Point-of-Decision Prompts to Increase Stair Use A Systematic Review Update Robin E. Soler, PhD, Kimberly D. Leeks, PhD, MPH, Leigh Ramsey Buchanan, PhD, Ross C. Brownson, PhD, Gregory W. Heath, DHSc, MPH, David H. Hopkins, MD, MPH, the Task Force on Community Preventive Services Abstract: In 2000, the Guide to Community Preventive Services (Community Guide) completed a systematic review of the effectiveness of various approaches to increasing physical activity including informational, behavioral and social, and environmental and policy approaches. Among these approaches was the use of signs placed by elevators and escalators to encourage stair use. This approach was found to be effective based on suffıcient evidence. Over the past 5 years the body of evidence of this intervention has increased substantially, warranting an updated review. This update was conducted on 16 peer-reviewed studies (including the six studies in the previous systematic review), which met specifıed quality criteria and included evaluation outcomes of interest. These studies evaluated two interventions: point-of-decision prompts to increase stair use and enhance- ments to stairs or stairwells (e.g., painting walls, laying carpet, adding artwork, playing music) when combined with point-of-decision prompts to increase stair use. This latter intervention was not included in the original systematic review. According to the Community Guide rules of evidence, there is strong evidence that point-of- decision prompts are effective in increasing the use of stairs. There is insuffıcient evidence, due to an inadequate number of studies, to determine whether or not enhancements to stairs or stairwells are an effective addition to point-of-decision prompts. This article describes the rationale for these systematic reviews, along with information about the review process and the resulting conclusions. Additional information about applicability, other effects, and barriers to implementation is also provided. (Am J Prev Med 2010;38(2S):S292–S300) Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine Introduction T he prevalence of overweight and obesity in the U.S. has increased over the past several decades. In 2003–2004, 66.3% of adults in the U.S. were overweight or obese, and 32.2% were obese. 1 Obesity increases the risk of many diseases and health conditions, including hypertension, type 2 diabetes, coronary heart disease, stroke, gallbladder disease, osteoarthritis, and some cancers. 2 The primary cause of overweight and obesity in the U.S. is energy imbalance. 2,3 Energy imbal- ance occurs when the number of calories used is not equal to the number of calories consumed. Energy expenditure has been on the decline in the U.S. for decades, due in part to increasing automation of previously manual activi- ties. In 1996, the U.S. Preventive Services Task Force (USPSTF) recommended that healthcare providers counsel all patients on the importance of incorporating physical activity into their daily routines. 4 One way to increase energy expenditure, and improve energy bal- ance, is to incorporate small bouts of physical activity into daily routines. 3 Many intervention approaches are available to increase engagement in physical activity by adults. 5 Each of these approaches has a set of advantages and disadvantages and can be applied, with differing degrees of success, to people with a variety of demographic characteristics and life- styles in diverse locations. As noted in an earlier review by From the Community Guide Branch, Division of Health Communications and Marketing Strategy, National Center for Health Marketing, (Soler, Leeks, Hopkins) and the Chronic Disease Nutrition Branch, Division of Nutrition, Physical Activity, and Obesity, National Center for Chronic Disease Health Promotion and Prevention (Buchanan), CDC, Atlanta, Georgia; St. Louis University, School of Public Health (Brownson), St. Louis, Missouri; and University of Tennessee at Chattanooga, Department of Health and Human Performance (Heath), Chattanooga, Tennessee Address correspondence and reprint requests to: Robin E. Soler, PhD, Community Guide Branch, Centers for Disease Control and Prevention, 1600 Clifton Road, MS-E69, Atlanta GA 30333. E-mail: [email protected]. 0749-3797/00/$17.00 doi: 10.1016/j.amepre.2009.10.028 S292 Am J Prev Med 2010;38(2S):S292–S300 Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine
Transcript
I
Toiid
FaLNDGLo
C1
S
Point-of-Decision Prompts to IncreaseStair Use
A Systematic Review Update
Robin E. Soler, PhD, Kimberly D. Leeks, PhD, MPH, Leigh Ramsey Buchanan, PhD,Ross C. Brownson, PhD, Gregory W. Heath, DHSc, MPH, David H. Hopkins, MD, MPH,
the Task Force on Community Preventive Services
Abstract: In 2000, the Guide to Community Preventive Services (Community Guide) completed asystematic review of the effectiveness of various approaches to increasing physical activity includinginformational, behavioral and social, and environmental and policy approaches. Among theseapproaches was the use of signs placed by elevators and escalators to encourage stair use. Thisapproach was found to be effective based on suffıcient evidence. Over the past 5 years the body ofevidence of this intervention has increased substantially, warranting an updated review. This updatewas conducted on 16 peer-reviewed studies (including the six studies in the previous systematicreview), which met specifıed quality criteria and included evaluation outcomes of interest. Thesestudies evaluated two interventions: point-of-decision prompts to increase stair use and enhance-ments to stairs or stairwells (e.g., painting walls, laying carpet, adding artwork, playing music) whencombined with point-of-decision prompts to increase stair use. This latter intervention was notincluded in the original systematic review.According to the Community Guide rules of evidence, there is strong evidence that point-of-
decision prompts are effective in increasing the use of stairs. There is insuffıcient evidence, due to aninadequate number of studies, to determine whether or not enhancements to stairs or stairwells arean effective addition to point-of-decision prompts. This article describes the rationale for thesesystematic reviews, along with information about the review process and the resulting conclusions.Additional information about applicability, other effects, and barriers to implementation is alsoprovided.(Am J PrevMed 2010;38(2S):S292–S300) Published by Elsevier Inc. on behalf of American Journal of PreventiveMedicine
