Skip to main content
NCBI home page
The Gerontologist logoLink to The Gerontologist
. 2022 Aug 2;62(10):1454–1465. doi: 10.1093/geront/gnac110

Work Expectations, Depressive Symptoms, and Passive Suicidal Ideation Among Older Adults: Evidence From the Health and Retirement Study

Briana Mezuk 1,2,✉, Linh Dang 3, David Jurgens 4, Jacqui Smith 5
Editor: Suzanne Meeks
PMCID: PMC9710239  PMID: 35914806

Abstract

Background and Objectives

Employment and work transitions (e.g., retirement) influence mental health. However, how psychosocial contexts such as anticipation and uncertainty about work transitions, irrespective of the transitions themselves, relate to mental health is unclear. This study examined the relationships of work expectations with depressive symptoms, major depression episodes (MDE), and passive suicidal ideation over a 10-year period among the “Baby Boom” cohort of the Health and Retirement Study.

Research Design and Methods

Analysis was limited to 13,247 respondents aged 53–70 observed from 2008 to 2018. Past-year depressive symptoms, MDE, and passive suicidal ideation were indexed using the Composite International Diagnostic Interview—Short Form. Expectations regarding working full-time after age 62 were assessed using a probability scale (0%–100%). Mixed-effect logistic regressions with time-varying covariates were used to assess the relationship of work expectations with mental health, accounting for demographics, health status, and functioning, and stratified by baseline employment status.

Results

At baseline, higher work expectations were inversely associated with depressive symptoms. Longitudinally, higher expectations were associated with lower odds of depressive symptoms (odds ratio [OR] = 0.93, 95% CI: 0.91, 0.94). This association was more pronounced among respondents not working at baseline (ORNot working = 0.93 vs ORWorking = 0.96). Greater uncertainty (i.e., expectations near 50%) was also inversely associated with depressive symptoms. Results were similar for past-year MDE and passive suicidal ideation.

Discussion and Implications

Expectations (overall likelihood and uncertainty), as indicators of psychosocial context, provide insight into the processes that link work transitions with depression risk.

Keywords: Cohort, Depression, Employment, Longitudinal, Uncertainty

The illusion that we understand the past fosters overconfidence in our ability to predict the future.―Daniel Kahneman

Depression is among the leading causes of disability worldwide, and is an established risk factor for suicide (U.S. Department of Health and Human Services, Office of the Surgeon General, 2021). However, the processes and mechanisms that both link, and differentiate, depression and suicide risk over the life course are not well understood (Franklin et al., 2017). An important element of a life course approach to suicide prevention is identifying “points of engagement” to promote and support mental health prior to the development of suicide crisis (Caine et al., 2011). Advocates and policy makers have called for research to identify salient factors related to the timing and context of major life transitions to inform these prevention efforts (U.S. Department of Health and Human Services, Office of the Surgeon General, 2021). Understanding factors that contribute to depression and suicide risk during major life transitions can inform the development of intervention programs and policies that support individuals during these periods of change, far upstream from episodes of crisis.

A Life Course Approach to Examining Employment Transitions and Suicide Risk

One common type of major life transition that may provide new avenues for suicide prevention among older adults pertains to work. While labor force status is generally operationalized as a categorical state (i.e., employed, retired, unemployed, etc.), each of these states represents a transition process that unfolds over time (Clark & Summers, 1978). Common work transitions in later life include changes in occupation (i.e., second careers) and the amount of time spent at work (i.e., reducing hours), as well as leaving the labor force through various mechanisms. These exits include planned retirements, unplanned job losses that eventuate in “retirement” due to inability to find another position at the desired pay, hours, health-related leaves-of-absence, or short-term disability (Ekerdt, 2010; Jacobs & Piyapromdee, 2016).

It is important to situate research on employment transitions and mental health within the life course framework (Amick et al., 2016). This framework focuses on the ways in which work-related factors (e.g., job strain, pay, hours) intersect with broader contexts, including individual (e.g., age, gender, income, health, and functioning), family (e.g., relationship status and quality, partner work, and health factors), and community (e.g., area-level unemployment, social safety-net programs) characteristics over time. This interdependence reflects the notion of embedded social relationships, that is, individuals are embedded within their relationships with family, friends, employers, and communities. These intersections in turn influence both the options individuals have, and the choices they make, regarding employment transitions across the life span (Amick et al., 2016). It illustrates how the same transition can have different implications for mental health, depending on these other factors.

A large body of research has explored how social, psychological, and economic aspects of these various work transitions relate to mental health from this life course perspective (Kim & Moen, 2002; Lee & Kim, 2017; Sjöberg, 2021). For example, unexpected job loss and involuntary retirement are established risk factors for depression and suicidal ideation (Abrams et al., 2022b; Artazcoz et al., 2004; Dooley et al., 1994; Mosca & Barrett, 2016). However, voluntary retirement, particularly from high-stress jobs, is generally associated with improvements in mental health (Fleischmann et al., 2020; Lahdenperä et al., 2022). Collectively, this research indicates that work transitions may be potential points of engagement for suicide intervention and prevention for older adults.

The Salience of the Expectedness of Major Life Transitions for Mental Health

One contextual factor that may influence the relationship of such work transitions and mental health is the expectedness of the event. Some transitions are anticipated (i.e., financial planning to retire by a particular age; quitting a job after lining up another) and some are unanticipated (i.e., being unable to return to work after a health event; being laid-off during a recession; Ekerdt, 2010). Expectations also provide a means to understand the role of non-transitions, such as continuing to work beyond an anticipated age due to uncertainty regarding health, finances, etc. or of on-time versus off-time transitions. In one of the few studies to address this issue, Abrams et al. (2022a) examined the relationship between “met” (e.g., reporting ≥90% probability of working after age 62 and actually working past that age) and “unmet” expectations (e.g., working at age 62 despite previously reporting there was “no chance” this would occur) among older workers. They found that “met” expectations (irrespective of whether the expectation had been to be working or not) were unrelated to depressive symptoms. However, those who had expected to be working, but were not, had elevated depressive symptoms (Abrams et al., 2022a). Similar findings have been found pertaining to met versus unmet work expectations and life satisfaction (Clarke et al., 2012). Collectively, these results suggest that expectations about future employment may have implications for mental health for older adults.

There are strong theoretical reasons for believing that such expectations, in and of themselves, relate to mental health. First, expectations are fundamentally subjective assessments about the likelihood of future events and related to several other future-oriented psychosocial constructs, particularly hopelessness and anxiety, which are established risk factors for depression and suicide risk behaviors. For example, leading psychological theories of depression and suicidal behavior emphasize the role of hopelessness, negative thoughts about the future, cognitive inflexibility, and low problem-solving ability (Beck et al., 1974; Dombrovski et al., 2019; Gibb et al., 2006; Giner et al., 2016; Liu et al., 2015). Hopelessness and cognitively inflexibility are correlated with subjective expectancies about future events (Bjärehed et al., 2010; Harris & Hahn, 2011; MacLeod & O’Connor, 2018; Thimm et al., 2013). Similarly, theoretical models of anxiety and depression emphasize the central role of uncertainty, including its implications for decision-making (Aberg et al., 2022; Grupe & Nitschke, 2013). Taken together, this suggests that not only the direction of the expectation, but also the certainty expressed by it (i.e., high certainty represented by both “almost certain” and “no chance” extremes) versus high uncertainty (i.e., “50/50 chance”) may have implications for mental health (Kiani et al., 2014; Zinn, 2004).

