43. BEYOND ENTERTAINMENT: MEDIA PATTERNS AND ASSOCIATIONS WITH DEPRESSION IN OLDER ADULTS

Abstract

Introduction

As individuals age, leisure time often increases, leading to greater engagement in less physically demanding activities like various forms of media, including television, books, and computer use (Matthews et al., 2008). While media use has been a significant area of research among younger populations, it remains underexplored in studies involving older adults. In particular, the connections between media usage and mental health—specifically, depression— in this demographic are even less understood. For example, do individuals with depression tend to favor certain types of media over others? It is essential to examine actual data and patterns of media usage in older adults, especially given that depression significantly increases mortality rates in elderly populations aged 65 and older compared to non-depressed counterparts (Schulz et. al., 2000). Although pharmacological treatments for depression are available, older adults often face increased sensitivity to side effects, potential interactions with other medications, and issues related to drug metabolism. Polypharmacy and the cumulative effects of medication usage over decades also warrant caution amongst older populations. As such, examining how behavioral preferences and variations in media consumption impact mental health among aging populations is important for developing more holistic guidelines to inform health delivery and policy. This research therefore aims to improve the quality of life for old age individuals by investigating behavioral patterns and associations with depression, particularly by addressing gaps in understanding media preferences among older adults.

Methods

This analysis utilized 2020 Public Survey data and 2021 Consumption and Activities Mail Survey (CAMS) from the Health and Retirement Study (HRS), sponsored by the National Institute on Aging (NIA U01AG009740) and conducted by the Univ. of Michigan. Data missingness ranged 0-2.4% and were multiply imputed (m=20) under the missing at random mechanism using random forest classification and regression (analytical sample N=3,710, > 55 years old). Respondents were categorized as having significant depressive symptoms (N=532; 14.7%) based on previously-validated CESD-8 score ≥4 cutoff. Media activities (MAs) included TV/movies, newspapers/magazines, books, music, communication (telephone, letters, e-mail, facebook, skype), and computer use. Covariates included age, gender, race, ethnicity, marital status, employment status, educational attainment, and degree of functional impairment (e.g., ADLs, IADLs). Negative binomial (NB) and binary logistic regression models were used to determine relationships between MAs, CESD score and depression category. Descriptive statistics, univariate comparisons, and regression model estimates and confidence intervals were obtained by pooling results across imputed datasets according to Rubin’s rules.

Results

Respondents with significant depressive symptoms were younger, more functionally impaired, less educated, and more likely to be female, non-White, Hispanic, non-married, and unemployed than those without. Univariate analysis indicated that those with depressive symptoms engaged in less leisure reading of newspapers/magazines (2.72±4.3 vs. 3.46±5.8 hrs/wk; Z=4.36, p < 0.005) and books (3.04±6.5 vs. 3.92±7.5 hrs/wk; Z=3.90, p < 0.005), and fewer hours of computer use (8.33±14.8 vs. 10.5±14.7 hrs/wk; Z=6.37, p < 0.005), but no differences in TV, music, or communication. MAs were weakly correlated (rho=0.00-0.28). After covariate adjustment in NB regression (R2=17.6%), TV use (but no other MA) was positively associated with a very small increase in depressive symptoms, such that a 10 hr/wk increase in TV use was associated with a 2.9% increase in CESD-8 score. Logistic regression did not indicate any covariate-adjusted associations between MA and the presence of significant depressive symptoms.

Conclusions

This study reveals a notable demographic finding: respondents with significant depressive symptoms were more likely to be non-white, Hispanic adults, in contrast to 2020 CDC data demonstrating higher rates of depression among non-Hispanic, white adults. Other findings were consistent with existing literature including that respondents with significant depressive symptoms were more likely to be female, functionally impaired, less educated, non-married, and unemployed. Future research should explore the reason for these differences and whether they reflect different sampled populations, cultural factors, or areas of disparity. Understanding the intersection of social factors and mental health outcomes may facilitate the creation of effective, targeted strategies in mental health care.

Regarding media usage and its association with depression, non-depressed respondents reported higher engagement in cognitively demanding media activities, such as reading magazines and books, and spent more time using computers. This study did not explore specific activities engaged in during computer use. However, if a detailed analysis is conducted, it may reveal support for certain software applications or the development of workshops to educate individuals with limited digital skills. Increased hours spent watching TV, traditionally viewed as less cognitively demanding, were associated with a very small increase in depressive symptoms. Although this study suggests different associations between cognitively active versus passive forms of media use, it does not formally differentiate these categories in a structured manner. We hope to determine a more systemic method to distinguish between active and passive media engagement and their respective impacts on mental health.

Notably, there was no significant difference between the two groups regarding hours/week spent on communication media usage. This finding is surprising as studies have shown online communication can be protective against developing depression in elderly populations (Nakagomi et al. 2022; Wiwatkunupakarn et al. 2022). However, this study only examined aggregated hours per week in communication use via telephone, letters, email, or social media, and did not examine each aspect individually or include in-person interactions.

Although this study does not examine the causal relationship between media use and depression, it provides valuable insights into behavioral patterns in elderly individuals with depressive symptoms, potentially aiding in the identification of early warning signs. Understanding patterns of media usage can also inform personalized behavioral approaches to mental health care providing increased options for multi-modal recommendations. By identifying patterns of media usage and demographic differences, these findings have the potential to inform future research and the development of health policies and guidelines tailored to the mental health needs of aging populations.