Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: J Appl Gerontol. 2023 Mar 29;42(7):1387–1396. doi: 10.1177/07334648231166289

Associations Between Cognitive Impairment Severity and Barriers to Healthcare Engagement Among Older Adults

Rebecca M Lovett 1,2,3, Julia Yoshino Benavente 2,3, Lauren A Opsasnick 2,3, Sophia Weiner-Light 1,2,3, Laura M Curtis 2,3, Michael S Wolf 2,3
PMCID: PMC10286119  NIHMSID: NIHMS1905766  PMID: 36987943

Abstract

Objectives:

To assess whether older adults with a cognitive impairment were more likely to report challenges interacting with medical providers, or to avoid needed medical care.

Methods:

Data for this exploratory, cross-sectional analysis were from older adults (N = 493) ages 60–82 participating in the “LitCog” cohort study. Multivariable generalized linear models compared cognitive impairment (none, mild, moderate, severe) with validated measures of healthcare engagement.

Results:

A moderate cognitive impairment was associated with delays in medical care due to embarrassment (RR 5.34.95% CI 1.30–22.0) and discomfort asking the doctor questions (RR 4.07, 95% CI 1.00–16.5).

Conclusions:

Intermediate cognitive deficits, such as with mild cognitive impairment (MCI) or mild dementias, may impact meaningful engagement with healthcare systems, potentially affecting timely detection and appropriate management of cognitive concerns and other chronic medical conditions. More research is needed to understand mechanisms underlying this relationship.

Keywords: cognitive impairment, Alzheimer’s disease and related dementias, healthcare engagement, healthcare avoidance

Introduction

Obtaining timely, appropriate medical care and engaging effectively with healthcare systems are essential components of maintaining health and independence among older adults, who are at increased risk of multimorbidity and disability (Salive, 2013). Yet, research has shown that as many as one-third of adults 65 and older report having delayed or avoided needed medical care (Federman et al., 2005; Kannan et al., 2014; Leyva et al., 2020). Consequences include missed or delayed diagnoses, increased risk and duration of hospitalization, worse physical and mental health, premature mortality, as well as higher healthcare costs (Byrne, 2008; Kraft et al., 2009; Ohl et al., 2010; Weissman et al., 1991). Various demographic, physical, psychological, and health-systems determinants have previously been linked to an increased risk of healthcare avoidance (Kannan et al., 2014; Leyva et al., 2020; Ng et al., 2020). However, few studies have examined how changes in cognitive function, which are increasingly prevalent in older age and can range from normal age-related decline to progressive syndromes such as Alzheimer’s Disease and Related Dementias (ADRD), may also be associated with barriers to recommended medical care.

A cognitive impairment, which includes deficits in one or more cognitive domains related to memory, executive function, language, attention, or processing speed, could undermine effective engagement with the healthcare system in several ways. For instance, many of the skills needed to effectively navigate healthcare systems and utilize healthcare services rely on cognitive abilities that may be compromised among those with a cognitive impairment (Lovett et al., 2020; Wolf et al., 2009). A cognitive deficit may therefore make it challenging for patients to understand or determine medical need, remember medical appointments and instructions, communicate effectively with clinicians and care teams, or problem-solve barriers to care. Furthermore, psychological aspects of disease, such as stigma or affective reactions to illness, including worry, fear, or shame related to diagnosis and/or losses in functional abilities or independence, are also common among individuals experiencing a cognitive impairment (Nguyen & Li, 2020), and are known to interfere with important health-seeking behaviors (Dolezal, 2022).

While barriers and facilitators surrounding healthcare engagement specific to ADRD-related concerns or a subjective cognitive complaint have been well-documented in the literature (Hill et al., 2021; Werner et al., 2014), the impact of a cognitive impairment on general, versus disease-specific, aspects of care has not yet been studied. Older adults with a cognitive impairment must often contend with other acute and chronic medical conditions in addition to their cognitive symptoms. Indeed, an estimated 86% of adults with a cognitive impairment report the presence of at least one other chronic medical condition (Centers for Disease Control and Prevention, 2020). Consequences of this comorbidity are bi-directional; the presence of multiple chronic conditions is linked with poor or worsening cognitive health, while the presence of a cognitive deficit may also contribute to the subsequent development of additional chronic conditions (Snowden et al., 2017). Increasing understanding of how a cognitive impairment may impact management of health more globally, including health-seeking behaviors and interactions with medical providers or systems, is therefore both an important individual and public health concern. This is especially relevant as rates of cognitive impairment, including ADRD, are expected to increase dramatically in the coming decades as the demographic composition of the United States continues to grow older and more medically complex (Alzheimer’s Association [Alz Assoc], 2022).

The objective of this exploratory investigation was therefore to examine whether older adults with a cognitive impairment, as measured via objective assessment, were more likely to report challenges interacting with medical providers, or to avoid needed medical care altogether. Furthermore, we also sought to determine if attitudes regarding patient-provider engagement or prevalence of reported reasons underlying healthcare avoidance might differ according to severity of cognitive impairment. We hypothesized older adults with an objective cognitive impairment would endorse greater barriers to healthcare engagement compared to cognitively normal older adults, and rates would be greater as severity of cognitive impairment increased. Findings may improve understanding of how the presence of a cognitive impairment may impact older adults’ perceptions of and participation in medical care more broadly, and can help to inform patient-centered interventions to optimize healthcare engagement within this growing population.

