A 2-Year Longitudinal Relationship Between Work–Family Conflict and Health Among Older Workers: Can Gardening Help?

Abstract

With the graying workforce worldwide, identifying factors that facilitate older workers’ health is critically important. We examined whether gardening mitigates the relationship of work–family conflict with disability, chronic conditions, depressive symptoms, and self-rated health among older workers. We drew a subsample of older workers aged 55 years and above from the Health and Retirement Study (N = 1,598). Our results indicate that the relationships of work-to-family conflict at baseline with disability and with poorer self-rated health at a 2-year follow-up were stronger for those who gardened less than those who gardened more. No significant interaction was found between family-to-work conflict and gardening in predicting the health outcomes. This study is the first to show that gardening may have a protective effect against the adverse impact of work-to-family conflict on older workers’ health.

Keywords

senior workers work–family interference disability chronic conditions depression self-rated health leisure activities

With the rising number of older workers in the workforce (Phillips & Siu, 2012), research efforts to identify factors that shape older workers’ health have continued. One such factor attracting growing attention is work–family conflict (WFC), a form of inter-role conflict occurring when work and family demands are mutually incompatible (Greenhaus & Beutell, 1985). WFC is associated with a host of psychological as well as physical health problems (Amstad et al., 2011). Previous research demonstrated that work interference with family is a significant predictor of comorbidity and disability among older workers (Cho & Chen, 2018). This stream of research emphasizes the need for interventions to alleviate the harmful effect of WFC.

This study examines the moderating role of gardening in the relationship between WFC and health among older workers. Gardening, which includes gardening and yard work activities, is one of the most popular leisure activities among older adults (Ashe et al., 2009). Based on the well-documented benefits of gardening (Roberson & Kudlacek, 2015; Wright & Wadsworth, 2014), we propose that engagement in gardening during leisure time has the potential to mitigate the link between WFC and health among older workers. To address this research question, we utilized data from the Health and Retirement Study (HRS), a longitudinal study with a representative U.S. sample. Older workers were defined as those aged 55 years and above (Centers for Disease Control and Prevention National Center for Chronic Disease Prevention and Health Promotion, 2012).

WFC and Health

WFC is a type of inter-role conflict experienced when participation in one role is made more difficult because of participation in another role (Greenhaus & Beutell, 1985). WFC occurs in two directions; work-to-family conflict (WTFC) occurs when work interferes with family (e.g., long work hours makes it difficult for a worker to attend family gatherings), whereas family-to-work conflict (FTWC) occurs when family interferes with work (e.g., fatigue from caregiving makes it difficult for a worker to concentrate on job tasks). A growing body of research suggests that WFC is a salient issue for employees across all ages (Allen & Shockley, 2012).

According to the conservation of resources (COR) model (Hobfoll, 2002), WFC is a stressor, an actual or anticipated loss of resources, which has adverse impact on health. Specifically, WFC represents the state in which personal resources are depleted in the process of managing work and family demands, and meta-analytic evidence shows that WFC, in both directions, is associated with psychological and physical health outcomes (Amstad et al., 2011). Particularly among older workers, WTFC is a significant predictor of comorbidity and disability over time, even after controlling for other known predictors of health (Cho & Chen, 2018).

In this study, we examine older workers’ functional (i.e., disability), physical (i.e., chronic conditions), mental (depressive symptoms), and self-rated health. These are important indicators of older workers’ health that merit further research because they are related to a heightened demands on health care costs (e.g., He & Larsen, 2014) and a greater mortality rate among older adults (e.g., Chamberlain et al., 2014). These conditions are related to retirement transitions among older workers (e.g., Nunes et al., 2016; Woo & Zajacova, 2016).

Gardening as a Moderator

Drawing on a resource perspective on the recovery from job stress (Hobfoll, 2002; Meijman & Mulder, 1998), we propose that gardening during leisure time may buffer the negative health effects of WFC by helping to replenish the personal resources that were lost due to the experience of WFC. That is, the WFC–health relationship may be less salient among older workers who spend more time on gardening because of recovery occurring during the gardening activities.

Recovery refers to the process by which the physiological and psychological systems return to pre-stressor levels (Meijman & Mulder, 1998). Exposure to stressors depletes resources and results in negative outcomes such as fatigue and negative affect. Of importance, proper recovery replenishes resources and negates the short-term reactions, thereby reducing the long-term impact of stressors on one’s health. In contrast, prolonged exposure to stressors without sufficient recovery may lead to chronic health problems and manifest diseases.

Previous research indicates that various leisure activities (e.g., social activities, physical activities, volunteering activities, and low-effort activities) can promote recovery (Sonnentag et al., 2017). It has been further shown that it is the underlying psychological experiences resulting from the leisure activities, not a specific activity per se, that facilitate recovery (i.e., recovery experiences; Sonnentag & Fritz, 2007). These recovery experiences include, but are not limited to, psychological detachment from work, relaxation, and mastery, and have been found to aid resource gain (e.g., energy, positive affect; Sonnentag & Fritz, 2007). In sum, a leisure activity that elicits recovery experiences has the potential to enhance recovery and mitigate the stressor-health link.

Ample research has documented that gardening helps accrue valuable resources, such as novel perspectives, the sense of fulfillment, supportive interpersonal relationships, and health. First, when gardening occurs outside in nature, it offers opportunities for introspection and reflection (Vaillant, 2008). Older workers may gain a new perspective toward life and reappraise stressors during gardening (Rodiek, 2002). Second, it has been reported that older adults engage in gardening to be creative and original (Ashton-Shaeffer & Constant, 2006), and that gardening allows them to learn new things (Wang & Glicksman, 2013). This suggests that gardening may engender the sense of fulfillment and satisfy the psychological need for mastery, both of which are critical resources that can facilitate recovery. Gardening also provides opportunities for interpersonal interaction and building friendship (Ashton-Shaeffer & Constant, 2006; Wang & Glicksman, 2013), and the social support that may come from this could also help older workers to effectively cope with stress. Finally, as a physical activity of moderate intensity (Park et al., 2008), gardening can promote and maintain health among older adults if engaged in regularly (Nelson et al., 2007). For instance, empirical evidence indicates that gardening has the potential to improve muscle mass, endurance, and dexterity (Park et al., 2016); enhance grip strength (Park et al., 2008); and prevent falls among older adults (Chen & Janke, 2012). Thus, the health benefits gained from gardening may cancel out negative effects of WFC on physical and psychological health.

