Article Text

Modelling the health impact and cost-effectiveness of lymphatic filariasis eradication under varying levels of mass drug administration scale-up and geographic coverage
1. Christopher M Stone1,2,3,
2. Randee Kastner1,2,
3. Peter Steinmann1,2,
4. Nakul Chitnis1,2,
5. Marcel Tanner1,2,
6. Fabrizio Tediosi1,2
1. 1Swiss Tropical and Public Health Institute, Basel, Switzerland
2. 2Universität Basel, Basel, Switzerland
3. 3Department of Statistics, North Carolina State University, Raleigh, North Carolina, USA.
1. Correspondence to Dr Fabrizio Tediosi; Fabrizio.Tediosi{at}unibas.ch

## Abstract

Background A global programme to eliminate lymphatic filariasis (GPELF) is underway, yet two key programmatic features are currently still lacking: (1) the extension of efforts to all lymphatic filariasis (LF) endemic countries, and (2) the expansion of geographic coverage of mass drug administration (MDA) within countries. For varying levels of scale-up of MDA, we assessed the health benefits and the incremental cost-effectiveness ratios (ICERs) associated with LF eradication, projected the potential savings due to decreased morbidity management needs, and estimated potential household productivity gains as a result of reduced LF-related morbidity.

Methods We extended an LF transmission model to track hydrocele and lymphoedema incidence in order to obtain estimates of the disability adjusted life years (DALYs) averted due to scaling up MDA over a period of 50 years. We then estimated the ICERs and the cost-effectiveness acceptability curves associated with different rates of MDA scale-up. Health systems savings were estimated by considering the averted morbidity, treatment-seeking behaviour and morbidity management costs. Gains in worker productivity were estimated by multiplying estimated working days lost as a result of morbidity with country-specific per-worker agricultural wages.

Results Our projections indicate that a massive scaling-up of MDA could lead to 4.38 million incremental DALYs averted over a 50-year time horizon compared to a scenario which mirrors current efforts against LF. In comparison to maintaining the current rate of progress against LF, massive scaling-up of MDA—pursuing LF eradication as soon as possible—was most likely to be cost-effective above a willingness to pay threshold of US$71.5/DALY averted. Intensified MDA scale-up was also associated with lower ICERs. Furthermore, this could result in health systems savings up to US$483 million. Extending coverage to all endemic areas could generate additional economic benefits through gains in worker productivity between US$3.4 and US$14.4 billion.

Conclusions In addition to ethical and political motivations for scaling-up MDA rapidly, this analysis provides economic support for increasing the intensity of MDA programmes.

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/

## Statistics from Altmetric.com

### Key questions

• A key challenge for the elimination of lymphatic filariasis (LF) is the expansion of geographic coverage of mass drug administration programmes. Without intense scale-up, elimination of LF will require both more time and more treatments.

• Prior studies have not considered the cost-effectiveness associated with scaling-up geographic coverage of an eradication programme, while accounting for progress made to date in eliminating LF.

#### What are the new findings?

• The faster geographic coverage of mass drug administration programmes is brought to scale, the greater the health benefits will be in terms of disability adjusted life years averted (DALY).

### Potential health system savings and worker productivity losses

Unsurprisingly, reaching LF eradication sooner was found to correspond to increased health systems savings, due to decreased morbidity management, ranging from US$140 million (95% CrI US$53.8–US$260.3m) in the eradication I scenario to US$483 million (95% CrI US$219.1–US$902.6m) in eradication III (figure 6).

Figure 6

Cumulative cost savings and averted losses over 50 years associated with LF eradication scenarios. Left: potential cost savings to LF endemic health systems due to decreased need for morbidity management practices; right: averted productivity losses due to eradication. LF, lymphatic filariasis.

Potential savings to the health system, however, were dwarfed by possible gains in worker productivity, which ranged from approximately US$3.4 billion (95% CrI US$2.03–US$5.36bn) under the eradication I scenario to US$14.4 billion (95% CrI US$8.58–US$22.02 billion) in the eradication III scenario (figure 6). Importantly, all increased with increasing rates of MDA scale-up, further supporting the conclusion from the cost-effectiveness analysis.

## Discussion

LF could become the first vector-borne disease to be eradicated. While the GPELF has made notable progress thus far, in order to achieve eradication, the programme needs to be extended to several endemic countries. Moreover, if the goal of global elimination as a public health problem by 2020, as specified in the London Declaration,32 is to occur, the scale-up of MDA to cover all populations at risk needs to be greatly intensified.

Here, we estimate that the impact on the health burden due to LF will increase with the rate of MDA scale-up, since DALYs averted have a longer time period to accrue when transmission is interrupted earlier. This highlights the importance of measuring costs and benefits of interventions over a long time horizon, as well as the benefits of integrating disease transmission, economic and demographic models.

