Changes in Causes of Death Among Persons with AIDS: San Francisco,California,1996

Abstract

The increased life expectancy among HIV-infected persons treated with combination antiretroviral therapy (ART), risk behaviors, and co-morbidities associated with ART place HIV-infected persons at risk for non-HIV-related causes of death. We used the San Francisco HIV/AIDS registry to identify deaths that occurred from January 1996 through December 2011. Temporal trends in AIDS- and non-AIDS-related mortality rates, the proportion of underlying and contributory causes of death, and the ratio of observed deaths in the study population to expected number of deaths among California men aged 20–79 (standardized mortality ratio [SMR]) of underlying causes of death were examined. A total of 5338 deaths were identified. The annual AIDS-related death rate (per 100 deaths) declined from 10.8 in 1996 to 0.9 in 2011 (p<0.0001), while the annual death rate from non-AIDS-related causes declined from 2.1 in 1996 to 0.9 in 2011 (p<0.0001). The proportion of deaths due to all types of heart disease combined, all non-AIDS cancers combined, mental disorders resulting from substance abuse, drug overdose, suicide and chronic obstructive pulmonary disease increased significantly over time. The SMRs for liver diseased decreased significantly over time but remained elevated. Our data highlight the importance of age-related causes of death as well as deaths from causes that are, at least in part, preventable.

Introduction

Highly active antiretroviral therapy (HAART) has led to a dramatic decline in deaths among HIV-infected persons,^1,2 a shift towards an increase in the proportion of deaths due to non-HIV causes,^1

–8 and increases in life expectancy.^9

–12 However, the increase in life expectancy in combination with co-infections and co-morbidities common among HIV-infected adults (hepatitis B and C, heart disease, non-AIDS cancers, and metabolic syndrome), and life style factors (alcohol, tobacco, and drug use) place them at higher risk for age-related and other non-HIV-related causes of death.^{13

–23} Side effects associated with some antiretroviral agents may also contribute to premature mortality from non-HIV causes such as cardiovascular, renal, and liver disease.^{24

–31} Deaths from these causes may occur more frequently than AIDS-related causes with prolonged exposure to these medications. Moreover, while AIDS-related mortality is largely attributable to immune deficiency, even in the context of viral suppression and immune-restoration, chronic inflammation and immune injury are likely to contribute to mortality among persons with HIV infection.^32

–35

As one of the earliest HIV epidemic centers of the United States, San Francisco has been disproportionately affected by HIV. As of December 31, 2010, the San Francisco, California metropolitan statistical area (MSA) had the seventh highest HIV prevalence of all MSAs in the United States.³⁶ Although the number of persons newly diagnosed with HIV in San Francisco has declined in recent years, the emphasis in San Francisco on access to antiretroviral therapy, including early initiation of HAART, has resulted in a total of 15,705 persons living with HIV in San Francisco as of the end of 2012, of whom 51% were aged 50 years or older.³⁷ The size and age of the population living with HIV and the completeness of AIDS surveillance data in San Francisco provided an opportunity to examine trends in causes of death and to compare these causes among older and younger persons.

Methods

Persons with HIV/AIDS who reside in San Francisco are legally required to be reported to the San Francisco Department of Public Health. Over 80% of cases in San Francisco were reported through active surveillance whereby health department staff members review the medical records of HIV/AIDS patients and complete the case report form. Information collected included sociodemographic and risk characteristics, AIDS-related clinical data, and vital status. The completeness of HIV/AIDS case reporting is evaluated annually and has consistently found to be above 90%.^37,38 Death information was ascertained through monthly review of San Francisco death certificates, reports from the California State HIV/AIDS registry, and annual matches with the Social Security Death Master File and the National Death Index. The most recent match with the Social Security Death Master File included deaths through 2012 and the most recent match with the National Death Index included deaths through 2011.

All AIDS cases that occurred among people aged >12 years and who died from January 1, 1996, through December 31, 2011, were included in this analysis. Individuals reported with HIV who did not have AIDS were excluded because name-based HIV reporting was not in effect until 2006 and because of the small number of deaths in HIV-infected persons without AIDS (207 deaths from 2006 through 2011). The underlying and contributory causes of death were classified using the International Classification of Diseases 9^th edition for deaths through 1998 and 10^th edition for deaths that occurred during or after 1999.^39,40 In our analysis we included frequently occurring non-AIDS-related causes, as well as causes with known or suspected relationship to antiretroviral therapy, chronic HIV infection, or life style risk factors associated with HIV.