In sum, we propose that gardening moderates the relationship between WFC and health observed among older workers. Specifically, we expect that the WFC–health relationship is weaker among those who spend more time in gardening, compared with those who spend less time in gardening:

Hypothesis 1: Gardening moderates the positive relationship of WTFC with (a) disability, (b) chronic conditions, (c) depressive symptoms, and (d) poorer self-rated health, such that each relationship is weaker for those who spend more time in gardening than it is for those who spend less time in gardening.

Hypothesis 2: Gardening moderates the positive relationship of FTWC with (a) disability, (b) chronic conditions, (c) depressive symptoms, and (d) poorer self-rated health, such that each relationship is weaker for those who spend more time in gardening than it is for those who spend less time in gardening.

Method

Data and Study Design

Data were drawn from the HRS (2006–2016) and its supplementary surveys, the Consumption and Activities Mail Survey (CAMS; 2007–2015), and the Psychosocial and Lifestyle Questionnaire (PLQ; 2006–2016). The HRS aims to understand aging-related changes in health and economic circumstances, the CAMS collects information on consumption and time use, and the PLQ covers six broad areas (i.e., subjective well-being, lifestyle and stress, quality of social ties, personality, work-related beliefs, and self-related beliefs). The HRS and the PLQ have been fielded in every even year, and the CAMS in every odd year. More information about these surveys can be found elsewhere (Hurd & Rohwedder, 2009; Smith et al., 2013; Sonnega et al., 2014). We used information on gardening from the CAMS and WFC from the PLQ. The rest of the variables came from the core HRS data set.

We examined each 2-year interval by pooling all data together such that participants’ information in 2006, 2008, 2010, 2012, and 2014 served as baseline and 2008, 2010, 2012, 2014, and 2016 served as the respective follow-up (Chen et al., 2017; Himes & Reynolds, 2012). Information on gardening (collected in 2007, 2009, 2011, 2013, and 2015) served as the moderator in each respective interval. This strategy was adopted to maximize the usage of existing data because only a half of random subset of participants from the HRS were asked to complete the CAMS and the PLQ for a given wave (i.e., missing at random; for more details see Smith et al., 2013).

Figure 1 presents the study design and exclusion flow. There were 213,665 observations after pooling the five 2-year intervals. Among these, 12,679 observations remained after excluding those who were not selected to answer the CAMS or the PLQ and who did not return these surveys. We further excluded individuals who were younger than 55 years old at baseline, not living in the community, and not working for pay. Those who had missing information on covariates and key variables and proxy responses were also excluded. The final sample included 1,598 observations. Compared with those who worked for pay but were excluded due to missing responses (n = 1,377), the current sample had a significantly lower percentage of females, 51.3% versus 58.2%, χ²(1, n = 2,975) = 14.27, p < .001. No significant differences were found in age, education, and race.

Figure 1.

Study design and exclusion flow diagram.

Measures

Work–family conflict

The HRS asked participants to rate WFC (in both directions) that occurred during last month on a 4-point scale (1 = rarely, 4 = most of the time) (Smith et al., 2013). Each direction was assessed by three items. An example item for WTFC is “My work schedule makes it difficult to fulfill personal responsibilities,” and one item evaluating FTWC is “My home life kept me from getting work done on time on my job.” For analyses, an average score was calculated for WTFC and FTWC, with a higher score indicating a greater conflict.

Gardening

Participants were asked how many hours they had spent on gardening or yard work last week. The reported hours (ranging from 0 to 30 hr) were used in the analysis.

Disability

Participants were asked whether they had difficulty (1 = yes, 0 = no) with activities of daily living (i.e., dressing, walking, bathing, eating, getting in/out of bed, toileting) and instrumental activities of daily living (i.e., using maps, preparing meals, grocery shopping, using phone, taking medications, and managing money). Following previous research (Spector & Fleishman, 1998), a composite score (0–12) was calculated, with a higher score representing a greater level of disability. Previous research has shown good agreement between disability based on self-report and performance-based disability measures (Nielsen et al., 2016).

Chronic conditions

Participants were asked whether doctors had told them that they had any of the following conditions: high blood pressure, diabetes, cancer (excluding skin cancer), lung disease, heart condition, stroke, and arthritis. A total score (0–7) was created for each individual, with a higher score indicating a greater number of chronic conditions. Previous research has shown good agreement between self-reported chronic conditions and objective medical records (Wolf et al., 2005).

Depressive symptoms

Depressive symptoms were assessed by the eight-item Center for Epidemiological Studies Depression scale (Karim et al., 2015). Participants were asked to rate how they felt or behaved in the past week for the following items (1 = yes, 0 = no): feeling depressed, happy (reversed coding), sad, lonely, restlessness sleep, enjoying life (reversed coding), everything was an effort, and could not get going. A total score (0–8) was created for each individual, with a higher score indicating more depressive symptoms.

Self-rated health

Participants were asked to rate their own health on a 5-point scale (1 = excellent, 2 = very good, 3 = good, 4 = fair, and 5 = poor). A higher score indicates poorer self-rated health. Self-rated health has been shown to associate with health (Xu et al., 2019).

Control variables

Sociodemographic characteristics (Schuch et al., 2016; Woo & Zajacova, 2016), total household income and wealth (Headey & Wooden, 2004), employment status (Rosenthal et al., 2012), and moderate-intensity physical activities (Schuch et al., 2016) were included as covariates given their relevance to health outcomes. Considering these covariates is important because previous research (Meesters, 2009) suggests that contextual and sociodemographic characteristics such as neighborhood type (i.e., rural/suburban vs. urban) are associated with engagement in gardening and might relate to the opportunity and ability for individuals to engage in these activities.