soathtt(cpiad
eacw
ntroductionhe prevalence of overweight and obesity in theU.S. has increased over the past several decades.In 2003–2004, 66.3% of adults in the U.S. were
verweight or obese, and 32.2% were obese.1 Obesityncreases the risk ofmany diseases and health conditions,ncluding hypertension, type 2 diabetes, coronary heartisease, stroke, gallbladder disease, osteoarthritis, and
rom the Community Guide Branch, Division of Health Communicationsnd Marketing Strategy, National Center for Health Marketing, (Soler,eeks, Hopkins) and the Chronic Disease Nutrition Branch, Division ofutrition, Physical Activity, and Obesity, National Center for Chronicisease Health Promotion and Prevention (Buchanan), CDC, Atlanta,eorgia; St. Louis University, School of Public Health (Brownson), St.ouis, Missouri; and University of Tennessee at Chattanooga, Departmentf Health and Human Performance (Heath), Chattanooga, TennesseeAddress correspondence and reprint requests to: Robin E. Soler, PhD,
ommunity Guide Branch, Centers for Disease Control and Prevention,600 Clifton Road, MS-E69, Atlanta GA 30333. E-mail: [email protected]/00/$17.00
sdoi: 10.1016/j.amepre.2009.10.028
292 Am J Prev Med 2010;38(2S):S292–S300 Published by El
ome cancers.2 The primary cause of overweight andbesity in the U.S. is energy imbalance.2,3 Energy imbal-nce occurswhen the number of calories used is not equalo the number of calories consumed. Energy expenditureas been on the decline in theU.S. for decades, due in parto increasing automation of previously manual activi-ies. In 1996, the U.S. Preventive Services Task ForceUSPSTF) recommended that healthcare providersounsel all patients on the importance of incorporatinghysical activity into their daily routines.4 One way toncrease energy expenditure, and improve energy bal-nce, is to incorporate small bouts of physical activity intoaily routines.3
Many intervention approaches are available to increasengagement in physical activity by adults.5 Each of thesepproaches has a set of advantages and disadvantages andanbe applied,with differing degrees of success, to peopleith a variety of demographic characteristics and life-
tyles in diverse locations. As noted in an earlier reviewby
sevier Inc. on behalf of American Journal of Preventive Medicine
Soler et al / Am J Prev Med 2010;38(2S):S292–S300 S293
F
heGuide to Community Preventive Services (Communityuide), which evaluated interventions designed to in-rease physical activity, “the role of community-basednterventions to promote physical activity has emerged ascritical piece of an overall strategy to increase physicalctivity behaviors among the people of the Unitedtates.”5 This 2002 review focused on community-basedntervention approaches, including:
Informational approaches to change knowledge andattitudes about the benefıts of and opportunities forphysical activity within a community;Behavioral and social approaches to teach people thebehavioral management skills necessary both for suc-cessful adoption and maintenance of behavior changeand for creating social environments that facilitate andenhance behavioral change; andEnvironmental and policy approaches to change thestructure of physical and organizational environmentsto provide safe, attractive, and convenient places forphysical activity.
This article reports the fındings from an update to the002 Point-of-Decision Prompts review, which is a be-avior and social approach as described above. The up-ated systematic review examines literature regarding theffectiveness of prompts on increasing stair use either byncreasing the number of actual stair users or increasinghe frequency of stair use through prompts that relate tooth of these foci, which can be implemented by commu-ities to help increase levels of physical activity. Point-of-ecision prompts can be used alone or with stairwellnhancements in an attempt to improve the effectivenessf the prompt (i.e., by making stairwells more attractiveo potential users).
uide to Community Preventive Services
he systematic reviews in this report present the fındingsf the independent, nonfederal Task Force on Commu-ity Preventive Services (Task Force). The Task Force iseveloping theCommunity Guidewith the support of theSDHHS in collaboration with public and private part-ers. The CDC provides staff support to the Task Forceor development of theCommunity Guide. The book,Theuide to Community Preventive Services. What Works toromote Health? (Oxford University Press, 2005; alsovailable at www.thecommunityguide.org) presents theackground and the methods used in developing theommunity Guide. The physical activity review notedbove was published in the American Journal of Preven-ive Medicine in 20025,6 and describes the broader ana-ytic framework used to evaluate the effectiveness of
ommunity-based physical activity interventions. r
ebruary 2010
ethodshis updated review was conducted according to the meth-ds developed for the Community Guide, which have beenescribed in detail elsewhere.5,7 As an update to an existingommunity Guide review,5 some information and guidanceas drawn from the previous review team and resultingocumentation. Inclusion criteria for studies in this reviewere: (1) primary research published in a peer-reviewedournal; (2) published in English before April 20, 2005;3) met the minimum research quality for study design andxecution7; and (4) evaluated the effects of point-of-decisionrompts to encourage stair use (with or without enhance-ents to the stairwell). The outcome measure remainedtair use, and the search strategywaswidened by inclusion ofdditional electronic databases. The systematic review teamthe team) accepted the broader conceptual approach of theriginal physical activity review5 but developed a new con-eptual framework for the interventions evaluated in thispdate. The team recalculated the original effect size mea-ure (relative change) and calculated a new summary effecteasure (absolute change); reexamined the evidence re-arding applicability of this intervention; and updated theverall conclusions based on the original six studies and andditional ten studies found through the updated literatureearch.
