Accepted Manuscript. An epidemiological analysis of dog behavior problems presented to an Australian behavior clinic, with associated risk factors

Accepted Manuscript An epidemiological analysis of dog behavior problems presented to an Australian behavior clinic, with associated risk factors R. Col, C. Day, C.J.C. Phillips PII: S1558-7878(16)30061-2 DOI: 10.1016/j.jveb.2016.07.001 Reference: JVEB 981 To appear in: Journal of Veterinary Behavior Received Date: 27 April 2016 Accepted Date: 18 July 2016 Please cite this article as: Col, R., Day, C., Phillips, C.J.C., An epidemiological analysis of dog behavior problems presented to an Australian behavior clinic, with associated risk factors, Journal of Veterinary Behavior (2016), doi: 10.1016/j.jveb.2016.07.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 2 An epidemiological analysis of dog behavior problems presented to an Australian behavior clinic, with associated risk factors 3 4 5 6 7 R. Col 1,2, C. Day 1,3 and C.J.C. Phillips 1*, 1 School of Veterinary Science, University of Queensland, Gatton, Queensland 4343, Australia 2 Department of Physiology, Faculty of Veterinary Medicine, Selcuk University, Konya, Turkey 3 Pethealth, Fairfield, Brisbane, Queensland, Australia 8 *Author to whom correspondence should be addressed; E-Mail: c.phillips@uq.edu.au 9 10 Abstract 11 Records of dogs presented to a behavior clinic servicing the eastern seaboard of Australia in and around 12 13 14 15 16 17 18 19 20 21 Brisbane and the Gold Coast were analysed for principal behavior problems and the related factors, from information recorded at the time that the owner approached the clinic for assistance. Data relating to 7,858 dogs presenting with 11,521 behavior problems between 2001 and 2013 were analysed. Twenty-two principal behavior problems were identified, of which the most common, in declining order, were aggression towards people, barking and anxious behavior. Male dogs were at greater risk of several behavior problems, compared to female dogs. Low socioeconomic status of owners and a short period spent at home each week were also associated with a greater risk of several behavior problems. The prevalence of breeds and breed groups presented to the clinic were compared with dog registrations at the local city council, and there were more working dogs, hounds and utility dogs and fewer terriers, toy dogs and non-sporting dogs in the clinic population. Uncommon breeds were over-represented in the clinic 22 23 population compared to local registrations. The risk of developing behavior problems is discussed in the light of evidence about the dogs and their owners. 1

24 25 Key words: dog; behavior; veterinary clinic; aggression; barking; anxious behavior 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 INTRODUCTION Dogs are important to human society for their companionship and the many other roles that they fulfill (Coppinger and Schneider, 1995; Jagoe and Serpell, 1996; Kobelt et al., 2003). However, behavior problems in companion animals present serious public health, economic and animal welfare concerns. We use the term behavior problems to describe any behavior exhibited by a companion animal that is unacceptable to the owner (Wells, 1996; Amat et al., 2009). Canine behavior problems break down the bond between dog and owner (Campbell, 1999), sometimes leading to the abandonment of the dog to a shelter (Patronek et al., 1996) and may ultimately lead to euthanasia of the dog (Patronek et al., 1995; Houpt et al., 1996). In both Australia and the UK about 80% of dogs have been reported as exhibiting behaviors undesirable to owners (Adams and Clark, 1989; O Farrell, 1992; Martınez et al., 2011). Approximately 50% of owners relinquishing their dogs around the world do so because of behavior problems (Patronek et al., 1996; Segurson et al., 2005). In Australia, there are an estimated 4.2 million dogs (ANKC, 2015) and in 2009-2010, 20,177 dogs were euthanized in RSPCA shelters there, 10,784 (53%) of whom met this fate because of behavior problems (RSPCA, 2010). Exhibition of canine behavior problems depends on many risk factors, including breed, age, sex, neuter status, relationship with owners, acquisition and diet (Voith et al., 1997; Wells and Hepper, 2000; Kim et al., 2009; Khoshnegah et al., 2011). The dwelling place is also important: most of the 63% of Australian households that have a pet dog (Animal Health Alliance, 2013) live in suburbia and keep the 45 46 dog confined to household backyards for much of the time (Kobelt et al., 2003). In addition, in modern society many owners work full time and therefore have to leave their dog at home alone for most of the 2

47 48 day, which may result in the dog experiencing anxiety and separation-related behavior problems (Rugbjerg et al., 2003; Rehn and Keeling, 2011; Scaglia et al., 2013). 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 Many researchers have reported associations between specific dog breeds and a high prevalence of certain behavior problems (Podberscek and Serpell, 1997; Bradshaw and Goodwin, 1998; Svartberg, 2006; Takeuchi and Mori, 2006). Although genetic factors undoubtedly play a major role in the underlying mechanism for some behavior problems (Overall 1994, 2005; Overall and Dunham 2002), it is the interaction between genetics, environment and experience that governs behavioral development (Plomin and Asbury, 2005). Breed-related risk factors can result from genetic predispositions to environmental influences (Mertens, 2002). Some epidemiological studies have found that breed and sex-related factors are affected by geographical and cultural differences (Landsberg, 1991; Bamberger and Houpt, 2006). Many environmental factors, and in particular owners characteristics (e.g. age, sex, level of education, levels of experience, extent of interaction with dogs, personality and area of residence) are risk factors for behavior problems (Podberscek and Serpell, 1997; Messam et al., 2008; PerezGuisado and Munoz-Serrano, 2009). Several canine behavior problems, including separation anxiety, can be attributed to responses to anxiety or stress (Horwitz et al., 2002). Dogs with separation anxiety frequently display excessive vocalization, destructive behavior, elimination in the house, pacing, salivating and trembling (McCrave, 1991; Voith and Borchelt, 1996; Overall, 1997). The most commonly reported clinical behavior problems in dogs include aggression, destructiveness, elimination, fearful behavior and barking (Marder and Voith, 1991; Mugford, 1995; Landsberg et al., 1997; Overall, 1997; Wells and Hepper, 2000; Kim et al., 2009). In the United States the most frequent behavior problem is reported to be aggression, which in one study was mostly towards 68 69 people, but dogs also displayed separation anxiety, house soiling, attention-seeking behavior, barking and destructive behavior (Bamberger and Houpt, 2006). Aggression has been cited as the most frequent 3