For sociodemographic characteristics, age and education were recorded in years. Sex (1 = female, 0 = male), race (1 = White, 0 = non-White), and marital status (1 = married, 0 = not married) were dummy-coded. Income and wealth was a sum of earnings reported from all members living in the same household based on the money they received from the following: employment (including self-employment), unemployment and workers compensation, trades, stocks, bonds, certified deposits, banks accounts, tips, bonus, commissions, social security and any supplementary security income, food stamps and welfare, veteran benefits, pension and annuity, and rental income. The sum of income and wealth was categorized into eight groups: 1 = US$0–US$10,000, 2 = US$10,001–US$20,000, 3 = US$20,001–US$35,000, 4 = US$35,001–US$50,000, 5 = US$50,001–US$75,000, 6 = US$75,001–US$100,000, 7 = US$100,001–US$150,000, and 8 = above US$150,000. Employment status was dummy-coded (1 = full-time, 0 = part-time). For moderate-intensity physical activities, respondents were asked to rate how often they took part in sports or activities that are moderately energetic on a 5-point scale (i.e., hardly ever or never, 1–3 times a month, once a week, more than once a week, and every day).

Analytic Approach

Descriptive analyses were used to summarize sample characteristics at baseline. Ordinary least squares regression analysis was conducted for hypothesis testing. First, the effects of WTFC, FTWC, and gardening at baseline on disability, chronic conditions, depressive symptoms, and self-rated health at follow-up were examined separately while adjusting for covariates (Model 1). Then, the interaction term between WTFC and gardening (Model 2) and the interaction term between FTWC and gardening (Model 3) were entered to the model. Simple slope analysis was performed for significant interaction effects following the method described by Dawson (2014). All regression analyses were performed using SPSS version 24 (IBM, 2016).

Results

Table 1 shows the results from the descriptive analysis at baseline. On average, respondents were 63.56 years old and had 13.63 years of education. The sample was mostly female (51.30%) and Whites (79.60%). The average time spent on gardening in the past week was 1.95 hr. Supplementary Table 1 shows the univariate associations of baseline variables with the health outcomes at follow-up. Overall, all relationships were in the expected directions.

Table 1.

Descriptive Statistics of Study Variables at Baseline.

Variables	Overall (N = 1,598)
Variables	M or %
Work-to-family conflict (1–4)	1.52 (0.57)
Family-to-work conflict (1–4)	1.16 (0.30)
Gardening (hours)	1.95 (3.04)
Disability (0–12)	0.21 (0.68)
Chronic conditions (0–7)	1.51 (1.16)
Depressive symptoms (0–8)	1.07 (1.71)
Self-rated health
Excellent	16.30%
Very good	39.40%
Good	29.90%
Fair	12.70%
Poor	1.70%
Moderate-intensity physical activities
Hardly ever or never	11.70%
1–3 times a month	12.20%
Once a week	18.20%
More than once a week	48.70%
Every day	9.20%
Age (years)	63.56 (6.76)
Sex: Female	51.30%
Race: White	79.60%
Education (years)	13.63 (2.63)
Married (yes)	53.60%
Income and wealth
US$0–US$10,000	3.90%
US$10,001–US$20,000	5.30%
US$20,001–US$35,000	9.60%
US$35,001–US$50,000	9.30%
US$50,001–US$75,000	13.00%
US$75,001–US$100,000	9.10%
US$100,001–US$150,000	13.40%
US$150,000 and above	36.40%
Employment status
Full-time	55.60%
Part-time	44.40%

Table 2 shows the results from the ordinary least squares regression analysis examining the effects of baseline WFC and gardening on disability at follow-up. WTFC at baseline was a significant predictor of disability at follow-up, such that one unit increase in WTFC is associated with a .107 increment in disability at 2-year follow-up (Model 1). In support of Hypothesis 1a, gardening was a significant moderator (Model 2). Specifically, the effect of WTFC at baseline on disability at follow-up was weaker among individuals who spent more time on gardening (simple slope analysis: B = .510, p < .001) than those who did not garden as much (simple slope analysis: B = 1.56, p < .001; Figure 2A). Hypothesis 2a was not supported, because the interaction between FTWC at baseline and gardening was not significant in predicting disability (Model 3).

Table 2.

Adjusted Effects of Work–Family Conflict at Baseline and Gardening on Disability at Follow-Up.

Variables	Model 1			Model 2			Model 3
Variables	B	SE B	β	B	SE B	β	B	SE B	β
Work-to-family conflict (1–4)	.107	.035	.069*	1.56	.040	.101***	.107	.035	.069**
Family-to-work conflict (1–4)	.104	.062	.036	.104	.062	.036	.120	.071	.041
Gardening (hours)	.004	.006	.015	.053	.019	.184**	.014	.022	.049
Self-rated health at baseline (1–5)	.036	.022	.040	.036	.022	.040	.036	.022	.040
Disability at baseline (0–12)	.651	.028	.509***	.649	.028	.507***	.652	.028	.509***
Chronic conditions at baseline (0–7)	.043	.018	.057*	.046	.018	.061**	.043	.018	.057*
Depressive symptoms at baseline (0–8)	.022	.012	.043	.023	.012	.044	.022	.012	.043
Moderate-intensity physical activities (1–5)	.023	.016	.031	.025	.016	.033	.024	.016	.031
Age (years)	.003	.003	.027	.003	.003	.026	.003	.003	.027
Sex: Female	.047	.039	.027	.049	.039	.028	.047	.039	.027
Race: White	−.066	.046	−.030	−.067	.046	−.031	−.067	.046	−.031
Education (years)	−.016	.007	−.048*	−.016	.007	−.050*	−.016	.007	−.048*
Married (yes)	−.043	.040	−.024	−.041	.040	−.023	−.043	.040	−.024
Income and wealth (1–8)	−.021	.010	−.052*	−.021	.010	−.052*	−.021	.010	−.052*
Employment status: Full-time	−.052	.039	−.029	−.047	.039	−.027	−.052	.039	−.030
Work-To-Family Conflict (1–4) × Gardening (hours)	—	—	—	−.035	.013	−.178**	—	—	—
Family-To-Work Conflict (1–4) × Gardening (hours)	—	—	—	—	—	—	−.008	.018	−.035
R ²	.377			.380			.377
F	63.92***			60.60***			59.91***

p < .05. **p < .01. ***p < .001.

