“Whoever takes the dog gets the house”: How older adults negotiate, budget, and deploy resources for multispecies family health and well-being

ABSTRACT In this study, we interrogate the ways that older adult pet owners exercise their agency within the limits of their social statuses to allocate their resources for the management of the health needs of their multispecies families. Guided by fundamental cause theory, we consider how the participants’ memberships in socially constructed categories can impact their experiences with household health management. We conducted focus groups with 30 community-dwelling older adults who owned pets. Using a thematic analytic approach, we found that the participants’ bonds with their pets led them to negotiate priorities and make concessions for the benefit of the household.


Introduction
Though companion animals (i.e., pets) are legally defined as property in the United States (U.S.), many scholars have argued that animal companions are more appropriately conceptualized as family members than inanimate objects to be owned.Recent sociological work has made a case for pets as family members as defined by their reciprocal roles with people (e.g., (Fox, 2006;Irvine & Cilia, 2017)).Irvine and Cilia argue that nonhuman family members take roles within family units that can either replace the roles traditionally "played" by other humans (e.g., a child) or contribute to the "doing of" family in a more-thanhuman manner (Irvine & Cilia, 2017).In an example of the former, a series of three 2017 articles by Andrea Laurent-Simpson shows that other people, primarily parents or spouses, help to reinforce pet owners' identities by validating reciprocal human-pet roles of parent-child or sibling-sibling (Laurent-Simpson, 2017b).Laurent-Simpson also demonstrates how individuals who choose not to have children use their identity of "pet parent" to reinforce this choice, explaining that their urge to nurture is fulfilled by the caregiving of their animals (Laurent-Simpson, 2017c), and thus providing a role identity (Laurent-Simpson, 2017a).This same concept can be extended to older adults.Though research about pet ownership in older adulthood has overwhelmingly focused on human health outcomes, rather than "doing family," a la Irvine and Cilia, the underlying assumptions of many of the studies rely on the provision of companionship and the presumption of mutually beneficial and contingent relationships between people and their pets (Irvine & Cilia, 2017).
A great deal of literature in recent years has been devoted to whether, to what extent, and why pets may provide a health benefit to older adults (e.g., (Gee & Mueller, 2019;Wells, 2019)).Often, the mechanisms presumed to be at play for any health benefit of pet ownership are primarily in the interest of the owner without much regard for the pets; e.g., pets provide stress relief to their owners or dog walking is good for an owner's physical health.However, the strength of the attachment bond between pet and owner is often implicated as a predictor and/or mechanism by which pets may provide health benefits (e.g., (McDonald et al., 2021;Wells, 2019)).Much of the pet attachment literature and measurement was borne out of human attachment theory, which requires a mutual, bidirectional social bond (e.g., (Kanat-Maymon et al., 2016;Keefer et al., 2014;Meehan et al., 2017;Zilcha-Mano et al., 2011, 2012)).
The human-animal bond is defined by the American Veterinary Medical Association as "a mutually beneficial and dynamic relationship between people and animals that is influenced by behaviors essential to the health and wellbeing of both" (American Veterinary Medical Association, n.d..).A 2019 paper by Bibbo et al. reports, among a nationally representative sample of older adults, that "companionship" was the most stated reason for having a pet (Bibbo et al., 2019).Furthermore, other research has shown that pet owners sometimes sacrifice their own health and well-being in order to prioritize the needs of their pets, such as delaying healthcare treatments if their pets are without alternative caregivers (Applebaum et al., 2020;Canady & Sansone, 2019;Polick et al., 2021;Ramirez et al., 2022), failing to evacuate during natural disasters if they cannot bring pets along (Brackenridge et al., 2012;Chadwin, 2017;Farmer & DeYoung, 2019;Heath et al., 2001), and forgoing entry to a domestic violence or homeless shelter if they don't allow pets (Komorosky et al., 2015;Labrecque & Walsh, 2015).This concept has been referred to as "more-than-human solidarity," when the navigation of social and economic constraints, such as economic insecurity, put pets and owners at risk for disrupting the human-animal bond (Rock & Degeling, 2015;Toohey & Rock, 2019).In the following paragraphs, we link this concept to fundamental cause theory.

