1
2
3
4
When would the distribution of tenofovir gel be cost
saving? Model projections of HIV impact and break-even
product prices in urban South Africa
5
Fern Terris-Prestholt1§*, Anna M Foss1*, Andrew P Cox1, Lori Heise1, Gesine Meyer-Rath
6
1,2,3
7
Charlotte H Watts1
, Sinead Delany-Moretlwe4, Thomas Mertenskoetter5, Helen Rees4, Peter Vickerman1 and
8
9
1
London School of Hygiene & Tropical Medicine, London, UK;
10
2
Health Economics and Epidemiology Research Office, Department of Medicine, Faculty of
11
Health Sciences, University of the Witwatersrand, Johannesburg, South Africa;
12
3
Center for Global Health and Development, Boston University, Boston, USA;
13
4
Wits Reproductive Health and HIV Research Institute, Johannesburg, South Africa;
14
5
International Partnership for Microbicides, Silver Spring, MD, USA.
15
16
*These authors contributed equally to this work
17
§
Corresponding author
18
19
Email addresses:
20
FTP: Fern.Terris-Prestholt@lshtm.ac.uk
21
AF: Anna.Foss@lshtm.ac.uk
1
22
AC: andrewpaulcox@gmail.com
23
LH: Lori.Heise@lshtm.ac.uk
24
GMR: gesine@bu.edu
25
SDM: sdelany@wrhi.ac.za
26
TM: tmertenskoetter@t-online.de
27
HR: hrees@wrhi.ac.za
28
PV: Peter.Vickerman@lshtm.ac.uk
29
CW: Charlotte.Watts@lshtm.ac.uk
30
2
31
Abstract
32
33
Background
34
CAPRISA 004 trial showed that a tenofovir-based vaginal microbicide had significant impact
35
on HIV incidence among women. This study uses the trial findings to estimate the
36
population-level impact of the gel on HIV and HSV-2 transmission and price thresholds at
37
which widespread product introduction would be cost-saving in urban South Africa.
38
39
Methods
40
dynamic HIV/STI transmission model, parameterised and fitted to Gauteng (HIV prevalence
41
of 16.9% in 2008), South Africa, was used to estimate the impact of gel use over 15 years.
42
Uptake was assumed to increase linearly to 30% over 10 years, with gel use in 72% of sex-
43
acts. Full economic programme and averted HIV treatment costs were modelled. These were
44
used to identify the break-even microbicide price, where the full costs of product delivery
45
would be offset by the cost savings associated with averting HIV infection.
46
47
Results
48
could lead to a 10.5% and 5.7% reduction in HIV and HSV-2 incidence respectively, by year
49
15. The break-even gel price was $0.41 per dose. Ceteris paribus, to be cost-saving, a more
50
effective microbicide (83%) could cost up to $0.79, while a 5% reduction in condom use
51
would almost halve the break-even price.
52
53
Conclusions
54
incidence by 10.5% in this setting, if current condom use is maintained. Even including full
55
economic programme costs of distribution, at projected prices Tenofovir gel is likely to be
56
cost-saving, if produced at scale.
There is urgent need for effective HIV prevention methods that women can initiate. The
The estimated ‘per sex-act’ HIV and HSV-2 efficacies were imputed from CAPRISA 004. A
Using plausible assumptions about product introduction, we predict that tenofovir gel use
These findings show that an effective coitally-dependent microbicide could reduce HIV
3
57
58
Key words: HIV, modelling, cost-effectiveness, microbicides, South Africa, introduction of
59
new technologies, economic analysis, tenofovir, ARV-based prevention, Pre-exposure
60
prophylaxis (PreP)
4
61
62
Background
63
After a decade of little progress in HIV prevention interventions, recent trials have shown
64
that antiretrovirals can be used to prevent HIV incidence and infectivity [1-4]. In 2010, the
65
CAPRISA 004 trial in Kwazulu Natal, South Africa, showed that the incidence of HIV
66
infection among women randomised to using a topical gel containing 1% of the antiretroviral
67
tenofovir before and after sex, was reduced by 39% (95% CI 6-60%). Women’s risk of
68
acquiring herpes simplex virus type-2 (HSV-2) was also reduced by 51%, (95% CI 22-70%)
69
[5-6]. This finding was groundbreaking, being the first clinical trial to show a statistically
70
significant impact of a microbicide product on HIV incidence, and providing a proof of
71
concept that a topically applied antiretroviral microbicide can reduce women’s risk of HIV
72
acquisition [7]. The FACTS 001 trial is now seeking to confirm the CAPRISA results, and
73
assess the generalizability of the findings across different settings [8]. Plans are underway to
74
prepare for the possible introduction and manufacture of tenofovir gel. Once this trial is
75
completed, its data will be combined with the CAPRISA 004 results and presented together
76
for regulatory approval. This study aims to inform this process and the field of HIV more
77
broadly.
There is an urgent need for effective HIV prevention methods that can be initiated by women.
78
79
Evidence from FACTS 001 and this cost-effectiveness analysis will be evaluated in the
80
context of VOICE trial results that showed that daily administration of the same product, and
81
oral dosing did not achieve a significant reduction in HIV incidence due to adherence issues
82
[9]. It appears that compliance to a daily regimen of gels or pills is not a viable option for
83
African women in particular for younger women. There is still hope however that a coitally
84
specific product may provide a feasible strategy for women who currently lack options they
5
85
can implement. Despite disappointing adherence in the VOICE trial, CAPRISA did show an
86
effect, especially among high adherers.
87
88
In parallel to clinical research, it is important to explore the potential population-level impact
89
that tenofovir gel introduction could have on HIV incidence in different epidemic settings,
90
and identify key factors that may influence the impact and cost-effectiveness of future
91
product introduction. Towards this end, this paper uses modelling and the CAPRISA trial
92
results to estimate the gel’s likely level of protection against HIV and HSV-2 during each
93
sex-act (here termed the ‘imputed efficacy’), and the potential population-level impact on
94
HIV and HSV-2 of gradual product introduction in an urban setting with high HIV
95
prevalence and incidence in South Africa. Using South Africa specific unit cost data, the full
96
cost of introducing microbicides into the public health sector are modelled, and the break-
97
even price, where the costs of introduction are offset by the associated treatment cost savings,
98
is explored. A range of scenarios are used to identify a threshold price under which
99
microbicide distribution would still be a cost-saving intervention.
100
101
102
Methods
103
104
Setting
105
urbanised province of South Africa with a population of reproductive age (aged 15 – 45
106
years) of 5.7 million. HIV prevalence among those aged 15-49 was 16.9% in 2008 [10], with
107
a higher prevalence among females (21.1%) than males (11.2%) documented in 2005 [11].