70 71 complaint regarding canine behavior presented to specialists, reported in 42 66% of all cases seen by small animal behaviorists (Borchelt, 1983; Overall, 1997; Beaver, 1999). In a survey of small animal 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 veterinary practices in Spain (Fatjó et al., 2006), most canine aggression was directed towards other dogs (39%), with 24% and 37% towards family and non-family members, respectively. Lindell (2002) reports that in an unpublished survey conducted in the US of 722 dogs from 502 different households, 76% of dogs showed aggressive behavior, 70% inappropriate elimination, 57% pulled on their lead, 48% were easily excitable and 23% behaved improperly when travelling. Epidemiological studies are important to understand the risk factors for behavior problems and to facilitate the use of the best preventive measures. Although there have been some surveys undertaken to gain information on dogs in Australian households (Adams and Clark, 1989; McHarg et al., 1995; Kobelt et al., 2003), to the best of our knowledge this is the first clinical epidemiological study of a large database of behavior problems in Australian dogs. The objective of the present study was to evaluate the risk factors for behavior problems in dogs. Specifically, we asked whether the following dog and owner characteristics, breed, breed group, pure vs. mixed/crossbred dogs, dog gender, neutered vs. entire status, postcode of the owner and time spent at home by the owner, were associated with increased risk for behavior problems. MATERIAL AND METHODS Participants in the study were dog owners who contacted a Brisbane-based companion animal behavior clinic (Pethealth.com.au). A total of 9,200 questionnaire responses from the self-selecting owners of dogs with behavior problems were downloaded from the database of the clinic. The owners included 'clients' (n 90 91 = 3,171), who paid for a face-to-face consultation, 'customers' (n = 533), who purchased products, e.g., training classes or pet accessories, but not a consultation, and 'prospective clients' (n = 5,496), who 4

92 93 completed the questionnaire but did not have a consultation. Consultations, all of which included treatment, were in one of four forms: a 30 minute telephone consultation 1, a 30 minute in-clinic 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 assessment, a 2 hour in-clinic consultation, and a 2 hour house call. The questionnaire URL was sent to prospective respondents (clients, customers and prospective clients) by e-mail after they had spoken to the staff of the clinic, and the questionnaire was also promoted in social media messages. In some cases (approximately 60% of the respondents), the questionnaire was completed by staff of the clinic during or after telephone conversations with people who called for advice, or during the face-to-face consultation. Method of contact did not influence the completion of the form, as the choice of responses was determined by the owners who dictated the information, not the six clinic staff who assisted with data entry. The clinical consults for the dog owners were provided by a professional veterinary behaviourist. 2 Data were collected from 2001 to 2013, using questionnaires completed during that period for the purpose of facilitating treatment. A total of 1,342 responses were discarded because of incomplete information, leaving 7,858 dogs analysed in this study. Of the 6,408 respondents that provided postcodes, most (79%) were from SE Queensland, with 85% from Queensland state, 5.9% New South Wales, 5.2% Western Australia, 2.8% Victoria, 0.8% South Australia and 0.4% Tasmania. The objective of the questionnaire was to record 1) the nature of the behavior problem(s), 2) information about the owner and 3) information about the dog. 1) The behavior problem. Respondents were first asked to provide details for the animal with the most significant problem, selected from a drop-down list of behaviors, but without the definitions (Table 1), with the option to additionally or alternatively enter descriptions of the 1 Indirect contact with the client long distance by telephone (who then talks to their vet) is a legal practice in Australia 2 Dr Cam Day, BSc, BVSc, MANZCVS (Veterinary Behaviour), Adjunct Associate Professor at the University of Queensland s School of Veterinary Science, and full-time small animal veterinary behavior consultant since 1995. 5

112 113 problem in their own words in a free-text field. Other behavior problems could be described in the free-text field. At the time of the consultation the identification of principal and secondary 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 behavior problems was verified by the consultant veterinary behaviorist and changed if necessary. 2) The owner s name, postcode, location and work routine. The questionnaire database was designed to accommodate one record per owner. Owner home postcode data was used as a correlate of socio-economic status, using Socio-Economic Indices for Areas (Pink, 2011). The Australian Bureau of Statistics (Pink, 2011) indicated that the median human population per postcode at the end of our study period was 21,232 (range 165-102,259 people). The following two indices were included in the logistic regression model: The Index of Relative Socio- Economic Disadvantage (RSED) indicates relative disadvantage, on a continuum from most disadvantaged (low values) to least disadvantaged (high values), based on income, educational attainment, unemployment, and dwellings in an area. The Index of Education and Occupation (EO) indicates educational and occupational variables of relative socio-economic advantage (high values) and disadvantage (low values) regarding the skills of the people in an area, from both formal qualifications and the skills required to perform different occupations. The work routine described their mean time away from home each day (options: mostly home, <20, 20-40, 40-60 h at work or variable (shift) work routine). 3) The dog s breed, purebred or mixed/crossbred status, sex, neuter status and age. Data on the dog s breed was nominated in a free-text field and in cases where no breed was entered the data was classified as unknown. The animal's approximate date of birth was also entered in a free- 133 134 135 text field, which was used in combination with the date entered in the database to estimate the animal's age at the time of reporting the problem behavior to the nearest half year. Breed descriptions were standardised by data cleaning to allow analysis. In the minority of cases 6