Figure 2.

Interaction effects between work-to-family conflict and gardening at baseline on disability (2A) and poorer self-rated health (2B) at follow-up.

Table 3 shows the results from the ordinary least squares regression analysis examining the effects of baseline WFC and gardening on chronic conditions at follow-up. After adjusting for covariates, WTFC, FTWC, and gardening were not significant predictors of chronic conditions at follow-up (Model 1). Hypothesis 1b was not supported because the effects of baseline WTFC on chronic conditions at follow-up did not differ by gardening (Model 2). Similarly, Hypothesis 2b was rejected as the interaction between baseline FTWC and gardening was not significant in predicting chronic conditions at follow-up (Model 3).

Table 3.

Adjusted Effects of Work–Family Conflict at Baseline and Gardening on Chronic Conditions at Follow-Up.

Variables	Model 1			Model 2			Model 3
Variables	B	SE B	β	B	SE B	β	B	SE B	β
Work-to-family conflict (1–4)	.037	.028	.017	.055	.031	.026	.036	.028	.017
Family-to-work conflict (1–4)	−.065	.049	−.016	−.065	.049	−.016	−.048	.056	−.012
Gardening (hours)	−.006	.005	−.015	.012	.015	.030	.004	.017	.010
Self-rated health at baseline (1–5)	.050	.017	.040**	.050	.017	.040**	.050	.017	.040**
Disability at baseline (0–12)	−.013	.022	−.007	−.014	.022	−.008	−.012	.022	−.007
Chronic conditions at baseline (0–7)	.904	.014	.864***	.905	.014	.865***	.905	.014	.864***
Depressive symptoms at baseline (0–8)	.008	.009	.011	.008	.009	.012	.008	.009	.011
Moderate-intensity physical activities (1–5)	−.007	.012	−.007	−.007	.012	−.006	−.007	.012	−.007
Age (years)	.005	.002	.026*	.005	.002	.026*	.005	.002	.026*
Sex: Female	−.004	.031	−.001	−.003	.031	−.001	−.004	.031	−.002
Race: White	−.005	.036	−.002	−.005	.036	−.002	−.006	.036	−.002
Education (years)	−.004	.006	−.010	−.005	.006	−.010	−.004	.006	−.009
Married (yes)	.056	.031	.023	.056	.031	.023	.056	.031	.023
Income and wealth (1–8)	−.007	.008	−.013	−.007	.008	−.013	−.007	.008	−.013
Employment status: Full-time	−.023	.030	−.009	−.021	.031	−.009	−.023	.031	−.009
Work-To-Family Conflict (1–4) × Gardening (hours)	—	—	—	−.013	.010	−.048	—	—	—
Family-To-Work Conflict (1–4) × Gardening (hours)	—	—	—	—	—	—	−.009	.014	−.026
R ²	.799			.799			.799
F	419.08***			393.127***			392.753***

p < .05. **p < .01. ***p < .001.

Table 4 shows the results from the ordinary least squares regression analysis examining the effects of baseline WFC and gardening on depressive symptoms at follow-up. WTFC at baseline was a significant predictor of depressive symptoms at follow-up, such that one unit increase in WTFC is associated with a .278 increment in depressive symptoms at follow-up (Model 1). Hypothesis 1c was rejected because the effects of WTFC on depressive symptoms at follow-up did not differ by gardening (Model 2). Hypothesis 2c was also rejected as the interaction between baseline FTWC and gardening was not significant in predicting depressive symptoms (Model 3).

Table 4.

Adjusted Effects of Work–Family Conflict at Baseline and Gardening on Depressive Symptoms at Follow-Up.

Variables	Model 1			Model 2			Model 3
Variables	B	SE B	β	B	SE B	β	B	SE B	β
Work-to-family conflict (1–4)	.278	.071	.094***	.315	.080	.106***	.275	.071	.093***
Family-to-work conflict (1–4)	.195	.125	.035	.195	.125	.035	.287	.143	.052*
Gardening (hours)	−.010	.012	−.019	.025	.039	.046	.046	.044	.083
Self-rated health at baseline (1–5)	.152	.044	.087**	.151	.044	.087**	.152	.044	.087**
Disability at baseline (0–12)	.172	.057	.070**	.171	.057	.070**	.177	.057	.072**
Chronic conditions at baseline (0–7)	.019	.036	.013	.021	.036	.014	.020	.036	.014
Depressive symptoms at baseline (0–8)	.378	.024	.386***	.378	.024	.387***	.377	.024	.385***
Moderate-intensity physical activities (1–5)	−.054	.032	−.038	−.053	.032	−.037	−.053	.032	−.037
Age (years)	−.011	.006	−.044	−.011	.006	−.044	−.011	.006	−.045
Sex: Female	.066	.080	.020	.068	.080	.020	.066	.080	.020
Race: White	−.044	.093	−.011	−.045	.093	−.011	−.051	.093	−.012
Education (years)	−.023	.015	−.037	−.024	.015	−.037	−.023	.015	−.036
Married (yes)	.010	.080	.003	.011	.080	.003	.010	.080	.003
Income and wealth (1–8)	−.041	.020	−.053*	−.040	.020	−.053*	−.041	.020	−.054*
Employment status: Full-time	−.107	.078	−.032	−.103	.078	−.031	−.109	.078	−.032
Work-To-Family Conflict (1–4) × Gardening (hours)	—	—	—	−.025	.026	−.068	—	—	—
Family-To-Work Conflict (1–4) × Gardening (hours)	—	—	—	—	—	—	−.048	.037	−.107
R ²	.303			.303			.304
F	45.84***			43.03***			43.11***

p < .05. **p < .01. ***p < .001.

