PLoS ONE plos plosone PLOS ONE 1932-6203 Public Library of Science San Francisco, CA USA 10.1371/journal.pone.0252400 PONE-D-20-18242 Research Article Medicine and health sciencesMedical conditionsInfectious diseasesViral diseasesCOVID 19 People and placesGeographical locationsOceaniaAustralia Social sciencesEconomicsHealth economics Medicine and health sciencesHealth careHealth economics Biology and life sciencesPopulation biologyPopulation metricsDeath rates Social sciencesEconomicsWelfare economics Medicine and health sciencesPublic and occupational health Social sciencesEconomics Medicine and health sciencesMedical conditionsInfectious diseasesInfectious disease controlSocial distancing Health and economic costs of early and delayed suppression and the unmitigated spread of COVID-19: The case of Australia Health and economic costs of early and delayed suppression and the unmitigated spread of COVID-19 https://orcid.org/0000-0002-0665-0966 Kompas Tom Conceptualization Formal analysis Methodology Validation Writing – original draft Writing – review & editing 1 * Grafton R. Quentin Conceptualization Formal analysis Writing – original draft Writing – review & editing 2 Che Tuong Nhu Data curation Formal analysis Writing – original draft 3 Chu Long Formal analysis Methodology Validation Writing – original draft 2 Camac James Formal analysis Investigation Writing – review & editing 4 Centre of Excellence for Biosecurity Risk Analysis, School of Biosciences and School of Ecosystem and Forest Sciences, University of Melbourne, Melbourne, Australia Crawford School of Public Policy, Australian National University, Canberra, Australia Australian Centre for Biosecurity and Environmental Economics, Crawford School of Public Policy, Australian National University, Canberra, Australia Centre of Excellence for Biosecurity Risk Analysis, School of Biosciences, University of Melbourne, Melbourne, Australia Devleesschauwer Brecht Editor Sciensano, BELGIUM

The authors have declared that no competing interests exist.

 * E-mail: tom.kompas@unimelb.edu.au 4 6 2021 2021 16 6 e0252400 15 6 2020 14 5 2021 2021 Kompas 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.

We compare the health and economic costs of early and delayed mandated suppression and the unmitigated spread of ‘first-wave’ COVID-19 infections in Australia in 2020. Using a fit-for-purpose SIQRM-compartment model for susceptible, infected, quarantined, recovered and mortalities on active cases, that we fitted from recorded data, a value of a statistical life year (VSLY) and an age-adjusted value of statistical life (A-VSL), we find that the economic costs of unmitigated suppression are multiples more than for early mandated suppression. We also find that using an equivalent VSLY welfare loss from fatalities to estimated GDP losses, drawn from survey data and our own estimates of the impact of suppression measures on the economy, means that for early suppression not to be the preferred strategy requires that Australia would have to incur more than 12,500–30,000 deaths, depending on the fatality rate with unmitigated spread, to the economy costs of early mandated suppression. We also find that early rather than delayed mandated suppression imposes much lower economy and health costs and conclude that in high-income countries, like Australia, a ‘go early, go hard’ strategy to suppress COVID-19 results in the lowest estimated public health and economy costs.

 The author(s) received no specific funding for this work. Data Availability Data Availability: All data and accompanying MATLAB model code can be obtained at: Kompas, T., Grafton, Q., Che, N., Chu, L., & Camac, J. (2020, December 5). COVID-19 Australia. Retrieved from osf.io/mn89p. We have also included the ‘Full Data’ set Excel file as a Supporting information file. Outbreaks COVID-19 1. Introduction

As of the end of May 2020, the global number of COVID-19 cumulative cases and reported fatalities, respectively, exceeded 6 million and 370,000 [1]. In response to the pandemic, many countries have imposed various types of suppression measures at different times to reduce the growth rate in COVID-19 infections [1]. Severe suppression measures (e.g., school and university closures, travel restrictions, mandatory social distancing requirements), when implemented in combination, have received the moniker, ‘lock-down’. Some have questioned whether the ‘cure’ (i.e., the lock-down) may be more costly in terms of the economy than no or limited suppression measures (e.g., [2–4]). This question can only be resolved with an epidemiological model of COVID-19 infections estimated from actual data combined with an empirical economic model of the costs of a lock-down. Using a combined epi-economic modelling approach we estimate and compare the public health and economic costs of early and late suppression measures as well as the impacts of unmitigated spread.

We evaluate whether a lock-down to suppress COVID-19 in Australia was justified by estimating the health and economic costs (including welfare costs of COVID-19 patients) for three scenarios. The first scenario is mandated ‘early suppression’ and is what actually occurred with the imposition of a lock-down that began in March 2020, and that we simulate continued until the end of May 2020. The second is mandated ‘delayed suppression’ or the effect of imposing a lock-down either 14, 21 or 28 days after it was actually imposed. For completeness, the third scenario is an ‘unmitigated spread’ counter-factual that assumes there were no Australian government (and voluntary) suppression measures. For each scenario, we estimate the number of fatalities, hospitalisations, and direct economy costs.

To obtain results we use a Susceptible, Infected, Quarantined, Recovered, and Mortality (SIQRM) compartment model we constructed for this purpose. Estimates of the welfare losses for COVID-19 patients are calculated using an Australian value of statistical life year (VSLY) and an age-adjusted value of a statistical life (A-VSL). Our estimates of the effects on the economy are drawn from established survey data and our own estimates of the impacts from losses on tourism and the effects of the extent or strictness of suppression measures on the economy.

 2. COVID-19 in Australia and policy responses

COVID-19 was first detected in Australia in January 25th, 2020, from a passenger who arrived in Melbourne from Wuhan on January 19th. As of May 30th, 2020, there were 7,173 reported cases and of which 6,582 have recovered with 102 deaths [5]. Fig 1 provides a summary of cumulative reported infected cases and the daily reported growth rate (3-day average) during this period.

10.1371/journal.pone.0252400.g001 COVID-19 infected cases and 3-day average growth rate.

The first recorded death from COVID-19 in Australia was on March 1st, 2020, when the total number of cases in the country was 29. During the initial stages of the infection, the total number of cases (including overseas arrivals) and the number of active cases grew exponentially, which prompted the Australian government to respond with a series of sequential and increasingly strict control measures.

Fig 2 summarizes the dynamics of COVID-19 and the main control responses in Australia since the first Australian death. Control responses varied by state and territory. From March 15th, the Australian federal government required international arrivals to self-quarantine for 14 days. Three days later, the government banned non-essential gatherings and encouraged social distancing (i.e., 1-5m distance and 4m2 per person). On March 21st, the federal government ordered non-essential businesses to close. Non-essential businesses included pubs, licensed clubs, hotels, places of worship, gyms, indoor sporting venue, cinemas, and casinos. Shortly thereafter, further tightening occurred with more restrictions on funeral and wedding attendance, fitness classes, and arcades [6].

10.1371/journal.pone.0252400.g002 Overview of COVID-19 in Australia.

The most severe mandated suppression measures were introduced on March 30th when public gatherings were reduced to a maximum of two people. From this date, Australians were required to stay at home unless shopping for essentials, receiving medical care, limited (30 minute) exercising, or traveling to and from work or for educational purposes. All suppression measures combined ‘flattened the curve’ and rapidly reduced the number of reported infected cases. Fig 2 shows that recoveries started exceeding new infections on April 4th 2020 when the number of recorded active cases peaked at 4,935.

As of 1st June 2020, Australian mandated suppression measures drastically reduced community transmission of the virus and the daily growth rate (the daily increase in the total number of cases over the total number of cases, on a three-day average) declined from around 25%, with 268 new recorded cases on 22nd March, to a daily growth rate of 0.26% and 11 new recorded cases on 30th May 2020 [5].

 3. Material and methods 3.1. Epidemiological model

We constructed an epidemiological model of COVID-19 infections in Australia that has five compartments, Susceptible, Infected, Quarantined, Recovered, and Mortality, using recorded data for Q, R, and M. Susceptible people can be infected with the virus via community transmission and become infectious. People who are infected can also come from abroad and transmit the disease before clinical signs appear. We assumed, as implemented in Australia, that all those who are officially reported to be infected are ‘quarantined’, recorded as active cases, in hospital or self-isolating, and either recover or die. Suppression measures mitigate the spread of the infection while the speed of recovery depends on treatment protocols.

Our model is formalized as follows: dSdt=-R0Ti1-ϵ×ISS+I+Q+R+μb S+I+Q+R-μd S dIdt=R0Ti1-ϵ×ISS+I+Q+R-μdI-1TiI+W dQdt=1TiI-χQ-fQ-μdQ dRdt=χQ-μdR dMdt=fQ where S, I, Q, R, M are the number of susceptible, infected, recorded active cases (i.e., quarantined), and eventually recorded as recovered and deceased, all functions of time; W is the number of infected people who arrive from abroad; R0 is the basic reproduction rate (see Wu et al. [7]); ϵ is the effectiveness of suppression measures; Ti is the infectious period; μd is the natural mortality rate (we estimate μd = 0.7% using the mortality data from the Australian Institute of Health and Welfare [8] and population data from ABS [9]); μb is the birth rate which we specify μb = 1.3%\ using the estimates from ABS [10]; χ is the rate at which active cases recover, which is observable (as in the top subplot in Fig 2); and f is the rate at which infected people die because of the virus, where f = 1.67%, as given from reported data in late April (noting that reported COVID-19 cases underestimated the true population infection rate (see Phipps et al. [11]), and varies daily over the model run from March 1st to April 20th [5]. Economy and cost impacts are also shown for the fatality rate in Verity et al. [12] of 0.7%.

The policy variable ϵ(t) represents mandated suppression measures and is delimited by 0 ≤ [1 − ϵ(t)] ≤ 1, for combined mandated and voluntary measures. Thus, without suppression measures [1 − ϵ(t)] approaches one (i.e., ϵ(t) approaches 0) and with the most restrictive suppression measures [1 − ϵ(t)] approaches zero (i.e., ϵ(t) approaches 1).