Agency in the management of multispecies household health: a fundamental cause perspective
According to Link and Phelan, socioeconomic status (SES) is a "fundamental cause" of disease and mortality because it determines 1.The context in which an individual is at risk of unhealthy behaviors ("at risk of risks") due to constrained agency and contextual or structural (macro-level) factors, and 2. Who has access to resources that can be used to bolster health, and both prevent disease and promote longevity (Link & Phelan, 1995).Fundamental cause theory postulates that the resources available to individuals of higher SES are dynamic or flexible: by their very nature, the components of SES (higher educational attainment, greater economic resources, greater power, greater occupational prestige, and greater social capital 1 ) will afford an individual access to information and new technologies that are not as readily available to individuals with lower educational attainment, fewer economic resources, less power, less occupational prestige, and lower social capital (Link & Phelan, 1995;Phelan et al., 2010).
Educational attainment is thought to be a flexible resource because it allows individuals to readily access and understand health information.Economic resources allow individuals to implement health knowledge by adjusting their behaviors and environments accordingly.Power, prestige, and social capital operate to allow individuals the ability to change their environments via influence within hierarchical systems and social networks (Phelan et al., 2010).Because the resources of higher SES individuals are dynamic or flexible, the intervening mechanisms by which individuals prevent disease or mortality will be replaced as new technologies and information emerge as those on the higher end of SES will access and implement the new information more rapidly than those of lower SES (Phelan et al., 2010).
Fundamental cause theory frames SES and racism as "causes of causes" or "risks of risks" as they cause the relationships between risk and protective factors and diseases by determining access to health-preserving or enhancing behaviors, as well as the contextual factors that shape exposure to risks (Clouston & Link, 2021;Phelan & Link, 2015;Phelan et al., 2010).While new health knowledge and technologies will typically benefit socially and economically privileged individuals first due to their resources: if a disease already has an existing distribution via social position (SES and race) and new knowledge about its prevention emerges, marginalized individuals will stand to benefit more than those with greater social status (Phelan et al., 2010).Conversely, if a disease is associated with the lifestyles of high-status individuals, when new knowledge about prevention emerges, it will benefit highstatus individuals more because they will alter their behaviors accordingly (Phelan et al., 2010).
Fundamental cause theory can be extended to more-than-human families (Irvine & Cilia, 2017) and more-than-human solidarity (Rock & Degeling, 2015;Toohey & Rock, 2019): multispecies families are not exempt from SES and racism as the cause of causes and/or risk of risks.It is hypothesized that social positions that are subject to marginalization in the U.S. social environment will transfer to pet health and welfare via the same mechanisms at play in human disadvantage.For example, Katja Guenther shines light on the ways in which power dynamics within an animal shelter, and the resulting fates of the animals contained within it, reflect back the broader problems within American society related to social inequalities (Guenther, 2020).Guenther conducted a multispecies ethnography at an open-access, municipal animal shelter in a low-income community with a largely Latinx population in Los Angeles County; in the book, she draws very clear lines between the ways that structural inequities among people trickle down to the health and welfare of the animals for which people are responsible (Guenther, 2020).In the chapter, "The Myth of the Irresponsible Owner," Guenther challenges a traditional belief in animal sheltering which places individual responsibility on people for the fate of their pets, without any acknowledgment of the larger structural issues that may make the idealized pet-keeping practices unattainable (Guenther, 2020).In challenging this "irresponsible owner" perspective, Guenther outlines the lives of precarity that many of the pet owners in question face, including eviction, foreclosure, homelessness, incarceration, threats of deportation, and unstable employment, among other circumstances that make caring for a pet in the idealized way more challenging (Guenther, 2020).Some of Guenther's outlined challenges to idealized pet-keeping may become more salient as pet owners age, which can become particularly problematic as pets are broadly believed to be beneficial to older adult health and well-being in a variety of domains, such as cognitive health (e.g., (Applebaum et al., 2022)), physical activity and physical function (e.g., (Dall et al., 2017;Friedmann et al., 2020Friedmann et al., , 2022;;Gee & Mueller, 2019;Peacock et al., 2020)), and mental health (e.g., (Applebaum et al., 2021;Bolstad et al., 2021;Gee & Mueller, 2019;Gee et al., 2017;Gee et al., 2021;Hui Gan et al., 2019;MacLean et al., 2017)).However, pet ownership is not without risks for older adults, including the increased risk of falling from pets being underfoot (Stevens et al., 2010), and potential zoonotic disease threats among older adults with weakened immune systems (Egorov et al., 2018;Obradović et al., 2020).Furthermore, for economically vulnerable older adults, the costs associated with pet ownership can compromise their ability to keep their pets, particularly if they need to enter supportive residential care with pet restrictions (Toohey & Krahn, 2018;Toohey & Rock, 2019).

The current study
The purpose of this study is to gain a greater understanding of the processes by which older adult pet owners negotiate, budget, and deploy their resources to manage and maintain the health and well-being of human and non-human animal household members.We primarily take a descriptive, interpretive approach to this exploratory focus group interview study (Battacharya, 2017).Building on the work of sociological animal studies scholars, this study interrogates how individuals make meaning of their relationships with their pets, and thus how these meanings influence their own behavior (Arluke, 2002(Arluke, , 2021;;Irvine, 2012aIrvine, , 2012b;;Irvine & Cilia, 2017;Laurent-Simpson, 2017a, 2017b, 2017c;Sanders, 2003).Additionally, a fundamental cause perspective is also deeply influential in this work: we seek to understand how social inequalities influence the daily lives of pet owners as they negotiate their resources and balance priorities.In doing so we consider the role that social structures maintain in human agency.We draw from fundamental cause theory, which offers a way to approach and make sense of the relative disadvantage or privilege that comes with various social identities that reflect power, or lack thereof, within the U.S. social environment, and how social position is thus implicated in health disparities (Link & Phelan, 1995;Phelan et al., 2010).Broadly speaking, in this study we aim to interrogate the ways that older adult pet owners exercise their agency within the limits of their own social positions and environments to allocate their resources for the management of the health needs of the people and pets in their family.Guided by a fundamental cause approach, we consider the various social statuses that come with the identities of the participants and how their group membership in socially constructed categories (e.g., race/ethnicity, SES, etc.) can impact their experiences in multispecies household health management (Link & Phelan, 1995;Phelan et al, 2010).Specifically, we seek to answer the following research questions: (1) How do older adults with pets budget resources for their health and well-being alongside the needs of their pets?(2) What resources are expended and negotiated in managing multispecies household health?(3) How do older adults with pets from varying social positions overcome barriers and solve problems related to household health and well-being?

