Five Ways that COVID-19 Diagnostics Can Save Lives : Prioritizing Uses of Tests to Maximize Cost-Effectiveness

Public Disclosure Authorized

Public Disclosure AuthorizedPublic Disclosure Authorized

No. 43

February 23, 2021

Five Ways that COVID-19 Diagnoscs Can Save Lives: Priorizing Uses of Tests to Maximize Cost-Effecveness

Tristan Reed, William Waites, David Manheim, Damien de Walque, Chiara Vallini, Roberta Ga, and Timothy B. Halle

Supplies of diagnosc tests for SARS-CoV-2, the virus that causes COVID-19, are sll limited in many countries, and there is uncertainty about how to allocate the scarce supply across alternave types of tesng (use cases). This Research & Policy Brief quanfies the cost-effecveness of five alternave diagnosc use cases in terms of tests required per death averted. Across use cases, a single death can be averted by administering 940 to 8,838 tests, implying a large and posive return on investment in all use cases-even assuming a very low value for loss of life. That is, all five use cases pay for themselves many mes over. When prevalence of SARS-CoV-2 is high, the most cost-effecve uses of SARS-CoV-2 diagnoscs seem to be clinical triage of paents, at-risk worker screening, and populaon surveillance. Test-trace-isolate programs and border screening are also worthwhile, although they are more resource intensive per death averted if done comprehensively. These laer two intervenons become relavely more cost effecve when prevalence is low, and can stop the virus from entering a community completely. While governments should seek widespread deployment of tests in all five use cases, priorizing them in this way is likely to maximize the cost-effecveness of their use. As more contagious strains emerge, each use case will become more valuable than ever.

Introducon

SARS-CoV-2, like many infecous diseases, can be transmied from persons who are not obviously ill. This presents a major challenge because it means that a public health response cannot rely upon symptoms to track and control its spread (Fraser et al. 2004). Indeed, even if everyone immediately self-isolated upon the onset of symptoms, and the self-isolaon eliminated any risk, this would reduce transmission by, at most, 50 percent (Grassly et al. 2020)-and that is not enough to avert an epidemic of SARS-CoV-2.

Public Disclosure Authorized

Use case	Tests per	COVID-19	Tests	Tesng
death averted	deaths averted	required	populaon
1	Clinical triage and cohorng	940	106	100,000	100,000 paents upon admission
2	At-risk worker screening	1,042- 5,208	19-96	100,000	100,000 workers for one week
3	Populaon surveillance to	1,611	175	281,884	Regular samples per 100,000 for one year
trigger or avoid lockdown
4	Test-trace-isolate	4,459	392	1,763,485	Regular samples per 100,000 for one year
5	Border screening	8,838	11+	100,000	100,000 border crossers

first quarter of 2020. In sengs where households lack buffer stocks of food and savings, and must leave their homes for their livelihoods in the absence of social protecon, lockdowns may not be feasible and there may be greater reliance on a set of "social distancing" measures. Such measures may have an important effect, but the available evidence does not suggest they would be sufficient to avert an epidemic (Flaxman et al. 2020), especially as contagiousness rises with new strains.

Because vaccinang enough of the populaon to reach herd immunity will take me, in both high-income countries (where a majority are hesitant to get the vaccine immediately) (Galewitz 2021) and in middle-income and low-income countries (where supplies could be constrained unl 2024, according to a risk-assessment by COVAX (Beaumont 2020), a global facility to provide universal vaccine access), countries connue to implement nonpharmaceucal intervenons, including social distancing and full or paral lockdowns. Lockdowns can arrest epidemic spread in many sengs (Flaxman et al. 2020) but carry enormous economic costs-perhaps 0.25 to 0.86 percent of GDP per week (see Acemoglu et al. 2020; Alon et al. 2020; de Walque et al. 2020; Eichenbaum, Rebelo, and Trabandt 2020). Notably, China's GDP declined by 0.86 percent per week in the

Fortunately, diagnoscs can be used to test persons for the presence of current infecon. This opens new approaches to control transmission and ease the trade-off between economic and health concerns with targeted rather than general containment measures. In combinaon with a range of other concerted acvies (self-isolaon and nonpharmaceucal intervenons), they could contribute enormously to saving lives and minimizing costly lockdowns.