The total number of susceptible people is assumed to be 70% of the total population. From the SIQRM model in Eqs 1–5, the total number of recorded cases (T) are calculated as the sum of active cases (in quarantine), recovered, and deaths, or T = Q + R + M, all of which were observed.

It is important to note that the model, although fit for purpose, is an abstraction of the relevant infection stages and does not include any explicit time lag between infection and infectiousness, which has impact on the practical timing of the predicted epidemic in Australia.

 3.2. Model parameters and forecasting

We estimated parameters and tested the model to actual recorded data from March 1st to April 20th, 2020 and test the model on recorded data thereafter. There are five parameters to estimate. The first is the infectious period (Ti), the period over which an individual can spread the virus (which can occur before developing symptoms) to ceasing to spread the virus because of quarantine, recovery, or death. The second is the basic reproduction number (R0). The other three parameters include the reduction in community transmission after Australian governments introduced each of the three principal suppression measures, i.e., March 19th -21st, March 22nd–28th, and after March 28th, all lagged before coming into effect by the appropriate number of days (see Fig 2).

We fitted the number of recorded cases projected by the model to the number of observed cases using a non-linear least squares technique [13] that estimated the parameters by minimizing the sum of the squared distance between the projected and actual values in Eq (6): β^yto,Xto=argmin<β>∑t=0Toyto-SIQRMtβXto2 where yto is the observed total number of cases (i.e., recorded cases) at time t, which differs from the actual number of infected people; Xto is other observable information at time t (e.g., what suppression measures have been introduced up to time t); β^(yto,Xto) is the (asymptotic mean of) the estimated parameters, conditioning on the observable data and information; To = 51 is the number of days between March 1st and April 20th, 2020; SIQRMtβXto is the total number of recorded cases at time t that is projected by the SIQRM model given observable information Xto and a set of parameters β.

Following Green [13], the estimated asymptotic variance-covariance (ESV) matrix of the non-linear least square estimators was calculated using Eq (7): ESVβ=∑t=0T0yt0-SIQRMtβ^Xt02T0∂SIQRMtβ^Xt0∂βTr∂SIQRMtβ^Xt0∂β-1 where Tr is the transpose matrix operator.

A computational issue arising from the non-linear least squares technique is that it depends on the starting point of each parameter value, and the optimization process may end up with a local rather than the global minimum. Our response was to use a multi-start optimization algorithm. Thus, we repeatedly solved the least-squares optimization process with 1,000 random starting ‘guess points’ and used the best point to simulate the projection outcomes. These starting ‘guess points’ were randomized within certain ranges, as suggested by parameter values in the literature. For example, the range for randomizing the initial guess for the basic reproduction rate was 1.4 to 3.25 [14], and the range for the initial guess of the infectious period was between 1 and 14 days [15]. The range for the effectiveness of suppression measures was set between 0 and 1, by the definition of that parameter.

The estimated parameters to forecast the dynamics of the spread of reported COVID-19 was undertaken by randomizing the parameters using their estimated distributions to generate 20,000 simulations with a parallel-processing routine. To test the predictive capacity of the model, we followed a standard procedure that used some part of the available data to fit the model, then projected forward to determine how the model matched the remaining available data. In particular, March 1st (the first recorded death) was considered as time zero, and we used the next 50-day interval (from March 2nd to April 20th, inclusive) to estimate model parameters, or to ‘fit the model’. Data from April 21nd were used to determine model ‘accuracy’. Suppression measures were introduced to account for the period of time (on average) before symptoms appear and thus when new cases were recorded in the data.

 3.3. Model fit

Fig 3 compares the mean-value projections generated by our SIQRM model and the actual data (relevant confidence intervals are provided in Fig 4). The three panels compare three observable indicators; namely, the total number of recorded cases, the number of active cases recorded, and the number of fatalities. The red line is model output, the blue line is the actual data, which we used to fit the model to April 20th, and the dotted line is actual post-fitted data. The model outcome for the peak timing of recorded active cases (roughly April 5th) is in the middle panel and closely matches actual data.

10.1371/journal.pone.0252400.g003 Projected versus actual data. 10.1371/journal.pone.0252400.g004 Australian COVID-19 dynamics: Early and delayed (21 days) suppression.

The two key estimated epidemiological parameters used to generate Fig 3, along with their 95% confidence intervals, are the (1) average infection period Ti in days, 7.00 [6.85–7.15] and (2) the basic reproduction rate R0, 2.48 [2.45–2.51]. The effectiveness of the three policy measures, estimated as percentage reductions in transmissions compared to the unmitigated spread scenario or, in effect, (1 − ϵ) in the SIQRM model, are: (3) bans on non-essential gatherings 92.19 [83.71–100], (4) non-essential business shutdown 38.14 [35.50–40.77] and (5) maximum two-person gatherings 3.5 [2.97–4.04]. A higher value of the policy parameter ϵ, or a lower (1 − ϵ), means a greater effect on reducing the rate of infection. Bans on non-essential gathering, for example, reduced transmission by roughly 8% with the most important mandated suppression measure coming from the limits on the number of people in a gathering.

Our results represent the impacts of a combination of suppression measures, especially non-essential business shut down and limits on gathering, noting that these measures likely reinforce each other. Combined with voluntary social distancing and home isolation, the three suppression measures decreased the extent of the transmission by an estimated (roughly) 96.5%. Our estimates are consistent with surveys by the Australian Bureau of Statistics [16] in the first week of April showing that 98 per cent of the population was practising social distancing and 88 per cent avoided public spaces and events.

There are two additional points to make. First, note that the significant drop in active cases at the beginning of April (and another drop in the third week of April) results from a spike in the number of cases which were classified as ‘recovered’—not all of which are predictable by epidemiological parameters alone because how a case was classified as recovered could depend on administrative and reporting procedures. For this reason, during the period where actual data was used to estimate the epidemiological model (i.e., March 1st to April 20th, 2020), we combine the fitted total number of cases (using epidemiological parameters only) and the daily recovery classification rate calculated from the actual data to estimate the number of cases which were reported as recovered. To test the prediction capacity or the accuracy of the epidemiological model, we used a period where the data was ‘new’ to the model (i.e., April 21st–May 11th, 2020), and in this testing situation, the number of ‘recovered’ cases was predicted by combining the epidemiologically predicted total number of cases with the average number of sick days of patients who recover, i.e., approximately 18.5 days [5].

Second, note that it is not just the mandated suppression measures that alters transmission, but how the people perceive them and change their social interactions. It is also the case that gathering bans and business shutdowns are highly correlated and subject to uncertainty. We do not account for either in the numerical calculations, given the nature of the compartment model, but we do pick up the point of the impacts of various social distancing outcomes in a related paper [17].

 3.4. Health care facilities

The requirements for health care facilities (e.g., hospitalization, mechanical, and non-invasive ventilators) are approximated via the following equations: H=h×Q ICU=icu×Q V=v×Q where H is the number of hospitalized patients and h is the hospitalization ratio (to the number of active cases). We used an estimate from the Australian Department of Health [18] and Garg et al. [19] to specify h = 4.6%; ICU is the number of patients who require an ICU bed and icu is the ratio of those patients to the number of active cases [20]. We specify icu = 1.5% using the estimate of Garg et al. [19] and Fox et al. [21]; V is the number of patients who require a ventilator, and v is the ratio of those patients to the number of active cases. The value of v is estimated to be 1.43% [22].

It is important to note that the unmitigated spread scenario assumes no capacity constraints in hospital beds or ICU units, despite the large increase in infections. This is clearly not realistic. At some point the capacity of the medical system is breached as indicated in the results below, and a lockdown or some severe government response to limit infections would likely ensue. We don’t account for this case. In this sense, a comparison of mandated early to delayed suppression is more practical and relevant.

To determine impacts on patient welfare, we used both a Value of a Statistical Life Year (VSLY) measure and an age-adjusted Value of Statistical Life (a-VSL). Note that we only have life expectancy for a given age as it relates to average life expectancy. Throughout, we follow the guidelines for health cost-effectiveness measures, section 3A, contained in [23].

In the first instance, we estimated welfare losses as the difference between normal life expectancy in Australia and average age at the time of death from COVID-19: L=VSLY(ts×R+M×ED) where VSLY is an estimate of the value society places on a year of life, in principle measured by estimating the marginal value or ‘willingness to pay’ (WTP) to reduce the risk of death. We use an updated VSLY [24], as applied by the Australian Government for public decision-making, of $213,000, independent of age [25]. ED in Eq (11) is the difference between normal life expectancy and the average age at death of patients who do not recover. We estimated ED = 6.9 years, using the average life expectancy in Australia of 82.5 years [10, 26], and estimate the average age at death from COVID-19 at 75.6 years [5]. The value ts is the average number of sick days of patients who recover, specified at 18.5 days on average [5].

The estimates in Abelson [24] are drawn from an extensive meta-analysis of prior VSLY and VSL estimates and includes an overall discussion on the major methods of valuation and empirical results for values of life, health and safety in Australia. It also suggests adjustments for those 70 years and older, although we rely instead on Alberini et al. [27] as more technically robust. As usual in cost-benefit studies, Abelson [24] adopts an average WTP value for life, adjusted for age, and thus the VSL is generally held constant regardless of the income of any social group either at any point in time or over time. Most importantly, as conventional, it assumes as a starting point the life of an adult of 40 years of age, likely to live for another 40 years, with again an adjustment for those over 70 years of age.

To avoid a problem with averaging and truncated values of VSLY estimates at the 82.5 average lifespan, we use a measure of VSL adjusted by age to account for the fact that most of the deaths in Australia from COVID-19 are among the elderly. Of the 103 fatalities as of 31st May, only 5 were less than 60 years of age, 10 were in their 60s, 34 in their 70s, 34 in their 80s and 20 in their 90s. The VSL was obtained from government estimates and given by $4.9 million [25]. We scaled this Australian VSL by .70 for those over 70 years of age, following Alberini et al. [27], who obtained age-adjusted WTP measures for reductions in mortality risk using contingent valuation surveys. We assumed that the age distribution is unchanged as the number of deaths varies.