We compared the temporal trends in the AIDS-related and non-AIDS-related annual mortality rates in which the numerator was the single underlying cause of death and the denominator was the number of persons living with AIDS plus the number of persons who died each year. For subsequent analyses, we divided the years into four time periods based upon the distribution of deaths: 1996–1997, 1998–2001, 2002–2005, 2006–2011. Then we compared the demographic characteristics of decedents over each of these time periods. Next, to assess trends in multiple causes of death, we measured changes in the proportion of selected underlying and contributory causes of death in the four time periods. For this analysis the numerator was the number of deaths due to the selected cause (as either the underlying or a contributory cause) and the denominator was the total number of deaths for that time period. We also repeated this analysis using the single underlying cause only. For this analysis the numerator was the number of deaths due to the underlying cause and the denominator was all deaths in the time period. We calculated the standardized mortality ratio (SMR) and 95% confidence intervals of selected non-AIDS-related underlying causes of death in each of the four time periods. The SMR, the ratio of observed to expected deaths, was calculated by dividing the observed number of deaths by the expected number of deaths. Because there were so few AIDS deaths among women, we only calculated age and race adjusted SMRs among men using the California population of men aged 20–79 years as our standard population.⁴¹ Lastly, we plotted the four most frequent non-AIDS underlying causes of death that occurred in any of the four time periods in three age groups (under 40 years, 40–49 years, and 50 years or more) to visualize these trends. Because the most frequently occurring deaths differed in the time periods, the number of causes examined in each age group differed between age groups. The chi square test was used to measure differences in the demographic characteristics of decedents using SAS version 9 (SAS Institute, Cary, NC). The Mantel-Haenszel chi square test for linear trend was used to measure changes in the proportions of causes of deaths across the four time periods (SAS version 9; SAS Institute, Cary, NC), in the annual death rates, and in the SMRs over time.^42,43

Results

There were a total of 5338 deaths reported in AIDS patients from 1996 through 2011. AIDS was the underlying cause of death in 3617 (68%) of the 5338 deaths. The annual death rate (per 100 deaths) from AIDS declined from 10.8 in 1996 to 4.0 in 1997, and to 3.0 in 1998, from which point the death rate slowly declined until 2010 when the death rate reached 0.9 where it remained in 2011 (p<0.0001 for linear trend, Fig. 1). The death rate from non-AIDS-related causes declined from 2.1 per 100 deaths in 1996 to 0.9 in 2011 (p<0.0001 for linear trend). In 2011, and for the first time, the number of AIDS-related deaths equaled the number of non-AIDS-related deaths.

FIG. 1.

AIDS-related and non-AIDS-related death rates among San Francisco AIDS cases, 1996–2011.

The sociodemographic and risk characteristics of decedents changed significantly over the observation period. The proportion of deaths among men, Whites, Latinos, persons less than 50 years old at death, men who have sex with men, and persons with private insurance decreased over time (Table 1).

Table 1.

Demographic and RIsk Characteristics of San Francisco AIDS Deaths by Calendar Period, 1996–2011 (N=5338)

