The three Andersen behavioral health service use factors—psychosocial, enabling, and need—remained factors in our analyses reflecting aspects of the older adults’ reasons for LTSS use. However, a few dimensions within the factors were more salient than others. We identified four new dimensions through conventional content analysis: losses and changes [psychosocial] and tangible support, capability to provide informal support, and accessibility of informal support [enabling]). Sixteen dimensions that were originally in the expanded model did not emerge during our directed analysis (dimensions that reflected <=1 respondent, e.g., interpersonal skill, home ownership [see Table
3 for entire list]). Nurses and other professionals were added as additional referents, caregiver expectations were modified to expectations of care, decision makers were split up into many layers based on who was making the decision or how the decision was being made and physical and cognitive need were merged to functional health.
Table 3Newly emergent and modified dimensions to Andersen’s expanded behavioral model (insert after third paragraph in results)
Psychosocial | Attitudes | Care Providers |
Technical Expertise |
Affordability |
Perceived expense |
Social Environment |
Social interaction |
Activity Level |
Familiarity |
Diversity |
Self-determination |
Independence |
Knowledge | Content and Amount of Information |
Sources of Information |
Family/friends/^self- |
Medical professionals |
Accessibility of Information |
Social Norms | Referents |
Spouse, ^family, children, ^self |
Friends, neighbors |
Social workers, ^nurse, clergy, doctors, lawyers, ^other professionals, ^they |
Relevant Norms |
Family burden |
*Expectations of care |
^Losses and changes |
Perceived Control- | Role of Choice |
Decision maker-split into 4 layers |
^Autonomous decision |
^Collaborative decision |
^Paternalistic decision |
^Placement |
^Unsure or forgot how decision was made |
Alternatives |
Planning for Future Needs Financial planning |
Psychological planning |
Enabling | Availability of Support | Formal Services |
Openings at facilities/waiting lists |
Proximity |
^Tangible Support |
Informal Support |
Ability to provide support- split into 2 layers |
Proximity |
Capability |
^Accessibility of informal support |
Financial Resources | Financial Well-being |
Assets |
Protection against Risk |
Insurance |
Need | Objective/Perceived | Degree of Disability |
*Functional health |
Duration of Disability |
*Functional health |
We created a conceptual framework depicting newly emergent (not present in expanded model) and modified dimensions (revised from expanded model) to Andersen’s expanded behavioral model after performing our conventional analysis (Table
3). Emerging data were not mutually exclusive to one dimension and could have been coded under multiple dimensions. Classical content analysis revealed the proportions of older adults within each LTSS and from each racial/ethnic background who identified with a specific dimension (Table
4). The classical content analysis results for Hispanic older adults vs. non-Hispanic older adults can be found in Supplemental Table 1. In the following sections, we discuss our findings within the context of LTSS use across racial and ethnic older adult groups.
Table 4Quantification of factors, domain, themes, and dimensions through classical analysis (insert after third paragraph in results)
PSYCHOSOCIAL | 50 | 99 | 7 | 227 | 5 | 6 | 43 | 63 | 42 |
Attitudes | 13 | 17 | 1 | 43 | 0 | 0 | 15 | 14 | 20 |
Care Providers | 9 | 11 | 1 | 16 | 0 | 0 | 10 | 6 | 18 |
Technical Expertise | 9 | 11 | 1 | 16 | 0 | 0 | 10 | 5 | 18 |
Interpersonal Skill | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
Affordability | 1 | 0 | 0 | 3 | 0 | 0 | 1 | 1 | 0 |
Perceived expense | 1 | 0 | 0 | 3 | 0 | 0 | 1 | 1 | 0 |
Home Ownership | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Social Environment | 2 | 6 | 0 | 11 | 0 | 0 | 2 | 6 | 1 |
Social interaction | 1 | 3 | 0 | 8 | 0 | 0 | 2 | 5 | 0 |
Activity Level | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
Familiarity | 0 | 1 | 0 | 3 | 0 | 0 | 0 | 0 | 0 |
Diversity | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Self-determination | 1 | 0 | 0 | 8 | 0 | 0 | 2 | 1 | 1 |
