Non-HIV-related causes of death in people w ith AIDS in New York - - PowerPoint PPT Presentation

▶
non hiv related causes of death in people w ith aids in
SMART_READER_LITE
LIVE PREVIEW

Non-HIV-related causes of death in people w ith AIDS in New York - - PowerPoint PPT Presentation

Jul 03, 2023 12 likes •376 views

Non-HIV-related causes of death in people w ith AIDS in New York City, 1999-2003 DB Hanna, MR Pfeiffer, LV Torian, JE Sackoff HIV Epidemiology Program NYC Department of Health and Mental Hygiene January 27, 2005 AIDS mortality rate by

slide-1
SLIDE 1

Non-HIV-related causes of death in people w ith AIDS in New York City, 1999-2003

DB Hanna, MR Pfeiffer, LV Torian, JE Sackoff

HIV Epidemiology Program NYC Department of Health and Mental Hygiene January 27, 2005

slide-2
SLIDE 2

AIDS mortality rate by underlying cause, NYC, 1988-2003

500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Deaths per 10,000 PWA (age-adjusted) All causes HIV-related Non-HIV-related

AZT, PCP prophylaxis (1987-1988) Combination therapy (1992) Protease inhibitors (1995) HAART (1996)

slide-3
SLIDE 3

AIDS mortality rate by underlying cause, NYC, 1988-2003

500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Deaths per 10,000 PWA (age-adjusted) All causes HIV-related Non-HIV-related

Kristal: AIDS w as leading cause of death for males 30-39 in NYC in 1984

slide-4
SLIDE 4

AIDS mortality rate by underlying cause, NYC, 1988-2003

500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Deaths per 10,000 PWA (age-adjusted) All causes HIV-related Non-HIV-related

Obiri et al: AIDS accounted for 32% of premature mortality (YPLL) in NYC in 1994

slide-5
SLIDE 5

AIDS mortality rate by underlying cause, NYC, 1988-2003

500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Deaths per 10,000 PWA (age-adjusted) All causes HIV-related Non-HIV-related

Chiasson & Wong et al: 72% decline in deaths due to HIV/AIDS 1995- 1998; odds of survival 8x greater w ith HAART

slide-6
SLIDE 6

AIDS mortality rate by underlying cause, NYC, 1988-2003

500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Deaths per 10,000 PWA (age-adjusted) All causes HIV-related Non-HIV-related

Fordyce et al: 2-year survival in PWA increased from 43% (1990-1995) to 76% (1996-1998)

slide-7
SLIDE 7

AIDS mortality rate by underlying cause, NYC, 1988-2003

100 200 300 400 500 600 700 800 900 1,000 1995 1996 1997 1998 1999 2000 2001 2002 2003 Deaths per 10,000 PWA (age-adjusted) HIV-related Non-HIV-related

slide-8
SLIDE 8

Non-HIV-related deaths among person w ith AIDS (PWA) NYC, 1988-2003*

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Proportion of deaths among PWA that are non-HIV-related

* Pre-1999 HIV -related deaths standardized to ICD-10 coding rules

slide-9
SLIDE 9

Non-HIV-related causes of death in PWA: reasons for concern

  • Smoking prevalence is higher in PWA than

in the general population

  • Abuse of alcohol and injection /

non-injection drug use is prevalent

  • Rates of hepatitis C (a major cause of

hepatocellular carcinoma) are ~80% among IDUs

  • Concern that complications secondary to

HAART increase CVD mortality

slide-10
SLIDE 10

The aging PLWA population, NYC, 1993 & 2003

2904 16900 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

1

# of PLWA 50+ 1993 2003 12.9 31.3 5

1

1993 2003 10 15 20 25 30 35 40 45 50 % of PLWA population 50+

slide-11
SLIDE 11

Objectives

  • To analyze non-HIV-related mortality

rates in the post-HAART period (1999- 2003) in NYC:

– overall, in specific subgroups, and for specific causes

  • To identify PWA at greatest risk of a

non-HIV-related death

  • To identify preventable non-HIV-

related illnesses that should be priorities for providers of health-care to PWA

slide-12
SLIDE 12

Population

  • NYC residents 13+ years old

– diagnosed w ith AIDS through 2003 – reported to DOHMH as of 9/30/04 – died 1999-2003 according to

