The effect of a hiding box on stress levels and body weight in Dutch shelter cats; a randomized controlled trial.

While staying in an animal shelter, cats may suffer from chronic stress which impairs their health and welfare. Providing opportunities to hide can significantly reduce behavioural stress in cats, but confirmation with physical parameters is needed. Therefore, the aim of this study was to determine the effect of a hiding box on behavioural stress levels (scored by means of the Cat-Stress-Score) and a physical parameter, namely body weight, during the first 12 days in quarantine for cats newly arrived cats at a Dutch animal shelter. Twenty-three cats between 1 and 10 years of age were randomly divided between the experimental (N = 12) and control group (N = 11) with and without a hiding box. Stress levels were assessed on days 1, 2, 3, 5, 7, 9 and 12 according to the non-invasive Cat-Stress-Score (CSS). Body weights were measured on days 0, 7 and 12. Finally, adoption rates and length of stay (LOS) were determined. Major findings of the study are: (1) the mean Cat-Stress-Score decreased with time for all cats, but cats with a hiding box showed a significant faster decrease in the CSS, reaching a lower CSS-steady state seven days earlier than the control group; (2) nearly all cats in both groups lost significant body weight during the first two weeks; (3) hiding boxes did not significantly influence weight loss; (4) no differences were found in the adoption rates and the LOS between both groups. Hiding enrichment reduces behavioural stress in shelter cats during quarantine situations and can therefore be a relatively simple aid to shelter adaptation. It offers no prevention however against feline weight loss, which indicates a serious health risk for shelter cats.

Introduction

Each year around 200 animal shelters in the Netherlands take in and rehome 27.000 stray and relinquished cats [1]. A shelter life is often associated with many stressors. Cats entering a shelter are introduced to a foreign environment with unfamiliar animals, people, sounds and smells. During these first days many of the cats struggle to adapt to these prolonged or repeated stressors and thus show stress responses [2-4]. Acute stress will encourage the animal to adapt to its surroundings. However, when prolonged aversive stimuli interfere with the adaptation ability, chronic stress will develop by causing a dysregulation of several major physiological systems like the hypothalamic–pituitary–adrenal (HPA) axis [5]. This may elicit clinical signs, such as hiding behaviour, defecating and urinating outside the litter box, decreased grooming or over-grooming behaviour and a loss of appetite [2,6-9]. Stress-induced long-term high cortisol levels can reduce the efficacy of the immune system against infectious diseases [1,6,7,9,10,11], and chronic stress can therefore harm a cat’s health [6,8,12,13].

When in a state of stress, the majority of cats will stop eating. Tanaka et al. found that stress elicited a decrease in food intake, negatively correlated with stress scores [14,15]. This stress response can have grave impact on cats: severe body weight losses in only a short period of time can induce feline hepatic lipidosis [6,16,17].

Several studies show that stressed cats display increased alert resting behaviour behind their litter box in an environment without hiding opportunities [12,18,19]. This is interpreted as alternative hiding behaviour for it offers some concealment [12,18]. Real concealment can be offered by providing a hiding box to shelter cats. A study conducted by Kry and Casey [19] demonstrated a decrease in stress, measured by the Cat-Stress-Score (CSS), when shelter cats were offered hiding boxes. Weight loss during quarantine is another phenomenon in shelter cats associated with stress [14]. However, little research has been done on the preventive effect of a hiding box on this stress induced weight loss.

A previous study conducted by Vinke et al. [12] has been the first step in gathering scientific data about the effect of a hiding box on stress levels of newly arrived cats in a Dutch animal shelter during the first 14 days in quarantine situations. The results show that cats with a hiding box recovered at least 4 days earlier from stress than cats without a hiding box [12]. The present study was designed with more frequent CSS scoring between Day 5 and 12, to gain greater insight into the feline recovery from stress and to relate these behavioural stress levels to a physical parameter, such as body weight.

The primary aim of this study was to determine the effect of a hiding box on behavioural stress levels and on body weight of newly arrived cats in a Dutch animal shelter during the first 12 days in quarantine. The additional aim was to compare the Length of Stay (LOS) of cats in both study groups. It was hypothesized that a hiding box would significantly reduce stress levels of newly arrived cats compared to the non-hiding box group, reflected in a lower CSS, less weight loss and a shorter LOS.

Materials and methods

The study was approved by the Animal Welfare Body Utrecht, after assessing the present study (Animal Welfare Body of Utrecht University, written declaration of 17 June 2016 by the Animal Welfare Officer). It was concluded that the study does not meet the definition of an animal experiment as defined in the Dutch Experiments on Animals Act and Directive 2010/63/EU because the animals encountered no discomfort.

