developed HIV infection stage 3 (AIDS) within 12 months of HIV diagnosis and persons who were homeless at diagnosis decreased (Table 3.1). Eighty-five percent of new cases were linked to care within three months of diagnosis. Timely treatment initiation is evident as the proportion of persons diagnosed with a higher CD4 count who initiated ART increased from 54% in 2010 to 66% in The proportions of persons who were virally suppressed increased slightly from 2010 to The mortality rates among persons living with HIV decreased between 2009 and While there is room for improvement, these indicators show that in San Francisco people living with HIV are accessing HIV care and that San Francisco is poised to measure and meet the goals of the NHAS. For example, the NHAS recommends that by 2015, 85% of all persons newly diagnosed with HIV are linked to a HIV care provider within three months of their diagnosis. We have already met that standard in San Francisco. We will continue to present these indicators each year in order to track progress in meeting the NHAS goals and to inform HIV prevention and care strategies in San Francisco. Table 3.1 Care and prevention indicators among new HIV diagnoses and living HIV cases, , San Francisco Year Indicators New HIV diagnoses N=451 N=423 Proportion developed AIDS within 12 months of diagnosis 32% 26% Proportion of homeless at diagnosis 12% 8% Proportion linked to care within 3 months of diagnosis 85% 85% Proportion starting treatment within 12 months of diagnosis among those diagnosed with a CD4 count>500 cells/ul 54% 66% Proportion virally suppressed 1 within 12 months of diagnosis 56% 57% Living HIV cases 2 N=14,747 N=14,999 Proportion of cases who had 1 CD4/viral load test 74% 73% Proportion received 2 tests among those with 1 test 80% 78% Proportion virally suppressed 1 among living cases 61% 62% Proportion virally suppressed 1 among those with 1 viral load test 84% 86% Mortality rates among HIV cases 3 1.6%(2009), 1.5%(2010) Not available Mortality rates among AIDS cases 3 2.3%(2009), 2.0%(2010) Not available 1 Defined as the latest viral load test during the specified period 200 copies/ml. 2 Includes persons who are alive as of the end of the year and have been diagnosed as of the end of the previous year. 3 Mortality rate was calculated as the number of HIV/AIDS cases who died during the year divided by the number of total HIV/AIDS cases alive anytime during that year.

30 Annual Report HIV Epidemiology San Francisco In- and out-migration for HIV care Out-of-jurisdiction HIV cases receiving medical care in San Francisco Routine HIV case surveillance assigns case ownership by residence at diagnosis. HIV cases residing in San Francisco at time of diagnosis are considered San Francisco cases. HIV cases currently receiving care in San Francisco but who were residing elsewhere at time of diagnosis are considered out-of-jurisdiction (OOJ) cases. We used two clinical datasets of patients who received HIV care in San Francisco from January-April 2012 to estimate the proportion of persons receiving care in San Francisco who are OOJ cases. First, we electronically linked data from the Medical Monitoring Project (MMP), a supplemental surveillance project that includes a representative sample of patients receiving outpatient HIV care in San Francisco (see Technical Notes Medical Monitoring Project ), to the HIV case registry. Second, we matched mandatorily reported electronic laboratory records of CD4 and HIV viral load tests from all persons tested at any clinical site in San Francisco to the HIV case registry. Twenty-seven percent (N=1,621) of the 6,103 patients sampled from the MMP medical sites and 27% (N=2,687) of the 10,015 cases from the SFDPH HIV laboratory database were determined to be OOJ cases (Figure 3.2). The similarity of these two results bolsters our confidence in estimating that approximately a quarter of the patients receiving HIV-related health care in San Francisco are OOJ cases. Figure 3.2 Comparison in proportions of cases receiving HIV care through January and April 2012 in San Francisco by case ownership Medical Monitoring Project Electronic Laboratory Reports (N=6,103) (N=10,015) OOJ cases 26.6% SF cases 73.1% Other 0.3% SF cases 73.2% OOJ cases 26.8%

31 16 Continuum of HIV Care We use data from the SFDPH case registry to assess 1) out-migration: San Francisco cases (residents of San Francisco at the time of HIV diagnosis) who moved outside of San Francisco following diagnosis, and 2) in-migration: non-san Francisco or OOJ cases (resident elsewhere at the time of HIV diagnosis, see Technical Notes Out-of-jurisdiction cases ) who received care in San Francisco. Information on residence is initially collected at the time of diagnosis and is updated through periodic medical chart review and from laboratory reports that include current address. To estimate out-migration, we included San Francisco HIV cases who were alive as of December 31, 2012 and were diagnosed through December 31, 2011; those who had at least one CD4 or viral load test in 2012 were considered to be in care. Among San Francisco cases not in care, we used the most current address to determine if they had moved out of San Francisco. There were 15,243 living San Francisco cases in 2012; 71% received care (including 63% who received care in the city and 8% who received care elsewhere) (Figure 3.3). Among the 4,396 San Francisco cases not in care, 1,102 moved outside San Francisco, 1,456 had a current San Francisco address and the remaining did not have current address information (lost-to-follow-up). To estimate in-migration, we used HIV-related laboratory tests performed in 2012 that matched OOJ cases to identify OOJ cases receiving care in San Francisco. Among the 12,845 cases that received care in 2012, 3,287 (26%) were OOJ cases. There are a significant number of OOJ cases who receive care in San Francisco. In addition, for San Francisco cases who are not in care, priority of re-engagement in care can be made to include those who have a known San Francisco address in recent years. To further understand HIV case migration and care utilization patterns, complete and current laboratory and residence information needs to be collected and shared between jurisdictions. Figure 3.3 Living San Francisco HIV cases* by care and current residence status, 2012 N=15,243 Received care elsewhere 8% Not in care, lostto-follow-up 12% Not in care, moved 7% Received care in SF 63% Not in care, current SF address 10% * Include persons who are alive as of December 2012 and have been diagnosed as of December 2011.

32 Annual Report HIV Epidemiology San Francisco Survival among Persons with HIV Infection Stage 3 (AIDS) The Kaplan-Meier survival curves in Figure 4.1 demonstrate the continual improvement in survival after HIV infection stage 3 (AIDS) since the 1980s. Survival was poor for persons diagnosed in the first ten years of the epidemic ( ) in which only 50% of persons with HIV infection stage 3 (AIDS) survived to 18 months (median survival time). Between 1990 and 1995, survival improved; median survival time was 38 months. Survival among HIV infection stage 3 (AIDS) cases diagnosed in the two time periods after 1995 continued to improve due to the widespread availability of highly active antiretroviral therapy (HAART). For HIV infection stage 3 (AIDS) diagnoses in 1996 to 2000, 79% survived 60 months (5 years) after diagnosis. Among persons diagnosed with HIV infection stage 3 (AIDS) in the most recent time period, 2001 to 2012, 84% were alive 60 months after diagnosis. Figure % Kaplan-Meier survival 1 curves for persons diagnosed with HIV infection stage 3 (AIDS) in four time periods, San Francisco 80% Survival Probability 60% 40% 20% 0% Months of Survival Diagnosed in (N=8,607) Diagnosed in (N=12,080) Diagnosed in (N=3,723) Diagnosed in (N=4,925) 1 See Technical Notes HIV infection stage 3 (AIDS) Survival.

33 18 Survival among Persons with HIV Infection Stage 3 (AIDS) Survival after HIV infection stage 3 (AIDS) diagnosis is worse for African Americans than other race/ethnic groups (Figure 4.2). Among African Americans diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012, 78% survived to 60 months (5 years), compared to 85% of whites, 86% of Latinos, and 90% of Asian/Pacific Islanders. Figure 4.2 Kaplan-Meier survival 1 curves for persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 by race/ethnicity, San Francisco 100% 80% Survival Probability 60% 40% 20% 0% Months of Survival White (N=2,672) Latino (N=916) African American (N=875) Asian/Pacific Islander (N=299) 1 See Technical Notes HIV infection stage 3 (AIDS) Survival.

34 Annual Report HIV Epidemiology San Francisco Survival after HIV infection stage 3 (AIDS) diagnosis has been better for MSM and heterosexuals compared to MSM IDU and non-msm IDU. For HIV infection stage 3 (AIDS) cases diagnosed from 2001 to 2012, the 5-year (60 months) survival probability was 88% for MSM, 83% for heterosexuals, 80% for MSM IDU, and 69% for non-msm IDU (Figure 4.3). Worse survival among IDU partly reflects higher death rates from causes associated with drug use such as overdose, liver disease, and other infections. Figure 4.3 Kaplan-Meier survival 1 curves for persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 by exposure category, San Francisco 100% 80% Survival Probability 60% 40% 20% 0% Months of Survival MSM (N=3,072) IDU (N=532) MSM IDU (N=965) Heterosexual (N=235) 1 See Technical Notes HIV infection stage 3 (AIDS) Survival.