This Brief considers five use cases (among many) to pinpoint the means by which tesng for SARS-CoV-2 can contribute to saving lives during the COVID-19 pandemic (table 1). These cases are provided as quantave illustraons of what may be possible. The precise way in which tests can be best put to use in any parcular seng, and the actual benefits derived, are highly sensive to many factors specific to the seng and mode of use. The Brief considers those tests that

Table 1. Summary of Diagnosc Cost Effecveness by Use Case

Source: Authors' calculaons.

Note: Tests per death averted may not match rao of tests required to COVID-19 deaths averted due to rounding. All scenarios consider the incremental value of tesng, compared to a scenario where individuals are isolated based on symptoms alone. The major effect of border screening is in minimizing the introducon of virus to the country and so contributes to making all the other aspects of migang epidemic more likely to work. " + " indicates that the esmate of deaths averted refers only to the number of infecons among those quaranned, who were not already infected.

Affiliaons: Tristan Reed, World Bank Development Research Group; William Waites, University of Edinburgh; David Manheim, University of Haifa; Damien de Walque, World Bank Development Research Group; Chiara Vallini, Boston Consulng Group; Roberta Ga, World Bank Human Development Global Pracce; and Timothy B. Halle, Imperial College London and Modelling Guidance Group of The Global Fund for the ACT Covid-19 Diagnoscs Accelerator. For correspondence: timothyy..hhaallelet@[e-mail]@worlldbbaannkk.o.orgrg.

Acknowledgements: We are grateful to Norman Loayza, Young Eun Kim, and members ohfttpAs:C//wTwwA.ficncdedxl.eorrga/ctoovidr-1D9/iaactg-anccoesleractosr-pProigllreasrs/ for helpful comments, and Nancy Morrison for excellent editorial assistant.

Objecve and disclaimer: Research & Policy Briefs synthesize exisng research and data to shed light on a useful and interesng queson for policy debate. Research & Policy Briefs carry the names of the authors and should be cited accordingly. The findings, interpretaons, and conclusions are enrely those of the authors. They do not necessarily represent the views of the World Bank Group, its Execuve Directors, or the governments they represent.

detect infecon 90 percent of the me (test sensivity), and correctly idenfy negave cases 97 percent of the me (test specificity), in line with available angen-based rapid diagnosc tests. The details of each use case are considered in turn in the discussion that follows, although it is important to note that these cases do in fact interlock and depend upon one another.

The analysis also computes the cost-effecveness of each use case, as measured by "tests per death averted" (see results in table 1). The incremental impacts of diagnoscs are considered, over and above self-isolaon and social distancing. Given the low cost of a single test, the results suggest benefits that substanally exceed costs: a death can be averted for the cost of $4,700 to $44,190, depending on diagnosc use case, or the cost of 940 tests to 8,838 tests at $5 each. Following others who have quanfied a return on investment (ROI) with respect to the recommended value of a stascal life, the ROI of one test ranges between 3.5 (for the use case of border screening) and 41.6 (for the use case of clinical triage and cohorng) when assuming a low value of stascal life of only $200,000- which is 2 percent (1/50) of the esmate for the United States derived by Robinson et al. (2019). Higher valuaons would lead to even higher returns.

In certain circumstances (described in the discussion that follows), these returns are in addion to other major benefits, such as avoiding lockdowns and increasing confidence among consumers and workers.

