UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Household pets and depression among urban adolescents Nevin-Haas, Mary M. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-ubc_1993_fall_nevin_haas_mary.pdf [ 4.33MB ]
JSON: 831-1.0086271.json
JSON-LD: 831-1.0086271-ld.json
RDF/XML (Pretty): 831-1.0086271-rdf.xml
RDF/JSON: 831-1.0086271-rdf.json
Turtle: 831-1.0086271-turtle.txt
N-Triples: 831-1.0086271-rdf-ntriples.txt
Original Record: 831-1.0086271-source.json
Full Text

Full Text

HOUSEHOLD PETS AND DEPRESSION AMONG URBAN ADOLESCENTSbyMARY MARGARET (MOLLY) NEVIN-HAASB.S.N., The University of British Columbia, 1987A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCE IN NURSINGinTHE FACULTY OF GRADUATE STUDIES(School of Nursing)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAOctober 1993© Mary Margaret (Molly) Nevin-Haas, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(SignatureDepartment of  Graduate Studies, School of NursingThe University of British ColumbiaVancouver, CanadaDate ^13 0c,T,c;>-b-e.■_,, 1q93 DE-6 (2/88)IIAbstractThis study was designed to describe the prevalence of depression amonga group of adolescents and examine the relationships between depression andpresence of pets in the home, primary ownership of pets and perceived degree ofattachment to the pets by the subjects. The coping conceptual framework ofFolkman and Lazarus (1988) provided the structure for the study. Subjects were401 grade nine males and females attending three Vancouver and three Victoriasecondary schools. The instruments utilized to gather the data were the Centerfor Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977) and theAdolescent and Pet Characteristics Questionnaire, developed by the researcher.The subjects completed the instruments anonymously in classroom settings.Among the grade nine students sampled, 48.4% demonstrated nodepressive symptoms, 41.4% demonstrated what were classified as mildsymptoms of depression, 8.7% demonstrated moderate symptoms of depression,and 1.5% of those sampled indicated severe depressive symptoms. Femaleadolescents were significantly more depressed than their male counterparts withmean depression scores being 18.7 and 15.1, respectively.Those subjects with pets in the home were significantly less depressedthan their counterparts without household pets. There was no relationshipbetween gender and pet ownership combined with level of depression. Ananalysis of primary pet ownership indicated that subjects who owned their ownpets were significantly more depressed on a whole than those in homes wherethe pet was owned by the entire family. Perceived degree of attachment to thehousehold pet was not related to depression among the subjects. Conclusionsare drawn from the findings and implications for nursing practice and researchare discussed.IIITable of ContentsAbstract^ iiTable of Contents^ iiiList of Tables viList of Figures^ viiAcknowledgements viiiChapter One: IntroductionBackground to the Problem^ 1Problem Statement^ 3Purpose^ 4Research Questions^ 4Conceptual Framework 5Definition of Terms^ 7Assumptions 8Limitations^ 8Significance of the Study^ 9Organization of Thesis 10Chapter Two: Review of the LiteratureAdolescent Depression^ 11Pets and Health in Humans 14Pets and Physical Indices of Stress in Humans^ 17Pets and Emotional Health in Humans^ 20Summary^ 29Chapter Three: MethodsInstruments^ 31Adolescent and Pet Characteristics Questionnaire^ 31Center for Epidemiologic Studies Depression Scale 32IVProtection of Human Rights^ 34Sample^ 35Data Collection^ 35Data Analysis 37Chapter Four: Presentation and Discussion of FindingsResponse Rate^ 41Demographic Characteristics^ 42Gender^ 42Age 43School Performance^ 43Family Composition 44Country of Origin^ 46Primary Language 48Residence^ 50Pet Demographic Information^ 52Presence of Household Pets 52Types of Pets^ 52Identified Favorite Pet^ 52Pet Names^ 53Duration of Pet Ownership^ 55Reason for Acquiring Pet 56Primary Ownership of Pet^ 56Pet Attachment^ 57Findings Related to Research Questions^ 59Question One: Prevalence of Depression 59Question Two: Relationship Between Gender, Pet Presence andDepression^ 61VQuestion Three: Relationship Between Pet Ownership and Depression 64Question Four: Relationship Between Pet Attachment and Depression^67Summary^ 68Chapter Five: Summary, Conclusions, and Implications for NursingSummary^ 70Conclusions 72Implications for Nursing^ 72Nursing Practice 73Nursing Research^ 76References^ 78AppendicesA. Adolescent and Pet Characteristics Questionnaire^89B. Center for Epidemiologic Studies Depression Scale 91C. Letter of School Consent and Consent Form^ 93D. Letter of Parental Informed Consent and Consent Form^96E. Information and Instructions for Questionnaire^ 99F. Density Distribution of CES-D Scores^ 101VIList of TablesTable^ Page1. Ages of Subjects^ 432. School Performance of Subjects^ 433. Parents Living at Home^ 444. Siblings Living at Home 455. Sibship Position of Subjects^ 466. Regions and Countries of Origin of Subjects and Parents^477. Subjects and Parental Birth Regions^ 488. Primary Languages Used at Home 499. Years of Residence in Canada and Current Neighbourhood^5110. Favorite Pet Categories and Types^ 5311. Pet Name Categories and Examples 5412. Categories of CES-D Scores^ 6013. ANOVA Summary Table for Gender, Pets, and CES-D Scores (N=401) 62List of FiguresFigure Page1. The Coping Conceptual Framework: Pets as Mediators of Emotion inAdolescent Depression (adapted from Folkman and Lazarus, 1988). 62. Numbers and Percentages of Subjects by School 423. Pet Ownership 574. Perceived Attachment to Pet 58VIIVIIIAcknowledgementsThis thesis could never have been realized in its current form were it notfor the kindness, assistance, and encouragement I received from many specialindividuals. Most importantly I would like to thank my thesis supervisors Dr.Marilyn Willman and Gloria Joachim. Dr. Willman accepted the position of chairof this committee in her final year as Professor and Director of the School ofNursing, her dedication and input to this project could not be overemphasized.Gloria Joachim has been with the project since its conception, herencouragement and clarity of vision often kept things going though obstaclessuch as unfavourable media coverage of school surveys and school boardstrikes. These individuals gave me the rare and valued combination of freedomof content, and direction in structure. I would also like to thank and acknowledgeDr. Anna Marie Hughes who offered valuable critique in the review of this work;as well as Janet Ericksen and Dr. Ann Hilton for their assistance in the proposalstage of this project.With regard to the long and frequently frustrating process of findingsubjects, Court Brousson, Nicole Chagnon, Eileen Eby, Andree Johansson,Gordon May, Ian McEwen, Patrick Mitchell, Rhonda Morrisson, Stephanie Nevin,Jack Nickolichuk, Gary Puder, Janice Phillips-Sim, Lee Simpson, Sharon Reid,Joyce Tinnion, Frank VanSoldt, and Sheila Wareing all deserve recognition fortheir assistance in the recruitment of subjects. Special acknowledgement isgratefully extended to Janice Phillips-Sim and Frank VanSoldt, whose extraefforts in promoting this study with their students were instrumental in achievingthe number of subjects for this study. I would also like to thank and acknowledgethe parents of the subjects for granting their consent for their children'sparticipation as well as the students who took the time and effort to transportconsent forms home and fill out the instruments for this study.IXStatistical methods and analyses of this project were patiently andgratefully clarified by Dana Atagi and Gordon Haas.Support and encouragement was also always afforded me by myemployers and supervisors throughout these years. For recognition of myresearch needs during these times I would like to thank Marsha Ablowitz, RheaArcand, Ralph Buckley, Lee Simpson, Sally Thorne, and Marilyn Willman.This work was specifically encouraged by Gordon Haas, Louisa Nicholls,and organizations such as the Human Animal Bond Association of Canada andthe Delta Society of the United States. It was inspired by past and presentanimals who have impacted my view of the world. For this, I would like to extenda unique thank you to Abigail, Amos, Baxter, Franklin, Gremlin, Greta, Oscar,Taco, and Wrangler. Finally, none of this work could have been enjoyable, muchless possible, without the continuous support, respect and encouragement Ireceived from my family, and at home from my husband Gordon Haas and ourown set of animal companions (Baxter, Franklin, and Oscar).1Chapter OneIntroductionBackground to the ProblemFossil remains indicate that the domestic relationship between humansand animals dates back approximately 12,000 years to the preagricultural periodof human history just following the end of the last global ice age (Davis & Valla,1978). Although there is controversy as to the motivation behind the originalformation of the relationship (whether it was one of companionship or one ofutilitarian purpose), there is general consensus that this relationship haspersisted throughout history and is strongly prevalent across a variety of agegroups and human cultures today (Bustad & Hines, 1984; Messant & Serpell,1980; Odendaal & Weyers, 1990; Ritvo, 1988; Serpell, 1988).Levinson (1962, 1965, 1970, 1972) was one of the first individuals to studythe nature of the relationship between humans and animals. He addressed thehealth value of human contact with the natural environment in an increasinglyindustrialized society. "When man is forced to live and work deprived of contactwith nature, he loses much emotional strength" (Levinson, 1972, p.6). Hefurthered stated that encounters between people and animals are a form ofcontact with nature that may influence emotional strength of humans:We need animals as allies to reinforce our inner selves. Wemust revive our intimate associations with nature and itsanimals if we are to survive as the dominant species onearth. It is of course possible that man can survive withoutanimals, but we would surely be a depleted race, shorn ofmost of our emotional strength (1972, p.29).Impairment in the emotional strength of individuals is a symptom which isfrequently associated the clinical disorder of depression. Depression has been2demonstrated to be more prevalent in urban as opposed to rural settings(Klerman & Weissman, 1989). Frequently called the "common cold of mentaldisorders" (O'Hara, 1984, p.46), depression rates among adults in urban settingshave been reported to be as high as 20% (Klerman & Weissman, 1989).The etiology of depression in adults is believed to be twofold. Bothphysiological genetic make-up as well as psychological response to losses andstressors contribute to its development (Klerman & Weissman, 1989; Lapierre,1988). In keeping with this bivariate foundation, the treatment of depression inadult populations has been shown to be equally effective with the use ofantidepressant medications or cognitive therapy (Beck, Holton, Young,Bedrosian, & Budenz, 1985; Murphy, Simons, Wetzel, & Lustman, 1984).Depression among adolescents is not as well understood as that amongadults. The reported prevalence of severe depression in the former group variesfrom 3% (Clarke, Lewinsohn & Hopps, 1990) to 16% (Pronovost, Cote & Ross,1990). In some studies, the use of antidepressant medications has been foundto be no more effective than placebos in treating adolescent depression (Kramer& Feiguine, 1981; MacLean, 1987; Puig-Antich, et al. 1987; Simeon, 1989).Such studies suggest that depression among adolescents may be morecommonly a result of psychological factors than of biological determinants.Despite the vague understanding of adolescent depression, it is well-documentedthat adolescents with depression are at increased risk for dropping out of school,substance abuse, acts of delinquency, non-consensual sexual activity, andsuicide than adolescents without depressive symptoms (Clarke, 1991;McDermott et al. 1990; Reinherz, Frost, & Pakiz, 1991; Simeon, 1989).Both of the aforementioned concepts of a human-animal relationship anddepression among adolescents have implications for nursing. Most models ofnursing include individuals' responses to their environment as a factor influencing3health (Riehl & Roy, 1980). Domestic animals in the form of household pets arean integral part of the environment of many individuals. Therefore, the impactpets may have on the well-being of their owners becomes a concern for allphases of the nursing process (McMahon, 1991). It remains to be determinedwhether the sole presence of a pet in the home may constitute the beneficialbonding referred to by Levinson (1961, 1972) or if, in fact, there are other factorswhich determine whether humans receive health benefits from pets in theirhomes (Johnson, Garrity, & Stallones, 1992).The second concept, depression among the adolescent population, alsohas implications for nursing. Nurses care for individuals across the lifespan. Thecare provided by nurses encompasses biological, psychological and socialcomponents of each individual's health (Johnson, 1980). The mental health ofadolescents is a specialized health focus of a population of individuals within therealm of nursing practice (Canadian Nurses Association, 1991).Problem StatementIt has been postulated that the alienation of humans in an urban settingfrom their natural environment may be related to the increased prevalence ofdepression (Klerman & Weissman, 1989) as well as an inability to recover quicklyfrom stressful events (Ulrich, Simons, Losito, Fiorito, Miles, & Zelson, 1991).Additionally, it has been noted that human contact with animals may beemotionally strengthening (Levinson, 1972). The presence of pets can potentiallyact as a link between people living in an urban setting and their naturalenvironment (Katcher & Beck, 1988; Phineas, 1974).The literature notes the effects of pets on the emotional well-being ofseveral different populations; however, there have been no studies whichdescribe the relationship between the presence of household pets and4prevalence of depression among urban adolescents. If there is such arelationship, and hence the possibility of altering depression and associatedfactors such as suicide, delinquency, and substance abuse, it behooves nursingto explore this potential. The identification of the possible implications of thecenturies old relationship between humans and animals for the mental health andfuture of our youth has significance for nursing knowledge, practice, andimproved health care.PurposeThe purpose of this study is to describe the prevalence of depression in asample of grade nine adolescents from two large urban areas and to examine therelationships between prevalence of depression, presence of pets in the home,primary ownership of the pets, and perceived degree of attachment to the pets bythe subjects involved.Research QuestionsThis study proposes to answer the following research questions:1. What is the prevalence of depression in a sample of urban adolescentmales and females currently attending grade nine in selected Vancouver andVictoria secondary schools?2. What is the relationship between the prevalence of depression and thepresence of pets in the homes of urban adolescent males and females currentlyattending grade nine in selected Vancouver and Victoria secondary schools?3.^What is the relationship between prevalence of depression and primaryownership of household pets among urban adolescents currently attending gradenine in selected Vancouver and Victoria secondary schools?54.^What is the relationship between prevalence of depression and perceiveddegree of attachment to pets among urban adolescents currently attending gradenine in selected Vancouver and Victoria secondary schools?Conceptual FrameworkThe theoretical framework of coping as proposed by Folkman & Lazarus(1988) is utilized in the conceptualization of this study (see Figure 1). Thisframework values an open and interactive system between people and theirenvironment and views coping as a multidimensional process inherent within thissystem. It describes coping as a constantly changing response to person-environment encounters and emotional appraisals. The use of such a copingframework is consistent with that used in other studies of human-animalinteractions (Carmack, 1988).The bidirectional relationship between emotion and coping is an importantaspect of the framework. The individual's appraisal of an event is related to anemotional response, which is related to a mediating method of coping. Thiscoping behaviour, in turn, leads to an emotional reappraisal of the same event oran alternate occurrence. Folkman and Lazarus (1988) stress the importance ofthe "mediating" act of coping. A mediating variable is a factor which changes therelationship between the antecedent and the outcome variable. Mediatingvariables are created within the context of the encounter. The response of"coping" is such a mediator. Coping is formulated during the encounter andchanges the original appraisal and its accompanying emotion in some way.Such a framework correlates well with the conceptualization of this study.The adolescent assumes the position of the individual in the model, with theperceived or actual stressors and/or losses representing the environmentalEMOTION(Study Measurement)6PERSON / ENVIRONMENT ENCOUNTERAdolescent and Etiologic Factorsof DepressionCOPING (Use of Pets)Problem Focussed^Emotion FocussedNew Person -EnvironmentEncounterFigure 1. The Coping Conceptual Framework: Pets as Mediators of Emotionin Adolescent Depression (adapted from Folkman and Lazarus. 1988). 7encounter(s). (The precise etiology of depression cannot be limited to oneoccurrence alone.) The preliminary emotional response to the appraisal of thesestressors may then be represented by depression. The presence of householdpets may act as a resource in the formation of mediating coping responses whichmay then interact with the depressive response to cause a reappraisal of theetiologic factor or the initial emotion and provide an alternate emotional response.Folkman & Lazarus's (1988) model is used in discovering the nature of therelationship between depression experienced by urban adolescents, pets as acoping resource, and an altered form of the original depressed emotionalresponse. As Figure 1 indicates, this model proposes a feedback loop. The useof mediating coping resources influences emotional responses by providing analternate appraisal of events than that which was obtained prior to the use of thecoping resources. The use of this mediating coping resource will then be utilizedby the individual on an ongoing basis in the face of similar emotions, therebyreplacing the original emotion with a more suitable emotional response.Definition of TermsFor purposes of this study, the following terms will be defined as statedbelow:Adolescent: Any youth between the ages of 13 and 16 currently attendinggrade nine in selected Vancouver and Victoria secondary schools;Attachment: An individual's subjective report of emotional closeness or affectiontowards an animal in her / his home, as measured by the Adolescent andPet Characteristics Questionnaire (Appendix A);Depression: An emotional state characterized by low mood, apathy, anhedonia,and lack of energy, as measured by the Center for Epidemiologic StudiesDepression Scale (CES-D) (Appendix B);8Pet: Any non-human, non-utilitarian animal which is currently living in the samehousehold as the subject and is cared for by some member of thathousehold;Primary Ownership: An acknowledgement by the subject as to whichhousehold member has recognized possession of the pet. This generallyimplies responsibility for pet care-giving;Urban: A city and surrounding geographic area characterized by its populationdensity and non-resource based nature of employment.AssumptionsThe study is based on two primary assumptions. The first assumption isthat those adolescents who are depressed at the time of the survey will have theconcentration and motivation to complete the instruments in their entirety. Thesecond assumption is that all the adolescents surveyed will respond truthfully tothe questions on the instruments.LimitationsThe limitations inherent within this study involve both design andenvironmental factors. The findings of the study are generalizable only tosubjects similar to those of the population sampled. The survey design accountsonly for contact that the adolescents may have with the pets that currently livewithin their households. The subjects were sampled over a four-month periodfrom October 1992 to January 1993. The study followed the implementation of asurvey (McCreary, 1993a, 1993b) that raised some parental objections (Lamb,1992a, 1992b; Odam, 1992; Truscott, 1992). It is possible that the negativepress from the previous survey may have had an impact on subjects' participationin this study.9Significance of the StudyThere is currently a lack of consistent information about the prevalence ofdepression in the general population of urban adolescents. Recently, muchattention has been paid to adolescent mental health because of the risingincidence of suicide (a potential consequence