Methods
We collected data for this study via four focus groups, which were conducted on Zoom 2 in the summer of 2021.Because the focus groups were originally intended to be in-person, recruitment was conducted locally in a mid-sized southeastern U.S. city.We used a purposive sampling strategy to identify both variation and commonalities in the experiences of the participants from differing backgrounds and social positions (Palinkas et al., 2015).We intended to recruit participants who represented a broad range of socioeconomic backgrounds and racial/ethnic identities.Participants were deemed eligible for participation if they were aged 65+, could speak, read, and write in English, and considered themselves to be the primary caregiver to a cat and/or dog.We recruited participants inperson via community pet-related organizations and clinics (e.g., a lowincome veterinary clinic, pet supply stores), community-based health and social organizations, and online through message boards and other social networks.Both in-person active and online passive recruitment methods were equally successful, whereas in-person passive recruitment (e.g., physical flyers at pet stores) did not yield many participants.Participants completed electronic informed consent documents and a short survey via Qualtrics for demographic information.They were mailed a gift card for their participation.Focus groups were audio recorded via Zoom, transcribed verbatim, and stored in Word documents.This study was granted ethical approval by the University of Florida Institutional Review Board, protocol number IRB202101238.

Focus groups
For this study, we chose to employ semi-structured interviews in small focus groups for data collection.Focus groups offer a unique opportunity for participants to hear one another recount their own experiences and thus can help with both recall and overcoming social desirability bias toward researchers (Morgan, 1996;Peek & Fothergill, 2009).For example, we were interested in some of the negative aspects of pet caregiving while aging; when we posed this topic, participants were reluctant to speak; however, once the ice was broken by one participant, many became comfortable in recounting similar negative experiences.A particular strength of focus groups lies in the interactions between participants (Morgan, 1996).The participants often elaborated on their points to better explain themselves to one another, and they at times engaged in asking for clarification from one another.Additionally, participants shared resources to help one another problem solve; Peek and Fothergill (2009) have pointed out that focus groups can be beneficial to participants in that they often provide social support to one another, particularly when they are experiencing similar circumstances.Though the focus groups were originally intended to be in-person, an advantage of the virtual Zoom format was the ability for participants to have their pets with them.Often, participants could be seen with their cats and small dogs in their laps.
In the interest of participant discretion, the participants were each assigned a unique letter and number (e.g., B4, D6), which was displayed with their video feed in place of their names.Zoom's automatic transcription displayed these code names for each speaker in the transcripts, effectively anonymizing the participants for analysis.Thirty-five participants were recruited and 30 ultimately participated in the focus groups, which ranged in size from three to ten participants each.Participants were assigned to groups randomly based on their availability.
Each participant completed a Qualtrics questionnaire with their demographic information prior to their focus group date.We collected the participants' age, income, homeownership status, educational attainment, race and ethnicity, marital status, and number and types of pets in their homes.This contextual and environmental information allowed us to analyze the focus group data with respect to each participant's social position and relative social advantage or disadvantage.By contextualizing the participants' words in their social statuses, we could garner broader information about the social structures and ecological factors influencing their experiences as older adult pet owners.Socioeconomic status is often operationalized via income or wealth, educational attainment, and occupational status.Because the target population in this study was retirement-age individuals, we did not collect occupational status, and we chose to use income instead of wealth to reduce participant burden and increase comfort with the disclosure of information.Homeowner status gave us an additional contextual measure to consider assets beyond income.
The focus group sessions were moderated by the second author, and the first author provided moderation support, such as offering additional prompts when necessary.Additionally, an undergraduate research assistant provided administrative and technical support.Each session lasted 90 minutes and focused on soliciting experiences of older adults who were aging in place (i.e., in their own home environment) with their pets.The format included prompts to solicit open discussion among the participants centering on topics of stress, health, the home environment, and animal welfare.For example, when eliciting responses related to stress, we asked participants, "Can you think of ways that your pet helps you cope with stress?" and "Can you think of ways that your pet causes or adds to your stress?"When asking about health behaviors and pets, we asked, "Have you or someone you know ever had to skip or delay any health care services or medical treatment, like picking up medication, going to the emergency room, or visiting your primary care physician, because of your pet?" and further prompted the participants with examples: "For example, not having anyone to care for your pet, or not being able to pay for boarding for your pet."

