(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



An economic evaluation of Wolbachia deployments for dengue control in Vietnam [1]

['Hugo C. Turner', 'Mrc Centre For Global Infectious Disease Analysis', 'School Of Public Health', 'Imperial College London', 'London', 'United Kingdom', 'Duong Le Quyen', 'World Mosquito Program', 'Ho Chi Minh City', 'Reynold Dias']

Date: 2023-06

From the health sector perspective, the Wolbachia intervention was projected to cost US$420 per disability-adjusted life year (DALY) averted. From the societal perspective, the overall cost-effectiveness ratio was negative, i.e. the economic benefits outweighed the costs. These results are contingent on the long-term effectiveness of Wolbachia releases being sustained for 20 years. However, the intervention was still classed as cost-effective across the majority of the settings when assuming only 10 years of benefits

Ten settings within Vietnam were identified as priority locations for potential future Wolbachia deployments (using a population replacement strategy). The effectiveness of Wolbachia deployments in reducing the incidence of symptomatic dengue cases was assumed to be 75%. We assumed that the intervention would maintain this effectiveness for at least 20 years (but tested this assumption in the sensitivity analysis). A cost-utility analysis and cost-benefit analysis were conducted.

Dengue is a major public health challenge and a growing problem due to climate change. The release of Aedes aegypti mosquitoes infected with the intracellular bacterium Wolbachia is a novel form of vector control against dengue. However, there remains a need to evaluate the benefits of such an intervention at a large scale. In this paper, we evaluate the potential economic impact and cost-effectiveness of scaled Wolbachia deployments as a form of dengue control in Vietnam–targeted at the highest burden urban areas.

Dengue is a major public health challenge and a growing problem due to climate change. The release of Aedes aegypti mosquitoes infected with the intracellular bacterium Wolbachia is a novel form of vector control against dengue. However, there remains a need to evaluate the health and economic benefits of such an intervention at a large scale, as well as its value for money. In this paper, we evaluate the potential economic impact and cost-effectiveness of scaled Wolbachia deployments as a form of dengue control in Vietnam–targeted at the highest burden urban areas. Ten settings within Vietnam were identified as priority locations for potential future Wolbachia deployments (using a population replacement strategy). We assumed that the effectiveness of Wolbachia deployments in reducing the incidence of symptomatic dengue cases would be 75%. We found that targeting high burden cities with Wolbachia deployments would be a cost-effective intervention in Vietnam and generate notable broader benefits besides health gains. Overall, this work highlights the value of investment in the scaled implementation of Wolbachia deployments as an effective and cost-effective tool for dengue control in Vietnam, and more generally for addressing the global challenge of dengue control.

Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: HCT received funding from the World Mosquito Program to conduct this analysis. DLQ, RD, CPS and KLA are employees of the World Mosquito Program. PTH has declared that no competing interests exist.

Funding: HCT has received funding from the World Mosquito Program (WMP) to conduct this analysis. HCT acknowledges funding from the MRC Centre for Global Infectious Disease Analysis (reference MR/R015600/1), jointly funded by the UK Medical Research Council (MRC) and the UK Foreign, Commonwealth & Development Office (FCDO), under the MRC/FCDO Concordat agreement and is also part of the EDCTP2 programme supported by the European Union. KLA, CPS, RD and DLQ acknowledge funding from the Wellcome Trust for this work (224459/Z/21/Z). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2023 Turner et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Dengue has been estimated to cause a substantial health and economic burden in Vietnam [ 11 ]. In Vietnam, wMel Wolbachia mosquito releases have been undertaken in several demonstration projects [ 12 ]. However, there remains a need to evaluate the health and economic benefits of such an intervention at a large scale, as well as its value for money.

