Skip to main content
NCBI home page
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences logoLink to The Journals of Gerontology Series B: Psychological Sciences and Social Sciences
. 2022 Oct 7;78(Suppl 1):S59–S70. doi: 10.1093/geronb/gbac150

Influence of Care Network Characteristics on Physician Visit Experiences for Black, White, and Hispanic Older Adults With Dementia

Mary F Wyman 1,2, Irving E Vega 3, Laura Y Cabrera 4, Reza Amini 5, Kyeongmo Kim 6, Wassim Tarraf 7,✉
Editor: Noah J Webster
PMCID: PMC10010481  PMID: 36205710

Abstract

Objectives

Most older adults with dementia are assisted by multiple caregivers, but the relationship of care network structure with health care access and quality is underexamined. We sought to test the associations of care network characteristics with the physician visit experience for older adults with dementia across diverse racial/ethnic groups.

Methods

We used data on Medicare beneficiaries (aged 65+) with dementia from the National Health and Aging Trends Study (2015–2019) to fit logistic regression models to test associations between physician visit outcomes and (a) size of the potential care network and (b) proportion of potential care network members (PCNMs) currently helping with daily functioning tasks. We also tested for modifications by race/ethnicity.

Results

Hispanic respondents had the largest potential care networks (M = 6.89, standard deviation [SD] = 3.58) and the smallest proportion of PCNMs providing help with daily functioning (M = 29.89%, SD = 22.29). In models adjusted for demographics and dementia classification, both network size and proportional involvement of PCNMs were positively associated with the presence of a PCNM and assistance during the visit. Associations remained significant at 4-year follow-up for the presence of PCNM at the visit and were robust to further adjustments for insurance type, income, and health factors. Associations were not modified by race/ethnicity.

Discussion

Larger networks and a higher proportion of PCNMs providing assistance predicted caregiver presence and assistance at the physician visit but not access to care. Findings suggest that strengthening care networks early in the disease may support improved health care outcomes for persons with dementia across diverse populations.

Keywords: Family caregiving, Health care disparities, Primary care

Dementia affects more than 6 million Americans (Alzheimer’s Association, 2022). Most persons with dementia (PwD) have comorbid medical conditions requiring outpatient management (Bell et al., 2015; Bunn et al., 2014). PwD are at risk for suboptimal health outcomes, including potentially avoidable hospitalizations and early institutionalization (Lin et al., 2013; Snowden et al., 2017). Regular physician visits are an important component of effective outpatient management and can improve care quality and reduce costs (Friedberg et al., 2010). However, these encounters can be challenging for both PwD and clinicians due to complex health profiles, communication difficulties, and worsening functional impairment (Bunn et al., 2017; Pratt et al., 2006; Prorok et al., 2017).

Caregiver Involvement in the Physician Visit

Caregiver involvement during the physician visit can help address these barriers, optimizing downstream outcomes for PwD (Wolff & Roter, 2011; Wolff et al., 2015, 2017). Health care providers typically rely on caregivers to provide effective care, including to obtain accurate histories and monitor treatment responses (Bunn et al., 2017; Green et al., 2019). Caregivers can provide familiar assistance during physical examinations and facilitate the PwD’s communication with the provider. Caregiver involvement in ambulatory care for PwD may improve health care resource use for comorbidities and improve outcomes (MacNeil-Vroomen et al., 2020; National Academies of Sciences et al., 2021). Despite this, numerous studies have reported caregiver dissatisfaction in their involvement in health care processes (Prorok et al., 2017; Sperber et al., 2019), sparking calls for health care systems to actively increase caregiver inclusion (Sperber et al., 2019). A growing literature has demonstrated that the care network as a whole must be considered, not just the primary helper (Ali et al., 2022; Andersson & Monin, 2018; Penrod et al., 1995). However, there is limited understanding of how care network factors associate with health care access and care quality for PwD.

Caregiving and Dementia Care in Underserved Populations

Clarification of the relationship of care network factors with dementia outcomes is particularly critical for historically underserved groups, including non-White populations (e.g., Alzheimer’s Association, 2021). A substantial literature has documented disparities in dementia care access and quality for minoritized older adults. Despite higher dementia prevalence (Mayeda et al., 2016), Black and Hispanic older adults with dementia are more likely than non-Hispanic Whites (NHW) to be undiagnosed and unaware of their condition (Lin et al., 2021). There are also racial and ethnic differences in the caregiving experience. Care networks for non-White PwD provide more intensive assistance and report more care demands than their NHW counterparts (Ali et al., 2022; Mehta & Yeo, 2018), but may use fewer formal services (Alzheimer’s Association, 2021). Hispanic older adults are more likely than NHWs to receive critical assistance from an adult child caregiver (Silverstein & Litwak, 1993). To help inform care models and caregiver policy and to address existing disparities, it is a priority to examine racial/ethnic differences in the link between care network factors and the outpatient care experience for PwD.

Theoretical Framework: Task-Specific Caregiving Model

The task-specific caregiving model (Litwak, 1985; Messeri et al., 1993) is one approach to describing how care networks for older adults form and function. The model holds that for any given caregiver, that role develops from an existing affiliation with the care recipient combined with the ability to meet a new task-specific care need. Different members of a care network provide different types of assistance and cannot be easily substituted for each other (Litwak, 1985). Having a larger potentially available network―including immediate family, as well as affiliated kin and nonkin―is therefore advantageous for receiving the full range of needed help (Silverstein & Litwak, 1993). Previous research has shown that structural aspects of the larger care network, including size, are related to older adult well-being (Andersson & Monin, 2018), care quality at end-of-life (Lei et al., 2021), and health-related behaviors (Lin et al., 2019). However, few population-level studies have examined associations between care network structure and physician visit experiences among PwD.

In the current study, we extend previous work on caregiving networks for PwD (Spillman et al., 2020) by examining how potential care network size affects outpatient medical care access and quality. We examine these associations both cross-sectionally and longitudinally, as care network structures are a response both to the initial development of dementia and to evolving care needs as the disease progresses. Because caregivers for PwD frequently provide help across both daily functioning and health care domains (Spillman et al., 2020; Wolff & Spillman, 2014) and because of reported racial/ethnic differences in task-specific involvement (Silverstein & Litwak, 1993), we include a measure of caregiver involvement with daily functioning tasks as a proportion of the potential care network. Measuring the relative extent of active network involvement allows us to separately consider the impact of these active caregivers, who may also be assisting at the physician visit (Spillman et al., 2020; Wolff & Spillman, 2014). We anticipate that a greater numbsssser of potentially available caregivers at baseline will be associated with having assistance at future health care visits. We further assess whether race/ethnicity moderates these associations. Given limited previous findings in the area of caregiving and primary care patient outcomes, testing for moderation by race/ethnicity was exploratory. Based on existing literature, we expected that racial/ethnic differences in care network size and composition may contribute to differential associations with physician visit outcomes.

