Bacterial Contamination Levels and Brand Perception of Sachet Water: A Case Study in Some Nigerian Urban Neighborhoods
, Chinonso C. Eddy-Ugorji 3, Egbe-Etu E. Etu 4, Nkpa M. Ogarekpe 5, Chikodinaka V. Ekeanyanwu 6, Oladapo A. Diekola 7, Oluwarotimi S. Oladele 8 and Obiora B. Ezeudu 9Round 1
Reviewer 1 Report
1. Please confirm whether the data in Table 1 is correct, such as 6(66.67), 9(100), 41(57.75), 67(94.37).
2. Please confirm whether the data in Table 2 is correct, such as 9(34.78), 41(53.85).
3. In line 370, "7-fold dilution" is incorrect, please correct it.
4. In line 382, "[38] also investigated" is an unconventional expression.
Author Response
The authors would like to appreciate the reviewers for their time and effort in thoroughly assessing our work.
The authors would like to appreciate the reviewers for their time and effort in thoroughly assessing our work.
Reviewer 1
- Please confirm whether the data in Table 1 is correct, such as 6(66.67), 9(100), 41(57.75), 67(94.37).
The numbers have been updated to represent the results more accurately.
- Please confirm whether the data in Table 2 is correct, such as 9(34.78), 41(53.85).
The same as point 1 applies here.
- In line 370, "7-fold dilution" is incorrect, please correct it.
The term 7-fold dilution on line 370 has been changed to seven-fold serial dilution.
- In line 382, "[38] also investigated" is an unconventional expression.
The sentence has been corrected as suggested.
Reviewer 2 Report
Review of Bacterial contamination levels and Brand perception of Sachet Water: A case study in Nigerian Urban Neighborhoods
This is an interesting study that adds to the small number of studies regarding this important topic.
Introduction
The introduction provides a general and focused review of the topic and the motivation for the study.
Methods
The methods cover the important aspects of the study but could be shortened (e.g., lines 158-162 are not required). The bacteriological methods are well-known and this section could be shortened.
Results
The results need a major revision. The most interesting results are the proportion f samples with each type and level of bacterial and chemical contamination. These are not clearly presented as they are just presented as part of an analysis of the association with perceived quality and storage time (and very few of these relationships are statistically significant). There does not seems to be much revealed by the PA?FA analysis. I suggest it is removed from the paper.
Discussion
There is a good comparison with other studies and a discussion of possible causes of the observed contamination. However there is not discussion of the strengths, limitations and future research needs.
Specific comments:
43: there are many comprehensive and authoritative per-reviewed references regarding contamination of water-bodies than a MS thesis. These should be used instead.
47-51: Same comment. There are far better references that would provide more comprehensive information for the readers.
63: ‘unimproved’ refers to a water system; contaminated is a more appropriate word.
79: Ire there any references of residual Cl in SW?
86: What is the source of references 22, 24 and 26?
120: Borehole is a well-understood term among the readers.
121: There are far better references that would provide more comprehensive information for the readers.
142-145: These are interesting results and should be presented in the Results section.
Figure 1 is not legible and needs to be revised.
165: What is the rationale for testing for Staphylococcus Aureus?
211: Which media we used?
222: The first analysis is not clear. Clearly state how the reputation was determined from users and how the categories were derived from the consumers answers. This should be in the Methods. This paragraph needs major clarification. A comparison of proportions would be a more appropriate statistical test.
251: What statistical test was used?
Table 2 would be best used as Table 1. For the bacterial concentrations, was the arithmetic or geometric mean and SD used? Geometric means/SD would be most appropriate.
298: the null hypothesis for a correlation coefficient is that the coefficient=0. Thus is is not surprising that all were significant. Some of the reported coefficients are low to moderate yet they are discussed as if they are strong.
350: Comments that should be in the Discussion.
Author Response
The authors would like to appreciate the effort of the reviewers.
Response to Reviewer 2
Methods
The methods cover the important aspects of the study but could be shortened (e.g., lines 158-162 are not required). The bacteriological methods are well-known and this section could be shortened.
The physico-chemical and bacteriological analysis sections have been shortened based on the reviewer’s suggestions. Adequate references have also been added for the sake of readers needing more detailed information.
Results
The results need a major revision. The most interesting results are the proportion of samples with each type and level of bacterial and chemical contamination. These are not clearly presented as they are just presented as part of an analysis of the association with perceived quality and storage time (and very few of these relationships are statistically significant). There does not seems to be much revealed by the PA?FA analysis. I suggest it is removed from the paper.
The results have been reorganized to provide clarity to the readers. Additionally, the PA/FA analysis was retained in the manuscript because it is aimed at predicting the source of / examining the relationship between contamination sources in the SW samples collected and tested.
Discussion
There is a good comparison with other studies and a discussion of possible causes of the observed contamination. However, there is not discussion of the strengths, limitations and future research needs.
The manuscript has been expanded to include discussions on strengths, limitations, and future research needs. More of these are captured in the conclusion part of the manuscript.
Specific comments:
- 43: there are many comprehensive and authoritative per-reviewed references regarding contamination of waterbodies than a MS thesis. These should be used instead.
The thesis document referenced in the manuscript has been replaced with relevant per-reviewed articles.
- 47-51: Same comment. There are far better references that would provide more comprehensive information for the readers.
The references have been updated to include those that would provide more comprehensive information for the readers.
- 63: ‘unimproved’ refers to a water system; contaminated is a more appropriate word.
The word “unimproved” has been removed from the sentence.
- 79: Are there any references of residual Cl in SW?
References discussing residual chlorine levels tested in sachet water have been included in the manuscript.
- 86: What is the source of references 22, 24 and 26?
The affected references have been updated.
- 120: Borehole is a well-understood term among the readers.
In the study area (Nigeria), a borehole is drilled by machine and is relatively small in diameter and a well is usually sunk by hand and is relatively large in diameter. However, the authors have chosen to retain the term borehole in the manuscript.
- 121: There are far better references that would provide more comprehensive information for the readers.
This statement has been reviewed for better understanding.
- 142-145: These are interesting results and should be presented in the Results section.
The observations in this section have been included in the results.
- Figure 1 is not legible and needs to be revised.
Figure 1 has been updated to reflect the sample collection points more legibly.
- 165: What is the rationale for testing Staphylococcus Aureus?
Studies conducted by other authors have shown that SW may contain Staphylococcus Aureus. See below for some references:
Igbeneghu, O.A., Lamikanra, A. The bacteriological quality of different brands of bottled water available to consumers in Ile-Ife, south-western Nigeria. BMC Res Notes 7, 859 (2014). https://doi.org/10.1186/1756-0500-7-859
Adegoke OO, Bamigbowu EO, Oni ES, Ugbaja KN. Microbiological examination of sachet water sold in Aba, Abia State, Nigeria. Global Research Journal of Microbiology. 2012;2:62-66
Barnabas, F., Ukpa, S., Obeta, U., Mantu, E., Nduke, S., & Muhammed, Z. (2021). Prevalent of Staphylococcus aureus from Sachet Waters Sold in Different Areas of Jos Terminus Market, Nigeria. Journal of Biochemistry, Microbiology and Biotechnology, 9(2), 40–42. https://doi.org/10.54987/jobimb.v9i2.616
- 211: Which media was used?
MacConkey Agar was used in the study. This has also been stated in the manuscript.
- 222: The first analysis is not clear. Clearly state how the reputation was determined from users and how the categories were derived from the consumers answers. This should be in the Methods. This paragraph needs major clarification. A comparison of proportions would be a more appropriate statistical test.
The brand perception was determined using an informal questionnaire as a guide. Then based on that, we organized focus group discussions, obtained retailer responses on most purchased brands and individual spoke to some consumer. The categories used in the manuscript – good and poor brand – were the same options presented to the consumers. Chi-square or cross tabulation which is a type of comparison of proportion was applied, and the result was reported as you have read in the manuscript. This clarification has also been further explained in the method section.
- 251: What statistical test was used?
Cross tabulation (chi square)
- Table 2 would be best used as Table 1. For bacterial concentrations, was the arithmetic or geometric mean and SD used? Geometric means/SD would be most appropriate.
The result section has been reorganized to clearly present the results of this study. In table 2 (now table 1), standard error of the mean was used, and this has been indicated in the table. In our study, we were concerned about the mean value of specific brands within specific location, so we decided that the error values would better tell us the deviation from the population mean (from all the brands). We wanted to see how accurate the mean value was rather than the deviation from the total value.
- 298: the null hypothesis for a correlation coefficient is that the coefficient=0. Thus, it is not surprising that all were significant. Some of the reported coefficients are low to moderate yet they are discussed as if they are strong.
Using the p values obtained from the analysis of our results, we decided not to neglect weak correlations since this study has implications on public health. We were more concerned about statistical significance rather that the strength of relationship between variables. We noticed the relationship might be weak or moderate at 5 percent but otherwise when confidence interval increases to 1 percent.
- 350: Comments that should be in the Discussion.
The relevant comments have been included in the discussion section.