35 20 Survival among Persons with HIV Infection Stage 3 (AIDS) Male HIV infection stage 3 (AIDS) cases have better survival than female and transfemale cases (Figure 4.4). The 5-year (60 months) survival probability among cases diagnosed 2001 to 2012 was 85% for men, 77% for women and 78% for transfemale persons. The differences in survival by gender are consistent with lower use of HAART among female and transfemale HIV infection stage 3 (AIDS) cases. Figure 4.4 Kaplan-Meier survival 1 curves for persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 by gender, San Francisco 100% 80% Survival Probability 60% 40% 20% 0% Months of Survival 2 Male (N=4,369) Female (N= 388) Transfemale (N=165) 1 See Technical Notes HIV infection stage 3 (AIDS) Survival. 2 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

36 Annual Report HIV Epidemiology San Francisco Overall, 84% of persons diagnosed with HIV infection stage 3 (AIDS) between 2001 and 2012 survived to five years (Figure 4.5). Differences in survival occurred across race/ethnicity, exposure category, and gender groups. African Americans, IDU, MSM IDU, women, and transfemale persons with HIV infection stage 3 (AIDS) had lower 5-year survival compared to other groups. Figure % 90% 80% 5-year Survival 70% 60% 50% 40% 30% 20% 10% 0% Five-year survival probability 1 after HIV infection stage 3 (AIDS) for persons diagnosed between 2001 and 2012 by race/ethnicity, exposure category, and gender, San Francisco 85% 78% 86% 90% 88% 69% 80% 83% 85% 77% 78% 84% 0% Race Exposure Category Gender Overall 1 Calculated from Kaplan-Meier method. 2 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

37 22 Trends in HIV Mortality As of December 31, 2012, the cumulative number of deaths that occurred among San Francisco HIV cases was 20,198 (Table 5.1). Death reporting for 2011 and 2012 is not yet complete. The proportion of deaths was stable by gender and race/ethnicity groups. By exposure category, the proportion of deaths that were heterosexual IDU increased from 2009 to Deaths continue to shift to older age groups, with more than two-third of deaths in persons age 50 years and older in The majority of deaths continue to occur among persons with HIV infection stage 3 (AIDS). Table 5.1 Deaths in persons with HIV infection, by demographic and risk characteristics, , San Francisco Year of Death Cumulative Totals as of Number (%) Number (%) Number (%) Number (%) 12/31/2012 Gender Male 222 ( 88 ) 211 ( 86 ) 184 ( 88 ) 159 ( 90 ) 19,186 Female 21 ( 8 ) 23 ( 9 ) 18 ( 9 ) 12 ( 7 ) 767 Transfemale 2 8 ( 3 ) 10 ( 4 ) 7 ( 3 ) 6 ( 3 ) 245 Race/Ethnicity White 155 ( 62 ) 147 ( 60 ) 111 ( 53 ) 109 ( 62 ) 14,823 African American 52 ( 21 ) 45 ( 18 ) 51 ( 24 ) 34 ( 19 ) 2,517 Latino 28 ( 11 ) 27 ( 11 ) 28 ( 13 ) 26 ( 15 ) 2,100 Other 16 ( 6 ) 25 ( 10 ) 19 ( 9 ) 8 ( 5 ) 758 Exposure Category MSM 141 ( 56 ) 129 ( 53 ) 115 ( 55 ) 100 ( 56 ) 14,871 IDU 28 ( 11 ) 34 ( 14 ) 35 ( 17 ) 32 ( 18 ) 1,603 MSM IDU 65 ( 26 ) 64 ( 26 ) 52 ( 25 ) 36 ( 20 ) 3,197 Heterosexual 13 ( 5 ) 9 ( 4 ) 3 ( 1 ) 4 ( 2 ) 213 Other/Unidentified 4 ( 2 ) 8 ( 3 ) 4 ( 2 ) 5 ( 3 ) 314 Age at Death (years) ( 2 ) 5 ( 2 ) 3 ( 1 ) 2 ( 1 ) 1, ( 12 ) 10 ( 4 ) 25 ( 12 ) 10 ( 6 ) 7, ( 29 ) 62 ( 25 ) 52 ( 25 ) 34 ( 19 ) 7, ( 32 ) 101 ( 41 ) 76 ( 36 ) 63 ( 36 ) 3, ( 25 ) 66 ( 27 ) 53 ( 25 ) 68 ( 38 ) 1,254 HIV Disease Stage Stage 1, 2, or unknown 31 ( 12 ) 40 ( 16 ) 33 ( 16 ) 29 ( 16 ) 388 Stage 3 (AIDS) 220 ( 88 ) 204 ( 84 ) 176 ( 84 ) 148 ( 84 ) 19,810 Total 251 ( 100 ) 244 ( 100 ) 209 ( 100 ) 177 ( 100 ) 20,198 1 Data in recent years are incomplete due to reporting delay. In addition, deaths that occurred outside of San Francisco are primarily identified through matching with the National Death Index (NDI) which is complete only through December 31, Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

38 Annual Report HIV Epidemiology San Francisco The trend in death rates in persons diagnosed with HIV infection was examined by the single, underlying cause of death for each person. The death rate due to HIV-related causes among persons with HIV infection stage 3 (AIDS) declined from 3.1 per 100 persons in 1999 to 0.9 per 100 persons in 2010 (Figure 5.1). Non-HIV-related causes of death among persons with HIV infection stage 3 (AIDS) fluctuated between 1.0 and 1.3 deaths per 100 persons from 2000 to When deaths in all stages of HIV infection are evaluated, rates of deaths for HIV-related causes were similar to rates for causes not due to HIV. Two thousand ten is the first year in which the death rate due to non-hiv-related causes of death exceeded the rate due to HIV-related causes of death for HIV infection stage 3 (AIDS) cases and for all HIV infection stages in aggregate. Figure Death rates 1 due to due to HIV-related and non HIV-related causes among persons with HIV infection, , San Francisco Death Rate (%) Year HIV causes of death among persons with HIV infection stage 3 (AIDS) Non-HIV causes of death among persons with HIV infection stage 3 (AIDS) HIV causes of death among persons with HIV infection Non-HIV causes of death among persons with HIV infection 1 Death rates are calculated as the number of persons with HIV infection stage 3 (AIDS), or all HIV infection stages, who died each year divided by the number of total HIV infection stage 3 (AIDS), or HIV infection all stage, cases alive during that year. See Technical Notes for Causes of Death.

39 24 Trends in HIV Mortality Underlying causes of death among persons with HIV continue to shift towards non-hiv related causes. The proportion of deaths in which HIV was listed as the underlying cause of death decreased from 70.5% of deaths among HIV cases occurring in to 48.2% in (Table 5.2). Other frequently occurring underlying causes of death in include non-aids cancers (11.2%), drug overdose (9.1%), heart disease (8.0%), and suicide (4.3%). The proportions of these non-hiv-related causes increased over the three time periods. Table 5.2 Underlying causes of death among persons with HIV infection 1, , San Francisco Year of Death N=1,357 N=1,300 N=1,060 Underlying Cause of Death 2 Number (%) Number (%) Number (%) HIV 957 ( 70.5 ) 821 ( 63.2 ) 511 ( 48.2 ) Non-AIDS cancer 89 ( 6.6 ) 126 ( 9.7 ) 119 ( 11.2 ) Lung cancer 25 ( 1.8 ) 41 ( 3.2 ) 40 ( 3.8 ) Liver cancer 20 ( 1.5 ) 24 ( 1.8 ) 19 ( 1.8 ) Anal cancer 7 ( 0.5 ) 5 ( 0.4 ) 8 ( 0.8 ) Colon cancer 2 ( 0.1 ) 6 ( 0.5 ) 7 ( 0.7 ) Hodgkins lymphoma 1 ( 0.1 ) 0 ( 0.0 ) 1 ( 0.1 ) Drug overdose 33 ( 2.4 ) 46 ( 3.5 ) 96 ( 9.1 ) Heart disease 64 ( 4.7 ) 87 ( 6.7 ) 85 ( 8.0 ) Coronary heart disease 43 ( 3.2 ) 51 ( 3.9 ) 40 ( 3.8 ) Cardiomyopathy 8 ( 0.6 ) 10 ( 0.8 ) 4 ( 0.4 ) Suicide 23 ( 1.7 ) 34 ( 2.6 ) 46 ( 4.3 ) Liver disease 34 ( 2.5 ) 27 ( 2.1 ) 24 ( 2.3 ) Liver cirrhosis 15 ( 1.1 ) 17 ( 1.3 ) 13 ( 1.2 ) Alcoholic liver disease 16 ( 1.2 ) 9 ( 0.7 ) 9 ( 0.8 ) Chronic obstructive lung disease 16 ( 1.2 ) 23 ( 1.8 ) 22 ( 2.1 ) Mental disorders due to substance use 24 ( 1.8 ) 32 ( 2.5 ) 20 ( 1.9 ) Cerebrovascular disease 13 ( 1.0 ) 10 ( 0.8 ) 11 ( 1.0 ) Viral hepatitis 14 ( 1.0 ) 9 ( 0.7 ) 9 ( 0.8 ) Renal disease 3 ( 0.2 ) 4 ( 0.3 ) 8 ( 0.8 ) Diseases of arteries 4 ( 0.3 ) 1 ( 0.1 ) 5 ( 0.5 ) Septicemia 2 ( 0.1 ) 4 ( 0.3 ) 3 ( 0.3 ) Diabetes 3 ( 0.2 ) 7 ( 0.5 ) 2 ( 0.2 ) 1 Deceased cases diagnosed with HIV infection that lack cause of death information are not represented in this table. 2 See Technical Notes Causes of Death. Aspergillosis was not an underlying cause of death among cases diagnosed with HIV infection from