Privacy | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Dignity | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Independence | 1 | 0 | 0 | 8 | 0 | 0 | 2 | 1 | 1 |
Knowledge | 0 | 3 | 0 | 4 | 0 | 0 | 7 | 11 | 0 |
Content and Amount of Information | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 |
Service types provided | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 |
Eligibility rules | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
Legal/regulatory issues | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Financial coverage | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Sources of Information | 0 | 3 | 0 | 3 | 0 | 0 | 6 | 9 | 0 |
Family/friends/SELF- | 0 | 3 | 0 | 3 | 0 | 0 | 1 | 4 | 0 |
Lawyers | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Medical professionals | 0 | 0 | 0 | 0 | 0 | 0 | 5 | 5 | 0 |
Clergy | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Accessibility of Information | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 |
Attainability | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 |
Comprehensibility | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Social Norms | 11 | 20 | 1 | 81 | 2 | 1 | 11 | 20 | 13 |
Referents- | 4 | 12 | 1 | 42 | 0 | 1 | 10 | 17 | 8 |
Spouse, *FAMILY, children, *SELF | 3 | 7 | 0 | 28 | 0 | 1 | 2 | 6 | 4 |
Friends, neighbors | 0 | 1 | 1 | 2 | 0 | 0 | 0 | 0 | 1 |
Social workers, *nurse, clergy, doctors, lawyers, *other professionals, *they | 1 | 4 | 0 | 12 | 0 | 0 | 8 | 11 | 3 |
Relevant Norms | 7 | 8 | 0 | 39 | 2 | 0 | 1 | 3 | 5 |
Family burden | 1 | 2 | 0 | 11 | 0 | 0 | 1 | 0 | 0 |
Expectations of care | 3 | 1 | 0 | 9 | 1 | 0 | 0 | 2 | 5 |
*Losses and changes | 3 | 5 | 0 | 19 | 1 | 0 | 0 | 1 | 0 |
Perceived Control | 26 | 59 | 5 | 104 | 3 | 5 | 10 | 18 | 9 |
Role of Choice | 25 | 59 | 5 | 89 | 2 | 4 | 5 | 11 | 7 |
Decision maker-split into 4 layers | 21 | 51 | 4 | 79 | 2 | 4 | 3 | 9 | 2 |
Autonomous decision | 3 | 1 | 1 | 32 | 1 | 0 | 1 | 4 | 0 |
Collaborative decision | 3 | 4 | 0 | 23 | 0 | 1 | 0 | 1 | 0 |
Paternalistic decision | 9 | 19 | 0 | 18 | 0 | 2 | 2 | 2 | 1 |
Unsure/forgot how decision was made | 3 | 3 | 2 | 3 | 1 | 1 | 0 | 2 | 0 |
Placement | 3 | 24 | 1 | 3 | 0 | 0 | 0 | 0 | 1 |
Alternatives | 4 | 8 | 1 | 10 | 0 | 0 | 2 | 2 | 5 |
Planning for Future Needs | 1 | 0 | 0 | 15 | 1 | 1 | 5 | 7 | 2 |
Financial planning | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Psychological planning | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
ENABLING | 10 | 19 | 1 | 37 | 2 | 1 | 9 | 12 | 9 |
Availability of Support | 8 | 16 | 1 | 37 | 3 | 1 | 8 | 10 | 9 |
Formal Services | 4 | 8 | 0 | 23 | 1 | 1 | 7 | 9 | 6 |
Openings at facilities/waiting lists | 2 | 0 | 0 | 4 | 0 | 0 | 0 | 2 | 0 |
Hours of operation | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Proximity | 2 | 4 | 0 | 10 | 0 | 0 | 0 | 0 | 0 |
Tangible Support | 0 | 4 | 0 | 9 | 1 | 1 | 7 | 7 | 6 |
Informal Support | 4 | 8 | 1 | 14 | 2 | 0 | 1 | 1 | 3 |
Willingness to provide support | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
Ability to provide support- split into 2 layers | 2 | 7 | 0 | 13 | 1 | 0 | 1 | 1 | 2 |
Proximity | 0 | 0 | 0 | 2 | 1 | 0 | 0 | 0 | 0 |
Capability | 2 | 7 | 0 | 11 | 0 | 0 | 1 | 1 | 2 |
Accessibility of informal support | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 0 | 1 |
Financial Resources | 2 | 3 | 0 | 0 | 0 | 0 | 1 | 1 | 0 |
Financial Well-being | 0 | 2 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
Income | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Assets | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
Protection against Risk | 2 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
Insurance | 2 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
NEED | 18 | 47 | 3 | 70 | 1 | 2 | 15 | 30 | 32 |
Objective/Perceived Need | 18 | 47 | 3 | 70 | 1 | 2 | 15 | 30 | 32 |
Degree of Disability | | | | | | | | | |
Functional health | 18 | 44 | 3 | 70 | 1 | 2 | 15 | 30 | 32 |
Duration of Disability | | | | | | | | | |
Functional health | 0 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Psychosocial factors for determinants of long-term services and support use
Attitudes (n = 123), knowledge (n = 25), social norms (n = 160), and perceived control (n = 239) are all constructed under the psychosocial determinant factor in Andersen’s expanded behavioral model. Each of these domains was cited by 5% or more of participants.