Data analysis
Initial transcriptions of the focus group recordings were conducted via automatic Zoom software transcription service.Two undergraduate research assistants corrected any discrepancies between the Zoom transcription and the audio recordings, which were then checked by both authors for accuracy.Once the transcripts were confirmed to be accurate, we uploaded the transcript documents to ATLAS.ti (Version 8.4.5) for data management.With the intention of ultimately translating our findings for a broad audience oriented toward the health sciences and public policy, we used an inductive thematic analytic approach to generate codes, categories, and ultimately, themes (Braun & Clarke, 2022;Saldana, 2021).We used an inductive approach, which permits the investigator to be theoretically flexible, however, we acknowledge that we are not impartial, nor can we be truly objective as we each have experiences and knowledge that inform our understanding of the world.Thematic analysis is thus an ideal approach as we approached this exploratory study data inductively, without the intention of adhering strictly to any particular theoretical framework (Braun & Clarke, 2006, 2019).Overall, our intention was to describe and interpret the meanings that participants were making of their experiences while also considering a critical approach by taking into account the social positions and environments the participants exist within and among, and how these social forces and structures shape their experiences.
We followed the six phases for thematic analysis as outlined by Braun and Clarke (2006): 1. Become familiar with the data: this was done by attending each focus group, reviewing the audio recordings, reading the text transcripts several times, and making notes of initial ideas.2. Generate initial codes: this step involved open coding to summarize chunks of data with words or phrases.We first employed descriptive and process coding to identify an initial code (Saldana, 2021).We used descriptive coding to summarize the topic that was being discussed by the participants (e.g., we assigned the code "finances" when participants discussed anything related to money), and process coding to assign actions that were being described (e.g., we assigned the code "household health budgeting" when participants described the ways that they budgeted resources) (Saldana, 2021).After entering the initial codes into a codebook, we systematically reviewed each transcript and recoded where necessary.3. Group codes into themes: themes were created iteratively alongside coding, as well as following the final round of coding in step two.Themes were created by reflecting upon the research questions while accounting for any new, unanticipated information to inform the findings.For example, we created the theme "enacting agency and resolving barriers" by revisiting our third research question (how do older adults with pets from varying social positions overcome barriers and solve problems related to household health and wellbeing?)and reflecting upon a code we called "workarounds," which we assigned to chunks of the transcript when participants described ways that they used creative problem solving to overcome a barrier or problem related to multispecies household health and pet caregiving.4. Review and confirm themes: this step was also iterative.Once themes were created, we went back through the data to confirm their appropriateness and fit with the codes and assigned quotations. 5. Define and name themes: this step was done in conjunction with steps three and four and involved analyzing the themes to create a coherent story from the findings.6. Select and report results: in this step, we carefully selected exemplary quotations from the transcripts to represent and illustrate each theme.We also drafted the results and completed the analysis during this step.We report the demographic characteristics of the participants when presenting their quotes to critically interrogate how their social position may have influenced their self-reported experiences.

Sample characteristics
Characteristics of the sample are presented in Table 1.Though we sought out to recruit participants who would represent variation in SES and race and ethnicity, our sample was ultimately over-representative of White and mid to high-SES individuals, as well as women.While we did have two participants who identified as Black and one who identified as Hispanic/Latinx, we were unsuccessful in recruiting pet owners of other races and ethnicities, such as Asian American and Pacific Islanders, or Native American and Indigenous peoples.This is a common problem in research related to pet ownership, which itself is not equally distributed among the U.S. population and tends to be highest among White Americans, compared to people of color (Applebaum et al., 2020;Mueller et al., 2021;Saunders et al., 2017).It is also important to acknowledge that the study's primary investigators both identify as White and we cannot fully appreciate nor understand the experiences of pet ownership for people of color, however, we hope to critically assess the focus group findings by taking into account the relative advantage and disadvantage via social position (e.g., race/ ethnicity, SES) in the U.S. Additionally, participants did reflect some variation in SES, but the mean income and education were relatively high, compared to the overall population.The inclusion of marginalized and under-represented groups in research about pet ownership and health is pertinent for future research -we elaborate more on this in the discussion section below.

Qualitative analysis
We identified three major themes: (1) negotiations and concessions made within the household to support family health, (2) taking account of resources and understanding limitations, and (3) enacting agency and resolving barriers.Table 2 displays themes, codes, and example quotations.

