cost-effectiveness of financial incentives for viral suppression · 2017-04-24 · financial...

Cost-Effectiveness of

Financial Incentives

for Viral Suppression

An Economic Model of HPTN 065

Blythe Adamson, MPH, PhD Candidate

University of Washington

Seattle, WA, USA

April 11, 2017

HPTN 065 aimed to assess the feasibility of

the “test and treat” model to decrease HIV

transmission at the community level.

To realize the benefits of anti-retroviral

therapy (ART), financial incentives promoted

linkage to care and viral suppression.

Background

TEST & TREAT FRAMEWORK

Population

DECREASE IN HIV TRANSMISSION

L2C Care site

TREAT

Counsel

TEST

Test site

Adherence & Suppression

VS

THERESA GAMBLE

TEST & TREAT FRAMEWORK

Population

DECREASE IN HIV TRANSMISSION

L2C Care site

TREAT

Counsel

TEST

Test site

Adherence & Suppression

VS

This analysis focuses on the value of

financial incentives (FI)

for viral suppression (VS)

THERESA GAMBLE

39 Clinics

The Bronx, New York Washington, D.C.

THERESA GAMBLE

39 Clinics Randomized

The Bronx, New York Washington, D.C.

To deliver standard HIV care OR care plus incentive for VS

THERESA GAMBLE

You

are

here

• $70 VISA gift cards

• Offered quarterly to patients on ART with viral suppression (VS <400 copies/ml)

• Duration: 2011-2013

• Supported startup costs, Financial Incentives Coordinator, and supplies at each site

Intervention

To evaluate the cost-effectiveness of

providing financial incentives for viral

suppression compared to standard HIV care

for patients using ART to inform public health

decisions in the United States.

Objective

valueassess

Conceptual Model of

Cost-Effectiveness

CHANGE IN HEALTH CARE COSTS ($)

CHANGE IN HEALTH STATUS (QALYs)

Incremental Cost-

Effectiveness Ratio

(ICER)

Based on “Cost-Effectiveness of Antiretroviral Therapy for Prevention” by Kahn et al., 2011

Conceptual Model of

Cost-Effectiveness

CHANGE IN HEALTH CARE COSTS ($)

CHANGE IN HEALTH STATUS (QALYs)

Incremental Cost-

Effectiveness Ratio

(ICER)

Increased

utilization

of services

Averted HIV

Infections

Based on “Cost-Effectiveness of Antiretroviral Therapy for Prevention” by Kahn et al., 2011

Effect of

intervention

Correlation with

healthcare costs

Program Costs

- Admin

- Gift cards

HPTN 065 / modeling

Published data / modeling

+ + -

More

ART use

Averted

Morbidity

+ -

Conceptual Model of

Cost-Effectiveness

CHANGE IN HEALTH CARE COSTS ($)

CHANGE IN HEALTH STATUS (QALYs)

Incremental Cost-

Effectiveness Ratio

(ICER)

More viral suppression

Increased

utilization

of services

Averted HIV

Infections

Averted HIV

Infections

Averted

Morbidity

More

ART use

Based on “Cost-Effectiveness of Antiretroviral Therapy for Prevention” by Kahn et al., 2011

Correlation

with health

Effect of

intervention

Effect of

intervention

Correlation with

healthcare costs

Program Costs

- Admin

- Gift cards

Longer

Survival

HPTN 065 / modeling

Published data / modeling

Averted

Morbidity

+ + + - -

+ ++

Measuring Health Benefits

Quality-Adjusted

Life Years (QALYs)

Health-related quality of life:

utility of health state between

1 (perfect health) and 0 (death)

birth death

1.0

0

Uti

lity

QALYs

HIV+QALYs gained

with intervention

age

QALYs = sum(life year * utility)

To capture the length and

quality of life for patients and

partners

Costs included:

• FI admin & gift cards

• Patient HIV-related healthcare costs– ART drugs

– Clinic visits

– Labs

• Other HIV-unrelated healthcare

• Partner health care costs

• Productivity: earnings

• Consumption: spending

Perspective

PAYER

Costs included:

• FI admin & gift cards

• Patient HIV-related healthcare costs– ART drugs

– Clinic visits

– Labs

• Other HIV-unrelated healthcare

• Partner health care costs

• Productivity: earnings

• Consumption: spending

Perspective

HEALTHCAR

E SECTOR

PAYER

Costs included:

• FI admin & gift cards

• Patient HIV-related healthcare costs– ART drugs

– Clinic visits

– Labs

• Other HIV-unrelated healthcare

• Partner health care costs

• Productivity: earnings

• Consumption: spending SOCIETAL

Perspective

HEALTHCAR

E SECTOR

PAYER

• Cohort-based semi-Markov model of HIV disease progression and primary transmission to sexual partners– Assumes financial incentive

effect diminishes to zero over 6 months after FI end

• Cost-effectiveness analysis: – Patient lifetime horizon

– 3% annual discount rate

Modeling

Approach

PATIENT COHORT

PARTNER COHORT

0% 20% 40% 60% 80%

VS

No

VS

Proportion of patients

<200

200-349

350-499

>500

CD4+ Count

Death

Healthy

<200200

-349

350

-499>500

dead

Program characteristics

and key model inputs Median (Range) Source

Average clinic size, number of patients in care 456 (43 – 2,262) HPTN 065

Baseline proportion of patients virally suppressed 61.9% (8.4 - 84.6%) HPTN 065

Effectiveness: change in viral suppression Average percentage points increase from baseline clinic proportion

3.7% (0.5 – 6.9%) HPTN 065

Increase in outpatient visits due to incentives, % 8.7% (4.2 - 13.2%) HPTN 065

HIV Utility, by CD4 strata 0.69 - 0.73 (0.58 - 0.83) Whitham 2016

Quarterly healthcare costs related to HIV, by CD4 strata Disaggregated by ART, outpatient visits and labs, and other costs [e.g.,

treatment of opportunistic infections]

$4,902 - $7,675

($4,736 – $8,348)Gebo 2010

Program characteristics

and key model inputs Median (Range) Source

Average clinic size, number of patients in care 456 (43 – 2,262) HPTN 065

Baseline proportion of patients virally suppressed 61.9% (8.4 - 84.6%) HPTN 065

Effectiveness: change in viral suppression Average percentage points increase from baseline clinic proportion

3.7% (0.5 – 6.9%) HPTN 065

Increase in clinic attendance, % 8.7% (4.2 - 13.2%) HPTN 065

HIV Utility, by CD4 strata 0.69 - 0.73 (0.58 - 0.83) Whitham 2016

Quarterly healthcare costs related to HIV, by CD4 strata Disaggregated by ART, outpatient visits and labs, and other costs [e.g.,

treatment of opportunistic infections]

$4,902 - $7,675

($4,736 – $8,348)Gebo 2010

Program characteristics

and key model inputs Median (Range) Source

Average clinic size, number of patients in care 456 (43 – 2,262) HPTN 065

Baseline proportion of patients virally suppressed 61.9% (8.4 - 84.6%) HPTN 065

Effectiveness: change in viral suppression Average percentage points increase from baseline clinic proportion

3.7% (0.5 – 6.9%) HPTN 065

Increase in clinic attendance, % 8.7% (4.2 - 13.2%) HPTN 065

HIV Utility, by CD4 strata 0.69 - 0.73 (0.58 - 0.83) Whitham 2016

Quarterly healthcare costs related to HIV, by CD4 strata Disaggregated by ART, outpatient visits and labs, and other costs [e.g.,

treatment of opportunistic infections]

$4,902 - $7,675

($4,736 – $8,348)Gebo 2010

Compared to standard care, for the FI program cohort the model projects:

• Participants on average survive 1 month longer

• Gain of 0.05 QALYs per patient

• 1 HIV infection avoided per 200 FI participants (9% reduction in primary HIV transmission)

RESULTS:

Health

Outcomes

Modeled

RESULTS: Projected lifetime

costs per patient

$0

$50,000

$100,000

$150,000

$200,000

$250,000

Life

tim

e C

ost*

per

Pa

tie

nt

*Limited societal perspective costs using lifetime horizon, discounted 3% annually, and adjusted to 2015 US$

CO

NT

RO

L

INT

ER

VE

NT

ION

Raised cost

Lowered cost

Effect of Intervention on

Lifetime Societal Cost

Total cost

$205,000

$210,000

$215,000

$220,000

Life

tim

e C

ost*

per

Pa

tie

nt

*Limited societal perspective costs using lifetime horizon, discounted 3% annually, and adjusted to 2015 US$

$706

CO

NT

RO

L

INT

ER

-

VE

NT

ION

Raised cost

Lowered cost

Effect of Intervention on

Lifetime Societal Cost

Total cost

$0

RESULTS: Projected lifetime

costs per patient

Cost-Effectiveness

QALYs Gained

Incre

men

tal C

ost

“DOMINANT”:COST SAVING &

GAINS HEALTH

?

