Transcript CDC Presentation - View the full AIDS 2016 programme

Anticipated Reductions in Long-Term
Tuberculosis Incidence and Associated Cost
Savings with Adoption of the Treat All
People Living with HIV Policy in Botswana,
2016–2035
Dr. Botshelo Kgwaadira
Manager, Botswana National Tuberculosis Program
Botswana Ministry of Health
PRESENTATION OUTLINE
•
•
•
•
•
•
•
BACKGROUND
OBJECTIVES
METHODS
FINDINGS
LIMITATIONS
CONCLUSIONS
NEXT STEPS
BACKGROUND
TB Services
• TB services delivered through
network of >630 health facilities
• MDR-TB care decentralized to 5
treatment initiation sites
• Laboratory Services
– AFB microscopy service points: 48
– Culture, Phenotypic and Genotypic
(LPA) Drug Sensitivity Testing: 1
– Xpert instruments: 34
MDR Treatment initiation sites
750
ARV Program Launched
50
600
40
450
30
300
20
150
17.1%
17.6%
18.5%
0
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
0
10
TB Notification Rate
HIV Prevalence Antenatal Sentinel Surveillance
HIV Prevalence AIDS Indicator Surveys
HIV seroprevalence (%)
TB Notification Rate (cases per 100,000)
Tuberculosis (TB) Notification Rate, 1990–2014
HIV Prevalence: Antenatal Surveillance and
AIDS Indicator Surveys
Integrated TB/HIV services
• Sustained scale up of integrated services for
TB patients living with HIV (PLHIV) (2014)
– HIV testing is close to universal (93%)
– Cotrimoxazole prophylaxis for TB/HIV patients is
close to universal (95%)
– ART uptake has significantly improved (80%)
1800
100
1600
90
1400
80
70
1200
60
1000
50
800
40
600
30
400
20
200
10
0
0
2010
2011
2012
2013
Year
TB incidence among PLHIV
TB incidence among persons without HIV
ART coverage
2014
PLHIV receiving ART (%)
Cases per 100,000 population
TB incidence estimates and ART coverage among PLHIV and
persons without HIV, Botswana, 2010-2014
End TB Strategy 2035
• To reduce TB deaths by 95%
and to cut new cases by 90%
between 2015 and 2035 and
to ensure that no family is
burdened with catastrophic
expenses due to TB.
Targets
2020
2025
2030
2035
Reduction in number of TB deaths 35%
compared with 2015 (%)
75%
90%
95%
Reduction in TB incidence rate
compared with 2015 (%)
50%
80%
90%
20%
Objectives
• To estimate the impact of ART on TB incidence
comparing two scenarios using the output of
the Botswana HIV Spectrum Model*
• To estimate the TB cost savings of moving
from Baseline Scenario to a Treat All Scenario
* Botswana Spectrum Model v5.41, 2016
Methods
•
•
•
•
Estimated the TB incidence assuming two scenarios
1. Baseline Scenario: Current ART eligibility (CD-350, B+, current prevention
coverage)
2. Treat All Scenario: Universal treatment using a Dolutegravir (DTG) regimen,
maximum prevention coverage
Used Spectrum module inputs for the two scenarios
– Number of total PLHIV per year
– Number of PLHIV on ART per year
– Median CD4+ cell count at ART initiation per year
TB Assumptions
– Module focuses only on TB among PLHIV
– Differential TB incidence among PLHIV on ART and not on ART
– Lower TB incidence among PLHIV in the Treat All scenario related to higher
CD4+ cell count
– Sensitivity Analysis
• Low, medium and high estimates of TB incidence on ART in the Treat All
scenario
Timeframe: 2016–2030
TB/HIV Epidemiology Assumptions
Model Input
Value
Source/Comment
TB Incidence among PLHIV
1.4%
Numerator: Estimated from TB notification adjusted
for case detection, 59% TB-HIV co-infection;
Denominator 390,000 PLHIV in 2014 (UNAIDS)
IRR: TB Incidence among PLHIV on ART/TB
incidence among PLHIV not on ART
0.39
Williams et al. PNAS (2010) 107 (45): 19485–19489
Increase in TB incidence rate per decrease by
100 CD4+ cells/μL
0.25
Williams et al. Curr HIV/AIDS (2015) 12:196–206
TB Incidence among PLHIV not on ART
2.3%
Derived from TB incidence among all PLHIV assuming
based on the proportion on ART and proportion not
on ART (program data) and IRR assumption above
TB Incidence among PLHIV on ART Baseline
Scenario
0.9%
Derived from TB incidence among all PLHIV assuming
based on the proportion on ART and proportion not
on ART (program data) and IRR assumption above
TB Incidence among PLHIV on ART Treat All
Scenario
0.38%
-0.9%
Derived from TB incidence among all PLHIV based on
the proportion on ART and proportion not on ART and
IRR assumption above, sensitivity analysis.