These esmates tend to be conservave, as they do not incorporate several factors that are likely to be important. First, several of the calculaons consider only the effect of the intervenon on reducing transmission from a single person, and do not account for the effect that this may have in stemming a whole chain of transmission. The analysis focuses only on deaths, whereas SARS-CoV-2 also causes substanal illness, and long stays in hospital for many paents are costly and may overwhelm the health system, increasing the risk of death overall. The benefits of controlling hospital-acquired (nosocomial) infecons- by placing paents known to be infected with SARS-CoV-2 in isolaon wards together and screening health care workers may have the addional benefit of lessening disrupon of other services and freeing up staff to aend to other forms of care; this could help avoid knock-on consequences of

2

the COVID-19 epidemic (Hogan et al. 2020). Other economic benefits may not be reflected in these calculaons: for instance, screening of key workers may enable more business to connue operang, and screening haulers at the border (rather than closing borders) helps internaonal trade to connue.

Five Use Cases

1. Clinical Triage and Cohorng

In a clinical seng, diagnosc tests may be used to confirm paents' infecon so that infected paents may be grouped together and isolated from other paents, a policy known as cohorng. Without a test for SARS-CoV-2, this may be done instead based on symptoms, whereby hospitals group all paents without symptoms together. As a result, those who are not infected may be grouped with those who are infected with SARS-CoV-2 but are asymptomac.

Authors' calculaon shows that cohorng using a diagnosc test rather than symptoms alone can reduce infecons originang from the hospital by 87 percent, resulng in 106 fewer deaths per 100,000 paents at the peak of the epidemic. It is assumed that 50 percent of paents entering the hospital have SARS-CoV-2 at the peak of an epidemic. To account for risk of in-hospital infecon, it is assumed that the potenal for transmission (i.e., R0, the basic reproducon number) is 2.5 in the hospital seng (meaning that each infecon leads to 2.5 more infecons), but this does not affect the proporonate reducon in transmission risk that is provided by the

use of tesng (see Excel spreadsheet for calculaon).

https://drive.google.com/file/d/1HJ04XH23Ak7h7ARvBEqVH-TtuGGbZA1e/view

2. At-Risk Worker Screening

The risk of any group of workers contribung to onward transmission can be reduced by screening them for SARS-CoV-2. Screening enables those persons who are infected but not symptomac to self-isolate (in addion to those who have symptoms) (Grassly et al.2020). The impact depends on how oſten they are screened and the delay unl they self-isolate (if the test is posive). If screening is done weekly and test results are provided before they risk infecng others, this could reduce the risk of onward transmission by about 32 percent.

The cost of at-risk worker screening depends on the number of tests needed to find one person who is currently infected who would not self-isolate without screening: that is, the prevalence of SARS-CoV-2 among those without symptoms in that group. Thus, screening is more cost effecve when it is done on higher-risk groups, such as health care workers, and those in public-facing roles. This prevalence will depend on the epidemic stage and other factors: for example, it was found to be as high as 10 percent among health workers in London at the peak of the epidemic (see, for example, Treibel et al. 2020), and 20 percent among those on COVID-19 wards in an Oxford hospital (Eyre et al. 2020), but 2 percent in frontline workers in the United Kingdom overall some weeks later (see, for example, the 2020 ONS survey in the United Kingdom). Between these high and low extremes, the number of tests per death averted ranges between 1042 and 5208, respecvely (figure 1). Thus, when deployed among a group at very high risk, this would appear to be one of the most cost-effecve and widely applicable use cases.

3. Populaon Surveillance to Trigger or Avoid Lockdown

If the policy is to implement a lockdown in an area (such as a large city) if there is a chance that the prevalence of SARS-CoV-2 infecon is more than 2 percent, random samples could be drawn from the populaon to inform that decision. This trigger value of prevalence is chosen somewhat arbitrarily. Smaller trigger values would require

larger sample sizes but lead to longer periods of lockdown. It is an open queson (not addressed here) as to the opmal triggers to use in this regard.

As an example, a daily random sample of 772 persons (or an equivalent) could be used to check if it can be ruled out (with 99 percent certainty) that prevalence exceeds 2 percent (see https:/o/driven.googlle.iconm/filee/d/1Pj-gaJaoepH_QEphjiURefHZdn6Me8dNaS2i-T7x/view?Ausp=shafringor derivaon) To have the same certainty that the prevalence is below 0.1 percent would require a random sample of 15,745 persons. The same-size sample is required to measure prevalence (with a given level of certainty) in any populaon, if the sampling is representave.