108
Commercial sex work is an important driver of the HIV epidemic in Gauteng [12], with
For the modelling analysis, we considered product introduction in Gauteng Province, a highly
6
109
behavioural surveys suggesting that 6.4% of female family-planning clinic attendees had
110
transactional sex in the last four weeks [13].
111
112
113
Microbicide introduction scenarios modelled
114
internationally were used to inform the microbicide introduction strategies considered in this
115
paper [14]. As described in more detail below, we considered the provision of a microbicide
116
through the public health system, assuming that mass media campaigns are used to increase
117
awareness of the product availability; health providers are trained to provide the product, and
118
they then routinely provide HIV testing, and counsel HIV negative clients about product use
119
and the importance of adherence and repeated HIV testing [14].
A review of evidence and discussions with key stakeholders in South Africa and
120
121
As in the CAPRISA 004 trial, we consider a coitally-dependent gel, with two doses used per
122
sex-act. We assumed that the mean consistency of gel use is the same as in the CAPRISA 004
123
trial (72% of sex-acts)1, irrespective of the type of partnership. In addition, a more pessimistic
124
scenario of 50% gel use was considered.
125
126
In the CAPRISA trial, reported condom use increased from 79% in the first six months to
127
84% in the last six months (P < 0.001) [5].2 In this analysis we considered two potential
128
scenarios firstly, that condom use is unaffected by gel use and HIV testing. Secondly, we
129
consider a more pessimistic assumption that there is a 5% and 10% absolute reduction in
130
condom use among microbicide users, and explored the thresholds of condom migration for
131
microbicide introduction to lead to additional HIV infections averted.
7
132
133
In this analysis, we consider the impact on HIV and HSV-2 incidence of population-level gel
134
distribution over 15 years, starting in January 2013. We assume that, following regulatory
135
approval, tenofovir is made available to HIV-negative sexually active women attending
136
public health facilities, alongside regular HIV testing.
137
138
Based on a review of the evidence on the rate of uptake of condoms and other new health
139
technologies [15], we assume that for the first 10 years the proportion of HIV-negative
140
women using the gel increases linearly, plateauing at 30%, half the rate at which urban
141
sexually active women are currently accessing modern contraceptives in Gauteng [16], for
142
the remaining five years. A more optimistic scenario of 60% uptake after 10 years was also
143
explored.
144
145
146
147
Imputed efficacy calculations
148
across all users in the active arm of the trial, including those who used the gel consistently or
149
intermittently, and those who did not use it at all [17]. As a result, for the modelling we need
150
to impute the level of protection provided against HIV and HSV-2 infection by the product in
151
each sex-act between a susceptible woman and an infected partner (how much it reduces the
152
probability of transmission during each act of intercourse). This was estimated by dividing
153
the trial effectiveness by the percentage of all reported sex-acts in which gel was used [5-6,
154
17-18]. Details of this calculation and the assumptions made are provided in the online
The CAPRISA 004 trial results represent an estimate of the average protection achieved
8
155
supplementary material. The imputed efficacy estimates were then used in the dynamic
156
impact model. We also investigated the contribution of the gel’s HSV-2 efficacy has on HIV
157
transmission and vice versa.
158
159
160
Description of the model structure
161
compartmental HIV/STI transmission model, adapted to enable the impact of the gradual
162
introduction of a microbicide to be explored [19]. The mathematics underlying the dynamic
163
transmission model are described in Vickerman et al. (2010) [20], and in the online
164
supplementary material. In brief, the model divides the heterosexual population into
165
subgroups, with stratifications by sex, level of sexual activity and HIV/STI infection status.
166
Using data on the probabilities of disease transmission and progression from the
167
epidemiological literature, established mathematical techniques are used to estimate how HIV
168
spreads between the different subgroups over time [21-22]. The transmission of HSV-2,
169
syphilis and gonorrhoea/chlamydia are simulated, including the facilitating effect of these
170
STIs on HIV transmission, and the effect of HIV on increasing the rate of HSV-2
171
symptomatic recurrences and HSV-2 infectivity [23]. The model also incorporates the higher
172
HIV infectivity associated with primary and late-stage HIV infection (in the absence of HIV
173
treatment), and the effect of ongoing STI treatment and the antiretroviral therapy (ART) in
174
the population on levels of HIV infectivity [24], of which more detail can be found in the
175
online supplement.
The HIV and HSV-2 impact projections were obtained using a population-level deterministic
176
9
177
178
Fitting the model to the setting
179
Gauteng, and complementary data from South Africa (Figures 1a,b,c and online
180
supplementary material). The approach used incorporates the uncertainty inherent in key
181
model parameters by running the model over a large number (almost 4 billion) of
182
permutations of model inputs. These are obtained by randomly sampling from the uncertainty
183
ranges for each model input parameter (supplement material Tables 2-4) [20]. As is standard
184
in HIV modelling, the model is seeded with a few HIV infected individuals (0.5% in the
185
general population and 5% among female sex workers) in 1980 and run until 2028. A ‘fit’ is
186
defined as a model simulation that lies within the 95% confidence intervals (CIs) of setting-
187
specific HIV and STI prevalence data. The model was fitted to HIV data from three national
188
surveys, using male and female prevalence for 2002, 2005 and a combined prevalence for
189
2008 (five HIV prevalence data points in total) [10-11, 25]; and HIV prevalence data from
190
female sex workers (FSWs) (1997) and their clients (2000). It was also fitted to syphilis and
191
gonorrhoea/chlamydia prevalence data from FSWs (1997) [12, 26]. HSV-2 data from the
192
CAPRISA trial [5], adjusting for HIV status [10], was used as a proxy for the general
193
population HSV-2 prevalence. Each combination of input values that fit to the observed
194
prevalence data is used in the final impact projections, with the best fit identified through
195
least chi-squared error.
The model was parameterised and fitted using behavioural and epidemiological data from
196
197
[INSERT FIGURE 1 HERE]
10
198
199
200
Methods used for the economic evaluation of gel introduction
201
the HIV care and treatment costs saved by averting HIV infections. This represents a scenario
202
of more public health benefits for the same public health cost and is a very conservative
203
approach. Alternative decision rules could be based on incremental cost-effectiveness ratios
204
(ICER); interventions with ICERs of less than per capita GDP ($7160 for South Africa) per
205
disability adjusted life year (DALY) averted are considered highly cost effective and worth
206
introducing, but may require additional resources.