onceptual Approach
oint-of-decision prompts are motivational signs, placed atr near stairwells or at the base of elevators and escalators,ncouraging people to use the stairs. These prompts areypically designed to change a behavior of interest by pro-iding information about a healthier alternative or establish-ng a deterrent to the behavioral standard (e.g., announcinghat an elevator is off limits to those capable of using stairs),ith the intended goal of motivating and enabling people tohange their behavior and maintain that change over time.tairwell enhancements improve the appearance of stair-ells by painting walls or laying carpet. A conceptual ap-roach was used to evaluate the effectiveness of point-of-ecision prompts and stairwell enhancements to increasetair use. The approach suggests that extended presence of aoint-of-decision prompt designed to increase stair useight work by changing individual knowledge or attitudesbout using the stairs. Information provided through stairrompts might also contribute to an individual’s change innowledge or attitudes about the value of physical activity ineneral. As a result, prompts are expected to increase the usef stairs as a mode of transportation and may change atti-udes toward or amount of engagement in physical activity.alking up or down stairs uses more energy than taking an
levator or escalator, and stair use requires bodily move-ent. The relationships between stair use and caloric expen-iture and between stair use and physical activity were not
eviewed. This conceptual approach suggests that the slight
S294 Soler et al / Am J Prev Med 2010;38(2S):S292–S300
ncrease in caloric expenditure (energy expenditure) result-ng from stair use, which serves to improve energy balancean, in combination with other forms of physical activity,ontribute to physiologic improvements that are, in turn,elated to longer-term health outcomes.
election of Outcomes for Review
he primary outcomes examined in this review were objec-ive measurements of changes in the use of stairs during twor more periods of time. Objective measurements were vi-ual counts of people using the stairs or electronic countsfrom devices such as motion detectors). Some of the quali-ying studies reported other outcomeswhichwere examinedut are not presented in this report.Selection of stair use as an outcome assumes that small
mounts of physical activity on a regular basis will helpmprove the energy imbalance that affects large numbers ofeople (particularly people who are sedentary and thoseho are obese). Stair use typically involves ascending orescending one to four flights per day. Using stairs expendswice as much energy as using elevators8 with each stairscended burning approximately 0.11 kilocalorie and eachtair descended burning approximately 0.05 kilocalorie.9
egular, substantial stair use (as many as six assents of 199teps per assent per day for 12 weeks) has been shown tomprove cardiovascular outcomes among previously seden-ary young women10 and Benn et al., in their study of a smallroup of older men found that
climbing only three to four flights of stairs at a mod-erate pace (approximately 50–70 s) elicits peak circu-latory demands similar to, but at a much more rapidrate of adjustment than, 10 minutes of horizontalwalking at 2.5 mph, intermittently carrying a 30-pound weight, or 4 minutes of walking up a moder-ately steep slope.11
Over the long-term, this added energy expenditure couldontribute to improved energy balance and longer-termealth outcomes such as weight control.
earch Strategy
he articles considered for this review were obtained fromystematic searches of multiple databases, reviews of biblio-raphic reference lists, and consultations with experts in theıeld. The team’s updated search for evidence encompassedhe period from 2000 to April 2005, which overlapped withhe search conducted for the original Community Guideeview of these interventions (search period 1980–2000).5
he original review used the following seven databases: En-iroline, MEDLINE, PsychInfo, Social SciSearch, Sociologi-al Abstracts, Sportdiscus, and Transportation Research In-ormation Services (TRIS). For the team’s updated search,he following 15 databases were examined: ArticleFirst,INAHL, EMBASE, Enviroline, Health Promotion and Edu-
ationDatabase,MEDLINE, Ovid, PsycINFO, PubMed, So- f
ial SciSearch, Social Science Citation Index, Sociologicalbstracts, SPORTDiscus, Transportation Research Infor-ation Services (TRIS), and WorldCat. This list includesome databases not available at the time of the originaleview.