136 137 where a dog was listed as mixed/crossbred, one or more breeds could be listed. If more than one predominant breed was provided, only the first breed description was used. We adopted the 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 breed type categorization of the Australian Kennel Club (including as a cross/mixed-breed), with groups as follows: Working, Terriers, Toys, Gundogs, Non Sporting, Utility, and Hounds (AKC, 2014). Separately, numbers of dogs of the different breeds that were registered with a large municipal authority servicing part of the catchment area of the practice (Gold Coast City Council) were obtained for 2010 in order to compare breed and breed group distribution with that in our survey. Statistical analysis The data were analysed in two stages. First, descriptive statistics on the behavior problems were inspected. For the purposes of maintaining adequate numbers of dogs in the different categories, aggression to people (82% of aggression cases), other dogs (16%), predatory (1%) and general (unattributed) aggression (1%) were merged for regression analysis. Logistic regression models were constructed by introducing possible combinations of predictor variables until those that best explained the variance in the dependent variable were obtained. All variables (age, sex, neuter status, breed group and breed) were entered into the model initially, with levels of variables with a binary input (neutered or entire, purebred or crossbred/mixed breed, male or female) compared directly, and the nominal variables with more than two input levels compared to a reference level. The reference group for breed was chosen to be dogs of unknown breed because this category contained a large number of dogs and did not selectively focus on one particular 156 157 breed. Results were discarded for individual breeds in which the number affected was less than 5, because the data was believed to be too susceptible to individual influences. Similarly, the reference group for work 7

158 159 routine and postcode were the dogs for which the level was unknown. Data were analysed by a regression model with logit function to determine relationships between independent variables and dog behavior, 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 using the statistical package Minitab, version 16. Coefficients, P values, odd ratios (OR) and 95% confidence intervals for the OR (CI 95) were calculated for the final risk factors associated with dogs exhibiting the behavior problems. Results were considered significant when P 0.05. The numbers of dogs in each breed and breed group was compared with the Gold Coast City Council (GCCC) registrations by chi square analysis. An over-representation coefficient was calculated by dividing the % of each breed in our study by the % of each breed in the GCCC registrations. Non-linear regression of over-representation of breeds, in relation to GCCC registrations, against breed % in our dataset was tested with asymptotic, exponential and power functions, using a Gauss-Newton algorithm and a maximum of 200 iterations. The resulting fits were compared with a Lack-of-Fit test, which produced the highest P value of 0.99 for the power function after 13 iterations. Results Owner characteristics Postcode information was obtained for 6,347 cases (81%), but the clients normal home/work routine was mostly unreported (n=5,932, 76%). Of those responding to the query about daily routine, the most common category was 20-40 h/week at work (n=647, 8.2%) or mostly home (n = 635, 8.1 %). Less commonly, respondents reported a variable work routine (shift-work) (n=372, 4.7%) or spending 40-60 h/week at work (n=269, 3.4%). Few responders reported spending less than 20 h/week at work (n=3, <0.01%). 178 8

179 180 Dog characteristics The most common principal or secondary problem was aggression (n = 2,928, 37% of dogs), followed by 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 barking (n = 1,101, 14%) and unspecified anxious behavior (n = 919, 12%). Other categorical behavior problems were reported in 5% of dogs or fewer (Table 1). A total of 34 % (n = 2,672) of the dogs were mixed/crossbred breeds, 58% (n = 4,539) were purebreds and the remainder were not specified (8%, n=647). The dogs in our sample were 46.9% males (n = 3,684), 40.4% females (n = 3,179) and 12.7% (n = 995) of unknown sex. Of the males, 24.5% were entire/intact (n = 904) and 75.5% were neutered (n = 2,780). Of the females, 21.4% were entire (n = 680), and 78.6% were neutered (n = 2,499). The dog s age was obtained for 7,001 cases (89.1%), with a mean age of 3.6 years (SEM 0.040) (Figure 1). The most common breeds were working dogs (n = 1,581, 20%), followed by terriers (n = 1,483, 19%), toy dogs (n = 1,151, 15%), gundogs (n = 899, 11%), non-sporting (n = 775, 10%), utility dogs (n = 768, 10%), and hounds (n = 490, 6%) (Table 2). The rest of the dogs were of unknown breed (n=711, 9%). The distribution of breed groups in our dog database was broadly similar to that of the Gold Coast City Council (GCCC), however, our sample had proportionately more working dogs, hounds and utility dogs and fewer non-sporting, terriers and toy dogs (P< 0.001). Apart from the dogs of unknown or other breeds, the most common breed in our sample was the Staffordshire bull terrier (n = 725, 9%), then the Maltese (n = 493, 6%), Border collie (n = 439, 6%), Labrador (n = 366, 5%), German shepherd (n = 361, 5%) and Australian cattle dog (n = 361, 5%) (Table 3). In the Gold Coast sample the most numerous breed was the Maltese, then the Labrador and Border collie, with considerably smaller numbers of Australian cattle dogs and German shepherds. The following breeds were most over-represented in our database, compared to the GCCC registrations, in declining order of over-representation: Mastiff, German shorthaired pointer, 200 201 202 Belgian shepherd, Maremma sheepdog, greyhound, Weimaraner and collie. The following breeds were most under-represented in our database compared to the GCCC registrations, in declining order of under- representation: Fox terrier, Pomeranian, shih tzu, bichon frisé and Lhasa apso. Breeds with small numbers 9

