Abstract
INTRODUCTION
Exposure to green spaces is beneficial to mental health in a variety of ways, ranging from stress reduction to increased attentiveness and elevated self-esteem. The impact of views of greenness, as opposed to direct exposure, has been examined, but the association between self-reported views and depressive symptoms is not known. The purpose of this study is to examine the relationship between views of greenness and Patient Health Questionnaire–9 (PHQ-9) score.
METHODS
Questionnaire responses from 191 participants in the Health, Environment, and Action in Louisville (HEAL) study were examined. Univariate statistical analyses included Mann-Whitney U, Kruskal-Wallis, and Spearman rank tests. Inferential statistical analysis was linear regression.
RESULTS
Participant satisfaction with residential greenness was significantly associated with reduced PHQ-9 score (partially adjusted: linear coefficient = −0.42; 95% CI: −0.70 – −0.14; fully adjusted: linear coefficient = −0.21; 95% CI: −0.44 – 0.02). Additionally, being satisfied with local greenness was significantly associated with having views of greenness from home (linear coefficient = 1.97; 95% CI: 1.23–2.68).
CONCLUSIONS
Though views of greenness were not directly associated with depression, satisfaction with local greenness was associated with reduced PHQ-9 score, and having views of greenness from home was crudely associated with increased greenness satisfaction. The findings suggest urban greening interventions that focus on greenness satisfaction may be a strategy to reduce depression. Further research is necessary to better understand these relationships.
Keywords: depression, urban greening, views of greenness, satisfaction with greenness
INTRODUCTION
Despite rigorous evidence, exposure to green spaces has long been thought to provide myriad mental health benefits. Practices such as ‘forest bathing’, or mindfully experiencing a forested environment, have been promoted for their positive impact on the human psyche, with benefits including reduced stress and anxiety, decreases in attentional fatigue, and improved developmental cognition1–4. In communities with elevated morbidities, increased availability of green space has been associated with lower cortisol levels, and therefore less stress5, perhaps reducing the health effects of illness. Further, exposure to the natural environment, such as through walks, can lessen the symptoms of depression, though the effects are often small and not consistently reported between studies6–8.
But for all of these benefits, what is it about green spaces and nature that can heal ailing minds? Is it because forests and other green spaces physically make our environments healthier9,10? Is it that green spaces encourage more exercise11? Is it the reduction of demanding stimuli otherwise omnipresent in our daily lives, thus allowing our minds to freely wander and heal12,13? The answer is likely to be found in a combination of factors. A question remains, however: Is the presence of green spaces in one’s surroundings sufficient to provide the observed benefits, or is physical interaction with these green spaces necessary? Is greenness more an abstract concept experienced at the personal level, with derived benefits depending on one’s perceptions, even at a distance14? Is it enough to merely witness greenness in daily life? Can serious and commonplace mental illnesses such as depression be treated or prevented by people coexisting alongside green spaces, even without direct contact?
Major depression is among the most common mental health disorders in the United States. The 2013–2016 National Health and Nutrition Examination Survey (NHANES) found that 8.1% of US adults experienced depression in any given two-week period, paralleling results from previous NHANES surveys and suggesting that depression has maintained a steady presence in the general population over time15. A 2018 study found that lifetime prevalence of a major depressive disorder, as defined in the Diagnostic and Statistical Manual of Mental Disorders, Version 5 (DSM-5), in US adults was 20.6%, with the majority of cases being classified as moderate or severe16. The hallmark of major depressive disorder, as outlined by the DSM-5, is the presence of at least five symptoms, such as depressed mood, loss of interest in and satisfaction with activities, disrupted sleep patterns, reduced capacity to concentrate, and irritability17. In the worst cases, depression can prove fatal, as a serious symptom of severe, untreated depression is thoughts of self-harm. The Centers for Disease Control and Prevention (CDC) estimate that, in the US, 47173 deaths occurred in 2017 as the result of suicide18. The customary approach to treat depression typically involves medication, therapy, or both. Less often is the social and physical environment considered in prevention or management. Given the prevalence and the impact of depression, it is essential to understand the ways in which depressive symptoms can be mitigated through environmental factors.
Whether at home, school or in the workplace, views of greenness have been found to benefit mental health in a variety of ways. A study, examining university students living in campus dormitories, found that those who had natural views from their room windows had significantly increased attention capacity compared with those who had views of built spaces19. In a controlled laboratory setting, adults who viewed slides depicting natural scenes prior to being exposed to a mental stressor had increased parasympathetic activity and improved recovery of autonomic nervous system function post-exposure compared to those who viewed slides of built environments20. Additionally, improvements in self-esteem were noted in adults exposed to scenes of nature21. Another set of studies found that workers in largely sedentary office settings who had views of greenness from windows had significantly increased workplace satisfaction and general wellbeing compared with those who did not22. A similar study found that residents who had windowed views of greenness from their low-rise apartments reported elevated satisfaction with their neighborhood and overall wellbeing, whereas those who had views of built environments had lower satisfaction23. Further, a study of young Japanese men determined that ‘forest bathing’ produced physiological and psychological responses indicative of improved physical and mental health when compared to the same measures taken in an urban environment4.
Though some benefits of natural views from indoor built settings have been documented, the impact of such views on depressive symptoms remains unclear. For example, one study found that those who moved to greener urban environments, compared to their previous residence, had significantly improved mental health; those who moved to urban areas that were less green experienced a pre-move decline in mental health that eventually shifted back to baseline24. In contrast, Tennessen and Cimprich19 found no significant differences on a depression-dejection scale between those who did and did not have views of greenness from their dormitory windows. Further, in terms of housing density, one study found that individuals living in areas with higher housing density had lower depressive symptoms and a borderline significant association between lower depressive symptoms and living in an area with moderate access to green space versus no access25.
