Cost-effectiveness of end-game strategies against sleeping sickness across the Democratic Republic of Congo =========================================================================================================== * Marina Antillon * Ching-I Huang * Samuel A. Sutherland * Ronald E. Crump * Paul E. Brown * Paul R. Bessell * Emily H. Crowley * Rian Snijders * Andrew Hope * Iñaki Tirados * Sophie Dunkley * Paul Verlé * Junior Lebuki * Chansy Shampa * Erick Mwamba Miaka * Fabrizio Tediosi * Kat S. Rock ## Abstract Gambiense human African trypanosomiasis (gHAT) is marked for elimination of transmission (EoT) by 2030. We examined the cost-effectiveness (CE) of EoT in the Democratic Republic of Congo, which has the highest global gHAT burden. In 166 health zones (HZs), we modelled the transmission dynamics, health outcomes, and economic costs of six strategies during 2024-40, including the cessation of activities after case reporting reduces to zero. Uncertainty in CE was assessed within the net monetary framework, presented as the optimal strategies at a range of willingness-to-pay (WTP) values, denominated in costs per disability-adjusted life-year averted. Status quo strategies, CE strategies (WTP=$500), and strategies with a high probability of EoT by 2030 are predicted to yield EoT by 2030 in 117 HZs, 130 HZs, and 138 HZs respectively, at a cost by 2040 of $159M (82M–266M), $175M ($98M–$285M), $206M ($114M–$339M). A more lenient timeline of EoT by 2040 could lead to EoT in 153 HZs at a cost of $189M ($105M–$311M), leaving 13 HZs shy of the goal. Investing in EoT by 2030 is predicted to reduce gHAT deaths from 34,770 (14,113–71,118) with status quo strategies to 8,214 (3,284–18,507). Keywords * sleeping sickness * elimination of transmission * economic evaluation * modelling * Democratic Republic of Congo ## Introduction *Gambiense* human African trypanosomiasis (gHAT), commonly known as sleeping sickness, is an enduring infectious disease that has afflicted certain West and Central African populations for centuries. Despite notable advancements in diagnosis, treatment modalities, and interventions, the disease continues to pose a significant threat to infected individuals lacking timely access to curative medications. Throughout the 20th century, gHAT witnessed three epidemics, with the most recent surge occurring during the 1990s, during which a staggering 37,385 cases were documented across Africa in 1998, 26,318 of which were in the Democratic Republic of Congo (DRC). Encouragingly, the period between 2008 and 2020 marked a transformative phase in gHAT control strategies, leading to a substantial reduction in reported cases. In 2022, only 799 cases were recorded globally, 516 of which originated in the DRC [1, 2]. The progressive decline in both the overall burden of gHAT and the share borne by the DRC offers optimism to the global health community, suggesting that the national control programme in the DRC has made significant strides towards achieving elimination [1, 2]. Aligned with the London Declaration on Neglected Tropical Diseases, a consortium targeting the elimination of various neglected tropical diseases, gHAT was identified as a key priority for achieving transmission elimination (EoT) by 2030 [3]. More recently, this target has been reiterated in the WHO’s 2030 roadmap for NTDs [4]. Previous studies have demonstrated that the observed reductions in reported cases are indicative of a decline in the underlying, albeit unobservable, transmission of gHAT, even in regions where screening coverage has diminished [5]. The currently available arsenal of tools exhibits promising potential in combating transmission throughout the DRC [6]. This manuscript aims to evaluate the feasibility and cost-effectiveness of achieving elimination of gHAT in the DRC. The study focuses on assessing six plausible strategies for gHAT control and elimination at the health zone level comprising different combinations of passive screening (PS) in fixed health facilities, active screening (AS) by mobile teams, treatment, and vector control (VC) using Tiny Targets which attract and kill tsetse via insecticide-impregnated fabric [7, 8]. Building upon prior health economic research, which concentrated on five health zones with different risk levels, this study expands the analysis to encompass health zones across the DRC which have reported cases [9]. To enhance the accuracy and reliability of the findings, the model fitting and projections are refined and compared to previous work, using an additional four years of gHAT case and screening data [5, 10]. Furthermore, the model accounts for epidemiological uncertainties related to potential non-human animal transmission by utilising an “ensemble model” approach. Stochastic projections are employed to better estimate variations in the timing of achieving elimination. In short, by using a comprehensive modelling framework, this study examines the intricate interplay among epidemiological dynamics, economic considerations, and temporal factors to inform effective decision-making regarding gHAT strategies for achieving EoT. ## Results The analysis process for each health zone is shown in Figure 1, consistent with our previous analyses [5, 6, 9]. In Step 1, the ensemble transmission model was fitted to 166 health zones. These health zones were distributed across 11 “coordinations”, which are subunits used by the national gHAT programme to coordinate intervention activities across the country. ![Figure 1:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/03/30/2024.03.29.24305066/F1.medium.gif) [Figure 1:](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305066/F1) Figure 1: a) Process of analysis. b) Future strategies simulated in most health zones. *Mean AS* is equal to the mean coverage of AS between 2016–2020, *Int. AS* is 30% or the historical maximum coverage between 2000–2020, whichever is higher. c) Future strategies simulated in Ango, Ganga, and Doruma health zones of the Bas Uélé region. We do not mean AS from 2016–2020 in these health zones as there was no AS during this period, hence *Restart PS* is effectively equivalent to this strategy. In these health zones, we take *Int. AS* to be the mean of 2010–2014 when MSF was operating in the region. In all health zones, the *Targeted VC* strategy only simulates VC along rivers in regions with high case density, and because the cases may be diffuse in some health zones, this strategy is not present in some health zones. Whereas the *Full VC* strategy involves the deployment of Tiny Targets throughout all large rivers in a health zone, regardless of the density of the cases. *Intervention cessation:* All strategies assume that AS will cease after 3 years of AS with zero cases in either AS or PS, followed by another AS in year 5 with no cases. RS is triggered if a case is found in PS and stops using the same 3+1 algorithm. VC stops after 3 years of no cases. PS is stopped 5 years after AS and RS have ceased. See the glossary and Table 8 for details of abbreviations. For Step 2, we simulated the current strategies and alternatives with the stochastic model as shown in Figure 1B. Strategies in the present study are comprised of AS at either a mean (*Mean AS*) coverage or an intensified level of AS (*Intensified AS*) which we assume is achieved by increasing turnout, screening neighbouring villages and/or screening in locations which historically had case reporting. The *Mean AS* coverage was set equal to the mean coverage of the last five years of data (2016–2020), and the *Intensified AS* coverage was set equal to the highest coverage within a health zone in the period of 2000-2020 or 30%, whichever is highest. Four out of the six strategies also include VC, combining *Mean AS* or *Intensified AS* with *Targeted VC* or *Full VC*. *Targeted VC* is the deployment of VC along large rivers (defined here as an average long-term discharge estimate for a river reach of *>* 20 m3/s [11]) with a case density of at least 1 case per 10 km of the treated riverbank. *Full VC* is the deployment of VC along all large rivers in the health zone. The *Full VC* strategy was considered in 155 of the 166 health zones we analysed, and the targeted strategy was considered in the 45 health zones where the amount of river meeting the above criterion was both non-zero and less than the extent considered as *Full VC*. For more details, see Supplementary Methods, Section A.8.1 and Section A.8. The status quo (comparator) strategy differs in some locations, as some locations have already conducted VC in addition to medical interventions. Furthermore, 10 health zones have no large rivers and therefore are not deemed suitable for deployment of VC. The breakdown can be found in Supplementary Methods, Section A.8. For Step 3, we take the simulated infections, cases reported, and undetected cases, and we apply them to a treatment outcome model, as shown in Supplementary Figure 8 and as done in previous work [9]. For each health zone, we have run the transmission model 20,000 times, constituting 10 stochastic iterations per draw for 2000 samples from a joint parameter distribution. Analogously, we sampled from the treatment costs and outcome parameters 20,000 times to propagate treatment- and cost-related uncertainty and calculate the costs of AS, PS, and VC (when applicable), which are detailed in Supplementary Methods, Section A.10. Lastly, we use the net monetary benefits (NMB) framework, which monetises DALYs averted (health benefits) according to a pre-defined willingness to pay (WTP) per DALY averted and subtracts the additional cost to deploy that strategy [12]. In this study, we have determined the optimal strategies at investment levels equal to $250, and $500, which encompasses the usual range of values considered to be “cost-effective” for a country with a GDP per capita of $577 in 2021 [13–15]. To determine the optimal strategy for a health zone across all iterations, one selects the strategy with the maximum NMB, averaged over all iterations. ### Strategies selected per health zone The optimal strategy for each health zone is shown in Figure 2 for the status quo, the minimum-cost strategy (WTP=$0), and the strategy that would be optimal at WTP=$250 and WTP=$500. The optimal strategies for each health zone are also shown when the goal is not to