UBC Faculty Research and Publications

Development and reliability of a streetscape observation instrument for international use: MAPS-global Cain, Kelli L; Geremia, Carrie M; Conway, Terry L; Frank, Lawrence D; Chapman, James E; Fox, Eric H; Timperio, Anna; Veitch, Jenny; Van Dyck, Delfien; Verhoeven, Hannah; Reis, Rodrigo; Augusto, Alexandre; Cerin, Ester; Mellecker, Robin R; Queralt, Ana; Molina-García, Javier; Sallis, James F Feb 26, 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12966_2018_Article_650.pdf [ 404.26kB ]
JSON: 52383-1.0364061.json
JSON-LD: 52383-1.0364061-ld.json
RDF/XML (Pretty): 52383-1.0364061-rdf.xml
RDF/JSON: 52383-1.0364061-rdf.json
Turtle: 52383-1.0364061-turtle.txt
N-Triples: 52383-1.0364061-rdf-ntriples.txt
Original Record: 52383-1.0364061-source.json
Full Text

Full Text

RESEARCH Open AccessDevelopment and reliability of astreetscape observation instrument forinternational use: MAPS-globalKelli L. Cain1,12*, Carrie M. Geremia1,12, Terry L. Conway1,12, Lawrence D. Frank2,3, James E. Chapman3, Eric H. Fox3,Anna Timperio4, Jenny Veitch4, Delfien Van Dyck5,6, Hannah Verhoeven5, Rodrigo Reis7, Alexandre Augusto8,Ester Cerin9, Robin R. Mellecker9, Ana Queralt10, Javier Molina-García11 and James F. Sallis1,12AbstractBackground: Relationships between several built environment factors and physical activity and walking behaviorare well established, but internationally-comparable built environment measures are lacking. The Microscale Auditof Pedestrian Streetscapes (MAPS)-Global is an observational measure of detailed streetscape features relevant tophysical activity that was developed for international use. This study examined the inter-observer reliability of theinstrument in five countries.Methods: MAPS-Global was developed by compiling concepts and items from eight environmental measuresrelevant to walking and bicycling. Inter-rater reliability data were collected in neighborhoods selected to vary ongeographic information system (GIS)-derived macro-level walkability in five countries (Australia, Belgium, Brazil,Hong Kong-China, and Spain). MAPS-Global assessments (n = 325) were completed in person along a ≥ 0.25 mileroute from a residence toward a non-residential destination, and a commercial block was also rated for eachresidence (n = 82). Two raters in each country rated each route independently. A tiered scoring system was createdthat summarized items at multiple levels of aggregation, and positive and negative valence scores were createdbased on the expected effect on physical activity. The intraclass correlation coefficient (ICC) was computed forscales and selected items using one-way random models.Results: Overall, 86.6% of individual items and single item indicators showed excellent agreement (ICC ≥ 0.75), and13.4% showed good agreement (ICC = 0.60–0.74). All subscales and overall summary scores showed excellentagreement. Six of 123 items were too rare to compute the ICC. The median ICC for items and scales was 0.92 witha range of 0.50–1.0. Aesthetics and social characteristics showed lower ICCs than other sub-scales, but reliabilitieswere still in the excellent range (ICC ≥ 0.75).Conclusion: Evaluation of inter-observer reliability of MAPS-Global across five countries indicated all items andscales had “good” or “excellent” reliability. The results demonstrate that trained observers from multiple countrieswere able to reliably conduct observations of both residential and commercial areas with the new MAPS-Globalinstrument. Next steps are to evaluate construct validity in relation to physical activity in multiple countries andgain experience with using MAPS-Global for research and practice applications.Keywords: Exercise, Walking, Built environment, Walkability, Measurement* Correspondence: kcain@ucsd.edu1Department of Family Medicine and Public Health, University of CaliforniaSan Diego, San Diego, CA, USA12Mary MacKillop Institute for Health Research, Australian Catholic University,Melbourne, AustraliaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Cain et al. International Journal of Behavioral Nutrition and Physical Activity (2018) 15:19 https://doi.org/10.1186/s12966-018-0650-zBackgroundRelationships between several built environment factorsand physical activity are well established [1]. Neighbor-hood environment features have been classified into twobroad categories. Macroscale features include larger,structural and urban form characteristics, such as streetconnectivity, land use mix, and residential density thatare not easily modifiable [2–5]. Microscale features, orsmaller details of environments, such as sidewalk orstreet-crossing quality and aesthetics, are believed toaffect people’s confidence, comfort, and safety for walk-ing [6, 7]. In contrast to their macroscale counterparts,microscale features generally can be modified more eas-ily as part of efforts to provide more supportive environ-ments for physical activity.Numerous observational measures of microscale envi-ronments with similar content but different formats havebeen published and showed good inter-observer reliability[3, 8]. These observational instruments have been devel-oped and used across a wide range of environment types,but they are tailored to local environments. However, wecould locate no measures that were designed for inter-national use or evaluated in multiple countries. Physicalinactivity is a global health problem that is not improving[9], and built environments have been related to physicalactivity internationally [10]. Therefore, a common reliabletool designed to capture the diversity of microscale envi-ronments found across the globe would foster inter-national comparisons and generate data to informinternational initiatives such as United Nations actions toreduce non-communicable diseases [11].The purpose of the present study was to describe thedevelopment and inter-rater reliability of a streetscapeobservation tool developed for international use andevaluated in several countries. The new measure wasbased on items, format, and scoring of the MicroscaleAudit of Pedestrian Streetscapes (MAPS) that was devel-oped in the United States, with several versions shownto be related to physical activity in multiple age groups,including the original 120-item version [6], a 54-item ab-breviated version [12], and a 15-item version suitable foruse by practitioners [7]. The name of the new measureis MAPS-Global.MethodsDevelopment