active living for rural children
TRANSCRIPT
I
Au
FaNMBNPMsE
tuv
©
Active Living for Rural ChildrenCommunity Perspectives Using PhotoVOICE
Erin Hennessy, PhD, MPH, Vivica I. Kraak, MS, RD, Raymond R. Hyatt, PhD,Julia Bloom, MPH, Mark Fenton, MS, Colby Wagoner, MS, Christina D. Economos, PhD
Background: Active living integrates physical activity into one’s daily routine. Current understanding ofactive livingamongchildrenandtheir families livinginruralcommunities is limited.Acommunityperspectiveis critical to understand the contextual factors that influence children’s physical activity in rural areas.
Purpose: The purpose of this studywas to identify the perceived environmental factors that supportor hinder physical activity among rural children to develop testable hypotheses to inform futureinterventions for reducing unhealthy weight gain and preventing chronic diseases associated withphysical inactivity.
Methods: PhotoVOICEwasusedtoexploreactivelivingopportunitiesandbarriersforchildrenlivinginfourlow-income, rural U.S. communities. In 2007, parents (n�99) and elementary school staff (n�17) receiveddisposable cameras todocument theirperspective.Using theirphotographs andnarratives, participantsdevel-oped emergent themesduring a facilitated groupdiscussion. In 2008, study authorsused theAnalysisGrid forEnvironments Linked toObesity (ANGELO) framework to categorize the themes.
Results: Microenvironment themes include physical (e.g., natural features, topography); sociocul-tural (e.g., isolation); policy (e.g., time for school recess); and economic (e.g., funding for physicalactivity programs). Macroenvironmental themes related to the built and natural environments andtransportation infrastructure.
Conclusions: This study identifıed rural environment elements that community members per-ceived as influencing children’s physical activity patterns. Certain aspects were unique to rural areas,whereas other urban and suburban factors may be generalizable to rural settings. PhotoVOICE wasa useful participatory researchmethod to gain insight into perceived factors affecting rural children’sphysical activity behaviors.(Am J Prev Med 2010;39(6):537–545) © 2010 American Journal of Preventive Medicine
pbfrtaTrmtpaa
mdtr
ntroductionctive living integrates physical activity into one’sdaily routine.1 Although active living researchhas increased,2 emphasis has primarily been on
rban populations. Because characteristics relevant to
rom the John Hancock Research Center on Physical Activity, Nutrition,nd Obesity Prevention, Gerald J. and Dorothy R. Friedman School ofutrition Science & Policy (Bloom, Economos), andDepartment of Familyedicine and Public Health, School of Medicine (Hyatt), Tufts University,oston; Cancer Prevention Fellowship Program,National Cancer Institute,ational Institutes of Health (Hennessy), Bethesda, Maryland; Deakinopulation Health Strategic Research Center, Deakin University (Kraak),elbourne, Victoria, Australia; Consultant (Fenton), Scituate, Massachu-
etts; Coordinated School Health (Wagoner), Kentucky Department ofducation, Frankfort, KentuckyAddress correspondence to: ErinHennessy, PhD,MPH,Cancer Preven-
ion Fellow, National Cancer Institute, National Institutes of Health, Exec-tive Plaza South, Suite 150E, 6120 Executive Boulevard, MSC 7105, Rock-ille MD 20892. E-mail: [email protected].
I0749-3797/$17.00doi: 10.1016/j.amepre.2010.09.013
2010 American Journal of Preventive Medicine • Published by El
hysical activity vary according to physical landscape,uilt environment design, social norms, and culture,3
ındings from urban and suburban studies may not beelevant or transferable to rural areas.4–6 Compared toheir urban counterparts, rural U.S. residents experiencehigher prevalence of physical inactivity and obesity.7,8
he economies and cultures of rural communities haveecently shifted toward more sedentary lifestyles,9 whichay be attributed to several demographic and societal
rends.10,11 Understanding the contextual elements thatromote active lifestyles is needed to help rural childrenchieve the recommended 60 minutes of physicalctivity/day.12,13
Qualitative research can help to identify rural environ-ental factors that influence physical activity behavior,14
evelop hypotheses for existing physical activity pat-erns,5 and fınd opportunities for children to engage inegular physical activity to support a healthy weight.15,16
t may also provide insight about environmental percep-
sevier Inc. Am J Prev Med 2010;39(6)537–545 537
thepcsoi
rPmmsntiovfcta
MS
T2(ttepsVssfNccottasscngmcd
rt
dsp(empvagpsp
P
Peteal
ef
sevfow
538 Hennessy et al / Am J Prev Med 2010;39(6):537–545
ions for further study.17 The fıeld of active living researchas developed rigorousmethods formeasuring the physicalnvironment, but it is less advanced for assessing humanerceptions of the environment. Research suggests that per-eptionmaymediate theassociationbetweenobjectivemea-ures and health outcomes,18 and that objective data conveynly partial information to understand contextual factorsnfluencing physical activity.19
Learning from community members about their envi-onments will help inform public health practice.19,20
articipatory research methods21,22 confront a funda-ental problem of traditional, nonparticipatory com-unity assessment: what professionals, researchers,pecialists, and outsiders believe is important mayot be congruent with community members’ perspec-ives.17,23–25 The purpose of the present study was todentify the perceived environmental factors that supportr hinder physical activity among rural children to de-elop testable hypotheses to inform future interventionsor reducing unhealthy weight gain and preventinghronic diseases associated with physical inactivityhrough a qualitative, participatory researchpproach.