Method

Data

Data come from the National Health and Aging Trends Study (NHATS), a nationally representative longitudinal survey of Medicare enrollees aged ≥65 years. The NHATS study started in 2011; in 2015, the sample was replenished. Non-Hispanic Black (NHB) and Hispanic respondents were oversampled (Freedman & Kasper, 2019). Annual, in-person interviews cover sociodemographic, economic, and health domains, and proxies report when respondents are unavailable (e.g., due to health conditions). For respondents in residential care, a facility questionnaire is completed. Details of study procedures are published elsewhere (Kasper & Freedman, 2020). The current study used data from 2015 (NHATS Round 5) to 2019 (Round 9) as baseline and follow-up, a time frame aligning with the 4–5-year average survival for individuals diagnosed with mild dementia (Mayeda et al., 2017). In 2015, approximately 18% were proxy respondents (Spillman et al., 2020).

Dementia Classification

NHATS investigators classified respondents into one of the three dementia categories (for details, see Kasper et al., 2013). Probable dementia was assigned to those who reported a clinician-determined diagnosis of Alzheimer’s disease or dementia; scored ≥2 on the AD8 questionnaire, administered to proxies (Galvin et al., 2006); or scored ≥2 standard deviation (SD) below the mean on tests assessing memory, orientation, and executive functioning. Possible dementia was assigned to individuals scoring ≤1.5 SD below the mean on any one cognitive domain. All other respondents were classified as having no dementia. Based on our interest in the full spectrum of cognitive impairment in community-dwelling adults and to enhance statistical power, we included participants with both probable and possible dementia at baseline.

Analytic Subpopulation

Of 8,334 (unweighted) NHATS respondents in 2015, most lived in the community (n = 7,070), followed by nonnursing home residential care settings (n = 565). For the current study, we excluded respondents who were residents of nursing home facilities (n = 403), who had only facility questionnaires (n = 136), or who were deceased (n = 296). Given our focus, we excluded 5,535 respondents with no dementia at baseline and those with missing data on covariates of interest (unweighted n = 28). We included only individuals who reported race/ethnicity as NHW, NHB, or Hispanic. The final analytic sample consisted of 1,520 respondents (unweighted n; NHW = 885, NHB = 478, Hispanics = 157), which is the weighted equivalent of 5,863,846 adults ≥65 years (4,215,941; 850,169; and 797,737; for NHWs, NHBs, and Hispanics, respectively).

At follow-up in 2019, 585 respondents were deceased, and 318 were lost to follow-up. Deceased or attrited individuals were more likely to be older, meet criteria for possible dementia, and endorse mental health symptoms (Supplementary Table 1). Deceased individuals reported more chronic conditions and were more likely to be NHW. Eighteen respondents completed the facility questionnaire only and were excluded. The analytic subpopulation for the end wave was n = 599.

Analyses accounted for the complex sampling design, including probability weighting, clustering, and stratification (Kasper & Freedman, 2020). We used baseline sampling weights to allow generalization to the target population of Medicare enrollees (age ≥65) in 2015.

Primary Outcomes (2015 and 2019; Rounds 5 and 9)

Access to a usual source of care was measured with interview questions (yes/no): (a) Is there a doctor that you think of as your regular doctor, that is, a doctor you usually go to when you are sick and need advice about your health? (b) Have you seen your regular doctor/another doctor within the last year? Respondents who had seen a provider in the last year were asked: (c) Did anyone sit in with you and your doctor during your visits? Affirmative responses were followed by a prompt to consider the person who most often accompanied them when responding to four questions (yes/no): During those visits in the last year, did that individual (d) Help you with getting on the exam table, dressing, and undressing? (e) Remind you about things you wanted to ask or tell the doctor? (f) Ask or tell the doctor things for you? And (g) Help you understand what the doctor was saying? Analytic Ns varied for each of the seven outcomes due to skipping patterns; assessment of outcomes d–g were conditioned on an affirmative response to item (c) (i.e., being accompanied to physician visit).

Primary Predictors (2015; Round 5)

NHATS respondents were queried throughout the interview to create a comprehensive roster of affiliated individuals, including spouses/partners, children, all household members, emergency contacts, persons cared for by the respondent, and up to five close confidantes. The roster further included any additional persons providing assistance, including those receiving payment (73% had 0 paid helpers; on average, 6.5% of helpers were paid). Rosters are cumulative across waves; therefore, the 2015 roster includes all individuals mentioned in interviews since 2011. We summed the number of unique individuals to calculate the size of the potential care network (“potential care network members” or PCNMs). Second, to derive the proportion of PCNMs currently providing assistance in daily living, we calculated the number of individuals helping in the past month with household tasks (e.g., medication and meal preparation), self-care (dressing and toileting), and/or mobility assistance (Freedman et al., 2011), then divided by the total potential care network size.

Covariates (2015; Round 5)

We adjusted for dementia classification at baseline using the criteria described earlier, and controlled for demographic, socioeconomic, and health characteristics in our models. Age was continuous. Race/ethnicity (NHW, NHB, and Hispanic), sex (male/female), and education (≤eighth grade, >eighth grade schooling without a high school diploma, high school diploma, some college, and college degree or beyond) were categorical variables. We adjusted for annual household income and enrollment in supplemental insurance (Medicaid, Tricare, Medigap, or Other). Physical health was assessed using a count of nine common conditions. Mental health was assessed by item means on the Patient Health Questionnaire two-item depression screen (PHQ-2; range 1–4) and the Generalized Anxiety Disorder two-item screen (GAD-2, range 1–4). Higher scores indicate more severe symptomatology (Kroenke et al., 2009).

Analytic Procedures

Analyses were conducted in three steps. First, we generated descriptive statistics to characterize the target population. To examine differences by race/ethnicity, we used survey-adjusted chi-squared tests for categorical variables and survey-adjusted t tests for continuous measures (Table 1). Second, we applied a series of logistic regression models to test partially adjusted and fully adjusted associations between the primary predictors and each of the seven outcomes, measured at baseline (Table 2) and again 4 years later (Table 3). For each outcome, we run two regression models to test the main effects of the predictors. Additionally, to test for modifications by race/ethnicity, we run two additional regression models adding interactions between race/ethnicity and each predictor, independently, to the fully adjusted models. For each outcome, the first model tested associations with network size and proportion of PCNMs involved in help and adjusted for sociodemographic characteristics and dementia severity. Model 2, the fully adjusted model, additionally accounted for income, insurance status, and physical and mental health factors. To facilitate the interpretation of findings, we used post hoc ANOVA techniques to estimate and plot average marginal probabilities for each outcome over the size of potential care network continua (Figure 1A and B). Finally, to assess for differential affects on the physician visit experience across racial and ethnic groups, we tested modification effects in Model 2, adding interaction terms between race/ethnicity and (a) care network size, and (b) proportion of PCNMs involved with help, independently. Results of interaction tests are provided in Table 4. All analyses were done using survey functionalities in Stata MP (version 17.0; StataCorp, 2021).