be cost-effective but to select the economically efficient strategy that maximises the probability of EoT by two dates: a more stringent goal date of 2030, and a more lenient goal date of 2040. For strategies in health zones that yield *<*90% probability of EoT, the strategy with the maximum probability of EoT is selected. Individual results by health zone are displayed in our custom-built graphical user interface [https://hatmepp.warwick.ac.uk/DRCCEA/v6/](https://hatmepp.warwick.ac.uk/DRCCEA/v6/), in addition to results for additional WTP values and aggregated by coordination and province. ![Figure 2:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/03/30/2024.03.29.24305066/F2.medium.gif) [Figure 2:](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305066/F2) Figure 2: Optimal strategies in each health zone of DRC compared to the status quo strategy (first column) depending on the level of investment (denominated by USD per DALYs averted), or by different target dates to maximize the probability of EoT. Colours represent the optimal strategy and numbers represent the probability of achieving EoT with that colour strategy by 2030. Strategies that do not reach a 90% probability of achieving EoT by 2030 are marked with an ‘x’. #### Strategies to maintain cost-effectiveness across willingness-to-pay levels To minimise costs, the analysis indicates that VC should be deployed in an additional 27 health zones, but it does not recommend additional AS anywhere. On the other hand, in Yasa Bonga, where VC has been in place since mid-2015, the analysis suggests that VC should cease to minimise costs with no sacrifice in the probability of elimination by 2030. In Bandundu health zone, where VC has been in place since mid-2019, the analysis suggests that VC should cease in order to minimise costs with only a one percentage point decrease in the probability of elimination from 0.99 to 0.98. In 9 and 16 health zones, the analysis recommends strategy changes at a WTP=$250 and WTP=$500, respectively. The difference between minimum cost strategies and strategies that are cost-effective at WTP=$250 would represent the addition of VC in six health zones (Wangata in Equateur Sud, Mushenge in Kasai Occidental, Isangi in Tschopo province, Maluku 2 in Kinshasa, Kasongo in Maniema-Katanga, and Minga and Tschumbe in Sankuru in Figure 2), the addition of AS in Doruma (in the Bas-Uélé region), and the addition of both AS and VC in Ango and Ganga (in the Bas Uélé region). To be more ambitious, the difference between strategies that are cost-effective at WTP=$250 and WTP=$500 would represent intensifying AS in three health zones (Nyunzu and Salamabila in Maniema Katanga and Tschumbe in Sankuru) and adding VC to the *Mean AS* strategy in four other health zones (Cilindu in Kasai Oriental; Tshitschimbi in Kasai Oriental; Lodja and Luseumbo in Sankuru in Figure 2). #### Strategies to maximise the probability of EoT The difference between the cost-effectiveness results at WTP=$500 and probability of EOT by 2030 means that the DALYs averted do not justify the additional costs avoided. In 69 health zones, this would mean more intense strategies than those that are CE at WTP=$500; in 31, 10, and 24 health zones there would be, a need for more AS, more VC, or both. Overall, the highest number of health zones where EoT-maximising strategies are not CE in Kasai Occidental (14), Kasai Oriental (15), and Maniema Katanga (10). There is one exception to the trends: in the health zone of Isangi, *Mean AS* will reach EoT with 0.91 probability and is the minimum-cost strategy, but the addition of targeted VC is cost-effective at WTP=$250 and WTP=$500. Overall, this means that 95 health zones need a change compared to *Status Quo* or 79 health zones would need different strategies to reach EoT by 2030 compared to the minimum-cost strategies. Of the 95 strategy changes needed, 16 of those changes will be cost-saving, and 10 of those changes will not be cost-saving but will be CE at WTP=$500; the remaining 69 changes will be neither cost-saving nor CE. The difference between striving for EoT by 2030 versus a more lenient timeline by 2040 is that 43 health zones need less ambitious strategies. Overall, this means that 72 health zones need a change compared to status quo strategies or 46 health zones would need different strategies to reach EoT by 2030 compared to the minimum cost strategies. Of the 72 strategy changes needed, 26 of those changes will be cost-saving, and 6 of those changes will not be cost-saving but will be CE at WTP=$500; the remaining 40 changes will not be CE. Notably, in 48 health zones, the strategy that maximises EoT by 2030 still has a *<*90% probability of reaching EoT by 2030. Twenty-two of the 48 health zones are in Kasai Occidental (12) and Maniema-Katanga (10), while the rest are in Kasai Oriental (6), Bandundu Nord (5), Bandundu Sud (5), Bas-Uélé region (all 3 health zones), and Equateur Sud (2). All the health zones in Equateur Nord, the portion of Isangi coordination in Tschopo province, Kinshasa, and Kongo Centrale have a *<*90% probability of reaching EoT by 2030. #### Geographic distribution of strategies and expected number of health zones reaching EoT Figure 3 displays the geographic distribution of optimal strategies at different investment levels and summarises the optimal strategy at WTP values of $0, $250, and $500, in addition to the 2030 and 2040 EoT targets. We expect that 117, 127, 129, and 130 health zones will reach EoT by 2030 with the status quo, the minimum cost, and the cost-effective strategies at WTP of $250–$500 (Figure 3). With strategies meant to maximise the probability of EoT, up to 138 health zones could reach EoT by 2030. However, our analysis suggests that, although 95 health zones would need to switch strategies to maximise the probability of EoT by 2030 this would increase the number of health zones that would reach EoT by 2030 by 21 for a final estimate of 138 health zones (Figure 3). ![Figure 3:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/03/30/2024.03.29.24305066/F3.medium.gif) [Figure 3:](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305066/F3) Figure 3: Maps and bar charts of optimal strategies according to economic or elimination goals for the whole of the DRC (time horizon 2024–2040 and 3% discounting, based on models estimated to 2000–2020 data). Maps show the optimal strategies depending on the status quo, minimum cost, and a WTP of $250 and $500, respectively. The final map shows the most efficient strategy with the maximum probability of EoT by 2030 and 2040 in each health zone of the DRC. Striped health zones are those where the probability of EoT by 2030 is less than 90%. Shapefiles used to produce this map were provided by Nicole Hoff and Cyrus Sinai under a CC-BY licence (current versions can be found at [https://data.humdata.org/dataset/drc-health-data](https://data.humdata.org/dataset/drc-health-data)). The table denotes the expected number of health zones that will meet the EoT goal by either 2030 or 2040 under the different objectives. An interactive map can be found at [https://hatmepp.warwick.ac.uk/DRCCEA/v6/](https://hatmepp.warwick.ac.uk/DRCCEA/v6/) If EoT were deferred until 2040, 136, 142, and 147 health zones would reach the goal with status quo, minimum-cost, and cost-effective strategies at WTP=$500. With EoT-maximising strategies, 154 health zones will reach EoT by 2040. Overall, 28 health zones will not reach EoT by 2030 and 12 health zones will not reach by 2040 under any approach, and these health zones are primarily concentrated in the Bas Uélé region, Kasai Occidental, Maniema Katanga, and Sankuru. ### Health outcomes and costs at different levels of investment Due to the uncertainty and the differential treatment of three health zones in the Bas Uélé region – Ango, Doruma, and Ganga – we show aggregate results that both include and exclude these health zones. Figure 4 shows the results by year, and Figure 5 shows the aggregated results for 2024-2040. ![Figure 4:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/03/30/2024.03.29.24305066/F4.medium.gif) [Figure 4:](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305066/F4) Figure 4: Projected total cases reported, deaths, DALYs and economic costs in all the DRC with optimal strategies at different investment levels over time. The left-hand column shows outcomes for the whole country and the right-hand side shows the whole country without the Bas-Uélé region (Ango, Doruma and Ganga health zones). Values here show actual numbers of projected cases, deaths, DALYs and costs (undiscounted) over the time horizon 2024–2040 based on fits to 2000–2020 data. ![Figure 5:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/03/30/2024.03.29.24305066/F5.medium.gif) [Figure 5:](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305066/F5) Figure 5: Projected total cases reported, deaths, DALYs and economic costs in all the DRC with optimal strategies at different investment levels between 2024 and 2040. The left-hand column shows outcomes for the whole country and the right-hand side shows the whole country without the Bas-Uélé region (Ango, Doruma and Ganga health zones). Values here show actual numbers of projected cases, deaths, DALYs and costs (undiscounted) over the time horizon 2024–2040 based on fits to 2000–2020 data. #### Status quo We expect 384 (95% PI: 2–460) reported cases in 2024 with status quo strategies, in addition to 1,788 (95% PI: 629–4,106) deaths, most of which would be undetected and occurring in the Bas Uélé region; excluding the Bas Uéle region, we expect only 680 (95% PI: 74–2,351) deaths. Because we simulated nothing being done in Bas Uéle, the reported cases will remain the same between the two estimates. With the status quo strategies, we expect 34,770 (95% PI: 14,113–71,118) deaths or 764,762 (95% PI: 316,897–1,554,736) DALYs, mostly in the Bas Uélé region. Outside Bas Uélé, we expect 8,256 (95% PI: 1,041–28,741) deaths or 188,545 (25,132–641,462) DALYs. Costs will be $15M (95% PI: 8M–23M) in 2024. After accounting for the fact that AS and VC operations will cease in health zones when no cases are reported, and PS clinics will be reduced to 1 in those same health zones 5 years after AS ceases, costs will be $159M (95% PI: $82M–$266M) by 2040 (undiscounted). #### Optimising