ethodstudy Sample
he current study was conducted in the spring of007 by Tufts University and Save the Childrenwww.savethechildren.org), a nonprofıt organiza-ion working in impoverishedU.S. rural communi-ies. It was part of a larger mixed methods research project toxplore activity-friendly rural environments for children. The sam-ling frame consisted of Save the Children’s rural U.S. partnerchools (children in grades K–5) categorized by region: Centralalley of California (n�11 schools);Mississippi RiverDelta (n�15chools); Southeast (n�9 schools); and Appalachia (n�48chools). Rural was de-ıned according to theational Center for Edu-ation Statistics localeodes, which are basedn the specifıc condi-ions of schools and refero very small geographicreas and circumstances,uch as population den-ity and size.26 Localeodes are often homoge-ous in small areas andenerally provide theost accurate type ofommunity where stu-ents reside.One school from each
egionwas selected topar-
Table 1. Demographic profile o
Location characteristics
Towna
Population size (n)
Household income (median)
Schoolb
Free–reduced lunch (%)
Enrollment (n)
Ethnicity (%)
aU.S. 2006 Census DatabNational Center for Education Statis
Wrelateat w
on
icipate using a random- B, black/African American; C, white/Cauc
igit sequence. Study staff met with school principals to secure theirupport. Table 1 provides demographic characteristics for the partici-ating townsandschools.Afterobtainingsupport, recruitment flyerstranslated into Spanish for California) were sent home with alllementary school–aged children. A total of 99 parents (88% fe-ale; n�25 in CA, n�27 in MS, n�20 in SC and n�27 in KY)articipated in the study. Teachers and school staff were recruitedia word of mouth, with a total sample of 17 (n�4 each in CA,MS,nd SC, and n�5 in KY). Each participating parent received a $100ift card, and school staff received a $50 gift card because theyarticipated in only the qualitative phase of the study. The currenttudy was approved by the IRB at Tufts University, and all partici-ants signed consent forms.
hotoVOICE
hotoVOICE was chosen because it is a qualitative, participatoryngagement approach that uses a specifıc photographic techniqueo explore community members’ perspectives.27 Disposable cam-ras allowed study participants to document their communitiesnd use their images with accompanying personal stories to shareocal knowledge and expertise to facilitate community change.28,29
Two workshops separated by group (parents vs school staff) forach school were held 1 week apart during after-school hours. Theırst detailed the study purpose and objective of the PhotoVOICE
process. PhotoVOICE is a naturalistic methodintended to haveminimal instruction; however,to provide a clear understanding of the tech-nique, a trained facilitator (the fırst author) led abrainstorming session about factors influencingchildren’s physical activity patterns at home,during school, and in their community and howto visually document these factors. Parents wereto photograph factors at home and in the com-munity while school staff photographed the
chool setting. Instructions for conducting this technique in anthical and sensitive manner (no faces photographed) were pro-ided. The facilitator demonstrated the process with photographsrom a pilot PhotoVOICE project. Participants received a tutorialn using a disposable fılm camera (27 exposures) and were given 1eek to take their photographs and return their camera to a drop-
rticipating communities
palachia(KY)
CentralValley (CA)
Mississippi RiverDelta (MS)
Southeast(SC)
1,738 3,466 2,102 1,220
5,923 25,313 17,972 26,937
100 100 100 91
196 472 319 399
100 C 93.7 H 99.7 B 94.0 B
Urban\Rural Classification Systems (nces.ed.gov/)26
chubcastajpm-.net.
f pa
Ap
1
tics.
