Study ID	Domain		Rating
Adekunle 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	It is not really clear what they did. They seem to have built a SEIR model of China or Wuhan; Next, a sample of the increasingly infected population flights across the world, using prepandemic flight levels until 24 January 2020. The model does not allow for infected (to a proportion) not to travel (e.g. for being hospitalized) after a while. They then report to have introduced flight restrictions, but it is not explained how this was done in the model. They report on time until community transmission, but it is not really clear what they did here.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	There are references provided for each parameter used in the model
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	The infectious period of 4 days is rather short. The mortality of 1.8% is too high.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Predicted number of imported cases modelled compared with those observed
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Predicted number of imported cases is generally consistent with reported number of imported cases
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Technical documentation available, code not shared
Anderson 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Well described adapted SEIR Model (SE2IQR) model. Travelers are introduced across all stages of infectiousness
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	Does not allow for asymptomatic infections; Developing symptoms does not lead to an increased probability of behavior change (physical distancing)
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	All parameters are well justified
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	The incubation period of 1.2 days and the latency period of 0.2 days seems very short; and do not match the cited references. The model assumes that ‐ across the time of infectiousness ‐ only 1/3 of infected are put into quarantine. This figure seems relatively low (but unclear)
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Figure S2: Model was fitted against historical model in the 12 regions
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Figure S2: Model was fitted against historical model in the 12 regions
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	The model explores the implications of 1 additional case (infected traveller) for an increase in contact rates by factors of 1‐2. Other parameters are not varied in the model
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code to reproduce analysis is provided
Anzai 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Part 1: Reduced Number of exported cases: Observed cases versus expected number (according to the model) for day 58 to day 67 following the lockdown of Wuhan Part 2: Reduced probability of a major overseas epidemic Part 3. time Delay to a Major Epidemic Gained from the Reduction in Travel Volume The structure of the model is simple; but not clearly described and not very well justified
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Part 1 and also part 2 and 3 of the model (as these are based on the output of part 1) rely on the number of cases averted by the travel restrictions for the period day 58 to day 67: The model is a simple counterfactual model. The interruption is the lockdown of Wuhan, which took place on day 58. The model assumes, that there would have been an exponential growth in cases outside China, if the lockdown would not have been initiated, assuming that the exponential growth rate would have remained constant. This is compared against the observed number of new cases (imported and local) The difference is attributed to the travel restrictions; primarily attributed to the cordon sanitaire around Wuhan.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	As the assumed model is very simple, the input parameters used are described and justified
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	The main input parameter is the exponential growth rate (r). It assumes, that r would remain constant for the period day 58 to day 67 and therefore does not assume any other potential confounders in China (e.g. hygiene, social distancing etc.) and outside China (travel related measures, quarantine of infected would have taken place The model relies on diagnosed cases outside china (using RT‐PCR). There is a high likelihood of infected cases missed in this approach The assumptions on the R0 and the dispersion parameters k are justified and reasonable; The assumption of detected cases outside China (day 0 ‐ 67) however is not, as we cannot be sure whether these figures are reliable; Additional testing (e.g. of repatriated passengers under quarantine) could have inflated this figure, while insufficient testing or inadequate tests could have led to a significant underestimation of h(t)
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Yes, the model has been fitted to data points for day 0‐58 outside china
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Data points up to day 58 was used to fit the Poisson model; and yes, it does fit the model. However, these data points are unreliable
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	The model provides a range of sensitivity analysis, e.g. regarding r or on contact tracing. But likely additional sensitivity analysis (e.g. on underestimated cases) would have been helpful
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available; formulas, input parameters, and program used for calculating the model are described
Ashcroft 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Structure is transparently described and justified based on empirically estimated transmission parameters from the literature.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Within the scope of the model the assumptions and structural decisions are reasonable.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described based on the literature. Concept of the paper is to compare different possible parameter values of transmission parameters which are displayed in graphs.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Shiny app is provided which can be used to test face validity.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Uncertainty was not assessed.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Transparent reporting and provision of R‐Shiny app.
Banholzer 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Model assumes equal intervention effectiveness across countries and that interventions have a unique effectiveness independent of time and place they are implemented in, which is not reasonable, given that a travel ban of different countries (different levels of disease importation pressure) and at different points of time is likely to have a different effect (as the number of imported cases changes with the course of the pandemic around them). It implicitly assumes that all changes in case numbers are due to the intervention (tried to handle with sensitivity analysis)
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	No disease transmission parameters used due to type of model. The model uses data points of cases in the respective countries
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	No disease transmission parameters used due to type of model. The model uses data points of cases in the respective countries
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Cross‐validation to estimate influence of single countries.
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code published and extensive supplementary material available.
Bays 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Assumptions are transparently reported.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Very simplistic model. Assumptions restrict relevance of the model to a large degree. Key mechanisms like false‐positive screening results or transmission among travellers are subject to unrealistic assumptions.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described and corresponding sources are cited.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Information about flight time distributions only based on assumptions.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Python code is available.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Only some scenario analyses but no formal uncertainty.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Python code is shared and transparent reporting.
Binny 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions and equations are described in the text, and are justified with the current scientific literature.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Input parameters and data are based on the available literature and country‐specific sources.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation procedures reported but model fits well to observed data (external publication).
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Uncertainty is assessed using 5000 realisations of the stochastic model and sensitivity analyses of assumptions.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Model only described in another external preprint publication and no code available.