Many past studies examining the impact of the natural environment on depressive mood have relatively small sample sizes, and predominately focus on the presence of greenness but less on how it is experienced at the personal level8,26. The current literature lacks thorough examination of the impact of greenness satisfaction on the overall benefit of greenness on depression. Additionally, the relationship between views of nature from home and depression does not appear to be known. Is abundance or presence of greenness in one’s residential environment sufficient to incur previously observed mental health benefits, or are factors of greenness preference, and ultimately satisfaction, equally as important?
The primary purpose of this study is to determine whether views of greenness from one’s home are associated with PHQ-9 score. Secondary objectives are to examine the relationship of abundance of greenness and satisfaction with greenness with PHQ-9 score.
METHODS
During the summers of 2018 and 2019, South Louisville residents were recruited for the Health, Environment, and Action in Louisville (HEAL) study, part of Green Heart Louisville (GHL), a controlled longitudinal study assessing the impact of urban greening interventions on various health and environmental outcomes. Though several methods were employed, recruitment largely occurred via mailing, door-to-door distribution of flyers, and participation in community events. Interested individuals were screened for eligibility by phone; eligibility criteria included being aged 25–70 years, living within the study area in South Louisville, KY, and meeting certain health requirements (e.g. not pregnant, not diagnosed with cancer). Those found eligible were invited to attend clinical events.
After participant consent was obtained at clinical events, demographic and socioeconomic information was gathered, then participants visited several stations for clinical measurements (e.g. blood pressure, vascular functions, body fat composition), and finally they completed questionnaires at their desired pace, with some questionnaires completed as they waited at clinical stations. One questionnaire was the Patient Health Questionnaire–9 (PHQ-9), administered to determine the burden of depression in the study sample27. Shortly after the clinical visit, a subset of participants answered additional questionnaires pertaining to perceptions of greenness.
All study methods and materials were approved by the University of Louisville Institutional Review Board (IRB), and participant health information protection guidelines were followed as outlined in the Health Insurance Portability and Accountability Act (HIPAA). Participation in the study was voluntary. As participant identifiers were gathered during data collection, data were deidentified prior to analysis.
In total, HEAL recruited 735 participants and 203 completed the additional perceptions of greenness questionnaire. After the removal of missing values and declined-to-answer responses, a final sample of 191 participants remained.
Measures
Depression status
Depression status was determined based on responses to PHQ-9 items and not clinically diagnosed27. Specifically, responses to nine statements (Supplementary file Table 1) were used to calculate an aggregate depression score: 0–4 minimal or no depression, 5–9 mild depression, 10–14 moderate depression, 15–19 moderately severe depression, and 20–27 severe depression. Continuous PHQ-9 score served as the outcome variable.
Greenness variables
Three variables pertaining to greenness in participants’ communities were examined. The primary exposure variable, view of nature from home, was defined as ‘yes/no’. Participants were asked to indicate level of agreement that there is ‘lots of greenery around my local area (trees, bushes, gardens)’, with response options: ‘strongly disagree’, ‘disagree’, ‘agree’, and ‘strongly agree’. Participants were also asked to indicate a number from 1–10 that represented ‘satisfaction with the level of greenness in your neighborhood’, with 1 meaning ‘not at all satisfied’ and 5 meaning ‘neither dissatisfied nor satisfied’, and 10 meaning ‘100% satisfied’.
Demographic and socioeconomic variables
Several demographic and socioeconomic variables were considered to determine their relevance as confounders and predictors of the study outcome. These variables included: age, sex, race, ethnicity, number of people living in household, and annual household income. Age was defined in years. Sex was defined as male or female. Race was redefined as White/Caucasian, Black/African-American, or Other (i.e. American Indian/Alaskan Native, Asian, Hawaiian/Pacific Islander, and other). Ethnicity was defined as Hispanic/Latino or Non-Hispanic. Household size was defined as one, two, three, four, or five or more. Annual household income before taxes was defined in US$ as: <20000, 20000–44999, 45000–64999, 65000–89999, 90000–124999, or ≥125000.
Other variables
We included other variables as potential predictors or confounders. These variables included body mass index (BMI), general health, self-esteem, regular exercise, frequency of alcohol consumption, and perception of area safety. BMI was defined continuously in terms of kg/m2. General health was categorized based on responses: ‘excellent’, ‘very good’, ‘good’, ‘fair’, or ‘poor’ to the question: ‘In general, would you say your health is …’. Perception of self-esteem was based on responses to the question ‘How would you respond to the following statement: ‘I have high self-esteem’, where a scale from one to five, with one meaning ‘not very true of me’, three meaning ‘somewhat true of me’, and five meaning ‘very true of me’, was used. Regular exercise status was based on the question: ‘Do you exercise regularly (more than 10 minutes each time)?’ and defined as ‘yes/no’. Frequency of alcohol consumption was defined based on the responses: ‘never’, ‘monthly or less’, ‘2–4 times per month’, ‘2–3 times per week’, and ‘≥4 times per week’. Perception of area safety was based on responses to the statement ‘My neighborhood is safe from crime’, where a scale of one to five was provided, with one meaning ‘strongly agree’, three meaning ‘neither agree nor disagree’, and five meaning ‘strongly disagree’.