  • match w ith the NYC Vital Statistics

Registry, or

  • field surveillance
slide-13
SLIDE 13

Coding cause of death: A primer

  • Death certificate may include up to

20 contributing or multiple causes

  • Nosologist assigns ICD code to each

contributing cause of death

  • All information on the DC used to

arrive at an underlying cause of death

  • Since 1999, ICD-10 scheme used to

code cause of death

slide-14
SLIDE 14

Source of cause of death

  • NYC Vital Statistics Registry,

1999-2003

  • Match w ith the

National Death Index (NDI), 1999 only

slide-15
SLIDE 15

Classification of HIV-related causes

  • ICD-10 codes for HIV-related deaths:

B20-B24

  • Based on ICD-10 rules, some

OI deaths are not initially coded as HIV-related, because HIV is not listed

  • n the death certifcate
  • Selik proposed expanded criteria to

include all CDC AIDS-defining conditions, if HIV status is know n

– HIV-related deaths ~4% – Our number is greater than w hat Vital Statistics reports

slide-16
SLIDE 16

Classification of non-HIV-related causes

  • Criterion:

Any other know n underlying cause of death

  • Categorized into major subgroups
  • Cause of death unknow n for only

183/11,388 total deaths (1.6%)

slide-17
SLIDE 17

Codes for major CV diseases

Category ICD-10 Chronic ischemic heart disease I20, I25 Acute myocardial infarction I21-I22 Hypertensive diseases I10-I13 Cerebrovascular diseases I60-I69 All other CV disease All other I00-I78, except I46

slide-18
SLIDE 18

Codes for non-AIDS-defining malignant neoplasms

Category ICD-10 Lip, oral cavity, and pharnyx C00-C14 Stomach C16 Colon C18 Rectum and anus C19-C21 Pancreas C25 Larynx C32 Lung, bronchus, and trachea C33-C34 Breast C50 Prostate C61 Leukemia C91-C95 All other sites All other C00-C97

Note: Liver cancer included in substance abuse category

slide-19
SLIDE 19

Codes for substance abuse

Vital Statistics definition

Category ICD-10 code(s) Mental/behavioral disorders due to alcohol F10 Accidental poisoning by alcohol X45 Mental/behavioral disorders due to psychoactive substance F11-F16, F19 Accidental poisoning by psychoactive substance X40-X42, X44

slide-20
SLIDE 20

Codes for substance abuse

“Clinical” definition

Category ICD-10 code(s) Mental/behavioral disorders due to alcohol F10 Accidental poisoning by alcohol X45 Mental/behavioral disorders due to psychoactive substance F11-F16, F19 Accidental poisoning by psychoactive substance X40-X42, X44 Alcoholic liver disease

(incl. alcoholic cirrhosis, alcoholic hepatic failure)

K70 Other alcohol-related conditions

(incl. alcoholic cardiomyopathy, alcohol-induced chronic pancreatitis)

E24.4, G31.2, G62.1, G72.1, I42.6, K29.2, K86.0

Hepatitis C B17.1, B18.2 Liver cancer C22 Other chronic liver disease

(incl. unspecific cirrhosis of liver)

K73-K74 Septicemia A40-A41 Aspiration pneumonia J69.0

slide-21
SLIDE 21

Analysis

  • Average mortality rates per 10,000

PWA (1999-2003), age-standardized to NYC population

– Overall and by sociodemographic factors – For non-HIV-related causes overall and specific causes

  • Multinomial logistic regression to

determine predictors of death from non-HIV-related causes

– Compared to living w ith AIDS – Independent variables: age, race, sex, transmission risk, borough

slide-22
SLIDE 22

Sociodemographic characteristics

PLWA 2003, N=53954 % Median age (years) 45 Sex Male 70.8 Female 29.2 Race Black 45.2 Hispanic 33.6 White 19.6 API/Nat. Am./Other 1.6 PLWA 2003, N=53954 % Borough Manhattan 33.3 Brooklyn 27.1 Bronx 22.6 Queens 15.2 Staten Island 1.9 Transmission risk MSM 25.6 IDU 31.2 Perinatal 0.6 Heterosexual 21.8 Other/Unknown 20.8