for cost-effectiveness The relationship between cases reported and higher WTP is not unidirectional because additional AS and the deployment of VC have different impacts on the cases reported. Absent any changes in AS, more VC has a downward effect on reported cases, as VC interrupts transmission but does not aid detection. For instance, with minimum-cost strategies, reported cases are expected to decrease, amounting to 2,874 (95% PI: 264–11,195) in 2024-2040 including and excluding Bas Uélé, as the analysis indicates that VC should be done in additional health zones but it does not recommend additional AS anywhere (Figure 2 and 3). Similarly, if we exclude Bas Uélé in our estimates, at WTP=$250 the estimated reported cases by 2040 is even lower than with minimum-cost strategies, 2,532 (95% PI: 211–10,336), as outside Bas Uélé the analysis indicates that the difference between WTP=$0 and WTP=$250 constitutes only additions of VC to the ongoing *Mean AS* strategy (Figure 2). In contrast, if we include Bas Uélé, with a WTP=$250 the estimated cases reported increase, rather than decrease, to 4,452 (1327–13,011) because the Bas Uélé region must increase its AS coverage (Figure 2) increasing in the cases reported in that region alone from 0 to 1,920 (95% PI: 1,116–2,975) (Figure 14). If instead WTP=$500 is adopted, the cases reported are higher still than with the strategies at WTP=$250, a finding that holds with or without Bas Uéle: 4,517 (95% PI: 1320–13,049) and 2,597 (95% PI: 205–10,074) cases would be reported including and excluding Bas-Uélé, respectively. The higher number of cases reported at WTP=$500 compared to the cases reported at WTP=$250 are due to the increase in AS in Nyunzu and Salamabila in Maniema Katanga and Tschumbe in Sankuru, as all other differences in the optimal strategy between the two WTP levels constitute only adding VC to the Mean AS strategy (Figure 2). A higher WTP would lead to higher numbers of deaths and DALYs averted (Figures 4 and 5). Nevertheless, the differences in magnitude due to the inclusion of Bas-Uele are substantial. Including Bas-Uélé, minimum-cost and cost-effective strategies at WTP=$250 and WTP=$500 will yield deaths totaling 33,211 (95% PI: 13,992–65,082), 11,145 (95% PI: 3,742–28,252), 10,156 (95% PI: 3,512–25,337), respectively, and will yield DALYs totalling 731,119 (95% PI: 314,174–1,427,363), 269,369 (95% PI: 94,269–662,107), 247,539 (95% PI: 88,913–599,623), respectively. Excluding Bas-Uélé, the patterns are similar but at lower numbers; minimum-cost and cost-effective strategies at WTP=$250 and WTP=$500 will yield deaths totaling 6,696 (95% PI: 920–22,705), 5,456 (95% PI: 725–18,927), and 4,468 (95% PI: 495–16,013), respectively, and DALYs totaling 154,902 (95% PI: 22,409–514,089), 128,898 (95% PI: 18,135–435,403), and 107,067 (95% PI: 12,779–372,919). Minimum-cost strategies are marginally less costly in the long run, but in the short run, these strategies represent small cost increases. In 2024, minimum-cost strategies will cost $16M (95% PI: $9M–$24M) compared to the cost of status quo strategies at $15M (95% PI: $8M–$23M), but by 2040, minimum-cost strategies will have cost $151M (95% PI: $82M–$249M) compared to the status quo, $159M (95% PI: $82M–$266M). At WTP=$250 and WTP=$500, the total cost will be $166M (95% PI: $93M–$271M) and $175M (95% PI: $98M–$285M), respectively, including Bas Uélé, and $156M (95% PI: $86M–$256M) and $163M (95% PI: $90M–$268M), respectively, excluding Bas Uélé. #### Optimising to maximise the probability of EoT We also considered how the total cases reported, deaths, DALYs, and costs compare when EoT is the goal, rather than when remaining cost-effective is the goal. Strategies that EoT-maximising strategies (by 2030) would see a rise in the cases reported to 4649 (95% PI: 1,301–13,587) and will bring the magnitude of deaths down to 8,214 (95% PI: 3,285–18,507), or 203,909 (95% PI: 83,607–450,290) DALYs, at a cost of $206M (95% PI: $114M–$339M). Compared to the status quo, EoT by 2030 would mean receiving an additional 962 reported cases, but averting 26,556 deaths, and 560,853 DALYs at just an additional $47M. If we exclude Bas-Uélé, strategies with a high probability of EoT by 2030 would see a decrease in cases reported to 2,766 (95% PI: 191–3,538), a decrease in deaths to 2654 (95% PI: 301–9,458), a decrease in DALYs to 66,026 (95% PI: 8,057–228,983), at a total cost of $193M (95% PI: $106M–$321M). Compared to the status quo, EoT by 2030 excluding Bas Uélé would mean seeing 921 fewer reported cases, averting 5,602 deaths and 122,519 DALYs, at a cost of $34M. Notably, however, the additional costs will have to be front-loaded; in the year 2024, the status quo will cost $15M (95% PI: $8M–$23M) while an EoT 2030 strategy would cost between $23M (95% PI: $14M–$33M) and $24M (95% PI: $15M–$35M). However, by 2035 EoT approaches will cost less than the status quo according to estimates including Bas Uélé, and excluding Bas Uélé EoT approaches will cost less than the status quo by 2033. Deferring EoT until 2040 saves only $17M and will still accumulate 4,670 (95% PI: 1,298–13,872) reported cases while incurring 8,421 (95% PI: 3,287–19,849) deaths and 208,791 (95% PI: 83,687–480,077) DALYs if we include Bas-Uélé. Compared to EoT by 2030, that would mean 30 more cases reported, 7 more deaths and 4,882 DALYs. If we exclude Bas-Uélé, delaying the EoT goal will save $16M, but the cases reported would be 2,796 (95% PI: 191–10,979) and the burden will be 2861 (95% PI: 303–10,799) deaths and 70,904 (95% PI: 8,115–258,782) DALYs. Compared to EoT by 2030, that would mean 30 more cases reported, 7 more deaths and 4877 DALYs. All results are available stratified by coordination in Supplementary Figures 16–21 and in our GUI [https://hatmepp.warwick.ac.uk/DRCCEA/v6/](https://hatmepp.warwick.ac.uk/DRCCEA/v6/), which also shows results by additional WTP values. Expected case trends throughout the country show a decline even with status quo or cost-minimizing strategies, with the notable exception of the Bas Uélé region, which will reap the largest health benefits if more ambitious strategies are implemented. ### Breakdown of costs and resources needed #### By coordination The largest portion of the costs at any level of investment will go to AS and PS screening and VC activities, although the distribution differs substantially by coordination (see Supplementary Figures 22 and 23). An effort to maximise EoT by 2030 would mean costs will be incurred primarily from Isangi, Kasai Oriental, and Maniema-Katanga. Bandundu Nord and Sud will see cost savings by 2040, and other coordinations may potentially break even by 2040. However, most coordinations are expected to need an infusion of resources soon to meet the EoT goal by 2030 or 2040, with the notable exception of Kongo Central. #### Resources needs We have estimated the number of drugs and screening tests that will be required (Supplementary Figures 24–25). NECT and fexinidazole will continue to figure prominently in treatment until acoziborole is approved, needing a mean of 174 doses of NECT and 211 doses of fexinidazole in 2024 under status quo strategies. Also in status quo, 2M tests will be needed for AS and 233 thousand tests will be needed for PS. If activities are expanded in line with EoT by 2030, a mean of 477 doses of NECT and 657 doses of fexinidazole will be needed, and screening tests will be used at a rate of more than 3M per year for AS for several years with a slow decline after 2030, and at a rate of about 400 thousand per year in PS, which is projected to decline more slowly than AS tests due to the need for long-term surveillance. ## Discussion This manuscript presents a comprehensive analysis evaluating the feasibility and cost-effectiveness of achieving the elimination of gHAT in the DRC. This analysis reveals that, overall, the EoT of gHAT by 2030 appears to be epidemiologically feasible with the existing toolkit in almost all health zones of DRC. However, while certain coordinations, such as Kongo Central and Equateur Nord are broadly on track to successfully achieve the elimination of transmission with the continuation of current strategies, 95 health zones are predicted to require strategy modifications to reach EoT by 2030 (Figure 2). In 16 of these 95 health zones, new strategies are expected to be cost-saving by 2040. To reach EoT by 2040, 72 health zones likely require strategy modifications and in 29 of those, strategies would be cost-saving by 2040. Notably, 48 health zones were projected to have a *<*90% probability of meeting EoT by 2030 even with a strategy change, and most of those health zones are in Kasaï Occidental and Maniema-Katanga. Compared to previous modelling studies [5, 6], this analysis demonstrates a more pessimistic outlook regarding EoT. Despite now having more data on cases from recent AS and PS which demonstrate a somewhat optimistic downward trend, our new reservations on the probability of EoT by 2030 arise from several factors related to modelling and analysis changes which we believe better capture the uncertainty in the predictions: a) previous analyses were performed with a deterministic model, while this analysis is performed with a stochastic model, which makes us less optimistic when considering the probability of EoT, b) our more refined estimates of the potential impact of VC based on case and river proximity, and c) 2024–2030 is a shorter time horizon than 2020–2030 – the previous time horizon in other analyses. Especially with regards to the importance of timing, our analysis shows how the proximity to 2030 is becoming a hindrance – with strategies aimed at maximising the probability of EoT by 2030, 138 health zones are expected to reach EoT by 2030 and an additional 16 health zones are forecast to reach EoT by 2040. The findings suggest that strategies likely to achieve EoT by 2030 will require an appreciable increase in economic resources. While status quo strategies have an expected mean cost of $15M in 2024, strategies to reach EoT have an expected mean cost of $24M in 2024, a 60% increase. However, by 2040, returns on investment from EoT by 2030 would mean that the accumulated cost of EoT will be only 30% higher than the status quo interventions. These expanded strategies are predicted to have the greatest impact in the