valuating and Summarizing the Studies
ach study that met the inclusion criteria was evaluated forhe suitability of the study design and study execution usinghe standardizedCommunity Guide abstraction form.12 Theuitability of each study design was rated as greatest, mod-rate, or least depending on the degree to which the designrotects against threats to validity. The execution of eachtudy was rated as good, fair, or limited on the basis ofeveral predetermined factors that could potentially limit atudy’s utility for assessing effectiveness. Each study waseviewed by at least two trained researchers. Concerns abouttudy design and execution were discussed with an expert inhysical activity interventions and differences in opinionere resolved by consensus among a team of three system-tic reviewers (the coordination team). Only studies ratedreatest or moderate in design suitability and good or fair inxecution were considered qualifying studies and includedn the team’s fınal assessment of the evidence in this review.tudies with limited execution are, by Community Guideethods, excluded from consideration, and studies of leastuitable design were excluded by the coordination teamecause the body of literature was adequately representedith moderate and greatest suitability study designs.
alculation of Effect Sizes
he qualifying studies provided measurements of change inhe number or proportion of people using the stairs beforend after the implementation of point-of-decision promptswith or without additional enhancements to the stairs ortairwells). To facilitate comparison across studies and anvaluation across the body of evidence, individual study armesults were converted (if necessary) into measurements ofoth absolute and relative percentage change. In addition,henever possible, a mean effect size was calculated on thentire sample in each study arm. Studies contained morehan one study arm when there were multiple locations orechanisms of implementation for the intervention. Inome cases, effect measures were reported for subgroupeans (e.g., one for men and one for women). For thesetudy fındings, the mean of the subgroups was incorporatednto the overall calculations for median and interquartilenterval (IQI), thus providing only one independent effectize per study arm (these are referred to as data points). Forime–series studieswithout a concurrent comparison group,he effect sizes (using pretest measurements and the lastostinterventionmeasurement provided) were calculated as
ollows:
www.ajpm-online.net
a
r
pw
a
r
gmFmeoaa
aptcpds
RPS
Tfpb
RP
Pola
hnolspstwsa
Eipao
F
Soler et al / Am J Prev Med 2010;38(2S):S292–S300 S295
F
bsolute percentage change (difference is described as “per-centage point change”),
Effect size � Ipost � Ipre;
elative percentage change (result is described as “percent-age change”),
Effect size � ((Ipost � Ipre)/Ipre) � 100.
For the study that included a concurrent comparisonopulation (not exposed to the intervention), the effect sizeas calculated as follows:
bsolute percentage change (difference is described as “per-centage point change”),
Effect size � (Ipost � Ipre) � (Cpost � Cpre);
elative percentage change (result is described as “percent-age change”),
For all calculations, I � intervention group; C � comparisonroup; and “pre” and “post” subscripts indicate measure-ents taken before and after intervention implementation.or studies in which multiple postintervention measure-ents were taken, the measurement most distant from thend of the intervention is used. In addition to the calculationf effect sizes for each study, an overall median effect sizend interquartile interval were determined for both absolutend relative percentage change.Throughout the results section effect sizes are presented
s both absolute and relative change. The original review ofoint-of-decision prompts5 reported relative change only;hus relative change is reported in this paper to allow foromparisons across reviews. Absolute change is also re-orted because it provides an estimate of change that is notependent on baseline rates (that may vary according toetting or other population characteristics).
esultsart I: Interventions to Increase the Use oftairs (Updated)
he team examined the evidence from qualifying studiesor two related interventions: (1) point-of-decisionrompts; and (2) stairwell enhancements when com-ined with point-of-decision prompts.
eview of Evidence: Point-of-Decisionrompts
oint-of-decision prompts are motivational signs placedn or near stairwells or at the base of elevators and esca-ators encouraging people to use stairs. These signs, such
s the one shown in Figure 1, inform individuals about a d
ebruary 2010
ealth or weight-loss benefıt from using stairs, about aearby opportunity to use stairs, or both. A few examplesf the content of the signs include “improve your waist-ine, use the stairs” or “your heart needs exercise, use thetairs.” Point-of-decisionsignsmaybecombinedwithotherrompts suchas footprintsplaced todirect individuals to thetairwell; the teamconsidered these additional effortswithinhis review. Point-of-decision prompts when combinedithmoreelaborate enhancements to the stairsor stairwells,uchaspaintingstairwellwallsorplayingmusic instairwells,re reviewed separately below.
ffectiveness. The literature search identifıed 15 stud-es that assessed the effectiveness of point-of-decisionrompts when used alone in changing the frequency ormount of stair use or the number of stair users.13–27 Fourf these studies were rated as having least suitable study
21–24
igure 1. Sample point-of-decision prompt
esigns and were excluded from further analysis.