slide-23
SLIDE 23

Leading causes of non-HIV-related death in PWA in NYC, 1999-2003

74.4% of all non-HIV-related deaths

Cause of death N % of all deaths (N=11,388) Age-adjusted death rate per 10,000 PWA (95% CI) ALL HIV-RELATED CAUSES 8,543 75.0% 334.3 (321.6, 347.0) ALL NON-HIV-RELATED CAUSES 2,662 23.4% 107.4 (100.8, 114.1) Substance abuse (alcohol and drug) 834 7.3% 23.6 (21.2, 26.0) Major cardiovascular diseases 613 5.4% 30.6 (26.8, 34.4) Malignant neoplasms 534 4.7% 25.5 (22.1, 28.8) Accidents 89 0.8% 2.7 (1.9, 3.5) Chronic lower respiratory diseases 67 0.6% 4.1 (2.5, 5.7) Diabetes mellitus 63 0.6% 2.7 (1.7, 3.7) Intentional self-harm (suicide) 63 0.6% 2.4 (1.3, 3.5) Assault (homicide) 52 0.5% 2.0 (1.0, 3.0) Nephritis, nephrotic syndrome and nephrosis 47 0.4% 1.8 (1.0, 2.6) All other non-HIV-related causes 271 2.4% 11.3 (8.8, 13.8)

slide-24
SLIDE 24

AIDS mortality rate by underlying cause, NYC, 1988-2003

20 40 60 80 100 120 140 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Deaths per 10,000 PWA Major CV diseases Malignant neoplasms Substance abuse

AZT, PCP prophylaxis (1987-1988) Combination therapy (1992) Protease inhibitors (1995) HAART (1996)

slide-25
SLIDE 25

Age-adjusted non-HIV-related mortality rates among PWA in NYC By demographic characteristics, 1999-2003

106.3 113.6 125.6 92.8 89.2 92.8 138.3 114.8 80.4 88.9 60.8 164.8 91.9 107.4 20 40 60 80 100 120 140 160 180 200

Overall Male Hispanic Brooklyn Queens IDU

Deaths per 10,000 PWA

Female Black White/other Manhattan Bronx Staten Island Heterosexual MSM

slide-26
SLIDE 26

Leading specific causes of non-HIV deaths among PWA, by gender

13.1 12.3 10.2 5.5 4.2 16.6 12.4 10.1 9.1 5.1 5.0

5 10 15 20 25 Chronic ischemic heart disease Drug abuse and dependence Lung cancer Hypertensive diseases Acute myocardial infarction Chronic ischemic heart disease Drug abuse and dependence Hypertensive diseases Lung cancer Acute myocardial infarction Diabetes mellitus Deaths per 10,000 PWA (age-adjusted)

Female Male

CV Cancer Substance abuse Other

slide-27
SLIDE 27

Leading specific causes of non-HIV deaths among PWA, by race 15.7 13.6 12.7 9.2 4.4 4.3 12.9 10.0 6.0 5.5 4.3 4.1 15.3 12.1 8.2 4.5 3.4

5 10 15 20 25

Chro nic is chemic heart dis eas e Lung cancer Drug abus e and dependence Hypertens ive dis eas es Chro nic o bs tructive pulmo nary dis eas e Acute myo cardial infarctio n Drug abus e and dependence Chro nic is chemic heart dis eas e Lung cancer Hypertens ive dis eas es Acute myo cardial infarctio n Alco ho l abus e and dependence Chro nic is chemic heart dis eas e Drug abus e and dependence Lung cancer Acute myo cardial infarctio n Intentio nal s elf-harm (s uicide)

Deaths per 10,000 PWA (age-adjusted)

Black Hispanic White and other

Alcohol abuse Chronic ischemic heart dis. Drug abuse Lung cancer Acute MI Suicide Acute MI Acute MI Chronic ischemic heart dis. Chronic ischemic heart dis. Drug abuse Drug abuse Lung cancer Lung cancer Hypentensive diseases Hypentensive diseases COPD

CV Cancer Substance abuse Other

slide-28
SLIDE 28

Leading specific causes of non-HIV deaths among PWA, by risk 10.2 6.1 5.3 3.7 3.3 14.3 14.0 8.6 6.9 6.8 6.7 15.3 11.1 6.9 5.7 5.1 35.8