Statistical analysis
The outcome variable, continuous PHQ-9 score, was tested for normality using the Anderson-Darling test for normality and was found not to have a normal distribution. Subsequent univariate analysis examined the relationship between all included variables and the outcome variable using non-parametric tests of independence (e.g. Mann-Whitney U test, Kruskal-Wallis test, Spearman rank test) (Table 1). Also, all non-outcome variables, categorized by dichotomous views of nature from home, were compared using chi-squared, Fisher’s exact, and t-tests of independence to examine their relationship with the primary exposure variable (Table 2). Variables included in the analysis were examined for confounder and predictor status. Confounders were those variables found to be significantly associated with both outcome and exposure in univariate analysis (Tables 1 and 2) and to noticeably impact the magnitude of the outcome coefficient upon inclusion (i.e. 10% rule). Detected confounders were income and satisfaction with level of greenness in the local area. Age was also included as a confounder as a matter of best practice, despite not meeting all criteria, given its common implication in almost all biological interactions. Predictors were those variables found to be significantly associated with the outcome variable in univariate analysis (Table 1) and that were not confounders. Detected predictors were sex, general health status, self-esteem, regular exercise, frequency of alcohol consumption, and perception of neighborhood safety. Multicollinearity among the included variables was examined based upon tolerance and variance inflation factors. Linear regression was used to examine the association between views of greenness from home and PHQ-9 score. Partially adjusted and fully adjusted models are reported, with the partially adjusted model containing the primary exposure variable and confounders only, and the fully adjusted model containing the primary exposure variable, confounders, and predictors. Results are reported significant at alpha <0.05, and 95% confidence intervals (95% CI) are also provided. Statistical analyses were carried out using SAS statistical software (version 9.4 with SAS/STAT 14.1, SAS Institute Inc., Cary, NC).
Table 1.
Descriptive statistics of study population by outcome – PHQ-9 score (N=191)
| Variables | n (%) or mean ± SD | PHQ-9 score mean ± SD | pa |
|---|---|---|---|
| Views of greenness from within home | 0.1673 | ||
| Yes | 81 (42.4) | 4.0 ± 4.2 | |
| No | 110 (57.6) | 5.3 ± 5.4 | |
| Age (years) | 50.7 ± 12.0 | 0.7579 | |
| BMI (body mass index, kg/m2) | 29.7 ± 6.1 | 0.0694 | |
| Sex | 0.0171 | ||
| Female | 116 (60.7) | 5.3 ± 5.0 | |
| Male | 75 (39.3) | 3.9 ± 4.7 | |
| Race | 0.1638 | ||
| White/Caucasian | 157 (82.2) | 4.7 ± 5.0 | |
| Black/African-American | 26 (13.6) | 4.5 ± 4.3 | |
| Other | 8 (4.2) | 7.4 ± 4.3 | |
| Ethnicity | 0.0972 | ||
| Non-Hispanic | 186 (97.4) | 4.7 ± 4.9 | |
| Hispanic | 5 (2.6) | 8.6 ± 5.3 | |
| Annual household income before taxes (US$) | 0.0069 | ||
| <20000 | 34 (17.8) | 6.2 ± 5.6 | |
| 20000–44999 | 55 (28.8) | 5.6 ± 5.3 | |
| 45000–64999 | 42 (22.0) | 3.8 ± 3.8 | |
| 65000–89999 | 31 (16.2) | 4.1 ± 5.3 | |
| 90000–124999 | 26 (13.6) | 3.6 ± 4.1 | |
| >125000 | 3 (1.6) | 1.3 ± 0.6 | |
| Number of people in household | 0.7630 | ||
| 1 | 53 (27.7) | 5.8 ± 5.8 | |
| 2 | 79 (41.4) | 3.8 ± 4.2 | |
| 3 | 27 (14.1) | 5.6 ± 5.1 | |
| 4 | 15 (7.8) | 5.5 ± 5.4 | |
| ≥5 | 17 (9.0) | 4.2 ± 4.0 | |
| General health status | <0.0001 | ||
| Poor | 1 (0.5) | 6.0 | |
| Fair | 28 (14.7) | 10.2 ± 5.2 | |
| Good | 83 (43.4) | 5.4 ± 4.8 | |
| Very good | 68 (35.6) | 2.0 ± 2.5 | |
| Excellent | 11 (5.8) | 3.3 ± 3.7 | |
| Self-esteem | <0.0001 | ||
| 1 (lowest) | 16 (8.4) | 10.6 ± 6.2 | |
| 2 | 15 (7.8) | 9.1 ± 4.5 | |
| 3 | 58 (30.4) | 6.4 ± 5.1 | |
| 4 | 46 (24.1) | 2.5 ± 2.3 | |
| 5 (highest) | 56 (29.3) | 2.1 ± 2.9 | |
| Exercises regularly | 0.0006 | ||
| Yes | 118 (61.8) | 3.5 ± 3.4 | |
| No | 73 (38.2) | 6.8 ± 6.2 | |
| Frequency of alcohol consumption | 0.0012 | ||
| Never | 59 (30.9) | 6.0 ± 5.2 | |
| Monthly or less | 50 (26.2) | 5.7 ± 5.4 | |
| 2–4 times per month | 30 (15.7) | 2.5 ± 2.5 | |
| 2–3 times per week | 27 (14.1) | 4.3 ± 5.2 | |
| ≥4 times per week | 25 (13.1) | 3.2 ± 3.9 | |
| Perception that neighborhood is safe | 0.0096 | ||
| Strongly agree | 4 (2.1) | 2.7 ± 1.5 | |
| Agree | 18 (9.4) | 2.2 ± 3.6 | |
| Neither agree nor disagree | 24 (12.6) | 3.5 ± 3.6 | |
| Disagree | 72 (37.7) | 5.4 ± 5.3 | |
| Strongly disagree | 73 (38.2) | 5.3 ± 5.1 | |
| Satisfaction with level of greenness in local area | 0.0062 | ||
| 1 (not at all satisfied) | 21 (11.0) | 6.9 ± 6.4 | |
| 2 | 13 (6.8) | 5.7 ± 7.2 | |
| 3 | 18 (9.4) | 7.6 ± 6.4 | |
| 4 | 30 (15.7) | 3.9 ± 4.0 | |
| 5 (neither dissatisfied nor satisfied) | 24 (12.6) | 5.9 ± 4.3 | |
| 6 | 27 (14.1) | 3.4 ± 3.5 | |
| 7 | 15 (7.9) | 4.5 ± 3.7 | |
| 8 | 19 (10.0) | 3.3 ± 3.6 | |
| 9 | 10 (5.2) | 3.5 ± 3.9 | |
| 10 (100% satisfied) | 14 (7.3) | 2.9 ± 4.1 | |
| Perception of abundance of greenery in local area | 0.1084 | ||
| Strongly disagree | 20 (10.5) | 4.9 ± 4.5 | |
| Disagree | 51 (26.7) | 5.6 ± 5.7 | |
| Agree | 94 (49.2) | 4.6 ± 4.7 | |
| Strongly agree | 26 (13.6) | 3.4 ± 4.1 |
Non-parametric (Mann-Whitney U/Kruskal-Wallis/Spearman rank) p for comparison of PHQ-9 scores between variable categories.