Boldog 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	clear & concise description
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Comprehensive description and justification of input parameters
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Input parameters seem to be reasonable; based on early data; change in R0 in China not adequately represented in the model
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation was conducted
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	the impact of variation in 3‐4 key model parameters on disease outbreak risk is assessed
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Comprehensive appendix, source code for the model is available on Github
Chen Y‐H 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Assumptions are transparently reported and the appropriate literature is cited and discussed.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	Within the scope of the SEIR model the assumptions and structural decisions are reasonable. The intervention is operationalised as changes in influx of infected individuals. The main issue is with the infectiousness: those who are infected without symptoms and those who are infected and show symptoms have the same transmission probability. e.g. individuals with symptoms do not change their behavior until they become hospitalized and those who show do not show symptoms do not get a lower transmission factor
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described and corresponding sources are cited. Some parameters are varied according to potential policy scenarios.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	The travel restriction measure is a hypothetical intervention: the study explores the implications for the local course of the pandemic if 10 versus 5 versus 1 infected individual per day arrive (in the sense of a 50% or 90% reduction). It does not address how this would take place or explores this measure in depth.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Process for internal validation not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Uncertainty for the individual parameters not reported. For the scenario with the different level of external infected influx (the travel restriction measure) no alternative scenarios or parameters are explored
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not reported, but inputs and assumptions reported sufficiently and should allow for replication
Chen T 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Structure is a relatively simple SEIR model; structure of model and relational assumption are adequately described; minor concerns as not the full model is provided in graphical form.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Simple model with reasonable assumptions on the topic at hand
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described based on the literature. Estimation of the importation of cases is reported in more detail in the appendix
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Operationalization of the measures is reasonable. Input parameters are sensible. Others (Rt) were derived from real world data Estimation of importations is reasonable
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Model was validated against the confirmed cases in the two countries
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	The simulated numbers of newly infected cases derived from the model based on estimated time‐varying Rt, actual intensity of entry restrictions and quarantine policy after adjusting a time lag between infection and reporting fit well with the official case figures. Likely, this should be regarded as a dependent validation.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Process for internal validation not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Authors conducted a Markov Chain Monte Carlo simulation by sampling from empirical distributions of these parameters to adjust for parameter uncertainty. Parameters were derived from the literature. This was the case for the proportion of presymptomatic travellers, travel probability, incubation period and serial interval.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not reported, but inputs and assumptions reported sufficiently and should allow for replication
Chinazzi 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Based on early data on transmission parameters; broad sensitivity analysis.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Limited approach to validate model projections against reported cases.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Limited approach to validate model projections against reported cases.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation described
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Established model, that is potentially validated, but this is not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Technical documentation in referenced methods paper; code not available
Clifford 2020a	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	Assume an exponential growth rate of r = 0.1 corresponding to an epidemic doubling time of 7.4 days based on data on early transmission in Wuhan. Also they assume a fixed travel time of 12 hours; The assumed sensitivity, cited from Quilty 2020 is questionably high. Quility data has major concerns regarding input parameters
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	The authors provide a shiny app which allows readers to assess the sensitivity of results to many parameter assumptions
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code available on GitHub
Clifford 2020b	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Assumptions are transparently reported and the appropriate literature is cited and discussed.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Within the scope of the model the assumptions and structural decisions are reasonable.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described and corresponding sources are cited. Some parameters are varied according to potential policy scenarios.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	R model code is available.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Uncertainty is assessed with bootstrapped confidence intervals and multiple scenarios are analysed and discussed.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code for analyses shared in Github repository
Costantino 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	The model does not consider other influencing factors e.g. the cordon around Wuhan as a reason for the difference and attributes it to the Australian travel ban.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Input parameters seem to be reasonable The mortality is relatively high; The assumptions on detected and isolated cases relatively low
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	The estimated incidence data in China was compared to the observed incidence data; no other external validation was done
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Limited approach to validate model projections against reported cases.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	A broad range of SE with alternative assumptions of different input parameters was done
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Supplementary material with further details of the model is supposed to available, but can't be opened; code is not available
Davis 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions are described and extensive information is available in the supplementary material and on the project website.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Some structural assumptions (e.g., homogenous travel probabilities) may not hold but generally minor concerns.
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	Not all input parameters are transparently described but the relevant literature is cited and calibration to observed data is described.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Pearson correlation in Figure‐1B
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Very limited external validation performed.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Model is continuously developed and provided as a stand‐alone software.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Uncertainty is assessed with confidence intervals and posterior probability distributions and sensitivity analyses in supplementary material.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Model is available as free, stand‐alone software but no accessible source code.
Deeb 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions and equations are described in the text, and are justified with the current scientific literature.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	With the information provided, the structural assumptions underlying the SEIR model appear to be valid.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Input parameters for the transmission‐related aspects underlying the model, as well as the running COVID‐19 cases and travel data are based on reported data.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Input parameters seem to be appropriate as far as it can be assessed.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Figure‐2  compares the cumulative number of infections predicted by the SEIR model with the observed data for the 130 retrospective days of the study period.
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Only dependent validation carried out; however, Figure‐2 shows that the model predicted the observed cases over 130 days well.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Additional analyses assess the impact of a variety of R values which represent varying levels of intervention; however, no analyses exploring how assumptions (such as starting values for R, E or I) may have influenced results.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Model formulas are rather explicit but neither the code nor the data are provided. Authors write that data are available upon request.
