TitleAnimal Companionship and Risk of Suicide.
Publication TypeJournal Article
Year of Publication2018
AuthorsBatty, GD, Bell, S
JournalEpidemiology
Volume29
Issue4
Paginatione25-e26
Date Published07/2018
ISSN1531-5487
Abstract
 

To the Editor:It has recently been advanced that animal companionship confers protection against leading causes of death, such as cardiovascular disease, potentially via the weight control associated with the ownership of pets whose care requires physical exertion.1 By means of other mechanisms, there are also reasons to anticipate that pet ownership may have an impact on additional health outcomes such as suicide, another major cause of premature mortality particularly in people under 50 years of age.2 Thus, human–animal interaction appears to have a favorable impact on selected risk factors for suicide, including interpersonal interactions, mood, anxiety, positive attention from others, and stimulation of social behavior.3 The benefits of animal contact also seem to extend to biomarkers of psychosocial stress, such as cortisol, heart rate, and blood pressure.3 Despite this circumstantial evidence, to the best of our knowledge, there has been no prospective examination of the link, if any, between animal companionship and suicide.Data were taken from the 1995 to 1997, 2001, 2002, and 2004 Health Surveys for England, a series of independent, U.K.-representative, near-identical surveys of individuals living in private households.4 Ethical approval for each survey was granted by local Research Ethics Committees, and study members provided informed consent.Study members were asked “Do you keep any household pets inside your house/flat?,” followed by inquiries about specific pets (dog, cat, bird, other furry pet, or “other” pet). Due to a low prevalence of ownership in the latter three groups, these data were collapsed. We used the following covariates, all of which were self-reported: age on leaving full-time education (categorized as: ≥ age 19/currently in education, age 17–18, age 16, and < age 16), mental health problems (self-reported mental health problems and/or use of psychotropic medication), cigarette smoking (nonsmoker, current smoker), somatic illness (one or more of the following physician diagnoses: neoplasms, diabetes, other endocrine disorders, cerebrovascular disease, myocardial infarction, angina, hypertension, any other heart disease, or respiratory disease), and marital status (married/cohabiting, other). As previously,5,6 suicide mortality was ascertained by linkage of cohort members to national cause-of-death registers until February 14, 2011 (International Classification of Disease codes in table footnote).A maximum duration of 17.1 years of follow-up (median, 10.4 years) gave rise to 47 deaths ascribed to suicide in an analytical sample of 67,441, which comprised people with data on pet ownership, covariates, and mortality. There was no clear relation between pet ownership and baseline covariates (eTable 1; http://links.lww.com/EDE/B331). Of these covariates, being male, having a more basic education, mental health problems, cigarette smoking, and not being married or cohabiting were related to an elevated rate of suicide death (eTable 2; http://links.lww.com/EDE/B331).In the main analyses, after basic adjustment, there was essentially no suggestion that animal companionship was related to suicide risk (Table). Disaggregating into the type of pet owned did not alter these conclusions. After adjustment for a range of covariates that included socioeconomic status and marital support, there remained no link between different types of animal companionship and suicide risk.Despite the reasonable prima facie case for a link between animal companionship and risk of suicide, we found no strong evidence in the first longitudinal study to examine this association. That we have been able to reproduce known risk factors for suicide (gender, smoking, mental health, marital status7) gives us some confidence in our findings for pet ownership.

DOI10.1097/EDE.0000000000000817
Alternate JournalEpidemiology
PubMed ID29489469
Grant List / / Medical Research Council / United Kingdom