Using the epidemiological model described inhttpos://drnive.gologilen.come/file/d/15aah-8rpDspbpkNzvteYh9AneVX3dJNIdjiJ1xzB/viewB?usp=,sharing this analysis calculates lives saved from a lockdown policy that triggers when more than 2 percent of tests are posive, and is released when less than 1 percent of tests are posive. The lockdown is calibrated to reduce interpersonal contact such that the Reproducon Number (R) falls below 1. If implemented for one year, the lockdown policy guided by random sampling is expected to avert 175 deaths per 100,000 people at the cost 281,884 tests per year (that is, 772 tests per day), leading to a cost-efficiency of 1,611 tests per death averted.

By the same token, because lockdown can be released quickly (or avoided altogether when unnecessary), days of unnecessary lockdown are avoided, and this would minimize disrupons to the economy. Box 1 describes how such an approach may have contributed to imposing and relieving lockdowns in Italy.

Box 1. Populaon Surveillance to Guide a Response in Pracce

Italy was the country second worst affected by COVID-19 aſter China by March 2020. When the country suffered its first death on February 22 in the small town of Vo (3,000 inhabitants) in Veneto, the whole town was put in quaranne and every inhabitant was tested. During the first round of tesng, 89 people tested posive. During the second round, 9 days later, only 6 were infected. Interesngly, the Italian authories found that at the me of the first symptomac case, about 3 percent of the populaon had already been infected and most of them were completely asymptomac. Through mass tesng and isolaon of those infected, the virus was eradicated from the town rapidly. At least 60 percent of all people infected by the virus were asymptomac. Mass tesng can give a clear picture of how many people are carrying the virus and can transmit it to others.

Source: Crisan and Cassone 2020.

4. Test-Trace-Isolate

Box 2. Test-Trace-Isolate in Pracce

Research & Policy Brief No.45

The Republic of Korea experienced a steep growth in COVID-19 cases early in the pandemic, but it quickly reduced rates of infecon and maintained low numbers of daily new cases. Korea did not implement strict lockdown measures, but focused on case-based contact tracing and cluster tesng and isolang. The country expanded tesng capacity from 3000/day on February 7, 2020 to 15,000/day to 20,000/day, with a turnaround me of 6 hours to 24 hours by the end of March. All suspected cases and paents under invesgaon are tested. Contact-tracing is performed through a mix of paent interviews and analysis of mobile phone locaon, credit card transacons, and health data.

Source: Dighe et al. 2020.

The analysis focuses on a highly effecve tesng and tracing system whereby tests are provided quickly (90 percent within 48 hours) to those who are self-isolang; the message to self-isolate is conveyed to all contacts rapidly (90 percent within 48 hours); and the intervenon begins from a point when incidence is low (0.01 percent of the populaon infected). It considers a city of 100,000 people in which other available intervenons have already been used. The analysis assumes that the R0 is 1.5, having been reduced through effecve social distancing measures, and further assumes that 25 percent of persons self-isolate effecvely upon the onset of symptoms that could be caused by COVID-19. In addion, persons may have symptoms that are caused by another pathogen but that are mistaken as being those of COVID-19, and this leads to unnecessary periods of isolaon for some.

Compared to a scenario without it, the test-trace-isolate intervenon could help avoid a wave of the epidemic, prevenng 392 deaths per 100,000. If a more general lockdown would have been ordered (as per the same criteria), the intervenon would also help avoid that lockdown. This would spare the economy a loss of 12 percent of GDP, assuming producvity under lockdown is 70 percent lower than normal.

The disncve contribuon of the tesng component (as opposed to the tracing and isolang components) is to allow persons who are not actually infected to avoid the period of isolaon (which they would endure if there were no tesng and they had symptoms or were in recent contact with someone who did and were traced). An intervenon that includes tesng leads to 29 percent fewer person-days spent in isolaon than one that does not include tesng.

