A Critical Review of Human Psychology and Behavior Research to Examine the Biological and Methodological Plausibility of Obesity Research Findings

1
A Critical Review of Human Psychology and
Behavior Research to Examine the Biological and
Methodological Plausibility of Obesity Research
Findings
Candace D. Rutt, Ph.D. Centers for Disease
Control and Prevention
2
The Importance of the Individual

• How does individual psychology and behavior
  interact with the environment to affect caloric
  intake and caloric expenditure?

3
Increasing Obesity

• Results about caloric intake over time have been
  mixed
• Over the last several decades daily energy
  expenditure has decreased
• Leisure-time activity
• Occupational physical activity
• Activities of daily living
• Increase in attractive sedentary activities
• (James, 1995 McArdle et al., 1997)

4
Caloric Intake

• Hunger is not determined by biological mechanisms
  alone
• Preferences
• Innate
• Experience
• Cultural
• (Abbot et al., 1998 Berthoud Seeley, 2000
  Birch Marlin, 1982 Birch et al., 1980 Birch
  et al., 1984 Birch et al., 1987 Grundy, 1998
  Lipsitt Behl, 1990 Weingarten, 1983)

5
Caloric Intake

• Portion sizes
• Increased variety
• Presence of others
• Availability of food
• More time in automobiles
• Preference for convenience foods
• Easy access while driving
• (Berry et al., 1985 Bureau of the Census, 1976
  2000 Center for Science in Public Interest,
  2003 Edelman et al., 1986 Polivy et al., 1979
  Porikos et al., 1982 Rolls et al., 1992 Rolls
  Hetherington, 1989)

6
Physical Activity

• 30 minutes of moderate to vigorous physical
  activity on most days of the week
• Community Guide
• Enhanced access with informational outreach
• Street-scale urban design
• Community-scale urban design

7
Physical Activity and the Environment

• Characteristics of the environment may have
  various impacts on different types of physical
  activities
• Variations of environmental impact on physical
  activity and obesity seen in different
  populations

8
Walking for Transportation

• Density ()
• Land-use mix ()
• Connectivity ()
• Sidewalks ()
• High walkable neighborhood ()
• Car ownership (-)
• Access to beach (-)
• (Frank Engelke, 2001 2002 Giles-Corti., 2002
  Moudon et al., 1997 Ross Dunning, 1997
  Saelens et al., 2003)

9
Walking for Leisure

• Density (na)
• Connectivity (na)
• Sidewalks ( na)
• Land-use ( and -)
• Convenient Facilities ( and na)
• Sprawl (-)
• High walkable neighborhood (na)
• Traffic (- and )
• Hills ()
• Safety ( and -)
• Aesthetics ()
• Costal location ()
• (Ball et al., 2001 Brownson et al., 2001
  de Bourdeaudhuij et al., 2003 Ewing et al.,
  2003 Hovel et al., 1989 1992 Humpel et al.,
  2004Rutt Coleman, in press Saelens et al.,
  2003)

10
Moderate Vigorous Physical Activity

• Density (na)
• Land-use (na)
• Connectivity (na)
• Sidewalks ( and na)
• Heavy traffic ()
• Access to parks ()
• Number of facilities (na)
• Availability of pay facilities ()
• Distance to facilities ()
• Slope (na)
• Hills ()
• (Brownson et al., 2001 King et al., 2000 Rutt
  Coleman, in press Saelens et al., 2003 Sallis
  et al., 1990)

11
Obesity

• Density ( and na)
• Connectivity (na)
• Land-use mix (- and )
• Sprawl ()
• Living in high walkable neighborhoods (-)
• Lack of sidewalks ()
• Time spent in car ()
• Owning a car ()
• (Bell Popkin, 2003 Ewing et al., 2003 Frank
  et al., 2004 Giles-Corti et al., 2003 Saelens
  et al., 2003 Reddy et al., 1997 Rutt Coleman,
  in press)

12
Reasons for Inconsistencies

• No standardized way to measure many environmental
  variables
• Poor agreement between subjective and objective
  measures of the environment
• High correlations between urban form variables
• Individual level characteristics
• (Kirtland et al., 2003 Sallis et al., 1990
  Troped et al., 2001)

13
Individual Level Differences

• Variations in environmental correlates seen
  across
• Gender
• Ethnicity
• Income
• (de Bourdeadhuij, 2003 Flynn Fitzgibbon, 1998
  Flegal et al., 1998 Humpel et al,. 2004 2004
  Voorhees Young, 2003 )

14
Individual Level Differences

• 46 of adults in the U.S. believe that their
  neighborhood is unsafe. This rate was double in
  minority households
• Residents of poor neighborhoods walk more despite
  the fact that they report more fear of being
  victimized (Princeton
  Survey Research Associates, 1994 Ross, 2000)

15
Individual Level Differences

• For those with low incomes, the most important
  environmental variable associated with walking
  was enjoyable scenery
• For those with higher incomes sidewalks were the
  most important variable (Brownson et al.,
  2001)

16
Reasons for Inconsistencies

• Most studies include some basic demographic
  variables
• Usually do not measure psychological or social
  variables that could affect behavior
• Self-selection into neighborhoods
• Individual level characteristics associated with
  certain types of neighborhoods

17
Reasons for Inconsistencies

• Households choose residential locations partly
  based on desired travel behavior
• Several researchers have found clusters of
  lifestyle and demographic variables that were
  linked to transportation choices
• Attitudinal and demographic variables have been
  found to be more strongly associated with travel
  than built environment characteristics
• (Boarnet Sarmiento, 1998 Boarnet Greenwald,
  2000 Cervero Duncan, 2003 Giles-Corti
  Donovan, 2002 Kitamura et al.,1997 Hanson,
  1982 Rutt Coleman, in press Salomon Ben
  Akiva, 1983 Zimmerman, 1982)

18
Mode choice

• Only one study which has examined travel behavior
  before and after a move
• Over half of the families moved to similar
  neighborhoods
• Decreases in walking and biking trips were seen
  in those that relocated from a low auto-dependent
  neighborhood to a medium auto-dependent
  neighborhood
• (Krizek, 2000)

19
Mode choice

• Travel time is the most important predictor of
  mode choice
• Out-of-vehicle travel time (walking, biking) is
  considered more costly than in-vehicle travel
  time
• However other components of the trips
  (aesthetics, safety, etc.) affect the price or
  utility of the trip
  (Handy et al., 2002)

20
Variance Explained

• Of the six classes of determinants (demographic,
  psychological, behavioral, social, environment),
  individual level variables had the strongest and
  most consistent associations with physical
  activity
• Studies usually explain only 30 of the variance
  in physical activity or travel behavior
  (Crane et
  al., 1999 Baranowski, 1998 Handy et al., 1996)

21
Environment and PA

• An environment that encourages physical activity
  is necessary but insufficient to increase
  physical activity
• We may need better measures of the environment
• (Trost et al., 1996)

22
Paradigm Shift

• Cannot examine how the environment influences
  physical activity, eating behavior, or obesity
  without examining the individual
• Researchers need to start thinking about
  interactions between variables as well as their
  independent effects
• Ecological Models
• Mediators
• Moderators
• (Barron Kenny, 1986 McLeroy, Bibeau,
  Steckler, Glantz, 1988 Sallis Owen, 1996
  Stokols, 1996)

23
Conclusions

• Individual level variables should not be simply
  viewed as covariates to be controlled for
  rather they should be examined as important
  predictors of travel behavior and physical
  activity
• If individual level variables are not included in
  the emerging literature, incorrect conclusions
  may be drawn about the relationship between the
  environment and obesity