Table 2.

Evidence table of the included studies.

Author (Country)	Intervention		Epidemiologic model	Health economic evaluation
	Population	Strategies	COVID-19 model	Primary outcome measure	Type of evaluation	Time horizon	Perspective	Author conclusion	CHEC
1. Protection
Savitsky and Albright30 (United States)	HCWs on labor and delivery	A.1) Universal COVID-19 screening if vaginal delivery (spontaneous labor) A.2) Universal PPE used if vaginal delivery (spontaneous labor) B.1) Universal COVID-19 screening if vaginal delivery (induced labor) B.2) Universal PPE used if vaginal delivery (induced labor) C.1) Universal COVID-19 screening if cesarean section C.2) Universal PPE used if CD	Decision tree to model transmission of SARS-CoV-2 to HCWs	ICER: cost/prevent one COVID-19 infection in an HCW (WTP = $25 000: estimate of immediate cost of a COVID-19 infection of a HCW)	Cost-effectiveness	NA	Not reported (only cost of testing + costs of PPE are included, limited/restricted healthcare perspective)	At relatively low prevalence of disease (<10%), universal screening is the preferred strategy for women presenting in spontaneous labor and for labor induction. Interestingly for a planned CD universal PPE was more often cost-effective, and therefore, the preferred strategy as long as the cost of PPE remained stable. At high disease prevalence, universal PPE is the best strategy to protect HCW.	13
Risko et al22 (139) (LMIC)	HCWs	1)Inadequate PPE: absence of one or more of the PPE elements 2)Full PPE on the basis of the WHO best practice guidelines (EFST): gloves, gown, face shield and masks for all encounters involving a suspected case and enhanced precautions for aerosol generating procedures	SIR (ESFT)—(S) susceptible, (I) infected, and (R) recovered	ICER: cost/HCW death averted; ICER = cost/HCW case averted (no WTP reported)	Cost-effectiveness; ROI analysis	30-week period	Societal perspective	Immediate investment in the wide-scale production and distribution of PPE for LMICs yields a significant benefit in lives saved and ROI. The authors also conclude that this public health strategy is required to prevent massive depletion of the healthcare workforce.	11
Ebigbo et al32 (Germany)	Patients presenting for endoscopy	1)No routine pre-endoscopy virus test; use of surgical masks, goggles, gloves, and apron for all procedures 2)No routine pre-endoscopy virus test; additional use of FFP-2 and water-resistant gowns for all procedures 3)Decentralized POC antigen test; use of surgical masks, goggles, gloves, and apron for all procedures 4)Decentralized POC antigen test; additional use of FFP-2 and water-resistant gowns for all procedures irrespective of test result 5)Centralized laboratory-based rapid PCR test; use of surgical masks, goggles, gloves, and apron for all procedures 6)Centralized laboratory-based rapid PCR test; additional use of FFP-2 and water-resistant gowns for all procedures irrespective of test result 7)Centralized laboratory-based standard PCR test; use of surgical masks, goggles, gloves, and apron for all procedures 8)Centralized laboratory-based standard PCR test; additional use of FFP-2 and water-resistant gowns for all procedures irrespective of test result	Decision tree to model transmission of SARS-CoV-2 from asymptomatic patients to HCWs in high-volume centers	ICER: cost/positive test (no WTP reported)	Cost-effectiveness	Not reported	Not reported (on the basis of included costs it can be considered a limited societal perspective)	ICER values for universal testing decreased with increasing prevalence rates. For higher prevalence rates (≥1%), ICER values were the lowest for routine pre-endoscopy testing coupled with the use of high-risk PPE, whereas cost per endoscopy was the lowest for routine use of high risk PPE without universal testing. In general, routine pre-endoscopy testing combined with high-risk PPE becomes more cost-effective with rising prevalence rates of COVID-19.	6
	2. Detection