Second, future-oriented constructs of hopelessness and anxiety are longitudinally related to depression and suicide risk, regardless of whether negative (or positive) events happen in the future. For example, Gibb et al. (2006) showed that cognitive inflexibility strengthened the relationship of experiencing daily hassles and depressive symptoms (Gibb et al., 2006). Consistent with the findings regarding work expectations and realizations described above, they also showed that cognitive inflexibility was significantly related to depressive symptoms even after accounting for experiences of daily hassles (Gibb et al., 2006). This is consistent with other work emphasizing the importance of personal characteristics related to cognition (e.g., problem-solving ability) and temperament (e.g., impulsiveness; Giner et al., 2016), as well as concepts that span the person and the event, such as vigilance (Helzer et al., 2009) and perceived control (Frazier et al., 2011), in and of themselves, as important determinants of depression and related outcomes in the context of life events.

Rationale and Focus of the Present Study

Several important gaps remain in our understanding of how work expectations relate to mental health. First, most existing studies of expectations are limited to people who were employed contemporaneously at the time the expectation was assessed (e.g., Abrams et al., 2022a; Clarke et al., 2012), which may reduce the amount of variation observed in the reported expectations. Second, most investigations of work expectations and mental health have focused primarily on the issue of future employment realizations conditional on anticipations (e.g., met vs unmet expectations). However, only examining expectations in relation to realizations implies a preferred state (i.e., mismatches between expectations and realizations are not preferred, and therefore have negative implications for mental health). However, expecting to work (or not) and wanting to work (or not) are not interchangeable anticipations about the future (Ekerdt, 2010). Finally, much of the research exploring the intersection of future-oriented constructs and mental health outcomes has been limited to younger samples (e.g., Frazier et al., 2011; Gibb et al., 2006; Helzer et al., 2009), and thus the relevance of these types of constructs to depression and suicide risk among older adults is not well understood.

Building on prior work, this paper explored two questions:

  1. What is the relationship of work expectations with past-year depressive symptoms, major depression episodes (MDE), and passive suicidal ideation over a 10-year period (2008–2018) among the “Baby Boom” cohort (born 1948 to 1965)?

  2. Does the relationship of work expectations with these mental health outcomes vary by work history?

We focused on the “Baby Boom” cohort for three reasons: (1) they experienced the Great Recession (2008–2009) during the period of their working lives directly preceding median retirement age, and (2) they were the first in the United States to have broad access to 401k and similar defined-contribution retirement plans rather than defined-benefit plans (i.e., traditional pensions; instituted in 1978; US Bureau of Labor Statistics, 2021; US Department of Labor, 2021). Both factors may influence expectations regarding working in later life. Finally, (3) over the past three decades, the Baby Boom cohort has demonstrated higher risk of completed suicide compared with other older generations (Hedegaard et al., 2020). In sum, this cohort has several important characteristics related to employment, retirement planning, and mental health that make it an important setting to explore the relationships of work expectations with suicide risk factors.

Method

Data Source and Study Population

The Health and Retirement Study (HRS) is a nationally representative longitudinal survey of U.S. adults aged 51 or older. Since 1992, approximately 20,000 respondents are interviewed biennially to assess a range of psychosocial characteristics, health history, economic circumstances, and employment history. The HRS employs a complex, multistage design, including oversampling Hispanics, African Americans, and Florida residents, and enrolls additional respondent cohorts every 6 years to remain representative of the current older adult population. Additional details of the study design are described elsewhere (Sonnega et al., 2014).

The present study focused on three “Baby Boom” cohorts: Early Baby Boomers (EBB, born 1948 to 1953), Mid Baby Boomers (MBB, born 1954 to 1959), and Late Baby Boomers (LBB, born 1960 to 1965). We included 13,247 nonproxy respondents (3,955 EBB, 4,879 MBB, and 4,413 LBB) who had complete data on work expectations, mental health outcomes, and potential covariates for at least one interview wave from 2008 to 2018. We chose these interview waves because the Composite International Diagnostic Interview (CIDI) has been included in the survey since 2008, and 2018 was the latest available interview wave at the time of this analysis. Of this analytic sample, 388 (2.9%) died prior to age 62. Supplementary Figure 1 details our analytic sample selection process. Supplementary Table 1 illustrates the differences between our analytic sample and the entire HRS cohort; most relevant to this investigation is that excluded respondents were more likely to not be working at baseline compared to the analytic sample.

The HRS is approved by the Institutional Review Board at the University of Michigan and all respondents provided written informed consent. This analysis used only publicly available data and was exempt from human participants regulation.

Work Expectations

The subjective probability of working full-time after age 62 was assessed by asking: “Thinking about work generally and not just your present job, what do you think are the chances that you will be working full-time after you reach age 62?” Original responses were recorded on the scale of 0–100. For this analysis, these responses were rescaled to 0–10 for a more meaningful interpretation of regression coefficients (i.e., one-unit increase in the scaled work expectation is equivalent to a 10-unit increase in the original variable). This question was asked at every interview wave and was modeled as a continuous time-varying exposure. We explored the possibility of nonlinearity (e.g., quadratic terms) in our model fitting, but none significantly improved model fit.

Mental Health Outcomes

Mental health outcomes were indexed by the CIDI—Short Form. The CIDI is a fully structured diagnostic interview that assesses the past 12-month depressive symptoms based on the Diagnostic and Statistical Manual of Mental Disorders—IV (DSM) criteria for major depressive episode (Kessler et al., 1998). Previous studies have shown that CIDI has moderate concordance with clinical psychiatric examinations (Eaton et al., 2007). The CIDI depression module assesses eight symptom domains of MDE (i.e., sadness, anhedonia/lost interest, sleeping disturbances, appetite changes, fatigue, guilt/worthlessness, concentration problems, thinking about death). Higher scores on the CIDI indicate more severe depressive symptomatology.

The present study examined three mental health outcomes during the past 12 months: (1) elevated depressive symptoms, indexed by screening into the full CIDI module (yes/no), (2) MDE (yes/no), and (3) passive suicidal ideation (yes/no). We defined screening positive on the CIDI as having reported feeling depressed or loss of interest for at least 2 weeks at sufficient frequency and intensity in the past 12 months (Dang et al., 2020). MDE were defined as endorsing five or more symptom domains on the CIDI, consistent with DSM diagnostic criteria. Passive suicidal ideation was assessed by the last item of the CIDI and asking respondents whether they “thought a lot about death—either your own, someone else’s, or death in general?” during the 2-week period during which they felt depressed or had low interest. These three outcomes were assessed at each wave and were treated as time-varying variables.