atd Pww.line
asian; H, Hispanic
www.ajpm-online.net
op
spetestcoepwopetfsPwdpwsaii
ct(sbuo
RAprftaatd
rPcatod
dclAbppt
odawtnwar(chbaPblpctihnh
iavbpeb(pgwtc
nthc
Hennessy et al / Am J Prev Med 2010;39(6):537–545 539
D
ff box at each school. A local photo facility developed thehotographs.During the second workshop, also held during the after-
chool and evening hours in a designated classroom, partici-ants’ photographs were returned. The facilitator instructedach participant to select 3–5 photographs that best representedheir perspective. Participants documented this on the back ofach photo by explaining why they took it and what it repre-ented. The facilitator then invited each participant to describeheir photographs with the larger group. The facilitator used ahalk board to create columns representing barriers to andpportunities favoring physical activity for children, and tapedach picture that was discussed to the board to reflect partici-ant views and opinions. During the discussion, a third columnas added to reflect participants’ ideas that represented both anpportunity and a barrier. After all participants shared theirhotographs, the group was invited to share any other experi-nces or pictures that were not yet represented. This informa-ion was added to the appropriate column on the board. Theacilitator reviewed the photographs and led a second discus-ion to identify the themes that the pictures reflected. BecausehotoVOICE is a participatory engagement tool, all themesere participant driven and identifıed rather than investigatorriven. The facilitator helped the group achieve consensus andarticipants were asked to confırm whether the photographicall accurately reflected their collective views. The entire ses-ion was documented through extensive notes and by retainingll photos. The sessions held in California were bilingual (Span-sh/English) and conducted by the facilitator and a trainednterpreter.In the spring of 2008, the study investigators used the ANGELO
onceptual framework30 to organize the participant-identifıedhemes. This framework arranges themes according to the sizeschool, community andhome) and type of environment (physical,ociocultural, economic, and policy). This framework was chosenecause it provides a simple characterization of the various datased in systematic reviews about diet,31 physical activity,32,33 andbesity.34
esultstotal of 98 parents participated in the PhotoVOICE
roject and 90 disposable cameras were returned by theequired deadline. Two cameras malfunctioned in theıeld, which resulted in data from88 different cameras. Ofhe 17 school staff enrolled, 17 cameras were distributednd returned. The response level for the PhotoVOICEssessment was 92% for the parent groups and 100% forhe school staff groups. In all, 441 photographs wereiscussed.Table 2 provides an example of the emergent themes
elated to the community setting. Tables showing thehotoVOICE themes from the home and school settingsan be found in Appendix A (available online at www.jpm-online.net). Indicator signs illustrate whether theheme was discussed as an opportunity (�); a barrier (–);r both (�/–) depending on the context in which it was
iscussed. For example, the location of a home was iecember 2010
iscussed as an opportunity if it was close to the townenter, school, or community amenities because it al-owed children to walk to school or other destinations.lternatively, a remotely located home was viewed as aarrier to children’s physical activity and active livingotential. Figures 1–3 provide examples of participanthotographs and accompanying stories shared duringhe discussions.Overall, participants identifıed several barriers andpportunities for their children to be physically activeuring school, in the community, and at home acrossll environments. Aspects specifıc to rural settings alsoere identifıed, with themajority representing barrierso children’s physical activity patterns. In the commu-ity (Table 2), many physical environment barriersere identifıed such as a lack of sidewalks or shouldersnd buffers, heavy commercial truck traffıc, unpavedoad surfaces, and few indoor recreation destinationse.g., youth centers). Specifıc regional features in-luded the natural environment (i.e., hiking in theills); presence of nearby state parks; and the towneing unincorporated. Long distances between homend any other destination were often cited as a barrier.arents recognized owning a vehicle as an opportunityecause they used their car to overcome the barrier ofong distances (i.e., for their child to be active, thearent had to drive the child to a park or play space). Inertain areas, high gas prices then became an addi-ional barrier for parents to support their child’s phys-cal activity. Having adequate green space around theome was viewed as an opportunity whereas commu-ity factors like social disorder (i.e., crime, vacantomes, loitering, and gangs) were seen as a barrier.School location and climate were identifıed as limit-
ng children’s physical activity (Appendixes A and B,vailable online at www.ajpm-online.net). Climate wasiewed as a barrier because weather conditions coulde extreme and restricted school space often did notrovide adequate indoor alternatives. Results revealedlements related to child physical activity that haveeen commonly described in the broader literaturee.g., urban/suburban areas). School physical activityolicies (i.e., time allowed for physical education); pro-ramming; and availability of recreational equipmentere consistent themes, whereas other factors such aseacher role-modeling were mentioned in specifıcommunities.Appendix B (available online at www.ajpm-online.et) illustrates factors supporting physical activity oppor-unities at home such as having equipment (i.e., bicycles);ousehold policies or rules; parental support (i.e., logisti-al); presence of siblings; and household pets. Barriers
ncluded sedentary leisure screen time and limited fı-
Table 2. Community PhotoVOICE results identifying physical activity opportunities and barriers for children living in rural communities as reported by parents
Environment
Community setting
Representative sample comment from parentsMS CA SC KY
Physical Roads: no sidewalkor shoulder (–)
Roads: no sidewalkor shoulder (–)
Roads: no sidewalkor shoulder (–)
Roads: no sidewalkor shoulder (–)
“See this road—there’s no space to walk. There are hardly anysidewalks around here so you have to walk in the road.”