These calculaons are sensive to many assumpons but are broadly consistent with other independent analyses of the impact of

Again using the epidemiological model outlined ihttnps://driove.gonoglel.ciomn/filee/d/15ahA-8rDsppbkpNzvtYeh9AneVX3dJNIdijJx1zB/viBew?u,sp=sharing contact tracing (Ferre et al. 2020; Grassly et al. 2020; Kucharski et al.

the analysis considers a "test-trace-isolate" (TTI) intervenon being rolled out, whereby an index person who has had symptoms (who is self-isolang) receives a test. If the test is posive, their recent contacts are instructed to self-isolate, and they, in turn, are tested. If their test is negave, they will cease self-isolang. If the test is posive, a new round of contact tracing occurs whereby their own recent contacts are instructed to self-isolate and then are tested. TTI is the most targeted form of lockdown; it is applied only to those who are infecous. Box 2 presents a the case study of this intervenon in pracce.

2020). The assumpon used here that half the people infected with SARS-CoV-2 have no symptoms but have the same risk of transmission as others makes our esmates of the impact of the "test-trace-isolate" intervenon conservave. However, the intervenon would be substanally less effecve if it was slower to test and trace persons than assumed here, or started from a point of a higher number of cases. The intrinsic value of tesng per se would also be lower if the frequency of symptoms not caused by SARS-CoV-2 (but rather by influenza) was lower, or adherence to self-isolaon did not increase even if those tested received a posive test result.

3

5. Border Screening

The number of infected persons entering a country can be reduced through tesng and quaranne of those tesng posive. Again, this approach is more effecve than solely quaranning persons who have symptoms. The proporonal reducon in the number of persons with SARS-CoV-2 who enter the country following screening is equal to the sensivity of the test: so, a 90 percent sensivity test implies 90 percent fewer introducons of infecon to the country, compared to screening with symptoms alone. The fewer persons who enter a country with infecon, the less likely it is that the infecon will spread widely and the more likely it is that effecve measures can be put in place to control it. Of course, the effecveness of border screening also depends on what coverage of border-crossers can be achieved.

There is further risk that, without tesng, quaranne of persons with symptoms crossing the border can result in addional infecons, if those quaranned are housed in the same locaon as those without SARS-CoV-2 (for example, Zimbabwe reports new infecons at quaranne facilies, ACF 2020). Authors' calculaons show that "border screening" could achieve 11 fewer deaths per 100,000 people crossing the border. Here, it is assumed that 1 percent of the populaon crossing the border has SARS-CoV-2, and that R0 is 3.0, to account for the fact that social distancing may not be adhered to at border crossings. A lower R0 would imply fewer deaths averted per

test (see Excel spreadsheet with the calculaon).

https://drive.google.com/file/d/1HJ04XH23Ak7h7ARvBEqVH-TtuGGbZA1e/view

Economic Consideraons

This Brief has illustrated five ways that SARS-CoV-2 diagnoscs can save lives, and has esmated the cost-effecveness of each use case using a standard epidemiological approach. There is a large and posive return on investment in diagnoscs in all five use cases, even for a very low value of a stascal life. Clinical triage and cohorng; populaon surveillance to trigger or avoid lockdown; and at-risk worker screening appear to be the most cost effecve, followed by test-trace-and-isolate programs and border screening. While governments should seek supplies to deploy diagnoscs in all use cases, priorizing them in this order is likely to maximize the cost-effecveness of their use.

To arrive at these results, the analysis has assumed that disease prevalence is very high-that is, the pandemic is out of control-as is the case in many countries. A caveat demonstrated in figure 1 is that deaths averted per test falls as disease prevalence falls, because with lower prevalence more tests will be used on those who are not infected. For countries with very low prevalence and where very few cases would be detected in a clinical seng, border screening may be more cost effecve than either clinical triage and cohorng or

screening of at-risk workers. Readers may use our Excel tool to

https://drive.google.com/file/d/1HJ04XH23Ak7h7ARvBEqVH-TtuGGbZA1e/view

compare cost-effecveness under alternave levels of prevalence.