Covariates

All models adjusted for covariates related to demographics, health behaviors, and health status that have been associated with expectations and mental health in prior work. These included interview wave (with the baseline wave coded as 1; respondents could have a maximum value of 6) and birth cohort (EBB, MBB, LBB); and sociodemographic characteristics such as age (years), sex (male, female), race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, other), education (less than high school, high school or equivalent, some college, college or above), marital status (never married, married/partnered, divorced/separated, widowed), household income (all USD: ≤$20,160, $20,161–$50,400, $50,401–$101,012, ≥$101,013), and total wealth (all USD: ≤$5,025, $5,026–$91,500, $91,501–$342,500, ≥$342,501). Health behaviors included smoking status (never, ever), alcohol use (never, ever); health characteristics included having at least one chronic condition such as hypertension, diabetes, cancer, chronic lung disease, heart attack, stroke, arthritis (any/none); and functioning, for example, need helps with basic activities of daily living such as dressing, bathing, eating, toileting, walking/transferring (any/none). All covariates were time-varying as a function of interview wave except for race/ethnicity, education, and sex.

Statistical Analysis

We compared the baseline sociodemographic and health characteristics as a function of CIDI screening status using Student’s t tests for continuous variables and chi-squared tests for categorical variables. Mixed-effects logistic regression with random intercepts was used to examine the relationships between expectations of working full-time after age 62 and the three mental health outcomes. This modeling approach accounts for the clustered nature of the data and the time-varying nature of the expectations and mental health outcomes, while also incorporating information from all assessments, not just those observations with complete follow-up over the 10-year period (Hardin & Hilbe, 2018). It also explicitly accounts for heterogeneity in the baseline expectation–depression relationship through the random intercept term (Rabe-Hesketh et al., 2004). Likelihood ratio test of the random intercept model versus model without was highly significant (χ 2 = 1,568.38, df = 1, p < .0001).

All regression analyses adjusted for interview wave, birth cohort, age, sex, race/ethnicity, education, marital status, household income, total wealth, health conditions, needs help with basic activities of daily living, smoking status, and alcohol use. We first examined the relationship of expectations with the three mental health outcomes in the entire analytical sample. We then fit models stratified by baseline working status (working for pay vs not), to examine whether employment status moderated the relationship between work expectations and the mental health outcomes. To visualize the findings, we estimated the mean predicted probability for each mental health outcome within strata of baseline employment status. To assess the robustness of our findings, we explored the mixed-effects models with a random intercept and a random slope for work expectations. This modeling approach accounts for how the relationship of work expectations and mental health differ across individuals and time. In addition, we also estimated the average marginal effects for each mental health outcome.

To explore the notion of uncertainty, three additional analyses were performed with alternative operationalizations of work expectations. In the first, work expectations were categorized into a five-level variable (0%, 1%–49%, 50%, 51%–99%, and 100%) with the value of 50% set as the reference to reflect “uncertainty,” that is, 50/50 chance. In the second, work expectations were categorized into a four-level variable: absolutely certain (0% or 100%), somewhat certain that will not be working (1%–49%), uncertain (50%), and somewhat certain that will be working (51%–99%), with the “absolutely certain” responses (0% and 100%) set as the reference. Lastly, we expanded the categories of certainty from the second analysis and instead categorized expectations as absolutely certain (0%–10% or 90%–100%), somewhat certain that will not be working (11%–49%), uncertain (50%), and somewhat certain that will be working (51%–89%), with the “absolutely certain” responses used as the reference.

Analyses were performed using StataSE version 16.1 (StataCorp, College Station, TX). Significance tests were two-sided and evaluated at p < .05.

Results

Descriptive Statistics

Among 13,247 respondents in the analytic sample, 29.9% were EBB, 36.8% were MBB, 56% were female, and 48.9% were non-Hispanic White. Respondents with elevated depressive symptoms (i.e., who screened positive on the CIDI) were more likely to be female, had less than college education, were less likely to be married, had lower household income and wealth, smoker, nondrinkers, had at least one chronic condition, needed help with basic activities of daily living, and less likely to be working at baseline (all p < .0001; Table 1).

Table 1.

Sample Characteristics by Depression Screening Status on the CIDI-SF at Baseline: Health and Retirement Study, 2008–2018 (N = 13,247)

Screened into CIDI-SF moduleb
Characteristics at baseline Totala (N = 13,247) Yes (N = 1,977) No (N = 11,145) p Valuec
Cohort, N (%)
 Early Baby Boomers 3,955 528 (26.7) 3,382 (30.4)
 Mid Baby Boomers 4,879 742 (37.5) 4,080 (36.6) .001
 Late Baby Boomer 4,413 707 (35.8) 3,683 (33.1)
Average number of interview waves (mean ± SD)d 3.6 ± 1.7 3.5 ± 1.7 3.6 ± 1.7 —
Total person-yearsd 95,308 13,894 80,436 —
Total person-years lost due to attritiond 18,594 2,690 15,680 —
Sociodemographics
Age (mean ± SD) 13,247 54.3 ± 3.0 54.5 ± 3.1 .02
Sex, N (%)
 Female 7,420 1,342 (67.9) 6,040 (54.2) <.0001
 Male 5,827 635 (32.1) 5,105 (45.8)
Race/ethnicity, N (%)
 Non-Hispanic White 6,478 1,025 (51.9) 5,388 (48.3)
 Non-Hispanic Black 3,399 413 (20.9) 2,963 (26.6) .53
 Hispanic, regardless of race 2,488 410 (20.7) 2,051 (18.4)
 Other 882 129 (6.5) 743 (6.7)
Education, N (%)
 Less than high school 2,055 367 (18.6) 1,659 (14.9)
 High school or equivalence 3,996 624 (31.6) 3,334 (29.9) <.0001
 Some college 3,890 622 (31.5) 3,238 (29.1)
 College or above 3,306 364 (18.4) 2,914 (26.2)
Marital status, N (%)
 Never married 1,217 223 (11.3) 993 (8.9)
 Married/partnered 8,822 1,053 (53.3) 7,648 (68.6) <.0001
 Divorced/separated 2,651 572 (28.9) 2,077 (18.6)
 Widowed 549 128 (6.5) 420 (3.8)
Household income, N (%)
 US$0–20,160 3,142 756 (38.2) 2,361 (21.2)
 US$20,161–50,400 3,336 524 (26.5) 2,786 (25.0) <.0001
 US$50,401–101,012 3,404 409 (20.7) 2,959 (26.6)
 ≥US$101,013 3,365 288 (14.6) 3,039 (27.3)
Total wealth, N (%)
 ≤US$5,025 3,642 813 (41.1) 2,804 (25.2)
 US$5,026–91,500 3,474 527 (26.7) 2,915 (26.2) <.0001
 US$91,501–342,500 3,298 369 (18.7) 2,894 (26.0)
 ≥US$342,501 2,833 268 (13.6) 2,532 (22.7)
Work characteristics
Working for pay, N (%) 9,113 906 (45.8) 8,115 (72.8) <.0001
Expectation of working full-time after age 62, % (mean ± SD) 12,878 34.4 ± 37.8 46.4 ± 37.5 <.0001
Categories of expectations, N (%)
 0% 3,289 791 (40.0) 2,498 (22.4)
 Between 0% and 50% 2,838 365 (18.5) 2,473 (22.2)
 50% 1,747 195 (9.9) 1,552 (13.9) <.0001
 Between 50% and 100% 3,222 373 (18.9) 2,849 (25.6)
 100% 1,782 206 (10.4) 1,576 (14.1)
Health behaviors and characteristics
Ever smoke, N (%) 7,310 1,323 (66.9) 5,920 (53.1) <.0001
Ever drink alcohol, N (%) 9,030 1,252 (63.3) 7,701 (69.1) <.0001
At least one chronic condition (hypertension, diabetes, cancer, chronic lung disease, heart attack, stroke, arthritis), N (%) 9,020 1,622 (82.0) 7,332 (65.8) <.0001
Need help with at least one (dressing, bathing, eating, toileting, walking/transferring), N (%) 805 327 (16.5) 474 (4.3) <.0001
Depressive syndrome
CES-D score (mean ± SD) 13,122 4.1 ± 2.7 1.3 ± 1.8 <.0001
Met criteria for major depressive disorder (CIDI ≥ 5), N (%)b 1,058 1,058 (83.3)
Open in a new tab

