Costing MAT for HIV Prevention among PWIDs in Georgia
07/20/12 Veena Menon & Irma Kirtadze 1
Outline
• Background • Study Design
• Findings
2
Background ● One of the first countries to adopt
Christianity, in the early 4th century, remains strongly Georgian Orthodox
● From 1921 to 1991, the Georgian Soviet
Socialist Republic was one of the 15 soviet republics of the USSR
● Suffered from civil war and economic crisis during the 1990s
● Rose Revolution of 2003, at which time the new government introduced democratic and economic reforms
● It covers a territory of 69,700 square
kilometers ● Its current population is almost 4.5 million
3
Evaluate costs of MAT programs in Georgia to improve budgeting and planning related to MAT programs
Study Goal
4
• Estimate unit costs of MAT provision for one patient (per month)
• Optimize current capacity to include more PWIDs
Study Objectives
5
Background
6
• PWID population approximately 40,000
• Estimated number of HIV+ people in Georgia - 4000
• Injecting drug use identified as primary cause of transmission
PWID 56% Hetero-
sexual 37%
MSM 3%
MTCT 2%
Other 2%
Source: AIDS and Clinical Immunology Research Centre (2011).
MAT providers
GFATM - Start year 2005 - 5 sites in total - Free for clients
MOHLSA - Start year 2008 - 11 sites in total
- 150 GEL/pm/patient
Program Characteristics
7
Program Characteristics (2)
8
GFATM MAT centers MOLHSA MAT centers GFATM prison centers
• MAT treatment protocol/personnel centrally-mandated and identical for all service providers – Key treatment parameters include
• Daily methadone dosage • STI testing (HIV, Hepatitis B/C) • Individual psychotherapy • Group counseling
Program Characteristics (3)
9
Program Characteristics (4) • Staffing at MAT sites is regulated by
Georgian law: – 3 medical doctors – 1 psychologist – 1 social worker – 1 head of department/chief doctor/
coordinator – 1-3 nurses – Other support staff (data entry specialist,
consultants, toxicologist, pharmacist and etc.) 10
Study Details • Study design – samples from each provider
– MOHLSA – 7 of 11 sites (sites chosen to ensure representation from small and large population centers)
– GFATM – 4 of 5 sites (excluded prison site) • Study period
– Qualitative data collected on treatment parameters – March, April and May 2011
– Financial program data – 2009, 2010
11
Study Details (2) • Collected cost and quantity data on:
• Personnel • Drugs and supplies • Furniture and equipment • Land and buildings • Consumables • Utilities
12
Results MAT Dosages Across Sites
13
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
100%
% o
f pat
ient
s w
ithin
eac
h do
se ra
nge Below 60 mg Above 60 mg
Distribution of HIV+ patients
14
87.3% 87.8%
98.9%
98.5% 12.7% 12.2%
1.1%
1.5%
0
100
200
300
400
500
600
700
800
GF 2009 GF 2010 MOLHSA 2009 MOLHSA 2010
Tota
l Num
ber
of P
atie
nts
HIV Negative HIV Positive
Increase in Average # Patients
15
0
100
200
300
400
500
600
700
800
GF—MAT MOLHSA—MAT
2010 $
GFATM Average number of PWID in 1 month 364
Unit cost for one month 229 GEL $133
MOHLSA Average number of PWID in 1 month 689
Unit cost for one month 236 GEL $137
Unit cost for 1 month treatment
16
Cost Drivers
17
78.9% 79.1% 75.4% 76.8%
21.1% 20.9% 24.6%
23.2%
0
50
100
150
200
250
MoLHSA 2009 MoLHSA 2010 GF 2009 GF 2010
Uni
t C
ost
(GE
L)
Direct Costs Indirect Costs
Share of direct and indirect costs in MAT facilities
Key cost drivers
18
Salaries of clinical & support
staff 51%
Regional monitoring
0%
Stationary office supply
0%
Drugs and medical supplies
28%
Rental cost 5%
Utilities 14%