Table 1.

Weighted Characteristics of Target Population at Baseline (Year 2015)

NHW
na = 885		 NHB
na = 478		 Hispanic
na = 157		 Total
na = 1,520		 p Value
% (95% CI)
Dementia classification
Possible dementia 50.68 (47.49; 53.87) 57.90 (52.28; 63.52) 46.60 (36.69; 56.51) 51.17 (48.66; 53.68) .148
Probable dementia 49.32 (46.13; 52.51) 42.10 (36.48; 47.72) 53.40 (43.49; 63.31) 48.83 (46.32; 51.34)
Sex
Male 47.61 (43.44; 51.78) 42.39 (37.06; 47.72) 44.96 (35.66; 54.27) 46.49 (43.39; 49.59) .446
Female 52.39 (48.22; 56.56) 57.61 (52.28; 62.94) 55.04 (45.73; 64.34) 53.51 (50.41; 56.61)
Education
≤Eighth grade 12.41 (9.09; 15.72) 26.70 (19.86; 33.54) 63.76 (54.16; 73.36) 21.47 (18.45; 24.48) <.0001
>Eighth grade, no high school diploma 12.25 (9.81; 14.69) 24.98 (20.73; 29.23) 13.79 (7.36; 20.22) 14.30 (12.30; 16.31)
High school diploma 36.94 (32.66; 41.23) 24.53 (19.54; 29.53) 10.92 (4.87; 16.96) 31.60 (28.28; 34.92)
Some college 20.04 (16.36; 23.73) 15.77 (11.31; 20.23) 8.06 (2.18; 13.94) 17.79 (14.99; 20.59)
College degree or beyond 18.35 (15.16; 21.55) 8.02 (4.80; 11.24) 3.48 (0.75; 6.21) 14.83 (12.59; 17.08)
Insurance
Medicaid 12.59 (9.71; 15.48) 41.74 (33.85; 49.62) 52.49 (39.55; 65.43) 22.24 (19.18; 25.31) <.0001
Tricare 6.46 (4.13; 8.79) 3.26 (1.17; 5.35) 0.00 (0.00; 0.00) 5.12 (3.39; 6.85)
Medigap 49.78 (45.85; 53.71) 23.37 (17.55; 29.18) 17.17 (9.39; 24.95) 41.52 (38.17; 44.86)
Other 31.16 (27.16; 35.17) 31.63 (24.80; 38.46) 30.34 (20.61; 40.07) 31.12 (28.09; 34.15)
Mean (95% CI)
Age (years) 80.98 (80.37; 81.59) 78.08 (77.28; 78.88) 78.83 (77.59; 80.08) 80.27 (79.79; 80.75) <.0001
Income (dollars) 48,653 (41,894; 55,412) 24,541 (21,287; 27,794) 17,363 (13,402; 21,324) 40,900 (35,931; 45,870) <.0001
Chronic conditions count 2.75 (2.62; 2.87) 2.71 (2.51; 2.9) 2.88 (2.59; 3.18) 2.76 (2.66; 2.86) .62
PHQ-2 item mean (range 1–4) 1.72 (1.65; 1.79) 1.78 (1.67; 1.89) 1.94 (1.77; 2.11) 1.76 (1.7; 1.82) .07
GAD-2 item mean (range 1–4) 1.64 (1.56; 1.72) 1.59 (1.5; 1.68) 1.86 (1.68; 2.04) 1.66 (1.59; 1.73) .03
Size of potential care network 5.74 (5.49; 5.98) 6.69 (6.35; 7.03) 6.89 (6.12; 7.66) 6.03 (5.82; 6.24) <.0001
Proportion of PCNMs involved with helpb 34.41 (32.31; 36.51) 30.22 (28.03; 32.41) 29.89 (25.31; 34.47) 33.19 (31.55; 34.82) .02
Open in a new tab

Notes: To examine overall differences, we used survey-adjusted chi-squared tests for categorical variables and survey-adjusted t tests for continuous measures. NHW = non-Hispanic White; NHB = non-Hispanic Black; CI = confidence interval; PHQ-2 = Patient Health Questionnaire two-item depression screening; GAD-2 = Generalized Anxiety Disorder two-item screening; PCNM = potential care network members.

ans are unweighted; all estimates, however, are weighted using National Health and Aging Trends Study (NHATS) sampling weights.

bTransformed to %.

Table 2.

Logistic Regression Models Showing Associations Between Potential Care Network Size, % Providing Help, and Physician Visit Outcomes Measured at Baseline

Baseline: 2015
Model 1 Model 2
OR (95% CI)
Have a USC (N_2015 = 1,517)
Size of potential care network 1.01 (0.88; 1.14) 0.98 (0.87; 1.11)
Proportion of PCNMs involved with helpa 0.99* (0.98; 1.00) 0.99 (0.98; 1.00)
Visit physician in last year (N_2015 = 1,514)
Size of potential care network 1.03 (0.92; 1.14) 0.98 (0.89; 1.08)
Proportion of PCNMs involved with help 1.00 (0.99; 1.01) 1.00 (0.99; 1.01)
Had someone sit with respondent at physician visit (N_2015 = 1,391)
Size of potential care network 1.22*** (1.14; 1.31) 1.22*** (1.13; 1.31)
Proportion of PCNMs involved with help 1.03*** (1.02; 1.04) 1.03*** (1.02; 1.04)
Provided help with getting on the exam table, dressing, and undressing (N_2015 = 934)
Size of potential care network 1.09** (1.03; 1.15) 1.07* (1.01; 1.14)
Proportion of PCNMs involved with help 1.02*** (1.01; 1.03) 1.02*** (1.01; 1.03)
Reminded about things respondent wanted to ask or tell the physician (N_2015 = 938)
Size of potential care network 1.03 (0.97; 1.10) 1.02 (0.96; 1.08)
Proportion of PCNMs involved with help 1.01 (1.00; 1.02) 1.01 (1.00; 1.01)
Asked or told the physician things for the respondent (N_2015 = 938)
Size of potential care network 1.12* (1.02; 1.23) 1.10* (1.00; 1.21)
Proportion of PCNMs involved with help 1.02** (1.01; 1.04) 1.02** (1.01; 1.03)
Helped respondent understand what the physician was saying (N_2015 = 941)
Size of potential care network 0.99 (0.92; 1.06) 0.96 (0.90; 1.03)
Proportion of PCNMs involved with help 1.01 (1.00; 1.02) 1.01 (1.00; 1.01)
Open in a new tab

Notes: Model 1: Adjusted for age, sex, education, race/ethnicity, and dementia classification (probable vs. possible dementia). Model 2: Additional inclusion insurance, income, count of comorbid conditions, PHQ-2, and GAD-2. OR = odds ratio; CI = confidence interval; PCNM = potential care network members; USC = usual source of care; PHQ-2 = Patient Health Questionnaire two-item depression screening; GAD-2 = Generalized Anxiety Disorder two-item screening.

aTransformed to %.