Dickens 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	The simulation of the passengers based on general epidemiological parameters is sensible in general; However, the assumptions regarding the estimation of number of infected travellers based on overall country parameter without taking further country and population characteristics into account is likely to distort the data
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	The model assumes the same time for illness onset to death for all countries included in the model. Given the heterogeneity of the countries (ranging from LIC to HIC), it is not reasonable. The estimation of infection burden within a country is based on the official case and death figures of the countries. Due to the different testing strategies in place, there is considerable heterogeneity in the relation of these figures in regard to the actual number of infections and COVID‐19 related deaths. The approach to estimate the number of those who are "mildly symptomatic and able to Travel at T" does not adequately capture this number and is likely highly distorted due to the general testing strategy and approach within the countries.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described based on the literature.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	The parameters in general have a reasonable foundation and are well justified in research. However, the generalisability of the parameters for CFR, IFR as well as time to death and time to hospitalisation for all countries is not reasonable. Regarding the PCR testing, the jump from 0% to 85% sensitivity at day two is likely inadequate; a less dichotomous development of the sensitivity is more likely.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Process for internal validation not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Uncertainty was not assessed systematically
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not reported, but inputs and assumptions reported sufficiently and should allow for replication
Gostic 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	Model assumes two modes of screening (self‐reporting / symptom screening) at arrival and departure, assuming the sensitivity of the scanners, the awareness about exposure, and the willingness to report symptoms. They furthermore assume that patients after a time period are being isolated/hospitalized in the departure country. The structural assumptions are reasonable, however the assumption, that infected are removed from the population and not able to travel seems quite arbitrary and ‐ given that 30 are asymptomatic ‐ the likelihood of 100% quarantine after 3‐7 days is not reasonable
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Authors provide references for most parameters. The parameters of awareness are assumed.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	The travel time of 24h hours seems to be rather long The assumption, the assumption, that infected are removed from the population and not able to ravel seems quite arbitrary and ‐ given that 30 are asymptomatic ‐ the likelihood of 100% quarantine after 3‐7 days is not reasonable
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No formal external validation conducted; however the results of the study are discussed in the light of findings of empirical studies coming to similar conclusions.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No formal external validation conducted; However the results of the study are discussed in the light of findings of empirical studies coming to similar conclusions. Our conclusion that screening would detect no more than half of infected travellers in a growing epidemic is consistent with recent studies that have compared country‐specific air travel volumes with detected case counts to estimate that roughly two thirds of imported cases remain undetected (Niehus et al., 2020; Bhatia et al., 2020). Furthermore, the finding that the majority of cases missed by screening are fundamentally undetectable is consistent with observed outcomes so far. Analyzing a line list of 290 cases imported into various countries (Dorigatti et al., 2020), we found that symptom onset occurred after the date of inbound travel for 72% (75/104) of cases for whom both dates were available, and a further 14% (15/104) had symptom onset on the date of travel. Even among passengers of repatriation flights, or quarantined on a cruise ship off the coast of Japan (who are all demonstrably at high risk), numerous cases have been undetectable in symptom screening, but have still tested positive for SARS‐CoV‐2 by PCR (Dorigatti et al., 2020; Hoehl et al., 2020; Japan Ministry of Health, Labor and Welfare, 2020; Nishiura et al., 2020; Hu et al., 2020). The onset of viral shedding prior to the onset of symptoms, or in cases that remain asymptomatic, is a classic factor that makes infectious disease outbreaks difficult to control (Fraser et al., 2004).
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Various sensitivity analysis were conducted to assess various parameters; the main parameter: number of asymptomatic cases is estimated to be 5%, 25%, and 50%
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code available
Grannell 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model setup and assumptions are elaborated and justified in the text.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	In their two region SEIR model, the authors account for the interaction between the population living on either side of the border strip between Ireland and Northern Ireland. Within each country, they make an assumption of homogeneous mixing. It is highly unlikely that infected individuals living on the border strip between Ireland and Northern Ireland have a constant contact frequency with susceptibles from all regions from their countries. In this vein, in the case of border interactions, it is reasonable to assume that the number of susceptibles in Northern Ireland will primarily affect the Irish population living on the border strip and only affect the rest of Ireland to the extent that infected individuals living on the border strip of Ireland interact with susceptibles from other regions. Therefore, while the assumption of homogeneous mixing is a standard assumption in SEIR models, it will likely lead to an overestimation of the effects of border interactions in the context of a two region SEIR model.
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	It is unclear how the authors determine some of the input parameters. For instance, they assume that cases will self‐isolate or self‐quarantine after 2 days but do not give any reference or justification for this assumption. They assume that the population of Ireland is 4, 900, 000 and state that 4, 977, 400 is a more accurate value from the Central Statistics Office but do not give any justification for not using the latter, the more accurate value.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Input parameters seem to be appropriate as far as it can be assessed.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Authors present a number of case studies in which some, but far from all, parameters are varied.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Major concerns
		10. Comments	Although the authors describe the model equations in detail, the methods used to fit these model equations are not described in sufficient detail to be able to reproduce the results.
James 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	Not all parameters are justified: for instance, no justification for adding a normally distributed random variable with mean 0 and standard deviation 1 to the date recorded by case recall to obtain symptom onset date or for adding a Gamma distributed random variable with mean 6.7 and standard deviation 5.4 to the reported date for cases with no onset date reported.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No sensitivity analyses or other assessments of uncertainty.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available, methods not described in sufficient detail to allow others to replicate the analysis well.
Kang 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Structural assumptions related to the choice of variables to construct the synthetic control are listed transparently, but are not well justified.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Factors considered important for the spread of SARS‐CoV‐2, and used to construct the synthetic control, included demographic factors, economic factors, the healthcare environment and the number of Chinese visitors. These factors are quite superficial and may not capture relevant differences across countries in constructing a valid control ‐ potentially important factors may not be covered, such as how governments otherwise reacted, e.g. how early testing or contact tracing strategies were implemented, if mask‐wearing was already part of established prevention.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Input parameters in this case are the observed data, and these are transparently reported.
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	No consideration is given to different rates of case ascertainment across countries, which, especially early in the pandemic, likely varied quite widely across the countries assessed.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	The synthetic control methodology uses observed data to construct a synthetic control; thus the overlay of the observed data and synthetic control data in the pre‐intervention period represents a form of external validation.
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	The various figures show that the counterfactual in the pre‐intervention periods fit the observed data well.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No concrete assessment of uncertainty; additionally, point estimates are provided with no measure of precision. Given the low numbers of cases in the study period, it is likely that there was substantial imprecision which is not clear.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Major concerns
		10. Comments	Code not available, methods for implementing the synthetic control design only poorly described; not likely that one could replicate the analysis well.
Liebig 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	The model to estimate traveller volumes assuming no travel restrictions is only described as a seasonal autoregressive integrated moving average model determined via step‐wise search over the model space. Moreover, it is unclear why the historical data on number of arrivals into Australia are Box‐Cox transformed to give the data a normal shape.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	The assumption that a country's incidence rate equals the percentage of observed COVID‐19 infections amongst travellers arriving into Australia from a given country is somewhat questionable, as individuals with COVID‐19 symptoms may both be less likely to travel and more likely to be infected. It is somewhat inconsistent that in the model for the expected number of importations, a Poisson variable is drawn instead of using directly the rate parameter of this distribution as this model also uses the probability that an individual is infected and the probability that an individual is infectious during the flight rather than sampling from a Bernoulli or Binomial distribution with these probabilities.