The World Mosquito Program (WMP) has partnered with governments and communities to deploy Wolbachia mosquitoes in 11 countries since 2011 [ 3 ]. A number of randomised and non-randomised field trials have been conducted [ 4 – 8 ] demonstrating successful establishment of wMel in local Ae. aegypti populations and the efficacy of the intervention in controlling dengue and other Aedes-borne diseases. The Vector Control Advisory Group of the World Health Organization concluded in December 2020 that the evidence for wMel introgression demonstrates public health value against dengue [ 9 ]. A past economic evaluation by Brady et al. [ 10 ] found that Wolbachia deployments (using a replacement intervention) were likely to be cost-effective for controlling dengue in urban areas of Indonesia.

The release of Aedes aegypti mosquitoes infected with the intracellular bacterium Wolbachia is a novel strategy for the control of dengue and other Aedes-borne diseases [ 2 ]. Mosquitoes infected with Wolbachia (wMel strain) 1) are less likely to transmit dengue, chikungunya, Zika, and yellow fever viruses [ 2 ] and 2) can suppress or replace the natural mosquito population due to fatal cytoplasmic incompatibility among Wolbachia wild-type mating pairs [ 2 ]. These mosquitoes can, therefore, be used to replace the existing Ae. aegypti population with a lower competence phenotype by releasing both females and males (known as an introgression or replacement intervention).

All costs and benefits are given in 2020 US$ prices and were discounted at 3% per year (in line with WHO-CHOICE recommendations) [ 20 ]. The different outputs and perspectives considered are summarised in Box 1 . The cost-effectiveness ratios were compared to a cost-effectiveness threshold of 0.5 times Vietnam’s GDP per capita (i.e. <US$1,760 per DALY averted) [ 21 , 22 ]. A CHEERS checklist [ 23 ] is provided in the S1 Checklist file. Univariate sensitivity analysis on several parameters summarised in Table 2 .

For the base case results, we assumed that the intervention would have its full impact for 20 years once the intervention becomes effective—six months after the completion of releases (therefore the total time horizon of the analysis was 22 years, to account for the 18-month preparation and release phases). We assumed the intervention was implemented in each setting independently, but for simplicity no specific assumptions were made on the sequence or timeline of releases across the ten settings.

To evaluate the intervention a scenario with scaled Wolbachia deployments in the selected settings was compared incrementally to an alternative scenario of continuing existing dengue control measures (i.e. the comparator was the status quo). It was assumed that in the absence of the intervention the baseline number of dengue cases occurring would increase annually based on the population growth rate of 1.14% (the average for Vietnam) [ 19 ]. The calculations compared the number of dengue cases (and their associated health and economic burden) that would be occurring under the status quo comparator compared to the number projected to be occurring in the presence of Wolbachia deployments. This was calculated with a static based approach, as the indirect effects of the intervention were accounted for within the population-level effectiveness metric.

We assume the Wolbachia intervention’s effect is area-wide and benefits the whole population within the administrative district boundary (see S1 Appendix ) and that the benefit of the intervention was equally experienced among the targeted populations.

Based on data from a cluster randomized trial and quasi-experimental studies in Yogyakarta, the effectiveness of Wolbachia deployments in reducing the incidence of symptomatic cases was assumed to be 75% [ 5 , 7 ]. It was further assumed that there is a lag of six months after the completion of releases before Wolbachia reaches a stably high prevalence in the local Ae. aegypti population, and for the full intervention effect to accrue.

Based on these assumptions the average total cost per person covered was assumed to be US$8.56 (discounted at 3% per year and expressed in 2020 US$ prices) (Table D in S1 Appendix ). The total implementation costs were based on the population within the release area for each setting. It was assumed that the full cost of the intervention was incurred by the health care provider and that it was not influenced by the chosen perspective.

The costs for preparation, release and short-term monitoring phases were informed by the WMP’s accounts from two project sites within Vietnam in 2022 (see S1 Appendix ). The costs of the long-term monitoring phase were based on WMP’s implementations in other countries (Table D in S1 Appendix ). These long-term monitoring functions are assumed to be integrated into routine public health activities and require no or only very occasional procurement of additional resources.