40 Annual Report HIV Epidemiology San Francisco Table 5.3 shows both underlying and contributory causes of death among persons diagnosed with HIV infection. Similar to the trend in underlying causes of death, the proportion of deaths with HIV-related causes fell from 85% in to 63% in Causes of death that had persistent proportional increases across the three time periods include heart disease, non-aids cancers, drug overdose, suicide, and chronic obstructive lung disease. Deaths due to drug overdose showed the largest percentage increase between recent time periods and Table 5.3 Multiple causes of death among persons with HIV infection 1, , San Francisco Year of Death N=1,357 N=1,300 N=1,060 Multiple Causes of Death 2 Number ( % ) Number ( % ) Number ( % ) HIV 1,157 ( 85.3 ) 1,040 ( 80.0 ) 662 ( 62.5 ) Heart disease 265 ( 19.5 ) 267 ( 20.5 ) 235 ( 22.2 ) Coronary heart disease 69 ( 5.1 ) 84 ( 6.5 ) 67 ( 6.3 ) Cardiomyopathy 31 ( 2.3 ) 26 ( 2.0 ) 17 ( 1.6 ) Non-AIDS cancer 127 ( 9.4 ) 179 ( 13.8 ) 157 ( 14.8 ) Lung cancer 27 ( 2.0 ) 45 ( 3.5 ) 44 ( 4.2 ) Liver cancer 24 ( 1.8 ) 27 ( 2.1 ) 24 ( 2.3 ) Anal cancer 9 ( 0.7 ) 9 ( 0.7 ) 10 ( 0.9 ) Colon cancer 4 ( 0.3 ) 6 ( 0.5 ) 8 ( 0.8 ) Hodgkins lymphoma 6 ( 0.4 ) 4 ( 0.3 ) 8 ( 0.8 ) Pneumonia 211 ( 15.5 ) 167 ( 12.8 ) 119 ( 11.2 ) Liver disease 224 ( 16.5 ) 178 ( 13.7 ) 116 ( 10.9 ) Liver cirrhosis 86 ( 6.3 ) 84 ( 6.5 ) 54 ( 5.1 ) Alcoholic liver disease 19 ( 1.4 ) 12 ( 0.9 ) 10 ( 0.9 ) Drug overdose 45 ( 3.3 ) 53 ( 4.1 ) 104 ( 9.8 ) Viral hepatitis 180 ( 13.3 ) 180 ( 13.8 ) 99 ( 9.3 ) Mental disorders due to substance use 83 ( 6.1 ) 105 ( 8.1 ) 87 ( 8.2 ) Renal disease 115 ( 8.5 ) 137 ( 10.5 ) 85 ( 8.0 ) Septicemia 133 ( 9.8 ) 149 ( 11.5 ) 79 ( 7.5 ) Chronic obstructive lung disease 46 ( 3.4 ) 67 ( 5.2 ) 65 ( 6.1 ) Suicide 23 ( 1.7 ) 34 ( 2.6 ) 46 ( 4.3 ) Diabetes 30 ( 2.2 ) 36 ( 2.8 ) 36 ( 3.4 ) Cerebrovascular disease 43 ( 3.2 ) 39 ( 3.0 ) 30 ( 2.8 ) Diseases of arteries 12 ( 0.9 ) 4 ( 0.3 ) 12 ( 1.1 ) Aspergillosis 17 ( 1.3 ) 7 ( 0.5 ) 4 ( 0.4 ) 1 Deceased cases diagnosed with HIV infection that lack cause of death information are not represented in this table. 2 Includes underlying and contributory causes of death. Individuals may have more than one cause of death. See Technical Notes Causes of Death.

41 26 Opportunistic Illnesses and Associated Mortality At the beginning of the HIV epidemic, opportunistic illnesses (OIs) were the most common cause of morbidity and mortality among HIV-infected persons (1). Despite the advent of azidothymidine (AZT) in 1987 and steady improvements in OI prophylaxis and treatment, OIs remained a serious health threat until the introduction in 1996 of combination antiretroviral therapy, after which there was a dramatic decline in OI incidence and substantial improvement in survival (2). Even with these gains, OIs remain a threat to HIVinfected persons. The SFDPH is unique insofar it is one, if not the only, health department in the country to have an HIV surveillance system that has collected initial and subsequent occurrence of OIs since We analyzed trends in the first OI diagnosis and survival thereafter among 20,860 adults and adolescents diagnosed with an OI between 1980 and Across three treatment periods , , and the majority of persons diagnosed with OIs were aged 25 and 44 years (78%, 73%, and 57%, respectively), white non-hispanic (85%, 76%, and 58%, respectively), male (99%, 97%, and 92%, respectively), and men who have sex with men (83%, 77%, and 60%, respectively) (Table 6.1). Over time, increases in the proportions of persons diagnosed with OIs occurred among persons 45 years and older, non-whites, females, and injection drug users. Table 6.1 Characteristics of persons diagnosed with HIV infection stage 3 (AIDS) and with OIs by calendar period of first OI diagnosis, , San Francisco ART-Naïve Period ART, Non-HAART Period ART, HAART Period Total (N=3,004) (N=14,097) (N=3,759) (N=20,860) Number (%) Number (%) Number (%) Number (%) Age at diagnosis (years) ( 2) 164 ( 1) 44 ( 1) 274 ( 1) ,342 ( 78) 10,301 ( 73) 2,149 ( 57) 14,792 ( 71) 45 or over 596 ( 20) 3,632 ( 26) 1,566 ( 42) 5,794 ( 28) Race/ethnicity Hispanic 245 ( 8) 1,612 ( 11) 629 ( 17) 2,486 ( 12) Black, non-hispanic 166 ( 6) 1,625 ( 12) 755 ( 20) 2,547 ( 12) White, non-hispanic 2,561 ( 85) 10,665 ( 76) 2,171 ( 58) 15,399 ( 74) Other 32 ( 1) 195 ( 1) 204 ( 5) 431 ( 2) Sex at birth Male 2,985 ( 99) 13,685 ( 97) 3,453 ( 92) 20,126 ( 96) Female 19 ( 1) 412 ( 3) 306 ( 8) 737 ( 4) HIV transmission category MSM 2,487 ( 83) 10,844 ( 77) 2,241 ( 60) 15,575 ( 75) IDU 32 ( 1) 876 ( 6) 519 ( 14) 1,427 ( 7) MSM & IDU 448 ( 15) 2,102 ( 15) 798 ( 21) 3,348 ( 16) Heterosexual 11 (<1) 126 ( 1) 129 ( 3) 266 ( 1) Other/Unknown 26 ( 1) 149 ( 1) 72 ( 2) 247 ( 1) ART: antiretroviral therapy; HAART: highly active antiretroviral therapy (1) Changes in AIDS-defining illnesses in a London Clinic, Mocroft A, Sabin CA, Youle M, Madge S, Tyrer M, Devereux H, Deayton J, Dykhoff A, Lipman MC, Phillips AN, Johnson MA. J Acquir Immune Defic Syndr Aug 15;21(5): (2) AIDS-defining opportunistic illnesses in US patients, : a cohort study. Kate Buchacz, Rose K. Baker, Frank J. Palella, Jr, Joan S. Chmiel, Kenneth A. Lichtenstein, Richard M. Novak, Kathleen C. Wood, John T. Brooks, HOPS Investigators AIDS June 19; 24(10):