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	In the importation model, the authors assume that arrivals spent an average of 15 days in the source country prior to arrival without further reference or justification. Uncertainty estimates are lacking for most parameters.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Input parameters seem to be appropriate as far as it can be assessed
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	The authors did not explore alternative input parameter values and model assumptions.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Major concerns
		10. Comments	Code is not reported. Only part of the data is available. Replication would be difficult.
Linka 2020a	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	The authors assume that the number of infections are equal to the difference between the confirmed cases and the recovered cases and deaths from data from the ECDC but from my understanding this assumption is not very reasonable as the reporting of recovered cases is unreliable and incomplete.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Based on data on the latent and infectious periods A 1⁄4 1=a 1⁄4 2.56 days and C 1⁄4 1=c 1⁄4 17.82 days from 30 Chinese provinces (Peirlinck et al. 2020). Yields estimates for R0 of 8.7 in Austria and 6.0 in Germany.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Limited approach to validate model projections against data used to build the model.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Simulated data seems to be quite far from the observed data (see Figure‐2).
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No sensitivity analyses or other assessments of uncertainty
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available
Linka 2020b	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model structure is well stated and seems to be reasonable
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	The authors assume that the number of infections is equal to the "difference between today's and yesterday's reported cases" thereby ignoring underreporting and reporting delay.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Within the scope of the analysis no concerns.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Input parameters seem to be appropriate as far as it can be assessed.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Limited approach to validate model projections against data used to build the model.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Only dependent validation. Figure 6 and 7 show a poor model fit to the data from Newfoundland and Labrador as 95% uncertainty intervals only cover a small percentage of the observed data points.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No sensitivity analyses or other assessments of uncertainty.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available. As the analyses are based on an open‐access modelled passenger flow matrix by Huang et al. (2010, Plos One) and a risk flow model proposed by Gilbert et al. (2020, The Lancet) it could in theory be possible to replicate the analyses. However, the model formulas are not explicit enough to understand exactly how to link these two sources. In particular, the authors perform simulations to understand how risk flow will change with travel restrictions and in some of the scenarios they assume that the transmissibility rate changes from 1 to 0.5 but it is unclear how this transmissibility rate intervenes in the model as the authors do not define a transmissibility rate in the method section and neither Huang et al. (2010, Plos One) nor Gilbert et al. (2020, The Lancet) define a transmissibility rate.
Mandal 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	The authors assume that all infections go through an asymptomatic stage and a symptomatic stage; they assume that the exposed are infected from the start (but with less severity as when showing symptoms). The model does somewhat mix the incubation period and the asymptomatic cases into one mixed stage
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	From my understanding, it is unclear how the authors determine the parameters of their SIR model concerning recovery and death and the parameter values are not given. The authors assume that asymptomatic cases are only 0.1 or 0.5 times as infectious as symptomatic cases but do not give any references for these assumptions.
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	The authors assume that all symptomatic COVID‐19 cases are identified and that zero, 50 or 90 percent of asymptomatic cases are identified. These values seem very optimistic (other authors, for instance Clifford 2020 or Gostic 2015 assume more realistic and justified parameter values concerning the sensitivity of screening measures).
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Limited sensitivity analyses on a few parameters
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Model formulas available but from my point of view, it is not very clear how the authors calibrated their SIR model. No code available.
McLure 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No sensitivity analyses or other assessments of uncertainty.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Data and code are available.
Nakamura 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	The authors combine methods of existing publications but the model formulas are not sufficiently explicit to understand exactly how these existing publications are linked. In particular, in the results section, the authors describe the results of scenarios in which the transmissibility rate is reduced but it is not clear how this parameter (which is not defined in the methods section of either the present article or the articles referenced as sources for the methodology) intervenes in the model.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	The risk flow model proposed by Gilbert et al. (2020, The Lancet) accounts for Aia, defined as the probability of travelling form i to a, conditioned on travelling internationally from i, while the authors of the present study define the quantity Aod as the probability of travelling from origin (o) to destination (d) conditioned on travelling internationally. It is unclear whether the omission of the conditioning of travelling from i is on purpose or not and it is questionable whether the normalization described on page 41 still holds if this conditioning is omitted.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Within the scope of the analysis no concerns.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	The number of confirmed cases in different countries are used as input data in the model without accounting for or discussing different ascertainment rates in these countries.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No sensitivity analyses or other assessments of uncertainty.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available. As the analyses are based on an open‐access modelled passenger flow matrix by Huang et al. (2010, Plos One) and a risk flow model proposed by Gilbert et al. (2020, The Lancet), it could in theory be possible to replicate the analyses. However, the model formulas are not explicit enough to understand exactly how to link these two sources. In particular, the authors perform simulations to understand how risk flow will change with travel restrictions and in some of the scenarios they assume that the transmissibility rate changes from 1 to 0.5 but it is unclear how this transmissibility rate intervenes in the model as the authors do not define a transmissibility rate in the method section and neither Huang et al. (2010, Plos One) nor Gilbert et al. (2020, The Lancet) define a transmissibility rate.
Nowrasteh 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Structural assumptions related to the choice of variables to construct the synthetic control are listed transparently, but are not well justified.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	Factors considered important for the spread of SARS‐CoV‐2, and used to construct the synthetic control, included immigrant population, the total population, population density, the percent of the population that is elderly, the share of the population that is urban, the median age, real GDP per capita (PPP), the absolute latitude or distance from the equator, the number of immigrants from China, and the number of airports with direct flights to China. This is a fairly comprehensive list, however some important aspects, such as how early testing or contact tracing strategies were implemented, if mask‐wearing was already part of established prevention, may not have been picked up.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Input parameters in this case are the observed data, and these are transparently reported.
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	No consideration is given to different rates of case ascertainment across countries, which, especially early in the pandemic, likely varied quite widely across the countries assessed.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	The synthetic control methodology uses observed data to construct a synthetic control; thus the overlay of the observed data and synthetic control data in the pre‐intervention period represents a form of external validation.
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	The various figures show that the counterfactual in the pre‐intervention periods fit the observed data well.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	For each outcome four additional specifications were conducted to test the robustness of the results; these included changing the pre‐ and post‐intervention periods.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code available and methodology described to an extent that the study could be replicated.