eastern coordinations, where limited activities have occurred in recent years and uncertainty in current transmission levels favours the adoption of more intense strategies. Whilst the treatment costs are quite small, timely access to treatment is pivotal for strategies to remain cost-effective, as treatment is the mechanism through which deaths and DALYs are averted and through which further cases are prevented in the health zones where no VC is implemented. While delaying the EoT target until 2040 could save an estimated $16M-$17M compared to implementing activities aimed at achieving elimination by 2030, it would incur an additional 7 deaths and almost 5000 DALYs; moreover, it is crucial to consider potential unforeseen events in the extra decade that may undermine the political will to sustain these activities. Operational feasibility emerges as a major challenge. Although effective tools exist and there is potential for increased donor funding to support expanded activities, the convergence of regions requiring new strategies with issues of inaccessibility and/or civil unrest presents a notable hurdle. Rapidly scaling up trained personnel to carry out these activities remains a crucial operational challenge that must be addressed. Despite operational challenges, this study presents a compelling case for the sensible utilization of resources in support of control efforts for gHAT that strive towards achieving elimination of transmission by 2030. By considering high-incidence regions, we demonstrate that additional investments in VC could potentially yield cost savings. Furthermore, we show that substantial investments aimed at achieving EoT do not need to be sustained for more than a decade in moderate and low-incidence regions. Additionally, this analysis confirms previous results, showing that VC could play a pivotal role in achieving elimination, as it is the key determinant in reducing the tsetse population. To alleviate the burden of disease, strengthening AS efforts is beneficial in the short term by identifying and treating individuals currently infected and ill, even if they are unlikely to transmit the infection whilst the tsetse population is suppressed via VC. The capacity to scale back AS without the need for RS is primarily influenced by the introduction of VC in locations where it is not yet implemented, as any undetected cases are wont to remain so in the context of imperfect screening. Lastly, this study highlights the importance of strategically allocating resources in gHAT control programmes to maximise cost-effectiveness and achieve the goal of EoT by 2030. By targeting high-incidence areas and implementing appropriate interventions, such as VC and intensified AS strategies, major progress can be made towards eliminating gHAT transmission, ultimately leading to substantial improvements in public health. ### Limitations This study has limitations inherent to its scope and methodology, categorised into three main areas: #### 1. Past data Whilst we account for the underreporting of cases, the study acknowledges the mechanistic models cannot extrapolate to regions with no or minimal historical case data due to insufficient surveillance. Alternative approaches (e.g. geostatistical modelling using satellite imagery) may offer better quantitative support for the absence of suitable tsetse habitat and gHAT transmission. In regions with sufficient data for modelling, including the Bas-Uélé region, there is a different data interpretation challenge as alternative screening algorithms were performed by MSF during 2007–2014 followed by virtually no subsequent surveillance leading to large uncertainty in modelled results. Exploratory screening in regions with little or no recent screening data could help refine model predictions, however, the strength of data to reduce uncertainty will be linked to the number of people screened. #### 2. Future strategies The study assumes the perfect application of tools and robust quality assurance in the future. In addition to video confirmation of case diagnosis which has been gradually rolled out in the DRC since 2015, further maintenance and development of the quality assurance system will be needed. For treatment, we did not simulate acoziborole, and all diagnosed cases must receive medication promptly, requiring sufficient stocks near where cases are diagnosed. If the same algorithm is used for diagnosis we would expect projected total treatments to remain the same if acoziborole becomes available. Operational feasibility is identified as a major hurdle, with regions such as Sankuru, Maniema Katanga, and the Bas-Uélé region posing significant challenges in terms of accessibility and security. These factors can impact the implementation of interventions and may result in overambitious projections for these regions. Throughout the country emphasis should be placed on the importance of population turnout in active screening to improve the efficiency of this intervention [16], particularly as community knowledge wanes with the increasing rarity of the disease; the model currently accounts for continued AS turn out based on past data. Other interventions that are under consideration for the future but do not feature in the present analysis are the community-based deployment of VC targets [17–19], a “screen-and-treat” intervention that would circumvent confirmation and staging if the new acoziborole treatment is approved and deemed suitably safe [20] and new laboratory-based diagnostics [21, 22]. These future strategies should be considered in future simulation studies, particularly when information on implementation algorithms for screen-and-treat and roll-out dates for acoziborole can be better defined. #### 3. Model formulation and use The analysis does not consider the cost of end-game diagnostics with higher specificity, which could improve the positive predictive and negative predictive value at low prevalence. As this diagnostic would be present in all possible strategies to the same degree, it is not expected to influence the selection of optimal strategies but would slightly affect the economic cost estimation. Likewise, post-elimination surveillance is not explicitly modelled, although we assume that PS will continue at reduced levels. One potential issue is treatment completion, as there have been instances where treatment access was *<*100% due to logistical issues or delays, which can impact the model fitting process and potentially contribute to ongoing transmission and disease burden. Resolving delayed or undertreatment in the future, however, is a matter of organisational capacity between pharmaceutical companies, the WHO, and the country programme. Ensuring sufficient spare stock at the right points in the supply chain is not considered in this analysis, however, this modelling work could be adapted to support stock planning. Future work will also encompass modelling at smaller spatial scales as, although analysis at the health-zone level aligns with the scale at which elimination as a public health problem is assessed for a country, health areas align more closely with operational decisions made for interventions. Health area modelling requires fitting to around 1,200 regions rather than 166 zones. Whilst this is the goal of future modelling work and the proof-of-principle for the modelling methodology has been demonstrated [23], it is beyond the scope of the present analysis to expand to this granularity. Overall, we believe our health zone-modelled trends provide very similar outputs to modelling at a smaller scale and aggregating [23]. Our predictions are potentially conservative, as the model assumes larger operations will remain in place longer versus reality. A cost-effectiveness analysis at the health-area level might, therefore, be more optimistic as additional interventions could be targeted at a smaller scale. ## Methods ### Data Historical case data was acquired from the WHO HAT Atlas, which included records of the populations seen in AS activities and clinics [1]. The HAT Atlas differentiates between cases detected during AS and PS and, during 2015–2019, includes staging information for the majority of cases. Before 2015, staging information was not routinely digitised even though it would have been performed to determine which treatment should be administered, and from 2020 many cases eligible for fexinidazole no longer required staging for treatment and therefore this information was never collected. We used location names (province, health zone, and health area) and geolocations (with two shapefiles provided by the American Red Cross [24]) to aggregate records including AS numbers and case detections from both AS and PS by year in 519 health zones. Out of 519 health zones in the DRC, 281 health zones reported one or more cases of sleeping sickness between 2000–2020. Of those, 190 health zones had 10 or more data points consisting of a) years of AS, yielding any number of cases, including zero, plus b) years with one or more cases of gHAT reported from fixed health facilities (PS) that have the capacity for parasitological (microscopy) confirmation. Following recommendations from PNLTHA, we excluded several health zones in urban areas, believing that case reports were likely due to importations from surrounding rural areas and the presence of a diagnostic hospital in those regions (see Table 1). After these exemptions, 166 health zones remained, encompassing 135,029 cases, or 96% of all cases reported in 2000–20. View this table: [Table 1:](http://medrxiv.org/content/early/2024/03/30/2024.03.29.24305066/T1) Table 1: Model Parameters. For further details and sources, see Supplementary Note 3. CIs: confidence intervals. AS & PS: active and passive screening, respectively, VC: vector control, PNLTHA: Programme de Lutte contre la Trypanosomie Humaine, NECT: nifurtimox-eflornithine combination therapy, CATT: card agglutination test for trypanosomiasis, S1 & S2: stage 1 & 2 disease, DALYs: disability-adjusted life-years, SAE: severe adverse events, RDT: Rapid Diagnostic Test The analysis is organised by coordination, but results are presented both for individual health zones and aggregated by coordinations, which are the units of administration of the programme. Coordinations were delineated in the early 2000s and are contiguous with or