Traffic (–) Traffic (–) Traffic (–) Traffic (–) “Trucks! This is what you see going past my house every day.And they go fast!”
Destinationproximity (�/–)
Destinationproximity (�/–)
Destinationproximity (–)
Destinationproximity (–)
“There are some places in town to go to, especially if you livecloser to the center you can walk to places.” (�/–)
— — — — “There’s nothing around—you have to drive to get to any store orpark.” (–)
Parks/rec areas(�/–)
Parks/rec areas (–) Parks/rec areas (–) Parks/rec areas(�/–)
“This is our park. It used to be nice until the gang-bangers came.And it’s right next to the train tracks so it’s not safe for kids.”(–)
— — — — There’s a park here where children can play, but it’s not always agood place to go—especially at night because there’s a lot ofdrug and gang activity here (�/–)
Climate (–) Climate (–) Climate (–) Climate (–) “The weather here isn’t always nice [too hot, too rainy, too cold]and we have no place to go.”
Street conditions (–) Street conditions (–) — — “The streets are terrible—there are giant holes everywhere.”
— — Road surface (–) Road surface (–) “A lot of the roads are dirt so it’s hard to do certain activitieslike rollerblading.”
— Street lights (–) — — “There are not enough street lights so once the sun starts to godown it’s not safe for my kids to play outside.”
— — Unused “open”space (–)
— “This is what you see when you drive around here. They just cutsome of the trees down and leave it like this. I don’t know whythey couldn’t do something with this area.”
— Stray dogs (–) — — “There’s a lot of dogs around—some of them are stray andmean. You have to watch out and stay away from them.”
— — — Natural environment(�)
“We live in such a beautiful area. When the weather’s nice youcan go to the National Park, swim in the creek, or hike in thehills.”
Sociocultural Crime/drugs(–) Crime/drugs (–) Crime/drugs(–) Crime/drugs(–) “This community has too many drugs for kids to get to and thatkeeps them away from everything. This is a HUGE barrier!”
Vacant homes (–) Vacant homes (–) Vacant homes (–) — “There’s a lot of vacant houses that have a lot of drug andcriminal activity going on. It’s unsafe for kids. And look howclose it [vacant home] is to the road.”
(continued on next page)
540Hennessy
etal/Am
JPrevMed
2010;39(6):537–545
www.ajpm
-online.net
Table 2. (continued)
Environment
Community setting
Representative sample comment from parentsMS CA SC KY
Loitering (–) Loitering (–) Loitering (–) — “This is the park where some of the older kids hang around. It’snot safe for my young kids because sometimes the older kidsget into fights.”
Trash/debris(–) Trash/debris (–) — — “There’s just debris everywhere. No one is coming to clean it upso it just stays here.”
— Gangs (–) Gangs (–) — “The gangs here can be bad sometimes. There was a shootingnot too long ago. There’s just not enough police and becausea lot of people here are immigrants no one likes to talk to thepolice anyway.”
Health promotion (–) — — Health promotion (–) “There’s just not enough information out there for parents oranyone in the community. It’s hard to know what’s available.”
— — Church physicalactivity programs(�)
Church physicalactivity programs(�/–)
“This is our church. They run programs for children andsometimes they’re more activity-based.” (�/–)
— Sports participation(�/–)
— — “A lot of kids play soccer around here, but that’s mostly for theboys. There are more sports teams for boys. There are feweropportunities for girls.”
Economic Funding forprograms (–)
Funding forprograms (–)
Funding forprograms (–)
Funding forprograms (–)
“I wish there were more programs but the community justdoesn’t have the money.”
Food landscape (–) Food landscape (–) Food landscape (–) — “Look at all this junk. This is one of the only stores in town andit’s nothing but junk. But it’s cheap and the kids like it, so it’swhat they buy.”
Closed hospital (–) — — — “This hospital closed a while back. We have a new one now butthis one is in the center of town. It looks terrible. I wish theywould turn it into a community center or something but there’sno money to do anything with it.”
— Unincorporated town(–)
— — “The town isn’t incorporated so we don’t have the money to docertain things.”
Policy Recreationdepartment (–)
— — — “The recreation department limits a lot of activities that can bedone in some of the parks.”
— Locked school yard(–)
— — “This is the school yard—you can see all the sports fields. It’sright in the middle of town for everyone to see but it has thisbig gate around it that’s locked at night and on the weekendsso we can’t use it. I wish they would open it for us to use.”