of MAPS-globalMAPS was originally developed as an observation toolbased on prior instruments [8, 13]. MAPS has beenshown to be a valid [6] and reliable [14] tool for survey-ing pedestrian environments and microscale urban formfeatures, with some coverage of macroscale attributessuch as land use. However, data on the validity and reli-ability of MAPS were collected in the United States only,and the tool was not designed for international use.The development of MAPS-Global was part of theInternational Physical Activity and the EnvironmentNetwork (IPEN) Adolescent study and led by the IPENCoordinating Center [15] (www.ipenproject.org). MAPS-Global was intended to be applicable for all ages, fromchildhood to older adulthood and drew from measuresdesigned for general populations and specific age groups.MAPS-Global was designed to have important physicalactivity-relevant attributes from every continent in oneinstrument to allow cross-country comparisons.To develop a version of MAPS appropriate for global use,the original MAPS and eight additional tools developed fordifferent countries and purposes were identified, and se-lected items and constructs were adapted to include inMAPS-Global: Bikeability Toolkit (Bicycle Federation ofAustralia) [16], Assessing Levels of PHysical Activity andfitness (ALPHA; Europe) [17], Environment in Asia ScanTool (EAST; Hong Kong) [18], Residential EnvironmentAssessment Tool (REAT; UK) [19], Forty Area Study StreetView tool (FASTVIEW; UK) [20], Systematic Pedestrianand Cycling Environmental Scan (SPACES; Australia) [21],Sport, Physical activity and Eating behavior: EnvironmentalDeterminants in Young people audit tool (SPEEDY; UK)[22], and International Study of Childhood Obesity, Life-style and the Environment audit tool (ISCOLE; inter-national) [23]. In addition, a self-report neighborhoodenvironment measure tailored to Africa was considered toenable the use of MAPS-Global in African environments[24, 25]. A document showing the source(s) for each itemin the MAPS Global tool can be downloaded at http://salli-s.ucsd.edu/measure_maps.html#MAPSGLOBAL [26].A draft of the MAPS-Global instrument was createdthrough a three-step revision process. First, items fromother tools that covered a similar construct as a MAPSitem were used to revise the MAPS item to reflect inter-nationally appropriate terms. Second, other modifica-tions were made to existing MAPS items to adapt tointernational settings, such as increasing the upper rangefor land uses and building heights. Third, items from alleight instruments were reviewed and considered for in-clusion if they met one of the following criteria: the itemwas found in more than one of the reviewed tools, wasconsidered policy relevant, or captured a feature uniqueto a region. As most previous tools focused on pedes-trian use, special attention was paid to incorporating abicycling component for MAPS-Global. Table 1 presentsa comparison between the original MAPS and MAPS-Global to highlight the changes.After this revision process, a draft of MAPS-Globalwas distributed to IPEN investigators from 15 countriesfor review and input. Recommendations for additionalitems were also solicited during this process. The toolwas then finalized for use in the current study and con-tained 123 items. The tool is available for download [26].Cain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 2 of 11Table 1 Summary of changes between the original MAPS and MAPS-Global for each subscaleItems deleted for MAPS Global (examples) Items added for MAPS Global Modifications for MAPS GlobalDestinations & Land Use (DLU)PositiveDLUBig box store, health/social services, library/museum, post office, senior center, parkingfacilities, retirement/senior living facility,community gardenTrail, bicycle shop, open air market, bakery Response scale increased from 2+ to 5+ asupper limit for all land uses except open airmarkets, strip malls, and shopping centers(shopping mall and shopping arcadecombined into 1 item)NegativeDLUWarehouse, casino, abandoned building,unmaintained lot/field, med-large parkinglotAge-restricted bar/nightclub NoneItems notin subscaleNon-residential buildings with parking lotsbetween walkway and entrance, non-residential buildings adjacent to walkwayPedestrian street or zone NoneStreetscape CharacteristicsPositivestreetscapePosted speed limit, drainage ditches,drinking fountain, public phonesBRT, train, subway, tram/streetcar, tuktuk/auto rickshaw, bicycle share, hawkers/shops/cartsModified number of driveways to becollected at the segment levelNegativestreetscapeDriveways/alleys, lack of street lights None NoneItems notin subscaleSenior transit, drainage ditches None NoneAesthetics & Social CharacteristicsPositiveaesthetics/socialNeighborhood watch signs, commercialsignageNatural bodies of water NoneNegativeaesthetics/socialRailroad tracks, extent social disorder,abandoned cars,Dog fouling Modified extent of graffiti and litter toinclude “dog fouling”; combined litter inyards and on street/sidewalk into 1 itemItems notin subscaleHistoric/cultural features, Beer/liquorbottles/cansNone NoneCrossings/IntersectionsPositivecrossingStop lines on road, one-way streetsthrough crossing, audible walk signal, dedi-cated turn arrowsCrossing on an overpass, underpass orbridge, bicycle signal, tactile paving, bikeboxNoneNegativecrossingGutters, steep slope, temporaryobstructions, poor visibility at cornersNone NoneItems notin subscaleCrosswalk timing, number of legs atintersection, poor condition of crossingsurfaceNone NoneStreet SegmentsPositivesegmentBuilding façade colors, building accentcolors, building materialsHawkers or shops, sidewalk or pedestrianstreet/zone obstructions, pedestrian bridge/overpass/tunnel, covered or air conditionedplace to walk along the street or connectingbuildings, quality of the bicycle lane/ zoneModified number of traffic lanes to includepedestrian street; modified coverage ofsidewalk/walkway to isolate trees (1 item)and overhangs/awnings (1 item); modifiedbuilding height to collect shortest (1 item)and tallest (1 item) and upper limitincreased from 11+ to 21+ stories; modifiedbicycle lane/zone with different options;modified street lights to specify for cars (1item) and for pedestrians (1 item)NegativesegmentCross-slope, minor trip hazards, how muchof segment at steepest levelGates/walls/tall fences around properties Modified sidewalk or pedestrian zoneobstructions to