*p < .05; **p < .01; ***p < .001.

Table 3.

Logistic Regression Models Showing Associations Between Potential Care Network Size, % Providing Help, and Physician Visit Outcomes Measured at End Wave (2019)

End wave: 2019
Model 1 Model 2
OR (95% CI)
Have a USC (N_2019 = 598)
Size of potential care network 1.01 (0.88; 1.16) 1.02 (0.89; 1.17)
Proportion of PCNMs involved with helpa 1.01 (0.99; 1.02) 1.01 (1.00; 1.03)
Visit physician in last year (N_2019 = 596)
Size of potential care network 1.14 (0.98; 1.32) 1.13 (0.96; 1.31)
Proportion of PCNMs involved with help 0.99 (0.98; 1.01) 0.99 (0.98; 1.01)
Had someone sit with respondent at physician visit (N_2019 = 548)
Size of potential care network 1.18** (1.07; 1.30) 1.17** (1.04; 1.30)
Proportion of PCNMs involved with help 1.02* (1.00; 1.03) 1.02* (1.00; 1.03)
Provided help with getting on the exam table, dressing, and undressing (N_2019 = 373)
Size of potential care network 1.09 (1.00; 1.18) 1.06 (0.97; 1.16)
Proportion of PCNMs involved with help 1.02** (1.01; 1.03) 1.02* (1.00; 1.03)
Reminded about things respondent wanted to ask or tell the physician (N_2019 = 374)
Size of potential care network 1.08 (0.99; 1.18) 1.08 (1.00; 1.18)
Proportion of PCNMs involved with help 1.00 (0.99; 1.01) 1.00 (0.99; 1.02)
Asked or told the physician things for the respondent (N_2019 = 375)
Size of potential care network 1.10 (0.99; 1.22) 1.09 (0.97; 1.22)
Proportion of PCNMs involved with help 1.01 (0.99; 1.03) 1.01 (0.99; 1.03)
Helped respondent understand what the physician was saying (N_2019 = 373)
Size of potential care network 1.08 (0.96; 1.21) 1.05 (0.94; 1.18)
Proportion of PCNMs involved with help 1.00 (0.99; 1.02) 1.00 (0.98; 1.01)
Open in a new tab

Notes: Model 1: Adjusted for age, sex, education, race/ethnicity, and dementia classification (probable vs. possible dementia). Model 2: Additional inclusion of insurance, income, count of comorbid conditions, PHQ-2, and GAD-2 OR = odds ratio; CI = confidence interval; PCNM = potential care network members; USC = usual source of care; PHQ-2 = Patient Health Questionnaire two-item depression screening; GAD-2 = Generalized Anxiety Disorder two-item screening.

aTransformed to %.

*p < .05; **p < .01.

Figure 1.

Figure 1.

Open in a new tab

(A) Predicted marginal associations (marginal probabilities × 100 to interpret in percent metric) of the size of the potential care network with physician visit outcomes of interest at baseline (2015) and end wave (2019). Model 1 (M1): adjusted for age, sex, education, race/ethnicity, dementia severity (probable vs. possible dementia) and proportion of potential care network members involved with help as %; M2: additional inclusion of insurance, income, count of comorbid conditions, PHQ-2, and GAD-2. (B) Predicted marginal associations (marginal probabilities × 100 to interpret in percent metric) of the size of the potential care network with four types of caregiver assistance during the physician visit at baseline (2015) and end wave (2019). USC = usual source of care; PHQ-2 = Patient Health Questionnaire two-item depression screening; GAD-2 = Generalized Anxiety Disorder two-item screening.

Table 4.

Tests for Modifications in Associations Between Potential Care Network Size and Physician Visit Outcomes Measured at Baseline and End Wave

Race/ethnicity × Size of potential care network
Baseline 2019
Fully adjusted model + Interaction
Have a usual source of care F(2, 55) = 3.99, p = .0241 F(2, 55) = 3.60, p = .0339
Visit physician in past year F(2, 55) = 0.52, p = .5999 F(2, 55) = 0.50, p = .6089
Had someone sit with respondent at physician visit F(2, 55) = 0.79, p = .4587 F(2, 55) = 0.89, p = .4181
Provided help with getting on the exam table, dressing, and undressing F(2, 55) = 0.05, p = .9475 F(2, 55) = 1.54, p = .2234
Reminded about things respondent wanted to ask or tell the physician F(2, 55) = 0.45, p = .6378 F(2, 55) = 1.39, p = .2585
Asked or told the physician things for the respondent F(2, 55) = 1.04, p = .3599 F(2, 55) = 0.05, p = .9501
Helped respondent understand what the physician was saying F(2, 55) = 0.32, p = .7262 F(2, 55) = 0.04, p = .9573
Race/ethnicity × Proportion of PCNMs involved with helpa
Baseline 2019
Fully adjusted model + Interaction
Have a usual source of care F(2, 55) = 2.90, p = .0635 F(2, 55) = 3.47, p = .0381
Visit physician in past year F(2, 55) = 6.44, p = .0031 F(2, 55) = 0.70, p = .4987
Had someone sit with respondent at physician visit F(2, 55) = 0.31, p = .7339 F(2, 55) = 1.83, p = .1700
Provided help with getting on the exam table, dressing, and undressing F(2, 55) = 0.16, p = .8563 F(2, 55) = 0.43, p = .6526
Reminded about things respondent wanted to ask or tell the physician F(2, 55) = 0.20, p = .8159 F(2, 55) = 0.65, p = .5284
Asked or told the physician things for the respondent F(2, 55) = 1.00, p = .3757 F(2, 55) = 0.56, p = .5762
Helped respondent understand what the physician was saying F(2, 55) = 0.11, p = .8979 F(2, 55) = 0.93, p = .3992
Open in a new tab

Notes: Models are adjusted for age, sex, education, dementia severity (probable vs. possible dementia), insurance, income, count of comorbid conditions, PHQ-2, and GAD-2. Models are adjusted for % of individuals in network providing help when race/ethnicity × size of potential care network is considered, and for size of potential care network when race/ethnicity × proportion of PCNMs involved with help is considered. PCNM = potential care network members; PHQ-2 = Patient Health Questionnaire two-item depression screening; GAD-2 = Generalized Anxiety Disorder two-item screening.

aTransformed to %.

Sensitivity Analyses

To assess the robustness of our findings, we replicated all analyses using a different NHATS cohort with data from respondents enrolled in earlier NHATS waves (baseline wave 2011, end wave 2015). Findings are presented as Supplementary Material.