3xGDP

1

Willingness to Pay

Cost-effectiveness threshold of

$50,000 - $150,000 per QALY

gained (1-3 x GDP per capita)1xGDP

Cost-Effectiveness

QALYs Gained

Incre

men

tal C

ost

“DOMINANT”:COST SAVING &

GAINS HEALTH

$4,600

Willingness to Pay

Cost-effectiveness threshold of

$50,000 - $150,000 per QALY

gained (1-3 x GDP per capita)

payer

Results reported per participant

$113,900/QALY

ICER, given perspective

HOW TO INTERPRET AN ICER: The incremental cost per

QALY gained from offering financial incentives for viral

suppression as compared to standard HIV care

0.4

Cost-Effectiveness

QALYs Gained

Incre

men

tal C

ost

“DOMINANT”:COST SAVING &

GAINS HEALTH

$4,600

0.5

Willingness to Pay

Cost-effectiveness threshold of

$50,000 - $150,000 per QALY

gained (1-3 x GDP per capita)

payer

health

sector

Results reported per participant

$51,600/QALY

$113,900/QALY

ICER, given perspective

$2,800

Includes

partner

benefit

HOW TO INTERPRET AN ICER: The incremental cost per

QALY gained from offering financial incentives for viral

suppression as compared to standard HIV care

0.4

Cost-Effectiveness

QALYs Gained

Incre

men

tal C

ost

“DOMINANT”:COST SAVING &

GAINS HEALTH

$4,600

0.5

Willingness to Pay

Cost-effectiveness threshold of

$50,000 - $150,000 per QALY

gained (1-3 x GDP per capita)

payer

health

sector

societal

Results reported per participant

Dominant

$51,600/QALY

$113,900/QALY

ICER, given perspective

-$3,500

Includes

partner

benefit

HOW TO INTERPRET AN ICER: The incremental cost per

QALY gained from offering financial incentives for viral

suppression as compared to standard HIV care

0.4

$2,800

• League tables

compare the value

of different

interventions

• Lower ICERs

correspond to

greater value

• Cost-effectiveness

depends on

willingness to pay

for health gains

Comparative Health

Interventions in US

ICER

(Cost/QALY)Source

Financial Incentives for

Viral Suppression

Lower cost &

Health gainsAnalyzed Here

HPV Vaccine$4,000 -

$14,000*Chesson 2008

Statins for Coronary

Heart Disease$22,000 Franco 2005

PrEP in US high risk$120,000 -

$600,000*

Gomez 2013, Paltiel

2009, Juusola 2012,

Desai 2008, and

Koppenhaver 2011

*2005 US$, **2012 US$,

Sensitivity Analysis

-$100,000 $ $100,000 $200,000 $300,000

base-case

ICER

Parameter Range

Low High

ICER, Cost per QALY gained

Effectiveness improving VS

Hazard non-AIDS death <500 CD5

Utilization increase among FI

Avg. rate of disease progression

Incentives not

good investment

in lower range of

VS effectiveness

confidence

0

Cost-Effective

3.7

%

0.5

%

6.9

%

• Financial incentives as used in HPTN 065 are likely to be cost-effective compared to standard HIV care in the US

• Limited by uncertainty in effectiveness

• Implications for global health

• NYC Housing Works Undetectables now provide financial incentives for viral suppression

Summary

The HIV Prevention Trials Network is sponsored by the

National Institute of Allergy and Infectious Diseases,

the National Institute of Mental Health, and the National

Institute on Drug Abuse, all components of the

U.S. National Institutes of Health.

ACKNOWLEDGEMENTS

The HPTN 065 Study team acknowledges

Dobromir Dimitrov, Fred Hutchinson Cancer Research Center

The HPTN Modeling Center

Louis Garrison, University of Washington

Josh Carlson, University of Washington

Ruanne Barnabas, University of Washington

Supplementary Material

Improved viral

suppression

Patient QALYs

Increased

utilization

of services

STEP 5 LONG-TERM

OUTCOMES

TOTAL COSTS

Financial

Incentives

Program Costs

Improved

adherence

Engagement

in HIV care

Coordinator

Supplies

Gift cardsHealthcare Costs

Life

Years

Quality

of Life

Fewer HIV

Transmissions

to Partners

Partner QALYs

Life

Years

Quality

of Life

More

ART Refills

STEP 1INTERVENTION

STEP 2 INTENTION

STEP 3 BEHAVIOR

STEP 4 SHORT-TERM

OUTCOMES

TO

TA

L Q

ALY

s

CONCEPTUAL FRAMEWORK

FOR COSTS & BENEFITS

Cost of Clinic

Visits and LabsDrug Costs

HPTN 065 / modeling

Published literature / modeling

Results from other perspectives

Disease Progression and

Predicted Survival

Modeling Outcomes

Probabilistic Sensitivity Analysis