Table 2.
Descriptive statistics of study population by exposure – views of nature from home (N=191)
| Variables | Total n (%) | Views of nature from home | pa | |
|---|---|---|---|---|
| Yes (N=81) n (%) | No (N=110) n (%) | |||
| Age (years), mean ± SD | 50.7 ± 12.0 | 53.6 ± 11.8 | 48.5 ± 11.8 | 0.0032 |
| BMI (body mass index, kg/m2), mean ± SD | 29.7 ± 6.1 | 30.0 ± 5.9 | 29.5 ± 6.4 | 0.5714 |
| Sex | 0.5107 | |||
| Female | 116 (60.7) | 47 (58.0) | 69 (62.7) | |
| Male | 75 (39.3) | 34 (42.0) | 41 (37.3) | |
| Race | 0.8251 | |||
| White/Caucasian | 157 (82.2) | 67 (82.7) | 90 (81.8) | |
| Black/African-American | 26 (13.6) | 10 (12.4) | 16 (14.5) | |
| Other | 8 (4.2) | 4 (4.9) | 4 (3.7) | |
| Ethnicity | 0.6520 | |||
| Non-Hispanic | 186 (97.4) | 78 (96.3) | 108 (98.2) | |
| Hispanic | 5 (2.6) | 3 (3.7) | 2 (1.8) | |
| Annual household income before taxes (US$) | 0.0450 | |||
| <20000 | 34 (17.8) | 22 (27.2) | 12 (10.9) | |
| 20000–44999 | 55 (28.8) | 19 (23.5) | 36 (32.7) | |
| 45000–64999 | 42 (22.0) | 17 (21.0) | 25 (22.7) | |
| 65000–89999 | 31 (16.2) | 11 (13.5) | 20 (18.2) | |
| 90000–124999 | 26 (13.6) | 12 (14.8) | 14 (12.8) | |
| >125000 | 3 (1.6) | 0 (0.0) | 3 (2.7) | |
| Number of people in household | 0.5909 | |||
| 1 | 53 (27.8) | 25 (30.9) | 28 (25.4) | |
| 2 | 79 (41.4) | 35 (43.2) | 44 (40.0) | |
| 3 | 27 (14.1) | 9 (11.1) | 18 (16.4) | |
| 4 | 15 (7.8) | 7 (8.6) | 8 (7.3) | |
| ≥5 | 17 (8.9) | 5 (6.2) | 12 (10.9) | |
| General health status | 0.3840 | |||
| Poor | 1 (0.5) | 1 (1.2) | 0 (0.0) | |
| Fair | 28 (14.7) | 12 (14.8) | 16 (14.6) | |
| Good | 83 (43.4) | 37 (45.7) | 46 (41.8) | |
| Very good | 68 (35.6) | 29 (35.8) | 39 (35.4) | |
| Excellent | 11 (5.8) | 2 (2.5) | 9 (8.2) | |
| Self-esteem | 0.3025 | |||
| 1 (lowest) | 16 (8.4) | 4 (4.9) | 12 (10.9) | |
| 2 | 15 (7.8) | 6 (7.4) | 9 (8.2) | |
| 3 | 58 (30.4) | 29 (35.8) | 29 (26.3) | |
| 4 | 46 (24.1) | 16 (19.8) | 30 (27.3) | |
| 5 (highest) | 56 (29.3) | 26 (32.1) | 30 (27.3) | |
| Exercises regularly | 0.1352 | |||
| Yes | 118 (61.8) | 55 (67.9) | 63 (57.3) | |
| No | 73 (38.2) | 26 (32.1) | 47 (42.7) | |
| Frequency of alcohol consumption | 0.1748 | |||
| Never | 59 (30.9) | 28 (34.6) | 31 (28.2) | |
| Monthly or less | 50 (26.2) | 21 (25.9) | 29 (26.4) | |
| 2–4 times per month | 30 (15.7) | 14 (17.3) | 16 (14.5) | |
| 2–3 times per week | 27 (14.1) | 13 (16.0) | 14 (12.7) | |
| ≥4 times per week | 25 (13.1) | 5 (6.2) | 20 (18.2) | |
| Perception that neighborhood is safe | 0.1779 | |||
| Strongly agree | 4 (2.1) | 3 (3.7) | 1 (0.9) | |
| Agree | 18 (9.4) | 10 (12.4) | 8 (7.3) | |
| Neither agree nor disagree | 24 (12.6) | 7 (8.6) | 17 (15.4) | |
| Disagree | 72 (37.7) | 34 (42.0) | 38 (34.6) | |
| Strongly disagree | 73 (38.2) | 27 (33.3) | 46 (41.8) | |
| Satisfaction with level of greenness in local area | 0.0002 | |||
| 1 (not at all satisfied) | 21 (11.0) | 3 (3.7) | 18 (16.4) | |
| 2 | 13 (6.8) | 3 (3.7) | 10 (9.1) | |
| 3 | 18 (9.4) | 2 (2.5) | 16 (14.6) | |
| 4 | 30 (15.7) | 13 (16.0) | 17 (15.4) | |
| 5 (neither dissatisfied nor satisfied) | 24 (12.6) | 10 (12.4) | 14 (12.7) | |
| 6 | 27 (14.1) | 12 (14.8) | 15 (13.6) | |
| 7 | 15 (7.9) | 7 (8.6) | 8 (7.3) | |
| 8 | 19 (10.0) | 14 (17.3) | 5 (4.5) | |
| 9 | 10 (5.2) | 8 (9.9) | 2 (1.8) | |
| 10 (100% satisfied) | 14 (7.3) | 9 (11.1) | 5 (4.6) | |
| Perception of abundance of greenery in local area | <0.0001 | |||