Nuckchady 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions and equations are sufficiently stated in text and supplement; appendix C provides the model data.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Simplistic model with reasonable assumptions on the topic at hand, but no adequate representation of reality (e.g. infected independent of severity of symptoms are equally infectious; homogenous risk of infection (no network effect); different population and age groups are not represented in spread and mortality data). Asymptomatic individuals do not get to the hospital and therefore do not get tested. Only 10% of symptomatic go to the hospital and have a chance of getting tested. No asymptomatic but infectious phase? Those without a test do not quarantine? All infected travelers who enter the country are assumed to be at the start of their incubation period. The model assumes that only infected individuals get tested, which has implications for the testing capacity.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described based on the literature, real world data or reasoning based on real world data or literature. Some are assumptions, but this is made transparent.
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	The test sensitivity is likely too low; given that primarily symptomatic individuals get tested. Assumptions regarding contact‐behavior is likely overestimated ‐ assumes no behavior change without intervention. With intervention, behavior change leading to a reduction of 20% is likely relatively low. Authors state: "These variables were manually modified to make the model fit the actual data." Indicating at an adjustment of model data post‐hoc being in line with the real world figures.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Model was assessed against case figures from Mauritius.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Model was assessed against case figures from Mauritius. While the real data is for the most part within the 95% confidence interval, the case estimates are no ideal fit. Author reports, that parameters were adjusted to fit the model: "These variables were manually modified to make the model fit the actual data" indicating issues with the external validation of the model
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Process for internal validation not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Author writes "multiple sensitivity analyses were conducted to ensure the results were robust." Unclear what this refers to; uncertainty for travel related outcome is not provided; insufficient variation of alternative scenarios / input parameters for the travel related measures.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not reported, but inputs and assumptions reported sufficiently and should allow for replication.
Odendaal 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Model intends to describe the impact of the travel restrictions in the US on 31 January 2020 (the interruption). It uses an exponential form to describe the cases outside China (as cases in China are considered unreliable).
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	It is a very simple model that intends to describe the impact of the travel restrictions in the US on 31 January 2020 (the interruption). It uses a simple exponential form to describe the cases outside China (as cases in China are considered unreliable). It finds, that the number of cases (largely) follows an exponential growth rate (data fits better starting from mid‐February 2020). The model actually does not predict the impact of the US travel restriction. The model does not conclude that the 26 day delay of community spread (the start of the exponential growth in the US) was caused by the US travel restrictions. They actually just postulate this: Observation 1: Community spread started around 26th of February 2020 (debatable); Starting here, the official number of cases in the US follow an exponential growth. Observation 2: the number of cases outside China largely follow an exponential growth rate Observation 3: the time between the implementation of the US travel restrictions (31 January 2020) and the beginning of community spread in the US is about 26 days. Conclusion: "The imposition of early travel restrictions from China into the USA slowed down the virus by containing it mainly to individuals who had been to infected areas. The model indicates that delay was 26 days before it reached “community‐spread” in the USA. A main issue in this model is that it does not account for other confounders here. In particular: the cordon sanitaire around Wuhan on the 23 January 2020. It does not take travel patterns into account; it does not take the probability of seeding (regular travel patterns) into account. There is no causal inference; the 26 days are observed; not a results from the model
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	The parameters of the exponential growth model are derived from the official number of cases outside China. This is well justified and reported; the parameters for the exponential growth rates are given and justified The start of the community spread is based on official numbers in the US; the parameters for the exponential growth rate (starting around 29th) in the US are justified
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	These parameters seem to be reasonable
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	The model for the exponential growth of global cases is based on exported cases; model checked against these figures. No external validation specific to the outcome of interest "delay of community spread was conducted"
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	The model for the exponential growth of global cases is based on exported cases. However, their model only starts to fit these figures starting from around mid‐February 2020. The "global model" is not fitted against the US figures, rather, it reports that there is exponential growth starting around end of February 2020 as well.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	The study does not assess uncertainties and does not take other explanations into account
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available; the "model" and the input parameters are given and it should be possible to replicate the analysis
Pinotti 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions and equations are sufficiently stated in text and supplement
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	There are no input parameters
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	There are no input parameters
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Dependent validation and predictive validation of the case arrival model
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Figure 3 shows good dependent validation and predictive performance of the case arrival model
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	The authors did not explore alternative input parameter values and model assumptions
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Model formulas are rather explicit but the code is not provided. Data is available
Quilty 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	The model is adequately described
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	The model assumes 2 types of infections: asymptomatic and symptomatic of which 100% will experience a worsening of symptoms after 9.1 (+/‐ 14 days). It is unreasonable to assume that all patients will be hospitalized. It could however be assumed, that a share of those experiencing symptoms will decide not to travel; however it is unclear, why this should take place after 9.1 days, Furthermore, the study does not allow for self‐reporting of symptoms but relies on the thermal scanners for detection.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	The study provides references almost all input parameters
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	The study assumes that 83 % of infected will develop symptoms. In the light of the current evidence, this figure seems to be too high. The model assumes, that 100% of the patients will develop fever, which is not the case (there are infections which are symptomatic but do not have fever or only elevated temperature below the level to be detected by the thermal scanners. The sensitivity of the scanners was assumed to be 86%. In the study referenced, the value is given as 0.86% (95%CI 0.75–0.97) for a temperature of 37.8°. The assumption of 100% of symptomatic infected will refrain from travelling after mean 9.1 days (+/‐ 14.7 days) is likely too high.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation was conducted; The study does not discuss the findings in the light of empirical studies assessing screening interventions or the findings from repatriation studies
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation was conducted; The study does not discuss the findings in the light of empirical studies assessing screening interventions or the findings from repatriation studies
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	While the study itself does not provide a sensitivity analysis, it provides an App which should allow the reader to conduct sensitivity analyses; however the link provided for the app does not work
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code to reproduce analysis is provided
Russell TW 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model structure is well stated and methods are found in another paper, also documented well. Model simplifications are discussed.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Correction for under‐ascertainment refines model results. Neglecting effects on local dynamics is reasonable if the relative number of imported cases is comparably low (which was found to be true for most countries). Treating all infected cases the same way does not account for symptom status of individuals. Other modes of travel will dominate in neighbouring countries. Both of these points were mentioned in the discussion.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Source for case and death data is cited. Air travel data for slightly different scenarios is stated by the respective organization, but no direct source is given. Duration of infectiousness is stated as 10 days. Case Fatality ratio is cited, but not given.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Duration of infectiousness not ideal when calculating the prevalence of infection (but probably no impact?); Reasonable otherwise.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	True case data is calibrated with available data (dependent validation), but calibration not illustrated.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No actual validation besides calibration is available.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Difficult part of code (under‐ascertainment) has been published in previous paper and reports some external validation, implying some sense of validity; Rest of the model not too difficult to allow for many mistakes.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Main results were reported with credible intervals. Structural uncertainties were discussed, but not analysed quantitatively although they could arguably have an impact. Origin of credible intervals not discussed, probably in methods paper? Sensitivity to flight data has been assessed by using different data sets. Sensitivity with respect to prevalence and incidence supposedly analysed, but not illustrated in paper.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Data sufficiently cited to reconstruct their origin. Code is available.