of adolescent depression) in thisage group (Children and Youth At Risk Steering Committee, 1991; Clarke, 1991;Conrad, 1991; McDermott et al., 1990; Pronovost et al., 1990;). This study willprovide information about the prevalence of adolescent depression in selectedVancouver and Victoria secondary schools. Also, there is a paucity of researchvalidating the benefits of pets on individuals' mental health as described in the layliterature (Robb & Stegman, 1983). This study will add to the body of knowledgerelated to the role which pets may play in adolescent mental health.The above is important for nurses who find themselves in contact withadolescent clients or families with adolescents. By understanding the prevalenceof depression in this population and its descriptive parameters, nurses canincrease their knowledge and ability to provide early detection and promptintervention for those individuals at risk for the more serious consequences ofthese pathological "blues" (McDermott et al., 1990; Offer & Schonert-Reichl,1992; Reinherz, Frost, & Pakiz, 1991). Further, by analyzing the nature of therelationship between prevalence of depression and pets in the homes of urbanadolescents, differences between the depressed and the mentally healthyadolescent may be identified. The results of this study may assist the practicingnurse in determining whether to include an assessment of pet ownership andinteraction (Davis, 1985a, 1985b) when analyzing the coping mechanisms ofyouth. As Levinson (1972) so clearly stated two decades ago:I am sure that the majority of my professional colleagueswould agree that our precious twentieth century youngsterswill probably mature in a chaotic, disturbed environment. Ifwe have a therapeutic tool such as the use of pets as amental hygiene adjunct, it behooves us to explore anddevelop fully the possibilities of this adjunct. If we do not, weas adults must take full responsibility for the consequencesof our neglect (p.11 ).Organization of ThesisChapter One has provided an introduction to the nature of the study,including the background to the problem, the purpose of the study, the researchquestions, conceptual framework, assumptions, limitations, and significance ofthe study. Chapter Two outlines the theoretical basis of the study with a reviewof selected literature pertaining to the two central themes of the study, namelydepression and the mental health benefits of pets. The research methods usedin the study are described in Chapter Three. In Chapter Four, the findings arepresented and discussed. Finally, Chapter Five presents the summary,conclusions, and a discussion of the implications for nursing practice, education,and research.1011Chapter TwoReview of the LiteratureThe literature review examines the current state of knowledge regardingadolescent depression as well as the nature of human and animal relationshipsand the potential mental health benefits of pets.Adolescent DepressionOnly within the past decade has increased attention been given to thedistinction between normal adolescent frustration or moodiness and itspathological variant, adolescent depression (Greydanus, 1986; MacLean, 1987;McDermott et al., 1990; Offer & Schonert-Reichl, 1992; Reinherz et al., 1991;Simeon, 1989). This recent interest may be largely attributed to the rapidlyincreasing numbers of adolescent suicides and a desperate societal attempt tounderstand some of the phenomena contributing to suicide, the second leadingcause of death among North American youth (Conrad, 1991). In the past twentyyears, the suicide rate for adolescents between the ages of 15 and 19 has risen44%, compared with a 2.6% increase for the general population (Conrad, 1991).In Canada, the suicide rate for 15 to 19 year old adolescents doubled in tenyears. For this age group, the suicide rate is 17.6 deaths per 100,000representing one out of five deaths (Pronovost et al., 1990).Depression has been correlated with suicide as well as with otherundesirable outcomes and behaviours among youth. Studies have indicated that40 to 60% of suicidal adolescents have feelings of depression (Pronovost et al.,1990). However, not all suicidal individuals are depressed nor are depressedindividuals necessarily suicidal (Greydanus, 1986; McDermott et al., 1990).Other undesirable outcome correlates of adolescent depression include anincreased risk for dropping out of school, substance abuse, non-consensual12sexual activity, and acts of delinquency (Clarke, 1991; McDermott et al., 1990;Reinherz et al., 1991; Rogers, 1987). Longitudinal studies are currentlyunderway to test these relationships (Weissman et al., 1987).Reinherz et al. (1991) studied a group of 386 adolescents from theirfreshman to their senior year in high school. They found a significant correlationbetween drop-out rates and depressed states in adolescent males over all othergroups. For adolescent females who were also depressed, substance abuserates were significantly higher than for non-depressed females or males of eithergroup. Thirty-three percent of these females reported "being high on marijuanaall day and in school" (p. 61). Depressed females also began drinking alcoholone year earlier (age 14) than did all other adolescents surveyed. Although thestability of depressive symptoms from grade 8 to grade 12 was only moderate(60%), individuals who were depressed in grade 8 were 9.3 times more likelythan their non-depressed peers to also be depressed in grade 12 (Reinherz etal., 1991). In addition to these behavioural concerns, poor academicperformance has also been frequently correlated with a high incidence ofdepression among adolescents (Baron & Perron, 1986; Blechman, McEnroe,Carella, & Audette, 1986; Reinherz et al., 1991; Weissman et al., 1987).McDermott et al. (1990) surveyed a random sample of 219 grade elevenand grade twelve students from a south central United States secondary school.The Center for Epidemiologic Studies Depression Scale (CES-D) was used tomeasure depression symptoms. Scores on the CES-D were then correlated witha variety of reported health practices. The investigators determined that suicidalideation was highly correlated with depression and was reported by 15.2% of therespondents. Health behaviours which were moderately correlated withdepression included using recreational drugs or alcohol, overeating, andengaging in non-consensual sexual activity. Health behaviours demonstrating a13weak correlation to adolescent depression were abstaining from physicalexercise and missing classes.Presently, scientific knowledge in regard to adolescent depressionremains at the descriptive level. The prevalence of depression reported for thegeneral adolescent population currently ranges from 3% (Clarke, 1991) to 16%(Pronovost et al., 1990). There is no consensus as to the clinical frequency ofadolescent depression (Siegel & Griffin, 1984). It is generally agreed that theprimary etiology of adolescent depression may be either endogenous or reactive(Greydanus, 1986; MacLean, 1987; Simeon, 1989) with a greater proportionbeing of the reactive nature than in the comparable adult population (Kramer &Feiguine, 1981; Puig-Antich et al., 1987; Simeon, 1989).Just as there have been conflicting findings about the prevalence ofadolescent depression, so too have there been significant discrepanciesregarding the demographic characteristics of depressed youth. Baron & Perron(1986), McDermott et al. (1990), Reinherz et al. (1991), Simeon (1989) andWeissman et al. (1987) have demonstrated significantly increased rates ofdepression among adolescent females over males. However, in a similarlydesigned study, using the same or a comparable diagnostic tool, Siegel & Griffin(1984) failed to demonstrate any gender differences.Baron & Perron (1986) examined the family setting and demonstrated nosignificant difference in level of depression and marital status of biologicalparents, type of family structure, employment of parents, and presence ofsiblings. Utilizing the same instrument and level of significance, Siegel & Griffin(1984) found a significantly higher level of depression among children of divorcedparents. As Baron and Perron's (1986) study involved a sample from westernQuebec, Canada and Siegel and Griffin's (1984) study was composed of asample from Missouri, U.S.A., this variation may be due to the differing value14each population places on the traditional nuclear family. Weissman et al. (1987)and Simeon (1989) demonstrated an increased prevalence of depression amongchildren and adolescents whose parents were depressed as compared to thosewho had non-depressed parents.Age and socioeconomic status have been found to be positively correlatedwith depression in some studies (Greydanus, 1986; Siegel & Griffin, 1984) andnegatively correlated in others (Baron & Perron, 1986; Reinherz et al., 1991).Cultural and other demographic differences have been found to exist amongurban and rural adolescents who were suicidal (Tonkin, 1984). No studies haveisolated similar factors among adolescents experiencing depression.Pets and Health in HumansSimilar to the research related to adolescent depression, the literature onpets and human health has increased greatly in the past several years. Anannotated bibliography of the human-animal bond lists 819 citations prior to 1983(Allen, 1985), and another lists 350 citations between 1987 and 1990 (Rowan,1992a). In 1987, Anthrozoos commenced publication as a multidisciplinaryrefereed journal concerned with the interaction of people, animals, andenvironment. This publication was joined in 1993 by Society and Animals, asimilar refereed journal committed to the scientific study of human and animalinteractions. A number of international organizations are devoted to the study ofhuman and animal interactions. Thirteen of these came together under theInternational Association of Human-Animal Interaction Organization (I.A.H.A.1.0.)for a conference in 1992 in Montreal, Canada.There are several levels of rigor and complexity in publications regardinghuman and animal interactions in a shared environment. At the most basic level,articles in the media and lay literature extoll the benefits of pets for one's mental15health in decreasing feelings of loneliness and depression and increasingfeelings of love and belongingness (Cohen, 1991; Humeston, 1983; Jaworski,1990; Meer, 1984; Schultz, 1990; Toufexis, 1987; Wyatt, 1988). These articleshave little scientific merit in most cases but may provide the impetus for empiricalexamination of popularly held beliefs.The next level of literature includes anecdotal case reports which aregenerally found in non-research-based journals of nursing or other healthprofessions. It is in this forum that "helpful hints" for improving patient care arepassed from one professional to another. Such references, like the lay literature,frequently cite the profoundly positive effects pets have had on the subjectivemeasurements of happiness and social behavior of various groups of individualsrequiring nursing care (Bibby & Posterski, 1992; Bikales, 1975; Blythe, 1980;Bossard, 1944; Brickel, 1979, 1980; Carbary, 1975; Davis, 1985a, 1985b; Davis,1986; Dolan, 1982; Fogel, 1983; Francis, 1981, 1991; Furstenberg, Rhodes, &Powell, 1988; Gammonley & Yates, 1991; Haggard, 1985; Lago & Knight, 1987;Manor, 1991; Manor, McCurdy, & Crain, 1988; Mayhew, 1988; Preshlock, 1985;Ruckert, 1987; Twiname, 1984; Wille, 1984).One well-known piece of anecdotal pet literature in nursing came fromFlorence Nightingale who recognized the effect that the pet owl she carried in herpocket had on her patients. She wrote, "A small pet is often an excellentcompanion for the sick..." (Nightingale, 1860/1967, p. 103). Like the popular layliterature, attention is best paid to these sources for their inquiry into observedclinical effects. It is from such sources that the nurse researcher may begin toplan a course of study to actively test some of the "hunches" held by practicingcolleagues.The remaining research-based literature on the use of animals in thehealth promotion of humans has been split into three distinct groups. The first16category is that examining the use of "helping animals" for the physicallydisabled. However, this study examines the relationship of pets with mentalhealth. Therefore, unless the emotional benefits of such a relationship are alsostudied, this body of literature is not addressed in the context of this review.The second category of pet research literature does measure theemotional or mental effects of pets on humans in pet therapy, or pet-facilitatedtherapy. This involves the use of an "outside" animal (which is usually thetherapist's but is almost always unfamiliar to the client) in therapeutic interactionsbetween a client and a health professional (Corson, 1977; Draper, Gerber, &Layng, 1990; Hundley, 1991; Levinson, 1962, 1965; Wolfe, 1977). The use ofanimal visitation programs with hospitalized or institutionalized individuals isconsidered under the category of pet therapy as it involves the use of an"unknown" animal (Fila, 1991: Francis, Turner, & Johnson, 1985; Kale, 1992;Kalfon, 1991; Lebeck, 1992; Lee, 1987; Mead, 1992; Moneymaker & Strimple,1991; Muschel, 1984; Ormerod, 1992; Robb, Boyd, & Pristash, 1980; Thompson,Kennedy, & Igou, 1983). This aspect of the literature is also not discussed in thereview as it does not directly apply to the proposed study.The third category of research-based literature discusses the nature of therelationship between individuals and their own pets. This literature occasionallyrefers to pets as "companion animals" to highlight the fact that it refers to ananimal well known to the subject, and not an external animal such as those usedin pet-facilitated therapy. In fact, there has been little clarification of theterminology differences between "pet" and "companion animal" in the literature(Veevers, 1992). In this study, the term "pets" refers to both "companionanimals" and "pets" as used in the literature. A definition of pets is included inChapter One.17The majority of the pet studies to date have been qualitative in design,examining the nature of pet ownership and/or interaction with a particular healthparameter of a given population. Like the aforementioned state of knowledgeregarding adolescent depression, the current state of knowledge in the study ofhuman health factors and pets is still in the descriptive stage.Pets and Physical Indices of Stress in HumansStress, one of the possible contributors to depression (Klerman &Weissman, 1989), is frequently demonstrated by an increase in blood pressure(Perko & Kreigh, 1988). In a within-subject experimental design study, Baun,Bergstrom, Langston, and Thoma (1984) examined the effect of pets on stress.This study demonstrated that stroking a pet dog (with whom the subject had anestablished bond) had a significant effect in lowering both the systolic anddiastolic blood pressure when compared to stroking an unfamiliar dog. Thedecreased blood pressure response of those subjects stroking a pet dogparalleled the relaxation response demonstrated by those asked to read quietlyalone. Unfortunately, there was no true control group in the latter study.A thesis completed by Oetting (cited in Baun, Oetting and Bergstrom,1991) examined the physiological indices of stress reduction (blood pressure,heart rate, and peripheral skin temperature) for groups solely stroking a pet dogor stroking a dog in combination with autogenic relaxation techniques. Althoughboth groups demonstrated a decrease in stress levels as measured by the aboveparameters, there were no significant differences in the stress reduction indicesof the two groups. The two aforementioned studies illustrate the differing findingsin studies of the relationship between people and the effects of pets on certainparameters (such as the physical indicators of stress) of mental health.18Blood pressure was also a variable used by Riddick (1985) in a pre-test,post-test, control group design that provided fish aquariums to non-institutionalized elderly. Significant decreases in diastolic blood pressure, andincreases in leisure satisfaction and in relaxation states were noted in themembers of the experimental group. It is unclear, however, if these results weredue to the actual nature of human-animal interactions or were an outcome of anew leisure hobby.Allen, Blascovich, Tomaka and Kelsey (1991) measured the physicalautonomic stress responses of 45 adult women asked to respond tomathematical skill testing questions (as a stimulus of stress) in a laboratorysetting with only the researchers present. These women were then asked torepeat the exercise two weeks later at home in the company of either their bestfemale friend (social support), their pet dog, or alone. The results wereconsistent across laboratory and field studies. The women with a friend presentat the time of testing scored less well and demonstrated significantly increasedautonomic stress responses. Those women with their dog present at the time oftesting demonstrated significantly less physiological reactivity on all autonomicstress measurements. The marked difference between support provided byfriends and that of pets was interpreted by the researchers as a result of the non-evaluative nature of pets as they relate to humans (Allen et al., 1991).Another study of physiological indices of relaxation and stress reduction inregard to the presence of animals is that of Friedmann, Katcher, Lynch andThomas (1980). These researchers followed 92 patients post-discharge from acoronary care unit. Findings demonstrated a significant positive relationshipbetween the presence of household pets and a one-year survival rate. Tocontrol for the physical requirements (and hence possible adherence torecommended exercise regimes) necessary in caring for a dog, all dog owners19were then omitted. The findings of the second analysis also demonstrated asignificant relationship between the presence of non-dog household pets and theone-year survival rate post-discharge from the coronary care unit. These findingswere consistent across age, gender, severity of illness, marital status, andpersonality traits of the subjects. Such findings again suggest the notable effecthousehold pets have on the health of their human companions.In a study examining the risk factors related to stress and cardiovasculardisease, Anderson, Reid, and Jennings (1992) studied 5,741 individualsattending a free clinic in Melbourne, Australia. Two matched groups wereestablished according to age and sex. Potentially confounding variables of diet,exercise, weight, socioeconomic status, alcohol intake, and smoking habitsappeared to be equally distributed throughout both the pet owning and non-petowning groups of all ages and genders. All the male pet owners demonstratedsignificantly lower plasma triglyceride levels, lower plasma cholesterol levels, andlower systolic blood pressure than did their non-pet owning counterparts. Thefemale pet owners in the 40 to 59 age group demonstrated significantly lowerplasma triglyceride levels and lower systolic blood pressures than did non-petowners in the same group. For all groups, the differences were independent ofthe type of pet owned. This is one of the first studies demonstrating a directcorrelation between pet ownership and physiological indices of stress related tospecific disease prevention.Serpell (1990, 1991) carried out a ten-month prospective study whichexamined changes in health status and behaviour in 71 adult subjects followingthe acquisition of a new pet (dog or cat). The pet owning groups reported asignificant reduction in minor health problems in the first month following petacquisition as compared to a control group of non-pet owners. For dog owners,the health status improvement was sustained for the entire ten-month study2 0period. Similarly, new pet owners also demonstrated a significant improvementin psychological well-being over the first six months. For dog owners, thisincluded an increased sense of self-esteem which was sustained for the entireten month study period. Serpell's (1990, 1991) findings suggest that acquisitionof a pet may have a causal influence on overall human physical andpsychological health and in some cases these effects may be relatively longlasting.Pets and Emotional Health in HumansThe research literature regarding the relationship between pet ownershipand the emotional well-being of the owners has been studied in a variety of wayson a vast number of populations. Carmack (1991) studied the lived experience of18 individuals with acquired immunodeficiency syndrome (AIDS) and the role of apet in their lives. Identified themes that persons with AIDS attributed to thepresence of their pets included an increased feeling of affection, feeling valuedand needed, an ability to have a confidant, a catalyst for facilitating discussion ofdifficult topics, an increased ability to focus on the present, a feeling of consistentsupport, and an increased ability to handle anger, as well as a decreasedperception of loneliness and stress.It becomes apparent that the value which persons with AIDS placed on thepresence of pets may, in fact, also be present among pet owners without AIDS.According to