Methods

Data for this cross-sectional, secondary analysis came from the ongoing, prospective Health Literacy and Cognitive Function among Older Adults cohort study (Wolf et al., 2012). LitCog is an observational cognitive aging study originally designed to investigate the relationship between health literacy, chronic disease self-management, and general cognitive abilities over time. Three study timepoints (T1–T3) have been completed, with an additional two timepoints currently underway (T4 and T5). Nine-hundred cognitively-intact older adults were recruited for baseline assessment between August 2008 and August 2015 using convenience sampling methods. English-speaking adults aged 55–74 receiving care at either one academic general internal medicine practice or six federally qualified health centers in Chicago, IL between August 2008 and August 2015 were identified through electronic health records as initially eligible by age, notified of the study by mail, and contacted via phone. Interested participants were further screened for any severe visual, hearing, or cognitive impairments (defined as ≤2 errors on the 6-item screener) (Callahan et al., 2002), that would significantly limit study participation. Participants completed a baseline assessment consisting of two structured interviews 7–10 days apart, each lasting 2.5 hr. Day 1 assessments included basic demographic information, socioeconomic status, comorbidity, health literacy, and assessments of chronic disease self-management. Day 2 interviews consisted of a comprehensive cognitive battery, detailed further below. Participants completing baseline assessment were invited to follow-up interviews every 3 years. Participants provided written informed consent at each study timepoint, and subsequent interviews followed similar procedures to baseline interviews, however, with additional measures added to the study battery at each timepoint. For this analysis, data from the third timepoint (T3, collected between August 11, 2014 and February 20, 2020) was used. T3 was selected as measures related to reasons underlying healthcare avoidance and attitudes towards patient-provider engagement were introduced into the study battery at this time, and all data collection activities have been completed for this timepoint. Out of the original 900 study participants, a total of 501 participants completed interviews at T3, representing a 56% retention rate. Cognitive testing data was missing for eight participants; thus, 493 participants were used for this analysis. The study was approved by the Northwestern University Feinberg School of Medicine Institutional Review Board (Approval Number: STU00026255-MOD0014).

Measures

Cognitive Impairment.

Cognitive impairment was our primary independent variable of interest. Thirteen cognitive tests were used to measure performance in five cognitive domains: processing speed (Digit Comparison, Pattern Comparison, Symbol Digit Modalities) (Salthouse, 1991, 1992; Smith, 1991), working memory (CANTAB Spatial Span Length—Reverse, CANTAB Spatial Working Memory, Size Judgment Span) (Cherry & Park, 1993; Robbins et al., 1994), delayed memory (CANTAB Delayed Verbal Memory, New York Paragraph—Delayed) (Kluger et al., 1999; Robbins et al., 1994), executive function (ETS Letter Sets, CANTAB Stockings of Cambridge, Ravens Progressive Matrices) (Ekstrom & Harman, 1976; Raven, 1976; Robbins et al., 1994), and language (CANTAB Graded Naming Test, Shipley Institute of Living—Vocabulary) (Robbins et al., 1994; Zachary, 1986). Determination of cognitive impairment within this sample has previously been described in detail (Lovett et al., 2020), and based on methods proposed by Shirk et al. (2011) for research purposes in which cognitive assessments have been adapted for practical use and/or when reliable norms are not available for a given population of interest. Briefly, individual age- and education-adjusted z-scores were calculated for each individual for each cognitive test using a modified regression-based approach, and recognizing participants were classified as having either mild (−1 to −1.49), moderate (−1.5 to −1.99), or severe (<−2) impairment if their performance on two or more tests within at least one cognitive domain met these thresholds (with higher domain severities superseding lower domain severities). This classification is closely aligned to methods proposed by Jak/Bondi, in which both a cut-off score and number of impaired tests in a domain are considered to balance both sensitivity and specificity (Bondi et al., 2014).

For diagnostic comparison, mild to moderate severities loosely correspond to DSM-TR identified neuropsychological threshold of 1–2 standard deviations below appropriate norms for Minor Neurocognitive Disorder (NCD) (also known as Mild Cognitive Impairment (MCI)) (American Psychiatric Association, 2022). Our severe classification most closely aligns with DSM-5-TR neuropsychological criteria (2 or more standard deviation below appropriate norms) for Major Neurocognitive Disorder (also known as ADRD).

Barriers to Healthcare Engagement.

Reasons for avoiding or delaying medical care were assessed using four items from the Access to Care Module of the validated Awareness and Beliefs about Cancer (ABC) measure (Simon et al., 2012). The ABC measure has previously been used to assess health-seeking behaviors among non-cancer populations (Jarbøl et al., 2021). All participants were provided the following prompt: “Sometimes people put off going to the doctor even when they have a symptom they think might be serious- …could you say if any of these might put you off going to the doctor?” Participants were then provided with four individual statements: “I would be too embarrassed,” “I would be worried about wasting the doctor’s time,” “I would be worried about what the doctor might find,” and “I am too busy to make time to go to the doctor.” Response options were the same for each item (Yes, often; Yes, sometimes; No), which were then dichotomized for the purposes of this analysis (Yes, often/Yes, sometimes vs. No).

Two individual items were also used to examine attitudes towards patient-provider interactions. Participants were first asked: “In general, how comfortable do you feel asking questions of doctors?” Original response options for this item (Not comfortable at all; A little comfortable; Somewhat comfortable; Very comfortable) were then dichotomized (Not comfortable at all/A little comfortable vs. Somewhat comfortable/Very comfortable). Participants were also asked: “In general, how nervous are you when you go to see the doctor?” Responses (Not nervous at all; A little nervous; Somewhat nervous; Very nervous) were also dichotomized (Very nervous/somewhat nervous vs. A little nervous/Not nervous at all).

Covariates.

As this was a secondary data analysis, covariate selection was based on a combination of the available data in addition to the prior literature. Age, sex, race/ethnicity, education, and number of chronic medical conditions were collected via self-report. Health literacy was measured using the Newest Vital Signs (NVS) (Weiss et al., 2005). Scores range from 0–6, and performance was categorized as limited (<2), marginal (<4), or adequate (≥4) health literacy. Mental health symptoms were measured using Patient-Reported Outcomes Measurement Information System (PROMIS) depression and anxiety short-form subscales (Cella et al., 2010). All PROMIS tests have a mean of 50 (±10) by definition and are normed against the U.S. general population. Higher scores indicate more symptoms.