Table 5 shows the results from the ordinary least squares regression analysis examining the effects of baseline WFC and gardening on self-rated health at follow-up. FTWC at baseline was a significant predictor of self-rated health at follow-up, such that one unit increase in FTWC is associated with a .161 increment in poorer self-rated health at follow-up (Model 1). In support of Hypothesis 1d, gardening was a significant moderator (Model 2). Specifically, the effect of WTFC at baseline on poorer self-rated health at follow-up was weaker among individuals who spent more time on gardening (simple slope analysis: B = .06, p = .18) than those who spent less time on gardening (simple slope analysis: B = .09, p = .04) (Figure 2B). Hypothesis 2d was not supported, as no significant interaction effect was found between baseline FTWC and gardening on self-rated health at follow-up (Model 3).

Table 5.

Adjusted Effects of Work–Family Conflict at Baseline and Gardening on Self-rated Health at Follow-Up.

Variables	Model 1			Model 2			Model 3
Variables	B	SE B	β	B	SE B	β	B	SE B	β
Work-to-family conflict (1–4)	.046	.035	.027	.092	.040	.054*	.044	.035	.026
Family-to-work conflict (1–4)	.161	.062	.050*	.161	.062	.050*	.214	.071	.067**
Gardening (hours)	.003	.006	.009	.048	.019	.152*	.035	.022	.110
Self-rated health at baseline (1–5)	.529	.022	.529***	.529	.022	.529***	.529	.022	.529***
Disability at baseline (0–12)	.031	.028	.022	.029	.028	.021	.033	.028	.024
Chronic conditions at baseline (0–7)	.126	.018	.151***	.128	.018	.154***	.126	.018	.152***
Depressive symptoms at baseline (0–8)	.029	.012	.051*	.029	.012	.052*	.028	.012	.050*
Moderate-intensity physical activities (1–5)	−.012	.016	−.014	−.010	.016	−.013	−.011	.016	−.014
Age (years)	−.004	.003	−.030	−.004	.003	−.030	−.004	.003	−.031
Sex: Female	−.104	.040	−.054**	−.102	.040	−.053*	−.105	.040	−.054
Race: White	−.045	.046	−.019	−.047	.046	−.020	−.049	.046	−.021
Education (years)	−.025	.007	−.067**	−.025	.007	−.068**	−.024	.007	−.067**
Married (yes)	−.025	.040	−.013	−.023	.040	−.012	−.025	.040	−.013
Income and wealth (1–8)	−.024	.010	−.055*	−.024	.010	−.055*	−.025	.010	−.056*
Employment status: Full-time	−.022	.039	−.012	−.018	.039	−.009	−.024	.039	−.012
Work-To-Family Conflict (1–4) × Gardening (hours)	—	—	—	−.032	.013	−.150*	—	—	—
Family-To-Work Conflict (1–4) × Gardening (hours)	—	—	—	—	—	—	−.027	.018	−.105
R ²	.479			.481			.479
F	96.87***			91.48***			91.03***

p < .05. **p < .01. ***p < .001.

Discussion

This study examined the moderating role of gardening—which includes gardening and yard work activities—in the relationship between WFC and health among older workers. We found a significant moderating effect of gardening in the relationships of WTFC with disability and with self-rated health. The observed interaction pattern was consistent for both outcomes; the positive relationship of WTFC with increased disability and with poorer self-rated health was stronger for those who gardened less than those who gardened more. Of importance, our analyses controlled for moderate-intensity physical activities, which suggests that gardening delivers unique health benefits for older workers. In sum, our study targeting senior workers replicates benefits of gardening reported in previous research, while indicating that the benefits of gardening may vary depending on health outcomes of interest.

We did not find a moderating effect of gardening for the relationship between FTWC and health. Previous research has demonstrated that FTWC tends to have a relatively weaker influence on health than does WTFC (Cho & Chen, 2018). Also, recent research suggests that individuals with a higher level of FTWC are more likely to exit work (Xue et al., 2020). Thus, the negligible effect of FTWC on health among the older workers remaining in the workforce may explain why this link did not differ as a function of the involvement in gardening. Alternatively, gardening, as a home-based leisure activity, may have exerted a limited buffering effect against FTWC because FTWC is experienced at work when family demands interfere with work.

Gardening did not buffer the negative effect of WFC on chronic symptoms and depressive symptoms. Although nonsignificant, further inspection of the data indicated that the pattern of interaction between WFC—in both directions—and gardening was in line with our expectation such that the relationships between WFC and these health outcomes were weaker for those who garden more than those who garden less. We presume chronic conditions inevitably increase with aging and the experience of WFC and gardening exerted only minimal effects on chronic conditions. In line with this notion, only age and baseline health outcomes (chronic conditions, self-rated health) were significant predictors of chronic conditions. Concerning depressive symptoms, previous research has shown that emotion and mood regulation improve with age (Carstensen & Mikels, 2005). Studies have also found that older workers are better at regulating their stress and negative affect compared with younger workers (Hertel et al., 2015). Likely, gardening did not exert moderating effects on the relationship between WFC and depressive symptoms among older workers who actively cope with their stress. Further research is needed to understand the weak moderating effect of gardening in the link from WFC to chronic conditions and depressive symptoms.

Theoretical and Practical Implications

Our research offers several important theoretical contributions. First, the current findings align with previous studies reporting the benefits of gardening for older adults’ physical and psychological health (Chen & Janke, 2012; Park et al., 2008; Wang & MacMillan, 2013; Wright & Wadsworth, 2014) and advance the literature on older workers’ health by identifying a boundary condition that has the potential to mitigate the adverse impact of WTFC. By focusing on older workers, we expand upon the gardening literature, as the majority of previous research in this area has been conducted with older retirees (e.g., Infantino, 2004).

Second, our study expands the literature on recovery from job stress by considering older workers. Senior workers represent a growing population that likely has a unique need for recovery and differential leisure time experiences due to their health and family status (Allen & Shockley, 2012; He & Larsen, 2014). Despite the voluminous body of research on recovery from work, no research has focused on older workers. Also, this study examined a novel activity from the recovery perspective, thereby reiterating the argument that it is the psychological attributes of the leisure activities, not a specific activity per se, that facilitate recovery (Sonnentag & Fritz, 2007).