"I've had to make adjustments": negotiations and concessions when balancing the health needs of multispecies household members
Within the multispecies household, older adults took account of their relationship with their pets and the ways in which they believed their pets impacted their health, which in turn influenced how the participants balanced resources and priorities among household members.The belief that pets were beneficial to the participants' health, as well as the bond the participants shared with their pets, led them to negotiate priorities and sometimes make concessions to their own health for the benefit of the household.
Overwhelmingly, the participants in the focus groups felt their pets had been beneficial to their health.For example, several participants discussed the ways their pets helped to both relieve stress and keep them physically active.One participant who was in their 60's and identified as White, with middle range  yearly family income, an owned single-family home, and high educational attainment said, "sometimes just walking the dog can take away stress, and it's something I might not do otherwise."The built environment and access to green space that often comes with this participant's social position likely allows them to take advantage of a health-promoting aspect of dog ownership.Perceived safety benefits to pet ownership also came up in the discussion: one participant who was in their 80's and identified as Black, with income below the federal poverty level, an owned single-family home, and low educational attainment said, "if anything comes, because we live in the country . . .I'm going to know it, because [my dogs are] gonna bark and let me know."This participant felt their dogs offered them protection in their home in a rural area where they were not in close proximity to other people.While the topics varied somewhat based on the social position of the participants (e.g., the community where they lived mattered), the perceived health and well-being benefits of pet ownership was a salient and recurring topic among the participants.
Several participants also discussed ways that their pets posed potential health threats in the household.The following quote is from a participant in their 80's who identified as White, with high income and high educational attainment, who rents a unit in a retirement community: Our dog gets [flea prevention] so ticks don't attach to him, but when I sit on the floor with him and pet him at night, sometimes the ticks get on me, even though they don't attach to him.This year's been a bad year for ticks.I've gotten three ticks from the dog, that I didn't have any ill effects from them, but that can be a potential problem.
Similarly, other participants discussed potential health threats such as slipping in water from dog bowls and tripping over pets.However, one participant in Note: n = number of times each code was assigned in the data by the researchers.Codes were not mutually exclusive.their 60's who identified as Latinx/Hispanic, with mid-range income, an owned single family home, and high educational attainment said that their pets motivated them to keep their home clean, "when you have pets in the house . . . it kind of focuses me more on keeping the house clean . . .healthythe quality of the environment -not just for the people, but also for the pets."For this participant, potential environmental health threats were reduced for the entire household because they were invested in the health of their animals in the home.
Participants also discussed ways that caregiving for their pets could cause or exacerbate stress, which the participants generally conceptualized as relatively minor inconveniences in the context of the overwhelming benefits.However, some stressors were evidently quite pervasive and specific to aging.For example, several participants mentioned their stress around planning for the future with their pets as they aged.One participant in their 70's who identified as White, with low income and high educational attainment who owns a condominium stated, " . . .if I had to go to assisted living or something like that, where they don't allow pets, I don't know what I'd do.I don't want to think about that."This topic was salient among many of the participants as they planned for aging while intending to maintain pet ownership.Similarly, planning for aging pets' care caused stress for many of the participants, especially for those with limited financial resources.Many participants explained that they experienced economic stress due to the high cost of veterinary care.A participant in their 80's who identified as White, had an associate's degree, owned a home in a retirement community, and chose not to disclose their income stated, "that's probably the most stressful, the vet bills."Despite the overwhelming consensus of the group participants that pets were a net positive to their health, many undeniably experienced stressors related to their pets.While most participants shared at least one stressor related to pet caregiving, stressors often related to the limitations of participants' social positions, such as economic insecurity.
Participants also described ways that they made negotiations and concessions to their own health to meet the needs of their pets.In particular, many participants explained that health was sometimes budgeted as a resource within their household.For example, one participant in their 80's who identified as White, had high educational attainment, and chose not to disclose their income nor housing circumstances explained that their cat caregiving responsibilities sometimes took priority over their own needs: Because I have a multitude of cats, I do have to put their needs before my own, which means getting pet food is even more difficult because, if you want to get the most for your money, you have to pick up 20 pounds of cat food and 18 pounds of litter.I have had some delays to health services, because having many cats I've not been able to schedule for some of my own needs.And also, never be able to take a vacation because I could never find anyone that could provide the care the cats would need.
This participant explained that their cats often took priority over their own needs, and this could become detrimental to their health.Not only did they describe delays to receiving healthcare due to scheduling conflicts with their caregiving responsibilities, but also touched on their inability to take a vacation due to pet caregiving, as well as making financial concessions.Other participants described ways that they made specific choices and concessions related to the pets they brought into their homes, such as deliberately choosing smaller dogs as they aged to avoid injury while taking care of them.Although the majority of the participants noted that they had low probability of falling in the next few months, pets can pose a serious fall risk for older adults and the participants were cognizant of this.This created a tension between the participants' desire to continue to have pets and the potential health threats the pets could create.
Longevity was a salient concern for the participants as they considered aging with their pets.For example, the same participant from the previous paragraph who discussed their cat caretaking responsibilities also stated: "[Having pets] has helped my health because I'm gonna try to outlive them.I'm trying to eat better, I'm exercising.I want to be there for them as long as I can."Taken together with this participant's previous discussion of health concessions, this exemplifies the dichotomy of pet caregiving as an older adult.The participants often internalized this belief that pets were beneficial to their health and often made related assertions with an air of pride.However, when pressed, they could often think of ways their pets could become detrimental to their health.Relatedly, a topic that was also salient to many of the participants was the impact of their own mortality on their pets.For example, a participant in their 60's who identified as White, with mid-range income and high educational attainment who owned their single-family home said: We had a kitten come up in last November, on Thanksgiving Day, and we initially said "no, no, we don't need it," because it could last for 20 years.We have been hesitant to adopt new animals at our age . . .because of that concern, because something could happen to one or both of us, when you're in your late 60's.If something happens to both of us, it is a real problem, and even if something happened to one . . .We used to have a lot more [pets] -we're limiting the numbers because, if one of us was left alone, it could be a problem.This concern about mortality as it relates to the ability to ensure pets are cared for was something that came up from many of the participants.Some talked about their plans for how they would ensure care for their pets when they were no longer able to care for them themselves, and others even wrote their pets into their wills.One participant explained that their friend had made plans for the care of a behaviorally challenging dog by ensuring in their will that the family member who took responsibility for the dog would also be given ownership of their home.There was often a sense of a balance between the participants' desire to have pets as they age for the benefit of their health, but also a self-limiting force because they truly cared for their pets' wellbeing and didn't want to take on a new pet if they weren't sure they could care for the pets for their entire lives.

