Balancing economic and epidemiological interventions in the early stages of pathogen emergence
This work develops a new epidemiological-economic approach to the epidemic control of emerging and reemerging diseases. This area has become central following the Covid outbreak in both disciplinary and interdisciplinary research for its large health policy implications.
The emergence of Covid-19 at the beginning of 2020 has had a profound impact on the world: over one hundred million people have been infected, two to five million have died, and the global economy is still recovering from the shocks produced by lockdowns imposed during the first year of the pandemic. Epidemiological models played a vital, but variable, role in determining national policy in response to Covid. The variability in the response of different nations was very much a consequence of there being no real quantitative understanding of the economic consequences of the epidemic and attempts to control it. Covid will not be the last virus to cross from wild reservoir species to humans. Novel pathogens enter human populations at least twice a year, and a major epidemic breaks out every four to five years (see e.g. Bernstein et al, 2022). It is therefore crucial to develop a novel epidemic-economic (epi-econ hereafter) approach to epidemic control policies with a close reproduction of the actual epidemic diffusion (main) features and a reasonable appraisal of the economic consequences of public health interventions. This task has been undertaken by an interdisciplinary group of Senior fellows of the Centre for Unframed Thinking (CUT) of Rennes School of Business led by Andy Dobson (Princeton, CUT Senior fellow), involving epidemiologists, ecologists, economists and mathematicians.
The epi-econ approach to Covid control
The epi-econ approach is extremely recent: it started in the late 90s with quite a few noticeable contributions till….the Covid crisis. As outlined above, the Covid outbreak raised and is still raising new important interdisciplinary questions, some of which are related to the intensive use of the so-called Non-Pharmaceutical Interventions (NPI) such as lockdowns, curfews or testing/isolating/tracing, not speaking about the sector-specific measures which have consisted in closing or strictly limiting the access to certain economic sectors (mostly in the tertiary sector). As a matter of fact, lockdowns of the magnitude and scale of Covid-times are unprecedented in the history of Humanity. One of the very few historical cases is the 27 kms long “plague wall” that was erected on the Vaucluse mountains in 1721 to protect the Comtat Venaissin from the diffusion of the plague which was raging at that time in Marseille (see below). Very far from the large scale lockdown measures which have been taken place in early Covid times, with the subsequent huge economic costs. This is precisely what led to the take-off of the epi-econ modelling since 2020.
Epi-econ models put together epidemiological models of disease diffusion and an economic block. The connection between the two components operates generally through the control policies: for example, lockdowns will reduce contagion, which is good for epidemic control, but the stronger and longer they are the more costly they reveal economically speaking. The epi-econ structure allows to determine the parameters of the lockdown policy which optimize the tradeoff between the human and economic costs.
The vast majority of the Covid epi-econ papers use very stylized epidemiological models which, in turn, lowers their relevance for control policy. Most of the latter literature uses indeed the so-called SIR (S for susceptible, I for infected and R for recovered) models which describe in a very rough way how epidemic diffusion operates between the 3 latter classes of individuals (using nowadays popular differential equations). This is of course interesting but does not fit enough into the Covid experience which is far more complex due, among others, to the role and size of the class of asymptomatic individuals. A second crucial shortcoming of this literature (see e.g. Acemoglu et al., 2021) is to assume that the authorities can adjust the parameters of, say, the lockdown policy in continuous time. In reality, this is far from possible, lockdown policy is only revised “occasionally”.
The “plague wall”
An enriched and more realistic epi-econ model
The two latter shortcomings have motivated the research conducted by Dobson et al. (2023). In particular, the epidemiological model has been considerably enriched to cope with some of the salient COVID features, which drastically increases the size of the model without obscuring the mechanisms at work. The model is represented below: it includes 7 epidemic classes (instead of the 3 in the earlier epi-econ work), two classes of exposed individuals, E1 and E2 , two classes of infected, I1 and I2 , and an additional category of contact persons, C. Exposed hosts, who are not yet infectious are classified as E1, while asymptomatic, contagious hosts are classified as E2. Infected hosts, I1, may become sick and get hospitalized, I2. These hosts have a higher mortality rate but are assumed to be in relative isolation and are thus unable to transmit the pathogen, except to unprotected health care workers. Class C are contacts of infectious hosts who do not develop infection. Contact tracing identifies C + E1 + E2 as contacts of infected hosts; testing is used to differentiate uninfected contacts, C, from exposed hosts (E1 and E2); the former can return to work, and the latter remain in isolation and go on to develop infection. This far more realistic Covid model is then calibrated using available data.
Next, the optimal policy choice is much wider, including testing/tracing/isolating in addition to lockdown, and also more realistic in that optimal policies are taken from a chosen discrete time to another as in reality, not in continuous time, the timing and duration of lockdown is optimized in this frame. Another important contribution of the paper is the joint optimization with respect to lockdown and testing/tracing/isolating. As such, one can evaluate a large number of policies, and identify the first-best.
Among many results, Dobson et al. (2023) obtain clear-cut findings for earliest epidemic stages where the only control tool available was either a lockdown or the laissez-faire response of doing nothing: when optimization takes into account economic and public health/mortality objectives, both economic damages and pathogen-induced mortality can be substantially reduced relative to the laissez-faire case if a lockdown is put in place as swiftly as possible. This is true even if we take into account the substantial adverse economic consequences resulting from a lockdown. The results also suggest that lockdown policies could be replaced by contact tracing and testing as soon as viable tests become available. Delays in initiating this transition in policy will always lead to higher economic damages and enhanced mortality. Finally, the desirable lockdown depth is clearly shown to be a function of the pathogen’s transmissibility. Accordingly, the duration of the lockdown would increase with the duration of infectivity. Establishing robust connections between pathogen characteristics and policy response is an important topic for future research.
The Dobson et al. epidemic model
This paper uses mathematical modelling of both the epidemiological and economic parts of the frame. The constructed epi-econ model is then calibrated on real data, and numerical optimization is finally used to identify the best policy-mix in terms of lockdown and testing/tracing/isolating policies.
Applications and beneficiaries
The model and methodology applied can be used (with or without- easy- adaptation) to assess epidemic control policies for a large number of pathogens. It’s therefore potentially useful for public health authorities and for practitioners in the field.
- Discipline: Epidemiology; economics; public health
- Keywords: Epidemic control; Non-Pharmaceutical Interventions (NPI); epidemic-economic assessment; Covid; optimal public health policy
- Reference to the research:
A.Dobson, C. Ricci, R. Boucekkine, F. Gozzi, G. Fabbri, T.Loch-Temzelides and M. Pascual (2023). Balancing economic and epidemiological interventions in the early stages of pathogen emergence (2023). Science Advances.9, Balancing economic and epidemiological interventions in the early stages of pathogen emergence | Science Advances
- Related useful references:
Acemoglu, D., V. Chernozhukov, I. Werning, M. D. Whinston (2021). Optimal targeted lockdowns in a multigroup SIR model. American Economic Review Insights 3, 487–502.
Bernstein, A. S., A. W. Ando, T. Loch-Temzelides, M. M. Vale, B. V. Li, H. Li, J. Busch, C. A. Chapman, M. Kinnaird, K. Nowak, M. C. Castro, C. Zambrana-Torrelio, J. A. Ahumada, L. Xiao, P. Roehrdanz, L. Kaufman, L. Hannah, P. Daszak, S. L. Pimm and A. P. Dobson (2022). “The costs and benefits of primary prevention of zoonotic pandemics (2022).” Science Advances 8(5): eabl4183.