 4. Results

Fig 4 shows both the effect of the 8-weeks lock-down from early suppression and the effect of an assumed 21-day delay in introducing the full range of mandated suppression measures. Early mandated suppression in Australia was highly effective at both ‘flattening and shortening’ the curve. Our model closely approximates the actual data and estimates the total number of active cases peaks at (roughly) 4,850 cases, with 100 deaths and 6,650 total cases (for a 95% confidence interval of [4,032–12,082] in case load).

With a 21-day delay mandated suppression strategy, the number of active cases peaks at 241,000 [132,000–405,000] and the number of deaths increases to 9,074 [5300–15,360]. The period over which severe suppression measures is extended runs until the third week of July, where there are roughly only 500 active cases remaining. With a 14-day delay the number of fatalities is 2,144 [1,312–3,334] and with a 28-day delay the number of fatalities is 35,374 [19,850–59,070].

Fig 5 shows the case of unmitigated spread, the extreme counterfactual. The total number of cases is nearly 16 million, with a peak of 5.7 million active cases. Fatalities without control, are roughly 260,000 (assuming Australia’s current reported mortality rate or reported deaths as a fraction of reported cases). Noting the difference in scale in the two panels in Fig 5, with unmitigated spread, the projected number of active cases still exceeds 2,100 in August 2020 and COVID-19 is not effectively suppressed until late December 2020. Using the lower fatality ratio of 0.7%, as per in Verity et al. [12], results in roughly 112,000 deaths with the unmitigated spread scenario. We include this case, along with the Australian reported fatality rate, in the economic estimates to follow because although the Australian fatality rate on recorded cases predicts well for the lock-down and delayed suppression cases (with their limited horizons), a different fatality rate may apply for the (out of sample) unmitigated spread case.

10.1371/journal.pone.0252400.g005 Australian COVID-19 dynamics: Early and unmitigated spread.

We have not examined or modelled excess mortality [28] compared to the previous year in Australia, but have a related paper which estimates the true (population) infection rate based on the confirmed number of cases obtained through RNA viral testing [11].

4.1. Public health outcomes

Using the results represented in Figs 4 and 5, we projected the demand for health care facilities for early and delayed suppression measures and unmitigated spread. With early mandated suppression, the number of hospital patients is 220 at its peak, with a 95% confidence interval of [130–380]. The peak number of ICU beds and ventilators are, respectively, about 80 and 70 with early mandated suppression. With unmitigated spread, assuming no capacity limits, the number of hospitalized patients peaks at more than 260,000, of which more than 80,000 would need to be placed in an ICU. This ICU use estimate is nearly 40 times higher than the capacity of Australia’s health care system, with only 2,229–2,378 ICU beds [29, 30].

With unmitigated spread, roughly 70,000 patients would require ventilators at the peak, while there are only around 4,820 machines capable of delivering invasive mechanical ventilation [30]. For 21-day delayed suppression, hospital beds peak at 11,100 [6,000–18,640], and ICUs at 3,585 [1,980–6,075], both also exceeding the capacity of the Australian medical care system.

 4.2. Welfare losses from Covid19 patients

Tables 1 and 2 provide welfare loss estimates of COVID-19 patients using an Australian VSL of $4.9 million, adjusted by 0.70 for those over 70 years of age, and a VSLY of $231,000 AUD. The VSL estimates in Table 1 vary substantially depending on the scenario. Welfare damages are minimized with early suppression. For the case of a fatality rate of 0.7%, the damage would be $401.6 billion.

10.1371/journal.pone.0252400.t001 A-VSL mortality estimates by scenario ($ billion AUD).
Early suppression 14 Day Delay 21 Day Delay 28 Day Delay Unmitigated Spread
0.37 7.9 34.1 129.5 956.2
(0.28–0.56) (4.7–13.1) (19.1–57.1) (69.5–221.2) (947.5–964.6)
10.1371/journal.pone.0252400.t002 Welfare losses of Covid-19 patients: Early versus unmitigated spread.
Early Suppression ($ million) Unmitigated Spread ($ billion)
Recovered 80.8 192.1
[58.5–113.7] [190.3–193.7]
Fatalities 148.2 380.8
[111.9–221.4] [337.3–383.8]
Total 229 572.8
[170.4–335] [567.6–577.5]

Table 2, with the VSLY measure, shows that, with mandated early suppression, the total welfare loss is approximately $229 million AUD with a 95% confidence interval of [$170.4–$335] million AUD. Around 30% of this total cost is attributed to those who recover and around 70% of the costs are from fatalities. For the unmitigated spread scenario, the losses are very large, amounting to $572.8 billion [567.7–577.5], or roughly 2,500 times the loss of early suppression. Losses with the 0.7% fatality rate are $240.6 billion, or 1,048 times larger than early suppression.

 4.3. Health care costs

Table 3 reports the estimate of hospitalization costs, using average measures of bed costs [31, 32], for mandated early suppression and unmitigated spread, using VSLY. For early suppression, the mean value of the total hospitalization costs is $9.8 million AUD of which $3.9 million is the cost of ICU services and $5.9 million is for occupied hospital beds. For the unmitigated spread scenario, the number of hospitalizations is more than 2,300 times higher and the projected hospitalization costs are at least $23.3 billion AUD, with a 95% confidence interval of $23-$27.3 billion AUD. For the 0.7% fatality rate, total costs are $5.88 billion for the unmitigated spread case, or nearly 1000 times larger. Hospitalization costs do not include pre-hospital costs (e.g., such as visits to the GP or ambulance services) or any costs occurring post-hospitalization (e.g., follow-up expenses, and related health risks from having contracted the virus, which are known to be significant).

10.1371/journal.pone.0252400.t003 Estimated hospitalization costs: Early versus unmitigated spread.
Early Suppression ($ million) Unmitigated Spread ($ billion)
Total Costs 9.8 23.3
(6–18) (23–23.7)
ICU beds 3.9 9.3
(2.4–7.2) (9.2–9.5)
Hospital beds 5.9 14
(3.6–10.8) (13.8–14.2)
4.4. Direct economy-wide costs

To measure the direct economy-wide costs from early mandated suppression, we used a survey conducted by ABS of business activity during the period from March 30th to April 3rd, after the March 21st shut-down of all non-essential businesses, and again in May [33, 34]. In total, 84 percent of businesses reported that mandated suppression measures affected their activity, with many operating at reduced levels, or not at all. Using the reduction of business activity provided by the ABS across all categories, we apportioned the economic cost per day of suppression measures as C by: C=1365∑t=1n∑j=1mϕ(i,j)GVA(i,j) where ϕ(i, j) is the loss of gross value added in region i, sector j given the suppression measures and GVA(i, j) is the gross value of production in 2019 [35].

The value of ϕ(i, j) was estimated as: ϕi,j=Yei,j-Yci,jYei,j where Ye(i, j) and Yc(i, j) are the expected and actual level of business before and after the full set of suppression measures were in place for COVID-19, or ϵ = 0.965 in our SIQRM model, drawn from the ABS surveys [33, 34]. It is important to note in these measures that we adjusted the ABS estimates on mining activity following Claughton et al. [36], and that we account for IMF forecasts [37] of a 6% fall in global economic activity, projected to occur (with or without the Australian lock-down).

The value of C is captured in an aggregated form (across the various ABS tables and adjustments) in Table 4. Total losses are $982.2 million per day, or roughly $6.5 billion per week. In terms of order of magnitude, $982.2 million per day is about a 17% fall in daily GDP (using an annual GDP in 2019 of $1,995 billion AUD). Fig 6 shows the impacts by state and territory in Australia, with NSW the most adversely affected.

10.1371/journal.pone.0252400.g006 Cost of COVID control measures by region ($million/day).

Note: NSW = New South Wales; VIC = Victoria; SA = South Australia; WA = Western Australia; TAS = Tasmania; NT = Northern Territory; ACT = Australian Capital Territory.

 10.1371/journal.pone.0252400.t004 Cost of control measures ($ million per day AUD).
Sector/Region NSW VIC QLD SA WA TAS NT ACT Sum
Mining 11.9 3.2 30.5 2.3 67.9 0.7 4.3 0.0 120.9
Manufacturing 29.2 23.4 19.0 4.9 10.3 1.8 1.1 0.4 89.9
Electricity, gas, water, waste 6.9 4.5 5.3 1.8 2.6 0.9 0.3 0.3 22.5
Construction 31.7 26.2 19.3 5.7 14.4 1.5 3.3 1.8 103.9
Wholesale trade 12.0 9.6 7.6 2.3 3.5 0.5 0.2 0.3 36.0
Retail trade 8.4 7.0 5.2 1.8 2.7 0.6 0.3 0.5 26.5
Accom & food services 6.5 3.9 3.0 0.9 1.6 0.3 0.1 0.7 16.9
Transport, postal, warehouse 10.5 9.8 8.4 2.5 6.8 0.9 0.7 0.5 40.2
Infor media & telecom 10.3 6.6 3.4 1.5 1.9 0.4 0.2 0.4 24.7
Financial & insurance 45.0 28.0 13.3 4.9 6.7 1.1 0.4 0.9 100.4
Rental, hiring, real estate 13.1 7.1 7.1 1.6 2.7 0.6 1.0 0.8 34.0
Professional, scientific & tech 34.4 24.8 13.2 4.0 8.9 0.7 0.8 2.4 89.2
Administrative services 17.6 11.4 9.6 2.8 5.4 0.4 0.5 0.6 48.3
Education & training 26.4 21.1 14.6 5.0 7.0 1.6 0.8 4.4 81.0
Health care, social assistance 22.5 17.6 14.6 5.7 8.5 1.4 0.9 1.6 72.7
Arts & recreation services 2.7 1.8 1.1 0.3 0.5 0.1 0.1 0.4 7.0
Other services 4.5 3.7 2.8 0.8 1.6 0.2 0.1 0.2 14.0
Sum 293.6 209.6 177.8 48.8 153.2 13.8 15.1 16.3 928.2

Note: Calculations based on source survey material on business activity from IMF [37], ABS [33, 34], and Claughton et al. [36]

 4.5. Transition costs following a lock-down

Our estimated economic losses from 8 weeks of early mandated suppression measures start from March 30th. These economic losses are the economic cost per day times the number of days under the sequence of the introduced control measures, or roughly $51.98 billion AUD over the eight-week period, including all international tourism (see Table 4). As suppression measures are relaxed, in the week of May 25th by our simulation, the economy does not ‘snap back’ but transitions to full economic activity over time, taking into account that international travel restrictions will still be in place for some time, or at least until 2021 by our calibration. The speed of transition also depends on how quickly government relaxes controls.