Tmtisetp
StaA1tqwntpdmswusstuss
O
Ea(ltifcw(ussu
quclq
o9si
dvrcp
deaaocTisan
aisfAvutsplch
AaudtdPstuswvss
S296 Soler et al / Am J Prev Med 2010;38(2S):S292–S300
wo of the studies19,25 were of good execution; the re-aining nine13–18,20,26,27 were rated as fair. One addi-
ional paper provided information on a study alreadyncluded in the review.28 Details of the 11 qualifyingtudies, including a summary of the content, delivery,valuation design, and outcomes, are available at www.hecommunityguide.org/pa/environmental-policy/odp.html.
tudy design and implementation characteris-ics. All 11 qualifying studies used time–series designs,nd were rated as being of moderate suitability.13–20,25–27
ll of the qualifying studies were conducted between980 and 2003, and measured stair use in adult popula-ions. The types of point-of-decision prompts used in theualifying studies were signs13–19,26,27 or banners,20
hich were distinctions used by the authors and notecessarily related to the size of the prompt, although inhe one study specifying stair banners, the messages werehysically placed on each stair, but like the signs, varied inesign and message. The 11 qualifying studies imple-ented a variety of point-of-decision prompts messagesuch as health benefıts and health promotion,13,14,16–18,25
eight control,14 and signs (in Spanish and English)sing either an individual or family perspective topecifıcally target the Hispanic community.19 Onetudy focused primarily on African-Americans, andhe point-of-decision prompt was tailored to this partic-lar community.15 Additionally, in one study a deterrentign was displayed that limited the elevator to use by thetaff and the physically challenged.26
utcomes Related to Stair Use
leven qualifying studies,13–20,25–27 consisting of 21 studyrms for stair use, provided evidence in terms of absolutei.e., percentage point) change. In these studies, the base-ine rates of stair use ranged from 1.7% to 39.7% of po-ential users (median�8.2, IQI�5.2, 21.2). Stair use dur-ng the intervention period in these study arms rangedrom 4.0% to 41.9% of potential users. The medianhange for the 21 study arms representing these studiesas an increase in stair use of 2.4 percentage pointsIQI�0.83, 6.7 percentage points). Increases in stairse in 15 of 21 study arms were reported as statisticallyignifıcant,14–20,22–28 while two study arms (from theame study) reported a signifıcant decrease in stairse.19
To examine effects relative to baseline stair use, elevenualifying studies that included 21 study arms for stairse were evaluated in terms of relative (i.e., percentage)hange.13–20,25–27 The majority of studies reported a lowevel of baseline stair use (�20%). Overall, in the 11
ualifying studies, the median relative improvement in i
bserved stair use was 50 percentage points (IQI�5.4%,0.6%) from baseline (Figure 2; note that data points forubpopulations and simplemeans for the total sample arencluded on this fıgure).The team examined how the effectiveness of point-of-ecision prompts, measured in units of absolute change,aried with baseline stair use and found no signifıcantelationship between baseline stair use and absolutehange (Spearman’s rho� �0.39, n�21 data points,�0.77).The team also examined the effectiveness of point-of-ecision prompts by the period of observation. Research-rs in nine of the studies (representing 18 studyrms)13,17–20,25–27 left point-of-decision prompts in placend observed passersby for different lengths of time, withbservation periods ranging from 1 week (relativehange�81.1%)26 to 12weeks (relative change�5.16%).25
he period of observationwas not reported for two qualify-ng studies representing three study arms.14,15 Therewas noignifıcant relationship between the period of observationnd relative change in stair use (Spearman’s rho� �0.12,�18 data points, p�0.65).Overall, 25 of the 28 data points representing 17 study
rms (ten studies) in this body of evidence reported fınd-ngs in favor of the intervention. For some studies thetatistical signifıcance of the results was not reported, andor some, the fındings differed bydirection for subgroups.mong those studies with fındings in favor of the inter-ention, at the individual level the actual increase in stairse was modest. Because using stairs is a physical activityhat can be done by most people in most places wheretairs are present, modest increases in stair use amongopulations of adults across settings (malls, worksites,ibraries, and other such facilities) and across time canontribute to or extend bouts of physical activity andmayave a positive effect on energy balance.
pplicability. The body of evidence used to evaluate thepplicability of this intervention was the same as thatsed to evaluate effectiveness. Seven studies were con-ucted in the U.S.,14,15,17–19,26,27 two were conducted inhe United Kingdom,13,20 and one study each was con-ucted in Scotland specifıcally16 and in Australia.25
oint-of-decision prompts were evaluated in a range ofettings, and two studies investigated the effectiveness ofhe same intervention in different locations.18,19 Baselinese of stairs differed across settings (e.g., buildings withingle or multiple flights of stairs, public locations andorksites), and the effectiveness of the intervention alsoaried across settings, suggesting that the goal (e.g., lei-ure activity or work), or type of dress (e.g., suit or workhoes) of people in certain types of locations may have an
Soler et al / Am J Prev Med 2010;38(2S):S292–S300 S297
F
The studies included in this review were conducted inhopping malls,14,18,20 train and bus stations,15,16,18 air-orts,19,26 an offıce building,19 a bank,19 a healthcare fa-ility,25 a medical school,13 a university,17 and a univer-ity library.19,27 Four studies13,17,19,25 specifıed thatorkers were included among those observed. Althoughome locations may have had a greater percentage oforkers present (e.g., offıce buildings and universities)han others (e.g., airports and malls), it is likely thatorkers were present in all places included in this review.Six studies, representing 13 study arms,14,16,19,20,26,27
easured effectiveness separately among men andomen (median relative percentage change equaled 33%nd 48%, respectively) and found that point-of-decisionrompts had similar effects for both groups. Addition-lly, age was measured in fıve studies representing tentudy arms.14,18,20,26,27 All studies included adult-onlyamples, and study authors grouped individuals into ei-her young or old adults (median relative percentagehange equals 51% and 65%, respectively). Age groupingsaried by study with three using age 30 years as a cutoffoint for the younger group14,18,27 and one using age 40ears as a cutoff point for the younger group.26 Two
igure 2. Relative change in percentage of people usingeasurements from 21 study arms in 11 qualifying studie
tudies used age 60 years to distinguish between their I
ebruary 2010
ounger and older groups.20,27 No studies examined theffectiveness of the intervention in changing the behaviorf children or adolescents. Four studies, with eight studyrms,14,15,18,20 measured effectiveness for whites and Af-ican Americans and found no difference between racialroups (median relative percentage change�53% forach group). The team therefore believed that this type ofntervention is likely to be effective across diverse settingsnd population groups, provided that the appropriateare is taken to adapt the messages15,19 for each setting oropulation. However, stair use may vary according tonvironmental characteristics (e.g., accessibility of stairs,umber of flights to destination, or cleanliness of stair-ell) and personal factors (e.g., body composition, pres-nce of children or heavy loads) of the targeted popula-ion, whichmay affect responsiveness to the intervention.