The investigated economic burden had two components. The first was the cost of illness associated with the dengue cases (this was stratified by direct medical costs, direct non-medical costs, and productivity costs). The second was the costs related to the government’s standard dengue prevention and control activities. Further details are provided in the supporting information (Tables B and C in S1 Appendix ).

The DALY burden of non-fatal cases was estimated using the overall disability weights from Zeng et al. [ 18 ] (Table A in S1 Appendix ). It was assumed that a self-managed case had the same disability weight as an outpatient case. The number of years of life lost per fatal case was estimated from the GBD 2019 study [ 13 ].

The total number of symptomatic dengue cases were broken down into the following severity categories; sought no formal treatment, outpatient cases, hospitalized cases, and fatal cases ( Table 1 ), based on the distribution reported previously from Indonesia [ 17 ] as no empirical data on this distribution were available from Vietnam.

The assumed overall incidence of dengue in Vietnam was taken from the model-based estimates from the Global burden of disease (GBD) 2019 study (1,047,320 symptomatic dengue cases occurring in Vietnam) [ 13 ]. We based the relative sub-national distribution of this country-level incidence on the mapping estimates of Bhatt et al. [ 14 ].

Ten high dengue burden urban settings within Vietnam, including four metropolitan and six provincial cities, were identified as priority locations for potential future Wolbachia deployments (described in and S1 Appendix ). These settings accounted for 38% of the dengue cases notified in Vietnam between 2016 and 2019, and approximately a quarter of the national population.

Cost per person covered: Intuitively the projected average cost-effectiveness and cost-benefit of the Wolbachia deployments were sensitive to assumptions regarding the cost per person covered ( S7 Table ). This highlights that opportunities for further reductions in the cost of the intervention would make it even more cost-effective.

Annual growth in case numbers: For the base case results we assumed a modest 1.14% increase in the baseline case burden of dengue (based on the average population growth rate for Vietnam). However, under a scenario of a higher increased per capita dengue incidence over time, this notably increases the projected value for money of the intervention

The DALY disability weights and the inclusion of persistent symptoms: Using disability weights that only accounted for acute symptoms of dengue notably increased the cost-effectiveness ratio from the health sector perspective. However, from a societal perspective, the ratios remained negative for the majority of the settings.

Time horizon and duration of the effectiveness: If the intervention was assumed to only generate 10 years of benefits, the projected long-term impact decreased. Under this scenario, the ICERs in the majority of individual settings as well as the overall cost-effectiveness ratios (i.e. when aggregated across all of the settings) remained under the cost-effectiveness threshold ( S6 Table ). However, under this assumption for some of the individual settings, the ICERs exceeded the cost-effectiveness threshold ( S6 Table ).

We performed univariate sensitivity analysis on several parameters ( Table 2 ). The results of this are shown in Fig 2 and S5 – S7 Tables. The results were most sensitive to the following parameters:

The projected cost per DALY averted was notably influenced by the perspective of the analysis ( Table 3 ). That said, regardless of the perspective used, all of the overall cost-effectiveness ratios (i.e. when aggregated across all of the settings) were under the cost-effectiveness threshold (<US$1,760 per DALY averted) ( S4 Table ). From the health sector perspective, the cost per DALY averted was US$420. From a societal perspective, the cost-effectiveness ratio was negative (i.e. the economic benefits outweighed the cost)–even when excluding the productivity gains related to prevented excess mortality. The societal benefit-cost ratio was 1.75, meaning that the valued economic benefits are larger than the cost of the intervention over a 22-year horizon ( Table 3 and Fig 1 ).

Assuming that the intervention has 20 years of impact once Wolbachia is established in the local mosquito population, we estimated that the intervention would avert 6.2 million dengue cases including over 784,000 hospitalisations, result in 153,285 DALYs being averted (i.e. healthy life-years gained), and generate US$299 million in economic benefits ( S3 Table ). The majority of the benefits were experienced by the patients (US$245 million) but the benefit to the health system was also notable (US $54 million). Setting-specific benefits are shown in S3 Table .