Neilan et al18 (United States)	People with COVID-19 symptoms	1)PCR testing only in patients with severe/critical symptoms warranting hospitalization. (Re = 0.9, 1.3, 2.0) 2)PCR testing for any COVID-19 consistent symptoms with self-isolation when positive. (Re = 0.9, 1.3, 2.0) 3)PCR testing for symptomatic patients and one-time PCR for entire population. (Re = 0.9, 1.3, 2.0) 4)PCR testing for all symptomatic persons and monthly re-testing for the entire population. (Re = 0.9, 1.3, 2.0)	Extended SEIRD (CEACOV)-(S) susceptible, (E) exposed (latent, noninfectious), (Ip/Ia) pre- and asymptomatic, (Im) mildly infected, (Is) severely infected, (Ic) critically infected, (Ir) recuperation after critical infection, (R) recovered, and (D) deceased	ICER: cost/QALY (WTP: $100 000/QALY)	Cost-effectiveness	180-day horizon	Healthcare system perspective	Testing people with any COVID-19-consistent symptoms would be cost saving compared with testing only those whose symptoms warrant hospital care. Expanding PCR testing to asymptomatic people would decrease infections, deaths, and hospitalizations. Despite modest sensitivity, at low-cost, repeated screening of the entire population could be cost-effective in all epidemic settings.	18
Jiang et al16 (China)	People suspected of having COVID-19	1)Two RT-PCR tests for diagnosing and discharging people with COVID-19 2)Three RT-PCR tests for diagnosing and discharging people with COVID-19	Extended SEIRD (SALIRD)-(S) susceptible, (A) asymptomatic, (L) presymptomatic, (I) infectious, (R) recovered, and (D) deceased	ICER: cost/QALY (WTP = CN¥64 644); NMB	Cost-effectiveness	January 2020 to March 2020 (43 days)	Healthcare system perspective	The three-test strategy is a dominant strategy in all scenarios.	18
Paltiel et al12 (United States)	Students (<30 years old and nonimmune, living in a congregate setting)	1)Weekly screening (Re = 1.5, 2.5, 3.5; sensitivity = 70%, 80%, 90%) 2)Every 3 days screening (Re = 1.5, 2.5, 3.5; sensitivity = 70%, 80%, 90%) 3)Every 2 days screening (Re = 1.5, 2.5, 3.5; sensitivity = 70%, 80%, 90%) 4)Daily screening (Re = 1.5, 2.5, 3.5; sensitivity = 70%, 80%, 90%)	Extended SEIRD—(S) susceptible, (E) exposed (latent, noninfectious), (Ia) asymptomatic infectious, (Is) symptomatic infectious, (R) recovered, (D) deceased	ICER: screening costs/infection averted (WTP = $100 000 per year-of-life gained; a maximum WTP to avert 1 infection ranging from $7500 (Re = 1.5) to $10 500 (Re = 2.5) to $13 500 (Re = 3.5) + budget impact)	Cost-effectiveness/budget impact	80 days	Not reported (only costs of screening were considered. On the basis of the included costs it can be considered a restricted approach/perspective)	There is a safe way for students to return to college in the Fall of 2020. The question is whether it is feasible today on a large scale. Coupled with strict behavioral interventions that keep Re below 2.5, a rapid, inexpensive and even poorly sensitive (>70%) test, conducted at least every 2 days, would produce a modest number of containable infections and would be cost-effective.	13
Paltiel et al19 (United States)	General population	A.0) Re = 0.9: no test scenario A.1) Re = 0.9: weekly home-based SARS-CoV-2 antigen testing B.0) Re = 1.3: no test scenario B.1) Re = 1.3: weekly home-based SARS-CoV-2 antigen testing B.0) Re = 1.7: no test scenario B.1) Re = 1.7: weekly home-based SARS-CoV-2 antigen testing B.0) Re = 2.8: no test scenario B.1) Re = 2.8: weekly home-based SARS-CoV-2 antigen testing	Extended SEIRD—(S) susceptible, (E) exposed (latent, noninfectious), (Ip/Ia) pre- and asymptomatic, (Im) mildly symptomatic, (Is) severely symptomatic, (Ic) critically symptomatic, (R) recovered, and (D) deceased	ICER: costs/infections averted and costs/deaths averted (Value of statistical life saved = $5.3 million)	Cost-effectiveness	60 days	Societal perspective	High-frequency home testing for SARS-CoV-2 with an inexpensive, imperfect test could contribute to pandemic control at a justifiable cost and warrants consideration as part of a national containment strategy.	15