203 204 in our database were over-represented, relative to the Gold Coast City Council registrations (Figure 2) by the following relationship: 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 y = 1.48 (SE 0.175) B -0.374 (SE 0.166) Equation 1 where y = over- representation coefficient, with 1 = proportion indicated for the region (Gold Coast City Council), > 1 indicating overrepresentation and < 1 indicating underrepresentation and B = breed % in our database Risk factors for behavior problems Table 4 shows the number of purebred and mixed/crossbred breed dogs in each of two behavior categories, for example, 15 dogs with the behavior problem chasing. Purebred dogs were at greater risk of presenting for chasing (P = 0.02) and boisterous behavior (P = 0.05), compared with mixed/crossbred dogs. Breed risks for each behavior problem (Table 5) include the number of dogs of each breed with each problem, the percentage of dogs with the behavior problems that is attributed to each breed and the percentage of total dogs of the breed that had the behavior problem. The last value is compared statistically to the percentage of dogs of unknown breed that have the behavior problem and an odds ratio for over- or underrepresentation is presented. Table 5 only shows the breeds with significant OR, hence the number indicated for chasing does not add up to the 15 identified in Table 4, in fact, only 2 are listed for specific breeds. Australian cattle dogs showed less pica, (OR 0.04, P = 0.04), and German shepherds were more at risk for obsessive behavior (OR 7.0, P = 0.03) relative to unknown dogs. Many breeds were at low risk of fear of noises, relative to unknown dogs, but this was especially true for Dobermans (OR 0.07, P = 0.01), 223 224 Rhodesian ridgebacks (OR 0.11, P < 0.01), Staffordshire bull terriers (OR 0.12, P < 0.001) and Australian cattle dogs (OR 0.13, P < 0.001). Two breeds, Rottweilers (OR 4.9, P < 0.01) and schnauzers (OR 4.1, P = 10

225 226 0.05), were at increased risk of house soiling. Only one breed was at increased risk of fearful behavior, bull mastiffs (OR 4.6, P = 0.05), and one at less risk, Australian cattle dogs (OR 0.35, P = 0.03). Many breeds 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 were at low risk of escape behavior, but this was pronounced for cocker spaniels (OR 0.04, P < 0.01), dachshunds (OR 0.06, P = 0.01) and German shepherds (OR 0.07, P < 0.01). Many breeds were at low risk for destructive behavior, particularly Rhodesian ridgebacks (OR 0.05, P < 0.01) and Australian cattle dogs (OR 0.09, P < 0.001). Two breeds, Australian cattle dogs (OR 0.02, P < 0.001) and Staffordshire bull terriers (OR 0.04, P < 0.001), were at particularly low risk of chasing behavior, and these two breeds were also at reduced risk of boisterous behavior (OR 0.28, P = 0.05) and circling behavior (OR 0.05, P = 0.05). Many dogs were at increased risk of barking, especially German shorthaired pointers (OR 7.6, P = 0.02), Rottweilers (OR 4.5, P < 0.01), great Danes (OR 4.1, P = 0.03), Jack Russell (OR 3.9, P < 0.01) and Staffordshire bull terriers (OR 3.4, P < 0.01). Breeds at low risk for anxious behavior included Alaskan malamutes (OR 0.12, P = 0.01), pugs (OR 0.16, P = 0.04), and Australian cattle dogs (OR 0.18, P < 0.001). Aggression was at particularly low risk in Weimaraners (OR 0.16, P < 0.01), Dalmatians (OR 0.19, P = 0.01), Alaskan malamutes (OR 0.25, P = 0.01), Australian cattle dogs (OR 0.28, P < 0.001) and dachshunds (OR 0.29, P < 0.01). Males (M) were generally at greater risk than females (F) for a number of behavior problems, in particular coprophagia (M 0.5%, F 0.2%, OR 0.36, P = 0.03), house soiling (M 5.8%, F 3.1%, OR 0.40, P < 0.01) and boisterous behavior (M 2.9%, F 1.7%, OR 0.44, P < 0.01) (Table 6). Neutered dogs (N) were generally at higher risk of developing many behavior problems compared with entire dogs (E): integration into the household (N 0.5%, E 0.2%, OR 0.15, P <0.01), obsessive behavior (N 0.5%, E 0.2%, OR 0.18, P = 0.01), digging (N 0.4%, E 0.2%, OR 0.21, P = 0.01), attention seeking (N 1.1%, E 0.5%, OR 0.22, P < 246 247 248 0.001) and grieving (N 0.5%, E 0.3%, OR 0.22, P = 0.01) (Table 7). Neutered dogs were at slightly lower risk of developing aggression (N 37.3%, E 37.9%, OR 0.47, P < 0.01) The dog s age was not associated with any reported behavior problems. 11

249 250 Seven behavior problems had associations with the socioeconomic indices (Table 8). 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 Socioeconomic disadvantage (low RSED) of the owner was associated with high levels of house soiling (OR 0.97, P< 0.01), escape behavior (OR 0.97, P=0.03), anxious behavior (OR 0.97, P< 0.01), destructive behavior (OR 0.96, P< 0.01), barking (OR 0.97, P< 0.01) and aggression (OR 0.97, P< 0.01). In the case of the last four this was accompanied by evidence of disadvantage in terms of educational and occupational variables (low E-O). For work routine, numbers returning < 20 h at work were too few for meaningful analysis (Table 9). Of the other options, several behaviors house soiling, destructive behavior, coprophagia, chasing, boisterous behavior, barking, anxiety-related and aggression showed similar patterns with work routine. Risks were greatest for dogs whose owners spent 20-40 h and 40-60 h at work, reduced for dogs with owners mostly home, with variable attendance intermediate or similar to mostly home. Of the other behaviors with significant connections to owners work routine, pica was particularly likely if owners were in variable attendance (OR 11.0, P = 0.01), obsessive behavior was particularly likely if owners were at work 20-40 h (OR 9.9, P < 0.01) or in variable attendance (OR 6.6, P = 0.03), fearful behavior was most likely for dogs whose owners were at work 20-40 h (OR 4.9, P < 0.01) or with variable attendance (OR 4.5, P <0.01), escape behavior was most likely for dogs with owners at work 40-60 h (OR 12.8, P < 0.001), and attention seeking was most likely for dogs with owners reporting variable attendance (OR 4.3, P = 0.04). DISCUSSION 270 271 Dogs exhibit a range of behaviors related to their emotional state (Beaver, 1982). Experience, genetics and environment are the three dominant influences on most aspects of behavior (Gottesman and Hanson, 12