MDR TB inputs
Derived in a similar fashion to TB inputs based on an
annual incidence of MDR TB of 0.053% among PLHIV
Annual Total TB HIV Cases by ART Treatment Scenario,
2016-2030 – sensitivity analyses
Cumulative
Difference in
TB Cases
2016–2030
No. cases
7000
6000
25,591 cases
5000
47,641 cases
55,906 cases
4000
3000
2000
1000
0
2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Baseline
Treat All_DTG_2016
Year
Treat All_DTG High TB Incidence Rate (0.9%)
Treat All_DTG Low TB Incidence Rate (0.38%)
Base Costs of TB Treatment*
Type of TB
Cost of Treating
one case*
New Drug Sensitive TB
$233
Previously Treated TB
$277
Drug Resistant TB (MDR or higher)
$9024
*True costs are much higher, NOT included here: human resource costs (clinic/hospital, lab or
program staff); administrative overhead costs; hospitalization/inpatient costs; DOT costs; clinical or
medical commodities apart from drugs and labs; sample transport costs; patient productivity loss;
patient transport costs.
Annual Base Costs of TB Treatment by ART Scenario,
2016-2030 – high, medium, low estimates
Annual TB Related Base
Costs
Cumulative
Cost
Difference
2016–2030
$4,000,000
$3,500,000
$3,000,000
$14.2 million
$2,500,000
$26.5 million
$31.1 million
$2,000,000
$1,500,000
$1,000,000
$500,000
$0
2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
Baseline
Treat All_DTG_2016
Year
Treat All_DTG High TB Incidence Rate (0.9%)
Treat All_DTG Low TB Incidence Rate (0.38%)
Base Cost by Type of TB
Cumulative 2016-2030
Low TB Rate
Estimate*
Drug Sensitive TB
Medium TB Rate
Estimate*
High TB Rate
Estimate*
$12,951,531 (42%)
$11,036,755 (42%)
$5,928,654 (42%)
$18,163,167 (58%)
$15,477,895 (58%)
$8,314,317 (58%)
$31,114,698
$26,514,650
$14,242,971
Drug Resistant TB
(MDR or higher)
Total
*Refers to the TB incidence on ART under the Treat All scenario. Due to uncertainty regarding the TB incidence
under the Test All strategy (based on immune function recovery), compared to the baseline strategy, we have
included high, medium and low incidence estimates
Model Limitations
• Dependent on HIV Spectrum Model estimates
• Crude model
– Does not use historical fit/TB epidemic dynamics to predict
future trends in TB incidence
– Uses crude estimates of incidence of TB on ART and not on
ART for the two different scenarios
– Does not currently take into account increased incidence
of TB during first year of ART
• Does not reflect
– The epidemiology of TB among non-HIV infected persons
– The TB transmission dynamics in the population
Conclusions
• Implementing Treat All policy rapidly will result in
a reduction in TB and drug resistant TB cases and
save costs:
Cumulative TB Base Cost Savings of >$25 million
(medium estimate) from 2016–2030
• Treat All policy will be an important intervention
for TB prevention and control, though, reaching
Botswana’s END TB goals by 2035 will require TB
specific interventions in addition to Treat All
Next Steps
• Estimation of overall impact of Treat All on TB
epidemic
• In-country modeling to see what it will take to
meet END TB goals
– TB prevention and control beyond Treat All
– Use results to adjust program policies and strategies
• Triangulate results with other established TB/HIV
models
• Refine HIV and TB Costing to identify
unaccounted for cost savings
– e.g., GDP, tax contributions
Acknowledgements
We thank all contributors from
• Botswana Ministry of Health
• National Tuberculosis and Leprosy Programme
• Department of AIDS Prevention and Care
• US Centers for Disease Control and Prevention
• Robert Koch Institute, Germany
• Careena Centre for Health, Botswana
• US Agency for International Development
• UNAIDS
• WHO
and Brian G Williams, Stellenbosch University
Ke a leboga
Thank You
School children in Gantsi District, Botswana
demonstrate good cough etiquette