Notes: CES-D = Center for Epidemiological Studies—Depression scale; CIDI-SF = Composite International Diagnostic Interview—Short Form for major depression; SD = standard deviation.

aTotal column reflected the number of respondents (out of 13,247) in each category of sample characteristics.

bOnly 13,122 respondents, rather than 13,247, had data on CIDI-SF screening questions at baseline

c p Value obtained from chi-squared test for binary and categorical variables, Student’s t test for continuous variables.

dEstimated for the entire analytical sample, Health and Retirement Study 2008–2018.

At baseline, screening positive on the CIDI was associated with lower work expectations compared to those who screened negative (mean expectation of working past age 62: 34.4% vs 46.4%, respectively; Table 1). We observed no significant cohort differences in baseline expectations of working full-time past age 62 (mean expectation of working full-time past age 62 was 44.6%, 43.7%, and 45.6% for EBB, MBB, and LBB, respectively). Non-Hispanic White adults had higher expectation of working compared to other racial and ethnic groups (Supplementary Table 2). Work expectations were not associated with age at baseline; however, respondents were less likely to expect working past age 62 as they got older (Supplementary Table 2).

Continuous Work Expectations and Mental Health

As shown in Table 2, higher expectations of working full-time after age 62 were associated with significantly lower odds of screening positive on the CIDI after adjusting for interview wave, demographics, health status, and functional impairment (odds ratio [OR] = 0.93, 95% confidence interval [CI]: 0.91, 0.94 for every 10% increase in expected likelihood of working full-time after age 62). In the models stratified by baseline employment status, there was an observed stronger inverse association between work expectations and screening positive on the CIDI among those not working for pay at baseline compared to those who were working (OR = 0.93 [95% CI: 0.90, 0.96] and OR = 0.96 [95% CI: 0.95, 0.98], respectively; Supplementary Table 3). We observed similar results for the MDE and passive suicidal ideation (Table 2 and Supplementary Tables 3–5). Post hoc tests of the interactions between expectations, race/ethnicity, and age showed that the relationships between expectations and the three mental health outcomes did not vary by race/ethnicity or age group (data not shown).

Table 2.

Odds Ratios and 95% Confidence Intervals for Association Between Work Expectation and Mental Health Outcomes: Health and Retirement Study, 2008–2018 (N = 13,247)

Odds ratios (95% CI)
Expectation of working full-time after age 62 Elevated depressive symptoms Major depressive episode Passive suicidal ideation
Continuous, 0–10 0.93 (0.91, 0.94)* 0.92 (0.91, 0.94)* 0.93 (0.91, 0.95)*
Variance for random intercept 4.44 4.92 4.41
0% 2.23 (1.89, 2.65)* 2.17 (1.81, 2.61)* 2.09 (1.71, 2.55)*
1%–49% 1.51 (1.27, 1.79)* 1.57 (1.31, 1.89)* 1.46 (1.19, 1.79)*
50% (ref) 1.0 1.0 1.0
51%–99% 1.04 (0.87, 1.23) 1.00 (0.83, 1.22) 1.01 (0.82, 1.25)
100% 1.11 (0.91, 1.35) 1.06 (0.85, 1.32) 1.10 (0.87, 1.39)
Variance for random intercept 4.41 4.90 4.39
0% or 100% (ref) 1.0 1.0 1.0
1%–49% 0.83 (0.73, 0.94)* 0.89 (0.78, 1.01) 0.83 (0.72, 0.96)*
50% 0.57 (0.48, 0.67)* 0.58 (0.49, 0.69)* 0.59 (0.49, 0.71)*
51%–99% 0.62 (0.54, 0.70)* 0.61 (0.53, 0.71)* 0.62 (0.53, 0.72)*
Variance for random intercept 4.45 4.95 4.43
0%–10% or 90%–100% (ref) 1.0 1.0 1.0
11%–49% 0.86 (0.74, 1.00) 0.90 (0.76, 1.06) 0.87 (0.73, 1.04)
50% 0.62 (0.53, 0.73)* 0.63 (0.53, 0.74)* 0.64 (0.53, 0.77)*
51%–89% 0.72 (0.63, 0.82)* 0.72 (0.61, 0.83)* 0.73 (0.62, 0.85)*
Variance for random intercept 4.46 4.96 4.44
Open in a new tab

Notes: All models adjusted for interview wave, birth cohort, age (continuous), sex, race/ethnicity, education, marital status, household income, total wealth, having at least one chronic condition, need helps with daily activities, smoking, and drinking.

*p < .05.

Figure 1A shows that higher expectations of working full-time after age 62 are associated with lower mean predicted probability of screening positive on the CIDI, a relationship that is more pronounced among respondents not working for pay at baseline. Similar results, albeit with smaller predicted probabilities, were observed for MDE and passive suicidal ideation (Figure 1B and C). However, there was no association between work expectations and probability of endorsing passive suicidal ideation when we conditioned on whether respondents had screened into the CIDI (Figure 1D), and there was also no substantial difference in probability of reporting passive suicidal ideation between respondents working for pay at baseline and those who were not.

Figure 1.

Figure 1.

Open in a new tab

Mean predicted probabilities of mental health outcomes by baseline work history. (A) For screening positive on the Composite International Diagnostic Interview—Short Form (CIDI-SF; sample size = 8,834); (B) for major depressive episodes (sample size = 8,834); (C) for passive suicidal ideation on the full analytical sample (sample size = 8,834); (D) for major depressive episodes conditioned on respondents screened positive on the CIDI-SF (sample size = 2,083).