Russell WA 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions and equations are sufficiently stated in text and supplement.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	The model only differentiates between infectious and non‐infectious; not taking the peak‐infectiousness (leading to super spreading events) in the time around symptom‐development into account. The sensitivity of the test only distinguishes between symptomatic/presymptomatic state, rather than being reflective of the development of varying sensitivity during the course of the pandemic.
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	Parameters are transparent; but limited justification and references for the selected parameters is provided.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Due to the fixed sensitivity; the presymptomatic phase is relatively high (assuming, that days 1‐3 of the infection PCR testing is very likely false negative). Sensitivity figures seem to refer to PCR testing (reference) although the figures in the cited publication do not match those reported here. If assumed for antigen‐rather than PCR testing, this seems to be acceptable. The assumption of 40% non‐infectiousness is likely too high (10.1371/journal.pmed.1003346). All figures can be changed in the online app.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Process for internal validation not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Model introduced uncertainty by varying the distributions’ mean and variance uniformly by ±20%, and sampled 1,000 parameter sets for the duration distributions. A number of alternative scenarios (high/low adherence; different levels of symptomatic status) are provided.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Authors provide an app which allows to vary parameters. Calculations on parts of the model are provided as a supplement.
Ryu 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	It is a relatively simple SEIR model; it is not discussed why a more complex model was not used
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	They assume all infected will become symptomatic; however this is not relevant for the model (as it is focused on an assumption of share of participants being quarantined); The model assumes, that those who will be quarantined, will be quarantined straight away, without having the ability to infect someone; The model does not allow for asymptomatic infectious; Model assumes random contacts within the whole population of Seoul; the interaction of the students is likely much more compartmentalized; Model assumes a perfect screening ‐ no symptomatic infectious arrive
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	Model is based on the assumption of 0.1%, 0.2%, or 1% of students being infected. This is justified by: "(i) were in the pre‐infectious period of COVID‐19 infection, based on previous literature reporting that 0.2% of individuals with contactees of SARS infection were asymptomatic [11]" It is not clear, how the authors came up with these parameters
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	The assumption of 70% ‐ 100% of all infected being quarantined and not breaking quarantine seems very high, given that around 30‐40% of all infections are asymptomatic The assumed incubation period seems a bit too long (6.5 days) The period from infectious to recovered is assumed to be 3.5 days. Within this model, the relevance of this parameters is the infectiousness period (only I are infectious). A period of 3.5 days infectiousness seems much too short.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	No external validation was conducted; The study does not discuss the findings in the light of empirical studies assessing screening interventions or the findings from repatriation studies
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation was conducted; The study does not discuss the findings in the light of empirical studies assessing screening interventions or the findings from repatriation studies
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	The study varies the compliance rate wit quarantine (80%,80%,90%,100%) and the share of arriving infectious students (0.1%, 0.2%, 1%) but no further sensitivity analysis is conducted
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available; the SEIR model and the input parameters are well described and it should be possible to replicate the analysis
Shi 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions are mostly established through existing literature. Method is based on another paper, but also summarized in this one. Choice of impact of travel restrictions with unclear formulation (75% of direct flights are cancelled). Missing explanation how other scenarios are evaluated, although this is probably just plugging in the estimated parameters.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	The concept of effective distance is non‐trivial and could therefore impact results if not implemented correctly. Assumption that effective distance only from Wuhan is considered might be bad with other outbreak locations contributing to international spread (but assumption might be consistent with data time span until end of February 2020). Increasing number of cases in China increase the risk of exporting the virus but cannot be covered by effective distance. Therefore, time‐constant hazard might be a poor assumption. Having many countries with increased risk of virus importation after imposing travel restrictions seems counter‐intuitive and it is not sufficiently explained why the model produces these results.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Table with survival data and dates of travel restrictions is provided. Data source for airline network is stated. Additionally, data is available in a repository.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	There are no concerns regarding the input data
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Model is calibrated using data, therefore dependent validation is available. But calibration is barely illustrated.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Although at least a dependent validation is available, there is no information about quality of estimates and model.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation is reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Model is simple in structure, not many concerns about internal validity. Similar results from the 25% and 50% travel reduction analysis provide some sense of validity.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No uncertainties for results are reported (providing quantiles is due to the nature of the results and not an analysis of uncertainty). There are many concerns whether structural assumptions are correct, analyses with alternative structures are necessary.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code and data are stored in accessible repositories.
Sruthi 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Much of the structure is hidden away in an AI‐type algorithm
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	As far as it can be addressed the assumed structure seems reasonable; Many of the assumptions is impossible to assess given the information in the study
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Algorithm parameters are specified Not many more parameters as it seems
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Since model inputs are fairly straightforward, there are barely any problems A minor concern would be the input of recovery time which scales the reproduction rate
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	5‐fold cross validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Cross‐Validity seems to suggest that weekly infection rates can be predicted well if case numbers are high enough No other forms of validation reported
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Functionality of cross‐validation suggests that model is at least function in some sense
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Uncertainties were reported, but they likely do not span varying structural assumptions which may have significant impact on the reproduction rate contributions
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code and source data available
Steyn 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	The model is described in adequate detail and had been used/described in a previous publication. The approach to simulate the passengers based on general epidemiological parameters is sensible in general
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Simplistic model with reasonable assumptions on the topic at hand.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are transparently described based on the literature.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	All parameters (e.g. for rate of asymptomatic cases, incubation period, sensitivity of the PCR test in relation to time) are well described and justified with adequate literature.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Process for internal validation not reported
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	The model does not explore alternative input parameter values and model assumptions
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not reported, but inputs and assumptions reported sufficiently and should allow for replication.