Characteristic	1996–1997 N=1395 N (%)	1998–2001 N=1394 N (%)	2002–2005 N=1209 N (%)	2006–2011 N=1340 N (%)	p Value ^a
Gender					<0.0003
Male	1325 (95.0)	1289 (92.5)	1103 (91.2)	1222 (91.2)
Female	70 (5.0)	105 (7.5)	106 (8.8)	118 (8.8)
Race/ethnicity					<0.0001
White	951 (68.2)	909 (65.2)	720 (59.6)	811 (60.5)
AfricanAmerican	211 (15.1)	292 (21.0)	270 (22.3)	276 (20.6)
Latino	182 (13.1)	138 (9.9)	151 (12.5)	157 (11.7)
Other	51 (3.7)	55 (4.0)	68 (5.6)	96 (7.2)
Age group (years)					<0.0001
13–39	568 (40.7)	362 (26.0)	221 (18.3)	127 (9.5)
40–49	567 (40.7)	613 (44.0)	514 (42.5)	448 (33.4)
50–59	195 (14.0)	318 (22.8)	324 (26.8)	481 (35.9)
60 and older	65 (4.7)	101 (7.3)	150 (12.4)	284 (21.2)
Risk group					<0.0001
Male–male sex	973 (69.8)	818 (58.7)	645 (53.4)	721 (53.8)
Heterosexual injection drug use	162 (11.6)	236 (16.9)	236 (19.5)	200 (14.9)
Male–male sex and injection drug use	223 (16.0)	298 (21.4)	277 (22.9)	360 (26.9)
Other	37 (2.7)	42 (3.0)	51 (4.2)	59 (4.4)
Insurance at time of diagnosis^b					<0.0001
Public	236 (16.9)	306 (22.0)	348 (28.8)	421 (31.4)
Private	571 (40.9)	481 (34.5)	369 (30.5)	449 (33.5)
None	375 (26.9)	462 (33.4)	384 (31.8)	373 (27.8)
Country of birth^c					0.4077
United States	1280 (91.8)	1300 (93.3)	1107 (91.6)	1203 (89.8)
Outside of USA	115 (8.2)	94 (6.7)	102 (8.4)	137 (10.2)

Chi square test; ^bmissing 563 records with unknown insurance status; ^cmissing 37 records with unknown country of origin.

The proportion of underlying and contributory AIDS-related causes of death was significantly different over the study period decreasing from over 90% in the period 1996 to 1997 to 72% in the period 2006–2011 (Table 2). The proportion of deaths due to all types of heart disease, coronary heart disease, all non-AIDS cancers, lung, liver, and anal cancers, mental disorders resulting from substance abuse, drug overdose, suicide, and chronic obstructive pulmonary disease were also significantly different, and increased over the four time periods. Deaths due to liver disease and viral hepatitis increased between each of the first three time periods and then declined in the most recent time period. Deaths from renal disease increased steadily over the first three time periods and remained relatively stable through 2006–2011. Among the non-AIDS causes of death, heart disease (n=1040), pneumonia (n=755), liver disease (n=665), non-AIDS cancers (n=595), and viral hepatitis (n=575) were the most frequent causes. When we restricted our analysis using underlying causes, only the trends in deaths due to anal cancer, renal disease, and septicemia were no longer statistically significant (p=0.1265, 0.2423, and 0.8203, respectively).

Table 2.

Underlying and Contributory Causes of Death Among Persons with AIDS, San Francisco, by Calendar Period, 1996–2011

Year
Cause of death ^a	1996–1997 N (%)^c	1998–2001 N (%)	2002–2005 N (%)	2006–2011 N (%)	p Value^b
AIDS-related deaths	1289 (92.4)	1188 (85.2)	1019 (84.3)	969 (72.3)	<0.0001
Non-AIDS-related deaths
Heart disease (all)	198 (14.2)	280 (20.1)	243 (20.1)	319 (23.8)	<0.0001
Coronary heart disease	16 (1.2)	62 (4.5)	67 (5.5)	103 (7.7)	<0.0001
Cardiomyopathy	20 (1.4)	34 (2.4)	24 (2.0)	25 (1.9)	0.6044
Non-AIDS-cancers (all)	97 (7.0)	129 (9.3)	143 (11.8)	226 (16.9)	<0.0001
Lung cancer	12 (0.9)	30 (2.2)	28 (2.3)	55 (4.1)	<0.0001
Liver cancer	7 (0.5)	24 (1.7)	21 (1.7)	34 (2.5)	<0.0001
Anal cancer	4 (0.3)	7 (0.5)	12 (1.0)	13 (1.0)	0.0109
Hodgkin's lymphoma	6 (0.4)	3 (0.2)	5 (0.4)	9 (0.7)	0.2469
Liver disease	94 (6.7)	197 (14.1)	194 (16.1)	180 (13.4)	<0.0001
Pneumonia	205 (14.7)	203 (14.6)	172 (14.2)	175 (13.1)	0.2135
Viral hepatitis	50 (3.6)	157 (11.3)	194 (16.1)	174 (13.0)	<0.0001
Renal disease	68 (4.9)	95 (6.8)	138 (11.4)	145 (10.8)	<0.0001
Septicemia	92 (6.6)	132 (9.5)	141 (11.7)	137 (10.2)	0.0002
Mental disorders due to substance abuse	34 (2.4)	62 (4.5)	88 (7.3)	126 (9.4)	<0.0001
Drug overdose	31 (2.2)	53 (3.8)	37 (3.1)	113 (8.4)	<0.0001
Suicide	13 (0.9)	25 (1.8)	24 (2.0)	45 (3.4)	<0.0001
Chronic obstructive pulmonary disease	22 (1.6)	44 (3.2)	59 (4.9)	87 (6.5)	<0.0001
Cerebrovascular disease	25 (1.8)	40 (2.9)	36 (3.0)	37 (2.8)	0.1168
Total number of deaths	1395	1394	1209	1340