42 Annual Report HIV Epidemiology San Francisco Between 1980 and 2012, the most common OIs diagnosed were Pneumocystis jirovecii pneumonia (PCP) (N=8,163), Kaposi s sarcoma (KS) (N=4,195), HIV wasting syndrome (N=1,864), candida of esophagus (N=1,381), and Mycobacterium avium complex (MAC) (N=1,069) (Table 6.2). The number of PCP diagnoses has been consistently the highest across the three treatment periods. Table 6.2 Adults/adolescents diagnosed with HIV infection stage 3 (AIDS), by first OI and calendar period of OI diagnosis, , San Francisco ART-Naïve Period ART, Non-HAART Period ART, HAART Period Total (N=3,004) (N=14,097) (N=3,759) (N=20,860) Number (%) Number (%) Number (%) Number (%) Initial OI 1 Pneumocystis jirovecii pneumonia (PCP) 1,640 ( ) 5,379 ( ) 1,144 ( ) 8,163 ( ) Kaposi s sarcoma (KS) 952 ( ) 2,768 ( ) 475 ( ) 4,195 ( ) HIV wasting syndrome 35 ( 1.17 ) 1,365 ( 9.68 ) 464 ( ) 1,864 ( 8.94 ) Candida, esophagus 100 ( 3.33 ) 910 ( 6.46 ) 371 ( 9.87 ) 1,381 ( 6.62 ) Mycobacterium avium complex (MAC) 80 ( 2.66 ) 834 ( 5.92 ) 155 ( 4.12 ) 1,069 ( 5.12 ) HIV encephalopathy 35 ( 1.17 ) 715 ( 5.07 ) 182 ( 4.84 ) 932 ( 4.47 ) Cryptococcosis 106 ( 3.53 ) 558 ( 3.96 ) 219 ( 5.83 ) 883 ( 4.23 ) Cryptosporidiosis 64 ( 2.13 ) 503 ( 3.57 ) 127 ( 3.38 ) 694 ( 3.33 ) Immunoblastic lymphoma 72 ( 2.40 ) 449 ( 3.19 ) 155 ( 4.12 ) 676 ( 3.24 ) Cytomegalovirus 91 ( 3.03 ) 455 ( 3.23 ) 97 ( 2.58 ) 643 ( 3.08 ) Recurrent pneumonia 8 ( 0.27 ) 307 ( 2.18 ) 226 ( 6.01 ) 541 ( 2.59 ) Toxoplasmosis brain 60 ( 2.00 ) 334 ( 2.37 ) 64 ( 1.70 ) 458 ( 2.20 ) Cytomegalovirus retinitis 19 ( 0.63 ) 378 ( 2.68 ) 56 ( 1.49 ) 453 ( 2.17 ) Mycobacterium tuberculosis, pulmonary 5 ( 0.17 ) 291 ( 2.06 ) 135 ( 3.59 ) 431 ( 2.07 ) Mycobacterium tuberculosis, disseminated 14 ( 0.47 ) 210 ( 1.49 ) 54 ( 1.44 ) 278 ( 1.33 ) Burkett lymphoma 26 ( 0.87 ) 144 ( 1.02 ) 69 ( 1.84 ) 239 ( 1.15 ) Progressive multifocal leukoencephalopathy 16 ( 0.53 ) 109 ( 0.77 ) 26 ( 0.69 ) 151 ( 0.72 ) Herpes simplex 49 ( 1.63 ) 148 ( 1.05 ) 54 ( 1.44 ) 241 ( 1.16 ) Mycobacterium other disseminated 9 ( 0.30 ) 94 ( 0.67 ) 20 ( 0.53 ) 123 ( 0.59 ) Primary brain lymphoma 14 ( 0.47 ) 81 ( 0.57 ) 24 ( 0.64 ) 119 ( 0.57 ) Candida trachea 7 ( 0.23 ) 49 ( 0.35 ) 30 ( 0.80 ) 86 ( 0.41 ) Histoplasmosis 3 ( 0.10 ) 60 ( 0.43 ) 15 ( 0.40 ) 78 ( 0.37 ) Isosporiasis 1 ( 0.03 ) 27 ( 0.19 ) 8 ( 0.21 ) 36 ( 0.17 ) Coccidoido mycosis 2 ( 0.07 ) 20 ( 0.14 ) 10 ( 0.27 ) 32 ( 0.15 ) Salmonella septicemia 3 ( 0.10 ) 14 ( 0.10 ) 2 ( 0.05 ) 19 ( 0.09 ) Cancer cervix 0 ( 0.00 ) 4 ( 0.03 ) 8 ( 0.21 ) 12 ( 0.06 ) 1 Cases may have more than one initial OI and are sorted in descending order of the total case count.

43 28 Opportunistic Illnesses and Associated Mortality We examined survival after first OI diagnosis and displayed Kaplan-Meier survival curves for the ten most diagnosed OIs. We found that survival time after first OI diagnosis in the HAART period ( ) significantly increased compared with survival time in previous years for all the OIs. In the HAART period, the median survival time for persons with the two most common OIs, PCP and KS, was more than 15 years compared with less than two years in the previous treatment periods (Figure 6.1). Although there has been an improvement in survival in the HAART period for MAC, HIV encephalopathy, and immunoblastic lymphoma, the median survival times for these OIs remain low at approximately five years. Figure 6.1 Survival time after first OI diagnosis for the 10 most frequently occurring OIs, , San Francisco a. Pneumocystis jirovecii pneumonia b. Kaposi s sarcoma c. HIV wasting d. Candida of esophagus

44 Annual Report HIV Epidemiology San Francisco e. Mycobacterium avium complex f. HIV encephalopathy g. Cryptococcosis h. Cryptosporidiosis i. Immunoblastic lymphoma j. Cytomegalovirus

45 30 Opportunistic Illnesses and Associated Mortality Table 6.3 shows the 5-year and 10-year survival probabilities for the HIV/AIDS patients with ten most frequently occurring OIs (presented in descending order by total case count). Persons with PCP experienced the greatest improvement in survival between and , with 5-year survival increasing from 1% to 69%. Persons with cryptosporidiosis experienced the largest survival probabilities compared with other OIs: in , the 5-year survival probability was 81% and the 10-year probability 74%. Persons whose first OI diagnosis was HIV wasting syndrome experienced the least improvement in survival probabilities over the three time periods. The survival improvements, especially in the most recent time period, , likely reflect a combination of factors: the advent of combination antiretroviral therapy, improved prophylaxis and treatment of OIs, and earlier diagnosis of HIV infection and linkage to effective care. Table 6.3 Five and ten-year survival probabilities after first OI diagnosis by calendar period, , San Francisco 5-year survival probability [95%confidence interval] 10-year survival probability [95%confidence interval] Pneumocystis jirovecii pneumonia (N=8,163) [ ] [ ] [ ] [ ] [ ] [ ] Kaposi s sarcoma (N=4,195) [ ] [ ] [ ] [ ] [ ] [ ] HIV wasting syndrome (N=1,864) [ ] [ ] [ ] [ ] [ ] [ ] Candida, esophagus (N=1,381) [ ] [ ] [ ] [ ] [ ] [ ] Mycobacterium avium complex (N=1,069) [ ] [ ] [ ] [ ] [ ] [ ] HIV encephalopathy (N=932) [ ] [ ] [ ] [ ] [ ] [ ] Cryptococcosis (N=883) * [ ] [ ] [ ] [ ] [ ] Cryptosporidiosis (N=694) [ ] [ ] [ ] [ ] [ ] [ ] Immunoblastic lymphoma (N=676) * [ ] [ ] [ ] [ ] [ ] Cytomegalovirus (N=643) [ ] [ ] [ ] [ ] [ ] [ ] * No persons first diagnosed with cryptococcosis or immunoblastic lymphoma between survived to ten years (ten-year survival probability = 0.00)

46 Annual Report HIV Epidemiology San Francisco Use of Antiretroviral Therapy among Persons with HIV Table 7.1 shows estimates of ART use among persons living with HIV as of December 31, Information on ART is obtained from medical chart review and persons whose medical record indicating that they were prescribed ART are assumed to have received it (see Technical Notes Estimate of ART Use ). The lower level estimate shown in the table was calculated among all cases living with HIV (N=15,705). The upper level estimate was calculated among living cases for whom a chart review was completed between January 2011 and March 2013 (N=8,777). Persons without follow-up information within the last two years or those known to have moved out of San Francisco were excluded from the upper level estimate calculation. Overall, 83%-91% of persons living with HIV received ART. ART use was lower among females, persons with race/ethnicity other than white, IDU, heterosexuals, homeless and under-insured persons. Table 7.1 Estimate of ART 1 use among persons living with HIV by demographic, risk, and socioeconomic characteristics, December 2012, San Francisco Percent Receiving ART Lower Level Estimate Upper Level Estimate Overall 83% 91% Gender Male 84% 91% Female 81% 85% Transfemale 2 83% 92% Race/Ethnicity White 86% 92% African American 79% 87% Latino 81% 90% Asian/Pacific Islander 79% 89% Native American 70% 80% Multiple race 79% 86% Exposure category MSM 84% 91% IDU 80% 90% MSM IDU 84% 91% Heterosexual 82% 86% Housing status, most recent Housed 85% 92% Homeless 60% 72% Insurance at HIV/AIDS diagnosis Private 89% 94% Public 83% 88% None 80% 90% 1 See Technical Notes Estimate of ART Use. 2 Transfemale data include all transgender cases. Transmale data are not released separately due to the potential small population size. See Technical Notes Transgender Status.