272 273 2005). In this study, the dog s breed, sex, neuter status and the owners socioeconomic status and work routine were all associated with some of the listed behavior problems assayed through an Australian 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 behavior clinic. In this population, behavior problems were most affected by neutering status (18 behavior problems affected), work routine (14 behavior problems), breed (13 behavior problems), gender (10 behavior problems), and socioeconomic status (7 behavior problems). Effects of work routine may be particularly important for future study since even though few respondents completed this section in our questionnaire, it was still clearly related to behavior problems. Our study did not compare the clinic population to a general one, so there may be effects of other animal-related factors (e.g., number of dogs in the household, exercise schedule, indoors vs. outdoor housing, enrichment) on behavior problems as reported by other authors (Kubinyi et al., 2009, Herron et al., 2014; Westgarth et al., 2015). Most dog owners in suburban Australia have their animal neutered, but in our study and that of Kobelt et al. (2003) there were fewer neutered males than bitches. Neutered dogs were over-represented for integration into the household, obsessive behavior, digging, attention-seeking and grieving. Although neutered dogs were generally at higher risk than entire/intact dogs of developing these behaviour problems, the latter were slightly more prone to aggressive behavior in our study, supporting Blackshaw s (1991) study. The logistic regression model that we created, with several thousands of dogs, is capable of detecting significant differences that are small and potentially of little practical significance. In other studies aggressiveness has been reported to be more frequent in neutered than entire females (Borchelt, 1983; Wright and Nesselrote, 1987; Wright 1991), at similar levels in both (van den Berg et al., 2006; Bennett and Rohlf, 2007; Martınez et al., 2011) and equally common in neutered males and females (Guy et al., 2001). 293 294 Breed-associated risk for behavior problems may be associated with the breed distribution in the geographic area (Overall, 1997; Reisner et al., 2005), the popularity of the breed (Svartberg, 2006), and 13

295 296 with the reasons the owners selected the dog. Different behaviour problems have been observed in surveys in Australia (Kobelt et al., 2003) and United States (Beaver, 1994). 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 Our study found that males in this consulting and survey population had a greater risk of several behavior problems house soiling, fearful and anxious behavior, destructive and boisterous behavior, coprophagia, barking and attention seeking. Some studies have shown males at a greater risk for many behavior problems than females (Borchelt, 1983; Wright and Nesselrote, 1987; Landsberg, 1991; Hsu and Serpell, 2003; Fatjó et al., 2007; Perez-Guisado and Munoz-Serrano, 2009). One study reported that phobias were more common in female dogs (Bamberger and Houpt, 2006). We found no significant effects of age, but some of the younger dogs exhibiting behavior problems may have been relinquished to shelters or put down, which may bias our results. Because people may try to cope with a dog s behavior problems for some time before seeking help, the time at which behavior problems are reported may not represent when the behaviour started or was at its most extreme. Low activity levels during separation and repeated, long term absence of both physical and mental stimulation when owners are away (Norling and Keeling, 2010) may be associated with separation-related behavior problems (Svartberg, 2006; Rehn and Keeling 2011). Our results suggest if time spent at home is an important factor in minimizing the risk of behavior problems. Breed, low socioeconomic status and work routine away from the house were correlated in our study. There is a positive relationship between low-income and working long hours (Mehdikarimi et al., 2015). Low socioeconomic status may increase hours spent away from the home, hence these two factors were correlated. Our second most common behavior problems was barking, which is commonly reported to veterinary behaviorists (Cross et al., 2009; Hassan et al., 2009; Rafie et al., 2011). Barking comprised 32% 316 of behavior problems in one survey (Kobelt et al., 2003) and is especially common in juveniles, rather than 14

317 318 puppies or adults (Wells and Hepper, 2000). Vocalization can be a normal response to stressors in dogs (Casey, 2002), and has been reported in hospitalized dogs who are left alone (Vaisanen et al., 2005). 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 Elimination (house soiling) was our fifth most common behavior problem and breed (Rottweiler and Schnauzer), male sex, socioeconomic disadvantage and work routines away from the house were risk factors. We had a group of less common behavior problems that were apparently related: destructive, boisterous behavior and escape behavior can be associated with excessive activity. In this study, breed and limited time spent at home were related to all three behavior problems. Expectations of the owners are clearly important to consider before dogs are adopted or purchased. Limitations of the study We made an assumption that the breed identification by the owner was accurate. This was not able to be verified in all cases. Inaccuracies were most likely to have been in the mixed/crossbreed dogs (34% of our dogs), which we treated as a separate group. We assumed that the first listed breed was dominant in the cross. Regardless, whether a dog was a purebred vs. a crossbred was not a significant risk factor for most behavior problems (20 out of 22) discussed. We also made assumptions that breed groups were relevant categories that can be judged according to the Australian Kennel Club classification. Caution is warranted as the behavior problem data were largely self-reported, with the assistance of a trained professional. Our comparison group was the Gold Coast sample, but our sample was from a broader 337 338 geographical range. The Gold Coast data was for registered dogs only, with registration being compulsory in the area but compliance not accurately known. 15

339 340 The choice of referent population for determination of risk factors for breeds in this study was problematic, since no single breed appeared suitable for this purpose. Thus we chose dogs for whom no 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 breed had been entered, unknown dogs, since this was a numerous category. We acknowledge that some known breeds were likely included in this category Respondents interpretation of the range of options for behavior problems may also have varied, and distinguishing similar or related behaviours (e.g., fearful vs. anxious behaviours) may have been difficult. Where possible, assistance was provided during consultations. Conclusions An analysis of dog and owner characteristics of cases at a dog behavior clinic has revealed the principal behavior problems in this region, in declining order, aggression, barking and anxious behavior. Breed information identified that more owners of working dogs, hounds and utility dogs, and fewer owners of terriers, toy dogs and non-sporting queried the clinic about their dogs behaviors. Breeds that were uncommon in our database were over-represented, compared to Gold Coast City Council registrations. Other risk factors included dog sex, with males being at greater risk for several behaviors, low socioeconomic status of owners and a short period spent at home each week. A greater understanding of the risk factors associated with behavior problems will enable them to be addressed and the relationship between dogs and humans improved. 16