In formal terms, the cost of transition is a generalised function of the extent of control measures: Ct=Cϵ1+δ=Cmaxϵt1-ϵmax1+δ where C(t) is the cost of control measure at time t; ϵ(t) is the policy measure at time t, ϵmax = .965 is the most restrictive policy measure, drawn from the SIQRM model, and δ is a cost-policy parameter. We calibrated Eq (14) with a knowledge of the cost of the lock-down above at $982.2 million per day and the economic costs of controls conditioned on an estimated value of ϵ(t) = 0.01, drawn from our SIQRM model, for the international travel ban, or the ‘arrival block’ on China, Iran, South Korea and Italy, prior to more severe domestic control measures being put in place. The losses from the arrival block, Cab, are given by: Cab=1365∑t=1zGVGVAi where GVA(i) is tourism revenue from region i and Z is number of blocked regions, initially China, Iran, South Korea, and Italy [38]. The total number of passengers from blocked regions is based on the Bureau of Infrastructure, Transport and Regional Economics (BITRE) (the Department of Infrastructure, Regional Development and Cities of Australia) [39]. Based on BITRE [39], the ban on travel from the arrival block reduced total inbound passengers, starting in January, by 17.44%. In 2019, 9.3 million international visitors arrived in Australia with a gross value of tourism of $45.4 billion [38]. In summary, we attribute all the costs of reduction in international tourism to the lockdown. This overstates the costs of lockdown because a decline in international tourism would have occurred even in the absence of stringent social distancing. This is because of Australia’s international border controls limit the number of arrivals into Australia and requires supervised quarantine of 14 days. These international border controls have been continuously maintained even when national lockdown ended in May 2020.

Results are summarized in Table 5 and using Eq (14) implies: Cab=39.45=928.20.010.9651+δ such that 1+δ=ln39.45928.2ln0.010.965 or a value of δ = −0.309.

10.1371/journal.pone.0252400.t005 Cost of international arrival block.
Contents Unit Value
International passengers from banned regions in Jan 2020 pers 386,278
Total international passengers in Jan 2020 pers 2,214,474
Share of arrival block regions in total passengers % 17.13
Share of arrival block regions in total tourism dollars % 31.7
GVA of tourism in 2019 $ billion/year 45.4
Average loss per day (banned regions) $ million/day 39.45

Note: Calculations based on source material from BITRE [39] and Tourism of Australia [38]

Using Eq (14), we ‘unwind’ the government control measures and allow for a return of economic activity until ‘full recovery’, so that, after transition, the remaining costs are those largely from losses in international tourism from all countries. If the transition time is as short as 1 month, the domestic economy recovers quickly and the losses in GDP are the smallest. If it takes longer for the economy to recover, modelled as a slower fall in ϵ in our formulation, then depending on the period of time over which the transition occurs, economy losses are larger.

 4.6. Public health and economy costs combined

Table 6 provides a summary of the total economic losses for four different scenarios, by months of transmission for 1, 2, 3 and 4 months duration, drawing on Australian Bureau of Statistics survey data [33, 34] on the cost of the 8-weeks lock-down itself, which we estimate at $51.98 billion. There are two issues to consider: how quickly suppression measures are relaxed and the time it takes for the economy to recover. With early mandated suppression, lock-down measures can be removed more quickly.

10.1371/journal.pone.0252400.t006 Direct economic costs with early suppression in $ billions and GDP loss, and health and welfare losses.
Recovery (months) Costs ($ billion AUD) Economy Costs (% annual GDP)
Lock-down Recovery Total Annual Loss GDP (%)
Early Suppression Measures for 8 Weeks from March 30th
Transition 1 1 51.98 14.39 66.37 3.33
Transition 2 2 51.98 30.39 82.37 4.13
Transition 3 3 51.98 48.27 100.26 5.03
Transition 4 4 51.98 68.59 120.57 6.04
Welfare Losses, Hospitalization Costs and Fatality Equivalents of Unmitigated Spread
Welfare Hospital Total Annual Loss GDP (%)
VSLY 572.8 23.3 596.1 29.8
VSLY* 240 23.3 263.3 13.1
A-VSL 956.2 956.2 47.9
A-VSL* 401.6 401.6 20.1
Fatality equivalent at %GDP** 30,491 (3.3%) 37,816 (4.13%) 46,057 (5.03%) 55,305 (6.04%)
Fatality equivalent at %GDP* 12,808 (3.3%) 15,882 (4.13%) 19,343 (5.03%) 23,228 (6.04%)

*VSLY, A-VSL and fatality equivalent measures using the fatality ratio in Verity et al. [12].

**Fatality equivalent is the VSLY-measured number of fatalities under the unmitigated spread scenario that equals the direct economy cost associated with an early 8-weeks lock-down (early suppression) for each % GDP loss (3.33, 4.13, 5.03, and 6.04). N.B. The estimated early suppression model fatalities are 100.

We assume that the cost of transition is roughly half the monthly losses in lock-down in the first month, with costs of transition convex in recovery time in the remaining months. We also note that the cost of the lock-down itself already includes losses in international tourism for all countries, which continue through transition. Given a convex cost of recovery, our simulation gives a range of annualized losses in GDP of early suppression from 3.33% to 6.04% (see Table 6). With no suppression measures, the welfare and hospitalization costs as a percentage of annual GDP range from 13.1% to 47.9%, depending on the welfare measure.

 5. Discussion

Whether early mandated suppression results in a higher overall economic cost than delayed or no suppression measures requires specification of the severity and duration of the lock-down and reliable estimates in relation to key public health parameters. Using a fit-for-purpose SIQRM model for Australia, we estimated the number of cases (cumulative and active), fatalities, hospitalization costs, welfare losses for COVID-19 patients and the loss in economic activity under early suppression and various delays (14, 21 and 28 days), including the case of unmitigated spread. Total estimated hospitalization costs, welfare losses and number of deaths range from more than 1,000 to 4,000 times larger with unmitigated spread compared to early suppression. Delays in suppression (14 to 28 days, or more) provide no economic gain, but increase fatalities and also lengthen the period over which suppression measures are required before active cases fall below 500. This is not to say that the economic cost of the suppression does not depend on the timing, only that fatalities are impacted by the timing of the suppression measures.

Direct economy costs of early mandated suppression depend on how quickly the economy recovers following a lock-down. In the fastest (1 month) assumed recovery after lock-down, the total economy-wide costs are about 3% of GDP. With the slowest assumed recovery (4 months), the total direct economy costs are some 6% of GDP. Current Reserve Bank annualized estimates for the period from January to September of 2020 indicate a fall in Australia-wide GDP of 3.8%. By contrast, the total welfare losses and hospitalization costs range from 12.5% to nearly 48% of GDP with no suppression measures in place. Our results indicate that the total economic costs of unmitigated spread are (roughly) between 4 and over 8 times larger than mandated early suppression (see Table 6).

We provide a number of important caveats to our work. First, our VSLY measure underestimates the welfare losses associated with unmitigated spread. This is because we only have limited information to benchmark on average life expectancy, noting that many of those who die of COVID-19 are older than 82.5 years. In our view, people in this older age bracket would have a both a willingness and ability to pay for additional life. Second, in the case of unmitigated spread, we assume no constraints on hospital and ICU capacity noting that a capacity constraint would likely increase fatalities, and thus welfare losses, should such a constraint be exceeded. This effect alone may mandate a lockdown at some point. Third, for the unmitigated spread scenario, we do not explicitly model the effects of voluntary physical distancing or a delayed lock-down that could occur with substantial community infections and deaths. We have a related paper that does so [17]. Instead, we estimate the welfare losses (VSLY of fatalities) equivalent to the GDP costs of an 8-weeks lock-down (see Table 6). We find that VSLY welfare losses of fatalities that would be equivalent to GDP losses from an early 8-weeks lock-down results in more than 12,500–30,000 Australian mortalities, depending on the fatality rate–fatalities comparable to countries (e.g., United Kingdom) that implemented delayed lock-downs in 2020. In other words, for early suppression not to be the preferred response to COVID-19, requires that Australia would have to incur more than 12,500–30,000 deaths from unmitigated spread (actual Australian deaths were 102 as of 31st May 2020) than the direct economic costs associated with an early 8-weeks lock-down. Fourth, data limitations preclude us from estimating indirect social costs of unemployment with early or delayed suppression (e.g., additional suicides, domestic violence and alcoholism), but note that many Australian workers have retained employment through a government subsidised and temporary ‘Job Keeper’ scheme that began on March 30th, 2020. We have also not accounted for morbidity effects in health outcomes associated with those who recovered form COVID-19.

Notwithstanding our caveats, from both a public health (mortality and morbidity) and direct economic costs (including public health costs) perspective, Australia’s mandated early suppression measures that began in March 2020, and which we assume would have continued until the end of May 2020 (suppression measures began to be progressively relaxed from early May), generate a very large economic payoff relative to alternatives of delayed suppression measures or unmitigated spread.

Our findings provide robust evidence that a ‘go hard, go early’ mandated suppression, at least in a high-income country like Australia, is the preferred approach from both a public health and an economy perspective–a result consistent with the non-technical discussion in Group of Eight Australia [40]. By comparison, some high-income countries adopted a delayed (or much less strict) lock-down with infection rates much higher than when Australia began its lock-down. Our model results suggest that if other high-income countries had imposed effective suppression measures earlier they may have reduced both their public health (including lower fatalities per 1,000 people) and economy costs.