Taylor 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model is basic enough to fully develop the structure step by step; Most assumptions intuitive; Many small details make it difficult to assess
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Some minor concerns; Example: Probability of getting infected abroad equivalent to prevalence? Model might be too detailed for comprehensive analyses to be possible
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameter table is given including sources Assumed values are clearly stated
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Most parameters seem reasonable But high asymptomatic proportion Important parameters are neither reported with uncertainties nor backed up with enough sources (disease courses, non‐compliance, efficiency of screening measures)
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Stochastic effects were considered Key uncertainties in disease characteristics, test efficiencies and structure of compliance were not investigated Existence of many parameter inputs requires a discussion of uncertainties
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code not reported, but inputs and assumptions reported sufficient to allow for replication
Utsunomiya 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Assume that the progression of COVID‐19 for each country can be fit to a sigmoidal function, at either the lagging, exponential, decelerating or stationary stage; they also added a fifth more flexible stage to allow for bi‐directional changes due to public health interventions, for example
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	It is unclear whether a completely data‐driven approach imposing a function onto the data is a reasonable approach; ignoring completely the transmission characteristics of COVID‐19 as well as human behavior and mobility may be problematic.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	No real input parameters required for this data‐driven approach, apart from the data on the daily cases of COVID‐19
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	External validity assessed through predicting ECDC data 1‐day in advance
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Figure 3 shows that the model was able to accurately predict the ECDC data
	Validation (internal)	7. Has an internal validation process been described?	Reported
		7. Comments	Internal validation through the conduct of a simulation study; accuracy of estimates obtained by model was evaluated
		8. Has the model been shown to be internally valid?	No to minor concerns
		8. Comments	Figure‐2 shows the results of the internal validity assessment
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No real assessment of uncertainty through sensitivity analyses
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code and further information available at a linked Github repository
Kwok 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Model structure was reported clearly and justified if necessary. Not exactly clear which local patches were modelled, by the information given seems like Hong Kong, Guangdong and China excluding Hubei are the patches considered. Temperature dependence of R0 needs more proof because it is quite controversial and has a high impact on results.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Model seems to be simple to capture important effects. Assumption about R0 decreasing linearly from some initial value to zero due to temperature is highly questionable. Basically SEIR model in Hong Kong with varying R0 and possible influx from outside cases.
	Input data	3. Are the input parameters transparent and justified?	Moderate concerns
		3. Comments	Input parameters seem to be mentioned across the document, but no complete list. Parameters are stated without uncertainties. Rates for movement between patches unclear (is there outward flux from Hong Kong?).
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Population of Hong Kong is wrong by a factor 10, which might be a typo? Inputs for uncertainty of effectiveness of screening measures are necessary because of their high impact on results.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Sensitivity in change of initial R0 was assessed. All other uncertainty analyses are missing (effectiveness of screening, change of dynamical model parameters). Since linear decrease of R0 is a critical but not sufficiently motivated assumption, modifications of the model structure should have been assessed due to the expected high impact of results.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not reported and model parameters are in some cases unclear.
Wells 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Assumed that no infected individuals travelled from Wuhan after the travel lockdown enforced on 23 January 2020 ​
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	Assumes that all symptomatic cases are identified in screening; Assumes a very high effectiveness for self‐reporting of exposure risk; Based on early data on transmission parameters; assumes that the maximum incubation period is 21 days; that all reported infected cases acquired infection within mainland China
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Predicted arrival times compared against observed
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	Predicted first arrival times are generally consistent with reported international importation arrival dates
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Code available on GitHub and extensive supplementary material available.
Wilson 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	SEIR based model; with different assumptions and modelled interventions (e.g. wearing masks based on flight) for the parameter of influx of infected cases to protected, disease‐free region (NZ)
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Model describes time until outbreak assuming that there is only one route of influx. The model does not assume any relevant countermeasures following a detected infection on a flight, which is unlikely. They assume that the only entry point to NZ would be Australia. And if: it is unlikely that the base risk is equal to the base risk of Australia
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	The study provides references for most assumptions
		4. Are the input parameters reasonable?	Major concerns
		4. Comments	While most assumptions on the variables are reported; there are a number of uncertainties in the underlying data. In particular, the assumptions for the effectiveness of the measures seem relatively arbitrary; e‐g‐ on the effectiveness of entry and/or exit screening
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation was conducted; The study does not discuss the findings in the light of empirical studies assessing screening interventions or the findings from repatriation studies
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation was conducted; The study does not discuss the findings in the light of empirical studies assessing screening interventions or the findings from repatriation studies
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation conducted
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation conducted
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	No to very limited sensitivity analyses
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	Code not available; the SEIR model and the input parameters are well described and it should be possible to replicate the analysis
Wong MC 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Authors describe their study methodology by relying heavily on citing another paper (Thompson et al. 2019); however, even after checking this publication it is not possible to understand completely what the structure of the model and analysis is.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	Given the poor description, it is not feasible to completely assess the methodology of this paper; assuming the authors remained close to the models described in Thompson et al. 2019, the structural assumptions are likely appropriate.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Inputs for this estimation comprise the early observed cases in Hong Kong as well as an estimated serial interval; these are described sufficiently.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	No concerns related to these inputs.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Figure‐2 shows the observed cases (with border control measures in place) versus the predicted cases (without measures in place), a comparison of the two curves in the period before control measures were in place serves as a form of dependent validation.
		6. Has the model been shown to be externally valid?	No to minor concerns
		6. Comments	The curves in Figure‐2 are mostly consistent through the early stage of the pandemic, before measures were in place, suggesting the model is externally valid.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No internal validation
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Authors write that the results are robust to the length of serial interval; however it does not appear that this is generalisable to the assessment of the impact of the control measures being assessed.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Major concerns
		10. Comments	Documentation of methods and code for the analysis is poor; authors reference another paper, however they provide very little detail on what they did, thus replicating this study would likely not be possible.