Analysis Plan

Differences between participant characteristics, cognitive impairment (none, mild, moderate, severe), and our outcomes of interest were first determined using a combination of chi-square tests and analysis of variance (ANOVA). Chi-square tests were then used to examine bivariate associations between cognitive impairment and barriers to healthcare engagement (reasons underlying healthcare avoidance, attitudes towards patient-provider interactions). Reasons underlying healthcare avoidance was assessed at both the item-level and overall (if any item was endorsed). Attitudes towards patient-provider interactions were examined only for each individual question item. Multivariable models using a Poisson distribution to estimate relative risk with 95% confidence intervals (CI) were then conducted for any bivariate associations found to be significant at the p < 0.05 level (Zou, 2004), controlling for race/ethnicity, income, number of chronic medical conditions, health literacy, and mental health symptoms (depression, anxiety). Covariates were included in multivariable analyses if they were significant with either cognition or our outcomes of interest. Education was not included in models due to concern for over-adjustment; previous studies have recommended inclusion of either education or health literacy, but not both (Wolf et al., 2005). All statistical analyses were conducted using Stata 15.1 (College Station, TX).

Results

Participant characteristics, overall and by cognitive impairment, are listed in Table 1. Participants were on average 68.6 (SD 5.2) years old. Over one-half (51.4%) of the sample were White. Approximately one-third (35.2%) had a graduate degree, and nearly one-half (47.7%) reported an income greater than $50,000. Participants had an average of 3.2 (SD 1.9) chronic conditions. Over one-half (51.3%) had adequate health literacy. Average PROMIS t-scores were 46.1 (SD 8.2) and 51.1 (SD) for depression and anxiety, respectively.

Table 1.

Participant Characteristics at LitCog T3, Overall and By Cognitive Impairment Severity.

Cognitive Impairment
Variables Overall (N = 493) None (n = 349) Mild (n = 81) Moderate (n = 34) Severe (n = 29) p value
Age, mean (SD) 68.6 (5.2) 68.4 (5.2) 69.0 (5.2) 68.9 (5.4) 69.0 (6.1) 0.79
Male, n (%) 138 (28.0) 95 (27.2) 27 (33.3) 10 (29.4) 6 (20.7) 0.56
Race, n (%) <0.001
 White 253 (51.4) 226 (64.9) 18 (22.2) 4 (11.8) 5 (17.2)
 Black 193 (39.2) 102 (29.3) 48 (59.3) 23 (67.7) 20 (69.0)
 Other 46 (9.4) 20 (5.8) 15 (18.5) 7 (20.6) 4 (13.8)
Educational attainment, n (%) <0.001
 High school or less 111 (22.6) 50 (14.4) 31 (38.3) 20 (58.8) 10 (34.5)
 1–3 years college/Technical degree 109 (22.2) 68 (19.6) 24 (29.6) 10 (29.4) 7 (24.1)
 College graduate 98 (20.0) 76 (21.9) 15 (18.5) 2 (5.9) 5 (17.2)
 Graduate degree 173 (35.2) 153 (44.1) 11 (13.6) 2 (5.9) 7 (24.1)
Income level, n (%) <0.001
 <$10,000 44 (9.3) 16 (4.7) 16 (20.8) 7 (23.3) 5 (18.5)
 $10,000–$24,999 109 (23.1) 52 (15.4) 28 (36.4) 17 (56.7) 12 (44.4)
 $25,000–$49,999 94 (19.9) 73 (21.6) 12 (15.6) 5 (16.7) 4 (14.8)
 ≥$50,000 225 (47.7) 197 (58.3) 21 (27.3) 1 (3.3) 6 (22.2)
Health literacy (NVS), n (%) <0.001
 Limited 129 (26.3) 39 (11.2) 44 (54.3) 28 (82.4) 18 (62.1)
 Marginal 110 (22.4) 82 (23.6) 16 (19.8) 3 (8.8) 9 (31.0)
 Adequate 252 (51.3) 226 (65.1) 21 (25.9) 3 (8.8) 2 (6.9)
# Of chronic conditions, mean (SD) 3.2 (1.9) 2.9 (1.8) 3.7 (1.8) 4.2 (2.1) 3.7 (2.1) <0.001
Anxiety (PROMIS T-score), mean (SD) 51.1 (8.9) 50.7 (8.5) 50.6 (9.8) 56.4 (7.5) 51.9 (9.6) 0.004
Depression (PROMIS T-score), mean (SD) 46.1 (8.2) 45.4 (7.9) 46.3 (8.2) 50.3 (9.0) 49.6 (8.0) <0.001
Open in a new tab

Note. NVS = Newest Vital Sign; PROMIS = Patient-reported Outcomes Measurement Information System. Bolded items indicate statistical significance.

Approximately one-third of our sample demonstrated an objective cognitive impairment (29.2%) (Table 1). Of these, the majority were mild (16.4%), followed by moderate (6.9%) and then severe (5.9%). Participants with a cognitive impairment were more likely to be of a racial/ethnic minority, report lower educational attainment and income levels, and have a greater number of chronic medical conditions (all p’s < 0.001). Those with a cognitive impairment were also more likely to have lower health literacy (p < 0.001), as well as more depressive (p < 0.001) and anxiety symptoms (p = 0.004).

In total, nearly half (41.0%) of older adults reported having delayed important medical care (Table 2). Of these, the majority reported delays due to worry about what the doctor might find (24.0%), followed by being too busy (16.5%), worry about wasting the doctor’s time (7.5%), and embarrassment (4.7%). Approximately 15% of our overall sample reported feeling nervous to go to the doctor; only a small proportion of adults overall endorsed discomfort asking the doctor questions (4.3%).

Table 2.

Bivariate Associations between Barriers to Healthcare Engagement and Cognitive Impairment Severity.