 Supporting information <p>(XLSX)</p> </caption> </supplementary-material> </sec> </body> <back> <ack> <p>The authors are grateful for helpful comments and suggestions provided by Emily Banks, John Baumgartner, Nathaniel Bloomfield, John Parslow, and Andrew Robinson on earlier versions of this manuscript.</p> </ack> <ref-list> <title>References Johns Hopkins University. COVID-19 Map—Johns Hopkins Coronavirus Resource Center. 2020. https://coronavirus.jhu.edu/map.html Frijters P. The Corona Dilemma. 2020. https://clubtroppo.com.au Singer P, Plant M. When Will the Pandemic Cure Be Worse Than the Disease? Project Syndicate. 2020. https://www.project-syndicate.org/commentary/when-will-lockdowns-be-worse-than-covid19-by-peter-singer-and-michael-plant-2020-04?barrier=accesspaylog van den Broek-Altenburg E, Atherly A. Economic Cost of Flattening the Curve. The Incidental Economist. 2020. https://theincidentaleconomist.com/wordpress/economic-cost-of-flattening-the-curve/ Covid19data. Covid19-Data Coronavirus (COVID-19) in Australia. 2020. https://www.covid19data.com.au/transmission-sources Australian Department of Health. Impact of COVID-19 in Australia–ensuring the health system can respond. 2020. https://www.health.gov.au Wu JT, Leung K, Leung GM. Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study. Lancet. 2020;395(10225): 689–97. doi: 10.1016/S0140-6736(20)30260-9 32014114 Australian Institute of Health and Welfare. Deaths in Australia. Australian Institute of Health and Welfare: Report and Data. 2019. https://www.aihw.gov.au ABS. Australian Demographic Statistics. ABS-31010-June 2017. 2017. https://www.abs.gov.au ABS. Births, Australia in 2018. ABS Stat. 2020. https://www.stat.data.abs.gov.au Phipps SJ, Grafton RQ, Kompas T. Robust estimates of the true (population) infection rate for COVID-19: a backcasting approach. R Soc Open Sci. 2020 Nov 18;7(11): 200909. https://dx.doi.org/10.1098/rsos.200909 33391791 Verity R, Okell LC, Dorigatti I, Winskill P, Whittaker C, Imai N, et al. Estimates of the severity of coronavirus disease 2019: a model-based analysis. Lancet Infect Dis. 2020 Jun;20(6). https://doi.org/10.1016/S1473-3099(20)30243-7 32240634 Green W. Econometric Analysis. 7th ed. Pearson Education. 2012. Australian Department of Health. Novel coronavirus (COVID-19): Information for Clinicians—Frequently Asked Questions. Australian Department of Health: Novel coronavirus (COVID-19). 2020. https://www.health.gov.au Lauer SA, Grantz KH, Bi Q, Jones FK, Zheng Q, Meredith HR, et al. The incubation period of coronavirus disease 2019 (CoVID-19) from publicly reported confirmed cases: Estimation and application. Ann Intern Med. 2020;172(9): 577–82. https://doi.org/10.7326/M20-0504 32150748 ABS. Household Impacts of COVID-19 Survey. 2020. https://www.abs.gov.au/ Grafton RQ, Parslow J, Kompas T, Glass K, Banks E, Lokuge K. Health and economic effects of COVID-19 control in Australia: Modelling and quantifying the payoffs of “hard” versus “soft” lockdown. medRxiv. 2020. Available from: https://doi.org/10.1101/2020.08.31.20185587 Australian Department of Health. IMPACT OF COVID-19: Theoretical modelling of how the health system can respond, Australian Department of Health: Impact of COVID-19. 2020. https://www.health.gov.au Garg S, Kim L, Whitaker M, et al. Hospitalization Rates and Characteristics of Patients Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019—COVID-NET, 14 States, March 1–30. MMWR Morb Mortal Wkly Rep 2020;69: 458–464. https://dx.doi.org/10.15585/mmwr.mm6915e3 32298251 Science Media Centre. Expert comment on ICU recovery data on COVID-19. 2020. https://www.sciencemediacentre.org/expert-comment-on-icu-recovery-data-on-covid-19/?cli_action=1621572363.241 Fox GJ, Trauer JM, McBryde E. Modelling the impact of COVID-19 on intensive care services in New South Wales. Med J Aust. 2020 Jun 8;212(10): 468–469. doi: 10.5694/mja2.50606 32383153 Grasselli G, Zangrillo A, Zanella A, Antonelli M, Cabrini L, Castelli A, et al. Baseline Characteristics and Outcomes of 1591 Patients Infected with SARS-CoV-2 Admitted to ICUs of the Lombardy Region, Italy. JAMA—J Am Med Assoc. 2020;323(16): 1574–81. Available from: https://jamanetwork.com/ 32250385 Australian Department of Health. Guidelines for preparing a submission to the Pharmaceutical Benefits Advisory Committee (PBAC Guidelines). 2016. https://pbac.pbs.gov.au/information/about-the-guidelines.html Abelson P. Establishing a Monetary Value for Lives Saved: Issues and Controversies. Working Papers in Cost benefit Analysis WP 2008–2. Department of Finance and Deregulation. 2007. https://www.pmc.gov.au Prime Minister and Cabinet. Best Practice Regulation Guidance Note Value of statistical life. 2019. https://www.pmc.gov.au ABS. Life Tables, States, Territories and Australia, 2016–2018. ABS 3302. 2019. https://www.abs.gov.au Alberini A, Cropper M, Krupnick A, Simon NB. Does the value of a statistical life vary with age and health status? Evidence from the US and Canada. J Environ Econ Manage. 2004;48: 769–792. Kontis V, Bennett JE, Rashid T, Parks RM, Pearson-Stuttard J, Guillot M, et al. Magnitude, demographics and dynamics of the effect of the first wave of the COVID-19 pandemic on all-cause mortality in 21 industrialized countries. Nat Med. 2020;26(12): 1919–28. Available from: https://doi.org/10.1038/s41591-020-1112-0 33057181 Australian and New Zealand Intensive Care Society. Centre for Outcome and Resource Evaluation—2018 Report. 2019. https://www.anzics.com.au Litton E, Bucci T, Chavan S, Ho YY, Holley A, Howard G, et al. Surge capacity of intensive care units in case of acute increase in demand caused by COVID-19 in Australia. Med J Aust. 2020;212(10): 463–7. Available from: https://www.mja.com.au/journal/2020/212/10/surge-capacity-intensive-care-units-case-acute-increase-demand-caused-covid-19 32306408 IHPA. National Hospital Cost Data Collection, Public Hospitals Cost Report, Round 18. 2016. https://www.ihpa.gov.au/publications/australian-public-hospitals-cost-report-2013-2014-round-18 IHPA. National Hospital Cost Data Collection Report, Public Sector, Round 22. 2020. https://www.ihpa.gov.au/publications/national-hospital-cost-data-collection-report-public-sector-round-22-financial-year ABS. Business Indicators, Business Impacts of COVID-19. ABS-5676055003 Reported on 30 March. 2020. https://www.abs.gov.au/ ABS. Business Indicators, Business Impacts of COVID-19. ABS-5676055003 Reported on 4 May. 2020. https://www.abs.gov.au/ ABS. Statistics of Australia. 2020. https://www.abs.gov.au/ Claughton D, Fowler C, Fitzgerald D. Mining exploration and service companies hit by coronavirus restrictions. ABC Rural. 2020. https://www.abc.net.au/news/rural/2020-04-07/covid19-mine-explorers-and-service-sector/12125518 IMF. World Economic Outlook. WEO Report. 2020. https://www.imf.org Tourism of Australia. The Economic Importance of Tourism. 2020. https://www.tourism.australia.com BITRE. Time Series Statistics of International Airline Activities. 2020. https://www.bitre.gov.au Group of Eight Australia. Roadmap to Recovery A Report for the Nation. 2020. https://go8.edu.au/research/roadmap-to-recovery 10.1371/journal.pone.0252400.r001 Decision Letter 0 Devleesschauwer Brecht Academic Editor 2021 Brecht Devleesschauwer 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. Submission Version 0

15 Oct 2020

PONE-D-20-18242

Health and Economic Costs of Early, Delayed and No Suppression of COVID-19: The Case of Australia

PLOS ONE

Dear Dr. Kompas,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Nov 20 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to EACH point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Brecht Devleesschauwer

Academic Editor

PLOS ONE

Additional Editor Comments:

In your revision note, please include EACH of the reviewer comments, provide your reply, and when relevant, include the modified/new text (or motivate why you decided not to modify the text). Note that failure to do so may result in a rejection of the manuscript.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide more details on data collection. Specifically, please list the date range for which you obtained data, any inclusion/exclusion criteria, and the categories of data that were extracted.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors tackle very relevant and timely issues on the public health and economic burden and benefits of COVID-19 suppression in Australia. The central question is whether the cure (i.e. the lock-down) may be more ‘costly’ in terms of the economy than delayed or no suppression measures. Therefore, the authors constructed a transmission model to fit the observed incidence by estimating epidemiological and suppression parameters. By delaying or relaxing the suppression, the authors calculate the corresponding burden of disease and health cost and make the comparison with the current situation.

I do have some comments on the manuscript and methods.

The transmission model is based on 5 compartments in terms of S-I-Q-R-M. The observed active cases in Australia show a significant drop at the beginning of April and the model fits this behavior surprisingly well. How did the authors capture this behavior with a homogeneous mixing model? And even more interesting, how come this drop is not present in their scenario analyses on delayed or no suppression measures?