Individuals may have more than one cause of death; ^bMantel-Haenszel chi-square test; ^cpercent of total deaths within each time period.

The SMRs were elevated for heart disease beginning in the 1998–2001 time period and remained so through the most recent time period, while the SMRs for all non-AIDS cancers, lung, liver, and anal cancers were elevated in all four time periods, with the highest SMR observed for anal cancer (Table 3). Although the number of observed deaths from liver disease was significantly higher than the number of expected deaths in all time periods, the SMRs declined from 5.29 (95% CI 3.02–8.58) in the first time period to 1.53 (95% CI 1.00–2.24; p=0.01) in the last time period.

Table 3.

Standardized Mortality Ratios (SMRs) for Underlying Non-AIDS-Related Causes of Death Among San Francisco Men with AIDS by Calendar Period, 1996–2011

Underlying cause of death	1996–1997	1998–2001	2002–2005	2006–2011	p Value ^a
Heart disease (all)
Observed	11	66	55	100
Expected^b	13.6	32.7	42.7	78.2
SMR (95% CI)	0.81 (0.40–1.45)	2.02 (1.56–2.57)	1.29 (0.97–1.68)	1.28 (1.04–1.56)	0.18
Cardiomyopathy
Observed	1	10	5	6
Expected	1.6	3.6	3.8	6.5
SMR (95% CI)	0.63 (0.02–3.52)	2.80 (1.34–5.14)	1.31 (0.42–3.05)	0.93 (0.34–2.02)	0.25
Coronary heart disease
Observed	6	34	35	51
Expected	7.5	22.4	30.3	55.2
SMR (95% CI)	0.80 (0.29–1.73)	1.52 (1.05–2.12)	1.16 (0.80–1.61)	0.92 (0.69–1.21)	0.41
Non-AIDS-cancers (all)
Observed	40	83	90	151
Expected	11.8	29.6	39.2	79.0
SMR (95% CI)	3.40 (2.43–4.62)	2.80 (2.23–3.47)	2.30 (1.85–2.82)	1.91 (1.62–2.24)	0.90
Lung cancer
Observed	9	26	24	47
Expected	3.5	8.5	10.6	20.0
SMR (95% CI)	2.57 (1.18–4.88)	3.06 (2.00–4.49)	2.27 (1.46–3.38)	2.36 (1.73–3.13)	0.52
Liver cancer
Observed	6	18	17	24
Expected	0.54	1.6	2.6	6.3
SMR (95% CI)	11.15 (4.09–24.27)	11.02 (6.53–17.41)	6.66 (3.88–10.66)	3.82 (2.45–5.69)	0.65
Anal cancer
Observed	4	4	7	8
Expected	0.02	0.08	0.08	0.20
SMR (95% CI)	179.63 (48.94–459.93)	51.57 (14.05–132.04)	86.64 (34.83–178.51)	40.81 (17.62–80.41)	0.47
Liver disease
Observed	16	25	26	26
Expected	3.0	7.1	9.1	17.0
SMR (95% CI)	5.29 (3.02–8.58)	3.50 (2.27–5.17)	2.85 (1.86–4.17)	1.53 (1.00–2.24)	0.01
Renal disease
Observed	5	3	1	8
Expected	0.20	0.78	1.4	3.1
SMR(95% CI)	25.46 (8.27–59.42)	3.87 (0.80–11.31)	0.71 (0.02–3.95)	2.61 (1.12–5.13)	0.26

Observed deaths are non-AIDS-related deaths that occurred among men with AIDS between ages 20–79 years during the years 1996–2011; expected deaths were derived using cause-specific death rates among California men, aged 20–79 years and adjusted for race and age.