207
The economic costs of a comprehensive microbicide distribution programme were estimated
208
from the provider’s perspective using an ingredients-based approach and unit cost from the
209
South African public sector. This includes HIV testing [27], facility based provider training
210
[28] and health facility costs, including staff, for the initiation visit and subsequent adherence
211
counselling and gel distribution visits [29-31], 10% product wastage [32], and periodic mass
212
media campaigns [33]. This analysis assumes general population HIV testing is in place and
213
the initial screening costs are not included. The current price a single dose of tenofovir gel as
214
produced for the FACTS 001 trial is $1.60 [34]. The predicted price of packaged and labelled
215
gel at low production scale (10 million doses) is $0.56, while this could be as low as $0.17 if
216
distributed at large scale with a tube of gel and fillable paper applicators [34]. Cost savings
217
are the estimated cost of HIV treatment and care that would not be spent thanks to averting
218
HIV infections. For the 52% of those in need who are on ART [35], 2012 annual treatment
219
costs are used ($562 for first line ART and $1094 for second line) [36], with an lifetime cost
220
of $14,310 assuming a near normal life expectancy due to ART [37]. It is worth noting that
221
the current tender in South Africa is substantially lower than in previous years and is thus
222
lower than the costs used in similar papers [38-39]. For those without ART access,
Break-even microbicide prices are estimated, where the full programme costs are offset by
11
223
discounted lifetime HIV care costs of $3080 are included [31]. This gives a discounted
224
weighted average HIV care and ART costs averted of $8,920 per HIV infection averted.
225
Though HSV-2 treatment with acyclovir is first line therapy for genital ulcer disease, these
226
averted costs have not been included and will render the analysis on the conservative side.
227
228
Break-even prices are estimated under a range of assumptions (number of annual HIV tests
229
and clinic visits (two to six times per year for both) and dosing (one to two doses per sex-act),
230
and epidemiological scenarios). All costs are presented in 2012 US dollars ($). All future
231
costs and effects are discounted to present values, using a 3% discount rate. A sensitivity
232
analysis is undertaken to understand the possible impact on the break-even price of: the prices
233
of ART (-25%, -50%) and other inputs (+/- 25%), discount rate (0%, 6%3), and treatment
234
uptake rates (80%). Further, an economic evaluation in terms of cost per DALY averted is
235
estimated at the projected product prices. The full details of the economic evaluation
236
methods are presented in the supplemental materials.
237
238
239
Results
240
241
Imputed efficacy and HIV incidence projections
242
probability of HIV acquisition by 54% (2.5%-97.5% credibility interval (95%CrI) from
243
model fits 8-83%) and the per sex-act probability of HSV-2 acquisition by 71% (95% CrI 30-
244
97%). Using these inputs, the projected impact on HIV incidence is shown in Figure 2. Table
245
1 presents the projected impact of microbicide introduction under a number of scenarios. In
246
the main intervention scenario with mean efficacies (30% uptake achieved over 10 years, gel
The imputed efficacy calculations suggest that tenofovir gel reduced the per sex-act
12
247
used in 72% of sex-acts and no reduction in condom use), we predict that the gel could lead
248
to a 10.5% (95%CrI 8.9-10.6%) relative reduction in HIV incidence and a 5.7% (95%CrI
249
5.3%-6.0%) relative reduction in HSV-2 incidence by year 15. This is a reduction in HIV
250
incidence in the whole population from 3.0 (95%CrI 2.7-3.2) per 100 person-years to 2.7
251
(95%CrI 2.4-2.9) per 100 person years, and a reduction in HSV-2 incidence from 9.9
252
(95%CrI 8.6-10.2) per 100 person-years to 9.4 (95%CrI 8.1-9,7) per 100 person years.
253
254
[INSERT TABLE 1 HERE]
255
256
These reductions in incidence translate into averting 2,143 (95%CrI 1,801-2,194) HIV
257
infections and 1,799 (95%CrI 1,589-1,875) HSV-2 infections per 100,000 population (Table
258
1) over 15 years. In this main scenario, one HIV infection and one HSV-2 infection is averted
259
for every 4,785 (95%CrI 4,416-6,389) and 5,699 (95%CrI 4,948-7,243) microbicide
260
protected sex-acts, respectively.
261
262
The impact projections are dependent upon the assumptions made regarding gel consistency
263
and uptake, and the degree to which condom use may or may not be affected by gel
264
introduction. For example, if the gel is used in 50% rather than 72% of sex-acts, then the
265
number of HIV infections averted is reduced by 31%.
266
267
Figure 2 shows how the HIV impact is affected by a number of modelling assumptions; the
268
dark bars represent the main intervention scenario. Gel efficacy is very important, with a very
13
269
highly efficacious gel potentially reducing HIV incidence by 16.4%, while a gel with poor
270
efficacy provides little population protection. Higher levels of uptake can have an important
271
impact with a doubling of uptake resulting in a more than doubling in impact (107%
272
increase). There is also a large reduction in impact if condoms are used in 5 percentage
273
points fewer sex-acts but uptake and gel use is maintained (dropping from 10.5 RRI to 7.4
274
RRI for HIV), dropping further to a 4.3% RRI at a 10 percentage points reduction in condom
275
use.
276
277
[INSERT FIGURE 2 HERE]
278
279
Interestingly the model projections suggest the HSV-2 efficacy of the gel contributes little to
280
HIV-impact. Without any HSV-efficacy, the relative reduction in HIV incidence is projected
281
to be 10.0% after 15 years for the main intervention scenario compared to 10.5% if the gel is
282
71% HSV-2-efficacious. However, the HIV efficacy has an important influence on HSV-2
283
incidence: the relative reduction in HSV-2 incidence is projected as 5.7% with 54% HIV
284
efficacy versus 2.9% if the gel has no HIV efficacy.
285
286
287
Break-even microbicide prices
288
intervention scenario (30% uptake, 72% gel use, 54% HIV efficacy) with two gels per sex-
289
act, three HIV tests and collection visits per year (after the initiation visit), the break-even
290
product price is $0.41 per filled and packaged applicator (Table 2). Given that the cost of the
291
gel makes up 63% of total costs, moving from two doses to a single dose would greatly
Table 2 presents the break-even prices for a single microbicide dose. For the main
14
292
decrease costs, allowing for a doubling of the price of gel to $0.82 to maintain cost neutrality.
293
HIV counselling and testing account for 29% of costs: each extra HIV test would require a
294
reduction of 15% in price in order for the intervention to remain cost-neutral. The health
295
facility costs of visits for adherence counselling and gel collection are relatively low,
296
contributing only 7% to programme costs: changes in collection frequency provide little
297
savings or extra costs. However, a more integrated approach, with adherence counselling and
298
gel distribution during the HIV testing visit, may be more attractive to clients and is cost
299
saving, with a break-even gel price of up to $0.45.
