Epidemiological data and baseline parameter values for modelled Bagalkot and Mysore scenarios and range for sensitivity analysis
| Definition of available data for model variables and model values used | Bagalkot | Mysore | Sources (refs) | |||
|---|---|---|---|---|---|---|
| Males | Females | Males | Females | |||
| *Karnataka Health Promotion Trust; †Sex work Demonstration Project-Bagalkot report and plan 2005–6;. ‡Note that single cofactor multiplies the risk of acquisition of HIV among STI positive individuals, Multiple cofactors are additive in the model for individual infected with two STIs. ¶An additional figure is in the STI website (www.stijournal.com/supplemental). | ||||||
| NA, not applicable; N–Sp, not specified*; NAv, not available; F, female; M, male. | ||||||
| Baseline scenario (baseline parameters) | ||||||
| (A) Epidemiological data | ||||||
| Initial HIV prevalence | FSW/clients: baseline model fit | 9.8% | 35.0% | 6.8% | 25.8% | Model fit |
| Data | NAv | NAv | NAv | 25.8% (21.7% to 30.3%) | 4, 8–13 | |
| General population (gen pop) baseline model fit | 2.7% | 2.2% | 0.9% | 0.7% | Model fit | |
| Data | 3.6% (95% CI 2.6% to 5.0%) | 2.9% (95% CI 1.9% to 4.0%) | NAv | 1.2% (0.6% to 2.1%) | 8, 11, 12 Mysore: KHPT* ANC survey | |
| Average duration of CT/GC (months) | FSW | NA | 1 | NA | 1.2 | 11, 12 |
| Gen pop | 1.5 | 5.2 | 1.7 | 5.2 | 11, 12 | |
| Average duration of HSV-2 (months) | FSW and gen pop | 480 | 480 | 480 | 480 | 45 |
| Year when HIV entered population (assumes 8 years between infection to AIDS) | 1977 | 1977 | 1977 | 1977 | 1 | |
| (B) Biological parameters | ||||||
| Average duration high viraemia phase (months) | 2 | 2 | 2 | 2 | 14 | |
| Mean time between end of high viraemia phase and morbidity (months) | 95 | 95 | 95 | 95 | 15–18 | |
| Fraction of males who are circumcised† | 15% | NA | 11.5% | NA | 19, 20 | |
| Probability of HIV transmission/sex act | 0.00087 | 0.0016 | 0.00087 | 0.0016 | 14, 21 | |
| Female to male (F to M), male to female (M to F) | F to M | M to F | F to M | M to F | ||
| Reduction in F to M HIV transmission/sex act because of male circumcision | 65% | 65% | 65% | 65% | 22 | |
| Probability of GC/CT transmission/sex act | 0.15 | 0.15 | 0.15 | 0.15 | 18, 23–31 | |
| Reduction in susceptibility to infection with GC/CT of sex workers | NA | 30% | NA | 30% | 32, 33 | |
| Probability of HSV-2 transmission/sex act | 0.0023 | 0.0063 | 0.0023 | 0.0023 | 29–31 | |
| F to M | M to F | F to M | M to F | |||
| Multiplicative HIV cofactor effect per sex act because of STI-1 (CT/GC)‡ | 2.0 | 2 | 2 | 2 | 18, 34–36 | |
| Multiplicative HIV cofactor effect per sex act because of STI-2 (HSV-2)‡ | 3.3 | 3.3 | 3.3 | 3.3 | ||
| Sexual transmission multiplicative factor during high viraemia phase | 8.7 | 8.7 | 8.7 | 8.7 | 14 | |
| Condom efficacy per sex act | 87% | 87% | 87% | 87% | 37 | |
| (C) Demographic parameters | ||||||
| Initial size of sexually active population | 473 833 | 456 450 | 737 014 | 690 828 | 11, 12 | |
| Sexually active life expectancy in years | 37 | 43 | 38 | 42 | 11, 12, 38, 39 | |
| (D) Baseline distribution in sexual activity | ||||||
| Fraction of the population per level of sexual activity (with none, low, or high numbers of casual/non-FSW partners or FSW) | None | 24% | 20% | 1% | 1% | 11, 12, 40 |
| Low | 40% | 64% | 79% | 78% | 11, 12, 40 | |
| High | 16% | 14.8% | 8% | 11% | 11, 12, 40 | |
| CL/FSW | 20.0% | 1.2% | 12.0% | 1.0% | 2, 11, 12, 41 | |