| Strongly disagree | 20 (10.5) | 6 (7.4) | 14 (12.7) | |
| Disagree | 51 (26.7) | 12 (14.8) | 39 (35.5) | |
| Agree | 94 (49.2) | 41 (50.6) | 53 (48.2) | |
| Strongly agree | 26 (13.6) | 22 (27.2) | 4 (3.6) | |
Chi-squared/Fisher’s exact/t-test p for comparison between views of nature from home categories.
RESULTS
The participant mean age was 50.69 years (Table 1; median 53.12; range 28.71–70.36). The majority of the sample was female (60.73%), Caucasian (82.20%), and Non-Hispanic (97.38%). Regarding greenness exposures, 81 (42.41%) stated that they had views of greenness from within their home, 120 (62.83%) agreed that there was an abundance of greenery in their neighborhood, and 85 (44.50%) indicated satisfaction with the level of neighborhood greenness. In terms of perceptions of health and health behaviors, 162 (84.82%) were of good or higher general health, 160 (83.8%) were of average or greater self-esteem, and 118 (61.78%) exercised regularly. In PHQ-9 responses, low energy and sleep issues were the most commonly reported concerns (Supplementary file Table 1). Few experienced thoughts of self-harm. Based on overall PHQ-9 scores, 77 (40.31%) participants were found to have depression of some form (Supplementary file Table 2). Of these individuals, 44 (57.14%) had mild depression, 22 (28.57%) had moderate depression, 7 (9.09%) had moderately severe depression, and 4 (5.19%) had severe depression. Multicollinearity was absent from the analysis, as tolerance values were all in excess of 0.57 and variance inflation factors were all 1.76 or below.
Results of the partially and fully adjusted linear regression models are presented in Tables 3 and 4. The primary exposure, views of greenness from within the home, was not found to be significantly associated with depression during univariate analysis; the protective effect of this exposure was also not significant in either the partially adjusted (linear coefficient = −0.80; 95% CI: −2.29 – 0.68) or fully adjusted (linear coefficient = −0.86; 95% CI: −2.07 – 0.35) models.
Table 3.
Linear regression examining association between PHQ-9 score and exposure to views of greenness from home – partially adjusted model
| Variables | Coefficient | 95% CI | p |
|---|---|---|---|
| Views of greenness from within home | −0.80 | −2.29 – 0.68 | 0.2874 |
| Age (years) | 0.03 | −0.03 – 0.87 | 0.3417 |
| Annual household income before taxes | −0.71 | −1.22 – −0.20 | 0.0063 |
| Satisfaction with level of greenness in local area | −0.42 | −0.70 – −0.14 | 0.0032 |
R2=0.1082. Adjusted R2=0.089.
Table 4.
Linear regression examining association between PHQ-9 score and exposure to views of greenness from home – fully adjusted model
| Variables | Coefficient | 95% CI | p |
|---|---|---|---|
| Views of greenness from within home | −0.86 | −2.07 – 0.35 | 0.1611 |
| Age (years) | 0.01 | −0.03 – 0.06 | 0.5913 |
| Annual household income before taxes | −0.26 | −0.69 – 0.17 | 0.2271 |
| Satisfaction with level of greenness in local area | −0.21 | −0.44 – 0.02 | 0.0758 |
| Sex | −0.44 | −1.59 – 0.71 | 0.4502 |
| General health status | −1.67 | −2.47 – −0.87 | <0.0001 |
| Self-esteem | −1.50 | −2.01 – −0.98 | <0.0001 |
| Exercises regularly | −0.78 | −1.99 – 0.43 | 0.2057 |
| Frequency of alcohol consumption | −0.04 | −0.47 – 0.39 | 0.8609 |
| Perception that neighborhood is safe | 0.31 | −0.23 – 0.86 | 0.2595 |
R2=0.4471. Adjusted R2=0.4164.
