or Prozac Regional planning natural amenities and psychological depression Depression Whats the problem Social impacts 2 nd leading cause of disability globallyleading source of years lived with a disability ID: 571039
Download Presentation The PPT/PDF document "Place" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Place or Prozac?
Regional planning, natural amenities, and psychological depressionSlide2
Depression: What’s the problem?*
Social impacts
2
nd leading cause of disability globally—leading source of years lived with a disability (Ferrari et al. 2010) Depression affects an estimated 350 million people worldwide (WHO 2010)
*
Economic impacts
3
rd
most costly medical condition for total expenditure
(AHRQ 2013
)
Total
cost of
all mental illnesses
in U.S. = $
317.6B
(
Insel
2008)
Slide3
The planner’s role?Slide4
Environmental context mental health and well-being
Environmental Context (IVs)
M
ental health (DV)
“Selection” or “drifter effect”
“Causation” or “breeder effect”
Physical activity
Environmental quality
Aesthetics
Opportunity
Access
Natural
amenities
Urban form
Water
resources
Air quality
Income
SES
Demographics
Time
Biophilia
Restoration (ART/SRT)
Positive Emotions
Soliphilia
/
solistalgia
TopophiliaSlide5
QuestionsAre county-level
measures of
urban form
and environmental context related to individual-level psychological depression?
If so,
what can planners do
to mitigate
or prevent psychological depression?Slide6
DataEnvironmental context
Natural Amenities Scale
(
McGranahan, USDA ERS) n=3000Public parks (2006 ESRI parks layer) n=3000Compactness (i.e. urban form) (Ewing & Hamidi 2010) n=967Mental IllnessP
sychological depression
(CDC—BRFSS 2012) n=447,000
Additional variables
Demographic and
socioeconomic (CDC—BRFSS 2012
) n=447,000Slide7
Individual-level (L1) variables mean (SD) or percentage (n=201,467)
Depression
(DV)
16%
Age
54.01 (17.20)
Married
55%
College educated
42%
Female
57%
Divorced
14%
Employed (any level)
58%
White
78%
Other
relationship status
31%
Income level (1-8)
5.99 (2.06)Black
10%
# of children0.59 (1.06)
Winter interview25%Asian
2%Tenure
76%#
of poor physical health days last month3.28 (7.56)
Other Race10%Veteran
13%#
of poor mental health days last month2.97 (6.96)“Good Health”
86%
Obese
64%
Respondent
p
hysically
active last month80%
County-level (L2)
variables mean (SD) (n = 945)
Natural
Amenities Scale
.28 (2.39)
Mean
Income (‘12)
$86,304
($20,302.91)
Park
fraction (land cover)
6.93
E+3 (1.70 E+2)
Park acres per
capita
7.28
E+3 (2.18 E+2)
Compactness Score
100.28 (24.98)Slide8
Natural Amenities Scale (1999)Slide9Slide10
Statistical method: 2-level binary logistic MLM
Nesting Structure
Level 1: Individual characteristics (BRFSS)
D
epression diagnosis = binary outcome variable
Demographics
Socio-economics
Level 2: County-level characteristics
Natural amenities scale
Public parks
Median Household income
County-level variables
Natural amenities
Parks
Sprawl
Individual
(BRFSS)Slide11
Tau
= .11130; likelihood function at iteration 2 = -2.800797E+005
(“
best fit”)Slide12
Selected & Significant Results
Fixed
Effect
CoefficientStandard ErrorP-valueOdds Ratio
Exp
(
dir&strength
)?
ANOVA
(“baseline” or “null” model)
… τ = 0.116
Intraclass Correlation Coefficient (ICC = .034) … L = -2.849 e
5
β0
γ00 (grand mean)
-1.8020.020.000
0.164Yes
Slopes- and Intercepts-as-outcomes (specified model) … τ = 0.111 … L = -2.801 e5
… McFadden Pseudo R2 = .001
β
0
γ00 (grand mean)
γ
01 (Natural amenities)-1.606-0.065
0.1310.0100.0000.000
0.2010.937Yes
Yes
β1 γ
10 (Poor phys
hlth days)
γ11 (Park fraction)
β
j
γ
30
(Age)γ40 (Female)γ
50
(Black)
γ
60
(Asian)
γ
80
(Divorced)
γ
80
(College)
γ
80
(Employed)
0.106
0.111
-0.006
0.681
-0.793
-0.879
0.389
0.119
-0.160
0.001
0.063
.001
0.032
0.058
0.128
0.041
0.028
0.030
0.000
0.027
.000
.000
.000
.000
.000
.000
.000
0.010
0.112
.994
1.975
0.453
0.415
0.389
1.127
0.852
Yes
Yes
No
(strength)
Yes
No (direction)
No (direction)
Yes
No (direction)
Yes Slide13
Results
Greate
r
natural amenities correspond to lesser odds of depression diagnosis
Each
unit increase in natural amenities
6.5% decrease in
likelihood
More park space corresponds to better physical health, which, in turn, leads to
lesser odds of psychological depression
L1 Control Variables of Interest
~96% of variation due to individual-level differences
Winter variable was intended to look at seasonal affective disorder Slide14
DiscussionCity and regional planners can and should work to address mental
illness
Place matters! Ecological planning can protect natural amenitiesParks/open space planning is an important tool: physical health mental healthCompactnes
s may become significant at smaller geographies
Evaluate difference between individuals in “most vs. least” compact placesSlide15
Limitations
Future Research
Geographic
scale of
BRFSS
public health data
Health
data at c
ensus
block or block group
Cross-sectional design
Longitudinal
design (mixed or panel data)
Regression
-based (MLM) correlative analysis
Structural
Equation Modeling & causal pathways
Operationalization
of environmental context
Include
additional context variables at both levels
Only
1 l
evel of nesting
Include
MSA and/or region-level IVs
Operationalization
of mediator/moderator IVs
Measure
interaction (use) and immersion (access)
Relationship
between variables & planning application
Relat
e mental illness to economic development
Time
: natural amenities change with climate change
Forecast
climate change impacts on depression ratesSlide16
Questions?Slide17
30 years of anecdotal, theoretical, and empirical researchEvolutionary affinity (
biophilia
)
Place attachment (Wilson 1984; Kaplan & Kaplan, 1989)Preference (population change, home value) (Herzog et al. 2000; McGranahan 1999; Wu & Gopinath 2008)Restorative benefits (cognitive)Attention Restoration Theory (Kaplan 1995, Kuo
2001,
Berto
2005)
Stress Recovery Theory
(Ulrich et al. 1991, 2003)
Well-being impacts (emotional)
Joy, happiness, self-confidence
(Kuo & Sullivan 2001)
What’s new here?
Operationalization of “nature”
Specificity of “mental illness”
Geographic scale (county)
Spatial planning view (in the US)Slide18
Selected Results
Tau
= 0.11632 (ICC=.034); likelihood function at iteration 2 = -
2.84947 E+5