| Number long term sexual partners/year for “low” and “high” sexually active males, females and sex workers | Low | 1 | 1 | 1 | 1 | Assumed |
| High | 1.00 | 1.00 | 1.25 | 1.25 | Assumed | |
| FSW | NA | 1.00 | NA | 1.25 | 11, 12 | |
| Average number sex acts/month for all long term partnerships | Low/High/FSW | 9 | 9 | 9 | 9 | 11, 12, 42 |
| Number of casual sexual partners/month for “high” sexually active males and females and FSW (averaged over all “high” people or all FSW) | High | 0.6 | 0.6 | 0.5 | 0.5 | 11, 12 |
| FSW | NA | 0.6 | NA | 0.5 | 11, 12 | |
| Average number sex acts/month: casual partnerships with “high” class or with FSW (among those having such a partner) | High | 6 | 6 | 5 | 5 | 11, 12 |
| FSW | NA | 6 | NA | 5 | 11, 12 | |
| Sex worker sexual behaviour: | ||||||
| Average number of clients per sex worker/month | NA | 27 | NA | 39 | 11, 12 | |
| Average number of sex acts between a client and a sex worker in one meeting | 1 | 1 | 1 | 1 | Assumed | |
| Average duration of sex work in months | NA | 180 | NA | 240 | 11, 12 | |
| Mixing parameter for long term partnerships (A) (1 = completely assortative, 0 = proportional mixing) | 0 | 0 | 0 | 0 | Assumed | |
| (E) Baseline STI treatment parameters before the intervention or among those not reached by the intervention: | ||||||
| STI clinic for FSW: | ||||||
| Proportion of FSW STI cases with symptoms | 40% | 40% | 40% | 40% | 6, 43–45 | |
| Proportion of FSW with symptoms that attend clinic/mth (fclinic) | NA | 70% | NA | 70% | 2 | |
| Proportion FSW that return for test results | 100% | 100% | 100% | 100% | Assumed | |
| Delay until obtaining test result | GC/CT | 0 | 0 | 0 | 0 | Assumed |
| PPT treatment: | ||||||
| Mean time between PPT doses from FSW STI intervention (month) (FreqPPT) | NA | NA | NA | NA | Assumed no PPT at baseline | |
| (F) Baseline condom parameter before the intervention or among those not reached by the intervention: | ||||||
| Proportion using condoms consistently in long term partnerships for individuals not reached by intervention or before intervention | 1% | 1% | 0.5% | 0.5% | 11, 12 | |
| Definition of consistency of condom use (ie, fraction of sex acts when condoms are used) for none, sometimes, and always categories | None (Dnone) | 0 | 0 | 0 | 0 | 11, 12 |
| Sometimes (Dsome) | 0.3 | 0.3 | 0.4 | 0.4 | 11, 12 | |
| Always (DAll) | 0.9 | 0.9 | 0.8 | 0.8 | 11, 12 | |
| Distribution of condom use in casual partnerships for individuals not reached by intervention | None | 85% | 90% | 63% | 36% | 20, 46 |
| Sometimes | 10% | 5% | 25% | 42% | SWDP-Bagalkot† | |
| Always | 5% | 5% | 12% | 22% | ||
| Distribution of condom use in FSW-client partnerships for individuals not reached by intervention | None (fcnone) | 21% | 21% | 5% | 5% | 11, 12 |
| Sometimes (fcsome) | 19% | 19% | 17% | 17% | 11, 12 | |
| Always (fcall) | 60% | 60% | 78% | 11, 12 | ||
| (G) Following the introduction of the intervention at year 2005: | Avahan targets | |||||
| Invariant parameters: | ||||||
| • Proportion of FSW STI cases with symptoms | ||||||
| • Proportion of FSW returning for test results. | ||||||
| • Delay until test result obtained. For individuals not reached by intervention: | ||||||
| • Proportion not reached by intervention using condoms consistently in long term partnerships | ||||||
| • Condom use distribution in casual partnerships by clients or FSW (except figs 3B and C) | ||||||
| • Definition of condom use categories “sometimes” and “none” | ||||||