The discounted total cost of Wolbachia implementation across all ten settings was projected to be US$171.3 million. The cost for each of the settings is shown in S3 Table . The majority of the cost was incurred during the preparation and release phases, occurring within the first 2 years of the intervention ( Fig 1 and Table D in S1 Appendix ).

The total corresponding baseline cost of illness associated with the dengue cases totalled US$23.69 million per year ( S1 Table ). Half of this was from productivity costs. In addition to this cost of illness, an estimated US$809,105 was spent each year on the government’s current dengue prevention and control activities–giving a total baseline economic burden of US$24.50 million per year ( S1 Table ). 19% of the economic burden was incurred by the health system and 81% by the patients.

Our projected number of hospitalized cases is generally comparable with the number reported to the Ministry of Health (45,749 vs 36,754 cases per annum) ( S2 Table ). Although this is promising as an average, there was variation across the different settings in the consistency between model-based and reported dengue case numbers ( S2 Table ).

It was projected that across the ten cities, an average of 363,086 symptomatic dengue cases occurred each year at baseline of which 42 (0.01%) would be fatal ( S1 Table ). Stratified by severity, we estimated 232,738 (64%) of the cases sought no formal treatment, 84,599 (23%) cases only sought treatment at outpatient facilities and 45,749 (13%) cases would be hospitalized. The corresponding baseline DALY burden was 13,674 ( S1 Table ).

Discussion

Dengue is a major public health problem worldwide and its burden is expected to increase even further due to climate change and urbanization [1]. It is therefore vital that new dengue interventions are developed and evaluated [29,30], such as the introduction of Wolbachia into Ae. aegypti mosquito populations. Health economic analyses play an important role here in evaluating the costs and benefits of new interventions to investigate their value for money and inform health policy. In this analysis, we investigated the cost-effectiveness of releasing Wolbachia mosquitoes throughout ten high burden cities in Vietnam as a form of dengue control. These settings account for approximately 40% of the country’s reported national dengue burden and we estimated that the Wolbachia intervention could avert 6.2 million dengue cases over the next two decades, generating substantial health and economic benefits. Overall this was found to be a cost-effective intervention and these results support previous findings that Wolbachia deployments can be a cost-effective intervention when targeted to high burden urban settings [10].

We estimated that the total cost of implementing Wolbachia in these ten cities would be US$171.3 million. Importantly, the majority of this cost was incurred during the preparation and release phases occurring within the first two years of the intervention. In contrast, the benefits accrued over 20 years (Fig 1). Our cost projections for the Wolbachia deployments were driven by the projected cost per person covered from two WMP demonstration projects in the south of Vietnam and were intended to be conservative. However, it is important to note that there is a degree of uncertainty regarding what the long-term costs of Wolbachia deployments would be in practice if it was implemented programmatically at this scale, and such cost estimates are sensitive to assumptions regarding the level of technical and implementation support that would be required from WMP global staff for subsequent releases. There is potential for the costs of Wolbachia deployments to be reduced over time, through advances in mass mosquito production, economies of scale, and alternative implementation models.

The estimated cost-effectiveness ratios were highly dependent on the perspective of the analysis and what savings were considered (Box 1). This needs to be considered when comparing the results to other studies. Averaged across all ten settings, the cost per DALY averted was US$420 from the health sector perspective. In contrast, from a societal perspective (which accounts also for averted productivity losses), the cost-effectiveness ratios were negative for the majority of the settings, meaning that the projected economic benefits generated by the intervention over 20 years outweighed its cost. Note that there are issues around potentially double counting benefits when including productivity gains within cost-effectiveness ratios [27,28]. Due to this we also calculated the societal perspective excluding the productivity gains related to prevented excess mortality and the results were still favourable.