47 32 Use of Antiretroviral Therapy among Persons with HIV Figure 7.1 shows ART use by the lowest CD4 count ever reported (nadir CD4) among persons living with HIV who have had follow-up information within the last two years and whose chart review was completed between January 2011 and March 2013 (N=8,777). The proportion receiving ART was greater among persons with lower CD4 count: 97% of cases with a CD4 count below 200 cells/µl, 92% with a CD4 count between cells/µl, 82% with a CD4 count between cells/µl, and 65% with a CD4 count above 500 cells/µl received ART. Figure % 80% Estimate of ART 1 use among living HIV cases with chart review by nadir CD4 level, December 2012, San Francisco 97% 92% 82% % Receiving ART 60% 40% 20% 65% 0% < >500 CD4 Count (cells/μl) 1 See Technical Notes Estimate of ART Use.

48 Annual Report HIV Epidemiology San Francisco In 2010, the SFDPH adopted a policy to offer ART to all its HIV-infected patients regardless of their CD4 cell count. We analyzed HIV surveillance data to examine trends in CD4 count at ART initiation, time from HIV diagnosis to ART initiation and characteristics of persons who initiated ART at higher CD4 count between 2007 and Among persons newly diagnosed with HIV, the median CD4 count at HIV diagnosis increased from 364 cells/µl in 2007 to 435 cells/µl in 2011 (Table 7.2). The median CD4 count at ART initiation increased from 347 cells/µl in 2007 to 404 cells/µl in 2011 among persons whose CD4 count at diagnosis was between cells/µl, and increased from 523 cells/µl in 2007 to 719 cells/µl in 2011 among persons whose CD4 count at diagnosis was >500 cells/µl. Among persons whose CD4 count at diagnosis was <351 cells/µl, the median CD4 count at ART initiation did not change significantly. Table 7.2 Year of HIV diagnosis Trends in median CD4 count at time of diagnosis and at time of ART initiation among persons newly diagnosed with HIV, (N=2,138), San Francisco Median CD4 count at HIV diagnosis Median CD4 count at ART initiation by CD4 count at diagnosis (cells/µl) among persons initiating ART (N=1,203) Number of total cases Cells/µL < >

49 34 Use of Antiretroviral Therapy among Persons with HIV The time between date of HIV diagnosis and date of ART initiation was calculated and compared using Kaplan-Meier product limit method. Among persons newly diagnosed with HIV whose CD4 count at diagnosis was <200 cells/µl, the time from HIV diagnosis to ART initiation did not change significantly over time (Figure 7.2a). In contrast, the time to ART initiation decreased significantly for persons who were diagnosed at a higher CD4 count. Among those diagnosed with a CD4 count between cells/µl, the median time to ART initiation decreased from eight months in 2007 to three months in 2011 (Figure 7.2b). Among newly diagnosed persons whose CD4 count at diagnosis was between cells/µl, the median time to ART decreased from 23 months in 2007 to five months in 2011 (Figure 7.2c). In the group diagnosed with a CD4 count >500 cells/µl, the median time to ART decreased from 37 months in 2007 to eight months in 2011 (Figure 7.2d). Figure 7.2 Kaplan-Meier estimates of time from HIV diagnosis to ART initiation by year of HIV diagnosis and CD4 count at diagnosis among persons newly diagnosed with HIV, , San Francisco a. CD4 count <200 cells/µl b. CD4 count cells/µl c. CD4 count cells/µl d. CD4 count >500 cells/µl

50 Annual Report HIV Epidemiology San Francisco Among persons who were living with HIV, we calculated the number and percent who initiated ART during each calendar year between 2007 and 2010 among those who were ART naïve at the beginning of each year and had at least one CD4 or viral load test during the year. We further stratified the number of persons living with HIV who started ART in each year by four levels of the lowest (nadir) CD4 count observed during the year (<200, , , and >500 cells/µl). For each of these four nadir CD4 groups, the denominator was the number of persons who were alive during the year and not known to have started ART at the beginning of the year. Among persons living with HIV whose nadir CD4 count was <351 cells/µl, the proportion initiating ART fluctuated between 2007 and 2010 (Table 7.3). Among persons living with HIV whose nadir CD4 was between cells/µl, the number and proportion of persons starting ART increased from 91 (17%) in 2007 to 141 (30%) in Among persons living with HIV whose nadir CD4 was >500 cells/µl, the number and proportion who starting ART increased from 46 (8%) in 2007 to 125 (18%) in We did not include data for the year of 2011 because ascertainment of ART initiation in 2011 is incomplete. Table 7.3 Trends in the number and percent of living HIV cases initiating ART each year among those who were ART naive at the beginning of each year, stratified by nadir CD4 count, , San Francisco Nadir CD4 count among living cases (cells/µl) < >500 Living cases Starting ART Living cases Starting ART Living cases Starting ART Living cases Starting ART Number Number (%) Number Number (%) Number Number (%) Number Number (%) Calendar year (55) (35) (17) ( 8) (57) (35) (19) ( 13) (55) (43) (21) ( 15) (50) (37) (30) ( 18)

51 36 Use of Antiretroviral Therapy among Persons with HIV Among persons living with HIV during 2007 and 2011 who were not known to have started ART before 2007, 43% did not initiate ART during this time period, 33% started treatment at CD4 count 350 cells/ µl, and 23% began treatment at a higher CD4 count (Table 7.4). There were socio-demographic differences in overall ART initiation as well as the CD4 count at time of initiation. Notably, a higher proportion of African Americans (49%), persons aged years (46%) and aged 50 years or above (47%), and persons with unknown health insurance (65%) did not initiate treatment. Initiation of ART at CD4 count >350 cells/µl was more frequent among whites, younger persons, MSM who did not inject drugs, the privately insured, and those who did not reside in an impoverished neighborhood. Table 7.4 Characteristics of persons living with HIV between 2007 and 2011 who did not initiate ART, initiated ART at CD4 count 350 cells/µl or initiated ART at CD4 count above 350 cells/µl, San Francisco Total cases Cases who did not initiate ART Cases who initiated ART at CD4 350 Cases who initiated ART at CD4 >350 Number % % % Total 4,223 43% 33% 23% Gender Male 3,785 43% 33% 24% Female % 34% 17% Transgender % 38% 20% Race/Ethnicity White 2,350 43% 31% 25% African American % 32% 18% Latino % 40% 24% Other % 37% 18% Age (years) % 29% 25% ,328 41% 33% 26% ,363 42% 35% 23% % 36% 18% Exposure category MSM 2,867 43% 32% 25% IDU % 39% 17% MSM IDU % 36% 22% Other/Unknown % 34% 20% Insurance at time of diagnosis Public % 42% 22% Private 1,702 43% 31% 26% None 1,268 43% 36% 20% Unknown % 12% 23% Homeless at time of diagnosis Yes % 30% 23% No 3,867 43% 34% 23% Resident of impoverished neighborhood 1 Yes 1,141 44% 36% 21% No 3,082 43% 33% 24% 1 Defined as a person whose most recent residence in a census tract where more than 20% of adults had a median household income below the U.S. poverty level.

52 Annual Report HIV Epidemiology San Francisco Health Insurance Status at Time of HIV Diagnosis Insurance status at time of initial HIV diagnosis differs between men, women, and transfemales. The proportion of men with private insurance was consistently higher than the proportions of women and transfemales with private insurance (Figure 8.1). The proportions of men with private or public insurance at diagnosis were fairly stable between 2008 and For females, public forms of insurance consistently accounted for the most frequent type of insurance at diagnosis each year. Insurance at diagnosis trends for transfemales fluctuate due to small numbers diagnosed each year. Compared to males and females diagnosed with HIV, transfemales had the highest proportion with no insurance coverage between 2008 and At the same time, the proportion of transfemales with public insurance increased and peaked in 2009 at 67%. In this time period, there were no transfemales with private insurance at diagnosis. Figure % Trends in health insurance status at time of HIV diagnosis by gender, , San Francisco Male Female Transfemale 1 100% 100% 80% 80% 80% Percent 60% 40% 60% 40% 60% 40% 20% 20% 20% 0% % % Public Private None Missing Public Private None Missing Public Private None Missing 1 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