Interestingly, it was not the abundance or views of greenness that was found to be significantly associated with depression, but rather participant satisfaction with the level of greenness in their neighborhood (partially adjusted: linear coefficient = −0.42; 95% CI: −0.70 – 0.14; fully adjusted: linear coefficient = −0.21; 95% CI: −0.44 – 0.02). Perhaps views of greenness are related to satisfaction with the level of greenness, and therefore indirectly associated with depression. For example, univariate (chi-squared, p=0.0002) and crude linear (linear coefficient = 1.97; 95% CI: 1.23–2.68: R2=0.1383) results suggest such an association between views of greenness and satisfaction with neighborhood greenness level. In the partially adjusted model, the confounder income was found to be significantly associated with the outcome (linear coefficient = −0.71; 95% CI: −1.22 – −0.20), but such a relationship was not observed in the fully adjusted model. In the fully adjusted model, the predictors general health status (linear coefficient = −1.67; 95% CI: −2.47 – −0.87) and self-esteem (linear coefficient = −1.50; 95% CI: −2.01 – −0.98) were significantly associated with the outcome as well.
DISCUSSION
We found that satisfaction with neighborhood greenness was associated with lower rates of depression in an urban population. Although the relationship between views of greenness and depression was not significant, a possible protective association was observed. Perhaps the significance of the association was diminished by the small sample size, reducing power and making type II error more likely; thus, potential associations should be investigated in future research. Additionally, the association between satisfaction with neighborhood greenness and depression highlights the importance of individual perceptions and thus the need to involve community members in urban greening and similar interventions.
In addition to the variable annual household income, which indicated significantly lower PHQ-9 scores in higher income brackets in the partially adjusted model, the predictors general health status and self-esteem scale were found to be significantly associated with the outcome. For each one-level increase in health quality above ‘poor’, PHQ-9 score was significantly reduced. It is possible that when asked ‘In general, would you say your health is …’, participants would factor mental health into their responses, but multicollinearity was not found, suggesting that responses to the general health status item were independent of the outcome variable. The significant association between greater health and lower PHQ-9 score suggests that interventions that improve general health could help in reducing the burden of depression.
For each one-level increase in self-esteem above ‘1’ on the 1–5 scale, PHQ-9 score was significantly reduced. The association between self-esteem and depression is complex; some researchers adhere to the vulnerability model, suggesting that low self-esteem leads to depressive symptoms; others adhere to the scar model, suggesting that depressive symptoms leave ‘scars’ that ultimately result in lower self-esteem; and others suggest that both models have merit28–31. Future longitudinal studies will be better equipped to address these complexities. Self-esteem was independent of the outcome, suggesting the possibility that interventions that build self-esteem could simultaneously reduce the burden of depression.
Although neither the primary exposure nor the perception of abundance of greenness was found to be significantly associated with the outcome, satisfaction with local greenness was significantly associated with reduced PHQ-9 score. Perhaps, rather than being determined by the quantity of greenness present in one’s local area or the simple act of viewing natural scenery, one’s personal satisfaction with local greenness is more instrumental in alleviating depressive symptoms. Additionally, univariate and crude linear analyses found that views of greenness from home were significantly associated with satisfaction with area greenness, suggesting the possibility that views of greenness may be indirectly associated with depression. Perhaps the primary exposure was not found to be significantly associated with depression due to undetected collinearity. This relationship will need to be further examined in future research focusing on the impact of satisfaction.
Urban greening interventions and exposure to natural settings have been shown to improve many facets of mental health, suggesting they could significantly reduce depression in affected populations6,26,32. Given that satisfaction with area greenness was significantly associated with lower PHQ-9 score and that views of greenness from home were significantly associated with increased satisfaction with area greenness, community participation in and feedback on planned interventions seem to be of significant importance. Inclusion of as many voices as possible in intervention planning and implementation may increase the likelihood of satisfaction. As public health officials, policy makers, and urban planners consider modifications to the built and natural environments, satisfaction with greenness should be considered.
Research examining views of greenness and depressive symptoms is sparse and inconclusive19,24. Although not significant, study findings point to the possibility of a protective effect between the primary exposure and depression; thus, re-examination of potential associations with larger samples is warranted. Our finding that higher self-esteem is associated with a reduction in PHQ-9 score, and therefore depressive symptoms, is supported by extant literature33,34. Though little research has examined the relationship between general health and depression, depression has been found to be higher in those with chronic comorbidities, supporting the finding that greater general health is associated with reduced depressive symptoms35,36. Thus, the findings of this study generally align with existing literature and point to areas for future inquiry.
Limitations
Some limitations were present. First, the study is cross-sectional, removing the possibility of assessing causality. Second, recruitment predominately occurred by convenience sampling, which does not ensure a representative sample. Third, questionnaire responses relied on self-reports, making recall bias or misreporting possible. In a related vein, factors, such as overall contentment, optimism, and depression, might have influenced evaluations of greenness. For example, previous work suggests that depressed individuals have lower satisfaction with life, which may, in turn, influence evaluations of greenness37. Fourth, weaknesses in the scale employed may have resulted in some cases of depression being undetected38. Fifth, aesthetic preferences may influence satisfaction with greenness and ultimately our findings; thus, such preferences may be considerations in future inquiry. Finally, research would benefit from the addition of an objective greening measure, such as the normalized difference vegetation index (NDVI). Despite these limitations, the study findings contribute to the literature on relationships between perceptions of greenness, health characteristics, and depression.
CONCLUSIONS
Although no significant association was found between the primary exposure and outcome, study findings suggest that satisfaction with local greenness is significantly associated with a reduction in PHQ-9 score. Further, greater self-esteem and general health were significantly associated with reduced PHQ-9 score. Though only indicated in the partially adjusted model, elevated income may also be associated with reduced depressive symptoms. Given the pervasiveness of depression in the US population, any interventions capable of lowering disease burden are essential, particularly those that can affect a large number of people. Though increasing overall urban greenness has benefits, the findings suggest that satisfaction with local greenness, rather than the level of greenness, may assist with alleviating depressive symptoms. Therefore, it is especially important that urban greening interventions consider the opinions of community members.