Cognitive Impairment
Variable Overall
(N = 493)		 None
(n = 349		 Mild
(n = 81)		 Moderate
(n = 34)		 Severe
(n = 29)		 p Value
Patient-Provider Interactions
 Uncomfortable asking doctor questions, n (%) 21 (4.3) 8 (2.3) 6 (7.4) 5 (14.7) 2 (5.9) 0.002
 Nervous to go to the doctor, n (%) 72 (14.7) 46 (13.3) 17 (21.0) 7 (20.6) 2 (6.9) 0.14
Healthcare avoidance
 Any delay in care, n (%) 201 (41.0) 146 (42.2) 26 (32.1) 19 (55.9) 10 (34.5) 0.09
  Delay care due to embarrassment, n (%) 23 (4.7) 11 (3.2) 3 (3.7) 8 (23.5) 1 (3.5) <0.001
  Delay care due to worry wasting doctor’s time, n (%) 37 (7.5) 32 (9.2) 3 (3.7) 2 (5.9) 0 (0.0) 0.13
  Delay care due to worry what doctor might find, n (%) 118 (24.0) 78 (22.5) 19 (23.5) 13 (38.2) 8 (27.6) 0.22
  Delay care due to being too busy, n (%) 81 (16.5) 62 (17.9) 9 (11.1) 8 (23.5) 2 (6.9) 0.15
Open in a new tab

Note. Bolded items indicate statistical significance.

Bivariate associations between cognitive impairment and the outcomes of interest are also listed in Table 2. Associations between cognitive impairment and the combined avoidance score were non-significant. However, on item-level analysis differences by cognitive impairment severity were observed (Mild: 3.7% vs. Moderate: 23.5% vs. Severe: 3.5% vs. None: 3.2%, p < 0.001). A cognitive impairment was also associated with higher rates of discomfort asking doctors questions (Mild: 7.4% vs. Moderate: 14.7% vs. Severe: 5.9% vs. None: 2.3%, p = 0.002). Rates of reported worry about what the doctor might find were in general higher for those with a cognitive impairment (Mild: 23.5% vs. Moderate: 38.2% vs. Severe: 27.6% vs. None: 22.5%), however, this association was not statistically significant. All other bivariate associations were non-significant.

In multivariable analysis (Table 3), compared to the group with no impairment and controlling for relevant covariates, only those with a moderate cognitive impairment remained more likely to endorse delaying medical care due to feelings of embarrassment (RR 5.34.95% CI 1.30–22.0). Independent associations between moderate cognitive impairment and discomfort asking the doctor questions were also observed (RR 4.07, 95% CI 1.00–16.5), and independent of observed associations with race and ethnicity (Black RR 5.30, 95% CI 1.25–22.5; Other RR 5.95, 95% CI 1.22–29.1). No significant associations were apparent among those with either mild or severe impairments in multivariable analysis.

Table 3.

Multivariable Models of Barriers to Healthcare Engagement.

Uncomfortable asking Doctor Questions Avoid Doctor due to Embarrassment
Variable RR (95% CI) p value RR (95% CI) p value
Cognitive impairment
 None REF
 Mild 2.29 (0.70, 7.45) 0.17 1.01 (0.24, 4.36) 0.98
 Moderate 4.07 (1.00, 16.5) 0.05 5.34 (1.30, 22.0) 0.02
 Severe 1.85 (0.34, 10.1) 0.48 0.83 (0.09, 7.99) 0.87
Race
 White REF
 Black 5.30 (1.25, 22.5) 0.02 0.69 (0.20, 2.38) 0.55
 Other 5.95 (1.22, 29.1) 0.03 1.30 (0.32, 5.37) 0.72
Income
 <$10,000 REF
 $10,000–$24,999 0.44 (0.13, 1.50) 0.18 0.61 (0.13, 2.81) 0.54
 $25,000–$49,999 0.21 (0.04, 1.17) 0.08 1.99 (0.41, 9.67) 0.39
 ≥$50,000 0.84 (0.19, 3.64) 0.81 3.14 (0.50, 19.6) 0.22
# Of chronic conditions 0.82 (0.62, 1.08) 0.15 1.12 (0.89, 1.42) 0.34
Health literacy (NVS)
 Adequate REF –
 Marginal 1.24 (0.34, 4.47) 0.74 0.64 (0.13, 3.12) 0.58
 Limited 0.91 (0.22, 3.70) 0.89 2.83 (0.67, 11.9) 0.16
Anxiety symptoms (PROMIS) 1.03 (0.97, 1.10) 0.34 0.97 (0.90, 1.03) 0.34
Depressive symptoms (PROMIS) 1.07 (0.99, 1.15) 0.07 1.09 (1.02, 1.17) 0.02
Open in a new tab

Note. NVS = Newest Vital Sign; PROMIS = Patient-reported Outcomes Measurement Information System. Bolded items indicate statistical significance.

Discussion

In this exploratory analysis among a sample of older, community-based adults, approximately four in 10 older adults reporting delaying needed medical care for reasons such as increased worry or embarrassment and/or due to time-constraints, regardless of cognitive status. Nearly a fifth of our sample reported feeling nervous or uncomfortable interacting with medical providers. Our findings also suggest that the presence of an objective cognitive impairment is independently associated with certain barriers (i.e., discomfort asking questions, feeling of embarrassment) to meaningful engagement in one’s medical care, above and beyond the influence of other known demographic and psychosocial factors. Furthermore, we also found that the degree of cognitive impairment may also be an important factor influencing such barriers. Specifically, those with a moderate cognitive impairment, likely corresponding to MCI or mild ADRD, were more likely to endorse these concerns when compared to adults without a cognitive impairment. No such relationship was observed for individuals with deficits categorized as either mild or severe.