Domain of attitudes
Attitudes consist of participants’ views on LTSS use related to the themes of: care providers, affordability, social environment, and self-determination.
Care providers were those who delivered formal or informal care. They were desired for their ability to provide technical expertise (e.g., based on prior training and experience or knowledge on how to handle emergencies) or interpersonal skill (trustworthiness, compassion, listening, and communication skills) to older adults in need of care. Over 20% of the 464 participants discussed their technical expertise needs that prompted LTSS use. The frequency of responses to this theme varied across racial groups; Other (n = 19/66, 29%); White (n = 35/236, 15%); and African-American older adults (n = 16/159, 10%). Twenty-nine percent of Hispanic older adults reported requiring some form of technical expertise (n = 26/91). Common among participants was a need for LTSS as their medical condition worsened. Specifically, one Hispanic participant using HCBS relayed the need for a provider who could “monitor [their] diabetes and hypertension…[and provide] physical therapy.”
Being able to afford services was also important to one’s actual use of LTSS. These views were guided by the perceived expense of services and the impact LTSS use had on home ownership. Concern about affordability was discussed primarily among those who were White and receiving care in AL (n = 3). For example, one White participant noted the desire and opportunity to save money as having the most influence on her decision: “AL was cheaper than independent [living] due to not needing a car.” Another White AL participant described the process of obtaining AL as “I sold my house and wanted to be in a medically safe, spiritual, and semi-affordable environment where people are comparable to me.”
In addition to affordability, the presence of a social environment was central to one’s longing for connections achieved by social interaction, participation in activities, familiarity in surroundings, and diversity of the environment. Among those who commented on social environment, social interaction was most commonly discussed and, particularly among White (n = 11/236, 5%) and African-American (n = 8/159, 5%) participants who were primarily receiving services via AL and HCBS, respectively. For example, having the opportunity to share spaces with other individuals and be active was necessary, as noted by an African-American participant receiving HCBS: “I love being around people, and to get up and know I’m going somewhere. I was living by myself and went to a community center which I loved. Then I moved and started to come here.”
Social interaction was additionally important when coping with the loss of loved ones, as a White participant receiving care in AL reported: “My husband passed away when we were in independent living. It was too hard to be there without him because I kept thinking of him…. I moved here to meet new people…. I just thought I’d be happier and I am.”
On the other hand, participants discussed their views on self-determination while using LTSS as a need to maintain independence. The majority of these views came from White older adults (n = 11/236, 5%) followed by the Hispanic ethnic population (n = 3/91, 3%) who primarily received services via AL and HCBS, respectively. For some participants, the choice of LTSS was based on their inability to live independently any longer, while for others, it was to maintain or regain their independence. A White participant in AL described the need for independence as “I was lonely living by myself; afraid of falling or experiencing pain. I wanted to make my own end of life decisions. I wanted to plan the rest of my future and make my own choices…. I like to live elegantly and want my independence.”
Domain of knowledge
Acquiring knowledge of the LTSS site was critical for making decisions about LTSS use and being aware of available options. While the content and amount of information did not emerge as a theme, what did emerge was who was providing information specific to LTSS (sources of information). Family, friends, study participants as well as health professionals were primarily discussed as key sources of information among the African-American older adult subgroup (n = 12/159, 8%). Family members typically visited different LTSS options and then shared what they learned with older adults. When describing this process, an African-American HCBS participant particularly noted, “My niece told me about it and I came and tried it and liked it.” Older adults across the White, African-American, and Hispanic racial/ethnic groups similarly reported receiving letters or cards or receiving information from their hospitals. Participants most commonly mentioned social workers and doctors as the health professionals who communicated with them about LTSS use.