Results

Descriptives

At baseline, the mean age was 80.3 years (SD 9.4), 53.5% were female (Table 1), and respondents reported an average of 2.76 comorbid medical conditions. Most respondents were NHW; 14.5% and 13.6% were NHB and Hispanic, respectively. Compared to NHB and Hispanic respondents, NHW had a higher average income and more formal education, were less likely to use Medicaid. Overall, Hispanic respondents had larger potential care networks (M = 6.89, SD 3.57, p < .0001) and a smaller proportion of PCNMs currently providing help (29.9%, compared to 34.4% for NHW and 30.2% for NHB, p = .02).

At baseline, 93.3% reported having a regular physician, 92.4% had seen a physician in the last year, and 65.3% of those respondents reported being accompanied to a visit. During these accompanied visits, 32.3% received help with mobility and dressing; 72% received reminders about things to ask or tell the physician; 78.5% said the caregiver communicated with the physician on their behalf; and 74.6% said the caregiver helped them understand medical instructions. These percentages were statistically consistent over time and across racial/ethnic groups. Weighted frequencies for these outcomes at baseline and follow-up, stratified by race/ethnicity, are visualized in Supplementary Figure 1A and B.

Cross-sectional Associations Between Potential Care Network Size and Outpatient Medical Care Access and Quality

In partially adjusted models including both primary predictors and adjusting for age, sex, education, race/ethnicity, and dementia classification, network size and the proportion of PCNMs helping with daily tasks were positively associated with being accompanied to a physician visit (odds ratio [OR]SIZE = 1.22; 95% confidence interval [CI] = [1.13; 1.31] and OR%help = 1.03 [1.02; 1.04]); receiving help with getting on the exam table and in dressing/undressing (ORSIZE = 1.09 [1.03; 1.15] and OR%help = 1.02 [1.01; 1.03]); and the PCNM conveying information to the physician on behalf of the respondent (ORSIZE = 1.12 [1.02; 1.23] and OR%help = 1.02 [1.01; 1.04]; Table 2).

Longitudinal Associations Between Potential Care Network Size and Outpatient Medical Care Access and Quality

In longitudinal analyses, these associations remained significant for being accompanied to the visit (ORSIZE = 1.18 [1.07; 1.30] and OR%help = 1.02 [1.00; 1.03]; Table 3). Additionally, the proportion of PCNMs assisting with daily tasks was associated with receiving help with getting on the exam table and in dressing/undressing (OR%help = 1.02 [1.01; 1.03]). Findings were robust to adjustments for type of insurance, household income, comorbid medical conditions, and physical and mental health status (Model 3 in Tables 2 and 3). To facilitate interpretation, plots of the estimated probabilities and 95% CI are shown in Figure 1A and B.

Modifications by Race/Ethnicity

Analyses revealed no consistent moderation (interaction) effects for race/ethnicity in associations of outcomes with network size or proportional involvement of PCNMs in help, across cross-sectional and longitudinal models. All tested interactions of statistically significant predictors (presented earlier) were statistically nonsignificant (Table 4).

Sensitivity Analyses

As shown in Supplementary Tables 2–4, the results of the validation analyses using an earlier NHATS cohort (2011–2015) were highly consistent with the primary results, providing further evidence for the robustness of our findings.

Discussion

Leveraging rich longitudinal data from a nationally representative survey, we identified associations of care network characteristics with key elements of the physician visit for PwD. Findings revealed independent effects of both predictors on aspects of caregiver assistance during the visit, both cross-sectionally and at 4-year follow-up, highlighting the importance of care network characteristics for the physician visit experience across disease stages. We add to existing literature through several primary findings.

First, in fully adjusted models, we found that the size of the potential care network is positively associated with having a caregiver present and with receiving specific types of assistance during the physician visit. Our findings are consistent with growing evidence for the importance of the potential care network for health care outcomes (Choi et al., 2021; Lei et al., 2021). At 4-year follow-up, network size was associated only with being accompanied to the physician visit. This suggests that as dementia worsens, having more potential helpers remains germane for health care access (e.g., providing transportation or managing appointments), but is less crucial for quality―that is, assistance during the medical encounter. It is plausible that as complexity increases with disease progression, nonprimary caregivers may be less familiar with important details of the patient and thus less able to assist in essential ways. Additionally, larger networks may complicate care coordination, leading to confusion and worse outcomes. Notably, a recent study found that a larger network size was linked to more depression and anxiety in older adults, especially as medical comorbidities increase, raising the possibility that having too many involved caregivers may undermine care recipient well-being (Andersson & Monin, 2018). Taken together, evidence suggests that care network size is an important structural characteristic in need of further investigation to guide policy and dementia care recommendations.

Second, these associations held for the proportion of PCNM helping the PwD with daily tasks and were independent of physical and mental health factors, suggesting that this finding is not due to disability necessitating assistance in daily life and during the medical encounter. Our findings are consistent with previous work showing that dementia caregivers, in particular, tend to be “generalists,” assisting with daily functioning as well as medical care tasks (Spillman et al., 2020; Wolff & Spillman, 2014). Importantly, care network characteristics were not consistently associated with every type of assistance received during the physician encounter; they are most critical for mobility help and managing communication with the physician, and less related to more collaborative forms of support (i.e., reminders and facilitating understanding). At follow-up, the adjusted association held between the proportion of daily involved PCNMs and mobility help, suggesting that having more involved caregivers early in the disease may help ensure the future availability of this intimate form of assistance.

Our findings have implications for clinical interventions with families to assist with planning and for caregiver policy to support this early, hands-on involvement, for example, flexible work schedules and parental care leave. Our results further highlight the critical need for research to describe how care networks can effectively function to help optimize health outcomes for the PwD. Still, characterization of types of support during the physician visit and the significance of each to health care outcomes for PwD remains a fruitful area for further research.

Surprisingly, neither potential network size nor the proportion of PCNMs helping with daily tasks were consistently related to measures of access to care, a finding inconsistent with previous work. For example, higher intensity of informal caregiving for PwD was associated with lower use of home and nursing care services, but increased use of outpatient visits across European Union countries (Bremer et al., 2017). Divergent findings may reflect differences among caregiving cultures and health care systems, and more research is needed to clarify how care network characteristics affect the utilization of health services, both primary care and specialty care.

Finally, while we found racial/ethnic differences in network characteristics, associations with physician visit experiences were not moderated by race/ethnicity. This was surprising, given the large body of literature demonstrating differences in family structures and informal caregiving arrangements by racial/ethnic group (Li & Fries, 2005; Lum, 2005; Maldonado, 2017). Our findings suggest that more research is needed prior to focusing on cultural tailoring of caregiving policies and interventions to improve the physician visit experience for PwD. Our findings confirm that the characteristics of the care network have critical implications for the physical and psychological health of the PwD across racial/ethnic groups (Epps et al., 2019). Future efforts should focus on understanding culturally specific family structures, preferences for caregiver involvement in medical care, and community health literacy needs (Andruske & O’Connor, 2020; Dilworth-Anderson et al., 2020; Hill et al., 2015) as they relate to health care outcomes. For example, future studies to clarify pathways to assuming caregiving tasks in the medical environment, and how pathways might differ by race/ethnicity, can help inform health literacy interventions and provider education to improve the ability to partner with caregivers (Wyman et al., 2020).