Animal shelter

This study was carried out at a Dutch animal shelter (Stichting Dierentehuis Arnhem en omstreken), a medium size animal shelter with an open intake of around 700 cats per year [20]. Cat housing is situated in five separate quarantine units, an isolation ward and an adoption unit, providing a maximum shelter capacity of 90 cats in total. Dutch legislation mandates that animal shelters must have a legal stray holding period of 14 days. Stray cats must be quarantined in solitary housing at intake and quarantine and isolation wards must be physically separated from the main shelter setting. Furthermore new cats must be vaccinated against FHV/FCV/FPV within five working days after intake [21]. The housing of the animals, their care and management provided by the shelter in this study is representative of the majority of Dutch shelters.

Informed consent was obtained from the shelter staff for this study. In order to relate this study to daily shelter management, the original shelter protocols about the intake of new animals, daily animal care and hygiene were generally accepted, and substantial adjustments were avoided.

Animals

For this study, 23 European short hair cats, 11 males and 12 females, were selected out of the cats entering the shelter between 4th November and 30th December 2015. Cats entering the shelter were examined at intake by the shelter staff for gender, breed and age and received a treatment against ecto- and endoparasites (Stronghold® and Milbemax®). Given that all the cats came in as strays, age was estimated in years. Within 5 days after intake, the shelter veterinarian performed a physical health check. During this veterinary check (during the morning hours), the cats were microchipped and vaccinated with an attenuated vaccine (Versifel CVR®) against feline panleukopenia virus (FPV), feline herpes virus (FHV-1) and feline calici virus (FCV). Intact cats were spayed or neutered after day 14.

Inclusion criteria for this study were based on breed (European shorthair), health status and age (between 1 and 10 years of age). When new cats showed no clinical signs of illness, obvious heat, pregnancy or signs of nursing during the physical examination at intake, they were included in this study. As it is not generally accepted practice in Dutch animal shelters to screen apparently healthy cats through diagnostic testing (e.g. FIV/FeLV) at shelter intake, apart from the physical examination, no additional information was available on the feline health status of the cats in this study.

All cats were observed for at least 12 days after intake.

Two cats participating in this study left the shelter before their last observation day: from the Hiding box group one cat went to a foster home, from the Control group one cat was released within a trap-neuter-release (TNR) programme. Data of both cats were excluded from this study. Two other cats were not included in data for the length of stay, but were included in data for the Cat-Stress-Score, body weight and the adoption rate. After the 12 days observation period, one of these cats (nr. 8, control group) proved to be infected with FeLV and was euthanized a few days after the quarantine period of 14 days, while another cat (nr. 19, control group), because of its semi-feral behaviour, was also released through the TNR programme. Given that shelters often take in these non-clinical, but infected, cats and stray cats being poorly socialized (and even rehome them), this study has included these two cats in three of the four measured parameters.

In this study, cats between approximately 1 and 10 years of age were included. The growth rate of cats younger than 1 year old [22] might interfere with the parameters of body weight used in this study, while the CSS of older cats (> 10 years) might be influenced by age related cognitive dysfunction [23].

As previous studies [19,24] found no gender-related significant differences in stress behaviour, both male and female cats were included in the present study. The 23 cats were randomly assigned to one of the two groups with and without access to a hiding box, using an online randomization tool [25,26].

Housing conditions

The cat housing in the two adjacent quarantine wards consisted of cages (L x W x H: either 84 x 95 x 80 cm with elevated perching shelves (at height of 28 cm) of 84 x 25 cm or 69 x 91 x 87 cm with shelves (equal height) of 69 x 25 cm) in which the cats were individually housed. Available floorspace per cage (cage floor plus perching shelf) was respectively 1.01 m2 and 0.80 m2.

Every cage was furnished with a food and water bowl, bedding of towels, and a litter box. The cages of the experimental group contained a hiding box that was placed at the right side at the back of the cage. To avoid place preference for towels as bedding, the towels covered the entire floor of the cage, including the shelf and the interior of the hiding box.

Cardboard boxes were used as hiding boxes and measured 44 x 31 x 26 cm (L x W x H). These boxes had two entrances (WxH 0.16 x 0.20 m) [12]. Hiding boxes were never reused.

Access to the cats in the quarantine wards was restricted to the caretakers and the observer. Natural daylight was provided through windows in both quarantine wards, combined with fluorescent lighting between 08:00 AM. and 5:00 PM. Daily temperatures in the quarantine wards ranged from 16.0 to 19.8 °C. In the quarantine wards, no dog vocalizations could be heard.

Daily animal care

The shelter staff cleaned the cages daily between 09:00 and 12:15 AM by removing waste and applying a spot-cleaning method [27]. During this procedure, cats remained in their cages. Litter boxes were cleaned daily with hot water and dried with clean paper towels. Cages were disinfected between cats or when indicated (e.g. diarrhea) with a chlorine disinfectant containing sodium dichloroisocyanurate (Halacid®).

Food was provided once daily between 9:30 and 10:00 AM and comprised of around 50g per day Adult Royal Canin® dry cat food (SC 365D) with a metabolizable energy content (ME) of 4066 kcal/kg (16.995 MJ/kg). Fresh water was provided ad libitum. Cats kept their own litter box for the duration of this study.