Mantel-Haenszel chi-square test for linear trend; ^bthe number of expected deaths was rounded to one decimal place except for when the expected number of deaths was less than one in which case the number was rounded to two decimal places.

The most frequently occurring underlying non-AIDS causes of death differed by age group and calendar period (Fig. 2). The proportion of deaths due to all non-AIDS causes combined increased with age. Among those who died before age 40, deaths from drug overdose and suicide accounted for higher proportions of deaths than most other non-AIDS causes, and both of these causes showed sharp proportional increases in the most recent time period. Among persons who died between ages 40 and 49 years, deaths from drug overdose were also proportionately higher in the most recent time period. In contrast to younger decedents, the proportion of deaths due to non-AIDS cancers and heart disease increased in older age groups. For persons who died at age 50 years and above, deaths from non-AIDS cancers and heart disease were proportionately greater than other causes and increased over time. As was observed with the other age groups, deaths due to overdose were relatively frequent and showed a sharp increase between the calendar periods 2002–2005 and 2006–2011.

FIG. 2.

The proportion of deaths due to non-AIDS causes among San Francisco AIDS cases by age group and calendar period, 1996–2011.

Discussion

We observed marked declines in the absolute number of deaths among persons with AIDS and in the rates of AIDS-related and non-AIDS-related deaths, with greater declines observed in AIDS-related deaths than non-AIDS-related deaths. The decline in deaths due to AIDS-related causes likely reflects better clinical management of HIV infection, including the widespread use of antiretroviral therapy. This has led to HIV-infected adults reaching ages at which non-AIDS causes of death, such as cancer and heart disease, are increasingly common. Our study, as well as others, demonstrated an increase in the proportion of deaths due to non-HIV causes.^1

–8

The elevated risk of death from non-AIDS cancers, heart, and liver disease across nearly all time periods among persons with AIDS compared to the general male population, merits careful examination to identify areas to target for prevention and early detection. Our finding of the increase in deaths from heart disease beginning in 1998 may be due, at least in part, to the use of protease inhibitors that have been shown to increase the risk of cardiac disease.^28,30,31 The elevated risk of death from lung, anal, and liver cancers, coronary heart disease, and liver disease are, at least to some degree, and either directly or indirectly, due to behaviors such as use of tobacco and alcohol, anal sex, and injection drug use that occur in higher frequency among HIV-infected adults than the general adult population,^19

–23 as well as from the lack of specific guidance for screening HIV-infected persons for non-AIDS cancers⁴⁴ and inadequate adherence to hepatitis screening and treatment guidelines.^45,46 Re-evaluation of the risks and benefits of cancer screening for HIV-infected persons, particularly those with substantial tobacco exposure and anal sex, as well as renewed focus on adherence to screening and treatment guidelines and on primary prevention through programs to prevent or eliminate tobacco and other substance use are warranted.

Non-AIDS causes of death accounted for a higher proportion of deaths in older adults over the duration of the study, with cancer and heart disease accounting for larger proportions of non-AIDS causes. The excess risk of these diseases, after adjusting for age and race, supports the hypothesis that HIV infection and its sequelae contribute to the development of non-AIDS-related diseases.