358 359 Acknowledgements No external funding was provided for this study. R. Col received a scholarship from Council of Higher 360 361 362 363 364 365 366 367 368 369 370 371 372 Education of Turkey to study at the University of Queensland. Conflict of Interest statement Dr. Cam Day manages the pet behavior consultancy company, Pethealth.com.au, which generated the data for this study. There are no other conflicts of interest. Authorship statement The idea for the study was conceived by Cam Day and it was designed by Clive Phillips. The study was performed by Ramazan Col and Clive Phillips. The data were collected by Cam Day. The data were analyzed by Clive Phillips and Ramazan Col. The paper was written by Ramazan Col and Clive Phillips, in consultation with Cam Day. 17

373 REFERENCES 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 Adams, G.J., Clark, W.T., 1989. The prevalence of behavioural problems in domestic dogs; a survey of 105 dog owners. Aust. Vet. Pract. 19, 135 137. Amat, M., Manteca, X., Mariotti, V., Torre, J.L.R., Fatjó, J., 2009. Aggressive behavior in the English cocker spaniel. J. Vet. Behav.: Clin. Appl. Res. 4, 111-117. Animal Health Alliance 2013. Pet Ownership in Australia Summary (http://petsinaustralia.com.au). Last accessed 1 June, 2015. Australian National Kennel Club (ANKC) 2015. A Forensic View of Puppy Breeding in Australia.http://ankc.org.au/media/1204/a-forensic-view-of-puppy-breeding-in-australiav2.pdf. Last accessed 29 September, 2015. Association of Pet Behaviour Counsellors (APBC), 2005. Annual Review of Cases 2005. Available online. http://www. apbc.org.uk/news_items/review_2005.htm. Last accessed 10 September, 2014. Australian Kennel Club (AKC) 2104. Dog breeds. http://ankc.org.au/. Last accessed 10 September, 2014. Bamberger, M., Houpt, K.A., 2006. Signalment factors, comorbidity, and trends in behaviour diagnoses in dogs: 1,644 cases (1991-2001). J.Am. Vet. Med. Assoc. 229, 1591-1601. Beaver, B.V., 1982. Distance-increasing postures of dogs. Vet. Med. Small Anim. Clin. 77, 1023-1024. 389 Beaver, B.V., 1994. Owner complaints about canine behavior. J. Am. Vet. Med. Assoc. 204, 1953 1955. 390 Beaver, B., 1999. Canine Behavior: A Guide for Veterinarians. Saunders, St. Louis, pp 137-199. 18

391 392 Bennett, P.C., Rohlf, V.I., 2007. Owner companion dog interactions: relationships between demographic variables, potentially problematic behaviours, training engagement and shared activities. Appl. 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 Anim. Behav. Sci. 102, 65 84. Blackshaw, J.K., 1991. An overview of types of aggressive behaviour in dogs and methods of treatment. Appl. Anim. Behav. Sci. 30, 351-361. Borchelt, P.L., 1983. Aggressive behaviour of dogs kept as companion animals: classification and influence of sex, reproductive status and breed. Appl. Anim. Ethol. 10, 45-61. Bradshaw, J.W.S., Goodwin, D., 1998. Determination of behavioral traits of pure-bred dogs using factor analysis and cluster analysis; a comparison of studies in the USA and UK. Res. Vet. Sci. 66, 73-76. Bradshaw, J.W.S., McPherson, J.A., Casey, R.A., Larter, I.S., 2002. Aetiology of separation related behaviour in domestic dogs. Vet. Rec. 151:43 46. Campbell, W.E. Behavior Problems in Dogs. BehavioRx Systems, Grants Pass, 1999. pp. 256-270. Cannas, S., Frank, D., Minero, M., Godbout, M., Palestrini, C., 2010. Puppy behaviour when left home alone: changes during the first few months after adoption. J. Vet. Behav.: Clin. Appl. Res. 5, 94-100. Casey, R., 2002. Fear and stress in companion animals. In: Horwitz, D., Mills, D., Heath, S. (Eds.), BSAVA Manual of Canine and Feline Behavioural Medicine. British Small Animal Veterinary Association, Gloucester, UK, pp. 144-153. 19

409 410 Coppinger, R., Schneider, R., 1995. Evolution of working dog. In: Serpell, J.A. (Ed.), The Domestic Dog: Its Evolution, Behaviour and Interactions with People. Cambridge University Press, Cambridge, pp. 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 21 47. Cross, N.J., Rosenthal, K.J., Phillips, C.J.C., 2009. Risk factors for nuisance barking in dogs. Aust. Vet. J. 87, 402 408. Duffy, D.L., Hsu Y., Serpell, J.A., 2008. Breed differences in canine aggression. Appl. Anim.Behav. Sci. 114, 441-460. Fatjó, J., Ruíz de la Torre, J.L., Manteca, X., 2006. The epidemiology of behavioural problems in dogs and cats: a survey of veterinary practitioners. Anim. Welf. 15, 179-185. Fatjó, J., Amat, M., Mariotti, V.M., Ruiz de la Torre, J.L., Manteca, X., 2007. Analysis of 1040 cases of canine aggression in a referral practice in Spain. J. Vet. Behav. Clin. Appl. Res. 2, 158-165. Flannigan, G., Dodman, N.H., 2001. Risk factors and behaviours associated with separation anxiety in dogs. J. Am. Vet. Med. Assoc. 219, 460 466. Gottesman, I.I. Hanson, D.R., 2005. Human development: biological and genetic processes. Ann. Rev. Psych. 56, 263-286. Guy, N.C., Luescher, U.A., Dohoo, S.E., Spangler, E., Miller, J.B., Dohoo, I.R., Bate, L.A., 2001. Demographic and aggressive characteristics of dogs in a general veterinary caseload. Appl. Anim. Behav. Sci. 74, 15-28. 20