Du et al20 (United States)	General population-households	A.1) Re = 1.2: daily test plus 1-week isolation A.2) Re = 1.2: daily test plus 2-week isolation A.3) Re = 1.2: test every 7 days plus 1-week isolation A.4) Re = 1.2: test every 7 days plus 2-week isolation A.5) Re = 1.2: test every 14 days plus 1-week isolation A.6) Re = 1.2: test every 14 days plus 2-week isolation A.7) Re = 1.2: test every 28 days plus 1-week isolation A.8) Re = 1.2: test every 28 days plus 2-week isolation B.1) Re = 2.2: daily test plus 1-week isolation B.2) Re = 2.2: daily test plus 2-week isolation B.3) Re = 2.2: test every 7 days plus 1-week isolation B.4) Re = 2.2: test every 7 days plus 2-week isolation B.5) Re = 2.2: test every 14 days plus 1-week isolation B.6) Re = 2.2: test every 14 days plus 2-week isolation B.7) Re = 2.2: test every 28 days plus 1-week isolation B.8) Re = 2.2: test every 28 days plus 2-week isolation	Extended SEIRD—(S) susceptible, (E) exposed (latent, noninfectious), (Ip/Ia) pre- and asymptomatic, (Im) mildly symptomatic, (Ih) hospitalized, (R) recovered, (D) deceased	Net benefit (WTP = $100 000 per YLL averted)	Cost-benefit	Not reported	Not reported (on the basis of included costs it can be considered a limited societal perspective)	Assuming a WTP of $100 000 per YLL averted and a price of $5 per test, the strategy most likely to be cost-effective under a rapid transmission scenario (Re = 2.2) is weekly testing followed by a 2-week isolation period subsequent to a positive test result. Under low transmission scenarios (Re = 1.2), monthly testing of the population followed by 1-week isolation rather than 2-week isolation is likely to be most cost-effective. Expanded surveillance testing is more likely to be cost-effective than the status-quo testing strategy if the price per test is less than $75 across all transmission rates considered.	11
	3. Prevention/containment
Miles et al24,25 (United Kingdom)	General population	0)Doing nothing (no change in behavior) 1)Lockdown	Extended SEIRD (Imperial College COVID-19 Response Team model)38 (Ferguson et al)	Total damage (WTP = £30 000/QALY)	Cost-benefit	March-July 2020	Not reported (on the basis of the included costs it can be considered as a partial societal perspective)	The costs of the 3-month lockdown in the UK are likely to have been high relative to the benefits. According to the authors there is a need to normalize how we view COVID-19, because its costs and risks are comparable with other health problems (such as cancer, heart problems, diabetes).	9
Zala et al10 (United Kingdom)	General population	0)No measures 1)Mitigation policy: individual case isolation, home quarantine (ie, quarantine of a household with a suspected case), and social distancing advice for people over 70 years of age 2)Suppression strategy 1 = mitigation policy plus general social distancing and closure of schools and universities: triggered “on” when there are 100 ICU cases in a week and “off” when weekly cases halve to 50 cases 3)Suppression strategy 2 = suppression strategy 1 triggered “on” when there are 400 ICU cases in a week and “off” when weekly cases halve to 200 cases	Extended SEIRD (Imperial College COVID-19 Response Team model)38 (Ferguson et al)	ICER (WTP = £20 000-30 000; according to NHS or more general estimates of the social value of a QALY between £10 000 and £70 000)	Cost-effectiveness	March-July 2020	Not reported (on the basis of the included costs it can be considered a societal perspective)	Suppression polices were compared with an unmitigated pandemic. Even the most pessimistic National income loss scenarios under suppression (10%), give ICERs below £50 000 per QALY. Assuming a maximum reduction in national income of 7.75%, the ICERs of suppression vs mitigation are below 60 000 per QALY.	17