"We've made arrangements": taking account of resources and understanding limitations
The focus group participants mentioned many types of resources or capital they used to manage multispecies household health.Financial resources were commonly discussed, particularly in relation to affording veterinary care and pet supplies.One participant in their 70's who identified as White with mid-range income, high educational attainment, and owned their single-family home discussed the difficulty of having to make end of life decisions for their pets while keeping finances in mind, "Sometimes you have to make tough decisions . . .we've probably all been in the situation where you had a very ill animal and you had to weigh out what the treatment might [cost] versus going ahead and euthanizing an animal . . .that's a very gut-wrenching experience."Because veterinary care can be cost-prohibitive, it can sometimes make pet owners conceptualize their pets in economic terms, such as this participant's example of assigning a monetary value to extending their pet's life.Similarly, time use was also discussed as a resource or form of capital for multispecies household health management.For example, one participant in their 60's who identified as White, owned their single-family home, had high educational attainment, and chose not to disclose their yearly family income described managing their pet's medication regimen, " . . .I thought, 'oh my gosh, I can't wait until I don't have to do eyedrops eight times a day . . .' I had to take off work in order to do it."Other participants also mentioned the ways that they had to budget their limited time for care of their pets, which tended to be a relatively bigger concern among those who were employed than the participants who were retired.While participants were not asked to describe their profession, individuals who work low-wage jobs likely would find the time burden of high-needs pet care to be relatively more difficult to manage than those who have more flexible positions with higher wages.
In addition to financial resources and time use, social relationships were discussed extensively as a resource for use in multispecies household health management.Many participants discussed their ability or inability to depend on their social network for care of their pets when they were unavailable.One participant in their 70's who identified as White with mid-range income, high educational attainment, and owned their single-family home discussed their concern about burdening family members with pet care if they were to become ill or pass away: I always think of people who get ill and then suddenly, if they pass away, who's going to take your animal?I come from a family of animal lovers, so I don't get too concerned, but still, it's a burden to them.[There's an] underlying background stress of getting older and thinking about what's going to happen to your animals . . .Many other participants shared this concern related to ensuring their pets would be taken care of when they were no longer available to care for them.This was especially distressing for those who didn't have family living nearby or were single, divorced or widowed.However, even for those who lived with their partner, arranging contingency care due to an injury became problematic.For example, one participant in their 60's who identified as White with mid-range income and mid-range educational attainment who owned their single-family home mentioned the stress they experienced when they had an injury and needed to rely on their partner for pet care: . . .We have lots of pets . . .and sometimes it's a little overwhelming . . .but yesterday, I broke a bone in my knee so I'm on crutches, and I'm trying to care for all the animals.My husband's here too, but I'm the animal person, so it's kind of stressful to have to ask him.You know, 'can you scoop the litter boxes, can you take the dogs out.'This participant, a woman who was the usual caregiver to the family pets, was even hesitant to ask their husband for help with caring for their pets.In this way, a live-in partner may not have necessarily provided the guarantee that pets would be taken care of if the responsible party (often the woman in the heterosexual partnerships represented in this study) were to become unavailable.
Another way that social relationships were framed as a resource or form of capital was related to social connections and personal knowledge of community resources to support multispecies household health.One participant in their 70's who identified as White with high family income and high educational attainment who rented their home in a retirement community described how their connections with their neighbors provided an opportunity to make their community more dog-friendly: "Where I live a bunch of dog owners got together and raised quite a bit of money to upgrade the dog park here."This participant engaged their social capital within their retirement community to change the build environment to make it more dog-friendly and thus more livable and convenient for the residents with dogs.This seemed to be a unique feature of pet-friendly retirement communities, where several of the White, high-SES participants lived -the residents who lived there were empowered to manipulate their space to accommodate their needs, as well as the needs of their pets.Relatedly, several participants mentioned various community organizations that can help with pet needs for those who need assistance in the community.For example, one participant in their 60's who identified as White with mid-range income and mid-range educational attainment who owned their single-family home described several local resources for low-income pet owners: The [animal rescue] has a wellness clinic with low-cost stuff.The [trap-neuter-return program] does spays and neuters, and they have their shelter diversion program -if you find kittens, they'll do all the vaccinations and surgeries at no cost.The [homeless shelter] has a partnership with [animal rescue] and several other groups that treat people who find themselves homeless, so their animals have access to care . . .But there's really nothing for people who are kind of in the middle [income].
This participant makes an important point that veterinary care and other related services are often inaccessible even to those who make too much money to qualify for the services, but do not make enough to afford regular veterinary and (often) human health care.
Many participants described social support as an important resource; participants described ways that they could rely upon social support, as well as when lack of social support created problems.One participant in their 80's who identified as White with high income and high educational attainment who rents an apartment in a retirement community explained, "We haven't had any problems [related to contingency pet care], we live in a retirement community, and we have a daughter nearby, so we always have someone we can rely on."This participant had a robust social network by way of family and community and thus were not concerned about finding someone to help with pets if they were to become unavailable for any reason.Another participant in their 70's who identified as White with high-income and high educational attainment who rents a home in a retirement community described their role in providing pet-related social support to a friend, "I had a friend who went to the hospital, and I got a call from her to go take care of her dog."Other participants similarly described the ways that they were reciprocal with their friends and family who also had pets in terms of providing and receiving instrumental support when necessary.Some participants, however, recounted negative social interactions related to pet caregiving that may have interfered with their health, or that of another (human or animal) household member.One participant in their 70's who identified as White with income below the federal poverty line and mid-range educational attainment who owned their home in a mobile home park explained that they had a poor relationship with their neighbor, "[My neighbor] lets his dogs run free and they do their business on my lawn and harass my cats and me.And one of the dogs is a menace, he tried to bite me one time."This example is an important counterpoint to the many discussions above regarding the beneficial community building and social capital that pets can provide within communities: these benefits are certainly not applicable across the board to all pet owners and can be dependent upon social and geographic context.Another participant who was in their 80's and identified as White with high educational attainment and chose not to disclose their income or housing situation recounted a poor interaction with law enforcement: I'm 84 and probably a stereotypical elderly person in a way . . .like the cat lady, feeding the community cats.I always got permission and I'm always respectful of the property.The police came and were very, say, non-informed, and told me they were going to put me in jail.They wouldn't tell me their names.I said, 'can I continue feeding?' and they said no, and one of them insinuated that I was one of those hoarders, in a derogatory sense.It took me over a year [to resolve with the police] . . .I think it had to do with my identity, being an old woman.
This participant shared this experience in response to a prompt that asked about any poor pet-related experiences they felt were due to their identity.The participant felt they were being discriminated against by the police due to the stereotypes associated with their age and gender, that the police assumed they were a cat hoarder or "cat lady" because they were an elderly woman who was feeding community cats.The participant recounted frustration that they were unfairly targeted while doing something benign.