Taken together, our findings provide additional support for the task-specific model of caregiving (Litwak, 1985; Messeri et al., 1993) and underline the crucial role of caregiver assistance during the medical visit (Vos et al., 2018). Consistent with this model, the number of available potential helpers is beneficial for some physician visit tasks (e.g., having someone at the visit) but not others (e.g., specific types of assistance). Our work is also in line with recent efforts to expand the caregiving research focus beyond one primary caregiver (e.g., Ali et al., 2022; Cleary et al., 2021; Spillman et al., 2020). A comprehensive and nuanced understanding of links between care network characteristics and health care utilization and quality is critical to develop theory and inform interventions to improve outcomes—particularly in PwD of color (Dilworth-Anderson et al., 2020)―and to inform the development of dementia care models and clinical guidelines appropriate for this complex population (Boyd et al., 2019).

Limitations

A few limitations are worth noting. First, NHATS dementia classification was derived with the primary goal of maximizing sensitivity in case identification (Kasper et al., 2013) and has been widely applied (e.g., Freedman et al., 2018; Kasper et al., 2015). Nevertheless, our sample likely includes respondents who do not meet formal diagnostic criteria for dementia. In sensitivity analyses using the probable dementia group only, results were largely unchanged but affected by limited precision of estimates (data available from the authors). Our models adjust for dementia classification, but future research incorporating criterion-standard clinical diagnoses will help clarify the impact of dementia severity on outcomes.

Second, our approach to identifying the potential care network may not have captured all potential helpers and could not account for the dynamics of care networks over time (Spillman et al., 2020). Furthermore, the network list included paid helpers. While excluding them did not change our findings (available from the authors), we acknowledge that understanding the relative effects of paid and unpaid caregivers on health outcomes remains an important area for future research.

Third, our sample was limited to three racial/ethnic groups; moreover, the NHATS population includes only Medicare-eligible older adults, which means that Hispanics who do not qualify for Medicare due to immigration status were not included. Future research in this area, including older adults from additional racial/ethnic populations and a more inclusive Hispanic sample is critical. Despite these limitations, our study contributes to the growing literature on dementia caregiving in the medical setting and serves to further illuminate the common, complex, and often hidden interactions between health care and caregiving for diverse older adults with dementia.

Conclusion

Larger networks and a higher proportion of PCNMs assisting with daily tasks predicted caregiver presence during the physician visit for older PwD across NHB, Hispanic, and NHW populations. Our findings have implications for health system policy and for interventions to improve caregiver involvement in care throughout the disease progression and highlight the need for continued research on care networks and health care for PwD.

Supplementary Material

gbac150_suppl_Supplementary_Material
Click here for additional data file. (390KB, pdf)

Contributor Information

Mary F Wyman, W.S. Middleton Memorial Veterans Hospital, Madison, Wisconsin, USA; School of Medicine & Public Health, University of Wisconsin, Madison, Wisconsin, USA.

Irving E Vega, Department of Translational Neuroscience, College of Human Medicine, Michigan State University, Grand Rapids, Michigan, USA.

Laura Y Cabrera, Department of Engineering Science and Mechanics, Rock Ethics Institute, and Huck Institutes of Life Sciences, Pennsylvania State University, University Park, Pennsylvania, USA.

Reza Amini, Department of Public Health and Health Sciences, College of Health Sciences, University of Michigan-Flint, Flint, Michigan, USA.

Kyeongmo Kim, School of Social Work, Virginia Commonwealth University, Richmond, Virginia, USA.

Wassim Tarraf, Institute of Gerontology, and Department of Healthcare Sciences, Wayne State University, Detroit, Michigan, USA.

Funding

This work was supported by the Michigan Center for Contextual Factors in Alzheimer’s Disease (MCCFAD) [P30 AG059300], funded by the National Institute on Aging of the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Department of Veterans Affairs. M. F. Wyman’s work was supported with resources and the use of facilities at the W.S. Middleton Memorial Veterans Hospital, Madison, WI (GRECC # 001-2022) and through a Career Development Award (IK2 HX003080) funded by the U.S. Department of Veterans Affairs Health Services Research and Development Service, VA Office of Research and Development.

Conflict of Interest

None declared.

Author Contributions

All authors jointly planned the study and revised the manuscript, approving the final version. M. F. Wyman, L. Y. Cabrera, I. E. Vega, and W. Tarraf drafted the initial manuscript. W. Tarraf supervised data analysis, and together with R. Amini and K. Kim performed all statistical procedures.