Asamoah et al26 (Ghana)	General population	1)Effective testing and quarantine when boarders are opened 2)Intensifying the usage of nose masks and face shields through education 3)Cleaning of surfaces with home-based detergents 4)Safety measures adopted by the asymptomatic and symptomatic individuals such as; practicing proper cough etiquette (maintaining a distance, cover coughs and sneezes with disposable tissues or clothing and wash hands after cough or sneezes) 5)Fumigating commercial areas such as markets 6)Combines the use of controls of strategy 1 to 5	Extended SEIR(D)—susceptible (S), exposed (E), asymptomatic (A), infectious (I), recovered (R), and virus on surfaces (V)	ICER: cost of control strategies/averted infections by control strategies (No WTP reported)	Cost-effectiveness	March 12 to May 7, 2020	Not reported (on the basis of included costs it can be considered a limited/restricted healthcare perspective)	Strategy 4 is the most cost-effective strategy: safety adopted by the asymptomatic and symptomatic individuals such as practicing proper cough etiquette by maintaining a distance, covering coughs and sneezes with disposable tissues or clothing and washing hands after coughing or sneezing.	5
Reddy et al17 (South Africa)	General population	1)HT (Re = 1.2; Re = 1.5) 2)= 1 + CT (Re = 1.2; Re = 1.5) 3)= 2 + IC (Re = 1.2; Re = 1.5) 4)= 3 + mass SxScreen (Re = 1.2; Re = 1.5) 5)= 3 + QC (Re = 1.2; Re = 1.5) 6)= 4 + quarantine center (Re = 1.2; Re = 1.5)	Extended SEIRD (CEACOV)—(S) susceptible, (E) exposed (latent, noninfectious), (Ip/Ia) pre- and asymptomatic, (Im) mildly infected, (Is) severely infected, (Ic) critically infected, (Ir) recuperation after critical infection, (R) recovered, and (D) deceased	ICER: the difference in healthcare costs divided by the difference in life years between strategies (WTP: $3250/YLS)	Cost-effectiveness	360 days	Public/private health sector perspective	A strategy combining all interventions would cost an additional $340 per year-of-life saved, which compares favorably with the cost-effectiveness of many established public health interventions in South Africa. With low epidemic growth (Re = 1.1-1.2): HT + CT + IC + QC was the optimal strategy; QCs remained cost-effective but adding MS was not cost-effective. With high epidemic growth (Re = 2.6), when the epidemic outpaced control measures and costs increased substantially, no combination of the modeled interventions was cost-effective compared with HT alone.	18
Khajji et al27 (not reported)	General population	1)Awareness campaign to protect susceptible individuals from contacting the infected individuals in the same region 2)Security campaigns and health measures protecting and preventing susceptible individuals from contacting the infected individuals in the same region or in other regions 3)Protecting susceptible individuals, preventing their contact with the infected individuals, and encouraging the exposed individuals to join QCs 4)Protecting susceptible individuals, preventing their contact with the infected individuals, encouraging the exposed individuals to join quarantine centers, and the disposal of the infected animals	SEIQRD with SI model for animals —susceptible (S), exposed (E), infectious (I), quarantined (Q) and recovered (R), (D) discrete time	ICER: delta cost/delta averted infections (No WTP reported)	Cost-effectiveness	Not reported	Not reported (unclear what costs are included)	Strategy 3 is most cost-effective.	6
Thunström et al23 (United States)	General population	1)Without social distancing 2)With social distancing	SIRD—susceptible (S), infectious (I), recovered (R), and deceased (D)	Net benefit (incremental GDP loss vs value of lives saved). Value of statistical life = $10 million	Cost-benefit	30 years	Not reported (on the basis of the included costs it can be considered a: limited societal perspective)	The authors conclude that social distancing likely generates net social benefits ($5.16 trillion).	9