partitioned the provinces before 2015 into approximately equal areas (see Supplementary Figure 1). Health zones were initially delineated to provide a unit of public health administration by the Ministry of Health. Health zones initially served approximately 100,000 people, but estimates of the population in 2022 show that a median of 210 thousand people live in each of the 166 health zones in our analysis, for a total of 37.7 million people (see Table 2). In terms of operations, the four major activities of the programme are AS with mobile units moving from village to village, PS in fixed health facilities with serological (RDT or CATT) tests and parasitology, VC via Tiny Targets, and treatment. First, the health zones are delineated to have a referral hospital, where historically gHAT treatment could take place before the availability of the simpler oral treatment of fexinidazole in the year 2020, and in many cases, the only health facilities where one would self-report for diagnosis with serological and parasitological tests and staging of the disease. Second, recent VC activities are organised at the health-zone level with coordinating support from the coordination and the national programme, in addition to international partners. Although AS is organised at the coordination level, these units are mobile, require limited start-up costs to expand, can be scaled back, and often serve more than one health zone, so considering the resource use by the health zone will accurately reflect the collective costs at the coordination level. For cost purposes, healthcare-seeking individuals screened during PS were deduced from two data sources: the historical records of clinics capable of screening with serological tests (RDT or CATT) [25], and records acquired from the PNLTHA for the years 2019–2020. Cost data was acquired from the literature and conversations with programme staff ([26, 27], among others) and the unit costs and cost functions are detailed in the Supplementary Methods. ### Transmission model fit to data For this study, we used two variants of the previously published Warwick gHAT model [5, 10] consisting of a mechanistic, deterministic modelling framework to explicitly simulate transmission between humans and possibly animals via tsetse vectors (see Supplementary Figure 2 and the Supplementary Methods, Section A.5). Model parameterization was performed individually for each model variant (models with and without possible animal transmission) and was updated compared to previous publications by fitting to WHO HAT Atlas data from 2000–2020 for each health zone in the DRC that had sufficient data—at least 10 or 13 data points for the models without and with animal transmission respectively (where any year with AS and any year with non-zero passive case detection count as individual data points). This gave 166 health zones with a fit using the model without animal transmission and 156 that were fitted to both model variants. More details of the statistical fitting procedure are provided in the Supplementary Methods A.5. The “ensemble” model consists of posteriors from both models. The proportion of samples from both models was determined by a statistical method (Bayes factors) that measures the relative goodness of fit of each model to the data. A glossary of epidemiology and health economic terms is found in the appendix, Supplementary Note 1. ### Epidemiological projections Stochastic simulations, which capture chance effects at very low prevalence, were performed rather than deterministic simulations, which would represent only average expected dynamics [6]. The exact list of strategies is refined compared to those presented in previous work [6]. We determined the strategies in the present study through conversations with PNLTHA-DRC and NGO or academic partners. Projections under up to eight plausible gHAT intervention strategies were simulated from 2024–2053 and are listed in Figure 1; furthermore, each activity is further explained in the following section, Supplementary Methods, Section A.8, and Supplementary Table 8. Each of our strategies represents a specific combination of three available interventions at different levels. AS has historically been the flagship activity of gHAT control, and therefore we simulate AS at two coverage levels: *Mean AS* which is the average coverage of the most recent five years in the data (i.e. 2016–2020), and *Intensified AS* which is the higher of the historical maximum (in 2000–2020) or 30% of the health zone. This *Int. AS* option was selected because the historical maximum is quite low in many health zones (less than 5% of people screened in 45 analysed health zones). PS is simulated to continue at estimated 2020 levels into the future; this level reflects some improvements over time, estimated by model fitting. We do not include additional future improvements to the PS system. We used the 2019 WHO survey of clinics to calculate the cost of PS with two exceptions: for coordinations Bandundu Nord and Bandundu Sud. In those coordinations, in the provinces of Mai-Ndombe, Kwilu, and Kwango, there were ongoing trials at the time to expand the network of health posts that would have RDTs available for serological diagnosis of HAT. Patients would then be referred to the health zone’s health centre for confirmation, staging (when necessary) and treatment. However, that programme has since ceased, and most of those clinics are no longer getting support. Therefore, we have used the number of PS clinics that were listed for Bandundu Nord and Sud in the survey that was published for the year 2012 [25]. For an overview of the operational inputs in each health zone, see Supplementary Table 9, and for the specifics in each health zone, see the Supplementary Note 1 and the GUI. VC activities have been shown to decrease both tsetse populations and case reports in other locations [28–30], and therefore we simulate the addition of VC activities at two levels. In this analysis any strategies including *Targeted VC* use an adapted algorithm based on that previously used by LSTM to identify areas with high case density at which to focus Tiny Target deployment efforts along large rivers. Strategies including *Full VC*, by contrast, involve a substantial expansion of VC interventions, considering the deployment of Tiny Targets throughout all large rivers in a health zone. A full algorithm for computing the case coverage and corresponding health-zone level tsetse population reductions is given in Supplementary Information A.8.1, however, we note that case coverage is highly variable depending on the recent case distribution and geographic proximity to large rivers suitable for Tiny Target deployments. Cessation of vertical interventions was modelled as stopping AS and VC based on consecutive years of zero detected cases by any screening modality. For AS this was three years of zero cases plus one additional AS in a fifth year in alignment with the WHO algorithm for stopping AS [3]. For VC, we used three years of no case detections (without the requirement for the fifth year). Following cessation, reactive screening (RS) is started if a new case presents to the fixed health facilities (PS) and more cases are found in the targeted screening that takes place in the next year. RS lasts until the same cessation has been met again (e.g. 3+1 years of no cases). To ensure *Mean AS* is a distinct strategy from *Int. AS* we force the model to simulate at least one year of *Int. AS* (in 2024) even if the cessation criterion has been met. For VC, we force operations to stay active for a minimum of three years of deployments from the start year and there must be at least one year of VC (in 2024) if a strategy name includes *+ VC*. Again, this ensures that there is a clear difference between strategies with and without VC, and the three-year rule is based on the assumption that to justify training local deployment teams there would need to be at least several years of the activity in the health zone. Strategies without VC in the name have no future VC in our model simulations, so the cessation criterion does not need to be met to stop any VC in places that previously started under these strategies. Reactive VC is not considered in our simulations. PS is assumed to remain constant for the duration of our transmission model simulations, however, in the health economic model we assume PS is scaled back to one health centre per health zone after 5 years with no case detections in that health zone. Fexinidazole is simulated as being available immediately, but acoziborole – which is not yet approved for use outside of clinical trials – is not included in this analysis. ### Cost-effectiveness analysis #### Health outcomes For individual disease outcomes, we used the outputs of the transmission model as inputs in a probability tree model of disease outcomes (see Figure 8). We simulated the disease process separately for stage 1 and stage 2 of the disease, including steps to sort patients into the type of care indicated by the WHO, treatment success or failure, diagnosis in the event of treatment failure, and progression to rescue treatment. Health burden is denominated in disability-adjusted life years (DALYs), but we also report cases and deaths as intermediate outcomes. All parameters are listed in Table 1. We assumed that the mean age of death from gHAT is 26.6 years (95% CI: 22.4–31.8, see Supplementary Note 3, the parameter glossary, for explanation). #### Costs We developed a cost function that incorporates the costs presented at each step of the transmission and the treatment models, including fixed and variable costs of operating preventive and therapeutic activities (see Supplementary Methods, Section A.10). Disease costs include diagnosis, confirmation, and staging (lumbar puncture), as well as the cost of the drug itself and the administration by health care personnel (nurse or community health care worker). We performed our analysis from the perspective of health- or intervention-delivery payers collectively (the government, the WHO, donors, etc), but excluding any research personnel or activities performed by academic partners; for that the value of their labour is converted to local costs. Costs were dominated in 2022 US dollars. #### Economic evaluation and investment horizon We adopted the