isolate cars; slope estimatedby sight rather than measured with aninclinometerCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 3 of 11Study design and citiesGlobal microscale environmental data were collected forthe purpose of this reliability study from November2014 – June 2015 as part of the IPEN Adolescent study,and included five cities: Melbourne (Australia), Ghent(Belgium), Curitiba (Brazil), Hong Kong (China), andValencia (Spain). Following IPEN protocol, all cities useda GIS-derived macro-level walkability index to selectneighborhoods defined as high versus low walkability,based on: net residential density, intersection density, andmixed land use [27, 28]. Neighborhoods in these citieswere selected to represent four neighborhood types cate-gorized as high/low-walkability by high/low-median socio-economic status (SES), to ensure the inclusion of a widerange of demographic and built environment attributes.The IPEN Adolescent study was approved for researchwith human subjects by the Institutional Review Boardsat Deakin University, Ghent University, Pontifical Cath-olic University of Parana, University of Hong Kong, andUniversity of Valencia.Route selectionTo identify routes for MAPS-Global assessment, eachcountry randomly selected 65 IPEN Adolescent studyparticipants, or randomly selected residences within po-tential study areas, (total n = 325) stratified by the fourwalkability-by-SES neighborhood types. The IPEN Co-ordinating Center identified each route’s destination asthe nearest commercial block. Routes were manuallycreated (0.25–0.45 mile (400–724 m) in distance) fromeach residence toward a commercial block using GoogleEarth. The routes were drawn along the road network,providing the most direct route from the residence towarda non-residential destination. Alleys, non-motorized, andinformal paths adjacent to the street network were noteasily identifiable using online images and were thereforenot used to create routes. However, these pedestrian facil-ities were coded within MAPS-Global when they were ob-served. MAPS-Global data were also collected along asingle road segment at the nearest commercial block toenhance the variety of environmental features assessed, asthe 0.25–0.45 mile routes did not always reach the enddestination, due to a cap on the maximum surveyed dis-tance (based on time and budget considerations).TrainingA research staff manager from the IPEN CoordinatingCenter was responsible for training, route creation, andquality control. Details about length of training and cer-tification can be found elsewhere [14]. Multiple raters ateach study site were instructed to use MAPS-Globalthrough an online webinar and were provided trainingmaterials including a manual with item definitions andphotos (see training manual online [26]). After the on-line training, each country’s team practiced rating streetsin the field and communicated with the IPEN Coordin-ating Center to clarify site-specific issues. To be certifiedto rate independently, raters were required to completeobservations of at least five routes with inter-rater reli-ability at 95% agreement or higher.Data collectionData were collected along a 0.25–0.45 mile route (n = 325residential routes) starting at a study participant’s home ora randomly selected residence and walking toward thenearest commercial destination. Data were also collectedalong 82 commercial blocks. Table 2 describes data collec-tion areas and sample sizes per country.Two raters in each country completed each MAPS-Global route independently. Residential routes took onaverage 26.1 min to complete (range = 2–100 min) andcommercial segments were completed in 15.8 min onaverage (range = 3–110 min). Raters and coordinatorsreviewed each tool for missing and discrepant items. Ifmore than 5% of items were missing, raters returned tothe route and completed the missing items.Table 1 Summary of changes between the original MAPS and MAPS-Global for each subscale (Continued)Items deleted for MAPS Global (examples) Items added for MAPS Global Modifications for MAPS GlobalItems notin subscaleSigns to discourage skateboard use, dead-end streetType of segment: residential or commercial NoneCul-De-SacsDiameter of cul-de-sac, slope, condition ofpavement, island, parkingSoccer goals, outdoor fitness equipment, NoneTable 2 Reliability pair sample sizes by countryCountry City Routesa Segmentsbc Crossingsc Cul-de-sacsAustralia Melbourne 65 218 108 10Belgium Ghent 65 236 156 6Brazil Curitiba 65 319 213 0China Hong Kong 65 224 145 9Spain Valencia 65 350 266 0Total 325 1347 888 25aresidential only, commercial blocks not includedbsegment defined as the area between crossingscboth residential and commercial includedCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 4 of 11Scoring and creating subscalesThe scoring of MAPS-Global largely followed the ori-ginal MAPS scoring structure which has been describedelsewhere [6, 14]. Briefly, the tool has six sections: desti-nations and land use (DLU), streetscapes, aestheticsand social, street segments (defined as the area betweenstreet crossings), street crossings, and cul-de-sacs/dead-ends. DLU, streetscape, and aesthetics/social items werecaptured at the route level, and these characteristicswere generally consistent throughout the route (e.g.,speed limit, aesthetics and social environment). Streetsegment variables, such as sidewalks, buffers betweenstreets and walking spaces, trees, and building setbackswere collected on each segment on the route. Street-crossing variables were measured at every intersectionor crossing on the route (e.g., crosswalks, signals). Cul-de-sac variables (e.g., size, amenities) were collectedwhen one or more cul-de-sacs or dead-ends werepresent within 400 ft (122 m) of the residential address.When multiple segments and crossings occurred alonga route, the respective segment and crossing variableswere averaged.This tiered scoring system summarized items into sub-scales at multiple levels of aggregation. Most sections in-cluded positive and negative valence scores based on theexpected effect on physical activity. Some items were ex-cluded from subscales due to being transitory (e.g., pres-ence of anyone walking), capturing a particularly importantelement of the environment (e.g., pedestrian street), or anunclear expected association with physical activity (e.g., seg-ment type). These became single-item indicators.A modification from the original