| Intervention parameter varied in each figure: | Figure 2B, C | Figure 3, supplementary figures B, C¶ | ||||
| Set 1 Simulations exploring coverage baseline 100% | Set 2 simulations with intervention parameters ⩾ targets | Set 3 Simulations with intervention parameters near targets | Set 4 | |||
| Fraction of FSW reached by STI clinic and condom intervention (freached) | 0–100% | 70–100% | 50–70% | 70% | ||
| STI treatment component: | ||||||
| Proportion of FSW with symptoms that attend clinic per month (fclinic) | 0–100% | 90–100% | 70–90% | 90% | ||
| Proportion of symptomatic FSW attending the STI clinic that are treated (ftreated-clinic) | 0–100% | 90–100% | 70–90% | 90% | ||
| Effectiveness of treatment regime for GC/CT (Efftreat) | 50–100% | 90–100% | 40–60% | 60% | ||
| PPT component: | ||||||
| Proportion of FSW reached by STI clinic intervention that receive PPT (fPPT) | 0–100% | 100% | 80–100% | 100% | ||
| Frequency of PPT (every x months) (FreqPPT) | 0–12 | 0–12 | 2–4 | 3 | ||
| Efficacy of PPT treatment for GC/CT (EffPPT-treat) | 75–100% | 75–100% | 75–95% | 95% | ||
| Condom use component: | 0–30% | |||||
| Average consistency of condom use amongst long term partnerships in those reached by intervention (CondomMain) | 0.5–100% | 30–100% | “All interventions” only; other interventions. as table 1F | 30% | ||
| Fraction of casual partnership in those “reached” where condom are used “always” (fcasual-always) | F: 22–100% | 50–100% | 40–60% | 50% | ||
| M: 12–100% | “All interventions” only; other interventions as table 1F) | |||||
| Fraction of FSW using condoms always with clients in those reached (fclient-always) | 80–100% | 90–100% | 80–100% | 90% | ||
| Definition of always condom use category (proportion of sex acts when condoms are used)Dall | Mysore: 0.8–0.9 | 0.9 | 0.8 | 0.9 | ||
| Bagalkot: 0.9 | ||||||
| (H) Definition of overall coverage parameters used in figures | ||||||
| Overall condom coverage = Fraction of FSW-client contacts protected among FSW reached and not reached by the intervention = freached (fcnone×Dnone+fcsome×Dsome+fcall×Dall)reached+(1−freached) (fcnone×Dnone+fcsome×Dsome+fcall x Dall)not reached | 69–81% | 80–90% | 69–81% | 81% | ||
| Overall STI coveragea = Fraction of FSW reached and cured by PPT at first visit = freached×fppt×fclinic×ftreated-clinic×EffPPT-treat | 0–62% | 0–100% | 15–54% | 54% | ||
| aThe effect of return visits at frequency DAll is not included | ||||||
| (I) Epidemiological parameters varied in specific figures: | Figure 2B, C (cont) | Figure 3, supplementary figures B, C (cont)¶ | ||||
| Set 1 (cont) | Set 2 (cont) | Set 3 (cont) | ||||
| Size of the FSW population | As baseline scenario | As baseline scenario | Figure 3 0.4–1.5%, B<0.9%, C>0.9% | |||
| Size of client population | As baseline scenario | As baseline scenario | 2.5–20% | |||
| HIV transmission probabilities per act | As baseline scenario | As baseline scenario | Normal distribution (mean F to M = 0.00087/act, 90% range = [0.0004, 0.00133].; mean M to F = 0.0016/act, 90% range = [0.0009 0.0023] | |||
| HSV-2 transmission probabilities per act | As baseline scenario | As baseline scenario | Normal distribution (mean F to M = 0.0023/act, 90% range = [0.013, 0.033], mean M to F = 0.0043/act., 90% range = [0.0021, 0.0068] | |||
| GC/CT transmission probabilities per act | As baseline scenario | As baseline scenario | Normal distribution (mean = 0.15/act; 90% range = [0.088, 0.212]. | |||