Note: The data collection period for this study was April–June 2007. All items reported by the participants are included in the table with a representative example quote. Items were organizedusing the ANGELO30 framework in the spring of 2008.CA, California; KY, Kentucky; MS, Mississippi; rec, recreation; SC, South Carolina; (�), opportunity; (–), barrier; (�/–), both
Hennessy
etal/Am
JPrevMed
2010;39(6):537–545541
Decem
ber2010
nata
DTrcAp
pett
tpttliaiaitwah
agccawqrsprtC
FokkP2SET
FtrgPSET
FbPSET
542 Hennessy et al / Am J Prev Med 2010;39(6):537–545
ances. In two communities, additional barriers includedfull-timeworkingmother. Figure 4 represents themajorhemes that emerged from the PhotoVOICE processcross the four rural regions.
iscussionhis research provides insight into the perceived envi-onmental determinants related to active living forhildren living in four rural regions of the U.S. TheNGELO framework30 was used to organize partici-ant-identifıed aspects of rural settings (e.g., topogra-
igure 1. “There’s a lot of vacant houses that have a lotf drug and criminal activity going on. It’s unsafe for theids. And look how close it is to the road. I don’t like myids walking by here.”HOTOGRAPHER: Mother in rural South Carolina, May007ETTING: CommunityNVIRONMENT: SocioculturalHEME: Vacant homes
igure 2. “Our kids can’t walk or bike to school. Look athe road, it isn’t safe. And some of these kids come fromeally far away—they spend almost an hour on the bus toet here.”HOTOGRAPHER: School staff in Kentucky, May 2007ETTING: SchoolNVIRONMENT: Physical
cHEME: Location
hy, lack of sidewalks, isolation). The common barri-rs identifıed across all four regions may be attributedo the study being conducted in similarly disadvan-aged rural communities.Many of the parent fındings focus on barriers rather
han opportunities for physical activity. This may be ex-ected given the existing research on environmental fac-ors influencing physical activity. Studies suggest thatravel distance has a negative impact on physical activityevels35 and this may be especially true for children livingn poor rural communities. Other factorsmentioned herend supported by previous research include school phys-cal activity policies33 and parental support for activity,36
s well as urban design characteristics including aesthet-cs,37 safety from traffıc,38 and lack of nearby destina-ions.37–41 A lack of sidewalks and high traffıc speedsere viewed as barriers to children’s physical activity,42
nd walking or cycling to school are known correlates ofigher physical activity levels for children.16,43
Participants in two communities reported a greaterbility to walk to destinations, which may be due toreater street connectivity and a more accessible townenter (data not shown).6 This aligns with research indi-ating that living in walkable neighborhoods44 and easyccess to recreation facilities37 are positively associatedith physical activity. Planning policies to reward or re-uire more connected village-style development and theoutine upgrading of roadways to include shoulders andidewalks connecting to trip generators (e.g., schools,arks) are appropriate interventions for study even inural settings. Some residents discussed school consolida-ion and families owning a motor vehicle as themes.ombined with the identifıed challenges to walking and
igure 3. “This is an opportunity for physical activityecause this is where several children play together.”HOTOGRAPHER: Mother in rural Mississippi, April 2007ETTING: HomeNVIRONMENT: PhysicalHEME: Physical activity equipment
ycling, this suggests that improved transportation (e.g.,
www.ajpm-online.net
sr
Foailirpfrcmso
li(pit
mnpacteur
FScpP
Hennessy et al / Am J Prev Med 2010;39(6):537–545 543
D
huttles, car pools) may allow more rural children toeach after-school physical activity programs.14
The current research has several notable strengths.irst, PhotoVOICE has been utilized in other areas,29 butnly recently to understand the determinants of physicalctivity behavior.45 It has the distinct advantage of engag-ng hard-to-reach (under-resourced, low-literacy) popu-ations. Second, the present study advances understand-ng of the environmental determinants that ruralesidents perceive as having an impact on children’shysical activity patterns. The literature provides con-licting evidence regarding physical activity and self-eport of environmental measures,46,47 and a recent arti-le noted incongruence between perceived and objectiveeasures of physical activity environments.48 An earliertudy49 also noted discordance between perceived and
p
COM
SC
H
E
Electronicequipment
Logisticalsupport
Si
Sedetim
Healthy eating
Layout
Location
C
Healthy eating
Recess time
PA programming
PE time
Crime/drugs
Roads: no sidewTraffic
Climate
Physica
Socio-cult
Policy environment
igure 4. Consistent PhotoVOICE themes across all ruraoutheast, Appalachia, and Central Valley region of Califohild physical activity are green; barriers for child physichysical activity are blue.A, physical activity; PE, physical education
bjective measures of an environment’s walkability, with a
ecember 2010