Behavioural observations

Cats were given a habituation period of 24 hours after shelter intake (= Day 0), before behavioural assessment was performed [19]. Behavioural data were collected on days 1, 2, 3, 5, 7, 9 and 12 between 12:30 and 5:15 PM, during which interactions with caretakers were avoided.

Each cat was observed for 20 minutes per day by using video-recording. Outside the cage, a video camera (H.264 DVR) was mounted on a tripod at cage height. For new observations the combination camera-tripod had to be readjusted to the new cat cage. Video recordings were viewed in real-time in an adjacent room and stored for subsequent analysis (Fig 1). Only one camera was used for recording.

Fig 1

Course of the Cat-Stress-Score in time of individual cats from the control group and the experimental group.

Line segments connect measurements within the same cat to show the change of CSS in course of time. Dotted lines: individual cats without hiding boxes (control group). Solid lines: individual cats with hiding box (experimental group).

Cat-Stress-Score (CSS)

Kessler and Turner [24,28] developed a 7-level Cat-Stress-Score (CSS), which has been used in several studies to estimate stress levels in confined cats [3,12,19,24]. This scoring system assesses the level of feline stress based on the posture of body elements (e.g. belly, legs, tail, head, eyes, pupils, ears, whiskers) and behaviour (vocalization and activity) as described in the ethogram of the UK Cat Behaviour Working Group [28]. The CSS ranges from 1 (fully relaxed) to 7 (terrorized).

One observer (LS) assessed the CSS score per cat on Day 1, 2, 3, 5, 7, 9 and 12. Intra-observer variation was minimized by observational training using (video) images of pre-described feline behaviours from previous experiments with shelter cats.

After the video camera had been positioned, the scan sampling started after 2 minutes during which the cat habituated to the novel situation. The cat was subsequently scored according to the Scan Sampling method, in which four scores (= four samplings) were made during the observation time (the 1st observation at 5 min, the 2nd at 10 min, the 3rd at 15 min and the 4th at 20 min) [29]. Imperceptible posture and behavioural elements were noted as missing values.

Each of the elements of the Cat-Stress-Score was scored separately. The scores of the four samplings were averaged to assign an overall CSS for each separate cat per day.

Body weight

During the study every cat was weighed on Day 0, 7 and 12 by using an electronic scale (accuracy ± 10 g). The standardized shelter feeding regime consisted of approximately 50g per cat per day of Adult Royal Canin (RC)® dry cat food, which equals 203.3 kcal or 849.8 kJ per cat per day. All cats were offered the same diet, with some individual temporal exceptions, which are mentioned in the text (oral medication would be given mixed with tinned food). The shelters’ feeding regime did not include the monitoring of the daily intake of food per cat and neither did our study design.

To secure adequate nutrition for the cats in this study, the daily caloric feline requirements (FEDIAF guidelines (80 kcal (335 kJ) ME per kg0.67)) were determined per individual cat [30].

Adoption rates and length of stay (LOS)

In order to determine the effect of a hiding box in quarantine situations on the subsequent adoption success, the adoption dates of the cats in this study were noted. Adoption rates (= # cats adopted / all cats in this study) and the length of stay (LOS: number of days between the shelter intake of the cats in this study and its day of adoption) was determined per cat. The LOS included the mandatory quarantine period of two weeks and only included adopted cats, excluding cats that were euthanized or returned to their outdoor environment after finishing this study.

Statistical analyses

A randomised controlled trial (RCT) design was used [31,32]. Data were stored in Microsoft Excel 2010 files (Microsoft Corp, Redmond, Wash.). Two statistical software programs were used for analysis of the data:

SPSS (IBM Corp, Armonk, NY version 25) for the two-sample T-test and chi-square test.

R (version 3.3.0) for the linear mixed regression models [33].

For the statistical analysis of effect of time and hiding box on the CSS (model ‘CSS-Time-Box’) a linear mixed regression model [34] was assumed, with the CSS as the outcome, while Time after arrival, the availability of a hiding box and the interaction between both were used as explanatory factors. CatID was used as the random effect to take the correlation between observations within cat into account. An AR1 correlation between the time points was added as well as a variance model to allow different variances for the separate time points. A maximum likelihood-based method was used to calculate the Akaike’s Information Criterion (AIC) to select the best model using a backward selection approach (smaller is better).

For the statistical analysis of effect of time and hiding box on the body weight, a linear mixed regression model [34] was used to analyse the weight as the outcome and Time after arrival, the availability of a hiding box and the interaction between both as explanatory factors. Although keeping the box in the linear mixed model resulted in a worse fit of the model, the availability of the box nevertheless was added in coherence with our primary aim. Also, in this model CatID was used for the random effect.

The validity of both models was confirmed by a visual inspection of the residuals for normality and constant variance.