Domain of social norms
Social norms consist of two themes: referents and relevant norms. Referents were defined in Andersen’s expanded behavioral model as decision makers who were considered legitimate sources of authority; however, in our analyses, this theme closely mirrored the sources of information theme. Therefore, to differentiate between the two, we further defined the referent dimension as a process that appears to be collective; that is, the participant was referred to a particular setting, it was suggested to them, and/or the referent had knowledge or experience with the LTSS site. Referents included friends, family, and members of the healthcare and professional team (i.e., social workers, clergy, doctors, and lawyers). “Nurses” and “Other Professionals” were added as additional referents from our analyses. White older adults more commonly reported spouses, family, and children serving as referents to services, while African-Americans discussed the healthcare and professional team (e.g., lawyers) serving as referents. For example, an African-American participant described a reference to their current NH along the lines of “The social worker said it was better for my family.”
Relevant norms were aspects that had direct bearings on one’s choice to use LTSS. The dimensions within this theme were comprised of the following norms: family burden, expectations of care, and losses and changes; the second norm was modified from care expectations, and the third norm was a newly emerging dimension. Relevant norms were primarily discussed by White participants. Regarding family burden, a concern was not just for burdening one’s family but also oneself and others outside of the family. A White participant using AL discussed her desire to avoid family burden as follows: “I wasn’t able to take care of my daily needs and my fatigue level caused me to sleep all of the time and miss meals and meds. It was taking too much of a toll on my daughter and it wasn’t fair to her.”
Expectations of care focused on participants’ goals while using LTSS. These goals were as generic as “to get better” or more specific as the need for “better communication.” The norm “losses and changes” was added as part of a relevant aging norm that consisted of experiences with death, illness, and decreased ability to care for the home. A White AL participant discussed the experience with losses and changes in functional status requiring subsequent LTSS use: “I became ill and my husband was sick then too. We couldn’t really care for each other then. Upon getting better, our kids felt it was time to move and give up the house that required too much care so we began making arrangements. We weren’t expecting to come to a place like this but needed it.”
Domain of perceived control
Perceived control describes the participants’ involvement in LTSS decisions made on their behalf. This domain specifically consists of the two themes: participants’ role of choice in LTSS use and prior planning for future needs and two dimensions: decision maker and alternatives. Because many players affected the older adults’ role of choice of LTSS use, the dimension of “decision makers” was broken into five subdimensions: autonomous, collaborative, paternalistic, placement (transferred/sent from facility/taken from home/put), and unsure or forgot.
One’s role in the decision making of LTSS use differed according to race/ethnicity and LTSS type. A greater proportion of White participants (n = 62/236, 26% [White older adults] vs. n = 11/159, 7% [African-American older adults]) discussed this decision as being autonomous or collaborative and used AL services, while a greater proportion of African-American participants (n = 45/159, 28% [African-American older adults] vs. n = 35/236, 15% [White older adults]) discussed this decision as being paternalistic or part of a “placement” process and used NH services.
In making the decision autonomously, a White participant discussed the process as being a combination of factors, but at the end of the day stated feeling this way: “I was having knee replacement surgery and expected that I wouldn’t recover enough to move back to independent living. The move was my choice. I didn’t want to be in independent living. I don’t think I would have moved if it hadn’t been for the surgery. I also knew that I could spend time post-surgery in [a] NH before being able to move back to AL.” Another White participant discussed the decision for LTSS use being collaborative with input from family; as she recalled, “I had a couple of falls in Florida and the kids felt it was time to move. We all agreed that this was a good place for my husband and I. We all agreed he was a little too much for me to handle alone.” Alternatively, an African-American participant in a NH discussed having no role in the decision making of LTSS use and the decision being paternalistic: “They felt that I needed to come here. This was not a decision I made. The doctors talked my family into moving me here.” Participants additionally discussed LTSS use as a process initiated and carried out by the hospital or medical professional coded in this study as “placement.” A NH African-American participant stated, “When I was hospitalized, they thought I should be in rehabilitation. When you can’t walk or do nothing, this is where they put you.”