Addional consideraons concern how transmissibility and virulence affect the calculaons. New, more contagious strains imply that any single social interacon is more likely to lead to infecon, which will only make tesng more valuable. Differences in the age distribuon and obesity rates across communies may affect the case fatality rate (Goldberg and Reed 2020) and thus deaths averted per test, though such differences will not materially affect the ranking of use cases by cost effecveness within a community. Moreover, even with a lower case fatality rate, each tesng use cases will sll have an ROI greater than 1, even for a very low value of a stascal life.

The effecveness of tesng is linked to how much sociees invest in it. If countries invest more in the research and development of more accurate and rapid tests and in the complementary skills required for delivery and tracing, this investment can help save even more lives and livelihoods.

Finally, social returns from tesng are higher than individual returns. Indeed, contagion implies a negave externality (a cost to others that is not internalized by the contagious individual), while tesng implies a posive externality (a benefit that extends to society beyond the contagious individuals' benefit). This wedge between private and social benefits is parcularly high for asymptomac but infecous people because they have no incenve to get tested or to isolate themselves. To solve this externality, governments should think about using incenves for people to get tested. Such incenves for tesng have been used in the past for tuberculosis (TB) and HIV (Geffen 2011; Keang 2013).

4

References

Acemoglu, D., V. Chernozhukov, I. Werning, and M. D. Whinston. 2020. "A Mul-Risk SIR Model withFraser, C., S. Riley, R. M. Anderson, and N. M. Ferguson. 2004. "Factors that Make an Infecous Disease

Opmally Targeted Lockdown." NBER Working Paper 27102, Naonal Bureau of Economic Research, Cambridge, MA.

Outbreak Controllable." Proceedings of the Naonal Academy of Sciences 101 (16): 6146-51.

Galewitz, P. 2021. "Poll: Nearly Half of American Adults Now Want the Covid Vaccine-ASAP" Kaiser Health

Acon Contre La Faim (ACF). 2020. "Zimbabwe: Under-equipped quaranne centers are increasing the spread of COVID-19" Press Release. 29 July.hps://www.aconcontrelafaim.org/en/press/zimbabwe-under-equipped-quaranne-centers-are-

News (KHN), January 27.hps://khn.org/news/arcle/poll-nearly-half-of-american-adults-now-want-the-covid-vaccine-asap/

hps://www.aconcontrelafaim.org/en/press/zimbabwe-under-equipped-quaranne-centers-are-increasing-the-spread-of-covid-19/

increasing-the-spread-of-covid-19/

Geffen, N. 2011. "What Does Science Show about Incenves for HIV Tesng?" Quackdown, December, 13.

Alon, T. M., M. Kim, D. Lagakos, and M. VanVuren. 2020. "How Should Policy Responses to the COVID-19

hps://www.quackdown.info/arcle/what-does-science-show-about-incenves-hiv-tesng/.

Pandemic Differ in the Developing World?" NBER Working Paper 27273, Naonal Bureau of Economic Research, Cambridge, MA.

Goldberg, P. K., and T. Reed. 2020. "The Effects of the Coronavirus Pandemic in Emerging Market and

Beaumont, P. 2020. "Scheme to Get Covid Vaccine to Poorer Countries at 'High risk' of Failure." The

Developing Economies: An Opmisc Preliminary Account." Prepared for the 50th Brookings Papers on Economic Acvity Conference, June.

Guardian, December 16.hps://amp.theguardian.com/world/2020/dec/16/scheme-to-get-covid-vaccine-to-poorer-countries-hps://amp.theguardian.com/world/2020/dec/16/scheme-to-get-covid-vaccine-to-poorer-countries-at-high-risk-of-failure.

Grassly, N., M. Pons Salort, E. Parker, et al. 2020. "Role of Tesng in COVID-19 Control." Imperial College

ahpts:/-/amhp.thiegguarhdian.c-omr/wiorsld/2k020-/doec/16f/s-chefmae-to-igelt-ucovidr-vaeccine.-to-poorer-countries-at-high-risk-of-failure.

