Background
Depression is an umbrella term that encompasses several constellations of symptoms and impairments that can affect an individual over their life course. As of 2021, 8.3% of Americans ages 18 and over meet the diagnostic criteria for major depression. Among adults 65 and older, 2.8% experienced an episode of major depression in 2021, while 13.2% experienced clinically significant symptoms of depression (also often referred to as minor depression) in 2020.1 Both major depression and clinically significant symptoms of depression (hereafter referred to as symptoms of depression) can result in significant impairment and impacts on functioning that would benefit from clinical intervention. Notably, while there is a growing recognition of the differences between major depression and symptoms of depression, and the negative impacts of both, there has been limited work on their presentation in older adults specifically.
An episode of depression has both immediate and long-run effects on an individual’s well-being. An episode of depression can affect an individual’s social connections and productivity, subsequently imposing disruptions to careers, lifetime earnings, and family life. Older adults experience the social and economic consequences of depression both through ongoing episodes and the downstream effects of past episodes.
In addition, depression is a recurring illness, with at least half of those who recover from one episode of major depression and 80% of those with 2+ episodes experiencing at least one other episode later in life. Those with a history of major depression experience between 5-9 episodes across their lifetimes with a recurrent episode typically beginning within 5 years of a previous episode. The effects of recurring episodes of depression can compound and further impair one’s well-being, especially in older adulthood. Aging often introduces new stressors including financial instability, loss of friends and family, and changes in physical and cognitive health that can harm one’s mental health. Among those 65 and older in 2020, data from the Health and Retirement Study (HRS) show that 61.6% of those with symptoms of depression had previously experienced symptoms of depression between ages 50 and 64.2
Thus, while depression is common and consequential at all ages, this paper highlights the unique manifestations of depression among older adults. We first examine how the prevalence of depression varies across adults with different characteristics at a point in time. We find that some groups – women, Hispanic individuals, people with limited financial resources, people with functional limitations, and people living alone – are much more likely to experience major depression or symptoms of depression. We also examine which factors are predictive of an individual’s likelihood of experiencing symptoms of depression in the future. We find that having a history of symptoms of depression or functional limitations are both highly predictive of experiencing depression late in life. Living alone and lower wealth during one’s working years are also predictive of symptoms of depression, but to a much lesser degree. We conclude by discussing how our results can be used to guide policy efforts aimed at reducing depression among older adults.
Data and methods
We measure rates of major depression and symptoms of depression using data from the National Survey on Drug Use and Health (NSDUH) and the Health and Retirement Survey (HRS) respectively. Major depression in the NSDUH is defined and measured according to the Diagnostic and Statistical Manual of Mental Disorders (DSM)-5 criteria. The survey identifies major depressive episodes (MDE) based upon nine symptoms associated with a MDE, such as depressed mood, diminished interest or pleasure, insomnia, fatigue, thoughts of death, and symptom duration.3 The HRS includes a widely-used and validated scale that measures symptoms of depression, known as the Center of Epidemiological Studies Depression Scale (CES-D). Based on research linking symptom levels and diagnostic criteria for depression, an individual is classified as having significant symptoms of depression if they score 4+ out of 8 depression symptoms. We consider the identification of those with symptoms that fall below the threshold for major depression to be a strength of this study as it permits us to assess the circumstances of older adults experiencing clinically significant symptoms of depression more completely.
Using the 2021 NSDUH and the 2020 HRS, we first present the patterns of depression across people with different characteristics at a point in time. We examine rates of depression by demographic characteristics, functional limitations, household income, household wealth, and living arrangements. Functional limitation is measured with both activities of daily living (ADLs) and instrumental activities of daily living (IADLs). ADLs include bathing, eating, dressing, toileting, walking across a room, and getting into or out of bed, while IADLs include using a telephone, taking medication, handling money, shopping, and preparing meals. We then make use of the longitudinal nature of the HRS to gain insight into which aspects of older adult’s past circumstances are the most predictive of symptoms of depression specifically at age 76. We examine symptoms of depression at age 50-54, income and wealth at age 50-54, and the presence of ADLs or IADLs at age 72. We focus on age 50-54 for the past symptoms of depression, income, and wealth measures because this is the earliest point in the HRS sample. We measure ADLs or IADLs at 72 as functional limitations were most common at this age prior to age 76. We use multivariate linear probability models to estimate the independent predictive power of each variable, and then conduct a decomposition of the variance analysis to understand which variables have the largest impact on the predictive power of the model.4
We report associations as it is difficult to establish causal links. While the literature has shown that depression at younger ages leads to lower income, savings, and less family stability, other linkages are less clear. The associations presented provide insights into how symptoms of depression and their potential correlates manifest.
Results and findings
Table 1 shows the rates of major depression and symptoms of depression by demographic groups. 2.8% of those 65 and older had major depression in 2021 and 13.2% of those 65 and older had symptoms of depression in 2020. Women, Hispanic individuals, and those below the federal poverty line have higher rates of both major depression and symptoms of depression.
Additional analyses of the HRS revealed that adults living alone were 54.3% more likely to have symptoms of depression compared to those not living alone, and retirees with no children and living alone were 46% more likely to have symptoms of depression compared to those not living alone. We also find that older adults with 2+ ADLs were 3.8 times as likely to have symptoms of depression compared to those with 0 or 1 limitations in ADLs. In addition, those in the first quintile of the income distribution of older adults ($0-$23,300) were 3.04 times as likely to have symptoms of depression compared to those in the top income quintile ($138,000-$5,700,000). Those in the first quintile of the asset distribution were also 3.07 times as likely to have symptoms of depression as those in the top quintile of the asset distribution.
