Background

HIV clinical and program outcomes among older patients with HIV enrolled in HIV care and initiating ART in sub-Saharan Africa

Matthew R. Lamb1,2, Eduard Eduardo1,2, Sasi Kandula1, Andrea Howard1,2, Veronica Mugisha1, Davies Kimanga3, Bonita

Kilama4, Wafaa El-Sadr1,2, Batya Elul1,2

For the Optimal Models of HIV Care and Treatment in sub-Saharan Africa consortium

1ICAP at Columbia University2Department of Epidemiology, Mailman School of Public Health, Columbia University

3National AIDS and STI Control Programme, Nairobi, Kenya4National AIDS Control Programme, Dar es Salaam, Tanzania

Background

• Antiretroviral treatment (ART) success results in more HIV-infected individuals living longer and growing older

• HIV-infected older adults (> 50 years of age) have higher mortality, more co-morbidities (noted in studies from resource-rich settings)

• Resource-limited settings: little reported information focusing on HIV+ adults >50 years

Study objectives

To compare HIV+ adults >50 years to those 15-49 years enrolling in HIV clinics in sub-Saharan Africa with respect to:

• Gender distribution

• Median CD4+ count at enrollment and ART initiation

• Mortality and loss to follow-up one year after enrollment and ART initiation

• CD4 response after ART initiation

Study population and setting

• 199 HIV clinics in Kenya, Mozambique, Rwanda, and Tanzania

• Supported by ICAP-Columbia University in partnership with health ministries

• Funded through PEPFAR

• Adults >15 years

• Enrolling into HIV care between Jan 2005-Dec 2010

• Clinical data recorded on national forms as part of routine care, then entered into electronic patient-level databases

• Data quality assessed at least annually

• Each quarter, databases are de-identified and combined across countries

• Approved under PEPFAR-supported public health evaluation “Identifying Optimal Models of HIV Care” (CDC PHE 07.0117)

Data sources

Statistical Methods

Compared adults >50 years vs.15-49 years with respect to:• Gender distribution & median CD4 count at enrollment & ART

initiation using descriptive statistics• Incidence of reported death & loss to follow-up one year after

enrollment and ART initiation: • Kaplan-Meier and log rank tests • Cox Proportional Hazards models, adjusting for clinic- and

patient-level characteristics• Change in CD4+ count after ART initiation:

• 79,087 (44%) of patients had required data• Time-dependent linear regression, adjusting for clinic- and

patient-level characteristics

Results

Population distribution

Country#

Clinics

Enrolling into HIV care (pre-ART and ART)

Initiating ART

Patients%

> 50 years Patients%

> 50 years

Kenya 69 88,008 12%

44,234 12%

Rwanda 44 38,038 10%

21,076 11%

Mozambique 34 211,860 9%

88,063 11%

Tanzania 52 54,253 10%

26,521 13%

Total 199 392,159 9.8%

179,894 11.3%

80% of adults >50 years of age are between 50 and 60 years old

2005

2005-32006

2006-32007

2007-32008

2008-32009

2009-32010

2010-30%

4%

8%

12%

16%

Newly enrolling into HIV care Active in HIV careNewly initiating ART Active on ART

Perc

enta

ge o

f adu

lts >

50

year

s

10.0% p = 0.33

Proportion of adults >50 years attending HIV clinics has increased over time

2005

2005-32006

2006-32007

2007-32008

2008-32009

2009-32010

2010-30%

4%

8%

12%

16%

Newly enrolling into HIV care Active in HIV careNewly initiating ART Active on ART

Perc

enta

ge o

f adu

lts >

50

year

s

14.4%

10.0%

p < .0001

Proportion of adults >50 years attending HIV clinics has increased over time

p = 0.33

12.0% p < .0001

Proportion of adults >50 years attending HIV clinics has increased over time

2005

2005-32006

2006-32007

2007-32008

2008-32009

2009-32010

2010-30%

4%

8%

12%

16%

Newly enrolling into HIV care Active in HIV careNewly initiating ART Active on ART