Folkman and Lazarus's (1988) model, several of these attributescould be included with the presence of pets as part of the coping response of anindividual experiencing depression. The use of pets (with the identified attributescited by Carmack, 1991) may enable individuals experiencing depression toreappraise that emotion into one of an alternate and perhaps more manageablenature.21Blenner (1991) interviewed 25 infertile couples in a qualitative study todetermine the effects of pets on this population. She concluded that using petsas therapeutic adjuncts greatly helped to break the cycle of depression,withdrawal, and stress or isolation in infertile patients' lives.The effects of pets on depressed individuals has been examined byMcCulloch (1981) who surveyed 31 individuals (ages 22 to 69) with a variety ofchronic physical health concerns and concurrent depression in order todetermine the role of their pets during the course of their physical illness. Themajority (71%) of the subjects indicated that their pets were an important sourceof companionship, as well as a valuable force in helping them cope with feelingsof loneliness (83%). A larger majority (94%) of McCulloch's (1981) samplestressed the importance of the pet in encouraging a sense of humour. Evenamong those in the sample who demonstrated social stability and satisfactorysupport systems, pets were perceived to provide invaluable additional copingresources to help the individuals deal with their physical illness as well as theconcurrent depression (McCulloch, 1981).Mugford and M'Comisky (1975) utilized a pre-test, post-test, control groupexperimental study design in which a sample of well elderly who were living alonewere given either begonias, caged birds, or nothing. The results demonstratedsignificant improvement in self-concept and attitude toward others for the groupwho received the birds for the five-month test period over those who received theplants or no item.In contrast, a more recent study by Robb and Stegman (1983) found nosignificant difference in measures of morale, locus of control, social interaction,and mental status among predetermined groups of pet owning and non-petowning veteran clients receiving home care services. Utilizing instruments withestablished reliability and validity, this study was one of the few to have2 2quantitatively measured the findings frequently cited in anecdotal reports. Robband Stegman (1983) recommended that further quantitative studies of factors inhuman-animal interactions be carried out utilizing established instruments. Inaddition, the researchers recommended that critical attention be paid tosituational and personal characteristics of the subjects.Baun, Cardiello, and Jassen (1992) utilized a pre-test, post-test controlgroup design in the measurement of depression, loneliness, and morale of olderadults transferred to a rehabilitation unit. Their study demonstrated a significantdecrease in depression among those individuals transferred to the rehabilitationunit with a pet bird over those in the control group who were not given a bird.The presence of a bird, however, made no significant difference in the morale orloneliness of the subjects.In a study of single seniors in their homes, Lane and De Gale (1992)measured self-perceptions of health, finances and living situations as well asindices of depression, loneliness, and life satisfaction among pet owners andnon-pet owners. Overall, pet owners viewed their health, finances, and livingaccommodations as significantly better than did the non-pet owners sampled.Objectively, there was a statistically significant difference between the two groupsonly on the life satisfaction scale (pet owners scored higher). Unlike thepreviously cited study of Baun et al. (1992), Lane and De Gale (1992) found thatthe depression and loneliness indices of their two study groups wereindistinguishable.In another study of perceived health status among the elderly, Siegel(1990) reported on a one-year prospective study of 938 individuals. When alldemographic characteristics were controlled for (including social support networkand chronic health problems), those with pets reported fewer physician contactsduring the year than did those without pets. The accumulation of stressful events2 3during the year was no different between the two groups; however, the eventsresulted in more physician contacts by the non-pet owners than by the petowners. Dog owners, in particular, did not utilize their physician in times ofstress. They reported that their pets provided them with companionship (75%),security (25%), and love (21%). It is hypothesized by the author that thosewithout dogs or other pets may be using their family physician to fill thiscompanionship role in times of stressful life occurrences (Siegel, 1990). Adescriptive study (Cookman, 1991) using grounded theory methodology to studyattachment of a similar population of pet owners demonstrated that elderly petowners used their animals to fill a perceived void in companionship, therebyconfirming Siegel's (1990) results.Stallones, Marx, Garrity and Johnson (1990) also studied use of healthcare services among pet and non-pet owners. They surveyed 1,300 adults (ages21 to 64) from the general population in the United States to determine therelationship between pet ownership and pet attachment with self-reported illnessbehaviour and depression. Their telephone survey used numbers of visits to thedoctor, use of prescription medication, and numbers of hospitalizations tomeasure illness-related behaviours. The respondents were also questioned as totheir self-perception of health, recent negative life events, support network andperceived attachment to their pets. The data analysis revealed no significantdifference between pet owners and non-pet owners on comparisons of illnessbehaviour, depression, life changes and support network. The results alsodemonstrated a significant inverse relationship between attachment to pets andpresence of a human support network. The absence of a human supportnetwork was associated with emotional well-being. On the basis of this study,Stallones et al. (1990) caution against advocating pet ownership for depressedindividuals as it may lead to a decrease of human supports and provide yet2 4"another stressor ... to increase rather than decrease emotional distress" (p.108). A review of the etiology of depression in adults and the use of a conceptualframework for the placement of pets in relation to the health parameters mayhave further assisted Stallones et al. (1990) in drawing conclusions from thelatter finding.The interrelationship between human-pet attachments and subsequenthuman-human interactions has been another subset of research into theemotional well-being of pet owners. In one of the original studies in this area,Brown, Shaw, & Kirkland (1972) reported a positive correlation between humanaffection for dogs and affection for humans. In opposition to this view, another ofthe original pet researchers determined that pet owners and men in general likepeople less than do non-pet owners and women in general (Cameron, 1966).They conclude that pets seem to function as a deterrent to effective socialrelationships and, consequently, to mental health (Cameron & Mattson, 1972).Given the nature of their sampling procedures and correlational analysis,Cameron and Mattson (1972) may be presumptuous in their statement ofcausation.More recently, the effect of human-pet relations on human-humanrelations has been tested using a variety of study designs. Robins, Sanders, andCahill (1991) used a participant observation research design to demonstrate thatdogs facilitate interactions with strangers in a park setting and help to establishtrust among the newly acquainted. Messant (1983) had 40 volunteer dogwalkers, and 40 control group walkers (no dog) observed with behaviouralmeasurements of the frequency and nature of interactions with other humans.He concluded that the presence of the animal acted as a "social lubricant"(similar to the findings of Mugford and M'Comisky, 1975). The presence of a dogincreased the likelihood of interactions with other people. These others usually2 5spoke to or touched the dog while also addressing its owner. Such anobservation reinforces the significance of the pet in the interaction in comparisonto those walkers who did not have a dog present.Rossbach and Wilson (1992) built on the previous research in their studyin which 34 subjects were asked to rate a series of 32 photographs ondimensions of approachability, happiness, relaxation, and "best" photo. Thephotographs consisted of the same four individuals in combinations of seatedand standing positions, alone, with flowers, with their own dog, and with astrange dog. In all cases, pictures which had dogs in them were rated higherthan all other photographs. In the second stage of the study, 45 subjects wereasked to rate slides of scenery, an individual alone in the scene, and the sameindividual with a dog in the scene. The findings demonstrated a preference forscenery alone on ratings of aesthetics, happiness, relaxation, safety, and "best"picture. Second to the images of scenery alone, were the scenes with anindividual and a dog. It was only when the subjects were asked to placethemselves in the photo, that the preference on all scales became the slide withthe dog. Rossbach and Wilson (1992) interpreted these results to indicate thattheir subjects would rather be with a dog than alone (companionship) and that anindividual with a dog is more appealing than one without (perhaps as a sociallubricant).Paul and Serpell (1992a) surveyed 385 university students regarding theirchildhood relations with pets and current attitudes towards animals and humans.Their results demonstrated a significant positive relationship betweeninvolvement with pets as children and both positive attitudes to animals ingeneral, and higher scores on a scale of empathy for other humans. Thisconfirms an earlier study of childhood pet ownership by Poresky (1990) whodemonstrated that empathy towards children by other children was correlated2 6with empathy towards pets. Children with a strong attachment bond to a pet hadhigher scores on empathy towards other children than did the children in thestudy who did not have pets (Poresky, 1990). This finding conflicts with those ofStallones et al. (1990) who concluded that pet attachment among adults wasnegatively correlated to attachment with humans.Gage (1992) studied 454 couples with a baby of one year or less. Aftercontrolling for factors of the parental childhood experience, and level ofsocialization for marital and parental roles, the findings demonstrated asignificant correlation between the extent of pet experience and competence inmarital and parental roles using established scales for self-esteem, lifesatisfaction, marital satisfaction, and parenting competence. Similar to theaforementioned studies, Gage (1992) demonstrates the effects relations withpets may effect other relations with humans.In regard to pets and the adolescent population, Covert, Whiren, Keith,and Nelson (1985) interviewed 285 urban and rural adolescents from 10 to 14years old about the nature of their relationship with their pets. Results indicatedthat 89% of those surveyed had cared for an animal. The benefits of petownership among those surveyed was found to be significant on the parametersof reported friendship (dog, fish, bird, and horse owners only), expandedknowledge (fish, and bird owners only), and acknowledged responsibility (smallmammal and horse owners only).In a study similar to that of Covert et al. (1985), Stevens (1989) surveyed490 urban and rural grade eight adolescents. The investigator determined on thebasis of the survey that urban adolescents were significantly more attached totheir pets than were their rural counterparts. There were no significantdifferences in the level of pet attachment among male or female adolescents orbetween dog or cat owning individuals. The study confirmed the more utilitarian2 7role animals play in rural families as well as the lack of gender bias or speciespreference in the pet attachment of urban adolescents.Much of the remaining literature concerning adolescents and pets, groupsadolescents together with children in an analysis of developmental taskachievement. Davis and Juhasz (1985) discuss the formation of a concept of"self" to be an important developmental task of the preadolescent. Pets maycontribute to adolescent confidence and sense of self by a facilitation of healthyexpression of feelings through play, a non-authoritarian nature, consistentavailability, maintenance of "confidences," and unconditional acceptance. Theyindicate that the pet's involvement as a developmental asset will fluctuate overtime and according to the individual needs of the adolescent and caution that "arigid relationship indicates limitations in growth potential relative to developmentalprogress" (p.91).Schaufl and Bergler (1992) found that among the 460 nine to fourteenyear olds they surveyed, dogs took on several roles and functions for theirsubjects. Namely, the pets represented fantasy, friendship, entertainment,facilitators of communication, topics of conversation, comforters, "nurses", andcatalysts for family togetherness.In addition, there are literature reviews and philosophical constructsdiscussing the roles played by pets in families with children. Albert and Bulcroft(1988) interviewed 320 pet owners and 116 non-pet owners to determine thedegree of attachment of pets have to the family system. Their findings suggestthat pets are viewed as family members who play significant emotional andpsychological roles in urban households. The nature of the role the pet holds inthe family system varies over the life cycle and type of family. Cain (1983)identified both positive and negative aspects of pet ownership dependent on thelife stage and the dynamics in individual families. The pets in the households2 8with adolescents frequently served as a confidant for the adolescent as well asteaching attributes of independence and responsibility and providing a source ofcompanionship (Albert & Bulcroft, 1988; Schaufl & Bergler, 1992).Robin and ten Bensel (1990) studied the role played by pets in thedevelopmental tasks of the adolescent:Pets function, particularly for adolescents, as transitionalobjects, much like the blanket or the teddy bear does forinfants. As transitional objects, pets help children feel safewithout the presence of parents. Pets are more sociallyacceptable as transitional objects for older children than areinanimate objects. Adolescence brings with it a changingrelationship to pets....At this period pets can be a confidant,an object of love, a protector, a social facilitator, or a statussymbol (p.17).The literature further indicates that, due to the intense relationship thatadolescents, (more so than elementary school children) have with their pets(Kidd & Kidd, 1990a, 1990b), it is members of this age group that generally havethe most profound and prolonged periods of grief in the event of loss of a pet(Robin & ten Bensel, 1990).In summary, a vast array of lay and anecdotal literature describes themental health benefits of pets for their owners. The research-based literature isas yet unclear as to the exact nature of the benefits of pets to humans.Consistent with the findings of Baun et al. (1984), Friedmann et al. (1980), andRobb and Stegman (1983), there appears to be a paucity of controlledquantitative investigations of particular health parameters in humans in relation tothe presence of household pets. The relationship between measurement ofdepression and pet ownership also appears to vary according to the age and2 9circumstances of the subjects. In general, the research literature confirms thatanimals are able to facilitate human-human contact. Pets in childhood areinstrumental in fostering feelings of empathy towards other humans in adulthood.However, as previously noted, despite the descriptions of the role pets may playin the developmental tasks of adolescents, there is little in the research literatureregarding the relationship of pets to the current mental health of adolescents.SummaryIn this chapter an overview of the relevant literature pertaining to theconcepts of adolescent depression and the health benefits of pets has beenpresented. The research literature regarding adolescent depression addressesthe demographic correlates of depressed youth as well as the potential etiologicbasis and behavioural representations. There is evidence in the literature thatadolescent depression is related to suicide, delinquency, dropping out of school,poor academic performance and substance abuse (Rogers, 1987). It is alsoclear from the literature that there are recognized differences in adolescentdepression between urban and rural populations and among different culturalgroups. The points on which the literature is not clear relate to important aspectssuch as overall prevalence in the population, as well as any demographiccorrelations of depression with gender, family structure, age, and socioeconomicstatus.The presence of household pets has not previously been identified as afactor in dealing with this health concern among adolescents. Literatureregarding the health benefits of pets ranges from lay anecdotal articles to reportsbased on rigorous experimental design. The areas pertinent to the concept ofmental health include the relationship of pets to stress responses, as well asmeasures of emotional well-being. The literature describes recognized3 0differences in the role played by pets for people living in urban and rural settings.The research literature also confirms that pets provide increased feelings of love,belongingness, and companionship for their owners. Despite the wealth ofanecdotal articles promoting other emotional health benefits of pets, the findingsof the research-based literature are conflicting as to whether pets are, in fact,correlated with a change in blood pressure, heart rate, loneliness, depression,affection for other people, morale, locus of control, and social interactivenesswhen examining groups of pet and non-pet owners.Although findings have indicated that adolescents are one of the agegroups most affected by the death of a pet, and rate as much more attached totheir pets than do younger children, there have been few studies regarding therelationship between mental health of adolescents and the presence of pets. Forthese reasons, this study helps to fill the void noted in the review of the literature.The study addresses the need for further clarification of the prevalence anddemographics of depression a selected group of adolescents. The study alsobuilds on the foundation knowledge in examining the correlation between thisspecific mental health concern and the presence of pets among members of thepopulation. The study adds to knowledge about both the emotional health ofurban grade nine adolescents and the potential correlation of emotional healthwith the environmental influence of household pets.31Chapter ThreeMethodsA multi-method quantitative design was used to address all four researchquestions. The first question relating to the prevalence of depression among thesubjects was addressed using a descriptive survey design. The second, third,and fourth questions concerning the relationships between pets, gender, anddepression, as well as primary ownership of pets and depression, and perceivedpet attachment and depression, were addressed by correlational study designs.InstrumentsThe instruments utilized in the study were the Adolescent and PetCharacteristics Questionnaire (see Appendix A) and the Center for EpidemiologicStudies Depression Scale (CES-D) (see Appendix B).Adolescent and Pet Characteristics Questionnaire The Adolescent and Pet Characteristics Questionnaire is an instrumentdesigned by the researcher to collect nominal level, descriptive demographicdata about the subjects (see Appendix A). The following criteria were utilized foritem inclusion on the Questionnaire:1. Each independent variable was identified in the literature as significant orpotentially significant (content validity);2. The inclusion of each variable appeared logically relevant in relation to theresearch questions and purpose of the study (face validity).The Questionnaire was submitted to a convenience group of 5 male and 6 femaleadolescents (aged 13 to 20) to assess for comprehensiveness, clarity, and facevalidity prior to its use. No changes to the instrument were made on the basis ofthis preliminary assessment.3 2The Questionnaire includes fourteen items related to gender, age, schoolperformance, current family members at home, immigration (of the subjects aswell as their parents), language, and presence of pets in the home. For thosesubjects who have pets, a further seven items (for which short responses arenecessary) gather additional information as to type and number of pets, identifiedfavorite pet, primary ownership, duration of ownership, and perceived degree ofattachment to the pet. Subjects complete the instrument by circling the nominalcategory or briefly describing that which best portrays their current situation.Center