Bank/ insurance
0%
Non medical
equipment 2%
MOLHSA—2010
Salaries of clinical &
support staff 55%
Regional monitoring
0.41%
Stationary office supply
1%
Drugs and medical supplies
21%
Rental cost 5% Utilities
13%
Bank/ insurance
fees 0.44%
Non medical
equipment 5%
GFATM—2010
Economies of Scale
19
GFATM Shida Kartli Narcological Center
GFATM Batumi Narcological Center
MoLHSA Division 3
MOLHSA Division 6
Average patients per month
45 117 18 142
Unit cost patient/ month
291 GEL ($169)
161 GEL ($93)
702 GEL ($407)
122 GEL ($71)
Potential for Service Expansion • 1382 maximum PWID capacity at MOLHSA sites • 500 maximum PWID capacity at GFATM sites
(currently at capacity) • Modelled 2 scenarios to assess potential for
expanding MAT coverage within existing facilities: – Scenario A: Increase coverage by 15 percent until
maximum available existing capacity is reached. – Scenario B: Graduated expansion in MAT coverage over
five years (five, seven, nine, 11 and 13 percent increases) using existing sites until maximum available capacity is reached. 20
Scenario A – 15% increase
21
MOLHSA Increase in average # of patients
2010—11 MAT sites 898
2011—15% increase 1,033
2012—15% increase 1,187
2013—15% increase 1,366
Scenario A – Decreasing unit costs
22
898 1033
1187
1366
235.8 213.21 193.58 176.5
0
50
100
150
200
250
300
350
400
450
0
200
400
600
800
1000
1200
1400
1600
2010 extrapolated 2011 2012 2013
Uni
t C
ost
(GE
L)
Num
ber
of P
atie
nts
number of patients unit cost
Scenario A – Share of direct/indirect costs
23
MOLHSA Direct costs (%)
Indirect costs (%)
2010—11 MAT facilities
79.10 20.90
2011—15% increase 79.90 20.10
2012—15% increase 80.75 19.25
2013—15% increase 81.64 18.36
Scenario B – graduated increase
24
MOLHSA Graduated increase in average # of patients
2010—11 MAT sites 898
2011—5% increase 943
2012—7% increase 1009
2013—9% increase 1100
2014—11% increase 1221
2015—13% increase 1379
Scenario B – Decreasing Unit Costs
25
897.9 942.8 1008.8
1099.6 1220.5
1379.2
235.8 225.6 213.2 200.7 189.1 179.4
0
50
100
150
200
250
300
350
400
450
0
200
400
600
800
1000
1200
1400
1600
2010 extrapolated
2011 2012 2013 2014 2015
Uni
t C
ost
(GE
L)
Num
ber
of P
atie
nts
Number of Patients Unit cost
26
MOLHSA Direct costs (%)
Indirect costs (%)
2010—11 MAT facilities 79.10 20.90
2011—5% increase 79.47 20.53
2012—7% increase 79.88 20.13
2013—9% increase 80.43 19.57
2014—11% increase 81.11 18.89
2015—13% increase 81.94 18.06
Scenario B – Share of direct/indirect costs
• We were unable to include certain costs in this analysis, primarily due to lack of data.
• Accurate patient tracking was a challenge in some
MOLHSA methadone facilities, particularly Tbilisi, where patients may have moved from one MAT site to another.
• Finally, this analysis focused on the costs incurred by
facilities offering MAT services. As such, it does not include the costs incurred by patients as they sought and received services or the opportunity costs of time spent travelling to and seeking services.
Limitations
27
• Specify clear indicators for successful treatment outcomes.
• Integration/co-location of PWID/HIV services for better patient follow-up (currently no data on recidivism)
• Greater autonomy for MAT sites on protocol/personnel
Recommendations
28
Acknowledgements USAID funded project HPI costing • Veena Menon, Kip Beardsley, Stiven Forsythe (Futures
Group) & data collectors team : Natia Topuria, Nino Balanchivadze, Tea Kordzade (Alternative Georgia), Lela Sturua (NCDC)
Staff of participant MAT centers’
Authors declaring no conflict of interest
29