"I had to non-comply with the vet": enacting agency, resolving barriers
This theme reflects the processes by which participants enacted agency to resolve barriers to multispecies household health and well-being related to their social position.Much of the discussion that was coded into this theme was related to the ways in which participants used other resources (e.g., education, time, social support, etc.) to resolve barriers related to financial constraints.Even participants who reported relatively higher income and greater markers of wealth (e.g., home ownership) reflected on workarounds they employed as strategies for maintaining their health as well as that of their pets.Often, they discussed ways their social and financial resources supplemented one another to resolve barriers.Most participants, regardless of access to resources, described how they felt empowered by creative problem-solving and often a great deal of self-efficacy.For example, one participant in their 70's who identified as White with mid-range income and high educational attainment who owned their single-family home explained: I find it's useful to have two vets.I like older vets sometimes because they've seen a lot of things, but you have every right to say, 'okay, do we have to do this, do we have to do that.' . . .I do my research . . . the Merck Veterinary Manual said that one of the biggest problems is that they undertreat pyoderma, this skin condition, where they don't leave the antibiotics long enough . . .This participant deployed their social and human capital to assess which veterinary treatments were necessary while avoiding costly bills for what they assessed to be potentially unnecessary treatments.However, it's important to note that this participant would likely need a great deal of free time and adequate financial resources to shop for the best deal or work with a veterinarian that would welcome their independent research about their pet's health condition.This same participant shared their knowledge of reducedcost veterinary medications from human-serving pharmacies, "If you've got a [pet] name that sounds like a family member, you can get away with it."They explained how they learned that their local pharmacy would dispense reducedcost medications for their pets if they believed the prescriptions were actually for human household members.Another participant in their 70s who identified as Black with income below the federal poverty level and low educational attainment who owned their apartment revealed a relationship with their veterinarian which was instrumental in providing pet care that the participant couldn't otherwise afford, "I had an emergency [out of town] and my dog's doctor knows she has to have her shots on time, so she volunteered to keep [my dog] for me for two weeks."This participant disclosed that they have a close relationship with their veterinarian such that they were willing to pet sit for free for an extended period when this participant was in need of contingency pet care.While financial resources were often instrumental to resolving many of the barriers related to the management of multispecies household health, even those with limited resources were driven to find workarounds to maintain their important relationships with their pets while attending to their own needs.