5 10 15 20 25

Chro nic is chemic heart dis eas e Intentio nal s elf-harm (s uicide) Lung cancer Acute myo cardial infarctio n Drug abus e and dependence Drug abus e and dependence Lung cancer Chro nic is chemic heart dis eas e As s ault (ho micide) Hypertens ive dis eas es Alco ho l abus e and dependence Liver cancer Lung cancer Chro nic is chemic heart dis eas e Acute myo cardial infarctio n Hypertens ive dis eas es P ro s tate cancer

Deaths per 10,000 PWA (age-adjusted)

MSM IDU Heterosexual

Hypentensive diseases Hypentensive diseases Chronic heart disease Chronic heart disease Chronic heart disease Lung cancer Lung cancer Lung cancer Suicide Acute MI Acute MI Drug abuse Drug abuse Alcohol abuse Homicide Liver cancer Prostate cancer

CV Cancer Substance abuse Other

slide-29
SLIDE 29

Predictors of non-HIV-related deaths

Gender (ref. = male) Female 0.93 (0.86, 1.02) 0.91 (0.83, 1.00) Race (ref. = white/other) Black 1.28 (1.15, 1.42) 0.99 (0.89, 1.12) Hispanic* 1.05 (0.94, 1.18) 0.80 (0.71, 0.91) Borough of residence (ref. = Manhattan) Bronx 1.14 (1.03, 1.27) 1.04 (0.92, 1.16) Brooklyn* 1.28 (1.16, 1.41) 1.22 (1.10, 1.36) Queens 0.90 (0.79, 1.03) 0.93 (0.81, 1.07) Staten Island 1.22 (0.92, 1.60) 1.11 (0.84, 1.47) Transmission risk (ref. = MSM) IDU* 3.43 (3.04, 3.86) 3.37 (2.96, 3.84) Heterosexual* 1.37 (1.18, 1.58) 1.36 (1.16, 1.60) Perinatal 0.07 (0.01, 0.51) 0.17 (0.02, 1.34) Other* 3.27 (2.20, 4.88) 2.90 (1.93, 4.37) Unknown* 1.43 (1.24, 1.65) 1.41 (1.21, 1.65)

  • Adj. OR (95% CI)

Crude OR (95% CI)

Odds ratio compared to PLWA at end of 2003. Logistic regression model also adjusted for age, poverty level, year of diagnosis. * Adj. p<0.001

slide-30
SLIDE 30

Conclusions

  • HIV-related causes continue to

dominate mortality among PWA in NYC

  • How ever, the proportion of deaths due

to non-HIV-related causes has increased and accounted for 26% of all PWA deaths in 2003

  • The risk of non-HIV-related deaths is

significantly higher among IDU, and varies significantly by age and borough

slide-31
SLIDE 31

Conclusions

  • Cardiovascular disease, cancer

and substance abuse account for ~75% of non-HIV-related deaths –Specific causes generally consistent by race, sex, risk

slide-32
SLIDE 32

Recommendations for Health Care Providers to PWA

  • Major causes of death are

amenable to prevention or intervention

– Screening for cancers, including prostate, prostate and colon – Providing support for lifestyle changes, e.g., nicotine patch, buprenorphine

  • Take Care New York priority areas

apply equally to PWA

slide-33
SLIDE 33

“Health care for people w ho are HIV-infected should shift from an [exclusive] focus on preventing HIV-related morbidity and mortality to include behaviors and conditions that are increasingly accounting for mortality.”

Primary Care Guidelines for the Management of Persons Infected w ith HIV, Recommendations of the HIV Medicine Association of the IDSA, 2004

slide-34
SLIDE 34

Limitations

  • Classification of cause of death not

validated by chart review

  • Substance abuse definition may have

missed some related deaths

  • Deaths occurring outside NYC (~6%)

are not ascertained

  • Small numbers for subgroup analyses

and trends for specific causes

  • Regression model did not include

laboratory values for CD4 count or viral load

slide-35
SLIDE 35

Next Steps

  • Further analyses w ithin subgroups

defined by geography and risk

  • Comparison to causes of death

among all NYC residents not know n to be HIV+

  • Comparable analysis of deaths

among people w ith HIV (not AIDS)

  • Analysis of economic impact of non-

HIV-related mortality using years of potential life lost (YPLL)

slide-36
SLIDE 36

Acknow ledgments

  • HIV Epidemiology Program

– Sonny Ly and Data Support Unit – Surveillance Unit

  • Office of Vital Statistics
  • Richard Selik, MD (CDC)