Per experimental group the number of adopted cats was analysed using the chi-square test, while the length of stay (LOS) was analysed using the two-sample T-test. The assumptions for these variables for equal variance (Levene’s test) and for normal distribution (Shapiro-Wilk’s test) were met.

We reported the estimated effects of the availability of a hiding box according the reporting guidelines for randomized controlled trials (www.reflect-statement.org).

Results

Characteristics of the study population

The experimental group consisted of 12 cats (6 males and 6 females) of which the estimated age ranged between 1 and 7 years (mean: 3.3 years, SD: 2.2). The control group consisted of 11 cats (5 males and 6 females) with estimated ages between 1 to 10 years (mean: 4.9 years, SD: 3.1, with n = 10: due to her semi-feral behaviour no age could be estimated of cat nr 19). At intake, the control cats were on average 300 grams heavier than those in the experimental group.

The cats in this study are presented in the appendix with their ID, experimental group, gender, age, bodyweight at intake (kg) and the quarantine wards they went to after intake.

Daily Cat-Stress-Score (CSS): Behavioural assessment

The time-dependent reduction of the individual CSS per cat in both groups is visualized in Fig 1.

Cats from the hiding box group reached a steady state sooner (on Day 2) than cats from the control group (on Day 9). The model results for the mean CSS are presented in Table 1. The estimated means of the CSS of the hiding box group (mean CSS = 2.7) and the control (mean CSS = 3.1) at Day 1 are similar as their difference is not significant (-0.4, 95% CI:-0.97 to +0.12). At all other days the mean CSS of the hiding box group is significantly lower than the mean CSS in the control group, largest at Day 2 (-0.99, 95%CI: -1.38 to -0.61) and decreasing in difference between the groups on Day 12 (-0.33, 95%CI: -0.57 to -0.08).

Table 1

Results of the model for the Cat-Stress-Score with 95% confidence interval, influenced by day and availability of a hiding box and interaction between both.

Time	Cat-Stress-Score (CSS)
	Control		Hiding box
days after Intake	Estimated mean	95% CI1	Estimated mean	95% CI1
Day 1	3.132	2.74–+3.53	-0.434	-0.97–+0.12
Day 2*	-0.113	-0.43–+0.20	-0.994	-1.38–-0.61
Day 3*	-0.543	-0.91–-0.17	-0.514	-0.79–-0.23
Day 5*	-0.763	-1.15–-0.37	-0.254	-0.47–-0.03
Day 7*	-0.823	-1.21–-0.42	-0.234	-0.40–-0.05
Day 9*	-0.923	-1.32–-0.53	-0.124	-0.24–-0.01
Day 12*	-0.913	-1.34–-0.49	-0.334	-0.57–-0.08

1 CI = Confidence Interval

2 Estimated mean CSS in cats in Control group on Day 1.

3 Estimated difference between mean CSS at specified day in Control group compared to mean CSS on Day 1 of same cats.

4 Estimated difference between mean CSS at specified day in cats of group with Hiding box compared to mean CSS of cats in group Control group at same day.

* Significant difference between Mean CSS of the hiding box group and the mean CSS in the control group of the same day.

For the comparison of both experimental groups, the absolute body weight was used. The initial weight difference of 300 grams between both groups reduced to 210 grams at Day 7 and Day 12. Cats in the control group lost overall 7.7% of their initial body weight, while cats with a hiding box lost 6.3% of their initial body weight during those 12 days (Table 2). The initial weight and weight reduction between the groups, however, proved not to be significant.

Table 2

Results of the model for body weight with a 95% confidence interval, influenced by day and availability of a hiding box and interaction between both.

Time	Body weight
	Control			Hiding Box
Days after Intake	% change from Day 0	Estimated mean (kg)	95% CI1	% change from Day 0	Estimated mean (kg)	95% CI1
Day 0	0	4.393	3.77–5.01	0	-0.305	-1.16–0.56
Day 7	-6.1 (SD2 4.1)	-0.254	-0.35–-0.15	-4.5 (SD2 3.9)	-0.215	-1.07–0.65
Day 12	-7.7 (SD2 5.1)	-0.324	-0.42–-0.22	-6.3 (SD2 4.4)	-0.215	-1.07–0.65

1 CI = Confidence Interval

2 SD = Standard deviation of the % change in body weight compared with Day 0.

3 Estimated mean Body weight of cats in the Control group on Day 0.

4 Estimated difference between the mean Body weight of cats in the Control group on the specified Day compared to the mean Body weight of the same cats on Day 0.

5 Estimated difference between the mean Body weight of cats in the Hiding box group and the mean body weight of cats in the Control group on the specified Day.