Mixed-effects models with random intercept and random slope for work expectations yielded similar results. We found that random slope had small variances in all models, which suggested that the relationship of work expectations with mental health differed across individuals mainly due to baseline differences rather than changes within individuals over time (Supplementary Table 6). In addition, we also observed inverse associations between work expectations and elevated depressive symptoms, MDE, and passive suicidal ideation using average marginal effects, consistent with the observed ORs from the logistic models (Supplementary Table 7).

Employment Uncertainty and Mental Health

As shown by Table 2, compared with those with the highest uncertainty about working full-time after age 62 (i.e., 50% likelihood), those who were more certain that they would not be working (<50% expectation) were more likely to screen positive on the CIDI (OR0% = 2.23 [95% CI: 1.89, 2.65] and OR1%–49% = 1.51 [95% CI: 1.27, 1.79]). This relationship was evident when analyses were stratified by baseline employment status, although not all point estimates were statistically significant among those not working at baseline. Greater certainty about working at age 62 (expectations of 51%–99% or 100%) was not associated with screening positive on the CIDI (OR51%–99% = 1.04 [95% CI: 0.87, 1.23] and OR100% = 1.11 [95% CI: 0.91, 1.35]). In general, greater uncertainty (i.e., expectations between 1% and 99%) was inversely associated with screening positive on the CIDI (ORs1%–49% ranged from 0.73 to 0.83; ORs50% ranged from 0.57 to 0.67; and ORs51%–99% ranged from 0.61 to 0.81 in Table 2 and Supplementary Table 3) using the alternate operationalization of expectations with “absolute certainty” (expectations of either 0% or 100%) as the reference group. However, when the “absolute certain” category was expanded to included responses of 0%–10% or 90%–100%, these inverse associations were slightly attenuated (Table 2 and Supplementary Table 3). In general, results were similar for the outcomes of MDE and passive suicidal ideation (Table 2 and Supplementary Tables 4 and 5); however, the associations were slightly attenuated for the latter.

Discussion

The primary finding from this study is that expectations about future employment (here indexed as the probability of working full-time at age 62) are longitudinally related to mental health including elevated depressive symptoms, experiencing an MDE, and reporting passive suicidal ideation among older adults. These relationships persist after accounting for medical burden and functional impairment, and they are more pronounced among people who are not working for pay. The nature of the relationships between these expectations and depressive symptomatology is complex and has both quantitative and qualitative elements. We found that expectations are inversely associated with these mental health outcomes, such that lower expectations of work are associated with higher odds of experiencing poor mental health in a quantitative manner; we also found that certainty about working in the future, regardless of the quantitative value (i.e., 0% or 100%) was also associated with higher odds of poor mental health relative to uncertainty (i.e., 50/50 chance). These findings have implications for both theory and research on how expectations of the future relate to depressive symptomatology and can inform efforts to promote mental health of older adults as they experience workforce transitions.

These findings should be situated within the life course framework (Amick et al., 2016), which emphasizes how temporal (e.g., experiences earlier in life; behavioral norms related to age and cohort) and contextual (e.g., family structure and social support; policies regarding retirement planning, social safety nets, and health insurance) factors shape mental health in later life. Expectations provide one element of context for understanding work transitions, for example, “on-time,” planned retirements, unplanned job losses that eventuate in “retirement,” leaves-of-absence, return to the workforce after retirement, etc. (Ekerdt, 2010; Jacobs & Piyapromdee, 2016). Beyond these actual transitions, the findings from this study indicate that expectations about working in the future, in and of themselves, are related to mental health; that is, they are not only associated vis-à-vis “met” or “unmet” realizations (e.g., achievements or disappointments) of working status as others have explored (Abrams et al., 2022a).

Implications for Theories Regarding Depression and Suicide Risk Behaviors in Later Life

The findings that work expectations, in and of themselves, are related to mental health is consistent with several theories about the causes of depression and suicidal behavior that focus on hopelessness (Liu et al., 2015). The construct of hopelessness inherently invokes perceptions of the future, for example, “I can’t imagine what my life would be like in 10 years” (Beck et al., 1974). Hopelessness is related to many factors that are tied to expectations, including locus of control (Hong et al., 2021) and sense of purpose (Chen et al., 2020). Hopelessness is also inversely associated with engagement in various aspects of social life, including religious involvement (Mitchell et al., 2020) and volunteering (Kim et al., 2020), which are protective for depression in later life.

These results also suggest qualitatively important elements regarding certainty about the future and its relation to mental health. In general, acute situations characterized by high levels of uncertainty (exemplified by the coronavirus disease 2019 pandemic) are positively correlated with psychological distress (Park et al., 2021). However, as it relates to longer-term expectations of future events, uncertainty may indicate flexibility or willingness to adapt to new situations and/or recognition that factors outside one’s control can affect the likelihood of events (Hershfield, 2011). Furthermore, uncertainty itself may not be protective for mental health, but rather one’s ability to cope with that uncertainty; the construct of tolerance for ambiguity refers to how individuals interpret and process situations that are complex or have incomplete information (Furnham & Ribchester, 1995). While there are many scales for assessing this construct, items often address beliefs around problem-solving and adaptation, for example, “There’s a right way and a wrong way to do almost everything” and “I feel anxious when things are changing” (Greco & Roger, 2001; Kirton, 1981). Consistent with this, viewing uncertainty as threatening has been associated with symptoms of depression and anxiety (Saulnier et al., 2019; Tanovic et al., 2018), although less research has been done on this construct in older adults.

Implications for Practice

In terms of practice, these findings emphasize the role of employers as stakeholders in supporting the mental health of older adults. Many workplaces offer “wellness” or employee assistance programs and retirement benefits, but those initiatives are often siloed. These findings suggest an opportunity for novel synergies between such programs to support financial and emotional health during retirement transitions (Vrkljan et al., 2019), including support groups for workers who are considering retirement and the use of digital tools to support older adults during this transition (Stara et al., 2020). Employers could also consider developing new ways for retirees to remain engaged in the workplace, such as mentorship programs that may provide both social connections and a sense of purpose (Ng et al., 2019). In addition, health shocks are a known factor leading to unexpected exits from the labor force; however, both individual factors (e.g., emotional resilience; Berthung et al., 2021; Jacobs & Piyapromdee, 2016) and work factors (e.g., flexible scheduling, ability to reduce hours, supportive workplace relationships; Barakovic Husic et al., 2020; Nahum-Shani & Bamberger, 2009) can reduce the likelihood of such exits. Such efforts are particularly important given the aging of the U.S. workforce (Barakovic Husic et al., 2020).