for Epidemiologic Studies Depression Scale The Center for Epidemiologic Studies Depression Scale (CES-D)(Radloff, 1977) was used to measure the prevalence of depression among thesubjects (see Appendix B). Unlike measures used in other studies (Beck, Ward,Mendelson, Mock & Erbaugh, 1961 ; Siegel & Griffin, 1984), the CES-D wasdesigned to measure depressive symptoms in the general population as opposedto the clinical case and is, therefore, particularly pertinent to the subjects in thestudy. The CES-D consists of 20 summated Likert-type statements on aunidimensional measurement of depression. The depression factors includemeasurement of depressed affect, positive affect, somatic concerns, slowedactivity, and interpersonal relations. The subjects circle the response which bestdescribes how frequently each statement pertained to them during the past week.There are four potential responses to each statement ranging from 0 to 3, with 0indicating "rarely or none of the time" and 3 indicating "most or all of the time"(Radloff, 1977). The subjects' circled responses are totalled to obtain an overallscore for the scale, which may range from 0 to 60. The CES-D scoring for adultpopulations uses a cut-off value of >15 to indicate the presence of depressionsymptoms (Radloff, 1977). Several authors have indicated, however, that this is3 3not an adequate indicator for an adolescent population and scores of >20(Pronovost et al., 1990), >24 (Ackerson, Dick, Manson, & Baron, 1990) or >28(Clarke, 1991; Manson, Ackerson, Dick, Baron & Fleming, in press) are betterindicators of specific depression-related symptoms in adolescents. Indeed, thoseadolescent studies which have utilized the adult cut-off point of >15 fordepression have demonstrated prevalence rates of depression such as 49% and58% of the population (McDermott et al., 1990). These rates are in excess of theadolescent depression prevalence rates reported in studies using alternatemeasurement instruments (Greydanus, 1986; Simeon, 1989). For the purposesof this study, no distinct cut-off point was utilized. Rather, a series of four scoringcategories, based on the cut-off points utilized by the above researchers and theprinciple of interval data, was used to indicate various degrees of depressivesymptoms. An overall score on the CES-D from 0 to 15 indicated no depression,a score from 16 to 30 indicated mild depression, a score from 31 to 45 indicatedmoderate depression, and a score from 46 to 60 indicated severe depression.The CES-D was designed for use with English-speaking, Caucasian orAfrican-American populations of both genders, a wide range of socioeconomicstatuses, and education levels and is freely available for public use. Althoughoriginally designed to measure depression in an adult population, it hasdemonstrated high concurrent validity (r = 0.73) and excellent internalconsistency (0.91 for suicidal youth, and 0.81 for non-suicidal youth) amongadolescent populations in both Canada and the United States (Clarke, 1991;Pronovost, 1990; Roberts, Andrews, Lewinsohn & Hopps, 1991). The exactreading comprehension level is not known; however the instrument was designedfor use with "less than high school" (Radloff, 1977, p.398) levels of education.Adolescent subjects complete the instrument in an average of five minutes.3 4Factor analysis of the internal sub-scales of the instrument revealed a highinterrelatedness that cautions against separating out any internal measures ofdepressed affect, positive affect, somatic retardation, and interpersonal factors.Rather, Radloff (1977) recommends using the tool only as a whole. The CES-Dalso has strong discriminant validity when compared to a series of 12 othersimilar scales (Radloff, 1977).Protection of Human RightsApproval of procedures for the protection of human rights was obtainedfrom the University of British Columbia Behavioural Sciences ScreeningCommittee for Research and Other Studies Involving Human Subjects prior tocommencing the study. Throughout the study, every effort was made to maintainanonymity of the subjects. Despite the fact that the students and teachers wereaware of who participated in the survey and who did not, no names wereaccepted on the completed instruments. The instruments were collected in anenvelope without any observation of results by the researcher or the teacher.Parents of the children were informed of the study and were offered theopportunity to view the instruments if they desired prior to providing their consent.None of the parents contacted elected to do so.Prior to conducting the study, the researcher, in conjunction with theschool counsellor or principal, identified an individual within the school to whomthe students could go for counselling if they found that the questions asked in thesurvey had caused them to think about problems they were experiencing. Thisinformation was verbally given to the students as a group after all the instrumentswere collected. Although some of the responses revealed severely depressedindividuals, no effort was made to seek out the identity of those individuals as thiswould have constituted a breach of anonymity. Rather, contact was made with35the principal or guidance counsellor of the three schools from which thecompleted instruments had been collected. Each was informed that there hadbeen responses indicating depressed students in the school. They were thenprovided with information about community resources in the area of the school toprovide class presentations or counselling to self-identified individuals.SampleAll secondary schools in urban Vancouver (eighteen) and Victoria (eleven)were approached for participation in the study. Only those in which approval wasreceived from both the principal and the classroom teacher involved wereincluded. These schools (three in Vancouver and three in Victoria) represented avariety of socioeconomic levels in the urban population (Capital Regional DistrictHealth, 1990; Rumel & Costanzo, 1992; Siegel & Griffin, 1984). Participationwas sought from all grade nine students in the six schools. Only those studentswho returned a signed parental consent form were included in the sample, whichconsisted of both males and females.The choice of grade nine students was made based on the literature whichdocuments an increase in the incidence of depression in the 13 to 16 year oldage group (Greydanus, 1986) as well as an increase in suicide rate in the 15 to18 year old age group (Conrad, 1991; Pronovost et al. 1990). Individuals 13 to16 years of age are at an optimal age for experiencing depression, but may notyet have dropped out of school (Reinherz et al. 1991) or been led to suicidal acts.Data CollectionApproval for completing the study was obtained from the VancouverSchool Board and the Victoria Regional District School Board. The VancouverSchool Board then approached the individual school principals to request theirconsent for school participation. The schools within the Victoria School District3 6were individually approached by the researcher through a letter introducing thestudy and requesting consent (see Appendix C). The principals were thencontacted by telephone to arrange a time to obtain the signed consent, answerany questions they might have, and complete the final arrangements for thestudy. For those principals who did not respond to the initial letter or telephonecall, a FAXed memo was sent requesting their participation.Once the schools were identified, letters were distributed explaining to thestudents and their parents the nature and purpose of the study, as well as thetime and effort required (see Appendix D). Active signed consent was requestedfrom the parents and required prior to student participation in the study. In four ofthe schools, parental consent forms were sent home with the subjects andreturned to class in the same manner. One school had a blanket consent onrecord which the parents had signed at the beginning of the school year givingtheir consent for voluntary participation by their child in any study approved bythe school board. In the sixth school, the principal requested that the studyinformation and consents be mailed home to the parents to avoid placing extraduties on the teaching staff. In the first four schools, the principals or classroomteachers volunteered to collect the parental consent forms from the students priorto the survey date. In the last school mentioned, the students either turned theirsigned consent form into the school office or brought it with them at the time thatthe instruments were completed.Once consent was obtained from the parents and subjects, the data werecollected. Each school varied slightly as to the procedure which the principalpreferred to have followed. One school with relatively few students gathered allof those with parental consent from their classrooms into a room at a pre-arranged time. Another preferred that the students respond to the instrumentsduring lunch time, and the four remaining schools had the instruments distributed37to participating subjects during a guidance class or, in the case of one school, anEnglish class.The researcher was able to personally distribute the questionnaires in fiveof the six schools. In each of these instances, the researcher introduced herselfas a nurse interested in learning how teenagers currently felt about themselvesand in the degree of involvement they had as individuals with pets in their homes.It was stressed that there were no right or wrong answers to the questions. Thesubjects were encouraged to express how they truly felt and not what theythought others might want to hear. Finally, it was pointed out again to thestudents that there were two pages to be completed and that their names werenot to be put on either form. The students were encouraged to respect eachother's privacy and look only at their own paper, fold it in half when they finished,and place it in the envelope held by the researcher. In the sixth school, theguidance classes were held at a time at which it was impossible for theresearcher to be present. The above directions were given to the guidancecounsellor who read the introductory statement and instructions to theparticipating subjects in each of his nine classes. In all cases, the subjects werepresented with the CES-D Scale first, followed by the Adolescent and PetCharacteristics Questionnaire.Data AnalysisData from the Adolescent and Pet Characteristics Questionnaire weretabulated as follows. The questions for which the potential responses were of alimited number were analyzed at face value as the nominal category which theyrepresented (gender, age, school performance, number of parents, number ofsiblings, birth country, years in current neighbourhood, primary language,presence of pets, type of pet, primary pet ownership, duration of pet ownership,3 8and attachment to pet). Four of the short-answer questions had numerouspotential responses (place in sibship order, overall numbers and types of allhousehold pets, name of pets, and occasion for pet acquisition). The responsesto these questions were categorically grouped according to commoncharacteristics. Sibship order was analyzed as eldest, youngest, and anywherein the middle of a group of siblings. Overall numbers and types of householdpets were analyzed in groupings of single pet, multiple pets of same species, andmultiple pets of different species. Names of pets were grouped in to categoriesof human names, names based on physical characteristics, names based onsocial or behavioural characteristics, humourous names, and miscellaneous.The reason for pet acquisition responses were grouped into categories of noreason, unplanned finding, gift, planned desire, replacement of a past pet, andreplacement of a past significant other.The first research question was answered with a prevalence ratio usingthe values from the CES-D scale (Glass & Hopkins, 1984; Munro, Visintainer, &Page, 1986; Radloff, 1977; Zar, 1984) to delineate those respondents who werenot depressed, mildly depressed, moderately depressed, and severelydepressed.The second research question was analyzed using a two-way fixed effectsanalysis of variance (ANOVA) design (Glass & Hopkins, 1984; Munro et al.,1986; Zar, 1984). In this case, the dependent variable was the interval levelmeasurement of depression as recorded by the CES-D score. The independentvariables of gender (female/male), and pets in the home (no/yes) were obtainedfrom the Adolescent and Pet Characteristics Questionnaire.In order to use the ANOVA statistical analysis procedure, three primaryassumptions about the data must be met (Glass & Hopkins, 1984; Munro et al.,1986; Zar, 1984). The first ANOVA assumption is that the dependent variable be3 9composed of normally distributed continuous data. Secondly, the sampledgroups must have equal (homogeneous) variances. Finally, the factor effectsmust be linear or additive in nature. To determine the distribution (normality) ofthe CES-D scores, the results were plotted graphically using a density distributionhistogram and a normal curve. The results showed a positively skeweddistribution with a long right-sided tail (see Appendix F) (Glass & Hopkins, 1984).To utilize parametric analysis procedures with these data as they existed in theirraw numerical form would be in violation of the assumption of normal distribution(Glass & Hopkins, 1984; Zar, 1984).A logarithmic (base 10) data transformation was used to correct for thepositively skewed distribution of CES-D scores. The conversion equation of X'=log 10 (x+1) was utilized due to the presence of several "0" scores in the data setand an inability to provide a log 10 value for 0 (Zar, 1984). This resulted in anormal distribution (see Appendix F). In addition to obtaining a normaldistribution of data, logarithmic transformations ensure data linearity or additivity(Zar, 1984). Therefore, the analysis of research questions three, and four(utilizing parametric analysis of variance procedures) also used logarithmic 10(x+1) transformed data.The third research question was addressed using a one-way fixed effectsANOVA design (Glass & Hopkins, 1984; Munro et al., 1986; Zar, 1984). Thisutilized a subset of the dependent variable of depression (CES-D score) whichincluded only those individuals who had pets in examination of combinations ofindependent pet ownership (individual / family / parent / sibling) as reported onthe Adolescent and Pet Characteristics Questionnaire. Post hoc analysis of thedata was completed using the Tukey test for honest significant difference asprovided by SYSTAT with Tukey-Kramer adjustment for unequal group numbers(Wilkinson, 1990). This adjustment is reported to provide more consistent4 0protection against type I and type II errors than the Newman-Keuls or Duncanpost hoc analyses (Wilkinson, 1990).The fourth research question was also addressed using a one-way fixedeffects ANOVA design (Glass & Hopkins, 1984; Munro et al., 1986; Zar, 1984).Once again the dependent variable was the subset of depression (CES-D)scores for those individuals with pets in examination of the relationship withperceived attachment as reported on a scale from 1 to 5 (not attached /somewhat attached / very attached / extremely attached) on the Adolescent andPet Characteristics Questionnaire. Post hoc analysis of the data was alsocompleted using the Tukey test for honest significant difference as provided bySYSTAT with Tukey-Kramer adjustment for unequal group numbers (Wilkinson,1990).41Chapter FourPresentation and Discussion of FindingsResponse RateGrade nine students of three Vancouver and three Victoria secondaryschools were the subjects for this study. Of a possible 710, 411 consent formswere returned for an overall response rate of 58%. A response rate between50% and 60% is expected when employing active parental consent procedures(Ellickson & Hawes, 1989). Such a consent procedure has been criticized for itspotential underrepresentation of ethnic minority groups, low achieving students,and "those at risk for engaging in problem behavior" (Ellickson & Hawes, 1989,p.46). Although it is probable that some parents did return the signed consentforms because they were unable to read English, a demographic review of thesubjects' ethnic origins later in this chapter will demonstrate that such anunderrepresentation was not likely the case in this study. This is likewise thesame for scholastic ability. Only eight of the 411 returned consents (1.9%)indicated parental refusal to grant permission for their child to participate in thestudy. As shown in Figure 2, the majority of the participants came from twoschools in which the teachers strongly encouraged and reminded the students toreturn their signed parental consent forms.Of the 403 completed instruments, 401 were usable. Two instrumentswere discarded; one was incomplete and the other had obviously fictitiousanswers. The remaining instruments were filled out in their entirety by thesubjects. Many were returned with elaborate explanations included in themargins to further describe the subject's circumstances to the researcher.Due to the profound differences in size among sample groups, before anyanalysis was completed, the differences between sample sites and depressionscores were analyzed to ensure homogeneity of variance (Zar, 1984). The13433.4%42results of this preliminary one-way fixed effects ANOVA (with log 10 (x+1) CES-Dscores as the dependent variable and school as the independent variable)indicated a probability statistic of p = 0.393. With statistical significance set at a <0.05 alpha level, this result clearly indicates that there is no significant differencebetween the depression values among the sampled school groups. Therefore,despite the selection bias of most subjects being from two schools, the analysisindicates homogeneity of variance across all sites. For the remainder of theanalyses, all school values were pooled and treated as one sample (Zar, 1984).250■- 51.1%205^ -2005003.7%151^I 4.7%^4.74^2.2%19^19 ^9 I^I 1^I^IA C D E FSchoolsFigure 2. Numbers and Percentages of Subjects by SchoolDemographic CharacteristicsGenderThe sample was composed of 53.6% female and 46.4% male subjects.43AgeAll subjects were in grade nine at the time of the study. Their ages rangedfrom 13 to 16, with almost 80% of the sample being 14 years of age (see Table1).Table 1Ages of Subjects Age (Years)Subjects (N=401)Number Percentage (%)13 12 3.014 316 78.815 70 17.516 3 0.7School Performance Almost 85% of the subjects reported receiving either B or C grades ( seeTable 2). These grades have been interpreted by the investigator as "high-average" or "average," respectively. The reported grades for the subjectsTable 2School Performance of Subjects GradesSubjects (N=401)Number Percentage (%)A 45 11.2B 193 48.1C 149 37.2D 14 3.54 4in this study are similar to those reported in the larger adolescent health survey inVancouver and Victoria which were generally above average (McCreary, 1993a,1993b).Family Composition Subjects reported the number of parents and siblings currently living athome with them and their position in the overall sibship order of the family. Themajority of subjects (78.8%) reported living with two parents, while 20.4%reported that they lived with only one parent and three subjects (0.7%) lived witholder siblings in lieu of their parents (see Table 3). During the completion of theinstruments, a total of six students asked if they were to count their parent'sspouse as a second parent on the instrument. The students were encouraged todo so only if they thought of that person as a parent. The percentage of twoparent families is slightly larger than reported in other studies: 75% for Canada(Bibby & Posterski, 1992); and 71% and 59% for Vancouver and Victoriarespectively (McCreary, 1993a, 1993b).Table 3Parents Living at Home Subjects (N=401)Parents at Home^Number^Percentage (%)0 3 0.71 82 20.42 316 78.8With regard to siblings living with the subject, the numbers ranged fromzero to seven (see Table 4). The largest representative group was 42.4% of thesubjects who reported one sibling living at home with them. Almost 13% reportedbeing the only child in the house, while nearly 28% reported having two siblings45at home. Seventeen percent of the respondents reported 3 or more siblings athome.Table 4Siblings Living at HomeSubjects (N=401)Siblings at Home Number^Percentage (%)0 51 12.71 170 42.42 112 27.93 42 10.54 18 4.55 5 1.26 2 0.57 1 0.2The number of siblings at home does not always correlate with the overallnumber of siblings in the family. This is illustrated by the figures which indicate12.7% of the subjects as the only child living at home, while only 10.5% of thesubjects were actually the only child in the family (see Table 5). Approximatelytwo-thirds of the subjects reported being either the eldest or youngest child, whilethe remaining 21% reported being somewhere in the middle of the sibship order.The presence of certain family members in the home may be beneficial in thatthey provide additional coping resources for the adolescent in emotionalreappraisals. The absence of certain family may also be beneficial by decreasingthe chance of environmental encounters which result in negative emotions for