The individual proportional decrease in body weight is visualized in Fig 2. All cats except one lost weight during both weeks. When weight loss at Day 12 was calculated as a percentage of initial body weight at intake, it was found that 7 of the 23 (35%) cats lost ≤ 5% of their body weight, whereas 15 of the 23 (65%) cats lost 5% or more of their weight. The maximum body weight loss was found in cat nr. 8 (control group), which lost 19% of its initial weight in 12 days and was diagnosed with an infection of FeLV a few days after completing this study.

Fig 2

The proportional change (%) in body weight in individual cats from the control group and the experimental group.

Line segments connect measurements within the same cat to show the change of body weight in course of time. Dotted lines: individual cats without hiding boxes (control group). Solid lines: individual cats with hiding box (experimental group).

Of the 23 shelter cats in this study, 21 were rehomed after the observation period was completed. In the control group, 9 out of 11 cats were adopted (82%), in the experimental group 12 out of 12 (100%). No significant difference was found in the adoption rate between the two groups (p = 0.55).

As we defined LOS as the number of days between the shelter intake of a cat and its day of adoption, 2 cats were not included in this data set, for they were not adopted.

The mean LOS for the control group (n = 9) was 24.1 days (SD 5.4, range 15–30 days) and for the hiding box group (n = 12) was 22.9 days (SD = 4.4, range 16–30 days). No difference in the mean LOS was found between the control and the hiding box group (p-value = 0.58).

Discussion

The aim of the present study was to determine the effect of a hiding box on behavioural stress levels and body weight in shelter cats during the first 12 days of quarantine. While in a previous study cats were monitored on days 1 through 5 and the 14th day [12], this new study added more insight about the differences in CSS between Day 5 and Day 12.

The most important findings of this study are:

The mean Cat-Stress-Score decreased with time for all cats, but cats with a hiding box showed a significant faster decrease in the CSS and recovered from stress seven days earlier than the control group.

Nearly all cats lost significant body weight during the first two weeks. On average, cats with hiding boxes lost 40 grams less of their initial body weight compared with cats without a box, although this difference was not significant.

The mean adoption rates and the LOS of cats with and without hiding boxes were equal.

The cages used for the cats in this experiment, varied in dimensions: the smaller cage offered 80% of the floor space of the larger one. Although Kessler and Turner [35] showed an effect of cage dimensions (0.49 m2 versus 1.02 m2) on stress levels in individual cats, they considered other qualitative aspects of the housing environment to be of importance as well. More recent research [36] studied the effect of cage dimensions (1.1 m2 versus 0.56 m2) combined with different feline care regimes. Doubling the cage sizes did not influence the pattern of feline responses, suggesting that physical space solely may be of less importance to cats, in contrast to managed and enriched environments that did significantly affect the cats. Therefore, in this present study, the effects of hiding enrichment were studied in spite of a minor size deviation between the individual cat cages.

Cat-Stress-Score (CSS): Behavioural assessment validation

In this study, cats with a hiding box showed a significantly faster decrease of behavioural stress compared to the control group, which was most prominent during the first observation days. These results were in line with a study of Gourkow and Fraser, in which the mean CSS of cats, housed in single barren cages without positive human-cat interaction, was higher compared to the other groups and only reached a similar CSS on Day 9 [15].

The findings of the present study complete the results obtained by Vinke et al., where the hiding box group recovered at least four days earlier. By increasing the number of observational days during the first 12 days, the current research provides more details about reaching the CSS-steady state, indicating that hiding boxes accelerate the recovery of behavioural stress by seven days. The hiding box clearly helps the shelter cat to adapt more quickly to a stressful new environment thus preventing the development of chronic stress [19].

This study shows a significant decrease in feline body weight during the first 12 days in an animal shelter. Approximately a third of the cats lost less than 5% body weight during the first 12 days, while two-third lost over 5%. These results agree with previous findings of Tanaka et al., in which 57% of their cat population showed 5% or more weight loss during their shelter stay [14]. Stressed cats are more likely to lose body weight. Cats in shelters [14], in laboratories [18], in boarding facilities [unpublished data] and even privately-owned cats [6] display this general stress response in challenging situations. However, when otherwise healthy cats lose weight unintentionally, it is a dramatic indicator of a health risk.

Weight loss can be caused by insufficient nutritional management (the shelter offers inadequate quantity and/or quality of food) and also by a decrease in feline appetite due to a physical stress response. Although food intake was not registered in the present study, it was observed that some cats were completely anorectic, especially during the first days. For the shelter this was the reason to standardize the feeding schedule of 50 g dry cat food per cat per day. According to the FEDIAF guidelines [30] for daily caloric feline requirements, during this study cats over 4.01 kg might have been offered an inadequate amount of food. With an individual requirement of 80 kcal (335 kJ) ME per kg0.67, 50 g dry cat food per day will meet maintenance energy requirements of cats up to a body weight of 4.01 kg. Cats weighing over 4.01 kg, need more Adult RC food daily. Of the 23 cats, 13 (57%) cats weighed more than 4.01 kg. The heaviest cat weighed 6.41 kg at intake and hence required at least 68.4 grams of cat food per day. During the daily observation, however, cats rarely finished their food rations during these first two weeks. An inadequate quantity of food was therefore not considered to be the cause of the observed body weight loss.