net-benefits framework, which expresses the probability that an intervention is optimal at a range of thresholds, known as the willingness-to-pay (WTP) thresholds. Unlike computing single ICERs, this takes into account all uncertainty and refrains the analysis from invoking any single WTP threshold as the cost-effective threshold. We discount both costs and health outcomes at a rate of 3% annually. We adopt a time horizon that is relatively long (2023–2040), as opposed to the convention (10 years) to assess the returns on investments in both augmented disease control and elimination after the 2030 goal has been reached. ## Supporting information Supplementary Information [[supplements/305066_file02.pdf]](pending:yes) ## Data Availability Information about the WHO HAT Atlas data used for fitting is described in the Supplementary Information. Data cannot be shared publicly because they were aggregated from the World Health Organisation’s HAT Atlas which is under the stewardship of the WHO; our data-sharing agreement does not allow us to share that data. WHO HAT Atlas data include identifiable data. Data are available from the WHO (contact neglected.diseases{at}who.int or visit [https://www.who.int/health-topics/human-african-trypanosomiasis/](https://www.who.int/health-topics/human-african-trypanosomiasis/)) for researchers who meet the criteria for access to confidential data, including secure computational facilities and an existing relationship to the national sleeping control program of the DRC. Time-frame for response would depend on the WHO’s timelines and workloads. Clinical outcomes and costs (listed in Table 1) were simulated using estimates from the literature and are described in Supplementary Note 3: Parameter Glossary. Assumptions and estimates were parameterised according to conventions in the economic evaluation literature [60]. All modelled outputs and code are available via Open Science Framework: [https://osf.io/ezjxb/](https://osf.io/ezjxb/). ## The role of the funding source This work was supported by the Bill and Melinda Gates Foundation ([www.gatesfoundation.org](http://www.gatesfoundation.org)) through the Human African Trypanosomiasis Modelling and Economic Predictions for Policy (HAT MEPP) project [INV-005121] (MA, SAS, CH, REC, PEB, EHC, KSR, and FT), and through the TRYP-ELIM projects [OPP1155293, INV-031337] (PRB, RS, AH, IT, SD, PV, and EMM) and by the Belgian Directorate-general for Development Cooperation and Humanitarian Aid (EMM, RS, and PV). The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. ## Author Contributions Statement MA, SAS, CH, REC and KSR developed the software and performed the analyses. MA and PEB visualised the results. EMM, JL, CS, SAS, PRB and RS curated the data. REC, SAS, CH and MA analysed the data. MA, RS, SAS, PRB, REC, and KSR developed the methods. MA, KSR, EHC and FT wrote the original draft. MA, KSR and FT conceptualised the study. All authors reviewed and approved the final version for publication. ## Competing Interests Statement The authors declare no competing interests. ## Ethical approval Ethics approval was granted by the University of Warwick Biomedical and Scientific Research Ethics Committee (application number BSREC 80/21-22) to use the previously collected DRC country HAT data, provided through the framework of the WHO HAT Atlas [2], in this secondary modelling analysis. No new data collection took place within the scope of this modelling study. ## Rights retention statement For the purpose of open access, the author has applied a Creative Commons Attribution (CC-BY) licence to any Author Accepted Manuscript version arising from this submission. ## Acknowledgements The authors thank PNLTHA for original data collection, WHO for data access (in the framework of the WHO HAT Atlas [2]), Cyrus Sinai and Nicole Hoff from UCLA Fielding School of Public Health for providing health zone level shapefiles (current versions can be found at [https://data.humdata.org/dataset/drc-health-data](https://data.humdata.org/dataset/drc-health-data)). Calculations were performed using the sciCORE ([http://scicore.unibas.ch/](http://scicore.unibas.ch/)) Scientific Computing Center at the University of Basel. The authors also thank Steve Torr and Alex Shaw for their valuable insights on specific topics in this article. For the purpose of open access, the authors have applied a Creative Commons Attribution (CC-BY 4.0) licence to any Author Accepted Manuscript version arising from this submission. * Received March 29, 2024. * Revision received March 29, 2024. * Accepted March 30, 2024. * © 2024, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution 4.0 International), CC BY 4.0, as described at [http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/) ## References 1. 1.World Health Organization. Global Health Observatory data repository. 2020. Available from: [https://www.who.int/data/gho/data/indicators/indicator-details/GHO/hat-tb-gambiense](https://www.who.int/data/gho/data/indicators/indicator-details/GHO/hat-tb-gambiense) 2. 2.Franco JR, Cecchi G, Paone M, Diarra A, Grout L, Ebeja AK, Simarro PP, Zhao W, and Argaw D. The elimination of human African trypanosomiasis: Achievements in relation to WHO road map targets for 2020. PLoS Neglected Tropical Diseases 2022; 16:1–19. doi: doi:10.1371/JOURNAL.PNTD.0010047. Available from: 10.1371/journal.pntd.0010047 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0010047&link_type=DOI) 3. 3.WHO Expert Committee on human African trypanosomiasis. Control and surveillance of human African trypanosomiasis: report of a WHO expert committee. Tech. rep. 2013 :237. Available from: [https://www.who.int/trypanosomiasis\_african/surveillance/en/](https://www.who.int/trypanosomiasis_african/surveillance/en/) 4. 4.World Health Organization. Ending the neglect to attain the Sustainable Development Goals: a road map for neglected tropical diseases 2021–2030. Tech. rep. Geneva, Switzerland, 2020 :196 5. 5.1. Perkins A Crump RE, Huang CI, Knock ES, Spencer SEF, Brown PE, Mwamba Miaka E, Shampa C, Keeling MJ, and Rock KS. Quantifying epidemiological drivers of gambiense human African Trypanosomiasis across the Democratic Republic of Congo. PLOS Computational Biology 2021 Jan; 17. Ed. by Perkins A:e1008532. doi: doi:10.1371/journal.pcbi.1008532. Available from: [https://dx.plos.org/10.1371/journal.pcbi.1008532](https://dx.plos.org/10.1371/journal.pcbi.1008532) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pcbi.1008532&link_type=DOI) 6. 6.Huang CI, Crump RE, Brown PE, Spencer SE, Miaka EM, Shampa C, Keeling MJ, and Rock KS. Identifying regions for enhanced control of gambiense sleeping sickness in the Democratic Republic of Congo. Nature Communications 2022; 13:1–11. doi: doi:10.1038/s41467-022-29192-w [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41467-022-29192-w&link_type=DOI) 7. 7.Liverpool School of Tropical Medicine. Use of Tiny Targets to control tsetse flies in Gambian HAT foci: standard operating procedures. Tech. rep. October. 2016 8. 8.1. Matovu E Tirados I, Hope A, Selby R, Mpembele F, Miaka EM, Boelaert M, Lehane MJ, Torr SJ, and Stanton MC. Impact of tiny targets on Glossina fuscipes quanzensis, the primary vector of human African trypanosomiasis in the Democratic Republic of the Congo. PLOS Neglected Tropical Diseases 2020 Oct; 14. Ed. by Matovu E:e0008270. doi: doi:10.1371/journal.pntd.0008270. Available from: [https://dx.plos.org/10.1371/journal.pntd.0008270](https://dx.plos.org/10.1371/journal.pntd.0008270) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0008270&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=33064783&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 9. 9.Antillon M, Huang CI, Crump RE, Brown PE, Snijders R, Miaka EM, Keeling MJ, Rock KS, and Tediosi F. Cost-effectiveness of sleeping sickness elimination campaigns in five settings of the Democratic Republic of Congo. Nature Communications 2022 Dec; 13:1051. doi: doi:10.1038/s41467-022-28598-w. Available from: 10.1101/2020.08.25.20181982%20[https://www.nature.com/articles/s41467-022-28598-w](https://www.nature.com/articles/s41467-022-28598-w) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41467-022-28598-w&link_type=DOI) 10. 10.Crump RE, Huang CI, Spencer SE, Brown PE, Shampa C, Miaka EM, and Rock KS. Modelling to infer the role of animals in gambiense human African trypanosomiasis transmission and elimination in the DRC. PLoS Neglected Tropical Diseases 2022; 16:1–23. doi: doi:10.1371/JOURNAL.PNTD.0010599 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/JOURNAL.PNTD.0010599&link_type=DOI) 11. 11.Lehner B and Grill G. Global river hydrography and network routing: baseline data and new approaches to study the world’s large river systems. Hydrological Processes 2013; 27:2171–86 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/hyp.9740&link_type=DOI) [GeoRef](http://medrxiv.org/lookup/external-ref?access_num=2013077575&link_type=GEOREF) 12. 12.Briggs AH. Handling uncertainty in combined endpoints. Pharmacoecnomics 2000; 17:479–500. doi: doi:10.2165/00019053-200017050-00006 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.2165/00019053-200017050-00006&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=10977389&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000087476400007&link_type=ISI) 13. 13.World Health Organization. Making Choices in Health: WHO guide to cost-effectiveness analysis. Tech. rep. Geneva, Switzerland, 2003 :318 14. 14.Marseille E, Larson B, Kazi DS, Kahn JG, and Rosen S. Thresholds for the cost–effectiveness of interventions: Alternative approaches. Bulletin of the World Health Organization 2015; 93:118–24. doi: doi:10.2471/BLT.14.138206 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.2471/BLT.14.138206&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25883405&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 15. 15.Woods B, Revill P, Sculpher M, and Claxton K. Country-Level Cost-Effectiveness Thresholds: Initial Estimates and the Need for Further Research. Value in Health 2016; 19:929–35. doi: doi:10.1016/j.jval.2016.02.017. Available from: 10.1016/j.jval.2016.02.017 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jval.2016.02.017&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=27987642&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 16. 