ther positive or negative effects. One study reportedonsignifıcant changes in elevator use consistent withhanges in stair use (lift use decreased as stair use in-reased),13 which may result in reduced electricity usend related costs. Potential harms of increased stair usenclude strains and sprains, as well as injuries due to falls.
irs when point-of-decision prompts are displayed (n�35Some studies have multiple study arms so are repeated.
sta
n addition, one author noted that “posting a sign extol-
lvsh
EoNi
Brmaowomomopsi
SoIrsiArsgvsTlto
iaerslespr
C
Arpotmce
REP
EptswTna
Eawcnpswttap
Steapotlweawawsro
S298 Soler et al / Am J Prev Med 2010;38(2S):S292–S300
ing the benefıts of climbing one flight of stairs may con-ey false information. It may lead people to believe that aingle 30-second climb will substantially improve theirealth.”29
conomic efficiency. For this updated review, a searchf literature on economic effectiveness was conducted.o studies were found that met the requirements fornclusion in a Community Guide review.30
arriers to intervention implementation. Few studieseviewed indicated specifıc barriers to successful imple-entation of the intervention. One author reported un-uthorized removal of prompts from stairwells.13 An-ther reported that the floor on which an employeeorked affected stair use, suggesting that the more stairsne has to ascend, the less effective the interventionight be.24 Additionally, some stairwells are locked andthers may be diffıcult to fınd, poorly lit, or not wellaintained.17 Some institutions may have fıre codes andther policies restricting the placement of prompts orosters in public areas. Choice of dress (e.g., high-heeledhoes) may also serve as barriers to stair use and mayncrease general risk of using the stairs.
ummary and Discussion: Effectivenessf Point-of-Decision Promptsn general, the qualifying studies identifıed in this revieweported a low level of observed baseline use of stairs, andmall but signifıcant increases in the use of stairs follow-ng the implementation of point-of-decision prompts.lthough absolute changes were small, these differencesepresent modest relative improvements in the use oftairs. In general, the lower the level of baseline use, thereater the improvements in use. The duration of obser-ation reported in the qualifying studies was relativelyhort, with a maximum observation period of 12 weeks.he team had little evidence with which to evaluate theong-term impact of these interventions on stair use, andhere was no signifıcant association between length ofbservation periods and changes in stair use.The venue in which the prompt is placed may also
nfluence the amount of exposure. Some locations, suchs malls and airports, have populations that (with thexception of a limited number of employees) likely do noteturn from one day to the next; whereas other locations,uch as offıce buildings and commuter train stations,ikely have populations that return—and therefore arexposed to the prompts—day after day. None of thesetudies examined the impact that repeated exposure toromptsmay have on stair use—clearly an area for future
esearch. s
onclusion
ccording to Community Guide rules of evidence,7 thiseview provides strong evidence that point-of-decisionrompts contribute tomodest increases in the percentagef people choosing to take the stairs rather than an eleva-or or escalator. The observed increases in the use of stairsay contribute to a modest improvement in daily physi-al activity that would have a cumulative effect on caloricxpenditure and, in turn, energy balance.
eview of Evidence: Stair or Stairwellnhancements when Combined withoint-of-Decision Prompts
nhancement of stairs or stairwells when combined withoint-of-decision prompts was also examined as part ofhis update review. This intervention includes modifyingtairwells through one or more of the following: paintingalls, laying carpet, adding artwork, and playing music.his intervention may indirectly increase the effective-ess of point-of-decision prompts by changing attitudesbout stair use (or a particular stairwell).
ffectiveness. The team identifıed two studies17,31 thatssessed the effectiveness of stairwell enhancementshen combined with point-of-decision prompts inhanging frequency of stair use, as measured by meanumber of trips per person per day and percentage ofeople using the stairs. Both of these studies used time–eries designs, were rated as moderate in suitability, andere evaluated as being of fair execution. Details of thewo qualifying studies, including a summary of the con-ent, delivery, evaluation design, and outcomes, are avail-ble at www.thecommunityguide.org/pa/environmental-olicy/podp.html.