The prevalence of older adults reporting delays in medical care in this study is somewhat higher than prior estimates in the literature, which have found rates range from 22–37% (Federman et al., 2005; Kannan et al., 2014; Leyva et al., 2020). This variability may be due to measurement differences; the items assessing healthcare avoidance from the ABC measure used in this study tended to be general in nature (e.g., “I would be worried about what the doctor might find.”) (Simon et al., 2012). Other studies assessing healthcare avoidance have often used Health Information Trends Survey Items, which has more specific terminology (e.g., “It makes me think about dying.”) (Kannan et al., 2014; Leyva et al., 2020). It should be noted that existing measurement tools for assessing avoidance of medical care, and barriers surrounding healthcare engagement more broadly, are often limited to a restricted number of items that may not comprehensively address factors associated with these concepts. To optimize understandings surrounding this important healthcare behavior within this population, future studies should be conducted to better operationalize this construct and develop more extensive, patient-centered, validated questionnaires to further improve this field of research.

Findings from this study also extend understanding surrounding determinants of certain barriers to healthcare engagement amongst older adult populations. Specifically, we found that individuals with a moderate, as opposed to mild or severe, cognitive impairment were more likely to report discomfort interacting with doctors or delay care due to embarrassment. This supports previous research conducted among the general population that psychological or emotional factors, both general and disease-specific, are often most predictive of poorer healthcare engagement (Kannan & Veazie, 2014; Taber et al., 2015). When considered in the context of a cognitive impairment specifically, an ADRD diagnosis can be devastating for patients and their families. While symptoms can be managed or even delayed with early identification and supportive care, no cure currently exists, and affected adults often live through many years of disability and dependence as the disease progresses. As a result of this high disease burden, internalized stigma, shame, and illness fears have been found to be common among those with ADRD or other cognitive concerns (Fowler et al., 2012; Nguyen & Li, 2020). Research in this area more broadly has shown stigmatized populations are more likely to report medical encounters as sources of shame and humiliation interfering with healthcare access (Dolezal, 2022), with commonly reported concerns including fears of judgment, disappointment, and blame (Northrop, 2017). Future research is needed to understand specific mechanisms underlying relationships between cognitive impairment and reduced healthcare engagement. While not a focus of this present study, it should also be noted other stigmatized populations endorsed discomfort interacting with medical professionals, specifically individuals who identified as a racial and ethnic minority. Racial disparities in patient-provider communication are well-established in the literature (Shen et al., 2018). Future studies should investigate how race and ethnicity moderates or confers an additive effect on the relationships between cognitive impairment and barriers to healthcare engagement.

The finding that moderate, versus mild or severe, cognitive impairments in our study, likely corresponding to intermediate stages of disease such as MCI or mild ADRD, were most associated poorer healthcare engagement, is also of particular interest. We expected that as severity of impairment increased so too would reported barriers to healthcare engagement, due to greater impact or consequences related to their symptoms. The severity effect observed here could be, in part, attributed to either presence or absence of insight or awareness related to their deficits. Theoretical models of healthcare delay posit that symptom appraisal is a necessary, but not sufficient, factor influencing action; that is, one must first perceive a symptom as being present prior to deciding to obtain treatment (Walter et al., 2012). At its most extreme end, lack of insight into one’s condition or its functional impact (i.e., “anosognosia”), is a common symptom of more severe impairments characteristic of ADRD (de Ruijter et al., 2020). Thus, as opposed to those classified as severe who may have lost awareness into their symptoms because of the disease process, participants with moderate impairments may still have been experiencing deficits that were impactful enough to be noticeable, but not yet significant enough to impair insight into these changes. The lack of association among those with milder impairments may also be due to the more subtle nature of their deficits, which may not have yet been significant enough to be perceived by the individual or detected by the healthcare system.

The possible consequences of underutilization of important healthcare services among individuals with a cognitive impairment are significant. Early detection and close management of chronic disease among the older adult population, in particular for progressive syndromes such as ADRD, are increasingly considered key to preventing or delaying further health declines, optimizing quality of life and preserving independence (Livingston et al., 2020). In time- and resource-constrained primary care settings where most older adults receive their medical care, identification and treatment typically relies on patient report. Yet, our findings suggest that many adults with a cognitive impairment may fail to report concerning signs or symptoms to medical providers or delay needed medical care, often for affective reasons such as discomfort or embarrassment, which may hinder formal detection or appropriate management of their cognitive conditions or other medical concerns. This may be particularly salient for older adults with more noticeable deficits that may have a larger impact on health self-management abilities, though may not yet be severe enough to have resulted in implementation of compensatory strategies or for caregivers to have assumed responsibility for these roles, which could further hasten health declines.

Previous research has demonstrated that increasing public health messaging and education surrounding cognitive impairment more broadly, and ADRD more specifically, may be one strategy to increase awareness and reduce both public, medical provider, and self/internalized stigma, which may have downstream effects on healthcare-seeking behavior (Herrmann et al., 2018). Medical providers should also be mindful that patients with a cognitive impairment may present with affective barriers impacting medical care engagement. Use of well-established patient-centered communication approaches, such rapport building, actively broaching sensitive topics, empathetic listening, and use of anecdotes, among other techniques, may help to further reduce and address these factors, when present (Storlie, 2015; National Institute on Aging, 2017). Identification and inclusion of care partners in the medical care and decision-making process for those with cognitive impairment is also known to promote greater healthcare engagement (Hill et al., 2021). Increasing uptake and use of reimbursable, person-centered assessment and care planning services, such as Medicare’s Annual Wellness Visit or the recently created CPT® coding for Cognitive Care Planning, may also help to facilitate symptom reporting among this at-risk population (Alz Assoc, 2016; Ganguli et al., 2017). Finally, emerging models of care based in primary care, including integrated, collaborative services that provide comprehensive, and ongoing education and counseling, care management and coordination, and personalized care plans have also shown promise in improving healthcare engagement and associated health outcomes (Boustani et al., 2019). Multi-faceted approaches combining individual, health system, and policy-level factors to optimize healthcare engagement within this population are likely needed.