The effect of stress on the body weight of shelter cats was first shown by Tanaka et al. [14], who found a negative correlation between food intake and stress scores of cats. The conclusion was that cats admitted to an animal shelter were likely to lose weight while in the shelter. These results are consistent with our findings, indicating that a decrease of feline appetite caused by a physical stress response, is most likely responsible for the weight loss.

Although the provided commercial food in this study was of a high quality, there is less understanding of the role of palatability of food for shelter cats in relation to weight loss. The only cat in this study to gain weight (cat 22, experimental group) received medication for diarrhea (fenbendazole 50 mg/kg, PO, q 24 h) mixed with canned food. This gives an indication of the importance of palatability of food for shelter cats.

Although the analysis of the effect of time and the presence of a hiding box on the body weight suggested that there was a difference between the two groups in body weight losses, as cats with hiding boxes showed approximately 40 grams less weight loss compared to the control group, this difference was not significant. For the individual cat, however, this could be biologically relevant, for weight loss due to feline anorexia has a serious impact on a cat’s health, increasing the risks of hepatic steatosis [6,16,17]. Therefore, more research is necessary to monitor these cats for a longer period of time, to register the process of adaptation to the new environment in correlation to the weight losses and to experiment with ways of preventing or reducing body weight losses in shelter cats.

Apart from stress, progressive weight loss can also be a sign of serious medical problems [37]. One of the cats from the control group showed a weight loss close to 20% in 12 days and was eventually diagnosed with FeLV. Shelters could use weight loss during quarantine time as an early warning sign for serious declines in physical conditions, but this would require weighing as a standard monitoring procedure.

Providing cats with hiding enrichment at any stage of their shelter stay requires an investment in shelters’ scarce time and money. Shelter staff sometime have their reservations about using hiding boxes, for it might decrease the visibility of cats to potential adopters and therefore slow down adoption rates (personal communications). Although Kry and Casey found no significant difference in the Length of Stay between cats with and without hiding boxes in the adoption ward [19], in the present study the hiding enrichment itself could not have influenced the adopters’ choices based on the (in)visibility of the cat. Hiding boxes were only present in 12 of the 23 cages during the first 12 days of quarantine time, while no hiding boxes were available in the adoption area.

Our study showed a decrease in the CSS during quarantine time, while eventually both groups had similar rates for adoption and LOS. In sum, providing hiding enrichment to shelter cats benefits the welfare of the animals without having negative consequences for the shelter (like an increased LOS per cat).

Finally, stressors versus signals of safety?

While this study proved again a significant decrease of the behavioural stress response when shelter cats were offered a hiding opportunity, the effects on body weight were minor. These results challenge our point of focus on stress in shelter animals: a shelter environment offers numerous stressors for which feline hiding behaviour appears not always sufficient enough to induce adaptation within the first 12 days after intake. New theories on human stress response mechanisms might shed some light on the feline stress response in these complex shelter environments and contribute to more practical tools for stress reduction. According to Brosschot [38], who introduced the Generalized Unsafety Theory of Stress (GUTS), ‘the stress response of the body is always “on” and it stays on as long as there is no obvious safety.’ This default response can only be inhibited when ‘signals of safety’ are perceived by the animal. We therefore should not look for the causation of a stress response, but rather ask ourselves ‘what stops the stress response?’. When present results are reviewed in the light of this GUTS, the hiding enrichment itself caused a decrease in feline behavioural stress scores, but did not provide an adequate signal of safety (SOS) to prevent weight loss in most cats. This GUTS approach asks for a comparison of the effect of distinct SOSs (such as hiding materials, food presentations, enriched feeding, feline pheromones, human contact, increased cage space, solitary housing, etc) and for the reinforcing effects of combining these signals on the majority of shelter cats. In addition to focusing on reduction of numerous stressors in the shelter environment, we should also search for SOSs that are strong enough to inhibit the stress response and thus create a situation which the majority of animals can perceive as safe.

Limitations

Conducting the present randomized controlled trial research in an operating animal shelter comes with its inherent limitations. Restrictions in research time consequently affects the number of animals available, shelter management does not always coincide with the formation of experimental groups in identical housing while shelter protocols for animal care need to be respected.

This study is based on a small sample of shelter cats and only captures the first 12 days after intake. The evaluation of stress impacting the body weight of cats likely requires a longer assessment period and a larger dataset to be able to apply these insights to a wider cat population in shelters. Despite these limitations, the current findings provide evidence that hiding boxes may not be able to prevent significantly weight loss in shelter cats and are instructive for our understanding of the effects of hiding enrichment on the feline behavioural parameters.