Study Limitations and Strengths

Findings should be interpreted considering study limitations and strengths. The HRS only asked about expectations regarding full-time employment, and therefore we could not examine the role of expectations for part-time employment preceding full retirement. Work expectations were assessed only on respondents younger than age 62 at time of interview, so our analytic sample was limited to those <62 years old at baseline. Early Baby Boomers were older and, thus, underrepresented in terms of person-years. The analytic sample had higher socioeconomic status and fewer health conditions than the HRS cohort generally, and thus our findings may not extend to older adults with lower socioeconomic status and more medical comorbidities. However, the younger age of the analytic sample also mitigated the potential for bias due to mortality. Personality characteristics (e.g., neuroticism) are likely related to the work expectation–depression relationship in complex ways. While not explored in this analysis, future research should investigate how these factors relate to how individuals anticipate and respond to work transitions in later life. Finally, while the focus of this analysis was to examine how work expectations relate to depressive symptoms, it is important to acknowledge that depression influences both expectations and participation in the labor force. Our modeling approach accounts for baseline depressive symptoms, we replicated our analysis across various indicators of mental health, and we conducted a stratified analysis conditioned on labor force status to assess how the influence of depression on work may influence our findings. However, given that the HRS does not have a measure of early-life depression, and most depression first onsets by age 30, we cannot fully explore how prior depressive episodes may have contributed to our findings. Despite the limitations, this study informs future research on this topic, including longitudinal investigations of bidirectional link between work expectations and mental health.

The study also has several strengths. First, to the best of our knowledge, this is the first longitudinal study that examined the relationship between work expectations and mental health outside the context of realizations and with attention to the role of uncertainty. Second, we focused specifically on Baby Boomers who currently have the highest rate of retirement (Fry, 2021), which has accelerated during the pandemic; indeed, in the United States the average expected likelihood of working past age 62 declined from 55.4% in March 2020 to 49.3% in November 2021 (Federal Reserve Bank of New York, 2022). Third, the study used a large, nationally representative sample with rich data on mental health including the CIDI, one of the most well-validated measures of depressive symptomatology.

Conclusion

Subjective expectations regarding the future, whether related to employment, health, or other outcomes, are shaped by both contextual and psychosocial factors. Additional research is needed to clarify the psychological and social mechanisms underlying the associations between expectations and mental health outcomes, including examining a broader array of outcomes such as anxiety and sleeping disorders. Such research can inform how to promote mental health of older adults, including during employment transitions in later life.

Supplementary Material

gnac110_suppl_Supplementary_Material
Click here for additional data file. (327.6KB, docx)

Contributor Information

Briana Mezuk, Center for Social Epidemiology and Population Health, Department of Epidemiology, University of Michigan School of Public Health, Ann Arbor, Michigan, USA; Research Center for Group Dynamics, Institute for Social Research, University of Michigan, Ann Arbor, Michigan, USA.

Linh Dang, Center for Social Epidemiology and Population Health, Department of Epidemiology, University of Michigan School of Public Health, Ann Arbor, Michigan, USA.

David Jurgens, School of Information, University of Michigan, Ann Arbor, Michigan, USA.

Jacqui Smith, Survey Research Center, Institute for Social Research, Ann Arbor, Michigan, USA.

Funding

This project was supported by a grant from the American Foundation for Suicide Prevention (DIG-1-110-19). The Health and Retirement Survey is sponsored by the National Institute on Aging (NIA U01AG009740) and the Social Security Administration.

Conflict of Interest

None declared.

Data Availability

The data used in this study are publicly available at https://hrs.isr.umich.edu/about. The analytic methods and statistical code used for this paper are available upon request to the author. This analysis was not preregistered.