Third, our research broadens the work–family literature by identifying a factor that modifies the health impact of WFC. In so doing, this study addresses the call for research that investigates moderators of the link between WFC and health among older workers (Cho & Chen, 2018). This study complements previous work–family research on individual (e.g., dispositional tendency; Cho et al., 2013) and organizational predictors of WFC (e.g., family-supportive supervisor; Kossek et al., 2011), which offered the knowledge about how to reduce WFC to avoid poor health outcomes, by demonstrating how leisure activities such as gardening have the potential to assist older workers in coping with WFC.

In terms of practical implications, our study provides preliminary evidence that gardening can be recommended to the growing number of older workers as a means to stay healthy, especially when facing the stressor of work interference with family. Also, older workers who are currently gardening could further modify their activities to maximize potential resources obtained during gardening activity. For instance, older workers may invite their friends and colleagues to engage in gardening, thus it might generate the opportunity to connect with others and receive social support. Furthermore, organizations may want to create a green environment at work where employees could have opportunities for gardening as a proactive and creative means of workplace health promotion. Of importance, organizations should consider ways to make such initiative as accessible and supportive as possible for older workers who may have barriers for participation (e.g., disabilities, mobility limitations; Evans et al., 2008). While it is unclear if the benefits of engaging in gardening meet or exceed those of other forms of moderate-intensity physical activity, research clearly indicates that gardening is good for individuals’ health and well-being (Soga et al., 2017). Thus, promoting gardening in the workspace may be a worthwhile investment, given other benefits of greenspace in this environment (e.g., workplace attitude; Lottrup et al., 2013). Supervisors should support older workers to achieve appropriate recovery and encourage them to engage in healthy leisure activities such as gardening.

Limitations and Future Research Directions

Limitations of this study should be acknowledged. First, the data were collected via self-report only, which may raise a concern for the validity, particularly for the health outcomes. Although all the measures used in the current research have been shown to be valid, future research may benefit from triangulation of self-reported measures with other data sources, such as medical record data. Second, we utilized pooled sample because the half of randomly chosen participants of the HRS were not invited to complete the PLQ and the CAMS in each wave (Hurd & Rohwedder, 2009; Smith et al., 2013; Sonnega et al., 2014). However, this approach prohibited us from conducting longitudinal analyses that examine changes over time. Future research should investigate the role of gardening in the stressor-health link in both macro and micro longitudinal studies. Third, details about gardening (e.g., occurs inside or out in nature, specific activities engaged, seasonal variation) were not available in the HRS. Therefore, we cannot differentiate yard work from gardening. This could have affected our findings because some people may not find yard work as enjoyable and leisurely as gardening. Moreover, as gardening and yard work activities have been found to vary seasonally (Mobily et al., 1995), future research may want to consider “the time of year” in the study design and analyses. Last, some covariates may affect older workers’ ability and opportunity to engage in gardening, but this was not explored further in this study. For instance, research has suggested that age and health might influence the amount of time spent in gardening (Scott et al., 2020). Further research should explore whether and how individual characteristics affect the engagement in and benefits from gardening.

Building on this study that focused on the time spent for gardening, future research may consider gardeners’ subjective experience during gardening. Because having to take care of a garden could be perceived as a burden, particularly for the very old (Fänge & Ivanoff, 2009), psychological experiences during the gardening activity may offer unique insights into the benefits received from gardening. Future research should also explore the role of gardening in facilitating well-being for the seniors during work transitions (e.g., retirement, re-employment). While our study adds to the literature on benefits of gardening that has rarely considered older workers, respondents who made work transitions during the study period as well as those who already retired were excluded from the analyses. Considering the significance of work transitions for the senior’ health and well-being (Harden et al., 2019), we call for future research that examines the benefits of gardening during work transitions in later life.

Conclusion

Considering the increasing number of older workers and the negative impact of WTFC on older workers’ health, it is imperative to understand ways to alleviate the harmful effect of WTFC. Using data from the Health and Retirement Study (HRS), we found that gardening moderates the relationship between WTFC and health—specifically disability and self-rated health—among older workers. WTFC was associated with more health problems and poorer perceived health among older workers who spend less time in gardening, whereas these relationships were less salient among those who spend more time in gardening. In sum, our findings suggest that gardening has potential as a leisure-based coping method for older workers who experience WTFC.

Supplemental Material

Supplementary_Table_1 – Supplemental material for A 2-Year Longitudinal Relationship Between Work–Family Conflict and Health Among Older Workers: Can Gardening Help?

Supplemental material, Supplementary_Table_1 for A 2-Year Longitudinal Relationship Between Work–Family Conflict and Health Among Older Workers: Can Gardening Help? by Eunae Cho, Tuo-Yu Chen and Megan C. Janke in Journal of Applied Gerontology

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Eunae Cho

Tuo-Yu Chen

Supplemental Material

Supplemental material for this article is available online.

References

Allen

T. D.

Shockley

K. M.

(2012). Older workers and work-family issues. In Hedge

J. W.

Borman

W. C.

(Eds.), The Oxford handbook of work and aging (pp. 520–537). Oxford University Press.