During this research, feeding protocols, based on long-term experience of the shelter staff, were unaltered. By doing so this study created a better understanding of the effect of shelter feline care on behavioural parameters and body weight of these cats. For subsequent research, however, it is recommended to gather data on the daily intake of food and water per individual shelter cat.

The behavioural stress parameter of the Cat-Stress-Score was combined with the physical parameter of body weight. Although stress has previously been shown to negatively affect feline body weight [14], the inclusion of a specific biochemical marker of stress would have strengthened the outcomes of this study. Preferably both physical and biochemical feline stress markers are combined in future studies.

The shelter cats were observed with a video camera mounted in front of their cages, a technique used in several studies [36,39,40,41]. In the previous [12] and current study cats were given a habituation time to positioning the camera of 2 minutes, before starting the observation. To the authors’ knowledge the stress inducing effect of placing a novel static object in front of a cat cage has never been studied in cats. Its effect on the loss of body weight in both groups cannot be ruled out and therefore new research is recommended to investigate the stress inducing effect of this observation method in shelter cats.

Conclusion

Providing hiding boxes can be a relatively simple way for cats to self-manage stress and to adapt faster to the shelter environment. The majority of the shelter cats, however, lost (considerable) weight during the quarantine period in an animal shelter. Providing them with hiding enrichment during that period, does not prevent weight loss. Nor do hiding boxes have effect on the adoption rates and the length of stay of both groups.

However, instead of keeping focus on identifying and reducing stressors in a very challenging environment like an animal shelter, an additional approach could be found in the application of ‘signals of safety’ (SOS), strong enough to attenuate or even inhibit the stress response and thus create a situation that animals can perceive as safe.

Diagram of the experimental set up, observer and both camera position in the quarantine wards in the animal shelter.

(BMP)

Click here for additional data file.

Baseline characteristics of treatment cohorts in a randomized field trial comparing quarantine cat housing with and without hiding opportunities in one Dutch animal shelter.

(BMP)

Click here for additional data file.

21 Jul 2019

PONE-D-19-15024

The effect of a hiding box on stress levels and body weight in Dutch shelter cats; a randomized controlled trial

PLOS ONE

Dear Mrs. van der Leij,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Sep 04 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Juan J Loor

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Yes

Reviewer #4: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

Reviewer #3: Yes

Reviewer #4: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The study confirms previous findings that a hiding box can reduce behavioral indicators of stress in shelter cats. It also attempted to determine if this behavioral change decreased the expected initial weight loss seen in shelter cats. It is not clear that enough animals were evaluated or that the amount of food ingested by each cat was measured. It was clear that not all cats had the same diet (one cat that gained weight was given wet food). So this goal was not reached by the study. Finally, the length of stay of the cats was compared. However, hiding boxes were not used in the adoption area. So the reasoning for how hiding boxes affect length of stay is not clear. Did the authors predict that the cats would be less social when they were removed from the "safety" of the box and transferred to the adoption area?

The study is relatively well-written with a few grammatical errors. However, the discussion should be shortened significantly.

Specific suggestions: Address the concern of low power (low N) for the statistics involving weight loss.

Lines 37-38: Grammar suggestion "...but cats with a hiding box showed a significantly faster..."

Line 101: Spelling "vaccine"

Line 110: Grammar suggestion "Two cats participating in this study left..."

Figure 1 is unnecessary.

Lines 187-189. Clarify the calculation of LOS. Which cats? All cats in the shelter? Only cats in the study?

Lines 222-228: The difference observed in the initial average weight of the groups would be beneficial to include here.

Lines 230-183: Report what the individual cats ate. Did they all finish the food they were given each day?

Shorten the discussion. A great deal of information provided is repetitious. Clarify what is known and what is presumed/thought/supposed.

Measurement of a specific biochemical marker of stress would have been helpful to report.

Reviewer #2: This is a relatively good study, however the housing used for cats in the study was not the same across the 2 groups. Housing may affect stress levels in individual cats and the differences in the size of the housing used looks to be significant - one cage type provided 1221cm squared of floor space and the other was 945cm squared of floor space. It was not indicated that this was controlled for in the study. Of the 12 cats housed with hiding structures 10 were in the larger housing units. The control group was more evenly distributed between small and larger cages with 6 controls in the smaller cage and 5 in the bigger cage. I am concerned this is confounding variable in this study and should be controlled for.

Reviewer #3: Reviewer report for PONE-D-19-15024

This interesting and well-written study that tested the hypothesis that the introduction of a hiding box for cats will decrease their stress. The authors approached the hypothesis by testing it in a shelter environment during the quarantine period, where the level of stress of newly incoming cats is high. The authors found that by providing a hiding box, cats achieved lower stress levels significantly earlier than control cats which did not have a hiding box. I think that the study is very important and its results critical for establishing new regulations for the management of stress levels in shelters across the world. Hence, I strongly support its publications.

Below please find my specific comments:

Lines 50-54: What is the definition of ‘chronic’? I think it will benefit your manuscript if you clearly define a ‘window of time’ that directly correlates with well-defined clinical outcomes.

