Integrating Spatial and Non-Spatial Dimensions to Evaluate Access to Rural Primary Healthcare Service: A Case Study of Songzi, China

Round 1

Reviewer 1 Report

 I very much enjoyed reading the paper and would love to see a revised version of it published in the International Journal of Geo-Information.

The paper has several strengths, such as strong and complex methodology, attention to detail, and the discussion of the relevance for policy. I had many thoughts and ideas as I read the paper, but many of them coalesced around the need to add conceptual richness and theoretical texture to two sections of the manuscript that clearly need some reframing: the methodology section and the discussion section.

1. To begin with the methodology section, I think the current writing misses the opportunity to articulate the project’s research design with the broader philosophy of space and spatial concepts developed by human geographers in the English-speaking world.

For sure, the authors keep repeating that their contribution is the integrated consideration of spatial and non-spatial variables, whereas most research treats them separately. But this integration has long been called for in human geography in seminal papers such as Simandan, D., 2011. The wise stance in human geography. Transactions of the Institute of British Geographers, 36(2), pp.188-192.  & Malpas, J., 2012. Putting space in place: Philosophical topography and relational geography. Environment and Planning D: Society and Space, 30(2), pp.226-242. These key theoretical contributions have paved the way for the methodological work of integration done in this manuscript, and therefore should be explicitly mentioned in a few lines in the methodology section.

2. With regard to the discussion section, I think the paper would benefit from adding a few sentences to connect the current interpretation more explicitly with political science discussions and sociological discussions of social exclusion and power dynamics in a rural context. I have in mind work by scholars such as Mark Shucksmith and/or Katherine Cramer Walsh, for example.

To sum up, I think the addition of a short paragraph for each of these two sections discussing the points above would make the manuscript much more compelling.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors aim at  measuring geographical accessibility and access to primary healthcare services in rural areas. This is a topic of interest both for the public health and the city management sectors.

The text is well written and well prepared. I have just a couple of notes that can help to improve the quality and interpretation if the text.

- the introduction is well written and very complete, I would also suggest to add information about organizational factors in the non geographical aspects.

- Fig1 - only 1 scale is present

- Generally, figures appear in low resolution, but i assume it is a particularity of the review platform.

- still, on this section I miss a table summarizing the socioeconomic and demographic conditions of the region, whit the factors that previously were reported to impact access (and after are indicates in table 1).

- As well as a brief description on the healthcare sector of the Region

- Figure3 and 4 can be arranged to be side by side? as done after

- the authors could consider to include as supp material the numerical values presented, for each administrative unit

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Author Response File: Author Response.pdf

Reviewer 3 Report

This review summarizes the following article: "Integrating spatial and non-spatial dimensions to evaluate access to rural primary healthcare service: A case study of Songzi, China." 

The article presents a novel approach that seemingly incorporates both the spatial and non-spatial aspects of access to clinics and hospitals in Songzi, China. The authors argue that previous research has typically examined spatial and non-spatial aspects of access as separate factors which undervalues both aspects for policy makers and others who would benefit from a combined analysis. 

My overall impression of the article is generally positive but has limitations that need to be addressed. It seems logical that spatial and non-spatial considerations should be treated as harmonious rather than disparate. While spatial and non-spatial aspects are treated in this article, my main concern is that it is not clear to this reader how the spatial and non-spatial aspects of access are combined together to create an overall access index. Below are a few concerns I had while reading through the article.

- It was not clear to this reader how the census population data was treated in the study. On the one hand, it seemed like population count was used but a more appropriate measure would be population density (people per square kilometer). Using population density would present a standardized metric of population concentration across the study area rather than simply a count.

- While the location of the clinics and hospitals were shown on the map in Figure 1, including these locations on the maps in the later figures (e.g., 3, 4, and 5) would be beneficial to the reader so that they could view any relationships between the different measures and the health care locations.

- For those individuals who have a color vision impairment (protanopia or deuteranopia) the green clinic dots do not visually contrast against the red-orange color fill for the upper right-hand map in Figure 1. A different dot color for the health care facilities is needed to provide an accommodating visual contrast to the dots so that the mapped pattern can be apparent. 

- The presentation of the geographic information shown in two left maps of Figure 1 could be improved. For example, the upper left-hand map should highlight Jingzhou rather than Songzi (the small red dot representing Songzi is barely visible on this map). The map in the lower left-hand corner should then label Songzi rather than Jingzhou.