Shlomai et al21 (Israel)	General population	1)National lockdown + Individuals who have essential occupations (as determined by government decisions) will not be quarantined and will be required to maintain social distancing. All known exposed individuals will be completely isolated for a 14-day period 2)Testing, tracing, isolation (focused isolation of individuals at high exposure risk who will return to the workforce under social distancing measures after a 14-day isolation period)	Extended SEIRD—(S) susceptible, (E) exposed, (Ia) asymptomatic, (Is) symptomatic and infectious, (R) recovered, (D) deceased	ICER: cost/life saved (WTP for statistical life saved = $10 000 000; WTP per QALY = $15 243-17 366)	Cost-effectiveness	200 day period	Not reported (on the basis of included costs it can be considered a limited societal perspective)	Over time a strategy of national lockdown is moderately superior to a strategy of focused isolation in terms of reducing death rates but involves extremely high economic costs to prevent 1 case of death. A national lockdown has a moderate advantage in saving lives with tremendous costs and possible overwhelming economic effects.	8
Zhao et al13 (China)	General population	1)Current practice: the real-world scenario in China, where the first movement restriction policies started on January 23, 2020 and ended on March 25, 2020 2)1-week delay in the imposition of movement restriction policies (MRPs) (MRPs end on the day when national newly confirmed cases reach zero) 3)2-week delay in the imposition of MRPs (MRPs end on the day when national newly confirmed cases reach zero) 4)4-week delay in the imposition of MRPs (MRPs end on the day when national newly confirmed cases reach zero)	Extended SIRD—(S) susceptible, (Ip) presymptomatic, (Im) infectious with mild symptoms, (Ih) hospitalized, (R) recovered, and (D) deceased	Net benefit (WTP = 70 892 RMB)	Cost-benefit	Period less than a year	Societal and healthcare perspective	Strategy A (“current practice”) dominates all other strategies, from both a healthcare perspective and societal perspective. At a WTP of 70 892 RMB per DALY averted, the probability that strategy A is more cost-effective compared with strategy B, C, and D is 96%, 99%, 100%, respectively. Delay in initiating MRPs leads to exponential growth in DALY loss and societal cost: a 4-week delay resulted in 3.7 million more DALYs and 2942 billion USD additional societal cost, compared with no delay.	16
Losina et al15 (United States)	College students	0)No intervention 1)Minimal social distancing + ResIsol (residence isolation in student dorm room) + self-screen 2)Masks + ResIsol + self-screen 3)Minimal social distancing + DesigIsol (student quarantine in separate location) + self-screen 4)Masks + DesigIsol + Self-screen 5)Minimal social distancing + DesigIsol + 1-time LT 6)Extensive social distancing + ResIsol + self-screen 7)Masks + DesigIsol + 1-time LT 8)Extensive social distancing + DesigIsol + self-screen 9)Extensive social distancing + masks + ResIsol + self-screen 10)Extensive social distancing + DesigIsol + 1-time LT 11)Extensive social distancing + masks + DesigIsol + self-screen 12)Extensive social distancing + masks + DesigIsol 1-time LT 13)Minimal social distancing + DesigIsol + RLTq14 (routine LT every X days) 14)Masks + DesigIsol + RLTq14 15)Extensive social distancing + DesigIsol + RLTq14 16)Campus closed 17)Extensive social distancing + Masks + DesigIsol + RLTq14 18)Minimal Social Distancing + DesigIsol + RLTq7 19)Masks + DesigIsol + RLTq7 20)Extensive social distancing + DesigIsol + RLTq7 21)Extensive social distancing + Masks + DesigIsol + RLTq7 22)Minimal social distancing+ DesigIsol + RLTq3 23)Masks + DesigIsol + RLTq3 24)Extensive social distancing + DesigIsol + RLTq3	Extended SEIRD (CEACOV)—(S) susceptible, (E) exposed (latent, noninfectious), (Ip/Ia) pre- and asymptomatic, (Im) mildly infected, (Is) severely infected, (Ic) critically infected, (Ir) recuperation after critical infection, (R) recovered, and (D) deceased	ICER: cost/QALY (WTP: $150 000/QALY)	Cost-effectiveness	One semester (105 days)	Modified societal perspective	Extensive social distancing with mandatory use of a mask could prevent 87% of COVID-19 cases on college campuses and be very cost-effective. Routine LT would prevent 96% of infections and require low-cost tests to be economically attractive.	18