300
301
[TABLE 2]
302
303
We also model the effect on the break-even gel price for variations to gel coverage, use
304
consistency, and HIV efficacy, linked with the different impact scenarios (Table 3). Condom
305
substitution of 5 percentage points would reduce the break-even microbicide price to just
306
$0.22, and twice the reduction in use (10 percentage points) would generate a break-even
307
price of just $0.04, but a more efficacious gel would increase the break-even price to $0.79.
308
Consistency of gel use (adherence) is also important, where only using gel in half the sex-acts
309
reduces effectiveness by more than the reduction in costs, with an estimated break-even price
310
of $0.31. This drop is attributable to maintaining the annual fixed costs per facility and
311
woman, while just reducing the gel use, and emphasises how important user behaviour will
312
be in such a products’ ultimate cost-effectiveness. Coverage, however, has a relatively small
313
impact on the break-even point: doubling coverage just increases the break-even price by 6%,
314
here both fixed cost of reaching women and their gel use is reduced.
15
315
316
[TABLE 3]
317
318
A sensitivity analysis of key cost assumptions is presented in the bottom half of Table 3. As
319
the cost of ART provision has decreased dramatically over the past decade, we also explore
320
the trade off between treatment and prevention for lower ART provision costs. If ART costs
321
were to drop by either 10% or 25%, the break-even microbicide price would also need to
322
decrease, by just over 13% and 33%, respectively. To account for variations in programme
323
costs we explored changes to all other input prices by +/- 25%, which would change the
324
threshold gel price by -/+ 23%, respectively. If ART coverage were expanded from 52% to
325
80%, in line with the 2016 national target [35], then the price of microbicides could be as
326
high as $0.63. In the most conservative case, where programme costs were increased by 25%
327
and ART costs decreased by 25%, the gel price would need to be $0.21 to be cost neutral.
328
Though cost-effectiveness analyses have traditionally assumed an international discount rate
329
of 3%, more recent analyses have started using local discount rates to account for local time
330
preferences which then better represent local decision making. The South African official
331
discount rate in 2012 was 5.5%. We estimate the impact of a 6% discount rate and show a
332
very large impact on the price threshold, reducing it from $0.41 per dose to just $0.10.
333
However, discounting is not without controversy. If no discounting were applied, the unit
334
product cost could be as high as $1.14.
335
However, cost neutrality is a very stringent condition. More traditionally, evaluation of value
336
for money would occur on the basis of the ICER; modelled ICERs are presented in online
337
Supplement Table 6.
16
338
339
340
Discussion and Conclusions
341
break-even price of tenofovir gel as introduced and distributed in urban South Africa. Our
342
findings suggest that, for the introduction scenario considered, gel introduction could be an
343
important addition to existing HIV prevention efforts, leading to a 10.5% (95%CrI 8.9-
344
10.6%) relative reduction in HIV incidence and a 5.7% (95%CrI 5.3-6.0%) relative reduction
345
in HSV-2 incidence over 15 years. The impact on HSV-2 is less than the impact on HIV,
346
despite the higher efficacy against HSV-2, due to higher risk of HSV-2-infection for women.
347
Indeed, using plausible assumptions about the full costs of programme delivery, our findings
348
suggest that the introduction of an efficacious gel that cost $0.41 or less would be cost-
349
saving. This is extremely promising, as this is above the expected cost range ($0.17-$0.27) at
350
large scale production of 100 million applicators [34]. In our model this scale of distribution
351
is achieved from year 5 onwards. Even at current FACTS 001 trial production prices ($1.60)
352
[34], the incremental cost per DALY averted would be less than $800 (see Supplement Table
353
6), which is well below the commonly applied cost-effectiveness threshold of 1*GDP ($7,610
354
for South Africa)[41].
We have used detailed mathematical and economic modelling to estimate the impact and
355
356
Using the trial estimates of 72% adherence, we impute that the HIV and HSV-2 per sex-act
357
efficacies of tenofovir gel are 54% (95% CrI 8-83%) and 71% (95%CrI 30-97%),
358
respectively. Although these estimates are based upon the trial measures of gel adherence
359
reporting, we considered that these are reasonably reliable, as gel applicator returns were
360
found to correlate well with vaginal tenofovir concentration measured in a pharmacokinetics
361
sub-study [42]. We are also reassured about the validity of our imputed efficacy estimates
362
because a secondary analyses of the effectiveness among ‘high adherers’ in the trial (i.e.
17
363
women who used both doses of the gel in over 80% of sex-acts) found that the risk of HIV
364
acquisition in this population was reduced by 54% (95%CI 4-80%)[5]. However, the wide
365
bounds on the imputed efficacy estimates of tenofovir gel, due to the wide confidence
366
intervals on the trial effectiveness estimates, lead to a wide range of population-level impact
367
estimates predicted by the model.
368
369
This analysis builds upon our previous modelling of the potential impact of microbicide
370
introduction in different settings [43-47], as well as the work by Williams et al. [49], Verguet
371
et al. [39] and Walensky et al. [38-39, 48]. Additional strengths include that the model is
372
carefully parameterised and rigorously fitted to a specific setting over multiple time points,
373
including fitting to HIV and STI infection status by sex, as well as incorporating the HSV-2-
374
efficacy of tenofovir found in CAPRISA 004 and the related bi-directional interaction
375
between HSV-2 and HIV and the dynamic effects of gel on the HIV and HSV-2 epidemic.
376
Walesky et al. [38] included only the first generation of HIV infections averted. Numerous
377
programme implementation components were also included in our analysis, such as training,
378
mass media and facility distribution costs, to provide a more comprehensive approach to
379
costs than comparable studies to date. However, we did not account for potential toxicity or
380
ART resistance as in Walensky et al. [38], because the concerns in this area are limited, as the
381
authors themselves acknowledge. Moreover, the scenarios that we have modelled are less
382
optimistic than those considered previously [48 ], with lower assumptions regarding the
383
likely coverage that can be achieved, more optimistic assumptions about the impact of ART
384
on life expectancy and the discounting of future DALYs averted, an omission in Verguet et
385
al.’s analysis [39], which in turn generates a far more conservative estimate of impact and
386
cost-effectiveness. Nevertheless, our findings still suggest that the introduction of an effective
18
387
gel could be cost savings at reasonable gel prices, in particular as ART access is expanded,
388
and highly cost-effective even at current gel prices.
