Cognitive impairment is a significant public health concern as the United States (US) population continues to age at unprecedented rates. It is well established that older adults are more likely to have higher rates of morbidity and mortality, in addition to greater healthcare utilization and costs. A parallel concern is the greater risk of cognitive deficits among this population. Impairments in cognitive function are known to exacerbate poorer health outcomes among older adults. Within the field of cognitive aging, there has also been increasing focus on impairments that are greater than would be expected for normal aging, yet not reaching thresholds for cognitive disorders such as dementia. Known broadly as cognitive impairment (CI) not meeting criteria for dementia, CI has been shown to be an ‘at-risk’ state for worsening cognition or dementia. However, gaps in knowledge about CI remain, especially regarding its potentially subtle yet clinically significant impact on chronic disease self-management. Furthermore, relatively few research studies have examined the extent and associations between CI and health outcomes, or subsequently explored causal mechanisms that could inform the development of health system responses to improve the identification and management of CI. These studies are especially needed among older adults diverse by socioeconomic status and race/ethnicity. The objectives of this secondary investigation were threefold. Using data from the National Institute of Aging (NIA) Literacy and Cognitive Function among Older Adults (i.e. ‘LitCog’) cohort study (R01AG030611; PI: Wolf), we investigated associations between varying severities of CI and multiple health outcomes, including health literacy and chronic disease self-management and measures of functional health status (i.e. physical function, depressive symptoms, and anxiety symptoms,) among a community-based cohort of older adults. We first sought to examine associations between CI and these factors cross-sectionally, and then longitudinally after 2-5 years of follow-up. Exploratory comparisons between identification of CI from objective, neuropsychological data obtained in this study and the presence of any cognitive concerns in the medical chart were also performed. Finally, we sought to investigate risk or protective factors that might mediate or moderate the expected associations between CI and physical health. We hypothesized that, compared to cognitively intact older adults, those with mild to moderate/severe CI would be more likely to demonstrate inadequate health literacy and chronic disease self-management skills and poorer physical and mental health at baseline, and also greater decline in these outcomes at follow-up. We also proposed that health literacy, chronic disease self-management skills, and mental health symptoms would mediate associations between CI and physical health, while measures of patient activation, physical activity, and social support would moderate associations. This research is significant as it aims to better characterize and understand the relationship between CI, health literacy and self-management skills, and prospective adverse health outcomes. This approach is innovative as few research studies have had the ability to investigate these pathways, let alone examine them longitudinally and as comprehensively. The long-term goals of this research are to ultimately develop health system-based, behavioral interventions to improve CI care management and health outcomes among this population. Findings from this investigation were variable. Foremost, CI was prevalent in our sample (~36%). While an exploratory analysis, there was considerable discrepancy between CI identified on our objective-based testing versus CI identified by the healthcare system. Less than 10% of our study-identified CI participants were identified as having a cognitive concern or deficit in their medical charts. In our primary analyses, we found that mild and moderate/severe CI was strongly associated with health literacy and self-managements skills. This effect was comparable to other well-established predictors, such as socioeconomic factors, and was dose-dependent, with individuals who demonstrated more significant impairments exhibiting the greatest risk. While adults with CI continued to demonstrate poorer health literacy and self-management skills compared to cognitively intact older adults at follow-up, rates of change remained relatively stable overall. Only when accounting for demographic covariates and other factors that may impact change over time, was a relationship between CI and greater decline on the Multimedia self-management task and Medication Dosing task observed. Conversely, performance on the Spoken subtest appeared to improve among older adults with CI compared to those with no CI, while no significant change was observed for the Print task or two health literacy measures. We observed no significant associations between CI and baseline functional health status, after controlling for relevant covariates. Average unadjusted mean scores over time remained relatively stable, or even improved. However, after adjustment for relevant demographic variables, time since baseline, and change in cognition, adults with CI reported a significantly greater increase in depressive symptoms, and at a magnitude which would be considered to be clinically significant. There was no significant change in anxiety symptoms observed from baseline to follow-up among participants with CI, when compared to more cognitively intact older adults. When accounting for older age ranges within our sample, however, we observed that older adults with milder CI and who were over the age of 70 reported a significant improvement in physical function over time. In mediation and moderation analyses, all proposed pathways were non-significant. In summary, CI not meeting thresholds for dementia appears to go relatively undetected within the medical system. Furthermore, CI appears to be a strong predictor of cross-sectional associations with health literacy and self-management skills. While associations between CI and poorer performance on these tasks were maintained 2-5 years later, skills within these domains appear to remain generally preserved over time. That is, older adults with CI do not appear to exhibit greater rates of declines in these constructs compared to cognitively normal older adults. When accounting for factors that may impact cognition, however, some trends emerge among skills requiring more cognitive demand, or delivered using more complex, potentially unfamiliar modalities. Finally, while associations between CI and measures of physical and mental health were non-significant initially, CI does appear to be a predictor of greater increase in depressive symptoms over time. Pathway analyses were non-significant, and likely require additional research. Overall, these findings suggest that CI may have a greater impact in chronic disease self-management than clinicians or current diagnostic paradigms may presently appreciate, and that changes in these skillsets may potentially used as a marker of impaired cognition among an older adult population. Increased awareness of the negative impact of CI on these domains may help identify more adults within community-based settings who struggle with CI. Interventions focused on better managing the self-care and emotional sequelae of CI may also be targets of increased research and intervention efforts among this population.

Creator

• Lovett, Rebecca Mullen

DOI

• https://doi.org/10.21985/n2-hh8r-5v48

Subject

• Clinical psychology

Language

• en

Alternate Identifier

• http://dissertations.umi.com/northwestern:14694
• etdadmin_upload_666549

Date created

• 2019-01-01

Resource type

• Dissertation

Rights statement

• In Copyright