Baggett et al11 (United States)	Homeless adults	A.0) Re = 0.9: No intervention: only basic infection control practices are implemented in shelters A.1) Re = 0.9: SxScreen/PCR/ACS: CDC-recommended SxScreen daily in shelters. Screen-negative individuals remain in shelters. Screen-positive individuals are sent to an ACS for people under investigation, where they undergo PCR testing and await results. PCR-positive individuals with mild/moderate illness are transferred to ACSs for confirmed COVID-19 cases. PCR-negative individuals return to shelter A.2) Re = 0.9: universal PCR/hospital: universal PCR testing every 2 weeks in shelters. Those with symptoms at the time of testing await results at the hospital; individuals without symptoms await results in shelters. PCR-negative individuals return to or stay in shelters. PCR-positive individuals, regardless of illness severity, remain in or are sent to the hospital A.3) Re = 0.9: SxScreen/PCR/hospital: CDC-recommended SxScreen daily in shelters. Screen-negative individuals remain in shelters. Screen-positive individuals are sent to the hospital for PCR testing. PCR-positive individuals remain in hospital; PCR-negative individuals return to shelter A.4) Re = 0.9: universal PCR/ACS: universal PCR testing every 2 weeks in shelters. Those with symptoms at the time of testing are sent to an ACS for people under investigation while awaiting results; individuals without symptoms await results in shelters. PCR-negative individuals return to or stay in shelters. PCR-positive individuals with mild/moderate illness are transferred to ACSs for confirmed COVID-19 cases A.5) Re = 0.9: universal PCR/TempHousing: All shelter residents are pre-emptively moved to TempHousing for the duration of the 4-month period. Universal PCR testing occurs every 2 weeks. PCR-positive individuals with mild/moderate illness remain in TempHousing and are transferred to the hospital if they progress to severe or critical disease A.6) Re = 0.9: hybrid/hospital: this includes the SxScreen/PCR/hospital strategy and adds shelter-based universal PCR testing every 2 weeks for those without symptoms A.7) Re = 0.9: Hybrid/ACS: this includes the SxScreen/PCR/ACS strategy and adds shelter-based universal PCR testing every 2 weeks for those without symptoms B.0) Re = 1.3: No intervention B.1) Re = 1.3: SxScreen/PCR/ACS B.2) Re = 1.3: Universal PCR/Hospital B.3) Re = 1.3: SxScreen/PCR/Hospital B.4) Re = 1.3: Universal PCR/ACS B.5) Re = 1.3: Universal PCR/TempHousing B.6) Re = 1.3: Hybrid/Hospital B.7) Re = 1.3: Hybrid/ACS C.0) Re = 2.6: No intervention C.1) Re = 2.6: SxScreen/PCR/ACS C.2) Re = 2.6: Universal PCR/Hospital C.3) Re = 2.6: SxScreen/PCR/Hospital C.4) Re = 2.6: Universal PCR/ACS	Extended SEIRD (CEACOV)—(S) susceptible, (E) exposed (latent, noninfectious), (Ip/Ia) pre- and asymptomatic, (Im) mildly infected, (Is) severely infected, (Ic) critically infected, (Ir) recuperation after critical infection, (R) recovered, and (D) deceased	ICER: cost/COVID-19 case prevented (whereby $1000/case prevented is approximately equivalent to $61 000/QALY gained)	Cost-effectiveness	4 months time horizon (April to August 2020)	Healthcare system perspective	Daily SxScreen and ACSs for sheltered homeless adults will substantially decrease COVID-19 cases and reduce costs compared with no intervention. In a surging epidemic, adding universal PCR testing every 2 weeks further decreases cases at modest incremental cost and should be considered.	18
Bagepally et al14 (India)	General population	0)Doing nothing 1)Surgical mask + hand hygiene 2)Hand hygiene 3)Surgical mask 4)N95 respirator (fit tested)	SQIRD—(S) susceptible , (Q) quarantined, (Im) mild infection, (Is) severe infection, (Ic) critical infection, (R) recovered, and (D) deceased	ICER: cost/QALY (WTP:INR 142 719 ($1921)/QALY gained)	Cost-effectiveness	1 year	Health system perspective	None of the interventions were cost-effective using the WHO WTP threshold. Among the interventions, hand hygiene appeared to be less expensive compared with other interventions but with similar effectiveness . The use of surgical mask with hand hygiene prevented the largest number of COVID-19 deaths.	15
	4. Treatment