Amstad

F. T.

Meier

L. L.

Fasel

Elfering

Semmer

N. K.

(2011). A meta-analysis of work–family conflict and various outcomes with a special emphasis on cross-domain versus matching-domain relations. Journal of Occupational Health Psychology, 16, 151–169. https://doi.org/10.1037/a0022170

Ashe

M. C.

Miller

W. C.

Eng

J. J.

Noreau

(2009). Older adults, chronic disease and leisure-time physical activity. Gerontology, 55, 64–72. https://doi.org/10.1159/000141518

Ashton-Shaeffer

Constant

(2006). Why do older adults garden? Activities, Adaptation & Aging, 30, 1–18. https://doi.org/10.1300/J016v30n02_01

Carstensen

L. L.

Mikels

J. A.

(2005). At the intersection of emotion and cognition: Aging and the positivity effect. Current Directions in Psychological Science, 14, 117–121. https://doi.org/10.1111/j.0963-7214.2005.00348.x

Centers for Disease Control and Prevention National Center for Chronic Disease Prevention and Health Promotion. (2012). Older employees in the workplace. https://www.cdc.gov/workplacehealthpromotion/tools-resources/pdfs/issue-brief-no-1-older-employees-in-workplace-07122012.pdf

Chamberlain

A. M.

Manemann

S. M.

Dunlay

S. M.

Spertus

J. A.

Moser

D. K.

Berardi

. . . Roger

V. L.

(2014). Self-rated health predicts healthcare utilization in heart failure. Journal of the American Heart Association, 3, Article e000931. https://doi.org/10.1161/JAHA.114.000931

Chen

T.-Y.

Janke

M. C.

(2012). Gardening as a potential activity to reduce falls in older adults. Journal of Aging and Physical Activity, 20, 15–31. https://doi.org/10.1123/japa.20.1.15

Chen

T.-Y.

Lee

Buxton

O. M.

(2017). A greater extent of insomnia symptoms and physician-recommended sleep medication use predict fall risk in community-dwelling older adults. Sleep, 40(11), zsx142. https://doi.org/10.1093/sleep/zsx142

10.

Cho

Chen

T.-Y.

(2018). The effects of work-family experiences on health among older workers. Psychology and Aging. https://doi.org/10.1037/pag0000293

11.

Cho

Tay

Allen

T. D.

Stark

(2013). Identification of a dispositional tendency to experience work–family spillover. Journal of Vocational Behavior, 82, 188–198. https://doi.org/10.1016/j.jvb.2013.01.006

12.

Dawson

J. F.

(2014). Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29, 1–19. https://doi.org/10.1007/s10869-013-9308-7

13.

Evans

D. M.

Conte

Gilroy

Marvin

Theysohn

Fisher

(2008). Occupational therapy–Meeting the needs of older adult workers? Work, 31, 73–82.

14.

Fänge

Ivanoff

S. D.

(2009). The home is the hub of health in very old age: Findings from the ENABLE-AGE Project. Archives of Gerontology and Geriatrics, 48, 340–345. https://doi.org/10.1016/j.archger.2008.02.015

15.

Greenhaus

J. H.

Beutell

N. J.

(1985). Sources of conflict between work and family roles. Academy of Management Review, 10, 76–88. https://doi.org/10.5465/amr.1985.4277352

16.

Harden

C. M.

Peppard

P. E.

Palta

Barnet

J. H.

Hale

Nieto

F. J.

Hagen

E. W.

(2019). One-year changes in self-reported napping behaviors across the retirement transition. Sleep Health, 5, 639–646. https://doi.org/10.1016/j.sleh.2019.08.005

17.

Larsen

L. J.

(2014). Older Americans with a disability, 2008-2012. US Census Bureau.

18.

Headey

Wooden

(2004). The effects of wealth and income on subjective well-being and ill-being. Economic Record, 80, S24–S33. https://doi.org/10.1111/j.1475-4932.2004.00181.x

19.

Hertel

Rauschenbach

Thielgen

M. M.

Krumm

(2015). Are older workers more active copers? Longitudinal effects of age-contingent coping on strain at work. Journal of Organizational Behavior, 36, 514–537. https://doi.org/10.1002/job.1995

20.

Himes

C. L.

Reynolds

S. L.

(2012). Effect of obesity on falls, injury, and disability. Journal of the American Geriatrics Society, 60, 124–129. https://doi.org/10.1111/j.1532-5415.2011.03767.x

21.

Hobfoll

S. E.

(2002). Social and psychological resources and adaptation. Review of General Psychology, 6, 307–324. https://doi.org/10.1037/1089-2680.6.4.307

22.

Hurd

Rohwedder

(2009). Methodological innovations in collecting spending data: The HRS Consumption and Activities Mail Survey. Fiscal Studies, 30, 435–459. https://doi.org/10.1111/j.1475-5890.2009.00103.x

23.

IBM. (2016). SPSS Statistics for Windows, Version 24. IBM Corporation.

24.

Infantino

(2004). Gardening: A strategy for health promotion in older women. The Journal of the New York State Nurses Association, 35, 10–17.

25.

Karim

Weisz

Bibi

ur Rehman

(2015). Validation of the eight-item center for epidemiologic studies depression scale (CES-D) among older adults. Current Psychology, 34, 681–692. https://doi.org/10.1007/s12144-014-9281-y

26.

Kossek

E. E.

Pichler

Bodner

Hammer

L. B.

(2011). Workplace social support and work-family conflict: A meta-analysis clarifying the influence of general and work-family-specific supervisor and organizational support. Personnel Psychology, 64, 289–313. https://doi.org/10.1111/j.1744-6570.2011.01211.x

27.

Lottrup

Grahn

Stigsdotter

U. K.

(2013). Workplace greenery and perceived level of stress: Benefits of access to a green outdoor environment at the workplace. Landscape and Urban Planning, 110, 5–11. https://doi.org/10.1016/j.landurbplan.2012.09.002

28.

Meesters

(2009). The meaning of activities in the dwelling and residential environment: A structural approach in people-environment relations (Vol. 27). IOS Press.

29.

Meijman

T. F.

Mulder

(1998). Psychological aspects of workload. In Drenth

P. J. D.

Thierry

Wolff

C. J. d.

(Eds.), Handbook of work and organizational psychology (Vol. 2, pp. 5–33). Psychology Press.

30.

Mobily

K. E.

Nilson

Ostiguy

L. J.

MacNeil

R. D.

Wallace

R. B.

(1995). Seasonal variation in physical activity in older adults. Topics in Geriatric Rehabilitation, 10, 11–22.

31.

Nelson

M. E.

Rejeski

W. J.

Blair

S. N.

Duncan

P. W.

Judge

J. O.

King

A. C.

Macera, C. A., & Castaneda-Sceppa

(2007). Physical activity and public health in older adults: Recommendation from the American College of Sports Medicine and the American Heart Association. Circulation, 116, 1094-1105.