Supplementary Material
FUNDING
This work was supported, in part, by grants from the National Institute of Environmental Health Sciences of the National Institutes of Health (NIH; Award Numbers R01 ES029846 and P42 ES023716); The Nature Conservancy (TNC); and the Christina Lee Brown Envirome Institute at the University of Louisville. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH, TNC, or the University of Louisville. The funding sponsors had no role in study design; data collection, analyses, or interpretation; manuscript preparation; or the decision to publish the results.
Footnotes
CONFLICTS OF INTEREST
The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none was reported.
ETHICAL APPROVAL AND INFORMED CONSENT
Ethical approval was obtained from the University of Louisville Institutional Review Board (IRB), and participant health information protection guidelines were followed as outlined in the Health Insurance Portability and Accountability Act (HIPAA). Participation in the study was voluntary and those who participated gave informed consent.
DATA AVAILABILITY
The data supporting this research are available from the authors on reasonable request.
REFERENCES
- 1.Chen HT, Yu CP, Lee HY. The Effects of Forest Bathing on Stress Recovery: Evidence from Middle-Aged Females of Taiwan. Forests. 2018;9(7):403. doi: 10.3390/f9070403 [DOI] [Google Scholar]
- 2.Park BJ, Tsunetsugu Y, Kasetani T, Kagawa T, Miyazaki Y. The physiological effects of Shinrin-yoku (taking in the forest atmosphere or forest bathing): evidence from field experiments in 24 forests across Japan. Environ Health Prev Med. 2010;15(1):18–26. doi: 10.1007/s12199-009-0086-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Louv R Last child in the woods: Saving our children from nature-deficit disorder. Algonquin Books; 2005. [Google Scholar]
- 4.Lee J, Park BJ, Tsunetsugu Y, Ohira T, Kagawa T, Miyazaki Y. Effect of forest bathing on physiological and psychological responses in young Japanese male subjects. Public Health. 2011;125(2):93–100. doi: 10.1016/j.puhe.2010.09.005 [DOI] [PubMed] [Google Scholar]
- 5.Ward Thompson C, Roe J, Aspinall P, Mitchell R, Clow A, Miller D. More green space is linked to less stress in deprived communities: Evidence from salivary cortisol patterns. Landsc Urban Plan. 2012;105(3):221–229. doi: 10.1016/j.landurbplan.2011.12.015 [DOI] [Google Scholar]
- 6.Berman MG, Kross E, Krpan KM, et al. Interacting with nature improves cognition and affect for individuals with depression. J Affect Disord. 2012;140(3):300–305. doi: 10.1016/j.jad.2012.03.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bratman GN, Daily GC, Levy BJ, Gross JJ. The benefits of nature experience: Improved affect and cognition. Landsc Urban Plan. 2015;138:41–50. doi: 10.1016/j.landurbplan.2015.02.005 [DOI] [Google Scholar]
- 8.Roberts H, van Lissa C, Hagedoorn P, Kellar I, Helbich M. The effect of short-term exposure to the natural environment on depressive mood: A systematic review and meta-analysis. Environ Res. 2019;177:108606. doi: 10.1016/j.envres.2019.108606 [DOI] [PubMed] [Google Scholar]
- 9.Selmi W, Weber C, Riviere E, Blond N, Mehdi L, Nowak D. Air pollution removal by trees in public green spaces in Strasbourg City, France. Urban For Urban Green. 2016;17:192–201. doi: 10.1016/j.ufug.2016.04.010 [DOI] [Google Scholar]
- 10.Nowak DJ, McHale PJ, Ibarra M, Crane D, Stevens JC, Luley CJ. Modeling the effects of urban vegetation on air pollution. In: Gryning SE, Chaumerliac N, eds. Air Pollution Modeling and Its Application XII. Springer; 1998:399–407. NATO • Challenges of Modern Society; vol 22. doi: 10.1007/978-1-4757-9128-0_41 [DOI] [Google Scholar]
- 11.Gladwell VF, Brown DK, Wood C, Sandercock GR, Barton JL. The great outdoors: how a green exercise environment can benefit all. Extrem Physiol Med. 2013;2(1):3. doi: 10.1186/2046-7648-2-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Song C, Ikei H, Miyazaki Y. Physiological Effects of Visual Stimulation with Forest Imagery. Int J Environ Res Public Health. 2018;15(2):213. doi: 10.3390/ijerph15020213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Berto R Exposure to restorative environments helps restore attentional capacity. J Environ Psychol. 2005;25(3):249–259. doi: 10.1016/j.jenvp.2005.07.001 [DOI] [Google Scholar]
- 14.Felsten G Where to take a study break on the college campus: An attention restoration theory perspective. J Environ Psychol. 2009;29(1):160–167. doi: 10.1016/j.jenvp.2008.11.006 [DOI] [Google Scholar]
- 15.Brody DJ, Pratt LA, Hughes JP. Prevalence of Depression Among Adults Aged 20 and Over: United States, 2013–2016. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2018. NCHS Data Brief, no 303. February 2018. Accessed June 15, 2021. https://www.cdc.gov/nchs/data/databriefs/db303.pdf [PubMed] [Google Scholar]