tudy design and implementation characteris-ics. Both studies reviewed investigated the impact ofnvironmental change on stair use. One study31 reportedlong-term evaluation during which a stairwell wasainted and carpeted, artwork was placed on the wallsf landings, point-of-decision prompts were postedhroughout the building and on the computer kiosk in theobby, and fınally, music was piped in. This interventionas implemented in stages where cumulative effects werexamined (effectiveness was evaluated after new carpetnd paint were added, and then again after adding art-ork). In the second study, the effectiveness of promptslone and the effectiveness of prompts plus adding art-ork andmusic to the stairwell were examined.17 For thistudy, the prompts-alone condition was included in theeview described above. One study was conducted in anffıce building31 and the otherwas conducted in a univer-
17
ity building. Both studies were conducted in the U.S.
Soler et al / Am J Prev Med 2010;38(2S):S292–S300 S299
F
utcomes related to stair use. There was not enoughvidence in this body of literature to draw conclusionsbout effectiveness. In the study conducted in an offıceuilding, all interventions (paint, carpet, art, signs, andusic) together led to a relative increase in stair use of.8% (baseline use:M�2.14 trips per day per occupant).31
he other study examined the effectiveness of point-of-ecision prompts with artwork andmusic and reported a9.6% relative increase in stair use (percentage of peoplesing stairs at baseline: 11.1%).17
arriers to intervention implementation. Fire codeegulations may limit or preclude enhancements to stairsnd stairwells. The qualifying studies did not providedditional information on barriers to implementation ofhese interventions.
onclusion
ccording to the Community Guide’s rules of evidence,7
here is insuffıcient evidence to determine the effective-ess of point-of-decision prompts in encouraging stairse when combinedwith stair or stairwell enhancements.wo studies of moderate suitability were identifıed. Al-hough both observed improvements in stair use over theeriod of observation (relative percentage changes of.8% in trips per person per day and 39.6% of peoplesing the stairs), more research is needed to determinehe effects of this intervention on stair use.
esearch Issues
nformational approaches to increasing physicalctivity.
ffectiveness. This review established the effectivenessf point-of-decision prompts to encourage stair use.owever, important research issues regarding the effec-iveness of these interventions remain. Many researchuestions from the fırst Community Guide review ofoint-of-decision prompts5 have been addressed inmoreecent studies. However, some questions have not beenddressed and others emerged from this update.
What effect does varying the message or format of theprompt have on providing a “booster” to stair useamong the targeted population?What type of prompt is most effective? What effectdoes format or size have, if any?Is there a “critical distance” from the elevator or esca-lator to the stairs, in which the effect of signage on stairuse is reduced?Are there a minimum or maximum number of flightsone must expect stair users to ascend in order for theprompt to be effective?
How many individuals read the point-of-decision w
ebruary 2010
prompt and react (i.e., increase their use of the stairs) asa result, as opposed to reacting to other knowledge thatthe intervention is occurring?What strategies can be used to maintain the interven-tion effect after the intervention ends? Are periodic“boosters” necessary or helpful?
conomic evaluations. The available economic dataere limited. Therefore, considerable research is war-anted on the following questions.
What is the cost effectiveness of each of these seem-ingly low-cost interventions?How can effectiveness in terms of health outcomes orquality adjusted health outcomes be better measured,estimated, or modeled?
ummaryn this article, the team reported results from an updatedeview of point-of-decision prompts that included andditional reviewof stair or stairwell enhancementswhensed with point-of-decision prompts. The inclusion ofore recent studies provides strong evidence of effective-ess of the point-of-decision prompt intervention in in-reasing the use of stairs. On average these improvementsepresent a modest improvement in stair use. Point-of-ecision prompts may represent a simple, lower-cost op-ion to increase physical activity in some settings. Thereas insuffıcient evidence to draw a conclusion regardinghe effectiveness of stair or stairwell enhancements whensed with point-of-decision prompts. Despite the inclu-ion of additional studies, there remain important gaps innderstanding of the effectiveness of these interventionsn some settings (such as worksites), and the contributionf these interventions to overall physical activity andhysical fıtness.
he team thanks the following individuals for their con-ributions to this review: Reba Norman, research librar-an; Kate W. Harris and Tony Pearson-Clarke, editors;nd the team’s Coordination Team: Nico Pronk, PhD,ealth Partners,MinneapolisMN;Dennis Richling,MD,orSolutions, Chicago IL; Deborah R. Bauer, RN, MPH,ational Center for Chronic Disease Prevention andealth Promotion, CDC, Atlanta GA; Andrew Walker,rivate Consultant, Atlanta GA; Abby Rosenthal,PH, Offıce on Smoking and Health, CDC, Atlanta GA;urtis S. Florence, II PhD, Emory University, AtlantaA; and Deborah MacLean, The Coca-Cola Company,tlanta GA.The names and affıliations of the Task Force members
Flegal KM. Prevalence of overweight and obesity in the U.S.,1999–2004. JAMA 2006;295(13):1549–55.