- The assignment of the parameters for G(d_kj, d₀) in equation 1 do not match what is discussed in the "where" text G(t_kj, t₀)." These parameters (d_kj, d₀ and t_kj, t₀) tend to be switched in other locations in the paper.

- The G(t_kj, t₀) parameters in equation 3 do not match the parameters G(d_kj, d₀) that appear in equation 1, but should. Or, if the G(t_kj, t₀) parameters are different, then those parameters should be identified and explained.

- From which equation are the "variations of standardized G-2SFCA access scores," discussed in section 3 Results, derived? Equation 3 simply states that Ai is the access score, not a standardized access score which is a term that is used later in the article.

- What values are used to calculate the HHNR in equation 4 (e.g., the number of children 0-4 in age or something else)? This same question applies to the variables in Table 1 (does this data constitute simply counts or something else?)

- It is not clear to this reader how the HHNR values are treated in this process. How are the HHNR values incorporated into the computations? This needs a better explanation.

- The discussion immediately preceding Table 1, lists "occupations" but this factor is not included in Table 1. Moreover, Table 1 lists health insurance but that factor is not mentioned in the discussion, but should be.

- Table 1 lists Income, Education level, Health insurance, and Satisfaction level as positive indicators. However, "Demographic" is the only negative indicator. It is not clear to this reader why this non-spatial factor has been uniquely assigned a negative indicator. This negative assignment needs explained. The other indicators being assigned + also needs explained. Moreover, the indicator "Demographic" is not defined. What is demographic, exactly? 

- The parameter in equation (7) "lnp_ij" is not defined but should be.

- The "method of information entropy" is not explained as to what it is or why it is applied to this computation. This new idea is not discussed in the beginning sections of this article. Also, provide a source for where the importance of this method was derived?

- The paragraph after equation 9 attempts to explain some of the parameters used in the previous equations but falls short of providing the reader with complete information. For example, what is the "index of demand locations" and the "index of spatial and non-spatial indicators?" Are "factors" and "indicators" synonymous? These words tend to be interchangeably used in the article. Be consistent with their usage.

- To help visualize the relationship between the clinic and hospital locations and the G-2SFCA scores for Figures 3 and 4, it would be helpful if the clinic and hospital locations were plotted on the maps in Figures 3 and 4 (be sure to avoid using a green hue for the location of the healthcare facilities).

- The authors make statements that are not well-supported by evidence. For example, the authors state that "there are substantial spatial disparities in economic, medical service satisfaction and education levels, as shown in Figure 6(b), (c), (d)." Yet, those substantial spatial disparities are not explained. What is a substantial spatial disparity and how it this condition manifested on or documented by the map(s)?

- After reading Table 2, it appears as if "Demographic" is the HHNR factor, but what exactly is "Demographic" is not very intuitive. Suggest a different name for "Demographic" such as "Age" or something more descriptive.

- Are the standardized comprehensive access scores the same CAS_ij values from equation 10? If so, then the naming convention in equation 10 needs to be adjusted accordingly. The paragraph on Page 13 interchangeably uses standardized comprehensive access scores and comprehensive access scores.

- It is not clear to this reader that the research results present “remarkable disparities” as used in the Discussion section in access to primary health care. What exactly is remarkable and how did the authors arrive at this remarkable conclusion? 

- A larger overriding concern is fully understanding how the non-spatial factors define access to primary health care facilities. This lack of understanding ties back to, I believe, using the “method of information entropy” that was mentioned on page 7. This method, its utility, and reasons for its inclusion in this paper needs to be better explained. It is not clear to this reader how access to health care facilities is determined based on various non-spatial factors. Such a discussion is missing from or not well articulated in the article.

- The authors state that “When combining spatial and non-spatial factors using a comprehensive evaluation method” but this comprehensive evaluation method is not discussed well enough for the reader to fully understand how the spatial and non-spatial methods are combined to create the index map shown in Figure 8. The equations for the spatial and non-spatial access do not to specifically share linkages. So, the combination of spatial and non-spatial access is not mathematically well-explained. Again, quantitatively, how are the spatial and non-spatial metrics integrated to demonstrate access to health care facilities?

There are quite a few passages where the phrasing is oddly worded, the phrasing does not make sense, or the wrong word is used or is missing. Collectively, these issues prevented clarity in communicating the article's information.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have reviewed the manuscript in light of my reviewer comments. I think the revised manuscript fits the aims and scope of IJGI and would be a beneficial addition to the field.

My recommendation is that the paper be accepted in its present form.

n/a