389
390
Our modelling stresses the importance of maintaining condom use following microbicide
391
introduction, to avoid large reductions in impact and cost-effectiveness. This issue is
392
particularly of concern if condom use is already high [53], and for a microbicide with lower
393
efficacy, as concomitant reductions in condom use could even lead to increases in the overall
394
number of infections [47]. Similarly, our analyses highlight the degree to which the costs of
395
gel introduction will be affected by programme costs, including the unit cost of gels, and the
396
required frequency of HIV testing. For this reason, further research is needed to identify the
397
optimal spacing between HIV testing, with the aim of achieving an appropriate balance
398
between what may be required to minimise the potential for resistance and be feasible for
399
women and health systems. Similarly, research to explore whether a single dose of tenofovir-
400
containing gel, applied prior to sex, could be effective at reducing a woman’s risk of
401
acquiring HIV is needed [54], to help reduce the product costs, as well as make gel use less
402
burdensome for women.
403
404
The fact that 17% of VOICE participants reported engaging in anal intercourse during the 3-
405
month period prior enrolment highlights an important area for future modelling work [55].
406
For such modelling, data is needed from different sites and population groups on the
407
frequency of anal sex over a specific timeframe (e.g. past 3 or 6 months), and whether a
408
condom and/or gel is used during the last anal sex-act.
409
19
410
411
Although it is tempting to draw broader, national or regional conclusions from this analysis,
412
our model projections may not be generalisable to other settings with different epidemic and
413
behavioural profiles and different price levels. However, the impact we predict compares
414
favourably with previous modelling of a hypothetical microbicide introduced into an area
415
within Gauteng [19], and with modelling studies of the impact of male circumcision on HIV
416
incidence in various settings of sub-Saharan Africa [49-51], matched to the same intervention
417
assumptions made here [52].
418
419
The results are also highly sensitive to the discount rate used. Our central estimate used a 3%
420
discount rate as commonly applied in international studies. However, when using the upper
421
bound of 6%, just above the South African national discount rate of 5.5%, the price threshold
422
drops to just $0.10. Though not a realistic price to expect, the programme’s incremental cost-
423
effectiveness, even at the current product cost of $1.60 per dose would still be highly cost
424
effective at $1,248 per DALY averted (using a 6% discount rate).
425
426
Some have questioned the viability of tenofovir gel given recent disappointing evidence of
427
low user adhernece in the VOICE trial. That users have a difficult time with adherence,
428
however, should come as no surprise. Inconsistent adherence is the norm for both
429
medications and any prevention methods that are user controlled. Nonetheless, every
430
indication is that the South African government and international donors are prepared to seek
431
licensure for tenofovir gel should the results of the CAPRISA trial be sustained in the on-
432
going FACTS 001 trial. Given that policymakers will soon face a key decision point
20
433
regarding the possible introduction and scale up of tenofovir gel, the findings of this article
434
are highly relevant. Despite intuitions to the contrary, it appears from the results of VOICE,
435
FEMPREP, and other trials of daily oral or vaginal prophylaxis, that women seem better able
436
to adhere to the coitally dependent BAT-24 regimen than daily use.
437
438
Moreover, even if the attention of the wider field has shifted toward long-acting modalities
439
such as the vaginal ring and injection, these products are still many years from licensure and
440
there is undoubtedly benefit of beginning to prepare consumers for using ARVs to prevent
441
infection. There will always be a market for methods like the vaginal gel that will likely not
442
require on-going HIV testing and that can be positioned as a sexual lubricant with additional
443
HIV fighting powers.
444
21
445
Abbreviations
ART
Antiretroviral therapy
CI
Confidence interval for data estimates
Crl
Credibility interval for model projections –
defined as the 2.5 to 97.5% percentile range
DALY
Disability adjusted life year
DfID
Department for International Development
FSW
Female sex workers
HIV
Human immunodeficiency virus
HSV-2
Herpes simplex type 2
ICER
Incremental cost effectiveness ratio
IPM
International Partnership for Microbicides
MDP
Microbicide Development Programme
446
447
22
448
449
Competing interests
450
received a grant from IPM in the initial development of these models. At the time, TM was
451
employed by IPM. IPM aims to promote the development and delivery of safe and effective
452
microbicides around the globe. LH was previously the director for the Global Campaign for
453
Microbicides.
I have read the journal’s policy and have the following conflicts: AF, FTP, AC, PV, CW
454
455
456
Author Contributions
457
AC; Undertook the impact modelling: AC and PV; Undertook the cost modelling: FTP;
458
Wrote the first draft of the manuscript: AF, FTP; Contributed to the writing of the
459
manuscript: LH, AC, GMR, SDM, TM, HR, PV, CW; Criteria for authorship read and met:
460
AF, FTP, AC, LH, GMR, SDM, TM, HR, PV, CW. Agree with manuscript results and
461
conclusions AF, FTP, AC, LH, GMR, SDM, TM, HR, PV, CW.
Conceptualised the study: AF, PV, FTP, TM, AC, LH,CW; Developed the model: PV and
462
463
464
Acknowledgements
465
Walker, Youssef Tawfik and Vimala Raghavendran. The authors are extremely grateful to
466
Christine Michaels, Chiweni Chimbwete, Emeka Okonji, Richard Hayes, Mitzy Gafos, Lilani
467
Kumaranayake, Jocelyn Moyes, Mags Beksinska, Sibongile Walaza, Claire von Mollendorf,
468
Jennifer Smit, Anna Vassall and Michelle Remme for their support and input into this work,
469
and to Salim S. Abdool Karim and Brian Williams for sharing trial data on condom and gel
470
use and. The authors are also indebted to all the participants of the various microbicide trials.
The authors are grateful for the ongoing support, advice and input from Céline Mias, Saul
471
23
472
473
Funding
474
Microbicides (IPM), funded by the European Union, and contributes to the Microbicides
475
Development Programme (MDP). MDP is a partnership of African and European
476
academic/government institutions and commercial organisations. MDP is funded by the
477
British Government Department for International Development (DfID) and the UK Medical
478
Research Council. AF, FTP, AC, PV, GMR and CW were also members of the DfID
479
Research Programme Consortium for Research and Capacity Building in Sexual and
480
Reproductive Health and HIV in Developing Countries of the LSHTM.
This research project was conducted in conjunction with the International Partnership for
481
24
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
References
1.
Cohen M, Chen Y, McCauley M, Gamble T, Hosseinipour M, Kumarasamy N, al. e: Prevention
of HIV-1 infection with early antiretroviral therapy. N Engl J Med 2011, 365:493-505.
2.
Grant R, Lama J, Anderson P, McMahan V, Liu A, Vargas L, al. e: Preexposure
chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 2010,
363:2587-2599.
3.
Baeten J, Donnell D, Ndase P, al. E: Antiretroviral prophylaxis for HIV prevention in
heterosexual men and women. N Engl J Med 2012.
4.
Thigpen MC, Kebaabetswe PM, Paxton LA, Smith DK, Rose CE, Segolodi TM, Henderson FL,
Pathak SR, Soud FA, Chillag KL, et al: Antiretroviral Preexposure Prophylaxis for
Heterosexual HIV Transmission in Botswana. N Engl J Med 2012.
5.
Karim QA, Karim SS, Frohlich JA, Grobler AC, Baxter C, Mansoor LE, Kharsany AB, Sibeko S,
Mlisana KP, Omar Z, et al: Effectiveness and Safety of Tenofovir Gel, an Antiretroviral
Microbicide, for the Prevention of HIV Infection in Women. Science 2010,
Epub:http://www.sciencemag.org/cgi/content/abstract/science.1193748v1193741.
6.
Karim SSA: Results of the CAPRISA 004 trial of tenofovir gel. In XVIII International AIDS
Conference 18-23 July; Vienna, Austria. 2010
7.
Cates W, Jr.: After CAPRISA 004: time to re-evaluate the HIV lexicon. Lancet 2010, 376:495496.
8.
FACTS 001 Study [http://www.facts-consortium.co.za/?page_id=83]
9.
Marrazzo J, Ramjee G, Nair G, Palanee T, Mkhize B, Nakabiito C, Taljaard M, Piper J, GomezFeliciano K, Chirenje M, VOICE Study Team: Pre-exposure Prophylaxis for HIV in Women:
Daily Oral Tenofovir, Oral Tenofovir/Emtricitabine, or Vaginal Tenofovir Gel in the VOICE
Study (MTN 003). In 20th Conference on Retroviruses and Opportunistic Infections; March 36; Atlanta, Georgia, USA. 2013
10.
Shisana O, Rehle T, Simbayi L, Zuma K, Jooste S, Pillay-van-Wyk V, Mbelle N, Van Zyl J, Parker
W, Zungu N, et al: South African national HIV prevalence, incidence, behaviour and
communication survey. A turning tide among teenagers? . Cape Town: HSRC Press; 2009
11.
Shisana O, Rehle T, Simbayi L, Parker W, Zuma K, Bhana A, Connolly C, Jooste S, Pillay V:
South African National HIV Prevalence, HIV Incidence, Behaviour and Communication
Survey.; 2005
12.
Dunkle K, Beksinka M, Rees H, Ballard R, Htun Y, Wilson ML: Risk factors for HIV infection
among sex workers in Johannesburg, South Africa. Int J STD AIDS 2005, 16:256 - 261.
13.
RHRU: Reproductive Health Research Unit Behavioural Sentinel Survey data report Clients of family planning - Gauteng.; 2000
14.
Watts C, Foss A, Kumaranayake L, Cox A, Terris-Prestholt F, Vickerman P: Identifying optimal
strategies for microbicide distribution in India and South Africa: Modelling and costeffectiveness analyses. A policy paper prepared by the International Partnership for
Microbicides with the London School of Hygiene and Tropical Medicine; 2008
15.
Rees H: Service delivery and demand challenges in introducing new health products. In
Book Service delivery and demand challenges in introducing new health products. City; 2008.
16.
Department of Health, Medical Research Council, OrcMacro: South Africa demographic and
health survey 2003. Pretoria: Department of Health Republic of South Africa; 2007
17.
Heise L, Watts C, Foss A, Vickerman P, Trussell J, Hayes R, McCormack S: Apples and
oranges? Interpreting success in HIV prevention trials. Contraception 2011, 83:10-15.
18.
Vickerman P, Foss A, Watts C: Using modeling to explore the degree to which a
microbicide's sexually transmitted infection efficacy may contribute to the HIV
effectiveness measured in phase 3 microbicide trials. J Acquir Immune Defic Syndr 2008,
48:460-467.
25
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
26
Vickerman P, Watts C, Delany S, Alary M, Rees H, Heise L: The importance of context: model
projections on how microbicide impact could be affected by the underlying epidemiologic
and behavioral situation in 2 African settings. Sex Transm Dis 2006, 33:397-405.
Vickerman P, Foss AM, Pickles M, Deering K, Verma S, Demers E, Moses S, Alary M, Boily MC: Is the Indian HIV epidemic driven by commercial sex? A modelling analysis from south
India. AIDS 2010, 24:2563-2572.
Garnett GP, Anderson RM: Strategies for limiting the spread of HIV in developing countries:
conclusions based on studies of the transmission dynamics of the virus. J Acquir Immune
Defic Syndr Hum Retrovirol 1995, 9:500-513.
Weinstein MC, Graham JD, Siegel JE, Fineberg HV: Cost-effectiveness analysis of AIDS
prevention programs: concepts, complications and illustrations. In Confronting AIDS. Edited
by Turner CF, Miller HG, Moses LE. Washington DC: National Academy Press; 1989: 471-499
Foss AM, Vickerman PT, Mayaud P, Weiss H, Ramesh BM, Reza-Paul S, Washington R,
Blanchard J, Moses S, Lowndes CM, et al: Modelling the interactions between HSV-2 and
HIV: implications for the HIV epidemic in southern India. Sex Transm Infect 2011, 87:22-27.
Wawer MJ, Gray RH, Sewankambo NK, Serwadda D, Li X, Laeyendecker O, Kiwanuka N,
Kigozi G, Kiddugavu M, Lutalo T, et al: Rates of HIV-1 transmission per coital act, by stage of
HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005, 191:1403-1409.
Shisana O, Simbayi L: Nelson Mandela/HSRC study of HIV/AIDS. South African National HIV
prevalence, behavioural risks and mass media. Cape Town: The human sciences research
council; 2002
Johnson LF, Coetzee DJ, Dorrington RE: Sentinel surveillance of sexually transmitted
infections in South Africa: a review. Sex Transm Infect 2005, 81:287-293.
Bassett IV, Giddy J, Nkera J, Wang B, Losina E, Lu Z, Freedberg KA, Walensky RP: Routine
voluntary HIV testing in Durban, South Africa: the experience from an outpatient
department. J Acquir Immune Defic Syndr 2007, 46:181-186.
Terris-Prestholt F, Kumaranayake L, Obasi AI, Cleophas-Mazige B, Makokha M, Todd J, Ross
DA, Hayes RJ: From Trial Intervention to Scale-Up: Costs of An Adolescent Sexual Health
Program in Mwanza, Tanzania. Sex Transm Dis 2006.