- 16.Hasin DS, Sarvet AL, Meyers JL, et al. Epidemiology of Adult DSM-5 Major Depressive Disorder and Its Specifiers in the United States. JAMA Psychiatry. 2018;75(4):336–346. doi: 10.1001/jamapsychiatry.2017.4602 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Uher R, Payne JL, Pavlova B, Perlis RH. Major depressive disorder in DSM-5: implications for clinical practice and research of changes from DSM-IV. Depress Anxiety 2014;31(6):459–471. doi: 10.1002/da.22217 [DOI] [PubMed] [Google Scholar]
- 18.Centers for Disease Control and Prevention. Suicide and Self-Harm Injury. Centers for Disease Control and Prevention. Updated March 1, 2021. Accessed June 15, 2021. https://www.cdc.gov/nchs/fastats/suicide.htm [Google Scholar]
- 19.Tennessen CM, Cimprich B. Views to nature: Effects on attention. J Environ Psychol. 1995;15(1):77–85. doi: 10.1016/0272-4944(95)90016-0 [DOI] [Google Scholar]
- 20.Gladwell VF, Brown DK, Barton JL, et al. The effects of views of nature on autonomic control. Eur J Appl Physiol. 2012;112(9):3379–3386. doi: 10.1007/s00421-012-2318-8 [DOI] [PubMed] [Google Scholar]
- 21.Brown DK, Barton JL, Gladwell VF. Viewing nature scenes positively affects recovery of autonomic function following acute-mental stress. Environ Sci Technol. 2013;47(11):5562–5569. doi: 10.1021/es305019p [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kaplan R The role of nature in the context of the workplace. Landsc Urban Plan. 1993;26(1–4):193–201. doi: 10.1016/0169-2046(93)90016-7 [DOI] [Google Scholar]
- 23.Kaplan R The nature of the view from home. Environ Behav. 2001;33(4):507–542. doi: 10.1177/00139160121973115 [DOI] [Google Scholar]
- 24.Alcock I, White MP, Wheeler BW, Fleming LE, Depledge MH. Longitudinal effects on mental health of moving to greener and less green urban areas. Environ Sci Technol. 2014;48(2):1247–1255. doi: 10.1021/es403688w [DOI] [PubMed] [Google Scholar]
- 25.Miles R, Coutts C, Mohamadi A. Neighborhood urban form, social environment, and depression. J Urban Health. 2012;89(1):1–18. doi: 10.1007/s11524-011-9621-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Fong KC, Hart JE, James P. A Review of Epidemiologic Studies on Greenness and Health: Updated Literature Through 2017. Curr Environ Health Rep. 2018;5(1):77–87. doi: 10.1007/s40572-018-0179-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.University of Washington. PHQ-9 Depression Scale. AIMS Center. Accessed June 15, 2021. https://aims.uw.edu/resource-library/phq-9-depression-scale [Google Scholar]
- 28.Sowislo JF, Orth U. Does low self-esteem predict depression and anxiety? A meta-analysis of longitudinal studies. Psychol Bull. 2013;139(1):213–240. doi: 10.1037/a0028931 [DOI] [PubMed] [Google Scholar]
- 29.Steiger AE, Fend HA, Allemand M. Testing the vulnerability and scar models of self-esteem and depressive symptoms from adolescence to middle adulthood and across generations. Dev Psychol. 2015;51(2):236–247. doi: 10.1037/a0038478 [DOI] [PubMed] [Google Scholar]
- 30.Orth U, Robins RW, Roberts BW. Low self-esteem prospectively predicts depression in adolescence and young adulthood. J Pers Soc Psychol. 2008;95(3):695–708. doi: 10.1037/0022-3514.95.3.695 [DOI] [PubMed] [Google Scholar]
- 31.Shahar G, Davidson L. Depressive symptoms erode self-esteem in severe mental illness: a three-wave, cross-lagged study. J Consult Clin Psychol. 2003;71(5):890–900. doi: 10.1037/0022-006X.71.5.890 [DOI] [PubMed] [Google Scholar]
- 32.Liu Y, Wang R, Xiao Y, Huang B, Chen H, Li Z. Exploring the linkage between greenness exposure and depression among Chinese people: Mediating roles of physical activity, stress and social cohesion and moderating role of urbanicity. Health Place. 2019;58:102168. doi: 10.1016/j.healthplace.2019.102168 [DOI] [PubMed] [Google Scholar]
- 33.Lakey CE, Hirsch JK, Nelson LA, Nsamenang SA. Effects of contingent self-esteem on depressive symptoms and suicidal behavior. Death Stud. 2014;38(6–10):563–570. doi: 10.1080/07481187.2013.809035 [DOI] [PubMed] [Google Scholar]
- 34.Dixon SK, Kurpius SER. Depression and College Stress Among University Undergraduates: Do Mattering and Self-Esteem Make a Difference? J Coll Stud Dev. 2008;49(5):412–424. doi: 10.1353/csd.0.0024 [DOI] [Google Scholar]
- 35.Bazargan M, Hamm-Baugh VP. The relationship between chronic illness and depression in a community of urban black elderly persons. J Gerontol B Psychol Sci Soc Sci. 1995;50(2):S119–S127. doi: 10.1093/geronb/50b.2.s119 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Schnittker J Chronic illness and depressive symptoms in late life. Soc Sci Med. 2005;60(1):13–23. doi: 10.1016/j.socscimed.2004.04.020 [DOI] [PubMed] [Google Scholar]
- 37.Gigantesco A, Fagnani C, Toccaceli V, et al. The Relationship Between Satisfaction With Life and Depression Symptoms by Gender. Front Psychiatry. 2019;10:419. doi: 10.3389/fpsyt.2019.00419 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Eack SM, Greeno CG, Lee BJ. Limitations of the Patient Health Questionnaire in Identifying Anxiety and Depression: Many Cases Are Undetected. Res Soc Work Pract. 2006;16(6):625–631. doi: 10.1177/1049731506291582 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting this research are available from the authors on reasonable request.