References

  1. Ali, T., McAvay, G., Monin, J. K., & Gill, T. M. (2022). Patterns of caregiving among older adults with and without dementia: A latent class analysis. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 77(1), S74–S85. doi: 10.1093/geronb/gbab237 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alzheimer’s Association. (2021). Special report: Race, ethnicity, and Alzheimer’s in America. Alzheimers and Dementia, 17(3), 71–104. [Google Scholar]
  3. Alzheimer’s Association. (2022). Alzheimer’s disease facts and figures. Alzheimers and Dementia, 18(4), 700–789. doi: 10.1002/alz.12638 [DOI] [PubMed] [Google Scholar]
  4. Andersson, M. A., & Monin, J. K. (2018). Informal care networks in the context of multimorbidity: Size, composition, and associations with recipient psychological well-being. Journal of Aging and Health, 30(4), 641–664. doi: 10.1177/0898264316687623 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Andruske, C. L., & O’Connor, D. (2020). Family care across diverse cultures: Re-envisioning using a transnational lens. Journal of Aging Studies, 55, 100892–100892. doi: 10.1016/j.jaging.2020.100892 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bell, J. F., Fitzpatrick, A. L., Copeland, C., Chi, G., Steinman, L., Whitney, R. L., Atkins, D. C., Bryant, L. L., Grodstein, F., Larson, E., Logsdon, R., & Snowden, M. (2015). Existing data sets to support studies of dementia or significant cognitive impairment and comorbid chronic conditions. Alzheimers and Dementia, 11(6), 622–638. doi: 10.1016/j.jalz.2014.07.002 [DOI] [PubMed] [Google Scholar]
  7. Boyd, C., Smith, C. D., Masoudi, F. A., Blaum, C. S., Dodson, J. A., Green, A. R., Kelley, A., Matlock, D., Ouelett, J., Rich, M. W., Schoenborn, N. L., & Tinetti, M. E. (2019). Framework for decision-making for older adults with multiple chronic conditions: Executive summary of action steps for the AGS guiding principles on the care of older adults with multimorbidity. Journal of the American Geriatrics Society, 67(4), 665–673. doi: 10.1111/jgs.15809 [DOI] [PubMed] [Google Scholar]
  8. Bremer, P., Challis, D., Hallberg, I. R., Leino-Kilpi, H., Saks, K., Vellas, B., Zwakhalen, S. M. G., Sauerland, D., on behalf of the RightTimePlaceCare Consortium (2017). Informal and formal care: Substitutes or complements in care for people with dementia? Empirical evidence for 8 European countries. Health Policy, 121(6), 613–622. doi: 10.1016/j.healthpol.2017.03.013 [DOI] [PubMed] [Google Scholar]
  9. Bunn, F., Burn, A. M., Goodman, C., Rait, G., Norton, S., Robinson, L., Schoeman, J., & Brayne, C. (2014). Comorbidity and dementia: A scoping review of the literature. BMC Medicine, 12, 192. doi: 10.1186/s12916-014-0192-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bunn, F., Burn, A. M., Robinson, L., Poole, M., Rait, G., Brayne, C., Schoeman, J., Norton, S., & Goodman, C. (2017). Healthcare organisation and delivery for people with dementia and comorbidity: A qualitative study exploring the views of patients, carers and professionals. BMJ Open, 7(1), 1–9. doi: 10.1136/bmjopen-2016-013067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Choi, H., Heisler, M., Norton, E. C., Langa, K. M., Cho, T.-C., & Connell, C. M. (2021). Family care availability and implications for informal and formal care used by adults with dementia in the US: Study examines family care availability and implications for informal and form care used by adults with Dementia in the United States. Health Affairs, 40(9), 1359–1367. doi: 10.1377/hlthaff.2021.00280 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cleary, J. L., Manalel, J. A., Ashida, S., Marcum, C. S., Rewley, J., & Koehly, L. (2021). Interpersonal correlates of dementia caregivers’ emotional support networks: Considering family history. Research on Aging, 44(5-6), 405–413. doi: 10.1177/01640275211026919 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dilworth-Anderson, P., Moon, H., & Aranda, M. P. (2020). Dementia caregiving research: Expanding and reframing the lens of diversity, inclusivity, and intersectionality. Gerontologist, 60(5), 797–805. doi: 10.1093/geront/gnaa050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Epps, F., Rose, K. M., & Lopez, R. P. (2019). Who’s your family?: African American caregivers of older adults with dementia. Research in Gerontological Nursing, 12(1), 20–26. doi: 10.3928/19404921-20181212-04 [DOI] [PubMed] [Google Scholar]
  15. Freedman, V. A., & Kasper, J. D. (2019). Cohort profile: The National Health and Aging Trends Study (NHATS). International Journal of Epidemiology, 48(4), 1044–1045g. doi: 10.1093/ije/dyz109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Freedman, V. A., Kasper, J. D., Cornman, J. C., Agree, E. M., Bandeen-Roche, K., Mor, V., Spillman, B. C., Wallace, R., & Wolf, D. A. (2011). Validation of new measures of disability and functioning in the National Health and Aging Trends Study. The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 66(9), 1013–1021. doi: 10.1093/gerona/glr087 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Freedman, V. A., Kasper, J. D., Spillman, B. C., & Plassman, B. L. (2018). Short-term changes in the prevalence of probable dementia: An analysis of the 2011–2015 National Health and Aging Trends Study. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 73(Suppl. 1), S48–S56. doi: 10.1093/geronb/gbx144 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Friedberg, M. W., Hussey, P. S., & Schneider, E. C. (2010). Primary care: A critical review of the evidence on quality and costs of health care. Health Affairs, 29(5), 766–772. doi: 10.1377/hlthaff.2010.0025 [DOI] [PubMed] [Google Scholar]
  19. Galvin, J. E., Roe, C. M., Xiong, C., & Morris, J. C. (2006). Validity and reliability of the AD8 informant interview in dementia. Neurology, 67(11), 1942–1948. doi: 10.1212/01.wnl.0000247042.15547.eb [DOI] [PubMed] [Google Scholar]
  20. Green, A. R., Lee, P., Reeve, E., Wolff, J. L., Chen, C. C. G., Kruzan, R., & Boyd, C. M. (2019). Clinicians’ perspectives on barriers and enablers of optimal prescribing in patients with dementia and coexisting conditions. Journal of the American Board of Family Medicine, 32(3), 383–391. doi: 10.3122/jabfm.2019.03.180335 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hill, C. V., Pérez-Stable, E. J., Anderson, N. A., & Bernard, M. A. (2015). The National Institute on Aging health disparities research framework. Ethnicity and Disease, 25(3), 245. doi:  10.18865/ed.25.3.245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kasper, J. D., & Freedman, V. A. (2020). National Health and Aging Trends Study user guide: Rounds 1–9 beta release. Johns Hopkins University School of Public Health. [Google Scholar]
  23. Kasper, J. D., Freedman, V. A., & Spillman, B. C. (2013). Classification of persons by dementia status in the National Health and Aging Trends Study. Technical Paper, 5, 1–4. [Google Scholar]
  24. Kasper, J. D., Freedman, V. A., Spillman, B. C., & Wolff, J. L. (2015). The disproportionate impact of dementia on family and unpaid caregiving to older adults. Health Affairs, 34(10), 1642–1649. doi: 10.1377/hlthaff.2015.0536 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kroenke, K., Spitzer, R. L., Williams, J. B., & Löwe, B. (2009). An ultra-brief screening scale for anxiety and depression: The PHQ–4. Psychosomatics, 50(6), 613–621. doi: 10.1176/appi.psy.50.6.613 [DOI] [PubMed] [Google Scholar]