MAPS was made toadapt to the more destination-dense environmentsfound internationally by increasing the upper range ofland use frequency response options to five or more foreach type of destination (only “two or more” was used inoriginal MAPS). Land use items were scored as 0, 1, 2,3, 4, or 5+. Other continuous and descriptive items weredichotomized or trichotomized based on their distribu-tions, theoretical relevance, and compatibility with otherscale items’ scoring. In several instances, related itemsneeded to be combined into single variables to be mean-ingful components of their respective subscales. For ex-ample, shortest and tallest building heights werecollected as two separate items, but for scoring theywere averaged into one variable for the subscale. In suchcases, the new variable was computed and then recodedfor scoring (e.g., di- or trichotomized) consistently withtheoretically related items to match scoring of otheritems within a subscale.After items were rescored as necessary, subscale scoreswere computed by summing the items’ scores. Valencescores were created by summing subscales that were ex-pected to have a positive or negative impact on physicalactivity based on the consensus of authors familiar withinterdisciplinary research, conceptual models, and guide-lines. For instance, the sum of the positive destinationsand land uses was thought to be positively associatedwith physical activity, and the presence of social disorderwas thought to be negatively associated. All of the posi-tive subscales within a section were summed to createthe positive valence score, and the negative subscaleswere summed for the negative valence score. The street-scape and cul-de-sac sections only contained positivelyrelated items. Finally, an overall section score (positive-minus-negative valence scores) was calculated for eachmain section that contained both of these valence scores.Overall valence scores were calculated by summing thesix main sections’ positive and negative scores. The over-all grand score was calculated by subtracting overallnegative from overall positive scores. The cul-de-sacscore was not included in overall valence scores due toan unclear expected association with physical activity.In addition to section-derived subscales, three newsubscales were created from items that were conceptu-ally related but collected within different sections of thetool (e.g., route and segment items). The three new sub-scales were pedestrian infrastructure, pedestrian design,and bicycle facilities. Detailed information about itemrecodes, transformations, and subscale creation can bedownloaded [26].AnalysisThe purpose of MAPS-Global was to represent the fullinternational variability in environments, so reliability re-sults were computed on the pooled international dataset.Country-specific reliability estimates would be misleadingbecause different attributes would be rare in each country,leading to reduced variability and low frequency of occur-rence of variables that would underestimate reliability. Toassess inter-rater reliability, the intraclass correlation coef-ficient (ICC) was calculated for the MAPS-Global com-puted scales and several single-item indicators (e.g., placeof worship, crossing overpass, etc.). IBM SPSS Version 21Scale/Reliability procedure was used to compute ICCsusing the one-way random model for average measures.A variety of numeric definitions and adjectival descrip-tors have been used to classify measures of inter-rateragreement using Cohen’s kappa coefficients for categor-ical variables and the ICC for test-retest of continuousmeasures [29–31]. For this study, Cicchetti’s [30] nu-meric ranges and descriptors were used. The ICC wasclassified to indicate test-retest reliability that was: ‘ex-cellent’ (ICC ≥ 0.75), ‘good’ (0.60–0.74), ‘fair’ (0.40–0.59),and ‘poor’ (< 0.40). Items with insufficient variability butpercentage agreement equal or higher than 75% wereconsidered to have good agreement [21].Cain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 5 of 11ResultsResults presented here were based on pooled analysesfor all five study sites. Table 3 summarizes reliabilityclassification levels for individual items that went intoscales, single-item indicators, subscales, and overallscores. Using Cicchetti’s criteria [30], 100% of the sub-scales and overall scores showed “excellent” agreement.Of the 112 individual items and single item indicatorsfor which ICCs or Kappa’s could be computed, 97(86.6%) had “excellent” reliability, and 15 (13.4%) had“good” reliability. Six of the tool’s 123 items (unantici-pated mid-segment crossing, bicycle locker or com-pound, basketball hoop in cul-de-sac, skateboard featurein cul-de-sac, soccer goal in cul-de-sac, and outdoor fit-ness equipment in cul-de-sac) were so rare that no ICCor Kappa could be calculated, yet all were retained inthe instrument due to their theoretical importance. Twoof the “good” agreement individual items (private out-door recreation and raised crosswalk) and two of the“good” agreement single item indicators (liquor/alcoholstore and presence of people walking) had relatively lowKappa’s (0.50–0.59) due to insufficient variability, buthad inter-rater agreements from 94.1%–99.9% so werecategorized as having “good” agreement [21].Table 4 provides more detailed results for the keyMAPS-Global constructs, including the number of itemsin subscales, range of scores, items and overall subscaledescriptions, and ICC’s/Kappa’s for single-item indica-tors, subscale, valence, and overall scores. The medianICC was 0.92, with a range of 0.50–1.0. Aesthetics andsocial characteristics showed lower ICC values thanother sections. Liquor/alcohol stores had the lowestICC, and crosswalk amenities had the highest. The ICCfor the overall grand score was 0.99.DiscussionTo facilitate international comparison of microscale envi-ronments relevant to physical activity, a new observationalmeasure (MAPS-Global) was developed by drawing onthe previously validated MAPS tool and eight other in-struments developed in and for a diverse set of countries.Evaluation of inter-observer reliability of MAPS-Global infive countries indicated all items and scales had “good” or“excellent” agreement. All of the summary scores had “ex-cellent” reliability, with an ICC of > 0.75, and the ICC forthe overall grand score was 0.99. The lowest reliabilitiesfor multi-item scales were for the three aesthetics and so-cial