427 428 Hassan, A.Z., Emmanuel, E.G., Awasum, C.A., Remi-Adewunmi, B.D., Hassan, F.B.,Mohammed, A., 2009. Behavioural conditions in dogs, a review: Part I Disagreeable normal behaviors. Nig. Vet. J. 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 30, 45 50. Horwitz, H., Mills, D.S., Heath, S., 2002. BSAVA Manual of Canine and Feline Behavioural Medicine. British Small Animal Veterinary Association, Gloucester, UK. Houpt, K.A., Honig, S.U., Reisner, I.R., 1996. Breaking the human companion animal bond. J.Am. Vet. Med. Assoc. 208, 1653-1659. Hsu, Y., Serpell, J.A., 2003. Development and validation of a questionnaire for measuring behavior and temperament traits in pet dogs. J. Am. Vet. Med. Assoc. 223, 1293 1300. Ingram, A., Jones, P., Heath, S.E., 2010. Study into the effects of attendance at formal traınıng on the presentatıon of aggressıon-related behavior problems in dogs in a referred populatıon. J. Vet. Behav.: Clin. Appl. Res. 5, 1. Jagoe, A., Serpell, J., 1996. Owner characteristics and interactions, and the prevalence of canine behaviour problems. Appl. Anim. Behav. Sci. 47, 31 42. Khoshnegah, J., Azizzadeh, M., Mahmoodi, G.A., 2011. Risk factors for the development of behavior problems in a population of Iranian domestic dogs: result of a pilot survey. Appl. Anim. Behav. Sci. 131, 123-130. Kim, Y.M., Abd El-Aty, A.M., Hwang, S.H., Lee, J.H., Lee, S.M., 2009. Risk factors of relinquishment 445 regarding canine behavior problems in South Korea. Berl. Munch. Tierarztl. Wochenschr. 122, 1-7. 21

446 Knol, B.W., 1987. Behavioral problems in dogs. Vet. Q. 9, 226 234. 447 Kobelt, A.J., Hemsworth, P.H., Barnett, J.L., Coleman G.J., 2003. A survey of dog ownership in suburban 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 Australia-conditions and behaviour problems. Appl. Anim. Behav. Sci. 82, 137148. Landsberg, G.M., 1991. The distribution of canine behaviour cases at three behaviour referral practices. Vet. Med. 86, 1081-1089. Landsberg, G., Hunthausen, W., Ackerman, L., 1997. Handbook of Behavior Problems of the Dog and Cat. Butterworths Heinemann, Oxford. Lindell, E., 2002. Control problems in dogs. In: Horwitz, D., Mills, D. and Heath, S. (eds) BSAVA Manual of Canine and Feline Behavioural Medicine. British Small Animal Veterinary Association, Gloucester, UK, pp 69-79. Lund, J.D., 2001. Forekomst af adlaerdsproblemer hos familiehunde i Danmark (in Danish). The Danish Animal Welfare Society and The Royal Veterinary and Agricultural University. Federiksberg Bogtrykkeri, Denmark, 78 pp. Marder, A.R., Voith, V.L., 1991. Veterinary Clinics of North America, Small Animal Practice, Advances in Companion Animal Behavior, vol. 21. Saunders, Philadelphia. Martınez, A.G., Pernas, G.S., Casalta, F.J.D., Rey, L.S., Palomino, L.F.D.C., 2011. Risk factors associated with behavioral problems in dogs J. Vet. Behav: Clin. Appl. Res. 6, 225-231. 463 464 McCrave, E.A., 1991. Diagnostic criteria for separation anxiety. Vet. Clin. North Am.: SmallAnim. Pract. 21, 247 255. 22

465 466 McHarg, M., Baldock, C., Headey, B., Robinson, A., 1995. National People and Pets Survey, a Report to the Urban Animal Management Coalition, Sydney. 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 Mehdikarimi, S., Norris, S., Stalzer, C., 2015. Regression Analysis of the Relationship between Income and Work Hours. Georgia Institute of Technology Econometric Analysis (ECON 3161), Dr. Shatakshee Dhongde. Mertens, P.A., 2002. Canine aggression. In: Horwitz, D., Mills, D., Heath, S. (Eds.), BSAVA Manual of Canine and Feline Behavioural Medicine, 1st edition. British Small Animal Veterinary Association Gloucester, England, pp. 195 199. Messam, L.L.M., Kass, P.H., Chomel, B.B., Hart, L.A., 2008. The human canine environment: a risk factor for non-play bites? Vet. J. 177, 205 215. Mills, D. S. Marchante-Forde, J., McGreevy, P.D., Morton, D.B., Nicol, C.J., Phillips, C.J.C., Sandøe, P. and Swaisgood, R.R. (Eds). 2010. The Encyclopedia of Applied Animal Behaviour and Welfare. Wallingford, Oxon, CABI Publishing. Mugford, R.A., 1995. Canine behavior therapy. In: Serpell, J.A. (Ed.), The Domestic Dog: Its Evolution, Behavior and Interactions with People. Cambridge University Press, Cambridge, pp. 140 152. Norling, Y., Keeling, L., 2010. Owning a dog and working: a telephone survey of dog owners and employers in Sweden. Anthrozoös 23, 157 171. O Farrell, V., 1992. Introduction. In: O Farrell, V. (Ed.), Manual of Canine Behaviour. British Small 483 Animal Veterinary Association, Gloucester, UK, pp. 11-14. 23

484 485 Overall, K.L., Dunham, A.E., 2002. Clinical features and outcome in dogs and cats with obsessive compulsive disorder: 126 cases (1989-2000). J. Am. Vet. Med. Ass. 221, 1445-1452. 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 Overall, K.L., 1994. Use of clomipramine to treat ritualistic stereotypic motor behavior in three dogs. J. Am. Vet. Med. Ass. 205,1733-1741 Overall, K.L., 1997. Canine aggression. In: Clinical Behavioral Medicine for Small Animals. Mosby-Year Book, St. Louis, pp 88-137. Overall, K. L., 2005. Veterinary behavioural medicine: a roadmap for the 21st century. Vet. J. 169,130-143. Patronek, G.J., Glickman, L.T., Beck, A.M., McCabe, G.P., Ecker, C., 1996. Risk factors for relinquishment of dogs to an animal shelter. J. Am. Vet. Med. Ass. 209, 572-581. Patronek, G.J., Glickman, L.T., Moyer, M.R., 1995. Population dynamics and the risk of euthanasia for dogs in an animal shelter. Anthrozoös 8, 31-43. Perez-Guisado, J., Munoz-Serrano, A., 2009. Factors linked to dominance aggression in dogs. J. Anim. Vet. Adv. 8, 336 342. Pink, B., 2011. Socio-economic indices for areas. Australian Bureau of Statistics, Canberra. http://www.ausstats.abs.gov.au/ausstats/subscriber.nsf/0/22ceda8038af7a0dca257b3b00116e 34/$File/2033.0.55.001%20seifa%202011%20technical%20paper.pdf). Last accessed 10 September, 2014. 24