16.Rock KS, Torr SJ, Lumbala C, and Keeling MJ. Predicting the Impact of Intervention Strategies for Sleeping Sickness in Two High-Endemicity Health Zones of the Democratic Republic of Congo. PLOS Neglected Tropical Diseases 2017; 11:1–17. doi: doi:10.1371/journal.pntd.0005162 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0005162&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=28056016&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 17. 17.Vander Kelen C, Mpanya A, Boelaert M, Miaka E, Pérez Chacón D, Pulford J, Selby R, and Torr SJ. Feasibility of community-based control of tsetse: A pilot project using Tiny Targets in the Democratic Republic of Congo. PLoS Neglected Tropical Diseases 2020; 14:e0008696 18. 18.Vander Kelen C, Mpanya A, Hasker E, Miaka E, Nzuzi R, Torr S, Perez D, and Pulford J. ‘Where are the dead flies!’: perceptions of local communities towards the deployment of Tiny Targets to control tsetse in the Democratic Republic of the Congo. BMJ Global Health 2022; 7:e006879 [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NToiYm1qZ2giO3M6NToicmVzaWQiO3M6MTE6IjcvMS9lMDA2ODc5IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDMvMzAvMjAyNC4wMy4yOS4yNDMwNTA2Ni5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 19. 19.Vander Kelen C, Mpanya A, Hasker E, Miaka E, Nzuzi R, Pulford J, Torr S, and Perez Chacon D. Community empowerment through participation in a tsetse control project in the Democratic Republic of Congo. PLOS Global Public Health 2023; 3:e0001325 20. 20.Kumeso VKB, Kalonji WM, Rembry S, Mordt OV, Tete DN, Prêtre A, Delhomme S, Kyhi MIW, Camara M, Catusse J, et al. Efficacy and safety of acoziborole in patients with human African trypanosomiasis caused by Trypanosoma brucei gambiense: a multicentre, open-label, single-arm, phase 2/3 trial. The Lancet Infectious Diseases 2023; 23:463–70 21. 21.Geerts M, Van Reet N, Leyten S, Berghmans R, Rock K, Coetzer T, Eyssen LA, and Büscher P. *Trypanosoma brucei gambiense*-iELISA: A promising new test for the post-elimination monitoring of human African trypanosomiasis. Clinical Infectious Diseases 2020 Aug; 73:e2477–e2483. doi: doi:10.1093/cid/ciaa1264. eprint: [https://academic.oup.com/cid/article-pdf/73/9/e2477/41059326/ciaa1264.pdf](https://academic.oup.com/cid/article-pdf/73/9/e2477/41059326/ciaa1264.pdf). Available from:10.1093/cid/ciaa1264 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/cid/ciaa1264&link_type=DOI) 22. 22.Priotto G, Franco JR, Lejon V, Büscher P, Matovu E, Ndung’u J, Biéler S, Mumba D, Van Reet N, Verlé P, et al. WHO target product profiles: four diagnostic tests needed in the effort to eliminate African trypanosomiasis. Bulletin of the World Health Organization 2023; 101:546 23. 23.Davis CN, Crump RE, Sutherland SA, Spencer SEF, Corbella A, Chansy S, Lebuki J, Miaka EM, and Rock KS. Comparison of stochastic and deterministic models for *gambiense* sleeping sickness at different spatial scales: A health area analysis in the DRC. MedRxiv 2023 24. 24.American Red Cross. DRC Health Zone and Health Area boundaries. 2020 Feb. Available from: [https://data.humdata.org/dataset/drc-health-data](https://data.humdata.org/dataset/drc-health-data) 25. 25.Checchi F, Funk S, Chandramohan D, Haydon DT, and Chappuis F. Updated estimate of the duration of the meningo-encephalitic stage in gambiense human African trypanosomiasis. BMC Research Notes 2015; 8:8–10. doi: doi:10.1186/s13104-015-1244-3 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s13104-015-1244-3&link_type=DOI) 26. 26.1. Ndung’u JM Snijders R, Fukinsia A, Claeys Y, Mpanya A, Hasker E, Meheus F, Miaka E, and Boelaert M. Cost of a new method of active screening for human African trypanosomiasis in the Democratic Republic of the Congo. PLOS Neglected Tropical Diseases 2020 Dec; 14. Ed. by Ndung’u JM:e0008832. doi: doi:10.1371/journal.pntd.0008832. Available from: [https://dx.plos.org/10.1371/journal.pntd.0008832](https://dx.plos.org/10.1371/journal.pntd.0008832) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0008832&link_type=DOI) 27. 27.Snijders R, Fukinsia A, Claeys Y, Hasker E, Mpanya A, Miaka E, Meheus F, and Boelaert M. Costs and outcomes of integrated human African trypanosomiasis surveillance system using rapid diagnostic tests, Democratic Republic of the Congo. Emerging Infectious Diseases 2021; 27:2144–53. doi: doi:10.3201/eid2708.202399 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3201/eid2708.202399&link_type=DOI) 28. 28.Mahamat MH, Peka M, Rayaisse JB, Rock KS, Toko MA, Darnas J, Brahim GM, Alkatib AB, Yoni W, Tirados I, Courtin F, Brand SP, Nersy C, Alfaroukh IO, Torr SJ, Lehane MJ, and Solano P. Adding tsetse control to medical activities contributes to decreasing transmission of sleeping sickness in the Mandoul focus (Chad). PLoS Neglected Tropical Diseases 2017; 11:1–16. doi: doi:10.1371/journal.pntd.0005792 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0005792&link_type=DOI) 29. 29.Rock KS, Huang CI, Crump RE, Bessell PR, Brown PE, Tirados I, Solano P, Antillon M, Picado A, Mbainda S, Darnas J, Crowley EH, Torr SJ, and Peka M. Update of transmission modelling and projections of gambiense human African trypanosomiasis in the Mandoul focus, Chad. Infectious Diseases of Poverty 2022; 11:1–13. doi: doi:10.1186/s40249-022-00934-8. Available from: 10.1186/s40249-022-00934-8 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s40249-022-00934-8&link_type=DOI) 30. 30.Courtin F, Camara M, Rayaisse JB, Kagbadouno M, Dama E, Camara O, Traoré IS, Rouamba J, Peylhard M, Somda MB, Leno M, Lehane MJ, Torr SJ, Solano P, Jamonneau V, and Bucheton B. Reducing human-tsetse contact significantly enhances the efficacy of sleeping sickness active screening campaigns: A promising result in the context of elimination. PLOS Neglected Tropical Diseases 2015; 9:1–12. doi: doi:10.1371/journal.pntd.0003727 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0003727&link_type=DOI) 31. 31.Office for the Coordination of Humanitarian Affairs. Journees Nationales de Vaccination (JNV) activities de vaccination supplementaire, RDC. Available from: [https://data.humdata.org/dataset/rdc-statistiques-des-populations](https://data.humdata.org/dataset/rdc-statistiques-des-populations) 32. 32.1. Matovu E Franco JR, Cecchi G, Priotto G, Paone M, Diarra A, Grout L, Simarro PP, Zhao W, and Argaw D. Monitoring the elimination of human African trypanosomiasis at continental and country level: Update to 2018. PLOS Neglected Tropical Diseases 2020 May; 14. Ed. by Matovu E:e0008261. doi: doi:10.1371/journal.pntd.0008261. Available from: [https://dx.plos.org/10.1371/journal.pntd.0008261](https://dx.plos.org/10.1371/journal.pntd.0008261) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0008261&link_type=DOI) 33. 33.Jamonneau V, Camara O, Ilboudo H, Peylhard M, Koffi M, Sakande H, N’Dri L, Sanou D, Dama E, Camara M, and Lejon V. Accuracy of Individual Rapid Tests for Serodiagnosis of Gambiense Sleeping Sickness in West Africa. PLOS Neglected Tropical Diseases 2015; 9:1–9. doi: doi:10.1371/journal.pntd.0003480 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0003727&link_type=DOI) 34. 34.Lutumba P, Meheus F, Robays J, Miaka C, Kande V, Büscher P, Dujardin B, and Boelaert M. Cost-effectiveness of Algorithms for Confirmation Test of Human African Trypanosomiasis. Emerging Infectious Diseases 2007 Oct; 13:1484–90. doi: doi:10.3201/eid1310.060358. Available from: http://wwwnc.cdc.gov/eid/article/13/10/06-0358_article.htm [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3201/eid1310.060358&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=18257991&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 35. 35.1. Arez AP Bessell PR, Lumbala C, Lutumba P, Baloji S, Biéler S, and Ndung’u JM. Cost-effectiveness of using a rapid diagnostic test to screen for human African trypanosomiasis in the Democratic Republic of the Congo. PLOS ONE 2018 Sep; 13. Ed. by Arez AP:e0204335. doi: doi:10.1371/journal.pone.0204335. Available from: [https://dx.plos.org/10.1371/journal.pone.0204335](https://dx.plos.org/10.1371/journal.pone.0204335) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pone.0204335&link_type=DOI) 36. 36.Keating J, Yukich JO, Sutherland CS, Woods G, and Tediosi F. Human African trypanosomiasis prevention, treatment and control costs: A systematic review. Acta Tropica 2015 Oct; 150:4–13. doi: doi:10.1016/j.actatropica.2015.06.003. Available from: 10.1016/j.actatropica.2015.06.003%20[https://linkinghub.elsevier.com/retrieve/pii/S0001706X15300267](https://linkinghub.elsevier.com/retrieve/pii/S0001706X15300267) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.actatropica.2015.06.003&link_type=DOI) 37. 37.Irurzun-Lopez M, Erondu NA, Djibo A, Griffiths U, Stuart JM, Fernandez K, Ronveaux O, Le Gargasson JB, Gessner BD, and Colombini A. The actual and potential costs of meningitis surveillance in the African meningitis belt: Results from Chad and Niger. Vaccine 2016; 34:1133–8. doi: doi:10.1016/j.vaccine.2015.10.045. Available from: 10.1016/j.vaccine.2015.10.045 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.vaccine.2015.10.045&link_type=DOI) 38. 38.Eperon G, Schmid C, Loutan L, and Chappuis F. Clinical presentation and treatment outcome of sleeping sickness in Sudanese pre-school children. Acta Tropica 2006 Jan; 101:31–9. doi: doi:10.1016/j.actatropica.2006.12.002. Available from: [https://linkinghub.elsevier.com/retrieve/pii/S0001706X06002439](https://linkinghub.elsevier.com/retrieve/pii/S0001706X06002439) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.actatropica.2006.12.002&link_type=DOI) 39. 39.Schmid C, Kuemmerle A, Blum J, Ghabri S, Kande V, Mutombo W, Ilunga M, Lumpungu I, Mutanda S, Nganzobo P, Tete D, Mubwa N, Kisala M, Blesson S, and Mordt OV. In-Hospital Safety in Field Conditions of Nifurtimox Eflornithine Combination Therapy (NECT) for T. b. gambiense Sleeping Sickness. PLOS Neglected Tropical Diseases 2012; 6. doi: doi:10.1371/journal.pntd.0001920 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0001920&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23209861&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 40. 