53 38 Health Insurance Status at Time of HIV Diagnosis Figure 8.2 shows the distribution of insurance types by gender for HIV cases diagnosed in Female and transfemale cases diagnosed during this time period had higher proportions that were underinsured (i.e., having no insurance or public insurance) compared to males. Females had the highest proportion using Medicaid (a state-sponsored insurance for persons meeting financial criteria) at time of HIV diagnosis followed by transfemales. In addition, Healthy San Francisco (HSF), the county-sponsored health care program for residents that became available in 2008, was utilized by close to 20% of females and transfemales at time of diagnosis. Figure 8.2 HIV cases by gender and health insurance status at time of HIV diagnosis, , San Francisco Male (N=1,128) Female (N=99) Transfemale 1 (N=19) VA 1.2% Medicare 0.9% Medicaid 7.1% Other Public (not specified) 4.0% Private 37.8% Medicaid 30.3% Other Public (not specified) 11.1% Medicaid 10.5% Medicare 5.3% Other Public (not specified) 15.8% HSF 10.8% Jail 1.0% Private 9.1% None 31.6% HSF 19.2% HSF 21.1% Missing 14.0% None 24.2% Missing 10.1% 1 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status. None 19.2% Missing 15.8%

54 Annual Report HIV Epidemiology San Francisco HIV among Men Who Have Sex with Men HIV surveillance data Whites accounts for the largest number of HIV cases by race/ethnicity among MSM in San Francisco (Figure 9.1). Newly diagnosed white MSM cases declined from 2006 to 2010 and leveled off since then. Diagnoses among Latinos, African Americans and other race/ethnicity groups were fairly stable over 2006 to Figure Cases diagnosed with HIV infection 1 among MSM by race/ethnicity, , San Francisco Number of Cases Year of HIV Diagnosis White African American Latino Other 1 Includes MSM and MSM IDU with HIV infection by year of their initial HIV diagnosis.

55 40 HIV among Men Who Have Sex with Men HIV sexual behavior data The STOP AIDS Project collects information on sexual behaviors and self-reported HIV status of MSM who participate in their outreach prevention activities in San Francisco. These data are collected anonymously and shared with the SFDPH to track trends in HIV-related risk behavior. Such data may not be representative of all MSM in San Francisco. In this section, trends in unprotected anal intercourse (UAI) in the past six months are assessed for MSM who are 18 years and older and reside in San Francisco. Figure 9.2 shows trends in reported UAI (either insertive or receptive) by self-reported HIV serostatus. Between 2006 and 2012, the percent of HIV-negative MSM who reported UAI steadily declined from 42% in 2006 to 32% in 2010, but rose to a six year high of 48% in 2011 and then slightly declined to 43% in Among HIV-positive men, the percent UAI fluctuated from a low of 45% in 2008 to a high of 63% in 2010 and dropped to 38% in Figure % 60% Percent of MSM reporting unprotected anal intercourse in the last six months by self-reported HIV status, the STOP AIDS Project, , San Francisco 57% 60% 60% 63% 56% Percent 50% 40% 30% 42% 36% 45% 37% 33% 32% 48% 38% 43% 20% 10% 0% Year HIV Positive HIV Negative

56 Annual Report HIV Epidemiology San Francisco Sexually transmitted diseases among MSM The impact of sexually transmitted diseases (STD) on HIV transmission is complex. On the one side, rises may be markers for increased potential to transmit HIV through increased unprotected sex and a synergistic effect of STD-HIV co-infection. On the other side, the potential increase in HIV transmission may be mitigated by higher ART coverage and serosorting. Figure 9.3 shows trends in male rectal gonorrhea and male gonococcal proctitis among MSM in San Francisco from 2003 through 2012 by HIV serostatus. Data on male rectal gonorrhea originate from case reporting by laboratories and health providers throughout the city. Data on male gonococcal proctitis originate from the municipal STD clinic only and represent men with symptomatic infection. Among men, rectal gonorrhea is a biological marker for unprotected receptive anal sex. The last four years has seen an increase in reported cases of male rectal gonorrhea. The stable numbers of cases of male gonococcal proctitis suggest that some of the increase in reported male rectal gonorrhea may be due to increased screening or reporting. Data may underestimate true levels of infections due to several factors, including lack of rectal screening by many health providers, under reporting, and a large proportion of cases who do not manifest symptoms. Figure Male rectal gonorrhea and male gonococcal proctitis among MSM by HIV serostatus, , San Francisco Number of Cases Year Male Rectal Gonorrhea (HIV+) Male Gonococcal Proctitis (HIV+) Male Rectal Gonorrhea (HIV-) Male Gonococcal Proctitis (HIV-)

57 42 HIV among Men Who Have Sex with Men Figure 9.4 shows trends in early syphilis cases (primary, secondary, and early latent) among MSM in San Francisco from 2003 through 2012 by HIV serostatus. Data originate from case reporting by laboratories and health providers throughout the city and from the municipal STD clinic, the site where most of the patients were diagnosed. Like gonorrhea, syphilis is a biological marker for unprotected sex. The number of early syphilis cases among MSM in San Francisco declined between 2003 and The increase from 2007 to 2012 in early syphilis is dramatic, especially among HIV-positive MSM who account for a greater proportion of early syphilis cases than HIV-negative MSM. Figure 9.4 Early syphilis among MSM by HIV serostatus, , San Francisco Number of Cases Year Primary (HIV+) Secondary (HIV+) Early Latent (HIV+) Primary (HIV-) Secondary (HIV-) Early Latent (HIV-)

58 Annual Report HIV Epidemiology San Francisco Substance use The STOP AIDS Project records substance use among San Francisco MSM. Figure 9.5 shows the percent of MSM who used methamphetamines, poppers, or cocaine in the past six months. Use of these substances has fluctuated over the period with little net change from 2006 to The most recent year however shows increase in poppers rising to 24.9% and cocaine to 15.5% while methamphetamine continues to trend downward to 6.0%. Figure 9.5 Substance use among MSM, the STOP AIDS Project, , San Francisco 30% 25% 24.9% 20% 20.2% 20.6% 19.9% 21.6% Percent 15% 10% 13.3% 13.0% 12.5% 10.3% 14.5% 8.9% 12.0% 9.6% 16.0% 10.8% 7.8% 13.6% 8.5% 15.5% 5% 8.2% 6.0% 0% Year Poppers Cocaine Methamphetamine

59 44 HIV among Injection Drug Users HIV surveillance data In recent years, the number of HIV diagnoses among IDU who are non-msm has been less than 10% of HIV cases diagnosed annually. Among this exposure category, whites accounted for the largest number of cases diagnosed with HIV infection each year between 2006 and 2008 (Figure 10.1). Beginning in 2009, the numbers of HIV cases among white and African American IDU s have been similar. The data in 2012 show similar numbers diagnosed for all race/ethnicity groups, although reporting is not complete. Figure 10.1 Cases diagnosed with HIV infection 1 among non-msm IDU by race/ ethnicity, , San Francisco Number of Cases Year of HIV Diagnosis White African American Latino Other 1 Includes persons with HIV infection by year of their initial HIV diagnosis.

60 Annual Report HIV Epidemiology San Francisco National HIV behavioral surveillance (NHBS) data NHBS has carried out three cross-sectional survey among IDU in San Francisco. HIV testing was conducted in 2009 and HIV prevalence among IDU in San Francisco appears to be stable at about 12%. However, self-reported HIV status and unrecognized HIV infection (i.e. HIV-positive persons who do not know they are HIV positive) suggest that a large proportion (41%) of IDU in San Francisco do not know that they are infected with HIV. Figure 10.2 HIV prevalence, incidence, self-report and unrecognized infection among non-msm IDU, , San Francisco 45% 40% 35% 36.0% 40.9% 30% 25% Percent 20% 15% 10% 5% 13.3% 13.7% 9.7% 11.6% 6.8% 0% IDU Survey Group in 2005 IDU Survey Group in 2009 IDU Survey Group in 2012 HIV-Positive Self-reported HIV-Positive Unrecognized HIV-Positive

61 46 HIV among Heterosexuals HIV surveillance data Trends in heterosexual HIV cases by race/ethnicity are difficult to characterize due to the small number of cases infected through heterosexual contact (Figure 11.1). African Americans accounted for the greatest number and proportion of heterosexual HIV cases from 2006 to 2012, although in recent years, whites, African Americans, and Latinos show similar numbers of diagnoses each year. Figure 11.1 Cases diagnosed with HIV infection 1 among heterosexuals by race/ ethnicity, , San Francisco Number of Cases Year of HIV Diagnosis White African American Latino Other 1 Includes persons with HIV infection by year of their initial HIV diagnosis.