Yang 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Model assumptions and equations are sufficiently stated in supplement. References for model analysis techniques are missing.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Unclear if time‐dependent parameters can be estimated reliably given the few observables. Travel restriction might be detrimental to country if there are strong internal infection dynamics; two‐sided travel restrictions are assumed.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Parameters are reported with sources and additionally described in the main text Parameter tale in the supplement
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Description of parameters suggest that they are reasonable Critical mobility data well enough described Questionable inputs in would have no impact in many instances since they are only initials and are adjusted in fitting process
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Dependent validation on model predictions exists by construction
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Bare minimum of validation available by dependent validation
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No internal validation reported
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	Analyses on simulated data would have been great to chow that time‐dependent parameter courses can indeed be estimated
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	No to minor concerns
		9. Comments	Stochastic and parameter uncertainties are well covered by stochastic approach with adjustable parameter values Uncertainties on trajectories clearly visualized No analyses on model structure or mobility data
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Python code and data are available Description of data analysis could have been more detailed
Zhang C 2020	Model structure	1. Are the structural assumptions transparent and justified?	No to minor concerns
		1. Comments	Linear Model is clearly stated and well explained. Implications of different outcome values are explained. There is some justification which time lag was assumed for different predictors. Fitting procedure has been described.
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Moderate concerns
		2. Comments	Model is motivated well enough to be reasonable. Parameters in the linear model are a bit confusing, but interpretations are given. Suspicious that the daily new infections from one day ago are a non‐significant predictor for the next day in too many cases. Results are by construction correlations, not clear to which extent causal relations can be extracted.
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Input data for flights, case data and country restrictions are stated.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	A minor issue would be that the analysis only accounts for confirmed cases. As discussed in the main text, data before the 22 January 2020 is missing for China. Incubation period of 14 days is quite long.
	Validation (external)	5. Has an external validation process been described?	Not reported
		5. Comments	No external validation
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	No external validation
	Validation (internal)	7. Has an internal validation process been described?	Reported
		7. Comments	Replication of results by use of other flight data and another case data source.
		8. Has the model been shown to be internally valid?	No to minor concerns
		8. Comments	Since data is probably quite similar, this is a check of internal validity. Model seems to describe the data well, high R‐Square (although some form of visualization would have been nice).
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Full table of all linear model results is given. P‐values for parameter were reported, although not according to best practices (only inequalities, different thresholds). Using other predictors for the model (different time lags) would have enhanced the model credibility.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	No to minor concerns
		10. Comments	Input sources have been cited and code for analysis is available in the supplement.
Zhang L 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Model is minimalistic, the few equations used are defined. Variables are defined confusingly, difficult to exactly understand what they mean. Since connectivity is the central variable, its properties should have been explained more (adopted from other publication).
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	Major concerns
		2. Comments	Given the information, it was unclear why several things were done (estimation of cases on day n?, sum over the past 13 days when calculating imported cases on day n). Risk of a traveller being infected seems to be proportional to the cumulative cases of that country? (Crucial since this affects all results).
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	Data sources have been cited, but it is seemingly a lot of data which should ideally be given in a supplement.
		4. Are the input parameters reasonable?	No to minor concerns
		4. Comments	Data sources seem to be appropriate as far as it can be assessed.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	Figure‐2 compares case risk index with imported cases.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	Unclear whether this is actual validation, since the imported cases seem to be estimated quite similarly as the case risk index. If the data of imported cases is actual data, this would be some form of validation, but this seems to be not the case.
	Validation (internal)	7. Has an internal validation process been described?	Not reported
		7. Comments	No form of internal validation was reported, but model is also quite simple.
		8. Has the model been shown to be internally valid?	Moderate concerns
		8. Comments	No large concerns because there is not much to validate.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Major concerns
		9. Comments	Uncertainty has not been considered.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Major concerns
		10. Comments	Code is not reported. Data must be aggregated from different sources. Replication would be difficult.
Zhong 2020	Model structure	1. Are the structural assumptions transparent and justified?	Moderate concerns
		1. Comments	Model structure is based on existing publication. Extensions are derived and explained in the supplement. Notation becomes complex but is summarized in a table. Arrival time and infected case reduction should have been defined more clearly. Sources are missing in supplement and are poorly cited in main document (Preprint version)
		2. Are the structural assumptions reasonable given the overall objective, perspective and scope of the model?	No to minor concerns
		2. Comments	Strong legitimation is given by a methods paper. Paper explains structure in detail, but reported analyses are a bit difficult to understand. Slope of linear relationship could change over time with more travel restrictions (was assumed constant?)
	Input data	3. Are the input parameters transparent and justified?	No to minor concerns
		3. Comments	No unreported parameters were noticed.
		4. Are the input parameters reasonable?	Moderate concerns
		4. Comments	There are some concerns with the nature of travel restriction parameters which have been assumed.
	Validation (external)	5. Has an external validation process been described?	Reported
		5. Comments	One small comparison of model prediction to independent value? Important model parameters were fitted to reproduce the linear relationship, some dependent validation.
		6. Has the model been shown to be externally valid?	Moderate concerns
		6. Comments	There was a comparison to real‐world data, although I could not reconstruct the argument. Nevertheless, it would be only a weak validation.
	Validation (internal)	7. Has an internal validation process been described?	Reported
		7. Comments	The model was able to reproduce the important features on simulated data.
		8. Has the model been shown to be internally valid?	No to minor concerns
		8. Comments	Approach seems to technically work as intended.
	Uncertainty	9. Was there an adequate assessment of the effects of uncertainty?	Moderate concerns
		9. Comments	Important assumption of unchanging slope was analysed in sensitivity analysis. Main results are stated with uncertainties. Many smaller results reported without uncertainties. Travel restriction parameters should have been explored in sensitivity analyses. Appropriateness of model structure was partially discussed when results needed further explanation. There was a discussion of further possible uncertainties at the end.
	Transparency	10. Was technical documentation, in sufficient detail to allow (potentially) for replication, made available openly or under agreements that protect intellectual property?	Moderate concerns
		10. Comments	No code was reported and replication should be difficult, but seemingly possible.