In terms of policy recommendations based on international standards for defining cost-effectiveness, the values were below the conservative 0.5 times GDP per capita threshold [21,22] in all of the settings when using the health sector perspective–and mostly negative (i.e. the economic benefits outweigh the cost) from a societal perspective (S4 Table). In addition, the overall average cost-effectiveness ratio using the health sector perspective was below the threshold of US$500 per DALY averted set by the third edition of the disease control priorities project to identify priority interventions for consideration in lower-middle-income countries [31]. The results were also favorable when compared to the cost-effectiveness of a diverse range of health interventions in Vietnam within the Global Health Cost Effectiveness Analysis (GH CEA) Registry [32]. This indicates that the intervention offers good value for money in this setting. The results were most sensitive to assumptions related to the duration of the impact of the intervention, the exclusion of persistent symptoms within the disability weights, the assumed growth in case numbers and the Wolbachia deployment costs.

Due to the sensitivity of these results to the assumed incidence of dengue and the cost of illness associated with dengue cases, the conclusions cannot be directly generalised to other country settings and further evaluations are needed. That said, the overall conclusions were similar to an economic evaluation of Wolbachia deployments in Indonesia [10], with the key difference being that the Indonesian study only considered benefits up to 10 years post completion of releases for their primary results whereas we assumed 20 years. This longer duration of benefit is supported by a modelling study [26] and by field evidence a decade after the first Wolbachia release sites in northern Australia showing stability of Wolbachia in the Ae. aegypti population and maintenance of its virus-blocking properties [33]. This assumption was varied in our sensitivity analysis, and even with a more conservative assumption of 10 years of benefits we predict that, overall, the Wolbachia intervention would remain cost-effective in Vietnam, though less so than predicted in the Indonesian setting. This could be because the cost per dengue case is higher in Indonesia, which therefore means averted cases generate larger savings to offset the cost of the intervention.

In addition, Box 2 highlights some of broader benefits of the intervention not accounted for within this analysis.

Box 2. Broader benefits of the intervention Our findings highlight the projected benefits to the health system of implementing Wolbachia for dengue control in Vietnam, through averted outpatient and inpatient dengue cases. There would also be notable social-economic benefits–including millions of productive days gained by both the patients and their caregivers. Because Wolbachia is deployed at a community level, all sub-populations benefit from the intervention, including socio-economically disadvantaged groups who are disproportionately affected by the economic burden of dengue. As well as the investigated benefits associated with controlling dengue related to averted cost of illnesses there are other potential economic benefits. For example, dengue can lead to lost revenue from tourism [34]. Controlling dengue with Wolbachia could therefore lead to economic benefits from increased tourism in a setting like Vietnam [11] and potentially offset some of the costs of the intervention. Dengue outbreaks are typically seasonal, with a notable proportion of the cases occurring over a period of several months. This means that hospitals, and particularly intensive care wards, may become congested during dengue outbreaks. It is possible that this could have negative consequences on care for patients with other conditions due to deterioration of overall service quality. In this economic evaluation, we have only considered the dengue burden avertable by Wolbachia deployments. However, Wolbachia mosquitoes are also refractory to Zika virus, chikungunya virus and, yellow fever virus [35,36]. This means that the overall public health impact of Wolbachia deployments will be larger, but the sporadic nature of these other epidemic arboviruses makes it challenging to quantify these projected benefits. It is also important to note that climate change is likely to expand the geographical distribution and transmission intensity of several vector-borne human infectious diseases–including dengue [37]. This, as well as population growth and other factors, could potentially increase the incidence of dengue and other epidemic arboviruses, increasing the burden that would be averted by Wolbachia deployments in Vietnam.

Although these findings are positive it is important to consider that there will be no one solution to controlling dengue and it remains vital to consider/evaluate other interventions (such as new vaccines as they become available [38]).

[END]
---
[1] Url: https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0011356

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.