RR:C19 Evidence Scale rating by reviewer:
Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.
***************************************
Review:
General Comments:
This is an interesting piece of work that explores several hypotheses related to green space use pre-and-post- COVID-19 pandemic peaks in UK local authorities. The authors caveat their findings, presenting several important limitations to the data available for analysis. The hypotheses of the study are of great interest to the current pandemic, however, and so I believe it is important for these questions to be explored in the capacity that it is possible to do so with limited data.
Abstract:
The abstract is clear, especially in terms of research questions and findings. However, some hint at the methodological approaches that were used (i.e., statistical methods) would be useful. Because the abstract is quite long, it might also help to have it structured into sections.
Introduction:
The literature cited in the introduction is relevant and from several settings, provides the necessary background to frame the work. Figure 1 is particularly helpful. The final paragraph speaks several times of predictions the authors are making, and these would be better expressed as distinct hypotheses that are explored.
Methods:
The authors begin by stating that they used visual inspection to support their decision to limit the dataset to include only authorities where new cases were reported in six out of seven days. It would be preferable to have a more rigorous methodological approach behind this decision beyond visual inspection. It is also somewhat unclear as to why the authors chose to further exclude local authorities whose period of case increase was less than 10 days. The authors should identify the limitations of assigning regional mobility estimates to local authorities. These areas are of vastly different size, with regions undoubtedly having heterogeneous spatial patterns in mobility within. This highlights a limitation with using publicly available mobility data, and it would be interesting to see the authors comment on the difficulties of accessing high spatial and temporal resolution mobility data (e.g. from mobile phone companies). The term “model” is used vaguely throughout the methods, whereas the authors should refer to the fact that they were conducting multivariate regression models. It is also unnecessary to explain the different parameters of a standard regression model before then explaining that they chose a linear mixed model. I’m also not sure that referring to their models as “baseline transmission models” is appropriate, as this gives the reader the impression that they are modelling the actual spread of cases, rather than examining drivers of spread.
Results:
It would be helpful to see some box plots, for example, of case numbers across the 98 local authorities during peak transmission, alongside mobility and park use
Discussion:
The discussion is thorough and provides several important caveats. It would be nice to see more discussion around future research. What type of data would be necessary to further support their findings, and how might we learn from this in terms of gathering data quickly in the event of a future pandemic?