The counter-factual scenario “no suppression” assumes that no voluntary behavioral changes occur. As such, “unmitigated spread” would be a better label for this scenario. In addition, the analysis assumes that cost of hospital and ICU admissions can be linearly extrapolated even if the need outreaches the capacity. If hospital beds or ICU’s are not available, other choices will be made which a different impact on the burden of disease and related costs. As such, the statement that the current suppression has saved XXX lives and XXX dollars is controversial. The comparison with a delayed of relaxed suppression seems more informative and reliable.

Not the suppression measures but how the people conceive them and change their social interactions alters the transmission. In addition, the estimated impact in the SIQRM model of gathering bans and business shutdowns is highly correlated and subject of uncertainty. This uncertainty aspect is missing in the numerical calculations.

The impact calculation of patient welfare is not clear to me. They use both the Value of a Statistical Life Year and an age-adjusted Value of Statistical Life. For the latter, they adopt a government estimate of $4.9 million. How is this calculated and which assumptions are made? This is a crucial parameter in a cost-benefit analysis. Please provide more insights on this.

Australia has clear guidelines for health technology assessments, which are not fully addressed here (https://pbac.pbs.gov.au/information/about-the-guidelines.html) Please make sure all aspects of section 3 “Economic evaluation” are included and reported in the study.

Reviewer #2: Abstract:

• Early (actual): not very clear what you mean with actual

• Second sentence is long (3/4 of the abstract) and hard to read; rephrase

• Clarify the methods used

Introduction

• As of the end of May 2020, the global number of COVID-19 cumulative cases and reported fatalities, respectively, exceed 6 million and 370,000 --> worldwide?

• Update the data -->� once accepted use the most recent available data

• (including the President of the United States on 22 March 2020) �--> not sure if this adds much. Others including scientists, were also posing this question.

• Several language issues: More costly in terms of the economy ; that we simulate continued until the end of May

• The structure is not logical. Research question comes rather early. Background info is limited. Last paragraph should be moved to method section, results are already given in introduction

• 2. COVID-19 in Australia and policy responses -->� include this in introduction section ; this is the background that we need in order to introduce research question

• Information included under heading 2 can also be shortened -->� too much detail about progress of pandemic. Main question is, why do we need these analyses?

Methods:

• excluded the below-20-year-old population group -->� most recent evidence indicated that you should lower the lower bound. Adolescents are also susceptible

• data source not clear

• I do not completely agree with QALY comment, luckily the adjustment for age covers this partially --> discuss this issue in more detail. + effect of age adjustement (scenario analyses of different adjustments)

• Life expectancy for given age should be used

• Why not using the GDP as VSLY?

• You only look at PRO of lock down (effect on covid deaths) but not at CON (effect of lock down on general population, delay in usual care, increase in more severe diagnosis after lock down, etc…)

• How did you calculate VSLY for recovered patients?

• Why not looking at (modelled) excess mortality compared to previous year?

Results

• It would be very informative to see the results of individual different measures included in lock down--> closure of schools, shut down of non-essential business, social distancing

DIscussion

• The discussion lacks comparison with other studies. It is merely a summary of own findings.

• Problem with +82 year olds could be solve by using life expectancy at given age

• What was the influence of time? We see that the number of deaths is now decreasing, even though less measures are in place compared to lock down? How should we look at upcoming months? Go hard, go early? Of a more adapted scenario?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

 10.1371/journal.pone.0252400.r002 Author response to Decision Letter 0 Submission Version 1

30 Dec 2020

Responses to Reviewers

PONE-D-20-18242

Health and Economic Costs of Early, Delayed and No Suppression of COVID-19: The Case of Australia

The authors of this paper sincerely thank the editor and two reviewers for their valuable comments and suggestions. The paper has been greatly improved as a result. Specifically, we have made the following changes in response to the Reviewers’ Comments:

Reviewer #1

The authors tackle very relevant and timely issues on the public health and economic burden and benefits of COVID-19 suppression in Australia. The central question is whether the cure (i.e., the lock-down) may be more ‘costly’ in terms of the economy than delayed or no suppression measures. Therefore, the authors constructed a transmission model to fit the observed incidence by estimating epidemiological and suppression parameters. By delaying or relaxing the suppression, the authors calculate the corresponding burden of disease and health cost and make the comparison with the current situation.

I do have some comments on the manuscript and methods.

Comment #1. The transmission model is based on 5 compartments in terms of S-I-Q-R-M. The observed active cases in Australia show a significant drop at the beginning of April and the model fits this behaviour surprisingly well. How did the authors capture this behaviour with a homogeneous mixing model? And even more interesting, how come this drop is not present in their scenario analyses on delayed or no suppression measures?

Response #1: Thank you. We have revised the paper, including the figure label to clarify this point, with the following in mind. The significant drop in active cases at the beginning of April (and another drop in the third week of April) results from a change in the administrative procedure for how a case was classified as ‘recovered’ during March-April 2020. For example, two active cases appear to have recovered (i.e., no longer contagious) on the same day, but were classified as recovered on different days, and some active cases on record appear to have recovered but were not yet classified due to variable administrative procedures early on and/or different testing rates and capacity during a given week – all of which are not predictable outcomes by our epidemiological parameters alone. For this reason, we estimated the number of cases which were classified recovered by combining the epidemiologically estimated total number of cases and the recorded daily recovery rate whenever actual data were available. As there were a spike in the daily recovery classification rates in the beginning and the third week of April 2020, the number of recovery classifications jumped and the number of active cases on record dropped. We have basically included this text verbatim in the revised paper.

Comment #2: The counter-factual scenario “no suppression” assumes that no voluntary behavioural changes occur. As such, “unmitigated spread” would be a better label for this scenario. In addition, the analysis assumes that cost of hospital and ICU admissions can be linearly extrapolated even if the need outreaches the capacity. If hospital beds or ICU’s are not available, other choices will be made which a different impact on the burden of disease and related costs. As such, the statement that the current suppression has saved XXX lives and XXX dollars is controversial. The comparison with a delayed of relaxed suppression seems more informative and reliable.

Response #2: We agree, this is a good suggestion. We have changed ‘no suppression’ to ‘unmitigated spread’ throughout, including in the title of the paper. We also highlight your point that ICU admissions and hospitalisations will reach a limit at some point and thus the comparison with ‘delayed suppression’ may be more informative. The text in section 3.4 reads: “It is important to note that the unmitigated spread scenario assumes no capacity constraints in hospital beds or ICU units, despite the large increase in infections. At some point the capacity of the medical system is breached as indicated in the results below, and a lockdown would likely ensue. In this sense, a comparison of early to delayed suppression is more practical and relevant.”

Comment #3. Not the suppression measures but how the people conceive them and change their social interactions alters the transmission. In addition, the estimated impact in the SIQRM model of gathering bans and business shutdowns is highly correlated and subject of uncertainty. This uncertainty aspect is missing in the numerical calculations.

Response #3: Thank you. We acknowledge both points in our revised paper, at the end of section 3.3 and indicate a related work that considers various social distancing outcomes. The text reads: “Second, note that it is not just the mandated suppression measures that alters transmission, but how the people perceive them and change their social interactions. It is also the case that gathering bans and business shutdowns are highly correlated and subject to uncertainty. We do not account for either in the numerical calculations, given the nature of the compartment model, but we do pick up the point of the impacts of various social distancing outcomes in a related paper (Grafton et al. 2020).”

Comment #4: The impact calculation of patient welfare is not clear to me. They use both the Value of a Statistical Life Year and an age-adjusted Value of Statistical Life. For the latter, they adopt a government estimate of $4.9 million. How is this calculated and which assumptions are made? This is a crucial parameter in a cost-benefit analysis. Please provide more insights on this.

Response #4: (1) Thank you, agreed, this is very important. We use both measures to partly account for age distribution and problems with averaging. VSLY is truncated at an average life expectancy of 82.5 years, which underestimates damages. A-VSL picks up this point and accounts for the fact that most patients who die from COVID in Australia are elderly. (2) We rely on Abelson (2007), the standard reference used by government in Australia to obtain the $4.9 million figure and take greater care to articulate the procedure and assumptions that are adopted here. Please see section 3.4, including the added text: “The estimates in Abelson (2007) are drawn from an extensive meta-analysis of prior VSLY and VSL estimates and includes an overall discussion on the major methods of valuation and empirical results for values of life, health and safety in Australia. It also suggests adjustments for those 70 years and older, although we rely instead on Albertini et al. (2004) as more technically robust. As usual in cost-benefit studies, Abelson (2007) adopts an average WTP value for life, adjusted for age, and thus the VSL is generally held constant regardless of the income of any social group either at any point in time or over time. Most importantly, as conventional, it assumes as a starting point the life of an adult of 40 years of age, likely to live for another 40 years, with again an adjustment for those over 70 years of age.”

Comment #5. Australia has clear guidelines for health technology assessments, which are not fully addressed here (https://pbac.pbs.gov.au/information/about-the-guidelines.html) Please make sure all aspects of section 3 “Economic evaluation” are included and reported in the study.

Response #5: Thank you. As you suggested, we have revised our paper to highlight the link you have checked to make certain that we have captured all aspects of section 3 of the guidelines. In particular, as listed in section 3A of the guidelines (Cost-Effectiveness Analysis), our revised paper includes an overview and rationale of the economic evaluation, and a description of the computational method, and population, including model variables and extrapolation. Our paper also includes and discusses health outcomes, health care resource uses and costs, model validation, results for the base-case (our ‘unmitigated spread’ or ‘delayed suppression’ case, depending on context), economic evaluation, as well as uncertainty analysis.

 

Reviewer #2

Comment #6: Abstract:

• Early (actual): not very clear what you mean with actual

• Second sentence is long (3/4 of the abstract) and hard to read; rephrase

• Clarify the methods used

Response #6:

• We have removed the term ‘actual’ from the abstract, it is confusing. Active cases are defined clearly in the paper in any case.

• Thank you, good point. We have rephrased the second sentence, re-worked the abstract and added detail on the methods used.

Comment #7: Introduction

• As of the end of May 2020, the global number of COVID-19 cumulative cases and reported fatalities, respectively, exceed 6 million and 370,000 --> worldwide?