ow-income, low–education level, and overweight partic-pants more likely to perceive high-walkability settingsbased on objective measures) as less walkable. This isarticularly relevant here as it supports the value of hav-ng participants use photographs to visually illustrateheir perceptions.Our study fındings also have some limitations. “Rural”ay be defıned inmany ways, and rural communities areot necessarily homogenous entities. Results from theresent study apply to the four focal areas andnot all ruralreas (such as the Southwest U.S. or rural areas in otherountries). All four communities met the study’s defıni-ion of underserved based on children’s participation lev-ls in the National School Lunch Program (Table 1). Thenderlying theme of poverty was discussed as a lack ofesources to support community programs or recre-
d cal ity
ITY
L
ent
Home location
Role modeling
Pets
Financial limitations
Recreation equipment
Climate
otivation
PA/PE programming
Funding for programs
Food landscape
nt es
Loitering
oulderDestination proximity
Parks/recreation area
ironment
Economicenvironment
nvironment
ttings and environments in the Mississippi River Delta,. Data collected in the spring of 2007; opportunities forctivity are red; both opportunities and barriers for child
Chilhysi
activ
MUN
HOO
OME
PA quipm
blings
ntarye
hild m
Vacahom
alk/sh
l env
ural e
l serniaal a
tional facilities and limited household income to sup-
ptmscm
FTfacisnaatfswscatfuarcsppigp
CTodfmtifhsecu
co
Ttdgltp
dTct
dHb
p
R
1
1
1
1
544 Hennessy et al / Am J Prev Med 2010;39(6):537–545
ort physical activity. The study illustrates the challengeso physical activity faced by children in poor rural com-unities, but fındings cannot be generalized to all ruralettings. Although the prevalence of poverty is highest forhildren living in rural communities,50 not all rural com-unities are poor.
uture Researchhis work suggests three broad areas for intervention andurther evaluation. First is the nexus between perceivednd actual environmental variables. Knowledge of per-eived environmental qualities can further understand-ng of how the rural environment affects active living, anduch knowledgemay allow for better evaluation andplan-ing of places that encourage physical activity.17,51 Itmaylso guide where physical and sociocultural interventionsre appropriate and how modifying community percep-ions may promote physical activity. Interventions to in-orm residents and change perceptions (e.g., way-fındingigns, active travel mentoring programs to teach safealking and cycling routes and skills, safety initiatives)hould be developed and tested. Second are actualhanges to the built environment, including the macro-ndmicro-environmental levels. Some evidence supportshe building of walking trails to increase physical activityor rural residents52 whereas other factors (such as landse and transportation policies that favor village centersnd traditional neighborhood designs or discourage seg-egated, single-use, low-density neighborhoods, largeonsolidated schools, and roads lacking shoulders oridewalks) should be evaluated. Third is studying therovision of programs and facilities for leisure-timehysical activity; although identifıed as a challenge here,t is questionable whether offering these types of pro-rams could increase physical activity levels in ruralopulations.
onclusionhis investigation illustrates the unique characteristicsf rural communities that may support or hinder chil-ren from being physically active. It provides datarom local communities to help develop high-qualityeasures of environmental perception for use in fu-
ure investigations. PhotoVOICE is an effective partic-patory research method to gain insight into perceivedactors affecting rural children’s physical activity be-avior. This research identifıed opportunities for de-igning effective obesity-prevention interventions,15
specially in areas where poverty levels are highest.53 Itan help inform other research approaches to fully
nderstand barriers and opportunities to influencehildren’s physical activity to reverse overweight andbesity trends in poor rural settings.
his study was funded by the Robert Wood Johnson Founda-ion’s Active Living Research Program (Grant No. 59458). Ad-itional support was provided by Save the Children’s U.S. Pro-rams and the New Balance Foundation. The authors wouldike to thank the families and school staff who participated inhis study and Save the Children’s fıeld staff who facilitated thearticipant recruitment and assisted with data collection.At the time of this study, E. Hennessy was a doctoral candi-ate at the Friedman School of Nutrition Science and Policy,ufts University, BostonMA. Dr. Hennessy is currently a Can-er Prevention Fellow at the National Cancer Institute, Na-ional Institutes of Health, Bethesda MD.At the time of this study, V. Kraak was with Save the Chil-ren, U.S. Programs. She is currently with Deakin Populationealth Strategic Research Center Deakin University, Mel-ourne, Victoria, Australia.No fınancial disclosures were reported by the authors of thisaper.
eferences1. Sallis JF, Cervero RB, AscherW, Henderson KA, Kraft MK, Kerr J. An
ecological approach to creating active living communities. Annu RevPublic Health 2006;27:297–322.
2. Saelens BE, Handy SL. Built environment correlates of walking: areview. Med Sci Sports Exerc 2008;40(7 S):S550–66.
3. Millington C, Ward Thompson C, Rowe D, et al. Development of theScottish Walkability Assessment Tool (SWAT). Health Place2009;15(2):474–81.
4. Nelson MC, Gordon-Larsen P, Song Y, Popkin BM. Built and socialenvironments associations with adolescent overweight and activity.Am J Prev Med 2006;31(2):109–17.