40.Mesu VKBK, Kalonji WM, Bardonneau C, Mordt OV, Blesson S, Simon F, Delhomme S, Bernhard S, Kuziena W, Lubaki JPF, Vuvu SL, Ngima PN, Mbembo HM, Ilunga M, Bonama AK, Heradi JA, Solomo JLL, Mandula G, Badibabi LK, Dama FR, Lukula PK, Tete DN, Lumbala C, Scherrer B, Strub-Wourgaft N, and Tarral A. Oral fexinidazole for late-stage African Trypanosoma brucei gambiense trypanosomiasis: a pivotal multicentre, randomised, non-inferiority trial. The Lancet 2018; 391:144–54. doi: doi:10.1016/S0140-6736(17)32758-7 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(17)32758-7&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=29113731&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 41. 41.Priotto G, Fogg C, Balasegaram M, Erphas O, Louga A, Checchi F, Ghabri S, and Piola P. Three Drug Combinations for Late-Stage Trypanosoma brucei gambiense Sleeping Sickness: A Randomized Clinical Trial in Uganda. PLOS Clinical Trials 2006; 1:e39. doi: doi:10.1371/journal.pctr.0010039 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pctr.0010039&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17160135&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 42. 42.Priotto G, Kasparian S, Ngouama D, Ghorashian S, Arnold U, Ghabri S, and Karunakara U. Nifurtimox-Eflornithine Combination Therapy for Second-Stage Trypanosoma brucei gambiense Sleeping Sickness: A Randomized Clinical Trial in Congo. Clinical Infectious Diseases 2007; 45:1435–42. doi: doi:10.1086/522982 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1086/522982&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17990225&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000251080800006&link_type=ISI) 43. 43.Checchi F, Piola P, Ayikoru H, Thomas F, Legros D, and Priotto G. Nifurtimox plus eflornithine for late-stage sleeping sickness in Uganda: A case series. PLOS Neglected Tropical Diseases 2007; 1:1–6. doi: doi:10.1371/journal.pntd.0000064 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0000064&link_type=DOI) 44. 44.Priotto G, Kasparian S, Mutombo W, Ngouama D, Ghorashian S, Arnold U, Ghabri S, Baudin E, Buard V, Kazadi-Kyanza S, Ilunga M, Mutangala W, Pohlig G, Schmid C, Karunakara U, Torreele E, and Kande V. Nifurtimox-eflornithine combination therapy for second-stage African Trypanosoma brucei gambiense trypanosomiasis: a multicentre, randomised, phase III, non-inferiority trial. The Lancet 2009; 374:56–64. doi: doi:10.1016/S0140-6736(09)61117-X. Available from: 10.1016/S0140-6736(09)61117-X [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(09)61117-X&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19559476&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000267768800034&link_type=ISI) 45. 45.Ngoyi DM, Lejon V, Pyana P, Boelaert M, Ilunga M, Menten J, Mulunda JP, Van Nieuwenhove S, Muyembe Tamfum JJ, and Büscher P. How to Shorten Patient Follow-Up after Treatment for Trypanosoma brucei gambiense Sleeping Sickness. The Journal of Infectious Diseases 2010; 201:453–63. doi: doi:10.1086/649917 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1086/649917&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20047500&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 46. 46.Burri C, Yeramian PD, Allen JL, Merolle A, Serge KK, Mpanya A, Lutumba P, Mesu VKBK, Bilenge CMM, Lubaki JPF, Mpoto AM, Thompson M, Munungu BF, Manuel F, Josenando T, Bernhard SC, Olson CA, Blum J, Tidwell RR, and Pohlig G. Efficacy, Safety, and Dose of Pafuramidine, a New Oral Drug for Treatment of First Stage Sleeping Sickness, in a Phase 2a Clinical Study and Phase 2b Randomized Clinical Studies. PLOS Neglected Tropical Diseases 2016; 10:1–18. doi: doi:10.1371/journal.pntd.0004362 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0004328&link_type=DOI) 47. 47.Pohlig G, Bernhard SC, Blum J, Burri C, Mpanya A, Lubaki JPF, Mpoto AM, Munungu BF, N’tombe PM, Deo GKM, Mutantu PN, Kuikumbi FM, Mintwo AF, Munungi AK, Dala A, Macharia S, Bilenge CMM, Mesu VKBK, Franco JR, Dituvanga ND, Tidwell RR, and Olson CA. Efficacy and Safety of Pafuramidine versus Pentamidine Maleate for Treatment of First Stage Sleeping Sickness in a Randomized, Comparator-Controlled, International Phase 3 Clinical Trial. PLOS Neglected Tropical Diseases 2016; 10:1–17. doi: doi:10.1371/journal.pntd.0004363 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0004328&link_type=DOI) 48. 48.Kansiime F, Adibaku S, Wamboga C, Idi F, Kato CD, Yamuah L, Vaillant M, Kioy D, Olliaro P, and Matovu E. A multicentre, randomised, non-inferiority clinical trial comparing a nifurtimox-eflornithine combination to standard eflornithine monotherapy for late stage Trypanosoma brucei gambiense human African trypanosomiasis in Uganda. Parasites and Vectors 2018; 11:1–11. doi: doi:10.1186/s13071-018-2634-x [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s13071-018-3228-3)&link_type=DOI) 49. 49.WHO Department of Control of Neglected Tropical Diseases. WHO interim guidelines for the treatment of gambiense human African trypanosomiasis. Tech. rep. Geneva, Switzerland: World Health Organization, 2019. Available from: [https://www.who.int/trypanosomiasis\_african/resources/9789241550567/en/](https://www.who.int/trypanosomiasis_african/resources/9789241550567/en/) 50. 50.Alirol E, Schrumpf D, Amici Heradi J, Riedel A, De Patoul C, Quere M, and Chappuis F. Nifurtimox-eflornithine combination therapy for second-stage gambiense human African trypanosomiasis: Médecins Sans Frontières experience in the Democratic Republic of the Congo. Clinical Infectious Diseases 2013; 56:195–203. doi: doi:10.1093/cid/cis886 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/cid/cis886&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23074318&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 51. 51.Balasegaram M, Harris S, Checchi F, Hamel C, and Karunakara U. Treatment outcomes and risk factors for relapse in patients with early-stage human African trypanosomiasis (HAT) in the Republic of the Congo. Bulletin of the World Health Organization 2006; 84:777–82. doi: doi:10.2471/BLT.05.028399 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.2471/BLT.05.028399&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17128357&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 52. 52.Bastide S, Priotto G, Ecochard R, and Etard JF. Effectiveness of short vs. long treatment schedules with pentamidine in first-stage HAT: a large field cohort study. Tropical Medicine & International Health 2011; 16:68–9. Available from: [https://onlinelibrary.wiley.com/toc/13653156/2011/16/s1](https://onlinelibrary.wiley.com/toc/13653156/2011/16/s1) 53. 53.Bertram MY, Stenberg K, Brindley C, Li J, Serje J, Watts R, and Edejer TTT. Disease control programme support costs: an update of WHO-CHOICE methodology, price databases and quantity assumptions. Cost Effectiveness and Resource Allocation 2017 Dec; 15:21. doi: doi:10.1186/s12962-017-0083-6. Available from: [http://resource-allocation.biomedcentral.com/articles/10.1186/s12962-017-0083-6](http://resource-allocation.biomedcentral.com/articles/10.1186/s12962-017-0083-6) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s12962-017-0083-6&link_type=DOI) 54. 54.Stenberg K, Lauer JA, Gkountouras G, Fitzpatrick C, and Stanciole A. Econometric estimation of WHO-CHOICE country-specific costs for inpatient and outpatient health service delivery. Cost Effectiveness and Resource Allocation 2018; 16:1–15. doi: doi:10.1186/s12962-018-0095-x. Available from: 10.1186/s12962-018-0095-x [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s12962-018-0095-x&link_type=DOI) 55. 55.Simarro PP, Diarra A, Ruiz Postigo JA, Franco JR, and Jannin JG. The human African trypanosomiasis control and surveillance programme of the World Health Organization 2000-2009: the way forward. eng. PLOS neglected tropical diseases 2011 Feb; 5:e1007. doi: doi:10.1371/journal.pntd.0001007 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0001007&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21364972&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 56. 56.Snijders R, Shaw APM, Selby R, Tirados I, Bessell PR, Fukinsia A, Miaka E, Tediosi F, Hasker E, and Antillon M. The cost of sleeping sickness vector control in the Democratic Republic of the Congo. medRxiv 2024 Jan :2024.02.02.24302172. doi: doi:10.1101/2024.02.02.24302172. Available from: [http://medrxiv.org/content/early/2024/02/03/2024.02.02.24302172.abstract](http://medrxiv.org/content/early/2024/02/03/2024.02.02.24302172.abstract) [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoibWVkcnhpdiI7czo1OiJyZXNpZCI7czoyMToiMjAyNC4wMi4wMi4yNDMwMjE3MnYxIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjQvMDMvMzAvMjAyNC4wMy4yOS4yNDMwNTA2Ni5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 57. 57.Hasker E, Lutumba P, Chappuis F, Kande V, Potet J, De Weggheleire A, Kambo C, Depoortere E, Pécoul B, and Boelaert M. Human African Trypanosomiasis in the Democratic Republic of the Congo: A Looming Emergency? PLOS Neglected Tropical Diseases 2012; 6:e1950. doi: doi:10.1371/journal.pntd.0001950 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pntd.0001950&link_type=DOI) 58. 58.United Nations Population Division. World Population Prospects 2019. 2019. Available from: [https://population.un.org/wpp/](https://population.un.org/wpp/) 59. 59.GBD 2017 DALYs and HALE Collaborators. Global, regional, and national disability-adjusted life-years (DALYs) for 359 diseases and injuries and healthy life expectancy (HALE) for 195 countries and territories, 1990-2017: A systematic analysis for the Global Burden of Disease Study 2017. The Lancet 2018; 392:1859–922. doi: doi:10.1016/S0140-6736(18)32335-3 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(18)32335-3&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=30415748&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) 60. 60.Briggs AH, Goeree R, Blackhouse G, and O’Brien BJ. Probabilistic Analysis of Cost-Effectiveness Models: Choosing between Treatment Strategies for Gastroesophageal Reflux Disease. Medical Decision Making 2002 Aug; 22:290–308. doi: doi:10.1177/0272989X0202200408. Available from: [http://mdm.sagepub.com/cgi/doi/10.1177/0272989X0202200408](http://mdm.sagepub.com/cgi/doi/10.1177/0272989X0202200408) [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1177/0272989X0202200408&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=12150595&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F03%2F30%2F2024.03.29.24305066.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000176914100001&link_type=ISI)