Terris-Prestholt F, Michaels C, Kumaranayake L, Meyer-Rath G, Watts C: Costing Microbicide
Distribution: Estimating cost-effectiveness & modelling potential role out. In Book Costing
Microbicide Distribution: Estimating cost-effectiveness & modelling potential role out. City;
2009.
Meyer-Rath G: Public sector salary costs. In Book Public sector salary costs. City; 2009.
Cleary S, Bouille A, McIntyre D, al. e: Cost-effectiveness of antiretroviral treatment for HIVpositive adults in a South African township. Cape Town: University of Cape Town, Health
Systems Trust, Medecins Sans Frontieres; 2004
Gilson L, Mkanje R, Grosskurth H, Mosha F, Picard J, Gavyole A, Todd J, Mayaud P, Swai R,
Fransen L, et al: Cost-effectiveness of improved treatment services for sexually transmitted
diseases in preventing HIV-1 infection in Mwanza Region, Tanzania. Lancet 1997, 350:18051809.
Bollinger L, Stover, J., Boulle, A., Cleary, S. : Database of Unit Costs for HIV and AIDS
interventions in South Africa: Resource Needs for HIV/AIDS: Model for estimating
resources needed for prevention, care and mitigation. In Book Database of Unit Costs for
HIV and AIDS interventions in South Africa: Resource Needs for HIV/AIDS: Model for
estimating resources needed for prevention, care and mitigation. City: Centre for Economic
Governance and AIDS in Africa.; 2006.
Friend D: Tenofovir 1% gel, current prices and expected prices at different scales. In Book
Tenofovir 1% gel, current prices and expected prices at different scales. City; 2012.
Johnson L: Access to antiretroviral treatment in South Africa, 2004-2011. South African
Journal of HIV Medicine 2012, march:22-27.
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
27
Meyer-Rath G: National ART Cost Model, South Africa. Johannesburg: Health Economics
and Epidemiology Research Office, Boston University/ University of the Witwatersrand; 2011
Johansson KA, Robberstad B, Norheim OF: Further benefits by early start of HIV treatment
in low income countries: Survival estimates of early versus deferred antiretroviral therapy.
AIDS research and therapy 2010, 7.
Walensky RP, Park JE, Wood R, Freedberg KA, Scott CA, Bekker LG, Losina E, Mayer KH,
Seage GR, 3rd, Paltiel AD: The cost-effectiveness of pre-exposure prophylaxis for HIV
infection in South African women. Clin Infect Dis 2012, 54:1504-1513.
Verguet S, Walsh JA: Vaginal microbicides save money: a model of cost-effectiveness in
South Africa and the USA. Sex Transm Infect 2010, 86:212-216.
Central Bank Rates [www.cbrates.com]
GDP per capits (current US$) [http://data.worldbank.org/indicator/NY.GDP.PCAP.CD]
Kashuba AD, Abdool Karim SS, Kraft E, White N, Sibeko S, Werner L, Mansoor LE, Gengiah T,
Sidhoo S, Abdool Karim Q: Do systemic and genital tract tenofovir concentrations predict
HIV seroconversion in the CAPRISA 004 tenofovir gel trial? In XVIII International AIDS
Conference 18-23 July; Vienna, Austria. 2010
Watts CH, Kumaranayake L, Vickerman P, Terris-Prestholt F: The public health impact of
microbicides: model projections. Report from the Public Health Benefits Working Group.
The Rockefeller Foundation Microbicide Initiative
(http://www.rockfound.org/Documents/488/rep1_summary.pdf ); 2002
Watts C, Vickerman P: The impact of microbicides on HIV and STD transmission: model
projections. AIDS 2000, 15:S43-S44.
Vickerman P, Devine A, Foss AM, Delany-Moretlwe S, Mayaud P, Meyer-Rath G: The costeffectiveness of herpes simplex virus-2 suppressive therapy with daily aciclovir for
delaying HIV disease progression among HIV-1-infected women in South Africa. Sex
Transm Dis 2011, 38:401-409.
Terris-Prestholt F, Kumaranayake L, MacPhail C, Rees H, Watts C: Determinants of South
African Women’s Demand for New Barrier Methods and their Distribution: Analysis of a
Discrete Choice Experiment. In IAEN; 1-2 August; Cuernavaca. 2008
Foss AM, Vickerman PT, Heise L, Watts CH: Shifts in condom use following microbicide
introduction: should we be concerned? AIDS 2003, 17:1227-1237.
Williams BG, Karim SSA, Karim QA, Gouws E: Epidemiological impact of tenofovir gel on the
HIV epidemic in South Africa. J Acquir Immune Defic Syndr 2011, 58:207-210.
Gray RH, Li X, Kigozi G, Serwadda D, Nalugoda F, Watya S, Reynolds SJ, Wawer M: The
impact of male circumcision on HIV incidence and cost per infection prevented: a
stochastic simulation model from Rakai, Uganda. AIDS 2007, 21:845-850.
Hallett TB, Singh K, Smith JA, White RG, Abu-Raddad LJ, Garnett GP: Understanding the
impact of male circumcision interventions on the spread of HIV in southern Africa. PLoS
ONE 2008, 3:e2212.
White RG, Glynn JR, Orroth KK, Freeman EE, Bakker R, Weiss HA, Kumaranayake L, Habbema
JD, Buve A, Hayes RJ: Male circumcision for HIV prevention in sub-Saharan Africa: who,
what and when? AIDS 2008, 22:1841-1850.
Cox AP, Foss AM, Vickerman PT, Chimbwete C, Okonji E, Terris-Prestholt F, Beksinska M,
Walaza S, von Mollendorf C, Smit J, et al: Making realistic impact projections for new HIV
prevention technologies: looking beyond a product’s efficacy. In 18th International Society
for Sexually Transmitted Diseases Research; 28 June - 1 July; London, UK 2009
Foss A, Vickerman P, Heise L, Watts CH: Shifts in condom use following microbicide
introduction: should we be concerned? AIDS 2003, 17:1227-1237.
MDP Statement on the discontinuation of tenofovir gel in the VOICE trial
[http://www.mdp.mrc.ac.uk/]
632
633
634
635
636
637
638
55.
56.
28
Marrazzo J, Ramjee G, Nair G, al e: Pre-exposure prophylaxis for HIV in women: daily oral
tenofovir, oral tenofovir/emtricitabine or vaginal tenofovir gel in the VOICE study (MTN
003). In Book Pre-exposure prophylaxis for HIV in women: daily oral tenofovir, oral
tenofovir/emtricitabine or vaginal tenofovir gel in the VOICE study (MTN 003). City; 2013.