Among adult AIDS cases under 40 years old, deaths from overdose accounted for a substantial proportion of non-AIDS deaths. In addition, we observed a sharp increase in this proportion between the calendar periods 2002–2005 and the most recent period. Though less dramatic, increases in deaths due to overdose were observed in the other age groups during the same periods. This mortality pattern among adults under 40 years old is also reflected in the general population of San Francisco, and may reflect increases in overdose from prescription opioids (P. Coffin, M.D., personal communication, January 7, 2013).^50–54

Limitations to this study must be considered. The mortality data come from death certificates; causes of death may be under-reported or misclassified, particularly those associated with social stigma. Additionally, contributory causes of death may have been incomplete if the person completing the death certificate was not familiar with the patient's full medical history. Our data also come from a single city where men who have sex with men account for the vast majority of cases. As such, our ability to characterize deaths among injection drug users, women, and heterosexual men is greatly limited. We examined deaths among persons with a prior AIDS diagnosis; while there may be important information obtained from studying deaths among HIV-infected persons prior to the development of AIDS, the number of deaths in San Francisco among HIV-infected persons who did not have AIDS was too small for meaningful analysis. In our analysis of multiple causes of death, the number of comparisons may have resulted in over estimating significance. To assess this, we conducted the analysis using the Bonferroni correction; only trends in anal cancer did not remain statistically significant.

Our findings contribute to the growing number of reports demonstrating decreases in deaths among adults with AIDS and a shift to increasing proportions of deaths due to non-AIDS causes and expand the literature by describing these non-AIDS causes and their frequency in younger and older persons. Although aging of the HIV-infected population contributes to the development of and death from non-AIDS causes, our data highlight the importance of deaths from causes that are, at least in part, preventable. As the prevalence and age of the HIV-infected population continues to increase in the United States and elsewhere, more can, and should be done to prevent premature death in this ever growing population.

Footnotes

Acknowledgments

The authors wish to thank Dr. George Rutherford for his critical review of the article and the San Francisco Department of Public Health HIV Epidemiology Section staff for their collection of HIV and AIDS surveillance data.

Financial support for this analysis was provided through the Centers of Disease Control and Prevention cooperative agreement number 5U62PS001000-02.

Author Disclosure Statement

No competing financial interests exist.

References

Palella

Jr , Baker

, Moorman

, et al. Mortality in the highly active antiretroviral therapy era: Changing causes of death and disease in the HIV Outpatient Study. J Acquir Immune Defic Syndr, 2006; 43:27–34.

Sackoff

, Hanna

, Pfeiffer

, Torian

. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City. Ann Intern Med, 2006; 145:397–406.

Lewden

, May

, Rosenthal

, et al. Changes in causes of death among adults infected by HIV between 2000 and 2005: The “Mortalite 2000 and 2005” Surveys (ANRS EN19 and Mortavic). J Acquir Immune Defic Syndr, 2008; 48:590–598.

Fazito

ELL

, Vasconcelos

NAM

, Pereira

. Causes of death among people living with HIV/AIDS in Brazil. Braz J Infect Dis, 2010; 14:558–553.

Crum

, Riffenburgh

, Wegner

, et al. Comparisons of causes of death and mortality rates among HIV-infected persons: Analysis of the pre-, early, and late HAART (highly active antiretroviral therapy) eras. J Acquir Immune Defic Syndr, 2006; 41:194–200.

French

, Gawel

, Hershow

, et al. Trends in mortality and causes of death among women with HIV in the United States: A 10-year study. J Acquir Immune Defic Syndr, 2009; 51:399–406.

Lau

, Gange

, Moore

. Risk of on-ART related mortality may exceed risk of AIDS-related mortality among individuals enrolling into care with CD4+ counts greater than 200 cells/mm³ . J Acquir Immune Defic Syndr, 2007; 44:179–187.

Simard

, Pfeiffer

, Engels

. Mortality due to cancer among people with AIDS: A novel approach using registry-linkage data and population attributable risk methods. AIDS, 2012; 26:1311–1318.

The Antiretroviral Therapy Cohort Collaborative. Life expectancy of individual on combination antiretroviral therapy in high income countries: A collaborative analysis of 14 cohort studies. Lancet, 2008; 372:293–299.

10.

May

, Gompels

, Delpech , et al. Impact of late diagnosis and treatment on life expectancy in people with HIV-1: UK Collaborative HIV Cohort (UK CHIC) Study. BMJ, 2011; 343:d6016.

11.

Wada

, Jacobson

, Cohen

, French

, Phair

, Munoz

. Cause-specific mortality among HIV-infected individuals, by CD4+ cell count at HAART initiation, compared with HIV-uninfected individuals. AIDS, 2014; 28:257–265.