References

  1. Aberg, K. C., Toren, I., & Paz, R. (2022). A neural and behavioral trade-off between value and uncertainty underlies exploratory decisions in normative anxiety. Molecular Psychiatry, 27(3), 1573–1587. doi: 10.1038/s41380-021-01363-z [DOI] [PubMed] [Google Scholar]
  2. Abrams, L. R., Clarke, P. J., & Mehta, N. K. (2022a). Unmet expectations about work at age 62 and depressive symptoms. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 77(3), 615–625. doi: 10.1093/geronb/gbab113 [DOI] [PubMed] [Google Scholar]
  3. Abrams, L. R., Finlay, J. M., & Kobayashi, L. C. (2022b). Job transitions and mental health outcomes among US adults aged 55 and older during the COVID-19 pandemic. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 77(7), e106–e116. doi: 10.1093/geronb/gbab060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Amick, B. C., McLeod, C. B., & Bültmann, U. (2016). Labor markets and health: An integrated life course perspective. Scandinavian Journal of Work, Environment & Health, 42(4), 346–353. doi: 10.5271/sjweh.3567 [DOI] [PubMed] [Google Scholar]
  5. Artazcoz, L., Benach, J., Borrell, C., & Cortès, I. (2004). Unemployment and mental health: Understanding the interactions among gender, family roles, and social class. American Journal of Public Health, 94(1), 82–88. doi: 10.2105/ajph.94.1.82 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Barakovic Husic, J., Melero, F. J., Barakovic, S., Lameski, P., Zdravevski, E., Maresova, P., Krejcar, O., Chorbev, I., Garcia, N. M., & Trajkovik, V. (2020). Aging at work: A review of recent trends and future directions. International Journal of Environmental Research and Public Health, 17(20), 7659. doi: 10.3390/ijerph17207659 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of pessimism: The Hopelessness Scale. Journal of Consulting and Clinical Psychology, 42(6), 861–865. doi: 10.1037/h0037562 [DOI] [PubMed] [Google Scholar]
  8. Berthung, E., Gutacker, N., Friborg, O., Abelsen, B., & Olsen, J. A. (2021). Who keeps on working? The importance of resilience for labour market participation. PLoS One, 16(10), e0258444. doi: 10.1371/journal.pone.0258444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bjärehed, J., Sarkohi, A., & Andersson, G. (2010). Less positive or more negative? Future-directed thinking in mild to moderate depression. Cognitive Behaviour Therapy, 39(1), 37–45. doi: 10.1080/16506070902966926 [DOI] [PubMed] [Google Scholar]
  10. Caine, E. D., Knox, K. L., & Conwell, Y. (2011). Public health and population approaches for suicide prevention. In Caine E. D., Knox K. L., & Conwell Y. (Eds.), Population mental health (pp. 303–338). Routledge. [Google Scholar]
  11. Chen, Y., Kim, E. S., Shields, A. E., & VanderWeele, T. J. (2020). Antecedents of purpose in life: Evidence from a lagged exposure-wide analysis. Cogent Psychology, 7(1), 1825043. doi: 10.1080/23311908.2020.1825043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Clark, K. B., & Summers, L. H. (1978). Labor force transitions and unemployment (Working Paper No. 277; Working Paper Series). National Bureau of Economic Research. doi: 10.3386/w0277 [DOI] [Google Scholar]
  13. Clarke, P., Marshall, V. W., & Weir, D. (2012). Unexpected retirement from full time work after age 62: Consequences for life satisfaction in older Americans. European Journal of Ageing, 9(3), 207–219. doi: 10.1007/s10433-012-0229-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dang, L., Dong, L., & Mezuk, B. (2020). Shades of blue and gray: A comparison of the Center for Epidemiologic Studies Depression Scale and the Composite International Diagnostic Interview for assessment of depression syndrome in later life. Gerontologist, 60(4), e242–e253. doi: 10.1093/geront/gnz044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dombrovski, A. Y., Hallquist, M. N., Brown, V. M., Wilson, J., & Szanto, K. (2019). Value-based choice, contingency learning and suicidal behavior in mid-life and late-life depression. Biological Psychiatry, 85(6), 506–516. doi: 10.1016/j.biopsych.2018.10.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dooley, D., Catalano, R., & Wilson, G. (1994). Depression and unemployment: Panel findings from the Epidemiologic Catchment Area study. American Journal of Community Psychology, 22(6), 745–765. doi: 10.1007/BF02521557 [DOI] [PubMed] [Google Scholar]
  17. Eaton, W. W., Hall, A. L. F., Macdonald, R., & McKibben, J. (2007). Case identification in psychiatric epidemiology: A review. International Review of Psychiatry (Abingdon, England), 19(5), 497–507. doi: 10.1080/09540260701564906 [DOI] [PubMed] [Google Scholar]
  18. Ekerdt, D. J. (2010). Frontiers of research on work and retirement. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 65(1), 69–80. doi: 10.1093/geronb/gbp109 [DOI] [PubMed] [Google Scholar]
  19. Federal Reserve Bank of New York. (2022). SCE labor market survey. https://www-newyorkfed-org./microeconomics/sce/labor#/expectations-retirement1
  20. Fleischmann, M., Xue, B., & Head, J. (2020). Mental health before and after retirement—Assessing the relevance of psychosocial working conditions: The Whitehall II Prospective Study of British Civil Servants. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 75(2), 403–413. doi: 10.1093/geronb/gbz042 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Franklin, J. C., Ribeiro, J. D., Fox, K. R., Bentley, K. H., Kleiman, E. M., Huang, X., Musacchio, K. M., Jaroszewski, A. C., Chang, B. P., & Nock, M. K. (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin, 143(2), 187–232. doi: 10.1037/bul0000084 [DOI] [PubMed] [Google Scholar]
  22. Frazier, P., Keenan, N., Anders, S., Perera, S., Shallcross, S., & Hintz, S. (2011). Perceived past, present, and future control and adjustment to stressful life events. Journal of Personality and Social Psychology, 100(4), 749–765. doi: 10.1037/a0022405 [DOI] [PubMed] [Google Scholar]
  23. Fry, R. (2021). The pace of Boomer retirements has accelerated in the past year. Pew Research Center. https://www.pewresearch.org/fact-tank/2020/11/09/the-pace-of-boomer-retirements-has-accelerated-in-the-past-year/ [Google Scholar]
  24. Furnham, A., & Ribchester, T. (1995). Tolerance of ambiguity: A review of the concept, its measurement and applications. Current Psychology, 14(3), 179–199. doi: 10.1007/bf02686907 [DOI] [Google Scholar]
  25. Gibb, B. E., Beevers, C. G., Andover, M. S., & Holleran, K. (2006). The hopelessness theory of depression: A prospective multi-wave test of the vulnerability–stress hypothesis. Cognitive Therapy and Research, 30(6), 763–772. doi: 10.1007/s10608-006-9082-1 [DOI] [Google Scholar]
  26. Giner, L., Blasco-Fontecilla, H., De La Vega, D., & Courtet, P. (2016). Cognitive, emotional, temperament, and personality trait correlates of suicidal behavior. Current Psychiatry Reports, 18(11), 102. doi: 10.1007/s11920-016-0742-x [DOI] [PubMed] [Google Scholar]
  27. Greco, V., & Roger, D. (2001). Coping with uncertainty: The construction and validation of a new measure. Personality and Individual Differences, 31(4), 519–534. doi: 10.1016/s0191-8869(00)00156-2 [DOI] [Google Scholar]
  28. Grupe, D. W., & Nitschke, J. B. (2013). Uncertainty and anticipation in anxiety: An integrated neurobiological and psychological perspective. Nature Reviews Neuroscience, 14(7), 488–501. doi: 10.1038/nrn3524 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hardin, J. W., & Hilbe, J. M. (2018). Generalized linear models and extensions (4th ed.). Stata Press. https://www.stata.com/bookstore/generalized-linear-models-and-extensions/ [Google Scholar]
  30. Harris, A. J. L., & Hahn, U. (2011). Unrealistic optimism about future life events: A cautionary note. Psychological Review, 118(1), 135–154. doi: 10.1037/a0020997 [DOI] [PubMed] [Google Scholar]
  31. Hedegaard, H., Curtin, S. C., & Warner, M. (2020). Increase in suicide mortality in the United States, 1999–2018. NCHS Data Brief, 362, 1–8. https://www.cdc.gov/nchs/products/databriefs/db362.htm [PubMed] [Google Scholar]