Perc

enta

ge o

f adu

lts >

50

year

s 14.4%

10.0%

p < .0001

p = 0.33

2005

2005-32006

2006-32007

2007-32008

2008-32009

2009-32010

2010-30%

4%

8%

12%

16%

Newly enrolling into HIV care Active in HIV careNewly initiating ART Active on ART

Perc

enta

ge o

f adu

lts >

50

year

s 14.4%

10.0%

12.0%

15.8%

p = 0.33

p < .0001

p < .0001

p < .0001

Proportion of adults >50 years attending HIV clinics has increased over time

Adults >50 years of age more likely to be male

15-24 yrs 25-39 yrs 40-49 yrs >50 yrs0%

25%

50%

75%

100%

15%

32%

45%50%

FemaleMale

p < .0001

Patients >50 years have lower CD4 counts at HIV care enrollment, similar CD4 at ART initiation

0

100

200

300

400

500

264224

158 163

Med

ian

(IQR)

CD4

coun

t (ce

lls/

mm

3) a

t enr

ollm

ent/

ART

initi

ation

Enrollmentp<0.001

> 50 yrs15-49 yrs

ART initiationp<0.001

Reported mortality is higher among adults > 50 years…

Time (yrs) since enrollment (solid)/ ART initiation (dotted)

Log-rank test:p < .0001

Prop

ortio

n su

rviv

ing

>50 yrs, since ART initiation15-49 yrs, since ART initiation

>50 yrs, since enrollment15-49 yrs, since enrollment

Since enroll-ment

Since ART initiation

>50 yrs 6.0% 6.0%15-49 yrs 4.1% 4.8%

..while loss to follow-up lower amongadults > 50 years

Time (yrs) since enrollment (solid)/ ART initiation (dotted)

Log-rank test:p < .0001

Enroll-ment

ART initiation

> 50 yrs 33% 15%15-49 yrs 42% 19%

Prop

ortio

n re

mai

ning

in c

are

0.6

0.8

1

1.2

1.4

Adjusted

1.18(1.09-1.26)

Adjusting for: country, facility type (10, 20, 30), location (urban, semi-urban, rural), CD4 at ART initiation, sex, and year of ART initiation

Increased mortality, decreased LTF among adults> 50 years remains after multivariate adjustment

DeathLTF

Haz

ard

Rat

io a

nd 9

5% C

I: >

50 y

ears

vs.

15-

49 y

ears

1 ye

ar a

fter A

RT

initi

atio

n

Crude Crude Adjusted

1.26(1.18-1.34)

0.83(0.80-0.86)

0.80(0.76-0.85)

All Patients

15-49 years> 50 years

Adults >50 years had lower CD4+ count response after ART initiation

WomenMen

Age group 95% CI

15-49 years> 50 years

Average one-year increase in CD4 (cells/mm3)

CD4+ count response after ART initiation:> 50 years vs. 15-49 years

Adjusting for: country, facility type (10, 20, 30), location (urban, semi-urban, rural), CD4 at ART initiation, timing of post-initiation CD4 measure(s), sex, and year of ART initiation

Reference-15.9 -4.6 to -27.1130

114.1

Relative one-year increase in CD4 (cells/mm3)

Conclusions• Proportion of active patients >50 years of age increased

steadily over time

• Compared to younger adults, substantially higher proportion of adults >50 years of age are male

• Adults >50 years had better retention, but inferior CD4+ count response and higher mortality after ART initiation

• Additional research needed in the older population:• Reduced CD4+ count response after ART initiation• Reasons for higher mortality and role of co-morbidities• Comparison of mortality with background mortality

Acknowledgements

• Ministries of Health in Kenya, Mozambique, Rwanda, Tanzania

• CDC in-country staff in Kenya, Mozambique, Rwanda, Tanzania

• President’s Emergency Plan for AIDS Relief

• Staff at ICAP-supported sites

• ICAP country and NY staff, especially Elaine Abrams

• People living with HIV