London, COVID-19 Report 16, April, 23.

Crisan, A., and A Cassone, 2020. "In One Italian Town, We Showed Mass Tesng Could Eradicate the

Hogan, A. B., B. L. Jewell, E. Sherrard-Smith, et al. 2020. "Potenal Impact of the COVID-19 Pandemic on HIV,

Coronavirus." The Guardian, March 20.hps://www.theguardian.com/commensfree/2020/mar/20/eradicated-coronavirus-mass-hps://www.theguardian.com/commensfree/2020/mar/20/eradicated-coronavirus-mass-tesng-covid-19-italy-vo.

Tuberculosis, and Malaria in Low-income and Middle-income Countries: A Modelling Study." The Lancet Global Health 8 (9): e1132-e1141.

tesng-covid-19-italy-vo.

Keang, J. 2013. "The Nudgy State." Foreign Policy, January, 2.

de Walque, D., J. Friedman, R. Ga, and A. Maoo. 2020. "How Two Tests Can Help Contain COVID-19 and

hps://foreignpolicy.com/2013/01/02/the-nudgy-state/

Revive the Economy." Research & Policy Brief 29, World Bank Group, Washington, DC.

Kucharski, A. J., P. Klepac, A. Conlan, S. M. Kissler, M. Tang, H. Fry, J. Gog, J. Edmunds, and CMMID COVID-19

Dighe, A., L. Caarino, G. Cuomo-Dannenburg, et al. 2020. "Response to COVID-19 in South Korea and

Implicaons for Liſting Stringent Intervenons." BMC Medicine 18 (321).hps://doi.org/10.1186/s12916-020-01791-8.

Working Group. 2020. "Effecveness of Isolaon, Tesng, Contact Tracing and Physical Distancing on Reducing Transmission of SARS-CoV-2 in Different Sengs: A Mathemacal Modelling Study." The Lancet Infecous Diseases 20 (10): 1151-60.

Eichenbaum, M. S., S. Rebelo, and M. Trabandt. 2020. "The Macroeconomics of Epidemics." NBER Working

Paper 26882, Naonal Bureau of Economic Research, Cambridge, MA.

Robinson, L. A., J. K. Hammi, and L. O'Keeffe. 2019. "Valuing Mortality Risk Reducons in Global

Benefit-Cost Analysis." Journal of Benefit-Cost Analysis 10 (S1): 15-50.

Eyre, D. W., S. F. Lumley, D. O'Donnell, et al. 2020. "Differenal Occupaonal Risks to Healthcare Workers from SARS-CoV-2: A Prospecve Observaonal Study." MedRxiv. doi: 10.7554/eLife.60675.

Treibel, T. A., C. Manisty, M. Burton, et al. 2020. "COVID-19: PCR Screening of Asymptomac Health-care

Workers at London Hospital." The Lancet 395 (10237): 1608-10.

Ferre, L., C. Wymant, M. Kendall, et al. 2020. "Quanfying SARS-CoV-2 Transmission Suggests Epidemic

Control with Digital Contact Tracing." Science 368 (6491): eabb6936.

United Kingdom, ONS (Office for Naonal Stascs). 2020. Coronavirus (COVID-19) Infecon Survey.

Flaxman, S., S. Mishra, A. Gandy, et al. 2020. "Esmang the Effects of Non-Pharmaceucal Intervenons on

hhpps:/s/:w/w/ww.owns.wgo.v.oukn/pse.ogpolepvo.puukla/poneanodpcolmempuoniptyu/hleaalthoanndasonciadlccaore/mcomndiuonnsiatnyd/dhiseeasaelst/dhaatansedts/scoorconiaavlicruascroevi/d1c9oinnfedcionosunrvseyadnatad.

COVID-19 in Europe." Nature 584 (7820): 257-61.

diseases/datasets/coronaviruscovid19infeconsurveydata.