characteristics subscales (ICCs = 0.78 to 0.81), thoughthey were still in the “excellent” category. Items dealingwith landscaping, water features, dog excrement, andhighway nearby may be more difficult to define and re-quire more subjective judgment than other types of items.In general, the results demonstrated that trained observersfrom multiple countries were able to reliably conduct ob-servations of both residential and commercial areas withthe new MAPS-Global instrument.The development process of MAPS-Global was guidedby two considerations. The first was to ensure internationalapplicability by including items relevant to physical activityon every inhabited continent. This was accomplished by in-cluding items from environmental measures developed inAfrica, the Americas, Asia, Australia, and Europe, as well asadding a bicycling environment subscale. Modificationswere also made to existing MAPS items and responsescales to capture a wider range of environments. Table 1summarizes these modifications. IPEN investigators from15 countries then reviewed, pilot tested, and providedfeedback to ensure MAPS-Global would be applicablein their countries. The second consideration was to en-sure comparability of measurement across countries.This was accomplished by producing a single instru-ment supported by a detailed and illustrated instructionmanual, delivering training from a central site, and re-quiring observers to complete an in-field certificationprocess. Although MAPS-Global does not include allpossible activity-relevant streetscape features, the in-cluded items were deemed most important by consen-sus of the IPEN Adolescent investigators. Though theTable 3 MAPS-Global ICC/Kappa reliability classifications of individual items, single-item indicators, computed scales, overall scores,and totalICC/Kappa Classifications“Excellent” agreement “Good” agreement(ICCs ≥0.75) (ICC = 0.60 to 0.74)Median (range) N (%) N (%) Total N (%)Individual items that make up scales 0.91 (0.50–1.0) 80 (87.0) 12 (13.0) 92a (100.0)Single-item indicators 0.88 (0.54–0.99) 17 (85.0) 3 (15.0) 20b (100.0)Subcales (sums of items) 0.94 (0.79–0.99) 21 (100.0) 0 (0.0) 21 (100.0)Overalls (sums of subscales) 0.96 (0.78–0.99) 13 (100.0) 0 (0.0) 13 (100.0)Total (items + single item indicators + subscales + overalls) 0.92 (0.50–1.0) 131 (89.7) 15 (10.3) 146 (100.0)a96 total individual items, but 4 items were too rate to compute Kappa’sb22 total single item indicators, but 2 items were too rare to compute Kappa’sCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 6 of 11Table 4 Inter-rater Reliability for MAPS-Global Single Item Indicators, Subscales and Overall Summary Scores# items(range of scores)Sample items and overall subscale description ICC/Kappa, Confidence IntervalDestinations & Land Use (DLU)Positive Destinations & Land UseResidential Mix 4 (0–3) Single family, multi-family, mixed, apartment over retail .83 (.79, .85)Shops 8 (0–28) Grocery, convenience store, bakery, drugstore, other retail,shopping mall, strip mall, open air market.97 (.96, .97)Restaurant-Entertainment 4 (0–20) Fast food, sit-down, café, entertainment .90 (.88, .92)Institutional-Service 3 (0–15) Bank, health-related professional, other service .94 (.92, .95)Worship 1 (0–5) Place of worship .91 (.89, .92)School 1 (0–5) School land use .98 (.97, .98)Public Recreation 4 (0–20) Public indoor, public outdoor facility, park, trail .83 (.80, .86)Private Recreation 2 (0–10) Private indoor, private outdoor facility .84 (.81, .87)Pedestrian Streeta 1 (0–5) Pedestrian street/zone .89 (.87, .91)Negative Destinations & Land UseAge-restricted bar or nightclub 1 (0–5) Age-restricted bar/nightclub .93 (.91, .94)Liquor or alcohol store 1 (0–5) Liquor or alcohol store .54 (.47, .61)Positive DLU 27 (0–111) Sum of the positive DLU subscales .96 (.95, .97)Negative DLU 2 (0–10) Sum of the negative DLU subscales .92 (.90, .93)Overall DLU 29b Positive DLU - Negative DLU .96 (.96, .97)Streetscape CharacteristicsTransit 9 (0–13) Number of stops, transit type and amenities(bench, shelter, and timetable).90 (.89, .92)Traffic calming 1 (0–5) Signs, circles, speed tables, speed humps, curb extension .90 (.88, .91)Roll-over curba 1 (0–1) Roll-over curbs .84 (.81, .87)Trash bins 1 (0–1) Public trash bins .89 (.87, .91)Benches 1 (0–1) Benches or other places to sit .81 (.78, .84)Bicycle racks 1 (0–1) Bicycle racks .83 (.80, .86)Bicyle lockers 1 (0–1) Secure bicycle access lockers or compounds Too rare to calculate KappaBicycle sharing 1 (0–1) Bicycle docking station for bike sharing .97 (.96, .98)Kiosks 1 (0–1) Kiosks or information booths .97 (.96, .98)Hawkers 1 (0–1) Hawkers/shops/carts .97 (.96, .97)Positive Streetscape 17c (0–25) Sum of positive streetscape .93 (.91, .94)Aesthetics & Social CharacteristicsPresence of anyone walkinga 1 (0–1) Presence of anyone walking .59 (.53, .65)Positive Aesthetics/Social 4 (0–4) Hardscape, water, softscape, landscaping .78 (.74, .82)Negative Aesthetics/Social 6 (0–6) Buildings not maintained, graffiti, litter, dog fouling,physical disorder, highway near.80 (.76, .83)Overall Aesthetics/Social 10d Positive Aesthetics/Social - Negative Aesthetics/Social .81 (.77, .84)Crossings/IntersectionsPositive Crossing SubscalesCrosswalk Amenities 7 (0–7) Crossing aids, marked crosswalk, high visibility striping,different material, curb extension, raised crosswalk,refuge islands.99 (.99, .99)Curb Quality & Presence 3 (0–6) Curb presence, curb ramps lined up, tactile paving .95 (.94, .95)Intersection Control & Signage 7 (0–7) Yield signs, stop signs, traffic signal, traffic circle,pedestrian walk signals, push buttons, countdown signal.97 (.96, .97)Bicycle Features 3 (0–3) Waiting area, bike lane crossing the crossing, bike signal .94 (.93, .95)Overpass 1 (0–1) Crossing on pedestrian overpass, bridge .80 (.77, .82)Mid-segment crossinga 1 (0–1) Unanticipated mid-segment crossing Too rare to calculate KappaCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 7 of 11Table 4 Inter-rater Reliability for MAPS-Global Single Item Indicators, Subscales and Overall Summary Scores (Continued)# items(range of scores)Sample items and overall subscale description ICC/Kappa, Confidence IntervalNegative Crossing SubscalesRoad Width 1 (0–2) Distance of crossing leg .99 (.98, .99)Positive Crossing 21 (0–24) Sum of the positive crossing subscales .98 (.98, .98)Negative Crossing 1 (0–2) Sum of the negative crossing subscales .99 (.98, .98)Overall Crossing 22e Positive Crossing - Negative Crossing .98 (.97, .98)Street SegmentsPositive Segment SubscalesBuilding Height-Setback 4 (0–10) Building height, smallest and largest setback .97 (.97, .97)Segment typea 1 (0–1) Segment type: residential or commercial .96 (.96, .97)Building Height-Road Width Ratio 5 (0–3) Building height, setback and road width .80 (.78, .82)Buffer 2 (0–5) Parking along street, buffer .92 (.91, .92)Bike Infrastructure 3 (0–5) Bike lane presence, quality, signage .95 (.94, .96)Shade 3 (0–6) Number of trees, sidewalk coverage, shade .93 (.93, .94)Sidewalk 2 (0–6) Sidewalk presence and width .93 (.92, .94)Pedestrian infrastructure 5 (0–5) Mid-segment crossing, pedestrian bridge, coveredplace to walk, street lights.93 (.92, .93)Building Aesthetics and Design 1 (0–2) Street windows .84 (.82, .85)Informal Path or Shortcut 1 (0–1) Informal path connecting to something else .86 (.84, .87)Hawkers/Shops 1 (0–2) Hawkers/shops on sidewalk/ped zone .73 (.71, .76)Negative Segment SubscalesSidewalk 7 (0–13) Non-continuous sidewalk, trip hazards, obstructions,cars blocking walkway, slope, gates, driveways.96 (.95, .96)Positive Segment 27 (0–45) Sum of the positive segment subscales .97 (.97, .98)Negative Segment 7 (0–13) Sum of the negative segment subscales .96 (.95, .96)Overall Segment 34f Positive Segment - Negative Segment .98 (.98, .98)Overall Summary and Grand ScoresOverall Positive 102 (0–210) Positive DLU, positive streetscape, positive aesthetics/social,positive segment (mean of all segments), positive crossing(mean of all segments)..96 (.95, .97)Overall Negative 16 (0–22) Negative DLU, negative aesthetics/social, negativesegment (mean of all segments), negative crossing(mean of all crossings)..92 (.90, .93)Overall Grand Score 118 Overall Positive - Overall Negative .99 (.99, .99)Cross-Domain SubscalesPedestrian Infrastructure 13 (0–27) Trail, pedestrian zone, sidewalk presence/width,buffer, shortcut, mid-segment crossing, pedestrianbridge, air conditioned place to walk, low lights,overpass, crosswalk, refuge island.96 (.95–.97)Pedestrian Design 13 (0–21) Open-air market, trash cans, benches, kiosks, hawkersand shops, setback, visibility, pedestrian walk signals,push buttons, countdown signals, ramps, crossing aids.98 (.97–.98)Bicycle Facilities 9 (0–11) Bike racks, docking stations, lockers, bike lane, bikelane quality, signs, bike signal, bike box, bike lanecrossing the crossing.97 (.96–.98)Cul-de-sacs/Dead-endsOverall Cul-de-sacg 6 (0–6) Closeness to participant’s home, total amenities, visibility .94 (.88, .97)anot included in subscaleb31 items in section, bicycle shops added to tool later, pedestrian zone not included in subscalesc22 items in this section, 4 new informal transit items added roll over curbs not included in subscalesd11 items in this section, presence of people walking not included in subscalese23 items in this section, mid- segment crossing not included in subscalesf30 unique items used in subscales, but 5 items (setback × 2, building height × 2, and sidewalk) were scored in more than one way for different subscales,segment type not included in subscalesgscore reported is based on 2 items as 4 items were too rare to calculate KappaCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 8 of 11instrument was developed as part of a study of adolescents,MAPS-Global was designed to be relevant to all ages.Strengths of the study were the wide variety of con-structs, clear scoring guidelines and training procedures,conceptually meaningful summary variables to use inanalyses, and good evidence of inter-observer reliabilitydocumented in the present paper. Weaknesses of themeasure and the study included the large number ofitems and need for training and ongoing supervision ofobservers that add to the costs and investigator burdenof data collection. Although MAPS-Global is conceptu-alized as a measure of microscale attributes, it also in-cludes variables such as land use that can be consideredmacroscale. The present method of assessing routesfrom residences toward destinations is not applicable forall purposes, such as evaluating microscale features foran entire neighborhood. However, a protocol has beendeveloped [7] for using MAPS-Global on all or selectedstreet segments by coding “route” items for each seg-ment. Although MAPS-Global was tested in five diversecountries, it has not been examined in low-incomecountries that may have distinct environmental featuresor rural areas where MAPS-Global may not be applic-able. Future studies using the MAPS-Global tool should in-clude study sites from even more diverse locations,especially low-income countries, to further assess inter-national comparability. Variability in frequency of occur-rence of items within countries reduced sample sizes andprecluded the presentation of country-specific reliabilityanalyses. Additional refinements may be needed to improvethe reliability performance among some of the items thatrequire subjective judgment in future iterations.ConclusionIt is important to improve understanding of how citiescan be built to support sufficient physical activity andother health indicators [32]. Microscale environmentdata are lacking internationally, so MAPS-Global prom-ises to fill a critical gap by providing measures of fea-tures such as sidewalks, safety of street crossings, andlandscaping that are more feasible and affordable tomodify than the macroscale layout of cities. Next stepsin the evaluation and application of MAPS-Global in-clude examining associations with physical activity (i.e.,construct validity), evaluating use of online imagery tofacilitate more efficient and cost effective data acquisi-tion, constructing more comprehensive observer trainingprograms, and eventually creating a shorter version ofthe instrument to encourage more widespread inter-national use. If MAPS-Global is shown to be valid andcomparable across countries, it could also be applied toprovide evidence for practice and policy, such as identi-fying strengths and weaknesses of activity-supportive en-vironments within and across cities to inform planningdecisions, and evaluating changes in built environments,especially those designed to improve physical activityand health.AbbreviationsALPHA: Assessing Levels of PHysical Activity and fitness; DLU: Destinationsand land use; EAST: Environment in Asia Scan Tool; FASTVIEW: Forty AreaStudy Street View; GIS: Geographic Information Systems; ICC: Intraclasscorrelations; IPEN: International Physical Activity and the EnvironmentNetwork; ISCOLE: International Study of Childhood Obesity, Lifestyle and theEnvironment; MAPS: Microscale Audit of Pedestrian Streetscapes;REAT: Residential Environment Assessment Tool; tool.; SES: Socio-economicstatus.; SPACES: Systematic Pedestrian and Cycling Environmental Scan.;SPEEDY: Sport, Physical activity and Eating behavior: EnvironmentalDeterminants in Young people.AcknowledgementsWe would like to acknowledge Casper Zhang, Kiko Leung, Ryan Lip, JoSalmon, Billie Giles-Corti and Karen Villanueva who provided feedback on thedevelopment of the MAPS Global tool.FundingFunding for this study was made possible by a grant from National Institutesof Health (NIH) R01 HL111378. EC is supported by an Australian ResearchCouncil Future Fellowship (FT #140100085). AT was supported by a NationalHeart Foundation of Australia Future Leader Fellowship (Award 100046). JVwas supported by a National Health and Medical Research Council EarlyCareer Fellowship (ID 1053426).Availability of data and materialsThe datasets used and/or analysed during the current study are availablefrom the corresponding author on reasonable request.Authors’ contributionsKLC conceived of study, participated in study design and coordination,drafted the manuscript and approved the final manuscript as submitted.CMG participated in study design and coordination, conducted analyses,drafted the manuscript and approved the final manuscript as submitted. TLCconceived of study, participated in study design, drafted the manuscript andapproved the final manuscript as submitted. LDF participated in studydesign, contributed to the manuscript review and approved the finalmanuscript as submitted. JEC participated in study design, contributed tothe manuscript review and approved the final manuscript as submitted. EHFparticipated in study design, contributed to the manuscript review andapproved the final manuscript as submitted. AT contributed to datacollection, reviewed and provided feedback to manuscript, and approvedthe final manuscript as submitted. JV contributed to data collection,reviewed and provided feedback to manuscript, and approved the finalmanuscript as submitted. DVD contributed to data collection, reviewed andprovided feedback to manuscript, and approved the final manuscript assubmitted. HV contributed to data collection, reviewed and providedfeedback to manuscript, and approved the final manuscript as submitted. RRcontributed to data collection, reviewed and provided feedback tomanuscript, and approved the final manuscript as submitted. AA contributedto data collection, reviewed and provided feedback to manuscript, andapproved the final manuscript as submitted. EC contributed to datacollection, reviewed and provided feedback to manuscript, and approvedthe final manuscript as submitted. RRM contributed to data collection,reviewed and provided feedback to manuscript, and approved the finalmanuscript as submitted. AQ contributed to data collection, reviewed andprovided feedback to manuscript, and approved the final manuscript assubmitted. JMG contributed to data collection, reviewed and providedfeedback to manuscript, and approved the final manuscript as submitted. JFSconceived of study, participated in the study design and coordination,drafted the manuscript and approved the final manuscript as submitted.Ethics approval and consent to participateAll investigators completed the San Diego State University InstitutionalReview Board training, the National Institutes of Health (NIH) FogartyInternational Center ethical requirements, and their own country’s ethicsCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 9 of 11requirements. All participants provided informed consent for participation intheir country-level study.Consent for publicationNot applicable.Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Department of Family Medicine and Public Health, University of CaliforniaSan Diego, San Diego, CA, USA. 2Health and Community Design Lab, Schoolsof Population and Public Health and Community and Regional Planning,University of British Columbia, Vancouver, BC, Canada. 3Urban Design 4Health, Inc., Rochester, NY, USA. 4Institute for Physical Activity & Nutrition,School of Exercise & Nutrition Sciences, Deakin University, Geelong, Australia.5Department of Movement and Sports Sciences, Ghent University, Ghent,Belgium. 6Research Foundation Flanders (FWO), Brussels, Belgium.7Prevention Research Center, Brown School, Washington University in St.Louis, St. Louis, USA. 8Research Group on Physical Activity and Quality of Life,Pontificia Universidade Catolica do Parana, Curitiba, Brazil. 9School of PublicHealth, University of Hong Kong, Hong Kong, China. 10Department ofNursing, University of Valencia, Valencia, Spain. 11Deparment of Teaching ofMusical, Visual and Corporal Expression, University of Valencia, Valencia,Spain. 12Mary MacKillop Institute for Health Research, Australian CatholicUniversity, Melbourne, Australia.Received: 19 September 2017 Accepted: 29 January 2018References1. Bauman AE, Reis RS, Sallis JF, Wells JC, Loos RJ, Martin BW. Correlates ofphysical activity: why are some people physically active and others not?Lancet. 2012;380(9838):258–71.2. Brennan Ramirez LK, Hoehner CM, Brownson RC, et al. Indicators of activity-friendly communities: an evidence-based consensus process. Am J PrevMed. 2006;31(6):515–24.3. Brownson RC, Hoehner CM, Day K, Forsyth A, Measuring SJF. The builtenvironment for physical activity: state of the science. Am J Prev Med. 2009;36(Suppl 4):99–123.4. Frank LD, Schmid TL, Sallis JF, Chapman JE, Saelens BE. Linking objectivephysical activity data with objective measures of urban form. Am J PrevMed. 2005;28(Suppl 2):117–25.5. Saelens BE, Handy SL. Built environment correlates of walking: a review.Med Sci Sports Exerc. 2008;40(Suppl 7):550–66.6. Cain KL, Millstein RA, Sallis JF, Conway TL, Gavand KA, Frank LD, Saelens BE,Geremia CM, Chapman J, Adams MA, Glanz K, King AC. Contribution ofstreetscape audits to explanation of physical activity in four age groupsbased on the microscale audit of pedestrian streetscapes (MAPS). Soc SciMed. 2014;116:82–92.7. Sallis JF, Cain KL, Conway TL, Gavand KA, Millstein RA, Geremia CM,Frank LD, Saelens BE, Glanz K, King AC. Is your neighborhood designedto support physical activity? A Brief Streetscape Audit Tool Prev ChronicDis. 