Discussion
In this study, we sought to explore the ways in which older adults described budgeting their resources to manage the health and well-being of the people and pets in their households.We also interrogated the ways in which the relative privilege or disadvantage of the older adults' social positions sometimes limited and sometimes bolstered their agency with respect to decisionmaking and problem-solving in the pursuit of multispecies household health and wellbeing.
Overwhelmingly, the participants thought of their pets as important, dependent family members that shaped their own behavior in profound ways, with implications for the whole household unit.Consistent with previous research about multispecies families, the participants generally conceptualized their pets as reciprocal beings with agency (Irvine & Cilia, 2017) and, at times, they practiced more-than-human solidarity (Rock & Degeling, 2015;Toohey & Rock, 2019).However, they often rationalized their love for their pets as serving a specific function, justifying their relationship with their pet and the associated expenditures and concessions as worthwhile because the pets were believed to be good for their health.This is perhaps not surprising given the widespread cultural belief, perpetuated by the media, that pets are akin to a therapeutic intervention (Herzog, 2011).This belief that pets were beneficial to the participants' health, as well as the bond the participants shared with their pets, led them to negotiate priorities and sometimes make concessions to their own health for the benefit of the household.Notably, those with greater advantage via their social position were often those who could take the most advantage of the mechanisms that pets are thought to be good for health, such as dog walking in nearby green spaces, and facilitating connections with other pet owners at community dog parks, thus bolstering social support and social capital (Bulsara et al., 2007;Wood et al., 2017;Wood et al., 2005).
The older adults in this study tended to make logical choices for their pets as they were aging, such as choosing smaller dogs or resolving not to have pets beyond a certain age for fear of outliving them.This is consistent with previous research on pets from a life course perspective (e.g., (Bibbo et al., 2019)), suggesting that pet ownership rates decrease with age, perhaps because of constrained agency due to aging processes.The participants in this study were often life-long pet owners who greatly valued the welfare of their pets and thus made personal sacrifices to maintain their pets' health and welfare.However, consistent with fundamental cause theory, social position played a distinct role in the extent to which the participants could exercise agency in support of their health (Link & Phelan, 1995;Phelan et al., 2010).Those with fewer resources were more vulnerable to poor health outcomes for themselves ("at risk of risks"), as well as their pets.For example, as individuals age, those with strong social ties and ample economic resources may be better able to stay in their homes with their pets, whereas those who need to enter residential supportive care are much less likely to be able to keep their pets with them (Toohey & Rock, 2019).If an older adult who lives alone is concerned that their pets may be relinquished to a shelter or let loose to fend for themselves, their own health may suffer because of delayed or forgone healthcare.Participants in this study with ample resources also harnessed them to ensure posthumous care for their pets, such as the participant who was offering their house as collateral to whoever would take responsibility for their pet when they die.In this way, health (dis)advantage and animal welfare (dis)advantage are inherently tied to one another: both health risks and relinquishment risks are accelerated in the absence of ample resources.
The results from this study also underscore a service gap among middleclass pet owners.Some of the participants were able to provide for all their pet needs without any concern for their economic circumstances, while others expressed that they experienced a great deal of stress in affording veterinary care.Many communities have made great strides in the provision of veterinary care and related pet services (e.g., grooming, boarding) for people whose incomes are very low, as well as individuals experiencing homelessness.There is still a need for the expansion of these types of programs, particularly with respect to making them more accessible to those for whom transportation can be a barrier, as well as in communities of color (Matheson & Pranschke, 2022;McDonald et al., 2022); however, a clear unmet need is among those who make too much income to qualify for subsidized or free veterinary and other pet services but still cannot afford them due to other expenses.

Limitations
While this study sample had some variation in social position among its participants, the majority were non-Latinx White and mid or high-SES due to recruitment limitations.Given the challenges highlighted in maintaining multispecies household health among those with social and economic disadvantage, it is likely a sample with a greater representation of marginalized and economically disadvantaged pet owners would reveal even greater hardships, sacrifices, and compounding limits to participants' agency.Future research should also consider the role of agency as it relates to structural inequalities, such as racism, in effectively maintaining multispecies family health.Additionally, we did not consider the health status of the participants in this study, which would likely impact the extent to which individuals can make concessions to their own health spending and budgeting: these questions should be asked among individuals who are managing a chronic illness.Finally, we chose not to ask about occupational status due to the age of our participants; however, many people over 65, particularly those of lower SES, do continue to work beyond retirement age.Future research on older adults who own pets should consider the occupational status of their participants.

Implications
There are several practice and policy implications that can be drawn from this research.We strongly caution against any interpretation of the findings in this study that could be misconstrued to assume that, because pets may interfere with the ability of older adults to manage their own health, the solution is to take away their pets.We contend that consistent with fundamental cause theory, the solutions to these issues of social inequalities in health and healthcare, the solutions are upstream.Keeping this in mind, we make the following recommendations for policy.
First, communities may consider partnerships between animal welfare organizations and community senior health resources to increase access to healthcare and veterinary resources for older adults and their pets.Ideally, community senior centers would partner with veterinarians to offer veterinary care alongside human healthcare and older adult social services (see (Hoy-Gerlach & Townsend, 2023;McLennan et al., 2022) for a review of community-based, volunteer-run organizations supporting older adults and pets together).Finally, removing pet restrictions in most senior living communities and assisted living facilities would provide more opportunities for older adults to stay with their pets without potentially compromising their own health. 3Ideally, pet-friendly housing would be the norm for older adults, while some facilities could retain pet restrictions to accommodate those who are at an increased risk for zoonoses (e.g., severely immunocompromised) and those who simply prefer to stay away from animals.Ultimately, housing should not be a barrier to preserving the human-animal bond.

Notes
1. Throughout this document we adhere to Bourdieu's conceptualization of social capital.
See (Bourdieu, 1985).2. Focus groups were initially planned to be conducted in-person at a community resource center, however, they were ultimately moved to Zoom for the safety of participants and researchers during the COVID-19 pandemic.3.For a discussion of potential issues related to pet ownership among older adults receiving care services, see (Bibbo et al., 2022).

Table 2 .
Themes, codes, and example quotations: codes were not mutually exclusive; we note in the table when a quote was double-coded.My budget is so tight that with this increase in [pet] food cost, I've just adjusted my own grocery costs . . .I've had to make adjustments so I can afford to buy even the substitute low fat food and the kidney food for the cat." [financial resources] "When I'm running late, it's stressful for me to think of the animals getting fed late and the dog being let out late." [cause or exacerbate stress] (Continued)