Intensifying the rate of MDA scale-up to eradicate LF is clearly supported on economic grounds. Our analysis suggests that above a willingness to pay threshold of US$71.5/DALY averted, pursuing eradication at the highest level of MDA scale-up is the most likely to provide the greatest net benefits and therefore provide the most value for money. To put this in perspective, a willingness to pay of US$150/DALY averted has been suggested for low and middle income countries as acceptable.33 While decision makers are not bound by this threshold, our analysis indicates that LF eradication would generally be considered cost-effective, assuming the rate of MDA scale-up is sufficient. If instantaneous scale-up (eradication III) is shown not to be feasible, the ICER of the eradication II scenario (rapid scale-up) remains low at US$121/DALY averted. Only at the slowest level of scale-up does the ICER fall above this threshold, adding further urgency to intensifying the rate of scale-up. Others have used the Gross National income per capita for low income counties of US$1035 as a threshold,34 by which measure all eradication scenarios are considered cost-effective. Cost-effectiveness as a measure of efficiency is typically applied to interventions or health programmes. Additionally, it has been suggested that less efficient programmes may be considered in the case of eradication (as opposed to disease control) programmes, due to a host of additional outcomes that are typically not captured in cost-effectiveness analyses.35 These could include the threats of resistance, insecurity of long-term funding, or implications for economic growth.36

Other considerations influence the cost-effectiveness of LF eradication. Depending on the perspective taken, the benefits that are expected to arise due to health systems savings and gains in worker productivity could be taken into account, which would further increase the dominance of the eradication III scenario. We did not consider certain aspects of morbidity management, such as the need for hydrocele surgeries, which would diminish over time as transmission is interrupted. The economic benefits of eradication could therefore be greater than estimated here. Likewise, our estimates of gains in productivity are likely conservative, because they were based on the time lost due to LF-related morbidity and agricultural wages, rather than on direct estimates of output and productivity loss (which are reported to be greater for LF, though data is scarce).29 By tracking morbidity only for hydrocele and lymphoedema, but not subclinical outcomes such as lymphatic dilation, or clinical manifestations such as ADL or tropical pulmonary eosinophilia, or a potential for excess mortality (either due to a lack of data or a lack of disability weights), we underestimate the true burden of disease.37 There are some epidemiological aspects that we did not consider, such as recrudescence of infections in areas following elimination due to migration. By ignoring this possibility, we made the implicit assumption that international movement among endemic populations was limited. Relaxing this assumption would require a metapopulation model and an investigation of human migration and commuting patterns in LF-endemic regions. However, previous studies in which similar mechanisms were considered have only added to the growing support for pursuing eradication.10 ,11

Further aspects which could interfere with the ability to maintain sufficiently high MDA coverage include insufficient political will, inadequate health infrastructure, logistical issues and the potential of systematic non-compliance. The development of drug resistance, as has been documented in animal systems,38 could also present complications. Further and equally important, in areas where W. bancrofti is co-endemic with L. loa, it remains to be seen how effective biannual distribution of ABZ by itself or together with long-lasting insecticidal nets will be. We have assumed that the strategy employed in these areas would be as effective as MDA with IVM and ABZ, and as unlikely to lead to resistance. However, if this is not the case, and an alternative strategy requires a larger investment or a prolonged campaign, the ICERs of the eradication scenarios will increase. We have likewise not accounted for any progress in interrupting LF transmission resulting from bed net programmes targeting malaria, although modelling suggests such methods are highly efficacious against LF.39 ,40 If vector control is going to be part of a strategy against LF in certain regions, cost estimates should likewise incorporate this intervention. Further, it is worth noting that our estimates of progress made to the current time were informed by the WHO PCT Databank, as described in Kastner et al.23 It has been pointed out that these self-reported values are sometimes overestimates of the true coverage.41 Accounting for this bias would likely decrease the ICERs of the eradication scenarios. Currently, data to improve on these estimates is unavailable but additional modelling work, more focused on individual districts based on local data, may be enlightening. Such work will be particularly valuable in identifying more effective strategies for dealing with endemic districts where progress seems to be lagging. Such strategies could potentially include novel technologies, or novel combinations, such as a proposed triple-drug treatment regime.42

Additionally, we assumed that endemic countries implemented MDA programmes for a fixed duration resulting in a high probability of achieving elimination (ie, where >97.5% of simulations reached elimination).23 A more dynamic decision process, whereby a shorter duration is followed by surveys and possible additional rounds of MDA until elimination is certified may be closer to reality, but beyond the scope of this global-level exercise.

Finally, our strategies assumed that all endemic countries included in the different scenarios are committed to elimination, and would not pursue a less ambitious goal, such as disease control only. It is plausible, however, for some countries to only target populations that live in moderate to high transmission zones, but not the greater number of people in low transmission areas where chronic disease is much less prevalent. A previous study indeed suggests that cost-effectiveness may improve if communities with microfilaria prevalence above 3.55% are first treated through a sequential strategy based first on control and a later shift of programme goals towards elimination.43 Ordering the treatment districts by intensity could thus lead to further increases in cost-effectiveness of our eradication scenarios.

In conclusion, this study suggests that eradication of LF is likely a cost-effective strategy and that if pursued, scaling up MDA as rapidly as feasible will result in increases in value.

View Abstract

## Footnotes

• Contributors CMS, RK and FT designed the study. CMS and NC developed the model. CMS and RK performed the analysis and wrote the first draft of this report. All authors contributed to interpretation of data, and writing of the report. All authors approved of the final version.

• Funding The Bill and Melinda Gates Foundation (grant number OPP1037660).

• Competing interests MT serves as board member for DNDi, and serves on the scientific advisory board of the Novartis Institute for Tropical Diseases.

• Provenance and peer review Not commissioned; externally peer reviewed.

• Handling editor Seye Abimbola.

## Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.