Line 66: The reference number for Vinke et al., is missing.

Line 79: Because the Ethics committee approved the study, then please provide the protocol number.

Line 102: ‘day 14’ not ‘Day 14’.

Lines 120-122: The structure of this sentence is unclear. It needs to be rewritten.

Line 124: Your manuscript will benefit from clearly defining the method of randomization. Was according to an online randomization tool? Coin toss? Etc.

Line 138: Do you think that the shelter in this study represents the setting in which many other shelters in the Netherlands and worldwide are set-up? In other words, do you know whether many / most other places have cats mixed with dogs and other stressors such as large traffic within the shelter that were not present in this shelter? You are attempting to test a method for stress prevention/reduction which is great, but if the setting of the study may not represent many other facilities, then the external validity of this study is in question. I am writing this from the standpoint of a reviewer who supports the publication of this study, so I am not looking for a defensive response… I am suggesting that you carefully think about this point and that in the revised manuscript you will discuss it and also point out how the results of this study could be leveraged to promote new regulation for shelter management and could be used as a stepping board for upcoming studies.

Lines 150-153: Are there any previous studies that indicated that the introduction of a new static object in front of a cat cage is not a stressful situation by itself. If there are studies that have shown it is not, then please cite them. Otherwise, in the limitation paragraph among the other things that I pointed out, you will need to discuss the possibility that mounting a camera in front of the cage could have induced stress and in theory, could be partially the reason for why there was no significant difference in body weight between the groups. Also, if it was not explored before, you should mention that it is something that needs to be studied in the future.

Line 170-171: This is related to my comment for lines 150-153; how do we know that 2 minutes is enough time. How do you know that it does not lead to prolonged stress…? If there is previous research to support that it is not, then cite it and it will make your study stronger. If you do not know, you will need to discuss it under the limitation paragraph.

Line 213: What did you mean when you used the word ‘constance’? is this a typo?

Figure 2: I strongly suggest that you will consider plotting the figure as a series of grouped boxplots with the CSS on the y-axis and the days it was assessed on the x-axis. Firstly, I think that it will be clearer than a bunch of lines that intersect and overlap with each other. Secondly, the CSS is not a continuum and therefore should not have a connecting line. For example, you do not know if there wasn’t a peak of stress in one of the days when it was not evaluated or during other hours on the days that you did…

Table 1: I strongly suggest that you will consider reconstructing your table such that the left column will have the days; the next column will have the control mean (95CI), and in the last column the hiding box mean (95CI).

Table 2: I strongly suggest that you will consider reconstructing your table such that the left column will have the days; the next column will have the control mean BW change from baseline (95CI) and in the last column the hiding box mean BW change from baseline (95CI). It would be much more meaningful for the reader to know the %change in BW from baseline. You can rerun your regression by incorporating it into your data file as a new column with the percent change for days 7 and 12 relative to baseline.

Figure 3: I strongly suggest that you will consider plotting the figure as a series of grouped boxplots with the %change in body weight on the y-axis and the days it was assessed on the x-axis — same reasons as for figure 1.

Line 298: please delete the word ‘however.’

Line 403: the limitation paragraph should include the limitations I have discussed throughout my reviewer response. In addition, you need to address the limitation that the CSS in this study was not coupled with biomarkers of stress, for example, cortisol.

Reviewer #4: This is an interesting and scientifically sound study that does contribute novel information in a research space increasingly convincing us that cats like boxes. Review the manuscript with a close eye to correct choice of words - getting it proofread by a third party may be helpful here. Examples include e.g. loose instead of lose; expresses instead of expressed.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-19-15024_reviewer(1).pdf

Click here for additional data file.

27 Aug 2019

Please see attached Word file: Response to reviewers PONE-D-19-15024

Submitted filename: Response to reviewers PONE-D-19-15024 - 27 aug 2019.docx

Click here for additional data file.

24 Sep 2019

The effect of a hiding box on stress levels and body weight in Dutch shelter cats; a randomized controlled trial

PONE-D-19-15024R1

Dear Dr. van der Leij,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Juan J Loor

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: All comments have been addressed

Reviewer #4: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #3: Yes

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I'm really sorry, but I just don't feel the study was controlled enough to be sure confounding factors (diet, initial weight, cage size, etc) didn't affect outcome. I recommend repeating the study and considering these factors before publishing any conclusions.

Reviewer #3: I am pleased with the revision that the authors made. I would like to thank the authors for their hard work.

Reviewer #4: The authors have successfully integrated reviewer feedback and strengthened their original submission.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

Reviewer #4: No

7 Oct 2019

PONE-D-19-15024R1

The effect of a hiding box on stress levels and body weight in Dutch shelter cats; a randomized controlled trial

Dear Dr. van der Leij:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Juan J Loor

Academic Editor

PLOS ONE