In Table 3, we examine the prevalence of symptoms of depression by income and functional limitations simultaneously. In both income categories, older adults with 2+ ADL limitations are about 30 percentage points more likely to have symptoms of depression compared to those with less than 2 ADL limitations. Within each ADL group, adults in the bottom 40% of the income distribution are around 9 percentage points more likely to experience symptoms of depression than adults in the top 60% of the income distribution. We find a similar relationship between symptoms of depression and IADLs. Older adults with 2+ IADL limitations, regardless of income, are far more likely to have symptoms of depression, but within functional limitation groups, those in the top three income quintiles were less likely to have symptoms of depression compared to their lower-income counterparts.
Now, we turn to a longitudinal analysis to better identify which factors of an older adult’s history are most predictive of symptoms of depression. Approximately 12% of those age 76 had symptoms of depression in 2020. Our estimates show that one very strong predictor of symptoms of depression at age 76 is prior symptoms of depression that occurred at ages 50-54. Holding other characteristics of individuals and their circumstances constant, having symptoms of depression between ages 50-54 is associated with a 14-16 percentage point increase in the probability of having symptoms of depression at 76, depending on the specification (Appendix A).
Our estimates also indicate that functional limitations are important predictors of symptoms of depression. We examine both ADL and IADL limitations. When we include an indicator for having 2+ ADLs, it predicts a 22-percentage point increase in the probability of having symptoms of depression at age 76; similarly, when we include an indicator for having 2+ IADLs, it predicts a 25 percentage point increase (Appendix A). Together, prior symptoms of depression and functional limitations are the strongest predictors of symptoms of depression among older adults accounting for 65% of the explained variation in depression at age 76 (Appendix B). In contrast, demographics (sex and race) account for about 4% of the explained variation in depression.
Wealth between 50-54 and living arrangements at age 72 are estimated to be predictive of depression at age 76, although to a much lesser degree than other predictors included in our multivariate model. Living alone at age 72 is associated with about a 5 percentage point increase in the probability of having symptoms of depression at age 76 (Appendix A). However, living arrangements only account for 5.1% of the explained variation in depression at age 76 (Appendix B). We also find increased wealth between 50-54 is negatively associated with having symptoms of depression at age 76. Still, we find that the reduction in probability of experiencing symptoms of depression at age 76 associated with an additional $100,000 in household wealth is only 2.8% as large as the increased in risk associated with having prior depression (Appendix A). Finally, our regression analysis does not produce consistent evidence of income being a significant predictor of symptoms of depression. However, income and wealth are highly related. Therefore, while our bivariate analysis showed overlaps between symptoms of depression and income, this could instead be capturing the relationship between wealth and symptoms of depression.
Thus, the main finding from our regression analysis is that measures of prior mental and physical health status are the strongest predictors of symptoms of depression among older adults.
Policy Implications
The strong predictive power of functional limitations and a history of depression can help guide efforts to reduce rates of symptoms of depression at older ages.
One avenue that our results suggest may merit greater attention is improved access to mental health care for older adults. There are likely major opportunities to expand access to care given that there are large gaps in treatment for depression: 31% of those ages 65 and over with major depression in 2021 did not receive any past-year treatment.5 Our results, notably our finding that functional capacity is a strong predictor of symptoms of depression, can also help guide screening efforts at older ages. Such predictors can help target care to those with the greatest risk of symptoms of depression.
Similarly, if association between prior and current depressive symptoms among older adults that we document here reflects a causal effect of past symptoms on current symptoms (as opposed to, for example, just a persistent higher susceptibility to these symptoms), then this fact would suggest that there are opportunities to address depression at older ages by intervening earlier in the life course. Several treatment options have been shown to prevent relapse including cognitive behavior therapy (CBT) and anti-depressant medications, although, there is disagreement about the strength of the evidence supporting some of these therapies. Still, it is clear there is considerable room for improvement: 34% of those with major depression between the ages of 35-64 did not receive any past-year treatment in 2021.6
While much attention has been paid to the mental health crisis in the U.S., there is less focus on the unique manifestations of depression among older adults. This analysis begins to hone in on which factors are predictive of an individual’s likelihood of experiencing symptoms of depression in the future with the hope that providers, caregivers, and policymakers alike can appropriately support the mental health of older adults. These considerations become particularly salient in light of the aging population.
-
Acknowledgements and disclosures
The authors thank Matthew Fiedler for careful review of this article. They also thank Ben Graham and Caitlin Rowley for excellent research and editorial assistance.
This work was supported by a grant from The SCAN Foundation.
The Brookings Institution is financed through the support of a diverse array of foundations, corporations, governments, individuals, as well as an endowment. A list of donors can be found in our annual reports published online here. The findings, interpretations, and conclusions in this report are solely those of its author(s) and are not influenced by any donation.
-
Footnotes
- Authors’ analysis of the 2021 National Survey on Drug Use and Health or NSDUH and 2020 Health and Retirement Study or HRS.
- This is likely an underestimate of lifetime recurrence since the HRS only follows individuals starting at age 50.
- Consistent with the DSM-5 definition, an individual is classified as having an MDE in their lifetime if they had at least five or more of the nine symptoms nearly every day in the same 2-week period, where at least one of the symptoms was a depressed mood or loss of interest or pleasure in daily activities. Among respondents who were identified as having a MDE in their lifetime, those who indicated a period of depression during the past 12 months were also classified as having MDE in the past year.
- We also validated that our findings remained consistent even if using logistic regressions rather than linear regressions.
- Authors’ analysis of the 2021 NSDUH.
- Authors’ analysis of the 2021 NSDUH.