Our study has several limitations which should be recognized. Foremost, our findings are limited to English-speaking, predominantly female, older adults who are connected to primary care practices in one urban city. However, the diversity of our sample regarding race/ethnicity should also be noted. Furthermore, this investigation was cross-sectional in nature, limiting interpretations regarding directionality and causality. Specifically, we are unable to determine if the presence of a cognitive impairment bestows an increased risk for avoiding needed medical care, or if increased barriers to healthcare engagement instead increases likelihood of developing a cognitive impairment. Our outcome measures addressing barriers to healthcare engagement were also reliant on self-report, which are frequently subject to recall or desirability biases, and limited to a restricted number of items that do not comprehensively address factors associated with this construct; as such, these measurement considerations likely limit a full understanding surrounding the prevalence of poor healthcare engagement and associated reasons within this population. Additionally, given the duration of data collection, changes in the healthcare environment or care practice models could have occurred over the course of the study period, potentially affecting access to care and interaction with healthcare providers. We also did not have proxy reports to compare with patient-endorsed outcomes, which limits our ability to confirm information accuracy, in particular for individuals with more severe impairments. Finally, as a secondary data analysis, this investigation was also limited by practical considerations including available measures and sample size. Specifically, the relatively small number of participants with a cognitive impairment in our sample, particularly those with moderate or severe deficits, likely affected our statistical power as well as the precision of any observed associations. A priori, longitudinal research studies utilizing larger samples, informant reporting, and comprehensive measurement of this construct are needed to help address these constraints.

In conclusion, the presence of a cognitive impairment negatively impacts adequate engagement in medical care through increased discomfort interacting with medical providers and feelings of embarrassment. Degree of cognitive impairment severity may also be an important consideration, with individuals demonstrating moderate cognitive deficits, likely characteristic of intermediate stages of disease such as MCI or mild ADRD, being more likely to report such barriers. Our findings underscore the need for multi-faceted, patient-centered solutions at the individual, health system, and policy-levels to help address affective factors such as stigma that may limit engagement with care. This exploratory analysis also highlights several directions for future research addressing this issue, in particular the need for further construct operationalization and measure development, investigations into the moderating or additive effect of other stigmatized identities, as well as longitudinal research to increase understanding of causality and mechanistic factors underlying the associations observed in this present study.

What this study adds

  • Cognitively-impaired older adults were more uncomfortable asking doctors questions and delayed medical care due to embarrassment.

  • Older adults with impairments of moderate severity, likely corresponding to MCI or mild dementia, were more likely to report these challenges.

Applications of study findings

  • Multi-faceted approaches are needed to enhance healthcare engagement in older adults with a cognitive impairment.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research reported in this publication was supported by the National Institute of Aging of the National Institutes of Health under award numbers R01AG030611 and P30AG059988. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr Lovett reports grants from the NIH (NIA, NINDS), and awards from the Health Assessment Lab. Ms Yoshino Benavente reports grants from the NIH and Gordon and Betty Moore Foundation. Ms Curtis reports grants from the NIH. Dr Wolf reports grants from the NIH (NIA, NIDDK, NINR, NHLBI, NINDS), Gordon and Betty Moore Foundation, and Eli Lilly, and personal fees from Pfizer, Sanofi, Luto UK, GlaxoSmithKline, University of Westminster, and Lundbeck. Ms Opsasnick and Ms Weiner-Light have no conflicts of interest to report.

Ethical Approval

This study was approved by the Northwestern University Feinberg School of Medicine Institutional Review Board (STU00026255).