5. Sallis JF, Owen N. Ecological models of health behavior. In: Glanz K,Rimer BK, Lewis FM, eds. Health behavior and health education. 3rded. San Francisco: Jossey-Bass, 2002:462–84.
6. Yousefıan A, Hennessy E, Umstattd R, et al. Development of the RuralActive Living Assessment tools: measuring rural environments. PrevMed 2010;50(S 1):S86–92.
7. Boehmer TK, Lovegreen SL, Haire-Joshu D, Brownson RC. Whatconstitutes an obesogenic environment in rural communities? Am JHealth Promot 2006;20(6):411–21.
8. Lutfıyya MN, Lipsky MS, Wisdom-Behounek J, Inpanbutr-MartinkusM. Is rural residency a risk factor for overweight and obesity for U.S.children? Obesity 2007;15(9):2348–56.
9. National Association of Counties. Rural obesity: strategies to supportrural counties in building capacity. Washington DC, 2008.
0. Rural Sociological Society. The changing face of rural America. RuralSociological Society 2006;1(1).
1. Johnson K. Demographic trends in rural and small town America.Durham: University of New Hampshire, 2006. Report No.: 1.
2. Physical Activity Guidelines for Americans Committee. Physical Ac-tivity Guidelines Advisory Committee Report. Washington DC:USDHHS, 2008.
3. Tester JM. The built environment: designing communities to promote
physical activity in children. Pediatrics 2009;123(6):1591–8.www.ajpm-online.net
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
A
S
S
Hennessy et al / Am J Prev Med 2010;39(6):537–545 545
D
4. Yousefıan A, Ziller E, Swartz J, Hartley D. Active living for rural youth:addressing physical inactivity in rural communities. J Public HealthManag Pract 2009;15(3):223–31.
5. Brug J, van Lenthe FJ, Kremers SP. Revisiting Kurt Lewin: how to gaininsight into environmental correlates of obesogenic behaviors. Am JPrev Med 2006;31(6):525–9.
6. Moore JB, Jilcott SB, Shores KA, Evenson KR, Brownson RC, NovickLF. A qualitative examination of perceived barriers and facilitators ofphysical activity for urban and rural youth. Health Educ Res2010;25(2):355–67.
7. Nasar JL. Assessing perceptions of environments for active living. Am JPrev Med 2008;34(4):357–63.
8. Roosa MW, Jones S, Tein JY, Cree W. Prevention science and neigh-borhood influences on low-income children’s development: theoreticaland methodological issues. Am J Community Psychol 2003;31(1-2):55–72.
9. Roosa MW, White RMB, Zeiders KH, Yun-Tein J. An examination ofthe role of perceptions in neighborhood research. J Community Psy-chol 2009;37(3):327–41.
0. McAllister C, Green B, Terry M, Herman V, Mulvey L. Parents, prac-titioners, and researchers: community-based participatory researchwith Early Head Start. Am J Public Health 2003;93(10):1672–9.
1. Viswanathan M, Ammerman A, Eng E, et al. Community-based par-ticipatory research: assessing the evidence. Evid Rep Technol Assess(Summ) 2004;(99):1–8.
2. Whitelaw S, Baxendale A, Bryce C, MacHardy L, Young I, Witney E.“Settings”-based health promotion: a review. Health Promot Int2001;16(4):339–53.
3. Borra ST,Kelly L, ShirreffsMB,NevilleK,GeigerCJ.Developing healthmessages: qualitative studies with children, parents, and teachers helpidentify communications opportunities for healthful lifestyles and theprevention of obesity. J Am Diet Assoc 2003;103(6):721–8.
4. EconomosCD, Irish-Hauser S. Community interventions: a brief over-view and their application to the obesity epidemic. J Law Med Ethics2007;35(1):131–7.
5. Nasar JL. Influence of familiarity on responses to visual qualities ofneighborhoods. Percept Mot Skills 1980;51:635–42.
6. National Center for Education Statistics. Urban\rural classifıcationsystems. Institute of Education Services: US Department of Education.nces.ed.gov/.
7. Wang C, Burris MA. Photovoice: concept, methodology, and use forparticipatory needs assessment. Health Educ Behav 1997;24(3):369–87.
8. Wang CC. Photovoice: a participatory action research strategy appliedto women’s health. J Womens Health 1999;8(2):185–92.
9. Hergenrather KC, Rhodes SD, Cowan CA, Bardhoshi G, Pula S. Pho-tovoice as community-based participatory research: a qualitative re-view. Am J Health Behav 2009;33(6):686–98.
0. Swinburn B, Egger G, Raza F. Dissecting obesogenic environments: thedevelopment and application of a framework for identifying and pri-oritizing environmental interventions for obesity. PrevMed 1999;29(6Pt 1):563–70.