the46adolescent. Obviously, the reverse of each situation is also plausible within theconceptual framework.Table 5Sibship Position of Subjects Subjects (N=401)Position^ Number^Percentage (%)Eldest 139 34.7Middle 85 21.2Youngest 135 33.7Only Child 42 10.5Country of Origin Cultural practices may be passed on to the children of immigrant parents,therefore, information regarding country of origin was gathered on both subjectsand their parents. The majority of the families originated in Canada with the nexttwo largest groups being from China and Hong Kong. However, as shown inTable 6, there were 49 different countries of familial origin, placed within eightregional categories. The subjects were highly multicultural as a sample group;however, the coping conceptual framework allows for differences associated witha variety of beliefs, values and resources. Cultural beliefs may influence howadolescents respond to environmental stressors and hence their establishment ofcoping responses (Bibby & Posterski, 1992).As Table 7 indicates, 296 (73.8%) of the subjects were born in Canada orthe United States. Seventy-seven (19.2%) were born in the Orient, and 28 (6.9%in total) in a variety of places across Asia, Australia/New Zealand, Europe, LatinAmerica, or the Middle East.47Unlike the subjects, the parents were not primarily from North America(see Table 7). The highest percentage of the parents were born in the Orient(average 37.4%). Those born in North America were second with an averagerepresentation of 31.9%. Twenty-one percent of the subjects' parents came froma variety of the countries listed in the European region. Parents of the remainderTable 6Regions and Countries of Origin for Subjects and Parents Countries of Subjects and All Reported ParentsAsia - Non-orientAsia - OrientRegionNorth AmericaLatin AmericaEuropeCanada^(505)Argentina^(4)Costa Rica (1)El Salvador (10)Honduras^(1)Austria^(1)Bulgaria^(3)CroatiaCzech Rep. r9England^(39)Finland^PFranceGermany^(7)Greece^(16)Holland^(2)Hungary^(9)Cambodia (3)China^(120)Hong Kong (103)Japan^(6)Korea (4)Brunei^(2)Fiji^(19)Middle East^Israel^(1)U.S.A. (42)Jamaica (2)Peru (2)Venezuela (4)Ireland (4)Italy (16)NorwayPoland r6Portugal (27)Romania (1)Scotland (7)Spain (1)Wales (2)Yugoslavia (6)Denmark (2)Malaysia (15)Philippines (23)Taiwan (42)Viet Nam (55)India (19)Lebanon (8)Africa Mauritius^(1)^South Africa (4)Australia /New Zealand^Australia^(8)^New Zealand (2)48were born in Africa, Asia, Australia New Zealand, Latin America and the MiddleEast regions (average total of 9.3%). Thus, almost half of the subjects were firstgeneration Canadian. This is important to note as it may indicate potential areasof conflict when the values and beliefs of the family are different from thoseencountered by the subject when among her / his peer group (Bibby & Posterski,1992).Table 7Subjects and Parental Birth Regions RegionNumbers and PercentagesSubjects (N=401)^Mothers (N=396)^Fathers (N=390)North America 296^(73.8%) 131^(33.1%) 120 (30.8%)Latin America 6^( 1.5%) 9 ( 2.3%) 9 ( 2.3%)Europe 14^( 3.5%) 80 (20.2%) 87 (22.3%)Asia - Orient 77^(19.2%) 149 (37.6%) 145 (37.2%)Asia - Non-orient 4^( 1.0%) 18 ( 4.5%) 18 ( 4.6%)Middle East 1^( 0.2%) 4 ( 1.0%) 4 ( 1.0%)Africa 0 2 ( 0.5%) 3 ( 0.8%)Australia /New Zealand 3^( 0.7%) 3 ( 0.7%) 4 ( 1.0%)Primary LanguageIn addition to reporting countries of origin, the subjects indicated whichlanguage was primarily spoken at home. Forty-six of the subjects (11%)indicated that more than one language was spoken at home. If they did notmake clear which was the language most frequently used, the first of thelanguages listed was recorded as the primary language used in the home. Intotal, 26 specific languages were reported by the subjects as shown in Table 8.49As were the countries of origin, the languages were categorized to facilitatepresentation.Only 58.4% of the subjects reported English as the primary languagespoken at home. The second most frequently spoken languages were in theOriental category with 28.2% of the subjects reporting the use of at least one ofthese nine languages. The remainder of the subjects (13.4%) reported speakinga variety of the remaining 16 Asian or European languages. This is a largerpercentage (41.6%) of non-English speaking households than indicated by acensus completed in 1986 in which 21.8% of households in Vancouver were non-English speaking (Child and Youth At Risk Steering Committee, 1991).Table 8Primary Languages Used At Home Subjects (N=401)Category Number^Percentage^LanguagesEnglish^234^58.4%^EnglishAsian - Oriental 113^28.2%^Cantonese (55) Korean^(2)Chinese^(16) Mandarin (20)Hakka^(2) Taiwanese (5)Japanese (1) Vietnamese (11)Tagalog^(1)^2.0%^Hindi^(4) Lebanese^(1)Punjabi^(3)Asian -^8Non-orientalEuropean -Latin-based^27^6.7%^French^(13) Portuguese (8)Italian^(1) Spanish^(5)European -Non-latin-based^19 4.7% BulgarianCroatianCzechFinnishGerman(1) Greek^(5)(4) Hungarian (2)(2) Norwegian (1)(1) Polish^(1)(2)5 0A reason for the difference in these numbers could be that the 1986 figureincludes all families regardless of composition, whereas this sample includes onlythose with a grade nine student in the home. It is also possible that multilingualfamilies make a concerted effort to speak their native language at home so thattheir children who are being educated in English will retain their non-Englishlanguage abilities, hence raising the statistical representation of non-Englishspeaking households for those families with children. It may also be that manymore immigrant families have settled in Vancouver and Victoria since the 1986census.ResidenceApproximately 70% of the subjects had lived their entire life in Canada(see Table 9). The remaining 30% of the subjects reported living in Canadaanywhere from two months to twelve years. Within this population, only about26% had spent their entire life in the same neighbourhood. Similar to the familiarbeliefs and practices found within cultural patterns, living in the same region for along period of time may permit the adolescent to cultivate a set of copingpractices which are effective in reappraising emotional circumstances.Twelve percent of the subjects had been in their current neighbourhood forless than a year. The remaining 61% of the subjects had lived in theirneighbourhoods anywhere from one to thirteen years (see Table 9). It should benoted that some of the subjects commented in the margins of the instrument thatthey had made several moves throughout their lives; however, always within thesame neighbourhood. It could be assumed, therefore, that support systems andresources would remain the same for these individuals.51Table 9Years of Residence in Canada and Current NeighbourhoodYearsNumbers and Percentages of Subjects (N=401)Canada^Current Neighbourhood< 1 9 ( 2.2%) 49^(12.2%)1 - 3 32 ( 7.9%) 89 (22.2%)4 - 6 17 ( 4.2%) 73 (18.2%)7 - 9 21 ( 5.2%) 40 (10.0%)10 - 12 40 (10.0%) 47 (11.7%)13 - 15 282 (70.3%) 103 (25.7%)In summary, the subjects in the study were generally similar to the cross-section of adolescent subjects from the same Vancouver and Victoria settingsparticipating in other studies. There were slightly more female than malesubjects and most were 14 years old at the time of the study. They attendedthree Vancouver and three Victoria secondary schools and their schoolperformance was generally in the average or high-average range. Three-quarters of the subjects lived with two parents and one to three siblings. Thesample was equally represented by subjects who were the oldest, youngest andmiddle in the sibship. Half of the subjects had lived their entire lives in Canadawith approximately half of the group remaining in the same neighbourhood. Themost remarkable characteristic of this sample was its multicultural nature,representing 49 countries and 26 languages.52Pet Demographic InformationPresence of Household Pets Within the group of 401 subjects, 226 (56.4%) reported having one ormore pets currently in the home. The remaining 43.6% of the sample had nopets. This is somewhat higher than the overall statistics which report theincidence of pet ownership to be approximately 50% of urban households in theUnited States and Britain (Beck, 1983; Rowan, 1992b) and approximately 50% ofall households in Canada (Bibby & Posterski, 1992). However, it is wellrecognized that households with children, especially between the ages of 6 and15, are more likely to have pets than those without children (Endenburg, Hart, &de Vries, 1990). A description of the pet demographics follows.Types of PetsOf the 226 subjects with pets in the home, 22.2% reported having severalpets of the same species. Examples of this are the subjects who reported havingtwo cats or sixteen fish. One subject reported having two adult and ten babyrats. Approximately 39% of the remaining sample reported having several petsof different species. Another 39% of the population reported having only one pet.Identified Favorite PetSubjects who reported having more than one animal at home were askedto answer the remainder of the questions on the Adolescent and PetCharacteristics Questionnaire with regard to their identified favorite pet. It wasassumed that subjects who had only one pet answered all questions in regard tothat single pet. Four subjects did not answer this question; however, they didrespond to other questions within the pet characteristics section so it is assumedthat they had a pet in the home and merely overlooked the question. Of thetypes of single pets identified, 32% were dogs, 31.1% were cats, 11.7% werefish, 11.3% were small mammals, 7.2% were reptiles or amphibians, and 6.7%53were birds (see Table 10). The categories of small mammal and reptile /amphibian were composed of the animals listed in Table 10.Table 10Favorite Pet Categories and TypesPet Owning Subjects (N=222)Category Number Percentage^TypesBird 15 6.7%Cat 69 31.1%Dog 71 32.0%Fish 26 11.7%Small Mammal 25 11.3% guinea pig (1) rabbit^(6)hamster (11) rat^(4)mice (3)Reptiles and 16 7.2% iguana (1) snake^(1)Amphibians newt (2) turtles (12)These findings are consistent with those of other studies involving youthand pets (Endenburg et al., 1990) but somewhat lower in dogs reported as petsthan in other studies (Kidd & Kidd, 1990b). This may be partially explained bythe fact that the subjects in the study were urban dwellers and there is a trendtowards cat ownership (58% increase in the past ten years) rather than dogownership (10% increase in the past ten years) in urban environments (Rowan,1992b).Pet Names Some pet theorists suggest that the type of name given to the pet mayreflect the type of role the pet owner expects the pet to assume within thehousehold (Fogel, 1983). The pet names provided by 91% of the subjects arecategorized with examples in Table 11.54Table 11Pet Name Categories and ExamplesPets (N=205)Category Number Percentage ExamplesHuman 92 44.9 Includes familiar first names andsurnamesHumourous 11 5.4 Biccardi, Cat (for a dog), Moose(for a hamster), Scat, Soup (for aturtle), Stew (for a rabbit), T-bone, YertleNone 27 13.0 The majority of fish ownersreported not naming their petPhysicalCharacteristics45 22.0 Bear, Blackie, Bubbleface,Chiquita, Diamond, Ginger, Jolie,Kiwi (for a newt), Pee Wee,Rabbit, Snowball, Spud, YellerSocial orBehaviouralCharacteristics19 9.3 Airwalk, Buddy, Cuddles,Eternity, Goofball, Hugs, Kisses,Spunky, Wacky, Undertaker (for asnake)Miscellaneous 11 5.4 Brax, Chichu, Cita, Csopi,Eleven, Mandu, Marmaul, NekoIn this study, the use of pets as coping mediators in the emotional reappraisal ofcertain situations might lead one to expect that the pets would be given eitherhuman names or names reflecting emotional or behavioural traits. The namemay reflect the expectation that the pet will act in an emotionally supportivemanner or even as a surrogate human.In fact, almost 45% of the subjects reported using a human name for theirpet, 22% described their pet by naming it with a physical characteristic, 9.3%used a social or behavioural trait as the name of their pet, 5.4% used obvioushumour when naming their pet, 5.4% named their pet by a word that could not becategorized usually because it reflected another language or an unusual concept,5 5and 13% reported not having a name for their pet. This last group was largelycomposed of fish or bird owners. It is possible that the 9% of the pet owners whodid not answer this question may not have had a name for their pet as many ofthem were also fish or bird owners. These names suggest that just over half thesubjects may have named their pet with regard to possible emotional copingstrategies.Duration of Pet Ownership More than half of the subjects (66.8%) reported having their pets for threeyears or less, with the largest proportion of the group (19%) having had their petsfor less than one year. The remaining 33.2% of the subjects with pets reportedhaving their pets anywhere from four to fifteen years.The framework for the study implies that the coping mediator must bepresent in order to be involved in the emotional reappraisal. This is the case forthe subjects in this study who currently owned pets. The study only accounts forpets presently owned by the subjects. However, there may be instances in whichsubjects had had a pet in the past and learned that walking the dog, for example,allowed them to cope effectively with emotional stress. The subjects may nolonger have their pets; however, they may be still using the coping mechanismwhich originated with pet ownership. In this example, however, the originalcoping mediator of pet ownership has been replaced within the framework bywalking.As the majority of subjects were relatively recent pet owners, thesedemographic data are best viewed while concurrently examining the reasons foracquiring the pets (most of which were acquired as the subjects enteredadolescence).5 6Reason for Acquiring PetThe majority of the subjects (51.5%) reported acquiring their pet for nospecific reason. Other subjects cited reasons behind acquiring their pet such asreplacing a previous pet (16.2%), as a gift (14.8%), having found it (7.9%),because they wanted it (6.9%), and to replace a family member who died (0.9%).Two subjects reported getting a pet for fun and entertainment (0.9%). Anadditional two subjects related other reasons such as "a fight between Mom andDad" and "we moved." These findings indicate no specific pattern as to thereasons for acquiring the pet. It could be that this may have been part of aparental decision and that the subjects were not generally privy to theinformation. The fact that most of these pets were acquired as the subjectsentered adolescence (and many of them may have had siblings who werealready adolescents), and that many of the pets were generally given humannames, may imply they were acquired to act as a non-judgemental support forthe subjects going through a period characterized by change such as the loss oftheir childhood, or of a family member, or previous pet. The coping conceptualframework demonstrates how this support may occur through the process ofemotional reappraisal with the use of a mediating variable such as the pet.Primary Ownership of PetAs Figure 3 indicates, the majority (43.4%) of subjects who responded tothis question, reported that the pet was owned by the whole family. The secondlargest group of subjects (35.6%) were the primary owners of the pets. Finally,14.2% of the subjects reported their parents as being the primary pet owners inthe family, and 6.8% reported siblings as being the primary pet owners. Of the226 subjects with household pets, 7 (3%) did not respond to this question. Theseresults were similar to those of Paul and Serpell (1992b) who noted primary petownership among adolescents to be somewhat lower than family pet ownership.57If the subject is to use the pet as an emotional coping mediator, this wouldbest be accomplished if the subject has frequent contact with the pet. This isrn10080CN1E(60040200 35.6%786.8%1595^43.4%14.2%31Family Parents^Self^SiblingOwnershipFigure 3. Pet Ownershipmost likely to occur if the pet is solely owned by the subject or if the subject is apart of family ownership of the pet. The presence of, for example, gerbils in asibling's bedroom, would not be expected to provide the coping mediationnecessary to effect emotional change unless the subject was frequently involvedin interactions with these pets. As well as interaction, pet ownership involvesresponsibility for providing care for the animal. In addition, there is often a senseof prestige which may accompany ownership of a pet.Pet AttachmentThe subjects reported perceived degree of attachment to their pets on afive point Likert-type scale from 1 to 5. On this scale, a rating of 1 was indicativeof no attachment and 5 was indicative of extreme attachment (see Appendix A).58Only 8 of the total 226 subjects with household pets did not respond to thisquestion. As indicated in Figure 4, of the subjects who did respond, 25 subjects(11.5%) reported that they were not at all attached to their pets. From a brief8033.5%► ► ►73706023.4%50E(z400302010011.5%25 10.5%21.1%465123►1^2^3^4^5Degree of AttachmentFigure 4. Perceived Attachment to Petreview of the completed instruments, the majority of these subjects were fishowners. Twenty-three subjects (10.5%) stated that they were only somewhatattached to their pets. Forty-six subjects (21.1%) rated their attachment as beingat the midline point on the scale. The remaining half of the subjects reported thatthey were either very (23.4%) or extremely (33.5%) attached to their pets. Inorder for the pets to be used as coping mediators it may not be entirelynecessary for the subjects to be attached to them. If they are attached, pets maybe utilized as confidants with whom adolescents may voice concerns and explorepossible solutions. If the subjects are not attached to the pets, they may still beused as emotional coping mediators. However, in this case, the pets may5 9become focal objects for the projection or displacement of the subjects' emotions(both positive and negative) through methods such as confrontation, distancing,self-control, planful problem solving, avoidance, and positive reappraisal (Pekar,1992; Puskar & Lamb, 1991).In summary, the subjects with pets in their homes displayed a broad rangeof characteristics. Those with pets were most likely to have either one pet ormany different species of animals in their household. The types of pets weregenerally one-third dogs, one-third cats, and one-third birds, fish, reptiles,amphibians and small mammals. Just under half of the subjects had humannames for their pets. The remainder named their pets a variety of humourousnames, names depicting physical or social/behavioural characteristics, orunknown names in their primary language. Well over half of the subjects hadhad their pets for three years or less and most had acquired them for noparticular reason. Three-quarters of the subjects reported either themselves ortheir whole family as the primary pet owner. Finally, approximately half of thesubjects felt they were very or extremely attached to their pets.Findings Related to Research QuestionsQuestion One : Prevalence of Depression The overall scores on the Center for Epidemiologic Studies DepressionScale (CES-D) were used to measure the prevalence of depression in the studygroup. The scoring with the CES-D is directly related to the level of depressivesymptoms. The scores ranged from 0 to 53 out of a potential 60 points. Themean score was 17.0, the median was 16.0, and the standard deviation was10.2. There was a difference in depression scores between adolescent females(mean score 18.7) and males (mean score 15.1).6 0Of the 401 subjects in the study, 194 (48.4%) obtained scores of 15 orbelow on the CES-D. These scores were interpreted as indicating no depressivesymptoms. Another 166 subjects (41.4%) scored between 16 and 30 and wereclassified as being mildly depressed. A third grouping of 35 subjects (8.7%)scored between 31 and 45 on the CES-D. These subjects were classified asbeing moderately depressed. Finally, 6 subjects (1.5%) scored 46 or above onthe CES-D. These individuals were classified as severely depressed, (see Table12). Thus, approximately half of the adolescents had some feelings ofdepression; however, only about 10% of the overall population scored above 30on the CES-D and could be rated as moderately or severely depressed.Table 12Categories of CES-D Scores Score Category of Symptoms Number (N=401) Percent (%)0 - 15 No 194 48.416 - 30 Mild 166 41.431 - 45 Moderate 35 8.746 - 60 Severe 6 1.5These findings are similar to those of previous studies which reportedmean values on the CES-D for adolescents of 16 (Manson et al.), 17.1 to 20.4(McDermott et al., 1990), 18 (Pronovost et al., 1991), or 19 (Clarke et al., 1990).Each of the studies utilized a different "cut-off" point to indicate the presence orabsence of depressive symptoms. The study in which the same cut-off point of>15 was used found prevalence rates similar to those in this study (McDermott etal., 1990). Studies using other cut-off points demonstrated prevalence rates fordepression among adolescents similar to those found in the moderately (8.7%)61and severely (1.5%) depressed categories