12.

Rodger

, Lodwick

, Schechter

, et al. Mortality in well controlled HIV in the continuous antiretroviral therapy arms of the SMART and ESPRIT trials compared with the general population. AIDS, 2013; 27:973–979.

13.

The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D.) Study Group. Factors associated with specific causes of death amongst HIV-positive individuals in the D.A.D. Study. AIDS, 2010; 24:153148–153157.

14.

The D.A.D Study Group. Liver-related deaths in persons infected with the human immunodeficiency virus. Arch Intern Med, 2006; 166:1632–1641.

15.

Jarrett

, Wanke

, Ruthazer

, Bica

, Isaac

, Knox

. Metabolic syndrome predicts all-cause mortality in persons with human immunodeficiency virus. AIDS Patient Care and STDs, 2013; 27:266–271.

16.

Jain

, Serenba

, Bhore

, et al. Change in fibrosis score as a predictor of mortality among HIV-infected patients with viral hepatitis. AIDS Patient Care STDs, 2012; 26:73–80.

17.

The Antiretroviral Therapy Cohort Collaborative. Mortality of HIV-infected patients starting potent antiretroviral therapy: Comparison with the general population in nine industrialized countries. Int J Epidemiol, 2009; 38:1624–1633.

18.

Phillips

, Neaton

, Lundgren

. The role of HIV in serious diseases other than AIDS. AIDS, 2008; 22:2409–2418.

19.

Page-Shafer

, Delorenze

, Satariano

, Winklestein

. Comorbidity and survival in HIV-infected men in the San Francisco Men's Health Survey. Ann Epidemiol, 1996; 6:420–430.

20.

Villanti

, German

, Sifkis

, Flynn

, Holtgrave

. smoking, HIV status, and HIV risk behaviors in a respondent-driven sample of injection drug users in Baltimore, Maryland: The BESURE study. AIDS Educ Prev, 2012; 24:132–147.

21.

Galvan

, Bing

, Fleishman

, et al. The prevalence of alcohol consumption and heavy drinking among people with HIV in the United States: results from the HIV Cost and Services Utilization Study. J Stud Alcohol, 2002; 63:179–186.

22.

Lifson

, Lando

. Smoking and HIV: Prevalence, health risks, and cessation strategies. Curr HIV/AIDS Rep, 2012; 9:223–230.

23.

Scott

, Vittinghoff

, Irvin

, et al. Age, race ethnicity and behavioral risk factors associated with per contact risk of HIV infection among men who have sex with men in the United States. J Acquir Immune Defic Syndr, 2014; 65:115–121.

24.

The SMART/INSIGHT and the D.A.D. Study Groups. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients. AIDS, 2008; 22:F17–F24.

25.

Crane

, Kestenbaum

, Harrington

, Kitahata

. Amprenavir and didanosine are associated with declining kidney function among patients receiving tenofovir. AIDS, 2007; 21:1431–1439.

26.

Goicoechea

, Liu

, Best

, et al. Greater tenofovir-associated renal function decline with protease inhibitor-based versus nonnucleoside reverse-transcriptase inhibitor-based therapy. J Infect Dis, 2008; 197:102–108.

27.

Islam

, Wu

Jianyun

, Jasson

, Wilson

. Relative risk of renal disease among people living with HIV: A systematic review and meta-analysis. BMC Public Health, 2012; 12:324 http://www.biomedcentral.com/1471-2458/12/234 (Last accessed April 4, 2014 ).

28.

Islam

, Wu

, Jansson

, Wilson

. Relative risk of cardiovascular disease among people living with HIV: A systematic review and meta-analysis. HIV Med, 2012; 13:453–468.

29.

Nunez

. Clinical syndromes and consequences of antiretroviral-related hepatotoxicity. Hepatology, 2010; 52:1143–1155.

30.

DAD Study Group. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med, 2007; 356:1723–1735.

31.

Lang

, Mary-Krause

, Cotte

, et al. Impact of individual antiretroviral drugs on the risk of myocardial infarction in human immunodeficiency virus-infected patients: A case-control study nested within the French Hospital Database ANRS Cohort CO4. Arch Intern Med, 2010; 170:1228–1238.