• Update the data -->� once accepted use the most recent available data

• (including the President of the United States on 22 March 2020) �--> not sure if this adds much. Others including scientists, were also posing this question.

• Several language issues: More costly in terms of the economy ; that we simulate continued until the end of May

• The structure is not logical. Research question comes rather early. Background info is limited. Last paragraph should be moved to method section, results are already given in introduction

• 2. COVID-19 in Australia and policy responses -->� include this in introduction section ; this is the background that we need in order to introduce research question

• Information included under heading 2 can also be shortened -->� too much detail about progress of pandemic. Main question is, why do we need these analyses?

Response #7:

• Yes, global data available at the time.

• We are going to retain the data relevant to the May 2020 period since it fits the context of the paper and the comparisons we make.

• We have removed the statement from the President, rightly so. Thank you.

• Thank you. We have carefully checked and revised the text.

• Thank you. We have revised the Introduction and, in particular, removed the results from this section.

• It appears to us that Section 2 follows logically from the Introduction, so we have left it as is and we prefer to keep all of the detail (on active case numbers and including the timing of the suppression controls) since it provides the needed context. We have shifted some of the material in the Introduction to this section.

Comment #8: Methods:

• excluded the below-20-year-old population group -->� most recent evidence indicated that you should lower the lower bound. Adolescents are also susceptible

• data source not clear

• I do not completely agree with QALY comment, luckily the adjustment for age covers this partially --> discuss this issue in more detail. + effect of age adjustment (scenario analyses of different adjustments)

• Life expectancy for given age should be used

• Why not using the GDP as VSLY?

• You only look at PRO of lock down (effect on COVID deaths) but not at CON (effect of lock down on general population, delay in usual care, increase in more severe diagnosis after lock down, etc…)

• How did you calculate VSLY for recovered patients?

• Why not looking at (modelled) excess mortality compared to previous year?

Response #8:

• We have revised this section to indicate that we have not excluded the under 20-year group in our modelling, while still noting throughout the paper that almost all of the fatalities are over 60 years old.

• We have double-checked to make certain that the references are clear. Data on all compartment categories is official government data, as indicated, mostly drawn from Covid-19-Data (2020).

• Thank you, we have added further discussion of ‘age adjustments’ to the welfare measures throughout but left the calculations as a .70 reduction for those over 70 years unchanged for the adjusted value of a statistical life, following Abelson (2007). Please also see section 3.4, including the added text: “The estimates in Abelson (2007) are drawn from an extensive meta-analysis of prior VSLY and VSL estimates and includes an overall discussion on the major methods of valuation and empirical results for values of life, health and safety in Australia. It also suggests adjustments for those 70 years and older, although we rely instead on Albertini et al. (2004) as more technically robust. As usual in cost-benefit studies, Abelson (2007) adopts an average WTP value for life, adjusted for age, and thus the VSL is generally held constant regardless of the income of any social group either at any point in time or over time. Most importantly, as conventional, it assumes as a starting point the life of an adult of 40 years of age, likely to live for another 40 years, with again an adjustment for those over 70 years of age.”

• We only have life expectancy for a given age as it relates to average life expectancy. We have clarified this point.

• We have measures for the fall in GDP but also included VSLY and A-VSL because these are standard measures.

• We have briefly highlighted the negative impacts of ‘lockdown’, while still noting that many of these impacts were mitigated through government assistance programs in the Discussion section. We have also re-emphasized that morbidity effects from COVID are not included. Unfortunately, there is no available quantitative measures to be able to include in our modelling.

• Thank you. We measure VSLY only for mortalities and account for hospitalisation and other costs for those patients who die and ultimately recover separately. We have clarified this point. We do account for the average number of sick days of patients who do recover, specified at 18.5 days on average.

• Thank you. We have provided additional references to acknowledge this point, just before section 4.1.

Comment #9: Results

• It would be very informative to see the results of individual different measures included in lock down--> closure of schools, shut down of non-essential business, social distancing.

Response #9: Agreed, unfortunately there is insufficient data for us to include in our modelling.

Comment #10: Discussion

• The discussion lacks comparison with other studies. It is merely a summary of own findings.

• Problem with +82 year olds could be solve by using life expectancy at given age

• What was the influence of time? We see that the number of deaths is now decreasing, even though less measures are in place compared to lock down? How should we look at upcoming months? Go hard, go early? Of a more adapted scenario?

Response #10:

• Thank you. To the best of our knowledge there is no published material or pre-print that combines an epidemiological model with the impacts on the economy from various mandated suppression measures (including the case of unmitigated spread) for Australia – save for a non-technical discussion in Group of Eight Australia (2020) and also Grafton et al. (2020) which are cited in the revised manuscript.

• We only have access to average life expectancy but note your suggestion/qualification in the revised paper.

• Thank you. We contend (based on our modelling results) that a ‘go hard, go early’ is the preferred public policy choice.

Submitted filename: Response to Reviewers.docx

 10.1371/journal.pone.0252400.r003 Decision Letter 1 Devleesschauwer Brecht Academic Editor 2021 Brecht Devleesschauwer 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. Submission Version 1

25 Feb 2021

PONE-D-20-18242R1

Health and Economic Costs of Early and Delayed Suppression and the Unmitigated Spread of COVID-19: The Case of Australia

PLOS ONE

Dear Dr. Kompas,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Apr 11 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Brecht Devleesschauwer

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments (if provided):

Thank you for addressing the reviewer comments. Reviewer #1 raised some further remarks, which can be addressed in a final, minor revision round. In your response to Reviewer #2, I had also noted that you had not always provided an adequate response -- for each comment, please make sure to refer us to changes made in the manuscript, or indicate why you decided not to change the manuscript. Only providing a reply in the revision note is not sufficient, because readers of the article will not have access to these notes.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors improved their manuscript and resolved many issues from both reviewers. Thank you for your gratitude and by taking our feedback seriously. The current message is more nuanced and thus informative, in my opinion. I still have a few comments.

The authors replied on my question (#1) why the reported cases dropped at the beginning of April. The added text (line 175-180) is rather long and difficult to read. However, changes in reporting strategy and other administrative issues have taken place worldwide and interfere with modelling, though my question was, “how did your SIQRM model capture this sudden drop”? What kind of temporal parameter(s) did you include to capture this observed behavior? How did you inform these parameters? And what consequences does this additional temporal “model intervention” has for making predictions?

The counterfactual scenario on “no suppression” is well explained in the revised manuscript, though the additional note that “a lockdown would likely ensure” seems to undermine the scenario. This is stated at line 206 and 440. In my comment #2, I referred to “a total breakdown of the medical system will be even more deadly than the estimates, since there are not enough e.g. ICU’s to save severe cases and all other care is not possible. The costs are uncertain, since you cannot pay for ICU’s or ventilators that are not there. Many patients will not be admitted to ICU, though you include unlimited ICU costs”. The bottom line of this scenario, is that there is no lockdown.

On line 211, the authors explicitly challenge the usefulness of QALYs for cost-effectiveness comparisons among alternative public health responses. Unless they have a vast amount of literature to backup this claim, I would just mention that the analysis is based on life-years-lost and not on QALYs, in line with the Guidelines.

On line 244, the authors report their scaling of the VSL of 4.9million dollar by 0.7 for people over 70 years of age. This is not clear to me. So, 70-year olds represents only/still 30% of their monetary value to society? Please explain this more for the reader.

The impact of mandated suppression on tourism seems not completely fair. It is sure that the measures in place prevented people from coming to Australia, and this has an economic cost. However, this pandemic also affects outgoing passenger flows in other countries. Even without lockdown, there will be economic loses for tourism. Hence, accounting the full economic loss for tourism to the lockdown requires some remarks.

Line 421 states that “delays in suppression provide no economic gain, but increased fatalities and …). This gives the impression that the economic cost of the suppression does not depend on the timing. Only the fatalities are impacted by the timing of the suppression measures. Would it be possible to elaborate more on this?

The following statement in the abstract and partly repeated in the discussion (line 449) is not clear to me:

"We also find that using an equivalent VSLY welfare loss from fatalities to estimated GDP losses, drawn from survey data and our own estimates of the impact of suppression measures on the economy, means that for early suppression not to be the preferred strategy requires that Australians prefer more than 12,500 - 30,000 deaths, depending on the fatality rate, to the economy costs of early mandated suppression. "

This reads like Australians have to choose between early suppression or 12500+ deaths. This seems not the take-home message from this manuscript. What about the nuance between delayed suppression? In addition, this assumes a fixed willingness-to-pay which is rather a complex concept and a subject for discussion.

Minor comments

- “Active cases” is rather uncommon. The number of active cases over time can be indicated by the prevalence.

- Line 60, how did the authors calculated the daily growth rate?

- The manuscript contains ‘Error! Reference source not found”

- Line 325: what do the authors mean with “Some 84 percent of businesses…” ?

- Line 344 misses the Figure number

- Line 436-437 is not clear to me.

- Line 432, what do the authors mean with “the spread of COVID-19” ?

- Line 471, please rephrase “public health costs and economy costs”.

- In the equations, the letter “T” is re-used in different equations. This is confusing.

- Line 105, R0 is defined as the basic reproduction number and not the “initial” reproduction number.

- The manuscript does not include any comment or remark that this model is an abstraction of the incubation, symptomatic, pre-symptomatic and asymptomatic infection stages. The model does not include any time lag between infection and infectiousness, which has impact on the timing of the predicted epidemic. The model structure is clearly stated and fits the purpose, though a remark on this issue would put the analyses in perspective.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

 10.1371/journal.pone.0252400.r004 Author response to Decision Letter 1 Submission Version 2

30 Apr 2021

Response to Reviewer(s)

Thanks again to Reviewer #1 for providing very helpful comments. We greatly appreciate the time she/he has spent with our paper! Our responses are below.

Reviewer #1: The authors improved their manuscript and resolved many issues from both reviewers. Thank you for your gratitude and by taking our feedback seriously. The current message is more nuanced and thus informative, in my opinion. I still have a few comments.