References

  1. Alzheimer’s Association (2016). Cognitive assessment and care planning services: Alzheimer’s association expert task force recommendations and tools for implementation Alzheimer’s Association. https://www.alz.org/careplanning/downloads/cms-consensus.pdf [Google Scholar]
  2. Alzheimer’s Association. (2022). Alzheimer’s disease facts and figures. Alzheimers Dementia, 18(4), 700–789. 10.1002/alz.12638 [DOI] [PubMed] [Google Scholar]
  3. American Psychiatric Association. (2022). Neurocognitive disorders. In Diagnostic and statistical manual of mental disorders (5th ed., text rev). American Psychiatric Publishing, Inc. 10.1176/appi.books.9780890425787 [DOI] [Google Scholar]
  4. Bondi MW, Edmonds EC, Jak AJ, Clark LR, Delano-Wood L, McDonald CR, Libon DJ, Au R, Galasko D, Salmon DP, & Nation DA (2014). Neuropsychological criteria for mild cognitive impairment improves diagnostic precision, biomarker associations, and progression rates. Journal of Alzheimer’s Disease, 42(1), 275–289. 10.3233/JAD-140276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boustani M, Alder CA, Solid CA, & Reuben D (2019). An alternative payment model to support widespread use of collaborative dementia care models. Health Affairs (Project Hope), 38(1), 54–59. 10.1377/hlthaff.2018.05154 [DOI] [PubMed] [Google Scholar]
  6. Byrne SK (2008). Healthcare avoidance: A critical review. Holistic Nursing Practice, 22(5), 280–292. 10.1097/01.HNP.0000334921.31433.c6 [DOI] [PubMed] [Google Scholar]
  7. Callahan CM, Unverzagt FW, Hui SL, Perkins AJ, & Hendrie HC (2002). Six-item screener to identify cognitive impairment among potential subjects for clinical research. Medical Care, 40(9), 771–781. 10.1097/00005650-200209000-00007 [DOI] [PubMed] [Google Scholar]
  8. Cella D, Riley W, Stone A, Rothrock N, Reeve B, Yount S, Amtmann D, Bode R, Buysse D, Choi S, Cook K, Devellis R, DeWalt D, Fries JF, Gershon R, Hahn EA, Lai JS, Pilkonis P, Revicki D, & Hays R (2010). The patient-reported outcomes measurement information system (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. Journal of Clinical Epidemiology, 63(11), 1179–1194. 10.1016/j.jclinepi.2010.04.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Centers for Disease Control and Prevention. (2020). Chronic diseases and cognitive decline-a public health issue [CDC Data Brief] U.S. Department of Health and Human Services. https://www.cdc.gov/aging/publications/chronic-diseases-brief.html [Google Scholar]
  10. Cherry KE, & Park DC (1993). Individual difference and contextual variables influence spatial memory in younger and older adults. Psychology and Aging, 8(4), 517–526. 10.1037//0882-7974.8.4.517 [DOI] [PubMed] [Google Scholar]
  11. de Ruijter NS, Schoonbrood AMG, van Twillert B, & Hoff EI (2020). Anosognosia in dementia: A review of current assessment instruments. Alzheimer’s and Dementia (Amsterdam, Netherlands), 12(1), e12079–e12079. 10.1002/dad2.12079 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dolezal L (2022). Shame anxiety, stigma and clinical encounters. Journal of Evaluation in Clinical Practice, 28(5), 854–860. 10.1111/jep.13744 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ekstrom RB, & Harman HH (1976). ETS kit of factor-referenced cognitive tests Educational Testing Service. [Google Scholar]
  14. Federman AD, Vladeck BC, & Siu AL (2005). Avoidance of health care services because of cost: Impact of the medicare savings program. Health Affairs (Project Hope), 24(1), 263–270. 10.1377/hlthaff.24.1.263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fowler NR, Boustani MA, Frame A, Perkins AJ, Monahan P, Gao S, Sachs GA, Hendrie HC, & Hendrie HC (2012). Effect of patient perceptions on dementia screening in primary care. Journal of the American Geriatrics Society, 60(6), 1037–1043. 10.1111/j.1532-5415.2012.03991.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ganguli I, Souza J, McWilliams JM, & Mehrotra A (2017). Trends in use of the US medicare annual wellness Visit, 2011–2014. The Journal of the American Medical Association, 317(21), 2233–2235. 10.1001/jama.2017.4342 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Herrmann LK, Welter E, Leverenz J, Lerner AJ, Udelson N, Kanetsky C, & Sajatovic M (2018). A systematic review of dementia-related stigma research: Can we move the stigma dial? The American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 26(3), 316–331. 10.1016/j.jagp.2017.09.006 [DOI] [PubMed] [Google Scholar]
  18. Hill NL, Bratlee-Whitaker E, Sillner A, Brautigam L, & Mogle J (2021). Help-seeking for cognitive problems in older adults without dementia: A systematic review. International Journal of Nursing Studies Advances, 3, 100050. 10.1016/j.ijnsa.2021.100050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jarbøl DE, Haastrup PF, Rasmussen S, Søndergaaard J,& Balasubramaniam K (2021). Women’s barriers for contacting their general practitioner when bothered by urinary incontinence: A population-based cross-sectional study. BMC Urology, 21(1), 99. 10.1186/s12894-021-00864-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kannan VD, & Veazie PJ (2014). Predictors of avoiding medical care and reasons for avoidance behavior. Medical Care, 52(4), 336–345. 10.1097/mlr.0000000000000100 [DOI] [PubMed] [Google Scholar]
  21. Kluger A, Ferris SH, Golomb J, Mittelman MS, & Reisberg B (1999). Neuropsychological prediction of decline to dementia in nondemented elderly. Journal of Geriatric Psychiatry and Neurology, 12(4), 168–179. 10.1177/089198879901200402 [DOI] [PubMed] [Google Scholar]
  22. Kraft AD, Quimbo SA, Solon O, Shimkhada R, Florentino J, & Peabody JW (2009). The health and cost impact of care delay and the experimental impact of insurance on reducing delays. The Journal of Pediatrics, 155(2), 281–285.e1. 10.1016/j.jpeds.2009.02.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Leyva B, Taber JM, & Trivedi AN (2020). Medical care avoidance among older adults. Journal of Applied Gerontology: The Official Journal of the Southern Gerontological Society, 39(1), 74–85. 10.1177/0733464817747415 [DOI] [PubMed] [Google Scholar]
  24. Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, Brayne C, Burns A, Cohen-Mansfield J, Cooper C, Costafreda SG, Dias A, Fox N, Gitlin LN, Howard R, Kales HC, Kivimaki M, Larson EB, Ogunniyi A, & Mukadam N (2020). Dementia prevention, intervention, and care: 2020 report of the lancet commission. Lancet, 396(10248), 413–446. 10.1016/s0140-6736(20)30367-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lovett RM, Curtis LM, Persell SD, Griffith JW, Cobia D, Federman A, & Wolf MS (2020). Cognitive impairment no dementia and associations with health literacy, self-management skills, and functional health status. Patient Education and Counseling, 103(9), 1805–1811. 10.1016/j.pec.2020.03.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. National Institute on Aging. (2017). Talking with older patients about sensitive topics National Institutes of Health. https://www.nia.nih.gov/health/talking-older-patients-about-sensitive-topics [Google Scholar]