  26. Lei, L., Gerlach, L. B., Powell, V. D., & Maust, D. T. (2021). Caregiver support and place of death among older adults. Journal of the American Geriatrics Society, 69(5), 1221–1230. doi: 10.1111/jgs.17055 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li, L. W., & Fries, B. E. (2005). Elder disability as an explanation for racial differences in informal home care. Gerontologist, 45(2), 206–215. doi: 10.1093/geront/45.2.206 [DOI] [PubMed] [Google Scholar]
  28. Lin, P.-J., Daly, A. T., Olchanski, N., Cohen, J. T., Neumann, P. J., Faul, J. D., Fillit, H. M., & Freund, K. M. (2021). Dementia diagnosis disparities by race and ethnicity. Medical Care, 59(8), 679–686. doi: 10.1097/MLR.0000000000001577 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lin, P.-J., Fillit, H. M., Cohen, J. T., & Neumann, P. J. (2013). Potentially avoidable hospitalizations among Medicare beneficiaries with Alzheimer’s disease and related disorders. Alzheimer’s and Dementia, 9(1), 30–38. doi: 10.1016/j.jalz.2012.11.002 [DOI] [PubMed] [Google Scholar]
  30. Lin, S., Faust, L., Robles-Granda, P., Kajdanowicz, T., & Chawla, N. V. (2019). Social network structure is predictive of health and wellness. PLoS One, 14(6), e0217264. doi: 10.1371/journal.pone.0217264 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Litwak, E. (1985). Helping the elderly: The complementary roles of informal networks and formal systems. Guilford Press. [Google Scholar]
  32. Lum, T. Y. (2005). Understanding the racial and ethnic differences in caregiving arrangements. Journal of Gerontological Social Work, 45(4), 3–21. doi: 10.1300/J083v45n04_02 [DOI] [PubMed] [Google Scholar]
  33. MacNeil-Vroomen, J. L., Thompson, M., Leo-Summers, L., Marottoli, R. A., Tai-Seale, M., & Allore, H. G. (2020). Health-care use and cost for multimorbid persons with dementia in the National Health and Aging Trends Study. Alzheimers and Dementia, 16(9), 1224–1233. doi: 10.1002/alz.12094 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Maldonado, L. (2017). Latinas and intergenerational caregiving: An integrative review of the literature. Journal of Transcultural Nursing, 28(2), 203–211. doi: 10.1177/1043659615623329 [DOI] [PubMed] [Google Scholar]
  35. Mayeda, E. R., Glymour, M. M., Quesenberry, C. P., Johnson, J. K., Pérez-Stable, E. J., & Whitmer, R. A. (2017). Survival after dementia diagnosis in five racial/ethnic groups. Alzheimer’s and Dementia, 13(7), 761–769. doi: 10.1016/j.jalz.2016.12.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Mayeda, E. R., Glymour, M. M., Quesenberry, C. P., & Whitmer, R. A. (2016). Inequalities in dementia incidence between six racial and ethnic groups over 14 years. Alzheimer’s and Dementia, 12(3), 216–224. doi: 10.1016/j.jalz.2015.12.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mehta, K. M., & Yeo, G. (2018). Incidence and prevalence of dementia in US race and ethnic populations. Ethnicity and the Dementias, 3,3–20. [Google Scholar]
  38. Messeri, P., Silverstein, M., & Litwak, E. (1993). Choosing optimal support groups: A review and reformulation. Journal of Health and Social Behavior, 34(2), 122–137. doi: 10.2307/2137239 [DOI] [PubMed] [Google Scholar]
  39. National Academies of Sciences, Engineering, and Medicine. (2021). Meeting the challenge of caring for persons living with dementia and their care partners and caregivers: a way forward. Washington, DC: The National Academies Press. doi:10.17226/26026 [PubMed] [Google Scholar]
  40. Penrod, J. D., Kane, R. A., Kane, R. L., & Finch, M. D. (1995). Who cares? The size, scope, and composition of the caregiver support system. Gerontologist, 35(4), 489–497. doi: 10.1093/geront/35.4.489 [DOI] [PubMed] [Google Scholar]
  41. Pratt, R., Clare, L., & Kirchner, V. (2006). “It’s like a revolving door syndrome”: Professional perspectives on models of access to services for people with early-stage dementia. Aging and Mental Health, 10(1), 55–62. doi: 10.1080/13607860500307530 [DOI] [PubMed] [Google Scholar]
  42. Prorok, J. C., Hussain, M., Horgan, S., & Seitz, D. P. (2017). “I shouldn’t have had to push and fight”: Health care experiences of persons with dementia and their caregivers in primary care. Aging and Mental Health, 21(8), 797–804. doi: 10.1080/13607863.2016.1159280 [DOI] [PubMed] [Google Scholar]
  43. Silverstein, M., & Litwak, E. (1993). A task-specific typology of intergenerational family structure in later life. Gerontologist, 33(2), 258–264. doi: 10.1093/geront/33.2.258 [DOI] [PubMed] [Google Scholar]
  44. Snowden, M. B., Steinman, L. E., Bryant, L. L., Cherrier, M. M., Greenlund, K. J., Leith, K. H., Levy, C., Logsdon, R. G., Copeland, C., Vogel, M., Anderson, L. A., Atkins, D. C., Bell, J. F., & Fitzpatrick, A. L. (2017). Dementia and co-occurring chronic conditions: A systematic literature review to identify what is known and where are the gaps in the evidence? International Journal of Geriatric Psychiatry, 32(4), 357–371. doi: 10.1002/gps.4652 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sperber, N. R., Boucher, N. A., Delgado, R., Shepherd-Banigan, M. E., McKenna, K., Moore, M., Barrett, R., Kabat, M., & Van Houtven, C. H. (2019). Including family caregivers in seriously ill Veterans’ care: A mixed-methods study. Health Affairs (Millwood), 38(6), 957–963. doi: 10.1377/hlthaff.2019.00012 [DOI] [PubMed] [Google Scholar]
  46. Spillman, B. C., Freedman, V. A., Kasper, J. D., & Wolff, J. L. (2020). Change over time in caregiving networks for older adults with and without dementia. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 75(7), 1563–1572. doi: 10.1093/geronb/gbz065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. StataCorp. (2021). Stata Statistical Software: Release 17. StataCorp LLC. [Google Scholar]
  48. Vos, J., Gerling, K., Linehan, C., Siriwardena, A. N., & Windle, K. (2018). Understanding care navigation by older adults with multimorbidity: Mixed-methods study using social network and framework analyses. JMIR Aging, 1(2), e11054. doi: 10.2196/11054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wolff, J. L., Clayman, M. L., Rabins, P., Cook, M. A., & Roter, D. L. (2015). An exploration of patient and family engagement in routine primary care visits. Health Expectations, 18(2), 188–198. doi: 10.1111/hex.12019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wolff, J. L., Guan, Y., Boyd, C. M., Vick, J., Amjad, H., Roth, D. L., Roter, G., & Roter, D. L. (2017). Examining the context and helpfulness of family companion contributions to older adults’ primary care visits. Patient Education and Counseling, 100(3), 487–494. doi: 10.1016/j.pec.2016.10.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wolff, J. L., & Roter, D. L. (2011). Family presence in routine medical visits: A meta-analytical review. Social Science and Medicine, 72(6), 823. doi: 10.1016/j.socscimed.2011.01.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wolff, J. L., & Spillman, B. (2014). Older adults receiving assistance with physician visits and prescribed medications and their family caregivers: Prevalence, characteristics, and hours of care. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69(Suppl. 1), S65–S72. doi: 10.1093/geronb/gbu119 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wyman, M. F., Voils, C. I., Trivedi, R., Boyle, L., Goldman, D., Umucu, E., Zuelsdorff, M., Johnson, A. L., & Gleason, C. E. (2020). Perspectives of Veterans Affairs mental health providers on working with older adults with dementia and their caregivers. Gerontology and Geriatrics Education, 18, 1–12. doi: 10.1080/02701960.2020.1764356 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gbac150_suppl_Supplementary_Material
Click here for additional data file. (390KB, pdf)

Articles from The Journals of Gerontology Series B: Psychological Sciences and Social Sciences are provided here courtesy of Oxford University Press

RESOURCES