Spatial Disparities and Correlated Variables of Community Care Facility Accessibility in Rural Areas of China
Round 1
Reviewer 1 Report
By applying the nearest distance method and GWR, it analyzes the spatial inequality of access to community care facilities (CCF) in Hubei Province of China. The article is generally well written and the results are informative. I have a few comments for minor revisions.
First, besides accessibility, is there any data on the quality of health care that can be utilized in rural areas?
Second, if possible, the authors may want to test if visits to those CCF facilities are from the nearest villages or human settlements. That being said, a supplemental analysis of actual usage data would be needed for this type of study.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper presents a straightforward spatial analysis procedure where geographical accessibility (to community care facilities) is calculated using nearest distance method, followed by an analysis of spatial clustering and then of geographical weighted regression.
Methodologically it appears sound, but being a rather small article, there is definitely room to increase both the theoretical background and the discussion, as well as clarify some points in the methodology. This is mostly an empirical paper, there is hardly any literature review on the theme, other than writing that ageing is growing in China, is high on the case study region and it’s a relevant theme. Further contextualization on ageing could be given, moving beyond Asian examples to other contexts (such as the European, where the average population over 65 is much higher than the Chinese), where papers have discussed geographical disparities on the access to health care and community care services, framed within territorial and social cohesion discourses.
Further contextualization on the Chinese context itself (which may be unfamiliar to an international readership) could also be helpful. For example, it is never fully explained, as CCF are “led by the government”, if there is any governmental criteria for their location/construction; how many are there or are supposed to be (for example per township); or indeed the difference between the village-level and the township-level (and Figure 2b is calculated not with Formula 1, but with a variation of it, right?). Sentences like “the rural areas in this paper are defined as all townships except for the subdistrict offices where urban communities are located” may be difficult to interpret, whilst others like “the main means of transportation in rural areas of China is the electric bicycle” just seem surprising without further contextualization or references.
In the analysis, it could be useful to clarify which distance bands were used and why (as they influence the outcome of Fig 3 for example), what is the AIC method, and also justify the selection of variables. Authors state they used seven variables (never explaining why these particular seven), none are described spatially, and immediately, in the following sentence, three are eliminated never to be mentioned again in the paper, because they failed (never shown) significance and covariance tests. In the Limitations section, authors further say that “some variables have not been able to be included”, but they never say which.
Finally, options in showing the results can generate confusion. The formula implies that “the lower the value of Ai, the better the accessibility of region I”; but in Figure 2 lighter blue colours show high accessibility regions (lower Ai values), which can be misleading. Also because Hot Spots in Figure 3 are of high values (of distance), meaning lower accessibility. This contrast is present, for example, in the analysis of population above 65, which has a “negative correlation with accessibility” but where “distance cost of accessibility to these areas decreases as the aging population increases”.
Discussion could also be augmented. Authors basically discuss where accessibility is high or low (to the non-Chinese reader this is of limited interest as the regions are never presented or contextualized), and then argue that facilities should be located where they are lacking. Conclusions related to the explanatory variables are similarly simple. Regarding “area” authors simply write: “the impact of regional size should be fully considered in these areas, and the layout of CCFs should be more reasonable”. Regarding “elevation” they write: “these areas should fully consider the driving force of elevation, and should further optimize the CCFs”. Concluding chapter repeats main findings of the previous section and nothing more is added. A more rounded discussion is recommended.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper contains an analysis of the accessibility of the Hubei Province in relation to Community Care Facilities.
The article advances the information and knowledge about the topic focusing the discussion into CCFs as a specific case of elderly care facilities and referencing already existing research.
Analysis and results are presented in an ordered way, but some improvements are required to make it more solid.
Below some general comments, while more specific comments in the attached document.