1. van der Horst K, Oenema A, Ferreira I, et al. A systematic review ofenvironmental correlates of obesity-related dietary behaviors in youth.Health Educ Res 2007;22(2):203–26.
2. Wendel-VosW, DroomersM, Kremers S, Brug J, van Lenthe F. Poten-tial environmental determinants of physical activity in adults: a system-atic review. Obes Rev 2007;8(5):425–40.
3. Ferreira I, van der Horst K,Wendel-VosW, Kremers S, van Lenthe FJ,Brug J. Environmental correlates of physical activity in youth—a re-view and update. Obes Rev 2007;8(2):129–54.
4. Kirk SF, Penney TL, McHugh TL. Characterizing the obesogenic envi-ronment: the state of the evidence with directions for future research.
Obes Rev 2009;11(2):109–17. oecember 2010
5. Oleson JJ, Breheny PJ, Pendergast JF, Ryan S, Litchfıeld R. Impact oftravel distance on WISEWOMAN Intervention attendance for a ruralpopulation. Prev Med 2008;47(5):565–9.
6. Davison KK, Cutting TM, Birch LL. Parents’ activity-related parentingpractices predict girls’ physical activity. Med Sci Sports Exerc2003;35(9):1589–95.
7. Humpel N, Owen N, Leslie E. Environmental factors associated withadults’ participation in physical activity: a review. Am J Prev Med2002;22(3):188–99.
8. Duncan MJ, Spence JC, Mummery WK. Perceived environment andphysical activity: a meta-analysis of selected environmental character-istics. Int J Behav Nutr Phys Act 2005;2:11.
9. De Bourdeaudhuij I, Sallis JF, Saelens BE. Environmental correlates ofphysical activity in a sample of Belgian adults. Am J Health Promot2003;18(1):83–92.
0. PikoraTJ,Giles-Corti B,KnuimanMW,Bull FC, JamrozikK,DonovanRJ. Neighborhood environmental factors correlated with walking nearhome: using SPACES. Med Sci Sports Exerc 2006;38(4):708–14.
1. Saelens BE, Sallis JF, Black JB, Chen D. Neighborhood-based differ-ences in physical activity: an environment scale evaluation. Am JPublicHealth 2003;93(9):1552–8.
2. CaseyAA, ElliottM,GlanzK, et al. Impact of the food environment andphysical activity environment on behaviors and weight status in ruralU.S. communities. Prev Med 2008;47(6):600–4.
3. Davison KK, Werder JL, Lawson CT. Children’s active commuting toschool: current knowledge and future directions. Prev Chronic Dis2008;5(3):A100.
4. Heath GW, Brownson RC, Kruger J, Miles R, Powell KE, Ramsey LT.The effectiveness of urban design and land use and transport policiesand practices to increase physical activity: a systematic review. TaskForce on Community Preventive Services. J Phys Act Health2006;3(1S):S55–76.
5. Kramer L, Schwartz P, Cheadle A, et al. Promoting policy and environ-mental change using photovoice in theKaiser PermanenteCommunityHealth Initiative. Health Promot Pract 2010;11(3):332–9.
6. Sallis JF, Hovell MF, Hofstetter CR, et al. Distance between homes andexercise facilities related to frequency of exercise among San Diegoresidents. Public Health Rep 1990;105(2):179–85.
7. Hoehner CM, Brennan Ramirez LK, Elliott MB, Handy SL, BrownsonRC. Perceived and objective environmental measures and physicalactivity among urban adults. Am J Prev Med 2005;28(2S 2):105–16.
8. Ball K, Jeffery RW, Crawford DA, Roberts RJ, Salmon J, Timperio AF.Mismatch between perceived and objective measures of physical activ-ity environments. Prev Med 2008;47(3):294–8.
9. Gebel K, Bauman A, OwenN. Correlates of non-concordance betweenperceived and objective measures of walkability. Ann Behav Med2009;37(2):228–38.
0. Economic Research Service. Rural America at a glance: 2005.Washing-ton DC: U.S. Department of Agriculture, 2005.
1. Humpel N, Owen N, Iverson D, Leslie E, Bauman A. Perceived envi-ronment attributes, residential location, and walking for particularpurposes. Am J Prev Med 2004;26(2):119–25.
2. Wiggs I, Brownson RC, Baker EA. If you build it, they will come:lessons from developing walking trails in rural Missouri. Health Pro-mot Pract 2008;9(4):387–94.
3. Levi J, Vintner S, Richardson L, St. Laurent R, Segal LM. F as in Fat2009: how obesity policies are failing in America. Washington DC:Trust for America’sHealth and the RobertWood Johnson Foundation,2009.
ppendix
upplementary data
upplementary data associatedwith this article can be found, in the
nline version, at doi:10.1016/j.amepre.2010.09.013.