1. The authors replied on my question (#1) why the reported cases dropped at the beginning of April. The added text (line 175-180) is rather long and difficult to read. However, changes in reporting strategy and other administrative issues have taken place worldwide and interfere with modelling, though my question was, “how did your SIQRM model capture this sudden drop”? What kind of temporal parameter(s) did you include to capture this observed behavior? How did you inform these parameters? And what consequences does this additional temporal “model intervention” has for making predictions?

Response: Thank you for this point. We have clarified the text to read: ‘There are two additional points to make. First, note that the significant drop in active cases at the beginning of April (and another drop in the third week of April) results from a spike in the number of cases which were classified as ‘recovered’-- not all of which are predictable by epidemiological parameters alone because how a case was classified as recovered could depend on administrative and reporting procedures. For this reason, during the period where actual data was used to estimate the epidemiological model (i.e., March 1st to April 20th, 2020), we combine the fitted total number of cases (using epidemiological parameters only) and the daily recovery classification rate calculated from the actual data to estimate the number of cases which were reported as recovered. To test the prediction capacity or the accuracy of the epidemiological model, we used a period where the data was ‘new’ to the model (i.e., April 21st – May 11th, 2020), and in this testing situation, the number of ‘recovered’ cases was predicted by combining the epidemiologically predicted total number of cases with the average number of sick days of patients who recover, i.e., approximately 18.5 days (Covid19-Data, 2020).’

2. The counterfactual scenario on “no suppression” is well explained in the revised manuscript, though the additional note that “a lockdown would likely ensure” seems to undermine the scenario. This is stated at line 206 and 440. In my comment #2, I referred to “a total breakdown of the medical system will be even more deadly than the estimates, since there are not enough e.g. ICU’s to save severe cases and all other care is not possible. The costs are uncertain, since you cannot pay for ICU’s or ventilators that are not there. Many patients will not be admitted to ICU, though you include unlimited ICU costs”. The bottom line of this scenario, is that there is no lockdown.

Response: Understood, thank you. Our only point is that if the capacity of the medial system is exceeded the government position would indeed be to enforce a severe lockdown. Our counterfactual ignores this, of course. We have adjusted the text (near line 206) to clarify the point.

3. On line 211, the authors explicitly challenge the usefulness of QALYs for cost-effectiveness comparisons among alternative public health responses. Unless they have a vast amount of literature to backup this claim, I would just mention that the analysis is based on life-years-lost and not on QALYs, in line with the Guidelines. On line 244, the authors report their scaling of the VSL of 4.9million dollar by 0.7 for people over 70 years of age. This is not clear to me. So, 70-year olds represents only/still 30% of their monetary value to society? Please explain this more for the reader.

Response: Thank you. We have removed this part of the discussion on QALYs.

Response: We have kept the text as is, although certainly acknowledge that the value of a statistical life can be calculated in different ways and, depending on the method used, may vary with a person's age. If based on future earnings, then persons who are young adults may have a higher relative VSL to an older demographic. If based on ability and willingness to pay (WTP) measures, middle-aged or older persons may have a higher relative VSL to a younger demographic. We followed the findings of Alberini et al. (2004) who found weak support that WTP declines with age, and only for older respondents (aged 70 and above) in their surveys of two populations (one in Canada and one in the US). They found that their Canadian survey respondents over the age of 70 were willing to pay about one-third less than their younger counterparts to reduce their risk of dying by 5 in 1000 over the next 10 years. Thus, we adopted the heuristic to reduce VSL of persons over 70 years of age by 0.30 noting that a similar heuristic (0.75 X VSL for persons 65 years and older) has previously been employed by Health Canada.

4. The impact of mandated suppression on tourism seems not completely fair. It is sure that the measures in place prevented people from coming to Australia, and this has an economic cost. However, this pandemic also affects outgoing passenger flows in other countries. Even without lockdown, there will be economic loses for tourism. Hence, accounting the full economic loss for tourism to the lockdown requires some remarks.

Response: We agree. The decline in tourism to Australia is primarily a result of the supervised 14-days quarantine on all arrivals, the cost of which must be paid by those arriving in Australia. This is approximately $3,500 per person. Supervised quarantine requires a large number of quarantine workers and suitable rooms to isolate travellers. Resourcing constraints, particularly a sufficient number of beds in appropriate hotels, has also meant the that the number of people permitted to come to Australia is about 20,000 per month. (see https://www.abs.gov.au/statistics/industry/tourism-and-transport/overseas-arrivals-and-departures-australia/latest-release) while pre-COVID in 2019 arrivals were, on average, 700,000 per month. Further, these constraints on arrivals continue despite the fact that the national lockdown ended in May 2020. In sum, attributing the full loss in tourism to the lockdown, as we have done, overestimates the lockdown cost. Thus, in the revised version we state: 'In summary, we attribute all the costs of reduction in international tourism to the lockdown. This overstates the costs of lockdown because a decline in international tourism would have occurred even in the absence of stringent social distancing. This is because of Australia's international border controls limit the number of arrivals into Australia and requires supervised quarantine of 14 days. These international border controls have been continuously maintained even when national lockdown ended in May 2020.' (line 376)

5. Line 421 states that “delays in suppression provide no economic gain, but increased fatalities and …). This gives the impression that the economic cost of the suppression does not depend on the timing. Only the fatalities are impacted by the timing of the suppression measures. Would it be possible to elaborate more on this?

Response: Thanks for this. We have added your qualification.

6. The following statement in the abstract and partly repeated in the discussion (line 449) is not clear to me: "We also find that using an equivalent VSLY welfare loss from fatalities to estimated GDP losses, drawn from survey data and our own estimates of the impact of suppression measures on the economy, means that for early suppression not to be the preferred strategy requires that Australians prefer more than 12,500 - 30,000 deaths, depending on the fatality rate, to the economy costs of early mandated suppression.” This reads like Australians have to choose between early suppression or 12500+ deaths. This seems not the take-home message from this manuscript. What about the nuance between delayed suppression? In addition, this assumes a fixed willingness-to-pay which is rather a complex concept and a subject for discussion.

Response: Yes, agreed, good point. The use of ‘Australians prefer’ is clearly bad form. We’ve reworded the material (in the abstract and conclusion) to pick this up and the point on unmitigated spread, indicating just model results. The abstract reads ‘ … from survey data and our own estimates of the impact of suppression measures on the economy, means that for early suppression not to be the preferred strategy requires that Australia would have to incur more than 12,500 - 30,000 deaths, depending on the fatality rate with unmitigated spread, to the economy costs of early mandated suppression. We also find that early rather than delayed mandated suppression imposes much lower economy and health costs and conclude that in high-income countries, like Australia, a ‘go early, go hard’ strategy to suppress COVID-19 results in the lowest estimated public health and economy costs.’ We have left aside your point of assuming a fixed willingness to pay as indeed too complex in this version of the paper – something for further research.

7. Minor comments

- “Active cases” is rather uncommon. The number of active cases over time can be indicated by the prevalence.

Response: We have kept it as is, as standard practice in Australia at the time. Active cases are the numbers of individuals identified and in quarantine.

- Line 60, how did the authors calculated the daily growth rate?

Response: We have clarified this in the text to read: ‘As of 1st June 2020, Australian mandated suppression measures drastically reduced community transmission of the virus and the daily growth rate (the daily increase in the total number of cases over the total number of cases, on a three-day average) declined from around 25%, with 268 new recorded cases on 22nd March, to a daily growth rate of 0.26% and 11 new recorded cases on 30th May 2020 (Covid19-Data, 2020).’

- The manuscript contains ‘Error! Reference source not found”

Response: Repaired. It came through this way in the PDF but is fine in the Word document. In any case, we have re-entered the source.

- Line 325: what do the authors mean with “Some 84 percent of businesses…” ?

Response: It is 84%. We have removed the ‘some’.

- Line 344 misses the Figure number

Response: Repaired, thanks.

- Line 432, what do the authors mean with “the spread of COVID-19” ?

Response: Thanks, we have removed this sentence. It wasn’t needed.

- Line 471, please rephrase “public health costs and economy costs”.

Response: Repaired, thanks.

- In the equations, the letter “T” is re-used in different equations. This is confusing.

Response: Thank you for this point. We have revised the paper to clarify that T (without the superscript) is the total number of observed cases, and the transpose matrix operator, which was also denoted as T, is now denoted as Tr.

- Line 105, R0 is defined as the basic reproduction number and not the “initial” reproduction number.

Response: We have revised this sentence as suggested, thank you.

- The manuscript does not include any comment or remark that this model is an abstraction of the incubation, symptomatic, pre-symptomatic and asymptomatic infection stages. The model does not include any time lag between infection and infectiousness, which has impact on the timing of the predicted epidemic. The model structure is clearly stated and fits the purpose, though a remark on this issue would put the analyses in perspective.

Response: Thank you. We have included a revised version of your qualifier in the technical section (near line 101), which reads as ‘It is important to note that the model, although fit for purpose, is an abstraction of the relevant infection stages and does not include any explicit time lag between infection and infectiousness, which has impact on the practical timing of the predicted epidemic in Australia.’

Submitted filename: Response to Reviewers.docx

 10.1371/journal.pone.0252400.r005 Decision Letter 2 Devleesschauwer Brecht Academic Editor 2021 Brecht Devleesschauwer 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. Submission Version 2

17 May 2021

Health and Economic Costs of Early and Delayed Suppression and the Unmitigated Spread of COVID-19: The Case of Australia

PONE-D-20-18242R2

Dear Dr. Kompas,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Brecht Devleesschauwer

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

 10.1371/journal.pone.0252400.r006 Acceptance letter Devleesschauwer Brecht Academic Editor 2021 Brecht Devleesschauwer 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.

25 May 2021

PONE-D-20-18242R2

Health and Economic Costs of Early and Delayed Suppression and the Unmitigated Spread of COVID-19: The Case of Australia

Dear Dr. Kompas:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Dr. Brecht Devleesschauwer

Academic Editor

PLOS ONE