* one important consideration from my perspective is that all the analysis and discussions are focused on numbers and classes that are always relative to the distribution of the variables in the county. What I mean is that it is of course important to understand unbalances and disparities between more accessible and less accessible areas, but it would be more useful to have some criteria guiding what a decent or good accessibility would be and targeting future improvements to that aim. The only place in the paper where I could find something similar is at lines 247-250 where the sentence " which means that all relevant staff can arrive within half an hour and provide appropriate services to the elderly in need within most villages and townships." is not clear and needs more description. This topic should be addressed in general in the discussions and conclusions.
* in addition to the previous point, at line 66 it is mentioned the opportunity to produce "recommendations for the optimized allocation of resources for elderly care facilities" but, in this paper, only general recommendations are provided. In the discussion and conclusions some more effort should be put in formulating some recommendations.
* one suggestion in the paper is at lines 346-348 "Government departments should pay attention to the lack of services in these areas and take substantial measures such as increasing the number of facility points to fill the gap between regions." Here an analysis of where the allocation of new CCFs would optimize an increase of accessibility would be really interesting and useful.
* about the accessibility in relation to the driving forces, it seems to me that the considerations in 3.3.2 and in 4.3 are somehow weak: the driving force for the high or low accessibility is essentially the availability of CCFs in an area. Using the area of the township is simply a way to aggregate/present the analysis, rather than a driving force. Regarding the elevation, was it considered in the calculation of the distances or distances were calculated only in 2D? Maybe the term "driving forces" could be substituted with "contributing factors" or "correlated variables" or similar? The real driving factor seems to be the variable "population aged 65 and more" and the fact that it is positively correlated with accessibility of CCFs seems to be somehow a logical consequence that CCFs are for elderly people.
* There are many words in the text where a space is missing after an ending point, a parenthesis, a comma, etc. That should be a general formatting issue, anyway to be solved.
* in general, numbers are presented with 2 or even 4 decimals: it seems that that precision in numbers is not needed and also is not helping in reading the results. I'd suggest to find a balance between the automatic numbers coming from the elaboration or classifications (e.g. Natural Breaks) and readibility/understandability of the information.
* in the text is frequently used the past perfect referring to the results of the analysis: I personally would prefer a present tense, but anyway check to be consistent throughout the article
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
The authors answered to some of the points raised in the first review, but some others still remain.
Point 1
I've understood the authors' point, but I think that some more analysis need to be undertaken to provide some guidance or at least to roughly estimate the effort to improve current situation. For example, authors can set as an exemplary goal a minimum target of a distance cost to be reached (maybe 10000, that's roughly in the middle of the Medium accessibility class, or 5000, that is in the Higher class?) for all the townships and estimate the number of CCFs to be added in general or at the township level to reach that goal, given the reference 25km/h reference speed mentioned.
Point 4
The authors answered, and accordingly modified the article, only in relation to the point related to "Maybe the term "driving forces" could be substituted with "contributing factors" or "correlated variables" or similar?"
What about the other important points?
* about the accessibility in relation to the driving forces, it seems to me that the considerations in 3.3.2 and in 4.3 are somehow weak: the driving force for the high or low accessibility is essentially the availability of CCFs in an area. Using the area of the township is simply a way to aggregate/present the analysis, rather than a driving force.
* Regarding the elevation, was it considered in the calculation of the distances or distances were calculated only in 2D?
* The real driving factor seems to be the variable "population aged 65 and more" and the fact that it is positively correlated with accessibility of CCFs seems to be somehow a logical consequence that CCFs are for elderly people.
Unfortunately, during the first review an attachment (also mentioned in the text of the review) with many minor corrections/suggestions was not included, so I'm including it now again. Lines numbers refers to the pdf provided in the first review.
Sorry for this misunderstanding.
Comments for author File: Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 3
Reviewer 3 Report
The authors have answered to all comments, trying to improve the article where possible.
I only have 2 minor corrections for the text:
- line 367: add " (Table 1)" after "distance cost" to help readers find the number in the text
- line 368: "This allows the relevant service personnel can arrive from CCFs" > "This allows the relevant service personnel to arrive from the CCFs"
Author Response
Please see the attachment
Author Response File: 
Author Response.doc
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