62 Annual Report HIV Epidemiology San Francisco Sexually transmitted diseases among heterosexuals Figure 11.2 shows the annual number of primary, secondary, and early latent cases of syphilis among heterosexual men in San Francisco from 2003 through Data originate from case reporting from laboratories and health providers throughout the city, although the majority are patients seen at the municipal STD clinic. Compared to MSM, syphilis among heterosexual men remains relatively low in recent years with fluctuation. Figure 11.2 Syphilis among heterosexual men, , San Francisco Number of Cases Year Primary Secondary Early Latent 10 Figure 11.3 shows the annual number of primary, secondary, and early latent cases of syphilis among women in San Francisco from 2003 through Data originate from case reporting from laboratories and health providers throughout the city, although the majority are patients seen at the municipal STD clinic. Among women, syphilis cases have been low and stable in recent years, with a potential increase in recent years. Figure 11.3 Syphilis among women, , San Francisco Number of Cases Year Primary Secondary Early Latent

63 48 HIV among Women White and African American are the two largest race/ethnicity groups among women diagnosed with HIV infection (Figure 12.1). Recent years data suggest slight increases in annual Latina cases diagnosed. From 2006 to 2012, whites and African Americans accounted for 33% and 35% of female cases diagnosed with HIV infection, respectively. Figure 12.1 Female cases diagnosed with HIV infection 1 by race/ethnicity, , San Francisco Number of Cases Year of HIV Diagnosis White African American Latina Other 1 Includes persons with HIV infection by year of their initial HIV diagnosis.

64 Annual Report HIV Epidemiology San Francisco Compared to the female population of San Francisco, African American women are disproportionately affected by HIV (Figure 12.2). Although African American women represent 6% of the female population, as of December 31, 2012 they accounted for 41% of the living female HIV cases in San Francisco. Figure 12.2 Living female HIV cases diagnosed through December 2012 and female population by race/ethnicity, San Francisco Living Female HIV Cases Female Population of San Francisco 1 Latina 19% Asian/ Pacific Islander 6% Native American 1% Other/ Unkown 3% Latina 14% Asian/ Pacific Islander 36% Native American <1% White 29% African American 6% Other 4% African American 41% White 40% 1 United States 2010 Census data. The current exposure category definition for heterosexual contact does not adequately describe transmission for a large number of women who were likely infected heterosexually. In 2010 the CDC HIV Incidence Case Surveillance Branch developed a definition for female presumed heterosexual contact to reclassify the exposure category for female cases who would otherwise be reported with no identified risk (see Technical Notes Female Presumed Heterosexual Contact ). Among all living female HIV cases diagnosed in San Francisco, half acquired HIV infection through IDU and 44% through heterosexual contact (Figure 12.3). Figure 12.3 Females living with HIV diagnosed through December 2012 by exposure category, San Francisco Transfusion/ Hemophilia 1% Presumed Other/ Heterosexual Unknown 7% 4% Heterosexual 37% IDU 51%

65 50 HIV among Children, Adolescents and Young Adults Persons living with HIV in San Francisco who are adolescents and young adults (current age years) make up 1% of all living HIV cases in the city. As of December 31, 2012 there were 8 adolescents (age years) and 150 young adults (age years) living with HIV. Among living young adult HIV cases, the majority were MSM (Table 13.1). Over one-third of living young adult cases were Latino, while whites and African Americans made up similar proportions (26% and 27%, respectively). Adolescent data are not shown due to small numbers. Table 13.1 Living young adult HIV cases by exposure category, gender, and race/ethnicity, December 2012, San Francisco Years Old Number (%) Total 150 ( 100) Exposure Category MSM 100 ( 67) IDU 2 ( 1) MSM IDU 17 ( 11) Heterosexual 6 ( 4) Perinatal 16 ( 11) Other/Unidentified 9 ( 6) Gender Male 122 ( 81) Female 20 ( 13) Transfemale 1 8 ( 5) Race/Ethnicity White 39 ( 26) African American 40 ( 27) Latino 51 ( 34) Asian/Pacific Islander 14 ( 9) Other/Unknown 6 ( 4) 1 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status. Table 13.2 compares cases diagnosed with HIV infection among San Francisco adolescents and young adults with those in the same age groups diagnosed nationally. Compared to the U.S. as a whole, San Francisco had lower proportions of adolescent and young adult HIV diagnoses. Table 13.2 Cases diagnosed with HIV infection among adolescents and young adults, , San Francisco and the United States Year of HIV Initial Diagnosis Number (%) Number (%) Number (%) Number (%) San Francisco HIV Cases Age years at HIV diagnosis 4 ( <1 ) 4 ( <1 ) 4 ( <1 ) 6 ( 1 ) Age years at HIV diagnosis 50 ( 11 ) 49 ( 11 ) 39 ( 10 ) 51 ( 12 ) U.S. HIV Cases 1 46,319 43,806 42,842 Age years at HIV diagnosis 2,195 ( 5 ) 2,114 ( 5 ) 1,997 ( 5 ) N/A Age years at HIV diagnosis 6,756 ( 15 ) 7,047 ( 16 ) 7,005 ( 16 ) N/A 1 U.S. data are based on reported case counts from the 50 states and 6 dependent areas with confidential name-based HIV reporting in CDC HIV Surveillance Report, 2011.

66 Annual Report HIV Epidemiology San Francisco As of December 31, 2012, a cumulative total of 60 cases were diagnosed with HIV among children in San Francisco (less than 13 years old and resided in San Francisco at time of diagnosis). The number of pediatric HIV cases peaked between 1986 and 1995, and has declined over the following years (Figure 13.1). No pediatric HIV cases have been diagnosed since Of the 60 pediatric HIV cases reported, 25 (42%) have died, 32 (53%) have survived beyond childhood, and three (5%) were living pediatric HIV cases at the end of Figure 13.1 Pediatric HIV cases by time period of HIV diagnosis, , San Francisco Number of Cases Year of HIV Diagnosis

67 52 HIV among the Aging Population With the advent of HAART, persons with HIV are living longer, and those aged 50 and older comprise a larger proportion of HIV cases over time. Those 50 years and over now represents more than 50% of the living HIV cases in San Francisco. Between 2008 and 2012, the number and proportion of persons living with HIV aged 50 and over increased from 5,881 (40%) to 8,063 (51%). The proportion of newly diagnosed persons aged 50 years and over had almost doubled from 2008 to 2009 but has been oscillating every year since then. In 2012, the proportion remained at 12% with 49 newly diagnosed cases (Figure 14.1). Figure 14.1 Number and percent of persons diagnosed with HIV infection 1 at age 50 years and older, , San Francisco 90 25% Number of Cases % 17% 61 14% 68 17% 49 12% 20% 15% 10% Percent 20 5% Year of HIV Diagnosis 50 and over at time of HIV diagnosis % 50 and over at time of HIV diagnosis 0% 1 Includes persons with HIV infection by year of their initial HIV diagnosis.

68 Annual Report HIV Epidemiology San Francisco Persons diagnosed with HIV in at the age of 50 years and older differed slightly in demographic distribution when compared to younger persons diagnosed during the same time. A greater proportion of persons diagnosed 50 years and older were women, African American, IDU, and heterosexuals in comparison to the younger diagnosed persons (Table 14.1). Table 14.1 Characteristics of persons diagnosed with HIV infection in by age at diagnosis, San Francisco Age 50 years Age < 50 years (N=303) (N=1,909) Number (%) Number (%) Gender Male 255 ( 84 ) 1,742 ( 91 ) Female 45 ( 15 ) 116 ( 6 ) Transgender 3 ( 1 ) 51 ( 3 ) Race/Ethnicity White 166 ( 55 ) 936 ( 49 ) African American 77 ( 26 ) 243 ( 13 ) Latino 45 ( 15 ) 453 ( 24 ) Other/Unknown 15 ( 4 ) 277 ( 14 ) Exposure Category MSM 169 ( 56 ) 1,374 ( 72 ) IDU 43 ( 14 ) 97 ( 5 ) MSM IDU 30 ( 10 ) 220 ( 12 ) Heterosexual 33 ( 11 ) 99 ( 5 ) Other/Unidentified 28 ( 9 ) 119 ( 6 ) The majority of persons aged 50 years and older living with HIV are male (93%), white (69%), and MSM (74%) (Table 14.2). The gender and exposure category characteristics of persons aged 50 years and older are similar to those under 50 years old. The 50 years and older population are more likely to be white whereas those aged under 50 have a higher proportion of Latinos. Table 14.2 Characteristics of living HIV cases by age group, December 2012, San Francisco Age 50 years Age < 50 years (N=8,063) (N=7,642) Number (%) Number (%) Gender Male 7,516 ( 93 ) 6,935 ( 91 ) Female 429 ( 5 ) 480 ( 6 ) Transgender 118 ( 1 ) 227 ( 3 ) Race/Ethnicity White 5,565 ( 69 ) 4,161 ( 56 ) African American 1,128 ( 14 ) 918 ( 12 ) Latino 978 ( 12 ) 1,721 ( 21 ) Asian/Pacific Islander 274 ( 3 ) 568 ( 7 ) Native American 31 (<1 ) 60 ( 1 ) Other/Unknown 87 ( 1 ) 214 ( 3 ) Exposure Category MSM 5,988 ( 74 ) 5,344 ( 70 ) IDU 667 ( 8 ) 485 ( 6 ) MSM IDU 1,007 ( 12 ) 1,177 ( 15 ) Heterosexual 206 ( 3 ) 291 ( 4 ) Other/Unidentified 195 ( 2 ) 345 ( 5 )