502 503 Plomin, R., Asbury, K., 2005. Nature and nurture: genetic and environmental influences on behavior. Ann. Am. Acad. Pol. Soc. Sci. 600, 86. 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 Podberscek, A.L., Serpell, J.A., 1997. Environmental influences on the expression of aggressive behaviour in English cocker spaniels. Appl. Anim. Behav. Sci. 52, 215 227. Rafie, S.M., Sattari, B., Naderi, M.M., Ale-Davoud, S.J., Seif, A.A., Bokaei, S., 2011. The prevalence of behaviour problems in domestic dogs in Tehran-Iran. Global Vet. 6, 333 338. Rehn, T., Keeling, L.J., 2011. The effect of time left alone at home on dog welfare. Appl. Anim. Behav. Sci. 129, 129-135. Reid, J.B., Chantrey, D.F., Davie, C., 1984. Eliminatory behaviour of domestic dogs in an urban environment. Appl. Anim. Behav. Sci. 12, 279-287. Reisner, I.R., Houpt, K.A., Shofer, F.S., 2005. National survey of owner-directed aggression in English Springer Spaniels. J. Am. Vet. Med. Ass. 227, 1594-1603. RSPCA, 2010, RSPCA Australia national statistics 2009-2010, RSPCA, Australia, http://www.rspca.org.au/assets/files/resources/rspcaannualstats2009-2010.pdf. Last accessed 6 March 2011. Rugbjerg, H., Proschowsky, H.F., Ersboll, A.K., Lund, J.D., 2003. Risk factors associated with interdog aggression and shooting phobias among purebred dogs in Denmark. Prev. Vet. Med. 58, 85-100. 519 520 Scaglia, E., Cannas, S., Minero, M., Frank, D., Bassi, A., Palestrini, C., 2013. Video analysis of adult dogs when left home alone J. Vet. Behav.: Clin. Appl. Res. 30, 1-6. 25

521 522 Scott, J.P., Fuller, J.L., 1965. Genetics and the Social Behavior of the Dog (1997 reprint). The University of Chicago Press, Chicago. 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 Segurson, S.A., Serpell, J.A., Hart, B.L., 2005. Evaluation of a behavioral assessment questionnaire for use in the characterization of behavioral problems of dogs relinquished to animal shelters. J. Am. Vet. Med. Assoc. 227, 1755-1761. Strandberg, E., Jacobsson, J., Saetre, P., 2005. Direct genetic, maternal and litter effects on behaviour in German shepherd dogs in Sweden. Livestock Prod. Sci. 93, pp 33-42. Svartberg, K., 2006. Breed-typical behavior in dogs - historical remnants or recent constructs? Appl. Anim. Behav. Sci. 96, 293 313. Takeuchi, Y., Ogata, N., Houpt, K.A., Scarlett, J.M., 2001. Differences in background and outcome of three behavior problems of dogs. Appl. Anim. Behav. Sci. 70, 297 308. Takeuchi, Y., Houpt, K.A., 2003. Behavior Genetics. Veterinary Clinics of North America: Small Anim. Pract. 33, 345-363. Takeuchi, Y., Mori, Y., 2006. A comparison of the behavioural profiles of purebred dogs in Japan to profiles of those in the United States and the United Kingdom. J. Vet. Med. Sci. 8, 789-796. Vaisanen, M.A.M., Valros, A.E., Hakaoja, E., Raekallio, M.R., Vainio, O.M., 2005. Pre-operative stress in dogs: a preliminary investigation of behavior and heart rate variability in healthy hospitalized dogs. Vet. Anaesth. Analg. 32, 158-167. 26

539 540 van den Berg, L., Schilder, M.B.H., de Vries, H., Leegwater, P.A.J., van Oost, B.A., 2006. Phenotyping of aggressive behavior in golden retriever dogs with a questionnaire. Behav. Genet. 36, 882 902. 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 Voith, V.L., Borchelt, P.L., 1996. Separation anxiety in dogs. In: Voith, V.L., Borchelt, P.L. (Eds.), Readings in Companion Animal Behaviour. Veterinary Learning Systems, pp. 2124 2139. Voith, V.L., McCrave, E.A., Marder, A.R., Lung, N., 1997. Environmental and behavioural profiles of dogs with separation anxiety. In: Proceedings of the First International Conference on Veterinary Behavioural Medicine, Birmingham, UK, p. 241. Voith, V.L., Wright, J.C., Danneman, P., 1992. Is there a relationship between canine behavior problems and spoiling activities, anthropomorphism and obedience training. Appl. Anim.Behav. Sci. 34, 263 272. Wacholder S., Hartge P., Lubin J.H., Dosemeci M., 1995. Non-differential misclassification and bias towards the null: a clarification. Occup. Environ. Med. 52: 557-558. Wells, D.L., 1996. The welfare of dogs in an animal rescue shelter. PhD thesis. School of Psychology, The Queen s University of Belfast, UK. Wells, D.L., Hepper, P.G., 2000. Prevalence of behavior problems reported by owners of dogs purchased from an animal rescue shelter. Appl. Anim. Behav. Sci. 69, 55 65. Wright, J.C., 1991. Canine aggression toward people. Bite scenarios and prevention. Vet. Clin. Small Anim. 21, 299-314. 27