of this study (Clarke et al., 1990;Pronovost et al., 1991). The results of this study indicate that 10.2% of thesubjects could be considered to be depressed, whereas 41.4% of the subjectsclearly have some mild symptoms of depression and 48.4% of the subjects haveno depressive symptoms. In relation to the study framework, clearly 10% of thesubjects are not coping effectively to reappraise the depressive emotions whichin adolescence may be a result of numerous environmental and somebiochemical sources. Despite the multicultural nature of the sample, thesefindings are similar to those in studies of more homogeneous ethnic populations.Therefore, although ethnicity was not a variable considered in this study, basedon comparisons with other studies, it does not appear to be a confoundingvariable which may have affected the results. Both genders were represented inall of the CES-D groupings. Males represented a higher proportion of thosescoring in the not depressed or mildly depressed categories, whereas femalesrepresented a higher proportion of those scoring in the moderately or severelydepressed categories.Question Two: Relationship Between Gender. Pet Presence and Depression The results of the two-way fixed effects ANOVA examine the relationshipbetween gender and depression, pets and depression and the combinedrelationships of gender with pets and depression (see Table 13). With regard tothe first correlational analysis, the results demonstrated a significant relationshipbetween gender and level of depression among the subjects (p = 0.000). FurtherTukey-Kramer post hoc analysis reveals that the female subjects weresignificantly more depressed than their male counterparts (p = 0.001). As theCES-D instrument has established reliability and validity across both genders(Radloff, 1977), it can be assumed that this difference in scores is truly6 2representative of the population sampled. This gender difference has been notedby other researchers (Baron & Perron, 1986; McDermott et al., 1990; Reinherz,1991, Schonert-Reichl & Offer, 1992; Simeon, 1989; Weissman et al., 1987).Table 13ANOVA Summary Table for Gender. Pets, and CES-D Scores (N=401) Source^Sum-of-Squares Degrees Freedom Mean-Square F-Ratio PGender^1.055 1 1.055 13.240 0.000Pets 0.489 1 0.489 6.142 0.014Pets*Gender 0.149 1 0.149 1.874 0.172Error 31.636 367 0.080Schonert-Reichl & Offer (1992) discuss several possible reasons whyfemale adolescents have higher prevalence rates of depressive symptoms thando males. One reason suggested is that male adolescents tend to exhibitdepression in the form of externalized behaviours such as delinquency andaggression whereas female adolescents express their disturbance in an inwardfashion. Hence, the CES-D tool may be inaccurate for use with maleadolescents as it emphasizes low mood and low energy symptoms as indicatorsof depression rather than high energy and destructive behaviour. In addition,some studies indicate that females report more symptoms than do males, andalso that expressing concerns is more acceptable for females than for males(Schonert-Reichl & Offer, 1992).Other influences which might explain the differences between prevalenceand severity of female and male depression symptoms are those of genderidentification and self-esteem. Schonert-Reichl and Offer (1992) report thatduring puberty, females and males begin to strongly identify with the extremes ofmasculine and feminine stereotypes. Research has demonstrated that6 3androgyny and masculinity are linked with higher levels of self-esteem and moreeffective coping mechanisms (Schonert-Reichl & Offer, 1992). Therefore,adolescent females suddenly confronted with extreme feminine stereotypes aremore likely to have lower levels of self-esteem than their male peers. In fact,among the adolescents in a recent study, Vancouver and Victoria females weretwo and four times, respectively, more likely to have low self-esteem than males,and males were almost twice as likely to have higher self-esteem than females(McCreary, 1993a, 1993b).Physical changes during puberty have also been reported to be moredistressing to females than to males (Schonert-Reichl & Offer, 1992). Of theVancouver and Victoria adolescents surveyed, 76% and 83%, respectively,measured within an average / healthy body mass ratio (McCreary, 1993a,1993b). However, females were two to three times more likely to be dissatisfiedwith their weight than males (McCreary, 1993a, 1993b). Given the supportingdata from other adolescent surveys in the Vancouver and Victoria population(McCreary 1993a, 1993b), it is likely that the differences in depressionprevalence among the females and males in this study are a result of actualdifferences in the population rather than instrument selection.With regard to the second relationship of this ANOVA correlationalcomparison, the analysis also revealed a significant relationship between thepresence of pets in the home and lower levels of depression among the subjects(p = 0.014). Such a result is consistent with the reports in the anecdotal reviewsindicating improved emotional well-being among those children and adolescentswith household pets (Albert & Bulcroft, 1988; Bibby & Posterski, 1992; Covert etal., 1985; Davis & Juhasz, 1985; Paul & Serpell, 1992a; Poresky, 1990; Robin &ten Bensel, 1990). The relationship between the presence of household pets andlower levels of depression in adolescents found in this study is precisely the6 4opposite of that found by Stallones et al. (1990) in a population of adultsresponding to the same instrument. It is likely that the reason for thisdiscrepancy is the reactive nature of depression in adolescents as compared tothe endogenous nature of adult depression (Simeon, 1989). As a result, pets inadolescent households may be viewed in the context of an emotional copingmediator as described in the conceptual framework.The third and final portion of this ANOVA analysis indicates that there isno significant difference in the interrelatedness among gender, pets and level ofdepression for the sampled subjects (p = 0.172). Males with pets scored thelowest on the CES-D instrument (indicating an absence of depressive symptoms)followed by males without pets, females with pets, and then finally femaleswithout pets. Despite the non-significance of these findings, it quite plausible thatmales and females utilize the presence of pets as emotional mediators indifferent ways.Question Three: Relationship Between Pet Ownership and Depression The third research question was answered using a sub-set of the overallCES-D scores for those 219 individuals who had pets. The ownership variablehas four distinct categories of family, parents, sibling, and self. There is asignificant relationship (p = 0.023) between pet ownership and lower levels ofdepression among the subjects. Post hoc Tukey-Kramer analyses of the findingsreveal the basis for this conclusion to be in the comparison between thosesubjects who owned their own pet and those subjects who were part of a familyownership of the household pet. Individuals who had sole ownership of thehousehold pet were significantly more depressed than those who were part of ashared family ownership (p = 0.028). There was no significant difference in the6 5depression scores for those whose household pets belonged to parents orsiblings as compared to either sole ownership or family ownership.The interpretation of these results involves addressing the value systemsof the adolescent population. Bibby and Posterski (1992) have established that"values that incite the spirit of individualism dominate what is important to youngpeople" (p. 289). The valued goals of "friendship" and "being loved" havedropped behind "freedom" in comparable Canadian teen polls of 1984 and 1992.In addition, the percentage of young people who value "working hard" hasdecreased from 69% in 1984 to 49% in 1992 (Bibby & Posterski, 1992). Giventhis value system, the findings become clearer. Of all those subjects with pets,those who had primary ownership of the pet were the most depressed and thosewith familial ownership were the least depressed. Values of adolescents havechanged since Levinson (1970) wrote "caring for a pet provides an opportunity forthe child to toughen his ego...acceptance of responsibility for the care of a pet willeventually lead to acceptance of responsibility for establishing meaningful,satisfying human relationships" (p. 1763). It is possible that the freedom fromtotal responsibility for the care of the pet, yet the recognition of shared ownershipof the pet within the family was the most desirable situation for the mental healthof the subjects.Indeed, Stallones et al. (1990) caution that pet ownership may createanother stressor for a depressed person rather than a facilitator to decreaseemotional distress. Such was the case with the subject who responded to thequestions of type of pet, pet name, primary ownership, reason for acquisition,and attachment with the following, "mouses...male: Hugs Eternity and female:Kisses Obsession....me....I begged for it and it was a big mistake!^ notattached" (subject #217). The names given to these mice imply that there mayhave been some behavioural expectations of affection from the mice; however, at6 6the time of the study, these expectations were clearly not being met. This couldhave been a result of the responsibility involved with caring for such pets.Considering the fact that the majority of subjects' pets were givenrecognizable human names, part of their role could be interpreted to be a "familymember" which is common for pets in urban settings (Albert & Bulcroft, 1988;Fogel, 1983). As a "family member" with joint ownership by all members of thefamily, the pet in this circumstance is in a pivotal position to act as a vector forintrafamilial communication (Albert & Bulcroft, 1988; Schaufl & Bergler, 1992),and as a buffer or recipient of emotions (Cain, 1991). Pets that are jointly ownedcan represent a shared family interest (Paul & Serpell, 1992b; Schaufl & Bergler,1992), thereby allowing adolescents to have the security of the family unit withoutbeing tied to the family as they strive for independence. As indicated by Robinand ten Bensel (1990), the pet is the ideal transitional object for the adolescent.These findings indicate the inappropriateness of implying that obtainingpets for depressed adolescents may help to "cheer them up," as the adolescentsin this study who owned their own pet were the most depressed among thosewith household pets. Rather, the usefulness of obtaining a pet for the familywhen one of its adolescents is depressed may be worthy of consideration. Itshould be noted, however, that several authors caution against suggesting petsto severely disturbed families, as the pet may be harmed or threatened as anabusive strategy directed at the child (Levinson, 1969; Robin, 1984; Robin, tenBensel, Quigley, & Anderson, 1981). Although having pets in the home is morehighly correlated with an absence of depression than not having pets, the leastfavoured situation regarding adolescent depression appears to be for theadolescent to own the pet.6 7Question Four: Relationship Between Pet Attachment and Depression The final research question was also answered using a sub-set of theoverall CES-D scores for the 218 individuals who had pets. This variable has fivedistinct categories of not attached, somewhat attached, attached, very attached,and extremely attached.The analysis indicates that depression levels are lower with increasedperceived attachment to the household pet; however, the relationship is notstatistically significant. This finding is difficult to interpret. First, one mustassume that the report of the subjects' perceived level of attachment (on a Liken-type scale from 1 to 5) adequately represents their actual perception ofattachment. It is important to note that this report does not attempt to representactual attachment as it is recognized in the literature as being a highly complexconcept (Johnson, Garrity & Stallones, 1992; Stallones et al., 1990; Zasloff,1992).The absence of any significant difference between those who reportedbeing extremely attached to their pets and those who reported not being attachedto their pets has provided an alternate view of the concept of attachment to thatreported by other researchers (Friedmann, 1990; Melson & Taylor, 1990;Michaels, 1989). The results receive support from the work of Albert and Bulcroft(1988) who report, "Pet ownership is particularly high among families withgrammar-school-age and teenage children. Yet, attachment to pets is relativelylow during this period." (p. 550). Kidd and Kidd (1990b) report that high schoolstudents indicate greater love for their pets than do elementary school students.However, the constructs of love and attachment can be only loosely compared.Similar to the results of this study, Zasloff (1992) determined no difference indegree of loneliness in owners according to attachment to pets.6 8These findings could be indicative of the mediating role which pets mayprovide in families with adolescents. It may be possible that adolescents receiveemotional support from sources outside the family system (Bibby & Posterski,1992) and, therefore, the household pet is used to meet intrafamilialcommunication needs rather than individual emotional needs. In the context ofthe conceptual framework, it is possible that the pet may serve as a copingmediator for emotional reappraisal through processes such as emotionalprojection, and displacement. The adolescent could utilize the pet (which, forexample, s/he may hate to walk or groom) as an outlet for frustration and angers/he may have for authority figures and/or peers. By expressing negativeemotions through the pet, the adolescent avoids internalizing these feelings aswell as any potential social consequences should s/he express these feelings tothe source of the conflict. These mechanisms of pet utilization would coincidewith the adolescent coping methods of confrontation, avoidance, distancing, andself-control discussed by Puskar and Lamb (1991). Subject #217 who was citedearlier as not being attached to the mice she begged to acquire, obtained a CES-D score of 4 (no depression). It is important to note that correlation does notimply causation and these results could simply be due to the fact that non-depressed individuals tend to acquire pets and whether they are attached tothose pets is irrelevant.SummaryThe sample in this study was equally divided between males and females,multicultural in nature, generally born in Canada, and was living at home with twoparents and some siblings. The subjects were either average or high-average intheir scholastic abilities. Pet owners and non-pet owners were equallyrepresented in the sample. Female subjects were more depressed than males,and non-pet owners were more depressed than pet owners. Subjects who6 9owned their own pet were more depressed than those in homes where the petwas owned by the entire family. In addition, there was no relationship betweenperceived attachment to the pet and level of depression among the subjects.This suggests that the manner in which the subjects used their pets as emotionalcoping mediators occurred in the context of complex family dynamics and maynot necessarily have reflected a supportive bond or attachment betweenadolescent and pet.Chapter FiveSummary, Conclusions, and Implications for NursingSummaryThe purpose of this study was to describe the prevalence of depressionamong a group of adolescents and to examine the relationships betweendepression and presence of pets in the home, primary ownership of pets andperceived degree of attachment to the pets by the subjects. Subjects were 401grade nine males and females attending three Vancouver and three Victoriasecondary schools over the period from October 1992 through January 1993.The instruments utilized to gather the data where the Center for EpidemiologicStudies Depression Scale (CES-D) (Radloff, 1977) and the Adolescent and PetCharacteristics Questionnaire, developed by the researcher. The subjectscompleted the instruments anonymously in classroom settings. The data weretransformed using logarithmic 10 (x+1) data transformation techniques to ensurenormality and linearity. The transformed data were analyzed using the SYSTAT5.01 computer programming for two-way and one-way fixed effects analysis ofvariance (ANOVA) and subsequent Tukey-Kramer post hoc analysis of pairwisecomparisons as necessary.The majority of the subjects surveyed were fourteen years old, hadaverage or high-average school performance, and generally lived at home withtwo parents and one to three siblings. The sample was definitely multicultural inits representation. Most of the subjects were born in Canada; however,approximately two-thirds of the subjects reported having parents who were notborn in Canada. Indeed, the sampled group included families which originated inforty-six different countries. Only half of the subjects reported English as theprimary language used at home. Twenty-seven different languages were7071reported to be used by the survey subjects. Just over half of the subjects hadhousehold pets at the time of the survey.Approximately two-thirds of those having household pets reported owningdogs or cats, with the remaining third of the subjects owning a variety of birds,fish, reptiles, amphibians and small mammals. Approximately half of the subjectshad human names for their pets and had had their pets for only three years orless. Most of the subjects acquired their pets for no particular reason. Three-quarters of the subjects reported either themselves or their whole family as theprimary pet owners. Finally, approximately half of the subjects felt they were veryor extremely attached to their pets.Of the Vancouver and Victoria grade nine students sampled, 48.4%demonstrated no depressive symptoms, 41.4% demonstrated what wereclassified as mild symptoms of depression, 8.7% demonstrated moderatesymptoms of depression, and 1.5% of those sampled indicated severedepressive symptoms. Findings indicated that female adolescents weresignificantly more depressed than their male counterparts (p = 0.001) with meanscores of 18.7 and 15.1, respectively.The findings also revealed that those subjects with pets in the home weresignificantly less depressed than their counterparts without household pets (p =0.014). There was no significant relationship between gender and pet ownershipcombined with level of depression (p = 0.172). An analysis of primary petownership indicated that subjects who owned their own pets were significantlymore depressed on a whole than those in homes where the pet was owned bythe entire family (p = 0.028). Perceived degree of attachment to the householdpet showed no relationship with depression (p = 0.725).7 2ConclusionsThe results of this study indicate the following five conclusions. The gradenine adolescents in this study demonstrated depression prevalence ratescomparable to those of other Canadian adolescents. Symptoms of depressionwere much higher among female subjects than among males. The presence ofpets in the household was directly related to decreased prevalence ofdepression. The strongest relationship between household pets and absence ofadolescent depression appears to occur when the pet is owned and cared for bythe entire family. The conceptual coping framework as proposed by Folkmanand Lazarus (1988) was useful in demonstrating the process by which thepresence of pets may influence the outcome measurement of depression in thissample of adolescents.Implications for NursingThe Canadian Nurses Association (1991) has officially recognized thatthere is a lack of information and resources for mental health promotion andillness prevention particularly regarding conditions encountered in adolescence.The Association's (1991) recommendations for reform include amongst others to:a) encourage research on the value of natural support networks and self-helpprograms in caring for persons with emotional or mental health problems;b) develop proposals for the implementation of prevention programs;c) encourage research to identify risk factors and groups at risk by virtue ofeither psychosocial or biological events;d) support policies and programs that enable infants, children, and adolescentsto accomplish normal developmental tasks;7 3e) support programs that strengthen the ability of individuals, families, andcommunities to cope or successfully negotiate situational stressthroughout the human life cycle (p.28-29).These recommendations are consistent with the direction provided by Health andWelfare Canada (1988) which emphasizes a commitment to generating newinterdisciplinary knowledge in the search for mental illness prevention and mentalhealth promotion practices in enhancing people's capacity to cope with stress.As demonstrated in this study, the concept of human and pet interactionhas a positive relationship to the emotional health of urban adolescents. Theincorporation of pets as an environmental influence in the coping practices andhealth behaviours of individuals and families has implications for both nursingpractice and research.Nursing Practice Florence Nightingale (1860 / 1967) noted the positive effects of a pet owlon the health of those for whom she was providing nursing care. As indicated inthe literature review for this study, nurses have utilized a variety of differentpractice routes to include pets in the provision of health care. Most nursingmodels contain the components of person, health, environment, and nurse intheir configuration (Reihl & Roy, 1980). In most totality systems models fornursing, the pet assumes the position of an external system, stimulus, therapy,resource, or force within the model (McMahon, 1991). For those utilizing aperson-environment simultaneity paradigm such as that of Rogers (Reihl & Roy,1980), the pet would be viewed as "a part of the sphere of health generatingenergies, forces, and stressors...as a vital, often symbiotic or energizingcomponent of the individual's state of being" (McMahon, 1991). Regardless ofwhere the pet is viewed in the system, there exist no models for nursing that do7 4not have a place for the inclusion of pets and their potential influence inpromoting the health of individuals.In a country where more than 50% of the homes with children also havehousehold pets, the inclusion of pet relationships in the assessment of theemotional health of adolescents should occur at all phases of the nursingprocess. In order to complete a comprehensive view of the biological,psychological, social and spiritual aspects of any individual or family system, anacknowledgement of the potential influence of household pets should be includedwhen assessing coping behaviours, supports, and forces. This study hasdemonstrated that almost 10% of the adolescents surveyed had symptoms ofmoderate or severe depression, and that the presence of pets in the household isrelated to decreased prevalence of depression. This finding alone shouldencourage nurses to incorporate an assessment of pet presence, ownership andcharacteristics when caring for the adolescent client. By asking about pets in thehome, the nurse may open lines of communication into traditionally unspokenareas of support which the patient may view as important. This will facilitate theestablishment of rapport as well as identify potential coping mediators for theindividual in regard to emotional reappraisal.The planning of any nursing care must be based on individualizedassessment and directed towards client-specific interventions. Nurses candevise strategies to include clients' pets in the establishment of coping mediatorsfor selected adolescents at risk for mental health concerns. This practice wouldbe in keeping with theory-directed nursing practice, as well as representingmovement towards the new vision for innovative, consumer-focussed mentalhealth promotion and illness prevention care. The practice of including pets inthe mental health care of adolescents would likely also be welcomed by clientsand their families, the exceptions being those families with adolescents who were7 5already severely depressed (Stallones et al., 1990) or those families where theremay be situations of abuse (Robin, 1984; Robin et al., 1981). The findings of thisstudy also direct the nurse to incorporate the entire family when working with theadolescent and pet, as the least depressed adolescents were from thosehouseholds where pet ownership was an acknowledged family venture. It is alsoimportant to remember, however, that even those who were not particularlyattached to their pets, derived benefit from the animals in the home. Therefore,planned care may include the involvement of animals to which the client does noteven seem to be particularly attached.Once the plans are made for integration of pets into the mental health careof the adolescent, implementation should proceed with the objectivity andcontinual evaluation and reassessment required for any nursing intervention.The descriptive parameters of nursing care applications of human - pet - healthinteractions must be documented to add to the existing knowledge base aboutthis relationship. It is important that all interventions be client-focused and family-focused and not the result of a program designed to meet nursing's goals.Evaluation of the interventions should focus on the health or illnessparameters of the adolescent as well as identifying the role which the pet playedin any change in health status. The use of familiar conceptual frameworks suchas coping, social support, family systems, attachment, loss, pain, stressmanagement, pain management, and adaptation may assist by providingguidance as to the process by which the pet may influence the adolescent'shealth system. The use of a model for nursing and conceptual framework for therole of the pet within the adolescent's system will provide both direction fornursing care and facilitation of growth in pet-related coping behaviours for theclient. The systematic inclusion of pet assessment in the parameters of76individualized health care will add to the knowledge of pet and human healthinteractions upon which nursing care can be based.Nursing Research With respect to nursing research, this study is in keeping with therecommendations of the Canadian Nurses Association (1991) regarding furtherstudy into the value of natural support systems to promote mental health, theidentification of risk factors for mental illness, and the exploration into theestablishment of coping resources for individuals and families. The studyprovides empirical validation of much of the previous anecdotal nursing literature.The findings provide a base for further exploration into the relationship betweenpets and the emotional health of urban adolescents. The first of further studiescould be a replication study using an alternate instrument to measure depressionin the adolescent population. Such an exercise would validate the findings of thisstudy and the use of this tool with such a culturally diverse adolescent population.Qualitative studies could be undertaken to determine the meaning of petsfor adolescents, the specific role of pets as coping mediators, the role of pets infamilies with adolescents, and the role of pets in first generation Canadianfamilies with adolescent members.Other studies could involve pre-test, post-test control group designs offamilies with adolescent members who had acquired a family pet for the first time.This would help to determine causal relationships between pets and theemotional health of adolescents and/or families.Through the process of studying the prevalence of depression in urbanadolescents and pets as a potential coping mediator, this study hasdemonstrated that the emotional relationship between humans and animalswhich dates back 12,000 years is still active in regard to the mental health of77urban adolescents. The credence that Florence Nightingale gave to the benefitof pets in the care of the sick over 100 years ago has relevance for the futuredirection of mental health nursing. The relationship between people, pets, andhealth can have a more central role in health care in the future.7 8ReferencesAckerson, L., Dick, R.W., Manson, S., & Baron, A. (1990). Properties of theinventory to diagnose depression in American Indian adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 29(2), 601-607.Albert, A., & Bulcroft, A. K. (1988). Pets, families, and the life course. Journal ofMarriage and the Family, 50(2), 543-552.Allen, K.M. (1985). The human-animal bond. An annotated bibliography. NewYork: Norris.Allen, K.M., Blascovich, J., Tomaka, J., & Kelsey, R.M. (1991). Presence ofhuman friends and pet dogs as moderators of autonomic responses tostress in women. Journal of Personality and Social Psychology, 1(4),582-589.Anderson, W.P., Reid, C.M., & Jennings, G.L. (1992, July). Pet ownership and risk factors for cardiovascular disease. Paper presented at The 6thInternational Conference On Human Animal Interactions: Animals and Us,Montreal, Canada.Baron, P., & Perron, L.M. (1986). Sex differences in the Beck DepressionInventory scores of adolescents. Journal of Youth and Adolescence, 15(2),165-171.Baun, M.M., Bergstrom, N., Langston, N.F., & Thoma, L. (1984). Physiologicaleffects of human/companion animal bonding. Nursing Research, 33(3),126-129.Baun, M.M., Cardiello, F., & Jassen, J. (1992, July). The use of avian companionship to alleviate the depression, loneliness, and low morale during translocation of the older adult into a skilled rehabilitation unit. Paper presented at The 6th International Conference On Human AnimalInteractions: Animals and Us, Montreal, Canada.Baun, M.M., Oetting, K., & Bergstrom, N. (1991). Health benefits of companionanimals in relation to the physiologic indices of relaxation. Holistic NursinqPractice, 5(2), 16-23.Beck, A.M. (1983). Animals in the city. In A.H. Katcher & A.M. Beck (Eds.), Newperspectives on our lives with companion animals (pp. 237-243).Philadelphia: University of Pennsylvania Press.Beck, A.T., Hollon, S.D., Young, J.E., Bedrosian, R.C., & Budenz, D. (1985).Treatment of depression with cognitive therapy and amitriptyline. Archives of Psychiatry, 42(2), 142-148.Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). Aninstrument for measuring depression. Archives of Psychiatry, 4, 53-63.79Bibby, R.W., & Posterski, D.C. (1992). Teen trends: A nation in motion. Toronto:Stoddart.Bikales, G. (1975). Dog as significant other. Social Work, 20, 150-152.Blechman, E.A., McEnroe, M.J., Carella, E.T., & Audette, D.P. (1986). Childhoodcompetence and depression. Journal of Abnormal Psychology, 9_5(3), 223-227.Blenner, J.L. (1991). The therapeutic functions of companion animals in infertility.Holistic Nursing Practice,  5(2), 6-10.Blythe, A. (1980). Patients' best friends...Companion animals. Health and Social Service Journal, 90, 1479.Bossard, J.H. (1944). The mental hygiene of owning a dog. Mental Hygiene, 28,408-413.Brickel, C.M. (1979). The therapeutic roles of cat mascots with a hospital-basedgeriatric population: A staff survey. The Gerontologist, 19(4), 368-372.Brickel, C.M. (1980). A review of the roles of pet animals in psychotherapy andwith the elderly. Internat ional Journal of Aging and Human Development,12(2), 119-128.Brown, L.T., Shaw, T.G., & Kirkland, K.D. (1972). Affection for people as afunction of affection for dogs. Psychological Reports, 31, 957-958.Bustad, L.K., & Hines, L.M. (1984). Historical perspectives of the human-animalbond. In R.K. Anderson, B.L. Hart, & L.A. Hart (Eds.), The pet connection: Its influence on our health and quality of life (pp. 15-29). South St. PaulMN: Globe.Cain, A.O. (1983). A study of pets in the family system. In A.H. Katcher & A.M.Beck (Eds.), New perspectives on our lives with companion animals  (pp.72-81). Philadephia: University of Pennsylvania Press.Cain, A.O. (1991). Pets and the family. Holistic Nursing Practice,  .5(2), 58-63.Cameron, P. (1966). Pet ownership and sex as determinants of stated affecttoward others and estimates of others' regard of self. PsychologicalReports, 19, 884-886.Cameron, P., & Mattson, M. (1972). Psychological correlates of pet ownership.Psychological Reports,  30, 286-288.Canadian Nurses Association (1991). Mental health care reform: A priority fornurses. Ottawa: Author.Capital Regional District Health (1990). Children. youth and their families: Acapital regional district profile. Victoria: Author.Carbary, L.J. (1975). What are those animals doing on the health care team?Journal of Practical Nursing, 25, 28-30.80Carmack, B.J. (1988, September). A multidisciplinary model of stress / coping: The human animal bond. Paper presented at the Delta Society SeventhAnnual Conference, Orlando, FL.Carmack, B.J. (1991). The role of companion animals for persons with AIDS/HIV.Holistic Nursing Practice, 5(2), 24-31.Children and Youth At Risk Steering Committee (1991). Children and youth atrisk: Towards a mental health plan for metropolitan Vancouver. Vancouver: City of Vancouver, Social Planning DepartmentClarke, G. (1991, February). Adolescent Depression. Proceedings from theProfessional Achievement Systems Workshop. Vancouver.Clarke, G., Lewinsohn, P., & Hops, H. (1990). Adolescent coping with depressioncourse. Eugene, OR: Catalina.Cohen, S.P. (1991, November). Hugs that help. The health benefits of pets.Newsweek, p.4.Conrad, N. (1991). Where do they turn? Social support systems of suicidal highschool adolescents. Journal of Psychosocial Nursing and Mental Health,29(3), 15-20.Cookman, C.A. (1991, October). Filling the void: A study of attachment between elders and their pet dogs. Paper presented at the Delta Society TenthAnnual Conference, Portland, OR.Corson, S.A. (1977). Pet dogs as nonverbal communication links in hospitalpsychiatry. Comprehensive Psychiatry, 18(1), 61-72.Covert, A.M., Whiren, A.P., Keith, J., & Nelson, C. (1985). Pets, earlyadolescents, and families. In M.B. Sussman (Ed.), Pets and the family(Special issue) Marriage and Family Review, 8(3), 95-108.Davis, J.H. (1985a). Children and pets: A therapeutic connection. PediatricNursing, 11(5), 377-379.Davis, J.H. (1985b). Implications of the human-animal companion bond in thecommunity. Home Healthcare Nurse, a(6), 11-15.Davis, J.H., & Juhasz, A.M. (1985). The preadolescent/pet bond andpsychosocial development. In M.B. Sussman (Ed.), Pets and the family(Special issue) Marriage and Family Review, $(3), 79-93.Davis, L. (1986). Down on the farm. Nursing Times, 82(48), 40.Davis, S.J.M., & Valla, F. (1978). Evidence for domestication of the dog 12,000years ago in the Natufian of Israel. Nature, 276, 608-610.Dolan, M. (1982). Rx - Wet nose, soft fur, and wagging tail. Nursing, 12(10).81Draper, R.J., Gerber, G.J., & Layng, M. (1990). Defining the role of pet animals inpsychotherapy. Psychiatric Journal of the University of Ottawa, 15(3), 169-172.Ellickson, P.L., & Hawes, J.A. (1989). An assessment of active versus passivemethods for obtaining parental consent. Evaluation Review,  no ), 45-55.Endenburg, N., Hart, H. & de Vries, H.W. (1990). Differences between ownersand nonowners of companion animals. Anthrozoos, 4(2), 120-126.Fila, D. (1991). The significance of companion animals to a geriatric vascularpatient: A case study. Holistic Nursing Practice, 5(2), 11-15.Fogel, B. (1983). Pets and their people. Markham, Ontario: Penguin BooksCanada.Folkman, S., & Lazarus, R.S. (1988). Coping as a mediator of emotion. Journal ofPersonality and Social Psychology, 51(3), 466-475.Francis, G.M. (1981). The therapeutic use of pets. Nursing Outlook, 2(6), 369-370.Francis, G.M. (1991). Here come the puppies: The power of the human-animalbond. Holistic Nursing Practice, 5(2), 38-41.Francis, G.M., Turner, J.T., & Johnson, S.B. (1985). Domestic animal visitationas therapy with adult home residents. International Journal of NursingStudies, 22(3), 201-206.Friedmann, E. (1990). The value of pets for health and recovery. In I.H. Burger(Ed.), Pets, benefits and practice (pp. 8-17). Waltham Symposium 20,distributed in conjunction with the Journal of Small Animal Practice.Friedmann, E., Katcher, A., Lynch, J.J., & Thomas, S.A. (1980). Animalcompanions and one-year survival of patients after discharge from acoronary care unit. Public Health Reports, 95(4), 307-312.Furstenberg, F.F., Rhodes, P.S., & Powell, S.K. (1988). Short-term value of pets.American Journal of Nursing, 88(2), 157.Gage, M.G. (1992, July). Companion-dog interaction and adult competence in marital and family roles: A life span analysis. Paper presented at The 6thInternational Conference On Human Animal Interactions: Animals and Us,Montreal, Canada.Gammonoley, J., & Yates, J. (1991). Pet projects, Animal assisted therapy innursing homes. Journal of Gerontological Nursing, 17(1), 12-15.Glass, G.V., & Hopkins, K.D. (1984). Statistical methods in education andpsychology (2nd ed.). Toronto: Prentice-Hall Canada.Greydanus, D.E. (1986). Depression in adolescence. A perspective. Journal ofAdolescent Health Care, 7(6S), 109S-120S.8 2Haggard, A. (1985). A patient's best friend. American Journal of Nursing, $5(12),1374-1376.Health and Welfare Canada (1988). Mental health for Canadians: Striking abalance. Ottawa: Minister of Supply and Services Canada.Humeston, B. (1983). How pets help people cope. Better Homes and Gardens,61(8), 61-66.Hundley, J. (1991). Pet project: The use of pet facilitated therapy among thechronically mentally ill. Journal of Psychosocial Nursing, 29(6), 23-26.Jaworski, M. (1990). Canine companions. Family Circle, 103(4),9.Johnson, D. (1980). The behavioral system model for nursing. In J.P. Riehl & C.Roy (Eds.), Conceptual Models for Nursing Practice  (2nd ed.), (pp. 207-216). New York: Appleton-Century-Crofts.Johnson, T.P., Garrity, T.F., & Stallones, L. (1992). Psychometric evaluation ofthe Lexington Attachment to Pets Scale (LAPS). Anthrozoos, 5(3), 160-175.Kale, M. (1992). Teens at risk: Working with animals to create a new self-image.InterActions, 10(4), 6-9.Kalfon, E. (1991). Pets make a difference in long term care. Perspectives. Journal of the Gerontological Nursing Association, 15(4), 3-6.Katcher, A.H., & Beck, A.M. (1988). Health and caring for living things. In A.N.Rowan (Ed.), Animals and people sharing the world (pp. 53-73). London:University Press of New England.Kidd, A.H., & Kidd, R.M. (1990a). Factors in children's attitudes toward pets.Psychological Reports, 66, 775-786.Kidd, R.M., & Kidd, A.H. (1990b). High school students and their pets.Psychological Reports, 0, 1391-1394.Klerman, G.L., & Weissman, M.M. (1989). Increasing rates of depression.Journal of the American Medical Association, 261(15), 2229-2235.Kramer, A.D., & Feiguine, B.A. (1981). Clinical effects of amitriptyline inadolescent depression. A pilot study. Journal of the American Academy ofChild Psychiatry, 20, 636-644.Lago, D., & Knight, B. (1987). Companion animals for the elderly. In P. Arkow(Ed.), The loving bond: Companion animals in the helping professions (pp.211-228). Saratoga, CA: R & E.Lamb, J. (1992a, April 11). Classroom survey an intimate probe. The WeekendSun, pp. A1, A6.Lamb, J. (1992b, April 16). School survey on intimate matters threatens teens'privacy. The Vancouver Sun, pp. B1, B6.83Lane, T.J., & De Gale, S. (1992, July). Single senior citizens / Pet ownership: Apreliminary study. Paper presented at The 6th International ConferenceOn Human Animal Interactions: Animals and Us, Montreal, Canada.Lapierre, Y.D. (1988, February). What causes depression?  Paper presented atcontinuing medical education conference: Treatment of Depression,Whistler, B.C., Canada.Lebeck, S. (1992). Healing without words: The effects of animals on depressedpopulations in psychiatric hospitals. Latham Letter, 13(2), 1-2.Lee, D. (1987). Companion animals in institutions. In P. Arkow (Ed.), The lovingbond: Companion animals in the helping professions (pp. 229-236).Saratoga, CA: R & E.Levinson, B.M. (1962). The dog as a "co-therapist." Mental Hygiene, 46(1), 59-65.Levinson, B.M. (1965). Pet psychotherapy: Use of household pets in thetreatment of behavior disorder in childhood. Psychological Reports, 17_,695-698.Levinson, B.M. (1970). Pets, child development, and mental illness. The Journal of the American Veterinary Association, 157(11), 1759-1766.Levinson, B.M. (1972). Pets and human development.  Illinois: Saunders.MacLean, G. (1987). Recent advances in child and adolescent psychiatry inCanada. Psychiatric Journal of the University of Ottawa, 12(3), 144-149.Manor, W.A. (1991). Alzheimer's patients and their caregivers: The role of thehuman-animal bond. Holistic Nursing Practice, 5(2), 32-37.Manor, W.A., McCurdy, J., & Crain, H.C. (1988). Pets visit patients. American Journal of Nursing, 88(8), 1061-1062.Manson, S., Ackerson, L.M., Dick, R.W., Baron, A.E., & Fleming, C.M. (in press).Depressive symptoms among American Indian adolescents: Psychometriccharacteristics of the CES-D. Psychological Assessment.Mayhew, P.B. (1988). No place for a dog? The Canadian Nurse/L'infirmiere canadienne, 5, 28-29.McCreary Centre Society (1993a). Adolescent health survey: Report for the Capital Region of British Columbia.  