  27. Ng BP, LaManna JB, Towne SD Jr., Peach BC, He Q, & Park C (2020). Factors associated with avoiding health care among community-dwelling medicare beneficiaries with type 2 diabetes. Preventing Chronic Disease, 17, E128. 10.5888/pcd17.200148 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nguyen T, & Li X (2020). Understanding public-stigma and self-stigma in the context of dementia: A systematic review of the global literature. Dementia (London), 19(2), 148–181. 10.1177/1471301218800122 [DOI] [PubMed] [Google Scholar]
  29. Northrop JM (2017). A dirty little secret: Stigma, shame and hepatitis C in the health setting. Medical Humanities, 43(4), 218–224. 10.1136/medhum-2016-011099 [DOI] [PubMed] [Google Scholar]
  30. Ohl M, Tate J, Duggal M, Skanderson M, Scotch M, Kaboli P, Vaughan-Sarrazin M, & Justice A (2010). Rural residence is associated with delayed care entry and increased mortality among veterans with human immunodeficiency virus infection. Medical Care, 48(12), 1064–1070. 10.1097/MLR.0b013e3181ef60c2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Raven JC (1976). Standard progressive Matrices: Sets A, B, C, D and E. Harcourt assessment Oxford psychologists Press [Google Scholar]
  32. Robbins TW, James M, Owen AM, Sahakian BJ, McInnes L, & Rabbitt P (1994). Cambridge neuropsychological test automated battery (CANTAB): A factor analytic study of a large sample of normal elderly volunteers. Dementia, 5(5), 266–281. 10.1159/000106735 [DOI] [PubMed] [Google Scholar]
  33. Salive ME (2013). Multimorbidity in older adults. Epidemiologic Reviews, 35(1), 75–83. 10.1093/epirev/mxs009 [DOI] [PubMed] [Google Scholar]
  34. Salthouse TA (1992). What do adult age differences in the Digit Symbol Substitution Test reflect? Journal of Gerontology, 47(3), P121–P128. 10.1093/geronj/47.3.p121 [DOI] [PubMed] [Google Scholar]
  35. Salthouse TA, & Babcock RL (1991). Decomposing adult age differences in working memory. Developmental Psychology, 27(5), 763–776. 10.1037/0012-1649.27.5.763 [DOI] [Google Scholar]
  36. Shen MJ, Peterson EB, Costas-Muñiz R, Hernandez MH, Jewell ST, Matsoukas K, & Bylund CL (2018). The effects of race and racial concordance on patient-physician communication: A systematic review of the literature. Journal of Racial and Ethnic Health Disparities, 5(1), 117–140. 10.1007/s40615-017-0350-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Shirk SD, Mitchell MB, Shaughnessy LW, Sherman JC, Locascio JJ, Weintraub S, & Atri A (2011). A web-based normative calculator for the uniform data set (UDS) neuropsychological test battery. Alzheimer’s Research and Therapy, 3(6), 32. 10.1186/alzrt94 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Simon AE, Forbes LJL, Boniface D, Warburton F, Brain KE, Dessaix A, Donnelly M, Haynes K, Hvidberg L, Lagerlund M, Petermann L, Tishelman C, Vedsted P, Vigmostad MN, Wardle J, & Ramirez AJ, ICBP Module 2 Working Group, ICBP Programme Board and Academic Reference Group. (2012). An international measure of awareness and beliefs about cancer: Development and testing of the ABC. BMJ Open, 2(6), Article e001758. 10.1136/bmjopen-2012-001758 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Smith A (1991). Symbol digit modalities test Western Psychological Services. [Google Scholar]
  40. Snowden MB, Steinman LE, Bryant LL, Cherrier MM, Greenlund KJ, Leith KH, Levy C, Logsdon RG, Copeland C, Vogel M, Anderson LA, Atkins DC, Bell JF, Fitzpatrick AL, & Fitzpatrick AL (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. 10.1002/gps.4652 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Storlie TA (2015). Patient-centered communication with older adults: The professional provider’s guide Academic Press. [Google Scholar]
  42. Taber JM, Leyva B, & Persoskie A (2015). Why do people avoid medical care? A qualitative study using national data. Journal of General Internal Medicine, 30(3), 290–297. 10.1007/s11606-014-3089-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Walter F, Webster A, Scott S, & Emery J (2012). The andersen model of total patient delay: A systematic review of its application in cancer diagnosis. Journal of Health Services Research and Policy, 17(2), 110–118. 10.1258/jhsrp.2011.010113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Weiss BD, Mays MZ, Martz W, Castro KM, DeWalt DA, Pignone MP, Mockbee J, Hale FA, & Hale FA (2005). Quick assessment of literacy in primary care: The newest vital sign. Annals of Family Medicine, 3(6), 514–522. 10.1370/afm.405 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Weissman JS, Stern R, Fielding SL, & Epstein AM (1991). Delayed access to health care: Risk factors, reasons, and consequences. Annals of Internal Medicine, 114(4), 325–331. 10.7326/0003-4819-114-4-325 [DOI] [PubMed] [Google Scholar]
  46. Werner P, Goldstein D, Karpas DS, Chan L, & Lai C (2014). Help-seeking for dementia: A systematic review of the literature. Alzheimer Disease and Associated Disorders, 28(4), 299–310. 10.1097/wad.0000000000000065 [DOI] [PubMed] [Google Scholar]
  47. Wolf MS, Curtis LM, Wilson EAH, Revelle W, Waite KR, Smith SG, Weintraub S, Borosh B, Rapp DN, Park DC, Deary IC, Baker DW, & Baker DW (2012). Literacy, cognitive function, and health: Results of the LitCog study. Journal of General Internal Medicine, 27(10), 1300–1307. 10.1007/s11606-012-2079-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wolf MS, Gazmararian JA, & Baker DW (2005). Health literacy and functional health status among older adults. Archives of Internal Medicine, 165(17), 1946–1952. 10.1001/archinte.165.17.1946 [DOI] [PubMed] [Google Scholar]
  49. Wolf MS, Wilson EAH, Rapp DN, Waite KR, Bocchini MV, Davis TC, & Rudd RE (2009). Literacy and learning in health care. Pediatrics, 124(0 3), S275–S281. 10.1542/peds.2009-1162C [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Zachary RA (1986). Shipley Institute of living scale: Revised manual Western Psychological Services. [Google Scholar]
  51. Zou G (2004). A modified Poisson regression approach to prospective studies with binary data. American Journal of Epidemiology, 159(7), 702–706. 10.1093/aje/kwh090 [DOI] [PubMed] [Google Scholar]

RESOURCES