Sheinson et al28 (United States)	Hospitalized patients with COVID-19	1)Treatment (no oxygen support; oxygen support without ventilation; oxygen support with ventilation) 2)Best supportive care	Acute care, short-term decision tree to model hospital treatment with 3 states: 1) No oxygen support, 2) oxygen support w/o ventilation, 3) oxygen support with ventilation, and 2 outcomes for every treatment: (A) alive or (D) deceased. Discharged patients advance to a long-term, post-discharge, life-table model with 2 states: 1) Alive and has not received ventilation during impatient stay and 2) alive and has received ventilation during impatient stay	ICER: cost/QALY (WTP: $50 000/QALY; $100 000/QALY; $150 000/QALY)	Cost-effectiveness	5 years	Healthcare payer perspective and societal perspective	Effective COVID-19 treatments for hospitalized patients may not only reduce disease burden but also represent good value for the health system and society. Post-COVID treatments were included.	10
Cleary et al29 (South Africa)	Hospitalized patients with COVID-19	1)General ward 2)General ward + ICU	Acute care, short-term decision tree to model health outcomes (recovered/deceased) of different hospital treatments	ICER: cost/DALY averted (WTP:38 465.46/DALY averted)	Cost-effectiveness	Not reported	Healthcare system perspective	ICU use for patients with COVID-19 was unlikely to be cost-effective on the margin, and therefore an expansion of ICU capacity during COVID-19 surges through government purchase of private services for use by public sector patients (at current prices and evidence of effectiveness) may not be the best use of limited health resources.	12
Gandjour31 (Germany)	Hospitalized patients with COVID-19	0)Maintaining ICU bed capacity (do nothing) 1)Expanding ICU bed capacity	Life years gained computed using life-table model	MCER of the last bed added to the existing ICU capacity (WTP: €101 493 per life-year gained); NMB; ROI	Cost-effectiveness	Lifetime	Societal perspective	Extending the existing ICU bed capacity seems acceptable on the basis of the MCER but also from a budgetary perspective. That is, extending capacity by more than 100% is forecast to result in a one-time increase in healthcare expenditure of 13%. If, however, the additional capacity remains entirely unused, the value of the investment becomes negative because of the presence of fixed costs. Nevertheless, it is reassuring that even a vacancy rate of 98% still allows for a positive return because of the low share of infrastructure costs. This is equivalent to a 2% probability of having full utilization.	6

ACS indicates alternative care site; CD, cesarean delivery; CEACOV, Clinical and Economic Analysis of COVID Interventions; CT, contact tracing; DALY, disability-adjusted life-year; DesigIsol, designated spaces; ESFT, Essential Supplies Forecasting Tool; FFP-2, filtering facepiece-2; GDP, gross domestic product; HCW, healthcare worker; HT, healthcare testing; IC, isolation center; ICER, incremental cost-effectiveness ratio; ICU, intensive care unit; INR, Indian rupee; LMIC, low- and middle-income country; LT, laboratory test; MCER, marginal cost-effectiveness ratio; NA, not applicable; NMB, net monetary benefit; PCR, polymerase chain reaction; POC, point of care; PPE, personal protective equipment; QALY, quality-adjusted life-year; QC, quarantine center; R_e, reproduction number; ResIsol, residence-based isolation; RLT, routine laboratory testing; ROI, return on investment; RT-PCR, real-time polymerase chain reaction; SALIRD, Susceptible-Asymptomatic-Presymptomatic-Infectious-Recovered-Deceased; SEIRD, Susceptible-Exposed-Infectious-Recovered-Deceased; SEIQR, Susceptible-Exposed-Infectious-Quarantined-Recovered; SI, Susceptible-Infected; SIR, Susceptible-Infected-Recovered; SIRD, Susceptible-Infectious-Recovered-Deceased; SQIRD, Susceptible-Quarantined-Infected-Recovered-Deceased; SxScreen, symptom screening; TempHousing, temporary housing; WHO, World Health Organization; WTP, willingness to pay; YLL, years of life lost; YLS, years of life saved.