Vancouver: Author.McCreary Centre Society (1993b). Adolescent health survey: Report for the Vancouver Region of British Columbia. Vancouver: Author.McCulloch, M.J. (1981). The pet as prosthesis - Defining criteria for theadjunctive use of companion animals in the treatment of medically ill,depressed outpatients. In B.Fogel (Ed.), Interrelations between people andpets (pp. 101-123). Springfield: Charles C. Thomas.84McDermott, R.J., Hawkins, W.E., Marty, P.J., Littlefield, E.A., Murray, S., &Williams, T.K. (1990). Health behavior correlates of depression in asample of high school students. Journal of School Health, fiQ(8), 414-417.McMahon, S. (1991). The quest for synthesis: Human-companion animalrelationships and nursing theories. Holistic Nursing Practice, 5(2), 1-5.Mead, A. (1992, July). Prisons - therapeutic benefits of birds. Paper presented atThe 6th International Conference On Human Animal Interactions: Animalsand Us, Montreal, Canada.Meer. J. (1984). Pet theories. Psychology Today, 18(8), 60-64.Melson, G., & Taylor, S. (1990, October). Pet ownership and attachment in young children: Relations to behavior problems and social competence. Paperpresented at the Ninth Annual Delta Society Conference, Houston, TX.Messant, P.R. (1983). Social facilitation of contact with other people by pet dogs.In A.H. Katcher & A.M. Beck (Eds.), New perspectives on our lives withcompanion animals (pp. 37-46). Philadelphia: University of PennsylvaniaPress.Messant, P.R., & Serpell, J.A. (1980). An historical and biological view of the pet-owner bond. In B. Fogel (Ed.), Interrelations between people and pets (pp.5-22). Springfield: Charles C. Thomas.Michaels, Y. (1989, November). The relationship between pet attachment and self-esteem in latency age children. Paper presented at the Delta SocietyEighth Annual Conference, Parsippany, NJ.Moneymaker, J.M., & Strimple, E.O. (1991). Animals and inmates: A sharingcompanionship behind bars. Journal of Offender Rehabilitation, 16(3/4),133-152.Mugford, R.A., & M'Comisky, J.G. (1975). Some recent work on thepsychotherapeutic value of caged birds with old people. In R.S. Anderson(Ed.), Pet animals and society (pp. 54-65). London: Baillere Tindall.Munro, B.H., Visintainer, M.A., & Page, E.B. (1986). Statistical methods for healthcare research. Philadelphia: Lippincott.Murphy, G.E., Simons, A.D., Wetzel, R.D., & Lustman, P.J. (1984). Cognitivetherapy and pharmacotherapy. Archives of General Psychiatry, 41(1), 33-44.Muschel, I.J., (1984). Pet therapy with terminal cancer patients. The Journal ofContemporary Social Work, 65(8), 451-458.Nightingale, F. (1967). Notes on nursing. New York: Dover. (Original workpublished in 1860).Odam, J. (1992, April 14). Sex and drugs survey ignites parents: School boardswamped with complaints. The Vancouver Sun, pp. B1.85Odendaal, J.S.J., & Weyers, A. (1990). A cross-cultural study of human-companion animal relationship. South African Journal of Sociology,  21(1),21-30.Offer, D., & Schonert-Reichl, K.A. (1992). Debunking the myths of adolescence:Findings from recent research. Journal of the American Academy of Child and Adolescent Psychiatry, 3_1(6), 1003-1014.O'Hara, J. (1984, March). The agony of depression. Maclean's, pp. 46-52.Ormerod, E. (1992, July). Animals in Scottish prisons. Paper presented at The6th International Conference On Human Animal Interactions: Animals andUs, Montreal, Canada.Paul, E.S., & Serpell, J.A. (1992a, July). Childhood pets and adult attitudes towards animals and people. Paper presented at The 6th InternationalConference On Human Animal Interactions: Animals and Us, Montreal,Canada.Paul, E.S., & Serpell, J.A. (1992b). Why children keep pets: The influence of childand family charcteristics. Anthrozoos, 5(4), 231-244.Pekar, J. (1992, July). Unconditional love in the person pet relationship. Paperpresented at The 6th International Conference On Human AnimalInteractions: Animals and Us, Montreal, Canada.Perko, J.E., & Kreigh, H.Z. (1988). Psychiatric and mental health nursing (3rded.). Toronto: Prentice-Hall Canada.Phineas, C. (1974). Household pets and urban alienation. Journal of Social History, 7, 338-343.Poresky, R.H. (1990). The young children's empathy measure: Reliability,validity, and effects of companion animal bonding. Psychological Reports,66, 931-936.Preshlock, K. (1985). Brandy: An incentive to recover. Home Healthcare Nurse,3(6), 16-17.Pronovost, J., Cote, L., & Ross, C. (1990). Epidemiologic study of suicidalbehaviour among secondary-school students. Canada's Mental Health,38(1), 9-15.Puig-Antich, J., Pere!, J.M., Lupatkin, W., Chambers, W.J., Tabrizi, M.A., King, J.,Goetz, R., Davies, M., & Stiller, R.L. (1987). Imipramine in prepubertalmajor depressive disorders. Archives of General Psychiatry, 44, 81-89.Puskar, K., & Lamb, J. (1991). Life events, problems, stresses, and copingmethods of adolescents. Issues in Mental Health Nursing, 12(3), 267-281.Radloff, L.S. (1977). A CES-D Scale: A self report depression scale for researchin the general population. Applied Psychological Measurement,  1, 385-401.86Reinherz, H.Z., Frost, A.K., & Pakiz, B. (1991). Changing faces: Correlates ofdepressive symptoms in late adolescence. Family and Community Health,14(3), 52-63.Riddick, C.C. (1985). Health, aquariums and the non-institutionalized elderly. InM.B. Sussman (Ed.), Pets and the family (Special issue) Marriage and Family Review, 8(2), 163-174.Riehl, J.P., & Roy, C. (Eds.). (1980). Conceptual models for nursing practice (2nded.). New York: Appleton-Century-Crofts.Ritvo, H. (1988). The emergence of modern pet keeping. In A.N. Rowan (Ed.),Animals and people sharing the world (pp. 13-31). London: UniversityPress of New England.Robb, S.S., Boyd, M., & Pristash, C.L. (1980). A wine bottle, plant, and puppy.Catalysts for social behavior. Journal of Gerontological Nursing, C12),721-728.Robb, S.S., & Stegman, C.E. (1983). Companion animals and elderly people: Achallenge for evaluators of social support. The Gerontologist, 2a(3), 277-282.Roberts, R., Andrews, J., Lewinsohn, P.M., & Hopps, H. (1991). Assessment ofdepression in adolescents using the Center for Epidemiologic Studies -Depression Scale (CES-D). Journal of Consulting and Clinical Psychology(in press).Robin, M. (1984). Abused children and their pets. In R.K. Anderson, B.L. Hart, &L.A. Hart (Eds.). The pet connection: Its influence on our health and quality of life (pp. 111-117). South St. Paul MN: Globe.Robin, M., & ten Bensel, R. (1990). Pets and socialization of children. TheLatham Letter, 11(2), 1, 17-23.Robin, M., ten Bensel, R., Quigley, J.S., & Anderson, R.K. (1981). Childhood petsand the psychosocial development of adolescents. In A.M. Katcher & A.M.Beck (Eds.), New perspectives on our lives with companion animals (pp.436-443). Philadelphia: University of Pennsylvania Press.Robins, D.M., Sanders, C.R., & Cahill, S.E. (1991). Dogs and their people: Petfacilitated interaction in a public setting. Journal of ContemporaryEthnography, 20(1), 3-25.Rogers, R.R. (1987). Assessment of depression in adolescent patients.Psychiatric Journal of the University of Ottawa, 12(1), 4-8.Rossbach, K.A., & Wilson, J.P. (1992). Does a dog's presence make a personappear more likeable? Two studies. Anthrozoos, 5(1), 40-51.Rowan, A.N. (Ed.). (1992a). New publications. Anthrozoos, 5(3), 205.Rowan, A.N. (Ed.). (1992b). Pet populations. Anthrozoos, 5(1), 57.87Ruckert, J. (1987). The four footed therapist: How your pet can help you solveyour problems. Berkeley: Ten Speed Press.Rumel, D., & Costanzo, G. (1992). The general index of health for the city ofVancouver. Canadian Journal of Public Health, 11%1), 82-83.Schaufl, A., & Bergler, R. (1992, July). Children and dogs. Paper presented atThe 6th International Conference On Human Animal Interactions: Animalsand Us, Montreal, Canada.Schonert-Reichl, K.A., & Offer, D. (1992). Gender differences in adolescentsymptoms. Advances in Clinical Child Psychiatry, 14, 27-60.Serpell, J.A. (1988). Pet-keeping in non-western societies: Some popularmisconceptions. In A.N. Rowan (Ed.), Animals and people sharing theworld (pp. 33-52). London: University Press of New England.Serpell, J.A. (1990). Evidence for long term effects of pet ownership on humanhealth. In I.H. Burger (Ed.), Pets. benefits and practice (pp. 1-7).Waltham Symposium 20, distributed in conjunction with the Journal ofSmall Animal Practice.Serpell, J.A. (1991). Beneficial effects of pet ownership on some aspects ofhuman health and behaviour. Journal of the Royal Society of Medicine, 84,717-720.Siegel, J.M. (1990). Stressful life events and use of physician services among theelderly: The moderating role of pet ownership. Journal of Personality and Social Psychology, 58(6), 1081-1086.Siegel, L.J., & Griffin, N.J. (1984). Correlates of depressive symptoms inadolescents. Journal of Youth and Adolescence, 13(6), 475-487.Simeon, J.G. (1989). Depressive disorders in children and adolescents.Psychiatric Journal of the University of Ottawa, 14(2), 356-361.Schultz, C. (1990). Peanuts. United Feature Syndicate, Inc.Stallones, L., Marx, M.B., Garrity, T.F., & Johnson, T.P. (1990). Pet ownershipand attachment in relation to the health of U.S. adults, 21 to 64 years ofage. Anthrozoos, 4(2), 100-112.Stevens, L.T. (1989, November). Attachment to pets among eighth graders. Paper presented at the Delta Society Eighth Annual Conference,Parsippany, NJ.Thompson, M., Kennedy, R.W., & lgou, S. (1983). Pets as socializing agents withchronic psychiatric patients: An initial study. In A.H. Katcher, A.M. Beck(Eds.), New perspectives on our lives with companion animals (pp.427-430). Philadelphia: University of Pennsylvania Press.Tonkin, R.S. (1984). Suicide methods in British Columbian adolescents. Journal of Adolescent Health Care, 5(3), 172-177.88Toufexis, A. (1987, March). Furry and Feathery Therapists. Time, pp.74.Truscott, B. (1992, September 23). VSB approves health survey: Controversialquestionnaire will be put to students, grades seven to twelve. The Vancouver Courier, pp. 11.Twiname, B.G. (1984). Having difficulty meeting your rehabilitation goals?Consider pet therapy. Rehabilitation Nursing, 2(6), 34-35.Ulrich, R.S., Simons, R.F., Losito, B.D., Fiorito, E., Miles, M.A., & Zelson, M.(1991). Stress recovery during exposure to natural and urbanenvironments. Journal of Environmental Psychology, 11(3), 201-230.Veevers, J.E. (1992, July). What is a pet? Defining the concept of companionanimal. Paper presented at The 6th International Conference On HumanAnimal Interactions: Animals and Us, Montreal, Canada.Weissman, M.M., Gammon, G.D., John, K., Merikangas, K.R., Warner, V.,Prusoff, B.A., & Sholomskas, D. (1987). Children of depressed parents.Increased psychopathology and early onset of major depression. Archivesof General Psychiatry, 44(10), 847-853.Wille, R. (1984). Therapeutic use of companion pets for neurologically impairedpatients. Journal of Neurosurgical Nursing,1Q(6),  323-325.Wilkinson, L. (1990). SYSTAT: The system for statistics. Evanston, IL: SYSTAT,Inc..Wolfe, J. (1977). The use of pets as transitional objects in adolescentinterpersonal functioning. Dissertation Abstracts International, 38, 2391 B.Wyatt, V. (1988). Pets. An amazing investigation. Toronto: Greey de PencierBooks.Zar, J.H. (1984). Biostatistical analysis (2nd ed.). Toronto: Prentice-Hall Canada,Inc.Zasloff, R. (1992, July). Friends, confidants. and companion animals: A study ofsocial support characteristics and psychological well-being among petowners and non-owners. Paper presented at The 6th InternationalConference On Human Animal Interactions: Animals and Us, Montreal,Canada.Appendix AAdolescent and Pet Characteristics Questionnaire8990Household Pets and Depression Among Urban AdolescentsAdolescent and Pet Characteristics QuestionnaireFor the following questions please circle the best option or write your answer in the spaceprovided. DO NOT PUT YOUR NAME ON THIS FORM.Personal CharacteristicsSex:^M^F Age: ^Average marks in school this year: A^B^C^D^FNumber of parents living with you: 1^2Number of brothers^ & sisters ^ living with you.Your position in the order of kids in your family (eg. oldest, youngest, second out of four, ninthout of ten):Country you were born in: ^Country your mother was born in:Country your father was born in: ^Number of years you have lived in Canada:^Number of years you have lived in your current neighbourhood:^Primary language spoken at home:^Do you have any pets in your home? ^Yes ^ No. (if no, then stop here).Pet CharacteristicsTypes of pets you currently have (and how many of each)dog(s)^ cat(s)^bird(s)fish reptile(s) insect(s)small mammal(s),(eg. hamster, gerbil, rabbit, rat) type:^other(s), type:^If you currently have more than one pet, for the following questions, answer in regard to yourfavorite pet:Type of pet: ^Name of pet:^Main owner of the pet:^ yourself^ parent(s)^ brother or sister^whole familyFor what occasion did you or your family get this pet? (eg. birthday, other pet died, brother lefthome, nothing special, other):How long you have had this pet:^Rate how close or emotionally attached you are to your favorite pet:1^ 2^3^4^5not attached^somewhat attached^attached^very attached extremely attachedAppendix BCenter for Epidemiologic Studies Depression Scale9192Household Pets and Depression Among Urban AdolescentsCenter for Epidemiological Studies Depression ScaleFor the following statements, circle the number for each statement which best describes howoften you felt this way DURING THE PAST WEEK.DURING THE PAST WEEK:1. I was bothered by things that usuallyRarely ornone ofthe timeSome or alittle ofthe timeOccasionally^Most oror a moderate all theamt. of time^timedon't bother me. 0 1 2 32. I did not feel like eating;my appetite was poor. 0 1 2 33.^I felt that I could not shake off the blueseven with help from my family or friends. 0 1 2 34.^I felt that I was just as good asother people. 3 2 1 05.^I had trouble keeping my mind on whatI was doing. 0 1 2 36.^I felt depressed. 0 1 2 37.^I felt that everything I did was an effort. 0 1 2 38.^I felt hopeful about the future. 3 2 1 09.^I thought life had been a failure. 0 1 2 310.^I felt fearful. 0 1 2 311. My sleep was restless. 0 1 2 312. I was happy. 3 2 1 013. I talked less than usual. 0 1 2 314.^I felt lonely. 0 1 2 315. People were unfriendly. 0 1 2 316.^I enjoyed life. 3 2 1 017.^I had crying spells. 0 1 2 318.^I felt sad. 0 1 2 319. I felt that people disliked me. 0 1 2 320. I could not get "going". 0 1 2 3Appendix CLetter of School Consent and Consent Form93THE UNIVERSITY OF BRITISH COLUMBIA 94School of NursingT. 206-2211 Wesbrook MallVancouver, B.C. Canada V6T 2B5Fax:(604) 822-7466Household Pets and Depression Among Urban AdolescentsDearYour consent is requested to enable a sample of grade nine students from your school toparticipate in a brief questionnaire survey to take place within school hours. This survey hasbeen approved by the Vancouver and Victoria School Boards. It addresses the important issueof depression among our adolescent population. Depression is known to be associated withincreased risks of dropping out of school, substance abuse, delinquency, and suicide.This survey will anonymously survey grade nine students in regard to certaindemographic characteristics (the presence of household pets being a major focus) in addition tomeasuring their current level of depression (if any). The survey takes approximately 10 minutesand would be ideally administered by myself in a pre-arranged portion of class time.(Homeroom, physical education, or health classes are preferable to increase the likelihood ofstudent attendance.)As the questionnaire is anonymous and will be collected by myself, student confidentialitywill be ensured. The analysis of this data may potentially reveal situations of a severelydepressed student(s). Should this be the case in your school, I will contact you with thisinformation as well as resource referrals for mental health education and consultation for yourstudents as a group.The completion of this project will contribute greatly to the understanding of the incidenceand characteristics of adolescent depression among grade nine students in selected Vancouverand Victoria high schools. I thank you in advance for your consideration and support in thisregard. Should you wish to give your consent, please sign the attached form and retain this pagefor your records. I will contact you by telephone within the next couple of weeks to arrange atime to meet with you and answer any questions you may have and pick up your consent form.Please do not hesitate to call me at the University of British Columbia, School of Nursing (822-7486). Thank you once again, I look forward to meeting you.Respectfully,Molly Nevin-Haas, R.N., B.S.N.Graduate StudentDr. Marilyn Willman, RN., Ph.D. & Gloria Joachim, R.N., M.S.N.Faculty Advisors, School of NursingUniversity of British Columbia (822-7417).^ page 1 of 2THE UNIVERSITY OF BRITISH COLUMBIA 95School of NursingT. 206-2211 Wesbrook MallVancouver, B.C. Canada V6T 2B5Fax: (604) 822-7466Household Pets and Depression Among Urban AdolescentsA questionnaire survey by Molly Nevin-HaasSigned School ConsentI have read the attached information sheet and hereby give / do not give ( please circle )my consent to have the above study take place at^ School. I understand thatindividual student and parental consent will be obtained by the researcher prior to surveying thestudents selected. I also understand that I may withdraw permission for this school to participatein the study at any time before or during the course of the sampling without experiencing anynegative ramifications from the researcher.Signed,(print name)(position)(date)page 2 of 2Appendix DLetter of Parental Informed Consent and Consent Form96THE UNIVERSITY OF BRITISH COLUMBIA g7School of NursingT. 206-2211 Wesbrook MallVancouver, B.C. Canada V6T 2B5Fax: (604) 822-7466Household Pets and Depression Among Urban AdolescentsLetter of Informed ConsentDear Parent:Your consent is requested in enabling your daughter or son to participate in a briefquestionnaire survey to take place at their high school.This survey has been approved by the Vancouver and Victoria School Boards andaddresses the important topic of teenage depression. It will determine the level of thisoccurrence among our young people as well as identify its relationship with a selection ofindividual characteristics. The presence of pets in the home will be the main focus of this survey.However, even if you have no pets in your household your child's opinion is important to thisstudy.Please be aware that the questionnaire is completely anonymous. Your child's namewill not appear on the questionnaire. The entire survey will take no longer than 10 minutesand will be completed during an allotted portion of school time. Should you choose not to allowyour child to participate in the questionnaire survey, there will be no impact on your child'seducation or treatment in class. Should you give your permission for your son or daughter toparticipate in the survey, your child will also be given an opportunity on the day of the survey toconsent to participate. Either decision you or your child makes will not affect his or her educationor treatment in class.Please sign the attached consent and have your son / daughter return it to school.Alternately send a note indicating your consent with your child to school. Retain this sheet foryour information. If you have any questions regarding this study, please do not hesitate to callme at The School of Nursing, University of British Columbia, (822-7486).Thank you for returning this signed form so promptly!Sincerely,Molly Nevin-Haas, R.N., B.S.N.Graduate StudentDr. M. Willman, R.N., Ph.D. & G. Joachim, R.N., M.S.N.Faculty AdvisorsU.B.C., School of Nursing (822-7417)^page 1 of 2THE UNIVERSITY OF BRITISH COLUMBIA 98School of NursingT. 206-2211 Wesbrook MallVancouver, B.C. Canada V6T 2B5Fax:(604) 822-7466Household Pets and Depression Among Urban AdolescentsA questionnaire survey by Molly Nevin-HaasSigned Parental ConsentI have read the attached information sheet describing this study. I give / do notgive (please circle your choice) permission for my child, ^  toparticipate in the study, Household Pets and Depression Among Urban Adolescents. Iunderstand that all results of the study are entirely anonymous. I also understand that I maywithdraw consent for my child to participate in this study at any time before or during the courseof the survey without any negative consequences towards myself or my child.Signed,(Parent or Legal Guardian).(date)page 2 of 2Appendix EInformation and Instructions for Questionnaire99THE UNIVERSITY OF BRITISH COLUMBIA noSchool of NursingT. 206-2211 Wesbrook MallVancouver, B.C. Canada V6T 2B5Fax: (604) 822-7466Household Pets and Depression Among Urban AdolescentsInformation and Instructions for QuestionnaireThis questionnaire is part of a survey to study the relationship between personalcharacteristics (and for those with a pet, also pet characteristics) and teenage depression. Byfilling out this questionnaire it is assumed that you are agreeing to have your answers used in ourstudy. If you do not wish to participate, do not fill out the questionnaire. Either choice you makewill not effect your school standing in any way.DO NOT PUT YOUR NAME ON ANY OF THESE PAPERS. Please answer all thequestions as honestly as possible. There are no right or wrong answers. The questionnaireconsists of two pages. Please be sure to complete both of them. The questionnaire will takeapproximately 10 minutes. When you are finished please fold your papers in half to conceal yourresponses, and bring it to the envelope held by the researcher.Thank you very much for agreeing to participate! Your information will greatly help add towhat is known about teenage depression.Molly Nevin-Haas, R.N., B.S.N.Graduate Student (822-7486)Dr. Marilyn Willman, R.N., Ph.D. & Gloria Joachim, R.N., M.S.N.Faculty AdvisorsU.B.C. School of Nursing, (822-7417)Appendix FDensity Distribution of CES-D Scores101P4 0.25 -r=4 0.20 -ra.4 0.15 -00.10 -r=4° 0.05 -a0 - 100-80- 60102- 40- 2035^70^105CESD ScoresDensity Distribution of Overall Raw CES -D Scores0.20 -- 70- 60- 50- 40- 30- 20- 10r:4 0.15 -aq4• 0.10-(O 0.05 -A.(0Pq0.0^0.5^1.0^1.5^2.0^2.5CESD Scores Using Log 10 TransformationDensity Distribution of Overall CES-D Scores UsingLogarithmic 10 (x+1) Data Transformation


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items