2015;12:E141.8. Brownson RC, Hoehner CM, Brennan LK, Cook RA, Elliott MB, McMullen KM.Reliability of two instruments for auditing the environment for physicalactivity. J Phys Act Health. 2004;1:189–207.9. Sallis JF, Bull FC, Guthold R, Heath GW, Inoue S, Oyeyemi AL, Perez LG,Richards J, Hallal P. For the lancet physical activity series 2 executivecommittee. Progress in physical activity over the Olympic quadrennium.Lancet. 2016;388:1325–36.10. Sallis JF, Cerin E, Conway TL, Adams MA, Frank LD, Pratt M, Salvo D,Schipperijn J, Smith G, Cain KL, Davey R, Kerr J, Lai P-C, Mitas J, Reis R,Sarmiento OL, Schofield G, Troelsen J, Van Dyck D, De Bourdeaudhuij I,Owen N. Urban environments in 14 cities worldwide are related to physicalactivity. Lancet. 2016;387(10034):2207–17.11. United Nations. Sixty-seventh session Political Declaration of the High Levelmeeting of the General Assembly on the Prevention and control of non-communicable diseases. http://www.un.org/ga/search/view_doc.asp?symbol=A/66/L.1. Accessed 28 August 2017.12. Cain KL, Gavand KA, Conway TL, Geremia CM, Millstein RA, Frank LD,Saelens BE, Adams MA, Glanz K, King AC, Sallis JF. Developing andvalidating an abbreviated version of the microscale audit for pedestrianstreetscapes (MAPS-abbreviated). J Transp Health. 2017;5:84–96.13. Kealey M, Kruger J, Hunter R, Ivey S, Satariano W, Bayles C, Ramirez B, BryantL, Johnson C, Lee C, Levinger D, Mctigue K, Moni C, Moudon AV, Pluto D,Prohaska T, Sible C, Tindal S, Wilcox S, Winters K, Engaging WK. Older adultsto be more active where they live: audit tool development. Prev ChronicDis. 2005;2(2) http://www.cdc.gov/pcd/issues/2005/apr/04_0142q.htm.Accessed August 28 201714. Millstein RA, Cain KL, Sallis JF, Conway TL, Geremia C, Frank LD, Chapman J,Van Dyck D, Dipzinski L, Kerr J, Glanz K, Saelens BE. Development, scoring,and reliability of the microscale audit of pedestrian streetscapes (MAPS).BMC Public Health. 2013;13:403.15. International Physical Activity and Enviornment Network. http://www.ipenproject.org. Accessed 28 August 2017.16. Bicycle Federation of Australia (BFA) for the Australian Greenhouse Office inthe Department of the Environment and Heritage, with the endorsement ofthe Australian Bicycle Council. http://www.travelsmart.gov.au/bikeability/index.html. Accessed 10 August 2017.17. Spittaels H, Verloigne M, Gidlow C, Gloanec J, Titze S, Foster C, Oppert JM,Rutter H, Oja P, Sjöström M, De Bourdeaudhuij I. Measuring physical activity-related environmental factors: reliability and predictive validity of theEuropean environmental questionnaire ALPHA. Int J Behav Nutr Phys Act.2010;7:48.18. Cerin E, Chan KW, Macfarlane DJ, Lee KY, Lai PC. Objective assessment ofwalking environments in ultra-dense cities: development and reliability ofthe environment in Asia scan tool–Hong Kong version (EAST-HK). HealthPlace. 2011;17(4):937–45.19. Dunstan F, Weaver N, Araya R, Bell T, Lannon S, Lewis G, Patterson J, Thomas H,Jones P, Palmer S. An observation tool to assist with the assessment of urbanresidential environments. J Environ Psychol. 2005;25(3):293–305.20. Griew P, Hillsdon M, Foster C, Coombes E, Jones A, Wilkinson P. Developingand testing a street audit tool using Google street view to measureenvironmental supportiveness for physical activity. Int J Behav Nutr PhysAct. 2013;10:103.21. Pikora TJ, Bull FCL, Jamrozik K, Knuiman M, Giles-Corti B, Donovan RJ.Developing a reliable audit instrument to measure the physicalenvironment for physical activity. Am J Prev Med. 2002;23(3):187–94.22. Jones NR, Jones A, van Sluijs EM, Panter J, Harrison F, Griffin SJ. Schoolenvironments and physical activity: the development and testing of anaudit tool. Health Place. 2010;16:776–83.23. Katzmarzyk PT, Barreira TV, Broyles ST, Champagne CM, Chaput J-P,Fogelholm M, et al. The international study of childhood obesity, lifestyleand the environment (ISCOLE): design and methods. BMC Public Health.2013;13:900.24. Oyeyemi AL, Conway TL, Adedoyin RA, Akinroye KK, Aryeetey R, Assah F,Cain KL, Gavand K, Kasoma SS, Kolbe-Alexander TL, Lambert EV, Larouche R,Moss SJ, Ocansey R, Onywera VO, Prista A, Tremblay MS, Sallis JF. ConstructValidity of the neighborhood environment walkability scale for Africa. MedSci Sports Exerc. 2017;49(3):482–91.25. Oyeyemi AL, Kasoma SS, Onywera VO, Assah F, Assah F, Adedoyin RA,Conway TL, Moss SJ, Ocansey R, Kolbe-Alexander TL, Akinroye KK, Prista A,Larouche R, Gavand KA, Cain KL, Lambert EV, Aryeetey R, Bartels C, TremblayMS, Sallis JFNEWS. For Africa: adaptation and reliability of a builtenvironment questionnaire for physical activity in seven African countries.Int J Behav Nutr Phys Act. 2016;13:33.26. MAPS-Global webpage for document downloads: http://sallis.ucsd.edu/measure_maps.html#MAPSGLOBAL. Accessed September 6, 2017.27. Frank LD, Sallis JF, Saelens BE, Leary L, Cain K, Conway T, Hess P. Thedevelopment of a walkability index: application to the neighborhoodquality of life study. Br J Sports Med. 2010;44:924–33.28. Kerr J, Sallis JF, Owen N, Bourdeaudhuij I, Cerin E, Reis R, Sarmiento O,Frömel K, Mitáš J, Troelsen J, Christiansen LB, Macfarlane M, Salvo D,Schofield G, Badland H, Guillen-Grima F, Aguinaga-Ontoso I, Davey R,Bauman A, Saelens B, Riddoch C, Ainsworth B, Pratt M, Schmid T, Frank LD,Adams MA, Conway TL, Cain KL, Van Dyck D, Bracy NL. Advancing scienceCain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 10 of 11and policy through a coordinated international study of physical activityand built environments: IPEN methods. J Phys Act Health. 2013;10:581–601.29. Cohen JA. Coefficient of agreement for nominal scales. Ed psych.Measurement. 1960;20:37–46.30. Cicchetti DV. The precision of reliability and validity estimates re-visited:distinguishing between clinical and statistical significance of sample sizerequirements. J Clin Exp Neuropsychol. 2001;23:695–700.31. Landis JR, Koch GG. The measurement of observer agreement forcategorical data. Biometrics. 1977;33:159–74.32. Giles-Corti B, Vernez-Moudon A, Reis R, Turrell G, Dannenberg AL, BadlandH, Foster S, Lowe M, Sallis JF, Stevenson M, Owen N. City planning andpopulation health: a global challenge. Lancet. 2016;388:2912–24.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Cain et al. International Journal of Behavioral Nutrition and Physical Activity  (2018) 15:19 Page 11 of 11


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items