32.

Nielsen

L. M.

Kirkegaard

Østergaard

L. G.

Bovbjerg

Breinholt

Maribo

(2016). Comparison of self-reported and performance-based measures of functional ability in elderly patients in an emergency department: Implications for selection of clinical outcome measures. BMC Geriatrics, 16, 199–199. https://doi.org/10.1186/s12877-016-0376-1

33.

Nunes

B. P.

Flores

T. R.

Mielke

G. I.

Thumé

Facchini

L. A.

(2016). Multimorbidity and mortality in older adults: A systematic review and meta-analysis. Archives of Gerontology and Geriatrics, 67, 130–138. https://doi.org/10.1016/j.archger.2016.07.008

34.

Park

S.-A.

Lee

A.-Y.

Son

K.-C.

Lee

W.-L.

Kim

D.-S.

(2016). Gardening intervention for physical and psychological health benefits in elderly women at community centers. Horttechnology, 26, 474–483. https://doi.org/10.21273/HORTTECH.26.4.474

35.

Park

S.-A.

Shoemaker

Haub

(2008). Can older gardeners meet the physical activity recommendation through gardening? Horttechnology, 18, 639–643. https://doi.org/10.21273/HORTTECH.18.4.639

36.

Phillips

D. R.

Siu

O.-L.

(2012). Global aging and aging workers. In Hedge

J. W.

Borman

W. C.

(Eds.), The Oxford handbook of work and aging (pp. 11–32). Oxford University Press.

37.

Roberson

D. N.

Jr. Kudlacek

(2015). The garden colony: Restorative ecology. Ecopsychology, 7, 12–19. https://doi.org/10.1089/eco.2014.0045

38.

Rodiek

(2002). Influence of an outdoor garden on mood and stress in older persons. Journal of Therapeutic Horticulture, 13, 13–21.

39.

Rosenthal

Carroll-Scott

Earnshaw

V. A.

Santilli

Ickovics

J. R.

(2012). The importance of full-time work for urban adults’ mental and physical health. Social Science & Medicine, 75, 1692–1696. https://doi.org/10.1016/j.socscimed.2012.07.003

40.

Schuch

F. B.

Vancampfort

Rosenbaum

Richards

Ward

P. B.

Veronese

. . . Stubbs

(2016). Exercise for depression in older adults: A meta-analysis of randomized controlled trials adjusting for publication bias. Brazilian Journal of Psychiatry, 38, 247–254. https://doi.org/10.1590/1516-4446-2016-1915

41.

Scott

T. L.

Masser

B. M.

Pachana

N. A.

(2020). Positive aging benefits of home and community gardening activities: Older adults report enhanced self-esteem, productive endeavours, social engagement and exercise. SAGE Open Medicine, 8. https://doi.org/10.1177/2050312120901732

42.

Smith

Fisher

Ryan

Clarke

House

Weir

(2013). Psychosocial and lifestyle questionnaire 2006-2010: Documentation report core section LB. http://hrsonline.isr.umich.edu/sitedocs/userg/HRS2006-2010SAQdoc.pdf

43.

Soga

Gaston

K. J.

Yamaura

(2017). Gardening is beneficial for health: A meta-analysis. Preventive Medicine Reports, 5, 92–99. https://doi.org/10.1016/j.pmedr.2016.11.007

44.

Sonnega

Faul

J. D.

Ofstedal

M. B.

Langa

K. M.

Phillips

J. W. R.

Weir

D. R.

(2014). Cohort profile: The Health and Retirement Study (HRS). International Journal of Epidemiology, 43, 576–585. https://doi.org/10.1093/ije/dyu067

45.

Sonnentag

Fritz

(2007). The recovery experience questionnaire: Development and validation of a measure for assessing recuperation and unwinding from work. Journal of Occupational Health Psychology, 12, 204–221. https://doi.org/10.1037/1076-8998.12.3.204

46.

Sonnentag

Venz

Casper

(2017). Advances in recovery research: What have we learned? What should be done next? Journal of Occupational Health Psychology, 22, 365–380. https://doi.org/10.1037/ocp0000079

47.

Spector

W. D.

Fleishman

J. A.

(1998). Combining activities of daily living with instrumental activities of daily living to measure functional disability. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, B, 53, S46–S57. https://doi.org/10.1093/geronb/53B.1.S46

48.

Vaillant

G. E.

(2008). Aging well: Surprising guideposts to a happier life from the landmark study of adult development. Little, Brown.

49.

Wang

Glicksman

(2013). “Being Grounded”: Benefits of gardening for older adults in low-income housing. Journal of Housing for the Elderly, 27, 89–104. https://doi.org/10.1080/02763893.2012.754816

50.

Wang

MacMillan

(2013). The benefits of gardening for older adults: A systematic review of the literature. Activities, Adaptation & Aging, 37, 153–181. https://doi.org/10.1080/01924788.2013.784942

51.

Wolf

M. S.

Gazmararian

J. A.

Baker

D. W.

(2005). Health literacy and functional health status among older adults. Archives of Internal Medicine, 165, 1946–1952. https://doi.org/10.1001/archinte.165.17.1946

52.

Woo

Zajacova

(2016). Predictive strength of self-rated health for mortality risk among older adults in the United States: Does It differ by race and ethnicity? Research on Aging, 39, 879–905. https://doi.org/10.1177/0164027516637410

53.

Wright

S. D.

Wadsworth

A. M.

(2014). Gray and green revisited: A multidisciplinary perspective of gardens, gardening, and the aging process. Journal of Aging Research, 2014. https://doi.org/10.1155/2014/283682

54.

Arling

Wang

(2019). A cross-sectional study of self-rated health among older adults: A comparison of China and the United States. BMJ Open, 9. https://doi.org/10.1136/bmjopen-2018-027895

55.

Xue

Fleischmann

Head

McMunn

Stafford

(2020). Work and family conflict in relation to work exit in later career stage. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 75, 716–727. https://doi.org/10.1093/geronb/gby146

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB