All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Dear Dr. Gunter, I congratulate you on the acceptance of this article for publication.
[# PeerJ Staff Note - this decision was reviewed and approved by Jennifer Vonk, a PeerJ Section Editor covering this Section #]
The manuscript is exceptionally well-written and logically structured. The introduction effectively establishes the gap in the literature (the need to study longer-duration fostering) and the methodology is described transparently and in sufficient detail, allowing for replication.
Tables and figures appropriately support the key findings. In summary, the document adheres to all fundamental standards of scientific reporting, and overall clarity is high.
The experimental design is robust and appropriate for answering the research questions.
The use of a multimodal approach, combining behavioral measures with physiological measures, is a significant strength, providing a more comprehensive assessment of dog welfare.The weeklong fostering intervention is a valuable extension of previous studies on temporary fostering, providing crucial data on the mid-term effects of leaving the shelter. The sub-study on co-housing is a very useful addition, offering practical recommendations that shelters can implement immediately. The statistical methodology employed is appropriate for the collected data.
The main conclusions are strongly supported by the presented data and are highly valid.
The evidence that mean urinary cortisol levels are significantly lower during the fostering phase provides compelling evidence of the positive impact of the home environment on welfare.
No Negative Return Effect: It is a crucial finding that mean cortisol levels were not significantly elevated in the post-fostering period compared to the pre-fostering period, suggesting that a one-week foster stay does not result in significant, lasting negative effects upon return to the shelter.
Minor Discussion/Improvement Points (Suggested Revisions):
High-Intensity Activity (Q5): Results indicate a significant increase in high-intensity activity (Q5) during fostering, even though rest (Q1) also increases. Please expand the discussion on whether the Q5 increase is interpreted as positive behavior (e.g., increased opportunity for play or exploration in a safe, larger environment) and strengthen the rationale for this apparent paradox.
Moreover, in the secondary study, co-housing with a familiar dog led to increased rest post-fostering, but did not influence cortisol levels. Please expand the discussion on why these two indicators (behavioral vs. physiological) diverge in this context, linking more explicitly to the literature on social buffering and the specific effects within the shelter environment.
This article provides a significant and immediately applicable contribution to the shelter dog welfare literature. The effect size found for the seven-day fostering period is notable and should encourage shelters to adopt longer-term foster programs.
I would suggest that the discussion be slightly expanded to include more speculation on the observed differences between the two shelters (CASPCA vs. PACC) regarding cortisol. The authors correctly attribute this to environmental differences; however, a brief reflection on the nature of these likely differences (e.g., intake policies, noise/crowding management, operational handling) would make the findings more actionable for industry professionals.
Final Recommendation: I recommend accepting the paper. The points listed above are suggestions to enhance the discussion and data interpretation but do not require new analyses or major modifications.
No comment
No comment
No comment
The authors have done a wonderful job of revising their paper, with attention paid to all comments, as indicated in their response table.
Dear Dr. Gunter, I request that you make corrections to the manuscript and hope that your responses to the reviewers will be comprehensive. I hope that the new version of this article will be approved for publication as soon as possible.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
The manuscript is well written in clear and professional English. The introduction provides a thorough overview of the literature and places the research question in a solid context. The structure of the paper follows scientific conventions and is easy to follow. Figures and tables are of good quality and support the results appropriately, although some captions could be made more self-contained so that readers can understand the main point of each figure without having to refer back to the text. The authors have also provided the raw data, which is crucial for reproducibility and transparency. Overall, the reporting is of a high standard.
The study is carefully designed and addresses a relevant question about the welfare of shelter dogs during weeklong fostering and co-housing. The use of two shelters, with different intake systems, adds strength to the design. The combination of physiological (cortisol/creatinine) and behavioral (activity monitoring) measures is a real strength, providing complementary perspectives on welfare.
That said, a few clarifications would improve transparency. The decision to use urinary cortisol instead of salivary cortisol should be explained more explicitly, as saliva is often considered more reflective of acute stress responses. Since the study relies on an immunoassay, it would also help to acknowledge the potential for assay interferences. In addition, because dogs were screened for behavioral suitability (e.g., no aggression, not overly fearful), the sample may not fully represent all shelter dogs, and this should be noted as a potential limitation. Finally, while the study details the protocol for the dogs, it does not specify the criteria used for selecting or training the foster families, which could have a significant impact on the outcomes.
The statistical analyses are appropriate and the conclusions are well linked to the results. Still, a few issues are worth addressing. About 4% of urine samples were collected outside the specified morning window, and since cortisol has circadian variation, it would be preferable either to exclude these samples or to confirm that their inclusion does not affect the results. The study also does not account for underlying medical conditions (e.g., Cushing’s disease) that might influence cortisol levels, and this should be acknowledged as a limitation.
The finding of different cortisol levels between shelters is interesting, but the discussion is quite brief. Expanding on possible environmental factors such as noise levels, kennel density, or daily routines would enrich the interpretation. Finally, the follow-up period after fostering was only five days, so the study speaks to short-term effects but not to longer-term outcomes. A brief discussion on the importance of future long-term studies would be valuable.
This is an excellent and valuable study with clear applied implications for shelter management. It shows that weeklong fostering and familiar co-housing can improve the welfare of dogs, both physiologically and behaviorally. My comments above are mostly minor clarifications that would make the manuscript even stronger.
In particular, I encourage the authors to expand slightly on the shelter-level differences, clarify the rationale for urine sampling, acknowledge the potential biases due to health conditions and sample selection, and discuss the long-term implications. I also think the results section would benefit from a few “plain language” interpretations alongside the statistical results, to make the practical implications clearer to a wider audience. A final point to consider is the cost and feasibility of implementing these programs on a larger scale, which could be briefly addressed to increase the applied relevance of the paper.
The article is well-written, without any issues pertaining to English.
The Introduction provides a comprehensive review of stresses experienced by dogs in shelters and how cortisol and activity measurements can be used to assess canine stress levels and welfare. Sufficient context for the study is provided.
The original data are available and easily understood.
The article is well-organized, and the structure conforms to the suggested journal format.
All figures and tables are essential to the paper, and the figures are of sufficient resolution. However, additional text is needed in table and figure titles to make these features stand- alone (see General Comments for details).
The submission represents an appropriate ‘unit of publication’, and I commend the authors for including the smaller study on co-housing in this paper; others might have pulled this study out as a separate publication.
I have no ethical concerns about the study because the research received IACUC approval (lines 288-289).
The research questions are clearly defined in the Introduction, and this study extends previous findings of this group by examining effects of weeklong fostering (previous research involved 1-3 days in foster homes). Additionally, a separate study on co-housing is included.
The methods are described in sufficient detail to allow replication by other researchers. However, I would like further information on the two study shelters (see General Comments).
Statistical analyses are well-described, but the particular software package(s) used (and versions) are not included.
The conclusions are clear, supported by the data, and tied to the research questions stated in the Introduction.
Abstract:
The Abstract clearly states what new information is provided by this study (weeklong durations of fostering away from the shelter) and provides an excellent summary of results.
Lines 22-24: Human-animal interaction, such as time spent outside of the shelter, has been shown to positively influence dogs’ cortisol and activity levels. Consider adding this phrase to the end of that sentence to be absolutely clear about the direction of the changes (positively might imply increases) “, leading to decreases in both measures.”
Lines 41-42: Insert “at the open admission shelter” after “(with or without another dog) to remind readers this study did not involve a between-shelter comparison.
Introduction:
Lines 53-57: The second and third sentences do not add anything meaningful to the paper and could be deleted. I would go straight from the first sentence to Line 57 “Once dogs arrive at an animal shelter…”
Lines 96-97: Additional reasons for single housing include risk of injury from dogs fighting, and greater challenges to staff, such as making it more difficult to monitor individual dogs access to food and water and manage more than one dog in a kennel (as detailed in Newbury et al. 2010 Guidelines for Standards of Care in Animal Shelters).
Lines 130-138: The authors provide a nice summary of hypotheses tested and predictions made. However, I wondered whether they had a prediction regarding activity levels in dogs co-housed with the same dog prior to fostering as compared to dogs living alone or with a new dog (lines 136-138; a prediction for cortisol levels is included).
Materials & Methods:
Lines 142-147: Given that you found a main effect of shelter in the cortisol analysis (line 325) and a three-way interaction of activity level, phase, and shelter for the activity analysis (lines 357-362), please consider providing more detail about the physical housing of dogs (e.g., size of enclosures) and enrichment activities at each shelter (e.g. number of walks/day or time spent in a play yard off leash/day, play groups, etc.). Even though you did not test for environmental factors, this would provide readers with a better understanding of the conditions at each shelter, especially because enclosure size and enrichment activities could impact activity levels and possibly cortisol measurements. This information could be covered in the Discussion (lines 449-459).
Table 1: Tables should be stand alone, able to be understood without reading the text. I would add some details to this table title, something as simple as adding “for dogs wearing collar-mounted activity monitors at two shelters” to the current title. Figures should also be stand alone, so I suggest some additions to Figure legends/titles below.
Results:
This section is well-organized by topic and very easy to follow.
Table 2: Consistent with my comments for Table 1, I would insert “two animal shelters,” before PACC and CASPCA
Lines 297-303: Here, the authors describe minor breaks with research protocol, which reflect the challenges of conducting research in animal shelters, where dogs may or may not be available on any given day of a multi-day research protocol.
Figures 1 and 2 are excellent, with nice resolution. However, I suggest one minor change: make the color scheme for PACC and CASPCA consistent across these figures. Right now blue denotes PACC in Figure 1 and CASPCA in Figure 2, and orange denotes CASPCA in Figure 1 and PACC in Figure 2.
Table 6: Consider inserting “(housed with or without a dog)” after “housing.”
Line 387: Include a brief description of the significant effect of length of stay on cortisol levels (i.e., what was the pattern?).
Figures 4 and 5: I suggest including in the titles a brief description of the activity levels (e.g., “with Q1 representing the lowest level, rest, and Q5 the highest level, high intensity activity”).
Discussion:
The effect size for weeklong fostering is reported in the Discussion (lines 430-434) – is it also reported in the Results section?
As a study limitation, the authors address the issue that the two shelters differed fairly dramatically in makeup of dog populations (Table 2; at PACC, most dogs were strays or surrenders whereas almost all dogs at CASPCA were transfers; and dogs at PACC tended to be older, larger, and with longer lengths of stay than those at CASPCA). How these differences might have affected between-shelter differences in cortisol levels remains to be determined.
In a separate paragraph, the authors should describe strengths of their study.
I enjoyed reading this paper and hope the authors find my comments helpful.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.