  32. Helzer, E. G., Connor-Smith, J. K., & Reed, M. A. (2009). Traits, states, and attentional gates: Temperament and threat relevance as predictors of attentional bias to social threat. Anxiety, Stress & Coping: An International Journal, 22(1), 57–76. doi: 10.1080/10615800802272244 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Hershfield, H. E. (2011). Future self-continuity: How conceptions of the future self transform intertemporal choice. Annals of the New York Academy of Sciences, 1235, 30–43. doi: 10.1111/j.1749-6632.2011.06201.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hong, J. H., Lachman, M. E., Charles, S. T., Chen, Y., Wilson, C. L., Nakamura, J. S., VanderWeele, T. J., & Kim, E. S. (2021). The positive influence of sense of control on physical, behavioral, and psychosocial health in older adults: An outcome-wide approach. Preventive Medicine, 149, 106612. doi: 10.1016/j.ypmed.2021.106612 [DOI] [PubMed] [Google Scholar]
  35. Jacobs, L., & Piyapromdee, S. (2016). Labor force transitions at older ages: Burnout, recovery, and reverse retirement.https://www.federalreserve.gov/econres/feds/labor-force-transitions-at-older-ages-burnout-recovery-and-reverse-retirement.htm
  36. Kessler, R. C., Andrews, G., Mroczek, D., Ustun, B., & Wittchen, H. U. (1998). The World Health Organization composite international diagnostic interview short‐form (CIDI‐SF). International Journal of Methods in Psychiatric Research, 7(4), 171–185. doi: 10.1002/mpr.47 [DOI] [Google Scholar]
  37. Kiani, R., Corthell, L., & Shadlen, M. N. (2014). Choice certainty is informed by both evidence and decision time. Neuron, 84(6), 1329–1342. doi: 10.1016/j.neuron.2014.12.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kim, E. S., Whillans, A. V., Lee, M. T., Chen, Y., & VanderWeele, T. J. (2020). Volunteering and subsequent health and well-being in older adults: An outcome-wide longitudinal approach. American Journal of Preventive Medicine, 59(2), 176–186. doi: 10.1016/j.amepre.2020.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kim, J. E., & Moen, P. (2002). Retirement transitions, gender, and psychological well-being: A life-course, ecological model. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 57(3), 212–222. doi: 10.1093/geronb/57.3.p212 [DOI] [PubMed] [Google Scholar]
  40. Kirton, M. J. (1981). A reanalysis of two scales of tolerance of ambiguity. Journal of Personality Assessment, 45(4), 407–414. doi: 10.1207/s15327752jpa4504_10 [DOI] [PubMed] [Google Scholar]
  41. Lahdenperä, M., Virtanen, M., Myllyntausta, S., Pentti, J., Vahtera, J., & Stenholm, S. (2022). Psychological distress during the retirement transition and the role of psychosocial working conditions and social living environment. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 77(1), 135–148. doi: 10.1093/geronb/gbab054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Lee, J., & Kim, M.-H. (2017). The effect of employment transitions on physical health among the elderly in South Korea: A longitudinal analysis of the Korean Retirement and Income Study. Social Science & Medicine (1982), 181, 122–130. doi: 10.1016/j.socscimed.2017.04.002 [DOI] [PubMed] [Google Scholar]
  43. Liu, R. T., Kleiman, E. M., Nestor, B. A., & Cheek, S. M. (2015). The hopelessness theory of depression: A quarter century in review. Clinical Psychology, 22(4), 345–365. doi: 10.1111/cpsp.12125 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. MacLeod, A. K., & O’Connor, R. C. (2018). Positive future‑thinking, well‑being, and mental health. In G. Oettingen, A. T. Sevincer, & P. Gollwitzer (Eds.), The psychology of thinking about the future (pp. 199–213). The Guilford Press. [Google Scholar]
  45. Mitchell, U. A., Gutierrez-Kapheim, M., Nguyen, A. W., & Al-Amin, N. (2020). Hopelessness among middle-aged and older Blacks: The negative impact of discrimination and protecting power of social and religious resources. Innovation in Aging, 4(5), igaa044. doi: 10.1093/geroni/igaa044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Mosca, I., & Barrett, A. (2016). The impact of voluntary and involuntary retirement on mental health: Evidence from older Irish adults. The Journal of Mental Health Policy and Economics, 19(1), 33–44. doi: 10.2139/ssrn.2543918 [DOI] [PubMed] [Google Scholar]
  47. Nahum-Shani, I., & Bamberger, P. A. (2009). Work hours, retirement and supportive relations among older adults. Journal of Organizational Behavior, 30(1), 1–25. doi: 10.1002/job.576 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Ng, S.-M., Leng, L., & Wang, Q. (2019). Active interest mentorship for soon-to-retire people: A self-sustaining retirement preparation program. Journal of Applied Gerontology, 38(3), 344–364. doi: 10.1177/0733464816687448 [DOI] [PubMed] [Google Scholar]
  49. Park, C. L., Finkelstein-Fox, L., Russell, B. S., Fendrich, M., Hutchison, M., & Becker, J. (2021). Psychological resilience early in the COVID-19 pandemic: Stressors, resources, and coping strategies in a national sample of Americans. The American Psychologist, 76(5), 715–728. doi: 10.1037/amp0000813 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Rabe-Hesketh, S., Skrondal, A., & Pickles, A. (2004). GLLAMM manual. U.C. Berkeley Division of Biostatistics Working Paper Series. https://biostats.bepress.com/ucbbiostat/paper160 [Google Scholar]
  51. Saulnier, K. G., Allan, N. P., Raines, A. M., & Schmidt, N. B. (2019). Depression and intolerance of uncertainty: Relations between uncertainty subfactors and depression dimensions. Psychiatry, 82(1), 72–79. doi: 10.1080/00332747.2018.1560583 [DOI] [PubMed] [Google Scholar]
  52. Sjöberg, O. (2021). Work-retirement transitions and mental health: A longitudinal analysis of the role of social protection generosity in 11 countries. Scandinavian Journal of Public Health. doi: 10.1177/14034948211042130 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Sonnega, A., Faul, J. D., Ofstedal, M. B., Langa, K. M., Phillips, J. W., & Weir, D. R. (2014). Cohort profile: The Health and Retirement Study (HRS). International Journal of Epidemiology, 43(2), 576–585. doi: 10.1093/ije/dyu067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Stara, V., Santini, S., Kropf, J., & D’Amen, B. (2020). Digital health coaching programs among older employees in transition to retirement: Systematic literature review. Journal of Medical Internet Research, 22(9), e17809. doi: 10.2196/17809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Tanovic, E., Gee, D. G., & Joormann, J. (2018). Intolerance of uncertainty: Neural and psychophysiological correlates of the perception of uncertainty as threatening. Clinical Psychology Review, 60, 87–99. doi: 10.1016/j.cpr.2018.01.001 [DOI] [PubMed] [Google Scholar]
  56. Thimm, J. C., Holte, A., Brennen, T., & Wang, C. E. A. (2013). Hope and expectancies for future events in depression. Frontiers in Psychology, 4, 470. doi: 10.3389/fpsyg.2013.00470 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. US Bureau of Labor Statistics. (2021). 67 percent of private industry workers had access to retirement plans in 2020: The Economics Daily. U.S. Bureau of Labor Statistics. https://www.bls.gov/opub/ted/2021/67-percent-of-private-industry-workers-had-access-to-retirement-plans-in-2020.htm [Google Scholar]
  58. U.S. Department of Health and Human Services, Office of the Surgeon General. (2021). The Surgeon General releases call to action to implement the national strategy for suicide prevention. https://www.hhs.gov/sites/default/files/sprc-call-to-action.pdf
  59. US Department of Labor. (2021). History of EBSA and ERISA. U.S. Department of Labor. https://www.dol.gov/agencies/ebsa/about-ebsa/about-us/history-of-ebsa-and-erisa [Google Scholar]
  60. Vrkljan, B., Montpetit, A., Naglie, G., Rapoport, M., & Mazer, B. (2019). Interventions that support major life transitions in older adulthood: A systematic review. International Psychogeriatrics, 31(3), 393–415. doi: 10.1017/S1041610218000972 [DOI] [PubMed] [Google Scholar]
  61. Zinn, J. O. (2004). Health, risk and uncertainty in the life course: A typology of biographical certainty constructions. Social Theory & Health, 2(3), 199–221. doi: 10.1057/palgrave.sth.8700033 [DOI] [Google Scholar]

Associated Data

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

Supplementary Materials

gnac110_suppl_Supplementary_Material
Click here for additional data file. (327.6KB, docx)

Data Availability Statement

The data used in this study are publicly available at https://hrs.isr.umich.edu/about. The analytic methods and statistical code used for this paper are available upon request to the author. This analysis was not preregistered.

Articles from The Gerontologist are provided here courtesy of Oxford University Press

RESOURCES