2. CDC. Overweight and obesity: an overview. www.cdc.gov/nccdphp/dnpa/obesity/contributing_factors.htm.
3. USDHHS. The Surgeon General’s call to action to prevent anddecrease overweight and obesity. www.surgeongeneral.gov/topics/obesity/.
4. U.S. Preventive Services Task Force. Guide to Clinical Preven-tive Services: report of theU.S. Preventive Services Task Force.2nd ed. Baltimore: Williams &Wilkins, 1996.
5. Kahn EB, Ramsey LT, Brownson RC, et al. The effectiveness ofinterventions to increase physical activity: a systematic review.Am J Prev Med 2002;22(4S):S73–107.
6. Task Force on Community Preventive Services. Recommen-dations to increase physical activity in communities. Am JPrevMed 2002;22(4S):S67–72.
7. Briss PA, Zaza S, Pappaioanou M, et al. Developing anevidence-based Guide to Community Preventive Services—methods. Am J Prev Med 2000;18(1S):S35–43.
8. Lanningham-Foster L, Nysse LJ, Levine JA. Labor saved, calo-ries lost: the energenic impact of domestic labor-saving de-vices. Obes Res 2003;11(10):1178–81.
9. Teh KC, Aziz AR. Heart rate, oxygen uptake, and energy costof ascending and descending the stairs. Med Sci Sports Exerc2002;34(4):695–9.
0. Boreham CAG, Wallace WFM, Nevill A. Training effects ofaccumulated daily stair-climbing excercise in previously sed-entary young women. Prev Med 2000;30:277–81.
2. Zaza S, Wright-De Aguero LK, Briss PA, et al. Data collectioninstrument and procedure for systematic reviews in the Guideto Community Preventive Services. Am J Prev Med 2000;18(1S):44–74.
3. Adams J,WhiteM. A systematic approach to the developmentand evaluation of an intervention promoting stair use. HealthEduc J 2002;61(3):272–86.
4. Andersen RE, Franckowiak SC, Snyder J, Bartlett SJ, FontaineKR. Can inexpensive signs encourage the use of stairs? Resultsfrom a community intervention. Ann Intern Med 1998;
129(5):363–9.
5. Andersen RE, Franckowiak SC, Zuzak KB, Cummings ES,Crespo CJ. Community intervention to encourage stair useamong African American commuters. Med Sci Sports Exerc2000;32:38.
6. Blamey A, Mutrie N, Aitchison T. Health promotion by en-couraged use of stairs. Br Med J 1995;311(7000):289–90.
7. Boutelle KN, Jeffery RW, Murray DM, Schmitz MK. Usingsigns, artwork, and music to promote stair use in a publicbuilding. Am J Public Health 2001;91(12):2004–6.
8. Brownell KD, Stunkard AJ, Albaum JM. Evaluation and mod-ifıcation of exercise patterns in the natural environment. Am JPsychiatry 1980;137(12):1540–5.
9. Coleman KJ, Gonzalez EC. Promoting stair use in a U.S.–Mexico border community. Am J Public Health 2001;91(12):2007–9.
0. Kerr J, Eves F, Carroll D. Gettingmore people on the stairs: theimpact of a new message format. J Health Psychol 2001;6(5):495–500.
1. Kerr J, Eves F, Carroll D. Encouraging stair use: stair-riserbanners are better than posters. Am J Public Health2001;91(8):1192–3.
2. Kerr J, Eves F, Carroll D. Posters can prompt less active peopleto use the stairs. J Epidemiol Community Health 2000;54(12):942–3.
3. Kerr J, Eves FF, Carroll D. The influence of poster prompts onstair use: the effects of setting, poster size and content. Br JHealth Psychol 2001;6(Pt 4):397–405.
4. Kerr J, Eves F, Carroll D. Can posters prompt stair use in aworksite environment? J Occup Health 2001;43(4):205–7.
5. Marshall AL, Bauman AE, Patch C, Wilson J, Chen J. Canmotivational signs prompt increases in incidental physical ac-tivity in an Australian healthcare facility? Health Educ Res2002;17(6):743–9.
6. Russell WD, Hutchinson J. Comparison of health promotionand deterrent prompts in increasing use of stairs over escala-tors. Percept Mot Skills 2000;91(1):55–61.
7. RussellWD,DzewaltowskiDA, RyanGJ. The effectiveness of apoint-of-decision prompt in deterring sedentary behavior.Am J Health Promot 1999;13(5):257–9.
8. Kerr J, Eves F, Carroll D. Six-month observational study ofprompted stair climbing. Prev Med 2001;33(5):422–7.
9. Reisman A, Gross CP. Increasing stair use. Ann Intern Med1999;130(7):616–7.
0. Carande-Kulis VG,MaciosekMV, Briss PA, et al. Methods forsystematic reviews of economic evaluations for the Guide toCommunity Preventive Services. Am J Prev Med 2000;18(1S):S75–91.
1. Kerr NA, Yore MM, Ham SA, Dietz WH. Increasing stair usein a worksite through environmental changes. Am J Health