Alternatives was characterized by changes to care needs, preferences, and options that influenced LTSS use. These alternatives ranged from the older adult having no other place to go, the older adult leaving unsatisfactory conditions and looking for a better way of life, or the older adult no longer being able to care for himself or herself. For example, an African-American participant in the NH described it as follows: “Because I knew I would get better services here than living on my own—doctors, laundry, food, it’s 24/7 here and you don’t have to worry about the aide not making it in.”
Planning for future needs comprised the second theme under perceived control. This theme described long-term decision making as a process unfolding over time and consisted of two dimensions: financial planning and psychological planning. Financial considerations such as affordability of LTSS and one’s state of finances were important to future planning, along with one’s psychological preparation that followed trends or beliefs in LTSS use as a preventive measure or a product of age. For one White participant, this trend in psychological preparation was described as a process dictated by the community: “The community pushed me to move. At my age, it seemed like the time had come to be sensible and accept more help. They had asked me a while ago and I wasn’t ready until now.”
Enabling factors for long-term services and support use
The enabling factor consisted of acquiring family and community resources and the accessibility of those resources, including availability of support and financial resources.
Domain of availability of support
Characteristics of the availability of support, both formal and informal, remained emergent with LTSS use. Support was not always available in the location when the older adult wanted it, however. To be considered for LTSS, some participants had to place their names on waiting lists, as one White AL participant noted: “My name was on the waiting list. They called me and offered me this. That made me start thinking perhaps it was a good time.”
Proximity was described in several forms and could have been related to the proximity of services, meals, activities, or care within or from the home, community, or facility. For example, a White participant in AL stated, “The independent housing I was in was very far away. It was hard to walk to the main building after I had my hip operation.” Proximity also took into account the proximity of family, as another White participant in a nursing home explicitly stated: “I wanted to be near my daughter. Family comes first.”
Tangible support (newly emerged) included the need for everyday physical support such as transportation, meals, and housework. An African-American participant using HCBS described this need for support as follows: “I have pain in my left knee and left side of hip. I’m always in pain and I feel very sad most of the time. It affects my ability to do housework and cook for myself so I’m losing weight. I’m afraid to travel by myself to the doctor. I need home assistance.”
The theme of informal support included willingness to provide support, ability to provide support (further broken down to capability [newly emerged] or proximity of the informal support), and accessibility of informal support (newly emerged). Willingness was described as a family member simply being okay with providing care to the older adult. A NH participant part of the Other racial group discussed her daughter’s unwillingness to provide support by stating, “My daughter wants to take her home back and I understand. I was in a room with my two grandkids and I wasn’t getting much sleep.” Regarding capability, an African-American participant residing in a NH discussed care limitations as follows: “Because my kids were working and couldn’t take care of me. We decided it was time for them to go on.”
For proximity of informal support, it was uncertain whether the family elected to serve as the participant’s caregivers or the participant simply wanted to be near them. Accessibility to informal support additionally influenced one’s decision for LTSS use. A White woman in AL described this lack of access as follows: “If I were to have gotten ill in independent living, I would have had no one but my younger sister to take care of me which wasn’t appropriate.” Or another participant who self-described as “Other-More than One Race” living in HCBS discussed the influence of children on LTSS use, reasoning: “Because I didn’t have children and I didn’t want [any].”
Domain of financial resources
Financial resources as they related to one’s financial well-being (i.e., income, assets) and protection against risks (i.e., insurance) were important in one’s ability to access services, but were less salient to actual LTSS use when compared to other dimensions.
Within this dimension, having insurance and considerations about income were most salient to LTSS use. One Hispanic older adult described the process of securing HCBS through insurance: “Because I have Medicare, they offered it to me when I was sick, they asked me if I would like to have a girl to help at home because I was ill because of my leg.” Alternatively, an African-American NH participant’s use of LTSS was influenced by income, as indicated by the statement, “I needed to be somewhere I could get [financial] assistance.”
Need
The need factor was the most commonly discussed factor attributed to actual LTSS use. Need primarily focused on the older adults’ objective or perceived need related to their functional health and the degree and duration of their disability. It originated from a wide range of ailments and included difficulty with vision, recent surgeries (e.g., heart, hip/knee replacements), strokes, heart disease, respiratory disease, or issues related to mobility deficits.
A Hispanic White/African-American participant described the complexity of functional need requiring LTSS use: “Because I can’t see, I can’t walk. I don’t have good balance. I need help at home and to be able to go out. I can’t check my blood sugar level because I can’t see.”