32.

Armah

, McGinnis

, Baker

, et al. HIV status, burden of comorbid disease and biomarkers of inflammation, altered coagulation and monocyte activation. Clin Infect Dis, 2012; 55:26–136.

33.

Neuhaus

, Jacobs

, Baker

, et al. Markers of inflammation, coagulation, and renal function are elevated in adults with HIV infection. J Infect Dis, 2010; 201:1788–1795.

34.

Effros

, Fletcher

, Gebo

, et al. Workshop on HIV infection and aging: What is known and future research directions. Clin Infect Dis, 2008; 47:542–553.

35.

Kuller

, Tracy

, Belloso

, et al. Inflammatory and coagulation biomarkers and mortality in patients with HIV infection. PloS Medicine, 2008; 5:e203.

36.

Centers for Disease Control and Prevention. HIV Surveillance Report, 2011; vol. 23. http://www.cdc.gov/hiv/topics/surveillance/resources/reports/. Published February 2013 (Last accessed March 12, 2014 ).

37.

San Francisco Department of Public Health. HIV/AIDS Epidemiology Annual Report, 2012. Available at: www.sfdph.org/comupg/oprograms/hivepisec/default/asp (Last accessed August 30, 2013 ).

38.

Chu

, Hsu

, Schwarcz

. Comparison of methods to measure the completeness of HIV/AIDS reporting in San Francisco. In: Oral Abstract Session: XVII International AIDS Conference: Abstract no. THAC0106. Mexico City August 3–8, 2008.

39.

World Health Organization. International Classification of Diseases and Related Health Problems. 9^th Revison:1979. Available from: ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Publications/ICD-9 (Last accessed January 25, 2012 ).

40.

World Health Organization. International Classification of Diseases and Related Health Problems. 10^th Revison:1999. Available from: http://apps.who.int/classifications/icd10 (Last accessed January 25, 2012 ).

41.

State of California, Department of Finance, Report P-3: State and County Population Projections by Race/Ethnicity, Detailed Age, and Gender, 2000–2050. Sacramento, California.

42.

Dean

, Sullivan

, Soe

. OpenEpi: Open Source Epidemiologic Statistics for Public Health, and according to tVersion. www.OpenEpi.com, updated 04/06/2013 (Last accessed December 20, 2013 ).

43.

Rosner

. Fundamentals of Biostatistics, 5 ^th ed., Duxbury, 2000, p. 606.

44.

Mani

, Aboulafia

. Screening guidelines for non-AIDS defining cancers in HIV-infected individuals. Curr Opin, 2013; 5:518–525.

45.

Hoover

, Butler

, Workowski

, et al. Low rates of hepatitis screening and vaccination of HIV-infected MSM in HIV clinics. Sex Transm Dis, 2012; 39:349–353.

46.

Jonckheere

, Vincent

, Belkhir

, Wilmes

, Vandercam

, Yombi

. Adherence to screening guidelines for hepatitis C among HIV-positive patients. AIDS Patient Care STD, 2013; 27:317–319.

47.

Office of the Chief Medical Examiner City and County of San Francisco Annual Report. 2005–2006, 63–65. Available at:http://sfgsa.org/index.aspx?page=943 (Last accessed February 22, 2013 ).

48.

Office of the Chief Medical Examiner City and County of San Francisco Annual Report. 2006–2007, 130–133. Available at:http://sfgsa.org/index.aspx?page=943 (Last accessed February 22, 2013 ).

49.

Office of the Chief Medical Examiner City and County of San Francisco Annual Report. 2007–2008, 135–138. Available at:http://sfgsa.org/index.aspx?page=943 (Last accessed February 22, 2013 ).

50.

Office of the Chief Medical Examiner City and County of San Francisco Annual Report. 2008–2009, 142–145. Available at:http://sfgsa.org/index.aspx?page=943 (Last accessed February 22, 2013 ).

51.

Office of the Chief Medical Examiner City and County of San Francisco Annual Report. 2009–2010, 144–148. Available at:http://sfgsa.org/index.aspx?page=943 (Last accessed February 22, 2013 ).

Changes in Causes of Death Among Persons with AIDS: San Francisco,California,1996–2011