CAPRISA Statement on the interim tenofovir gel results in the VOICE trial
[http://www.caprisa.org/SitePages/Home.aspx]
639
Figure Legends
640
Figure 1: Model-projected HIV prevalence and incidence trends over time and data
641
used to fit model*
642
Figure 1a: HIV prevalence trends among general population males and females in 2002 and
643
2005
644
Figure 1b: HIV prevalence trends among the general population and among female sex
645
workers in 2008
646
Figure 1c: HIV incidence trends among general population males and females
647
648
* The points with error bars for males and females in 2002 and 2005 represent the mean and
649
95% confidence intervals derived from data [11, 25]. The dotted line represents the model-
650
projected female HIV prevalence while the solid line represents the model-projected male
651
HIV prevalence. The shaded areas represent the range of prevalence estimates spanned by all
652
model fits. In addition, the model was fitted to the overall HIV prevalence data for 2008 (see
653
online supplementary material) [10].
654
655
Figure 2: Projected population-level HIV impact in Gauteng for different assumptions
656
about gel efficacy, uptake and use, and its influence on condom use**
657
**Main shaded bars show best-fit model projections and error bars indicate range spanned by
658
95% of all model fits (2.5% to 97.5% percentile range or 95% credibility interval), with the
659
dark bars illustrating the main scenario (54% HIV efficacy and 71% HSV-2 efficacy, gel
660
uptake reaching 30% by year 10, gel used in 72% of sex-acts, and no reduction in condom
661
use).
662
1
Adherence estimates based on applicator returns for the remaining 884 women indicate that, on average,
72.2% (median = 60.2%) of self-reported sex-acts in the last 30 days were covered by two doses of gel.
2
It should be noted that condom use outside the trial setting is lower and our approach will therefore
underestimate the impact of microbicides on HIV incidence.
3
The 6% discount rate is consistent with the South African central bank discount rate of 5.5% 40. Central
Bank Rates [www.cbrates.com].
29
663
Tables
664
Table 1: Projected impact from microbicide intervention.*
Mean efficacies
Low efficacies
High efficacies
(54% HIV efficacy and
(8% HIV efficacy and
(83% HIV efficacy and
71% HSV-2 efficacy)
30% HSV-2 efficacy)
97% HSV-2 efficacy)
Trial consistency of
Reduced
Trial consistency
Reduced
Trial consistency of
Reduced
gel use (72%)
consistency of of gel use (72%) consistency of gel
gel use (72%)
consistency of gel
gel use (used in
use (used in 50%
use (used in 50% of
50% of sex-acts)
of sex-acts)
sex-acts)
HIV impact projections*:
Percentage reduction in population HIV
10.5%
7.2%
1.6%
1.1%
16.4%
11.2%
incidence by year 15
(8.9-10.6%)
(6.1-7.3
(1.4-1.7%)
(0.9-1.2%)
(14.0-16.5%)
(9.5-11.3%)
Cumulative number of HIV infections
122,352
84,103
18,982
13,152
190,463
130,255
averted over 15 years
(102,840-125,293) (70,599-86,164) (15,974-19,428) (11,060-13,459) (160,180-195,145) (109,394-133,405 )
Cumulative number of HIV infections
2,143
1,473
332
230
3,336
2,281
averted per 100,000 population
(1,801-2,194)
(1,236-1,509)
(280-340)
(194-236)
(2,805-3,418)
(1,916-2,336)
Number of sex-acts protected by gel per
4,785
4,838
30,843
30,915
3,074
3,121
HIV infection averted
(4,416-6,389)
(4,466-6,460) (28,324-40,813) (28,397-40,917)
(2,832-4,101)
(2,882-4,170)
HSV-2 impact projections*:
Percentage reduction in population HSV5.7%
3.9%
1.6%
1.1%
8.4%
5.7%
2 incidence in year 15
(5.3-6.0%)
(3.6-4.1%)
(1.5-1.7%)
(1.0-1.2%)
(7.9-9.0%)
(5.3-6.1%)
Cumulative number of HSV-2 infections
102,740
70.088
29,262
20,169
152,112
102,968
averted over 15 years
(90,741-107,061) (61,892-73,026) (26,491-30,976) (18,772-21,347) (133,638-158,419) (90,436-107-203)
Cumulative number of HSV-2 infections
1,799
1,227
512
353
2.664
1,803
averted per 100,000 population
(1,589-1,875)
(1,084-1,279)
(464-542)
(320-374)
(2,340-2,774)
(1,584-1,877)
Number of sex-acts protected by gel per
5,699
5,801
20,008
20,159
3,849
3,949
HSV-2 infection averted
(4,948-7,243)
(5,036-7,369)
(17,304-25,310)
(3,346-4,896)
(3,432-5,021)
* Assuming gel only used by HIV-negative women, and levels of condom use are maintained after microbicide introduction.
665
666
Table 2: Break-even gel prices under variations to distribution service assumptions [2012 USD].
Baseline programme: 3 annual HIV tests, 3 gel
collection visits
Break-even
prices of gel
Break-even
prices of gel
2 gels per sex-act
1 gel per sex-act
0.41
0.81
Variations to distribution services
HIV tests per year
Adherence counselling and gel collection visits
2
0.47
15%
0.28
-65%
4
0.35
-15%
0.69
-15%
6
0.22
-45%
0.45
-45%
2
0.42
4%
0.84
4%
6
0.36
-11%
0.73
-11%
667
Integrated adherence counselling and gel collection
into HIV testing (3)
0.45
11%
0.90
11%
*Assuming 30% uptake after 10 years and trial consistency of gel use and mean imputed efficacy estimates, 10% product wastage during distribution, and a
668
3% discount rate on costs and effects
669
670
671
672
Table 3: Break-even gel prices under variations to introduction scenario assumptions and sensitivity analysis [2012 USD].
Reduction in condom use
Variations to introduction scenario
Baseline
0%
5%
10%
0%
0%
0%
Cost sensitivity analysis
ART costs 10% lower
ART costs 25% lower
All other input costs 25% higher
All other input costs 25% lower
input +25%, ART-10%
input +25%, ART-25%
ART coverage 80%
Discount rate 0%
Discount rate 6%
673
Coverage Consistency of gel use
30%
30%
30%
60%
60%
30%
72%
72%
72%
72%
50%
72%
HIV efficacy
54%
54%
54%
54%
54%
83%
Break-even price
$
%change
0.41
0.22
0.04
0.43
0.31
0.79
0.35
0.27
0.35
0.32
0.29
0.21
0.63
1.14
0.10
-45%
-90%
6%
-23%
95%
-13%
-33%
-15%
-20%
-28%
-48%
56%
180%
-76%