69 54 HIV among Transgender Persons Transgender status relies on review of information in medical records. Information on transgender status has been collected since During 2006 to 2012, there were a total of 89 transgender persons diagnosed with HIV in San Francisco (Table 15.1). All of the transgender persons diagnosed in this time period were transfemale. Transgender cases comprised approximately 3% of all HIV cases diagnosed in this time period. Compared to all HIV cases diagnosed in the same time period, transgender cases were more likely to be non-white, IDU, and younger. As of December 31, 2012, there were 350 transgender persons living with HIV in San Francisco (Table 15.2). African Americans and Latinos made up the majority of living transgender HIV cases, and 45% of living transgender cases were IDU. Two-thirds of living transgender cases were age 40 years or older at the end of Table 15.1 Characteristics of transgender 1 HIV cases compared to all HIV cases diagnosed in , San Francisco Transgender HIV Cases HIV Cases Diagnosed Diagnosed Number (%) Number (%) Total 89 3,262 Race/Ethnicity White 19 ( 21 ) 1,662 ( 51 ) African American 24 ( 27 ) 472 ( 14 ) Latino 31 ( 35 ) 714 ( 22 ) Other/Unknown 15 ( 17 ) 414 ( 13 ) Injection Drug Use Yes 26 ( 29 ) 617 ( 19 ) No 63 ( 71 ) 2,645 ( 81 ) Age at Diagnosis (Years) ( 0 ) 11 (<1 ) ( 22 ) 367 ( 11 ) ( 21 ) 482 ( 15 ) ( 31 ) 1,062 ( 33 ) ( 17 ) 906 ( 28 ) ( 8 ) 434 ( 13 ) 1 See Technical Notes Transgender Status. Table 15.2 Characteristics of transgender 1 persons living with HIV, December 2012, San Francisco Number (%) Race/Ethnicity White 75 ( 21 ) African American 121 ( 35 ) Latino 109 ( 31 ) Asian/Pacific Islander 34 ( 10 ) Other/Unknown 11 ( 4 ) Injection Drug Use Yes 159 ( 45 ) No 191 ( 55 ) Age in Years (at end of 2012) ( 0 ) ( 2 ) ( 7 ) ( 23 ) ( 33 ) ( 34 ) Total Number 350 ( 100 ) 1 See Technical Notes Transgender Status.

70 Annual Report HIV Epidemiology San Francisco HIV among Homeless Persons A case is classified as homeless if, at the time of HIV or HIV infection stage 3 (AIDS) diagnosis, the medical record states that the patient is homeless or the patient s address is one of the following: (1) a known homeless shelter, (2) a health care clinic, or (3) a free postal address not connected to a residence ( general delivery ). Cases with missing information on residence are not classified as homeless. Among all cases diagnosed with HIV infection, the number of homeless cases fluctuated each year between 2006 and 2012 (Figure 16.1). After reaching a peak of 14% of all cases diagnosed in 2010, the proportion of homeless persons diagnosed each year has decreased to between 8% and 9% in recent years. Figure 16.1 Number and percent of homeless cases diagnosed with HIV infection 1 by year of diagnosis, , San Francisco 90 25% 80 Number of Homeless Cases % % 10% 54 12% 59 14% 32 8% 37 9% 20% 15% 10% 5% Percent Year of HIV Diagnosis 0% Homeless Cases Homeless % 1 Includes persons with HIV infection by year of their initial HIV diagnosis.

71 56 HIV among Homeless Persons Compared to all HIV cases diagnosed in 2006 to 2012, persons who were homeless at their HIV diagnosis were more likely to be women, transfemale, African American, and IDU (Table 16.1). Table 16.1 Characteristics of homeless HIV cases compared to all HIV cases diagnosed in , San Francisco Homeless HIV Cases HIV Cases (N=335) (N=3,262) Gender Male 77% 90% Female 16% 7% Transfemale 1 7% 3% Race/Ethnicity White 44% 51% African American 29% 14% Latino 18% 22% Other/Unknown 9% 13% Exposure Category MSM 33% 71% IDU 24% 6% MSM IDU 29% 13% Heterosexual 9% 6% Other/Unidentified 4% 4% Age at Diagnosis (years) % <1% % 11% % 15% % 33% % 28% % 13% 1 Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical Notes Transgender Status.

72 Annual Report HIV Epidemiology San Francisco Sexually Transmitted Diseases among Persons with HIV The occurrence of STD diagnosis among persons living with HIV is a marker for unprotected sex, which depending upon HIV treatment status and partner HIV serostatus may gauge an increased potential for HIV transmission. Diagnosis of STD among persons with HIV was determined through a computerized match of the HIV and STD case registries. Data from STD registry included persons reported with gonorrhea, chlamydia, non-gonococcal urethritis, or infectious syphilis. The number of STD cases among persons living with HIV continues to rise from 774 cases in 2007 to 992 cases in 2011 (Figure 17.1). The increase coincided with the trend shown in early syphilis reported from 2007 through 2011, especially among MSM diagnosed with HIV (Figure 9.4). In 2011, rectal gonorrhea and male gonococcal procititis increased among HIV-positive MSM (Figure 9.3). All STD occurred after the HIV diagnosis, indicating unprotected sex among persons with known HIV infection. Figure 17.1 Number of HIV cases diagnosed with an STD by year of STD diagnosis, , San Francisco No. of Cases w/ STD Year of STD Diagnosis MSM Non-MSM

73 58 Sexually Transmitted Diseased among Persons with HIV Table 17.1 shows the demographic characteristics of HIV cases diagnosed with an STD from 2007 through The majority of cases were male, white, and aged years at the time of STD diagnosis. Gender and race/ethnicity distributions are similar across five years. The proportion of HIV cases diagnosed with an STD in the years age group decreased from 2007 to In 2011, the proportion of HIV cases diagnosed with STD in years increased slightly from previous years. Table 17.1 HIV cases diagnosed with an STD by demographic characteristics, , San Francisco Year of STD diagnosis Number (%) Number (%) Number (%) Number (%) Number (%) Gender Male 755 ( 98) 904 ( 97) 799 ( 97) 907 ( 97) 960 ( 97) Female 8 ( 1) 12 ( 1) 8 ( 1) 8 ( 1) 11 ( 1) Transfemale 1 11 ( 1) 12 ( 1) 16 ( 2) 16 ( 2) 21 ( 2) Race/Ethnicity White 500 ( 65) 549 ( 59) 487 ( 59) 590 ( 63) 598 ( 60) African American 69 ( 9) 91 ( 10) 80 ( 10) 76 ( 8) 89 ( 9) Latino 143 ( 18) 216 ( 23) 179 ( 22) 199 ( 21) 215 ( 22) Asian/Pacific Islander 43 ( 6) 51 ( 6) 53 ( 6) 48 ( 5) 60 ( 6) Other/Unknown 19 ( 2) 21 ( 2) 24 ( 3) 18 ( 2) 30 ( 3) Age at STD diagnosis ( 10) 106 ( 11) 97 ( 12) 101 ( 11) 110 ( 11) ( 35) 286 ( 31) 267 ( 32) 252 ( 27) 245 ( 25) ( 39) 400 ( 43) 321 ( 39) 415 ( 45) 415 ( 42) ( 14) 108 ( 12) 113 ( 14) 128 ( 14) 172 ( 17) ( 2) 28 ( 3) 25 ( 3) 35 ( 4) 50 ( 5) Total Transfemale data include all transgender cases. Transmale data are not released separately due to potential small population size. See Technical notes Transgender Status.

74 Annual Report HIV Epidemiology San Francisco Geographic Distribution of HIV Map 18.1 illustrates the geographic distribution of the number of living HIV cases in San Francisco by census tract and neighborhood as of December 31, The most current residence collected through prospective chart review and laboratory reporting were used to map the living cases. Census tract level data provide an enhanced understanding of varying number of cases within a given neighborhood. The seven census tracts in the Castro all ranked in the ten most populous census tracts among HIV cases in the City, ranging from 214 to 406 cases. Map 18.1 Geographic distribution of persons living with HIV, December 2012, San Francisco * Living homeless HIV cases (N=560) are not displayed on this map.

75 60 Geographic Distribution of HIV Map 18.2 shows the rates of newly diagnosed cases in San Francisco from by neighborhood. The highest rate of new diagnoses occurred in the Castro (462 per 100,000) followed by South of Market (252 per 100,000) and Tenderloin (227 per 100,000). New diagnoses were relatively low in the outer neighborhoods while the central neighborhoods exhibited higher rates. Map 18.2 Geographic distribution of HIV rates for newly diagnosed in per 100,000 population, San Francisco