Built environment assessment: Multidisciplinary perspectives.

CONTEXT: As obesity has become increasingly widespread, scientists seek better ways to assess and modify built and social environments to positively impact health. The applicable methods and concepts draw on multiple disciplines and require collaboration and cross-learning. This paper describes the results of an expert team׳s analysis of how key disciplinary perspectives contribute to environmental context-based assessment related to obesity, identifies gaps, and suggests opportunities to encourage effective advances in this arena. EVIDENCE ACQUISITION: A team of experts representing diverse disciplines convened in 2013 to discuss the contributions of their respective disciplines to assessing built environments relevant to obesity prevention. The disciplines include urban planning, public health nutrition, exercise science, physical activity research, public health and epidemiology, behavioral and social sciences, and economics. Each expert identified key concepts and measures from their discipline, and applications to built environment assessment and action. A selective review of published literature and internet-based information was conducted in 2013 and 2014. EVIDENCE SYNTHESIS: The key points that are highlighted in this article were identified in 2014-2015 through discussion, debate and consensus-building among the team of experts. Results focus on the various disciplines׳ perspectives and tools, recommendations, progress and gaps.
CONCLUSIONS: There has been significant progress in collaboration across key disciplines that contribute to studies of built environments and obesity, but important gaps remain. Using lessons from interprofessional education and team science, along with appreciation of and attention to other disciplines׳ contributions, can promote more effective cross-disciplinary collaboration in obesity prevention.

Introduction

The built environment related to obesity can be thought of as the totality of places built or designed by humans, including buildings, grounds around buildings, layout of communities, transportation infrastructure, parks and trails (Anonymous, 2005, Sallis et al., 2012), and features of locations where food is marketed, sold and served (Glanz et al., 2005, Glanz, 2009). Built environments and the policies that shape them are increasingly considered key determinants of health behaviors related to obesity and other chronic diseases (Anonymous, 2001, Koplan et al., 2005, Parker et al., 2009). Thus, an improved understanding of built environments – and built environment measures – is critical to population health.

A variety of measures now exist that allow researchers and practitioners to plan and evaluate changes to the built environment (Brownson et al., 2009, McKinnon et al., 2009, Story et al., 2008). The measures establish the foundations for interventions to improve health by changing the built environment and for evaluating those interventions. Important advances in assessing the built environments related to eating and activity have drawn on multiple disciplines that have not traditionally worked together. These disciplines include nutrition, exercise science, public health, epidemiology, social and behavioral sciences (psychology, sociology, anthropology), urban planning, transportation, economics, and other disciplines (e.g. law, informatics/computer science, geography, policy studies). The research traditions, core concepts, metrics, and analytic methods from these different disciplines vary greatly from each other. The cross-disciplinary nature of these methodologies has created challenges to using a wide range of measurement strategies, because researchers, practitioners, and policymakers have tended to be rooted in single disciplines. Despite substantial progress in working across disciplines, silos and obstacles to collaboration remain.

The Built Environment Assessment Training (BEAT) Institute Think Tank was convened in 2013 to enable top scholars and practitioners to discuss the contributions of their respective disciplines to research and practice on assessing built environments that are relevant to obesity prevention. The 2-day invitation-only meeting, held in the summer of 2013 in Philadelphia, brought together 29 nationally recognized faculty, highly-cited authors in related fields, and exceptional alumni from the preceding five years׳ BEAT Institutes (Glanz, Sallis, & Saelens, 2015) [see Supplementary Appendix].

This article is based on discussions that began at that meeting and continued through early 2015. The purpose of this article is to highlight examples of key concepts that are defined and viewed differently through the lens of various disciplines, to describe selected successful collaborations across disciplinary lines, and to identify examples of and needs for better cross-disciplinary training and research. Although this article is not exhaustive and is a selective examination of published literature, it covers a number of key issues that the expert team identified. The article offers recommendations and highlights opportunities for successful collaborations.

Participants and methods for reviewing and synthesizing evidence

The expert team members were trained in two key “content-oriented” disciplines: nutrition and food sciences, and exercise science and physical activity research (Sallis, Carlson, Mignano, Lemes, & Wagner, 2013); and five other disciplines: urban planning, transportation, public health and epidemiology, behavioral and social sciences, and economics. Some have interdisciplinary backgrounds in additional relevant disciplines. Each expert reviewed evidence in her discipline; summarized key concepts, definitions and measures from that disciplinary tradition; and identified representative examples of built environment assessment related to obesity. Brief coverage of other disciplines, not explicitly represented by the team, was also compiled. A selective review of published literature and internet-based information on training programs was conducted in 2014 and 2015. The team also sought ideas for successful collaboration from the fields of interprofessional education, organizational development, and team science. The key points highlighted in this article were identified in 2014–2015 through discussion and consensus-building among the team of experts.

The foundation of built environment assessments and intervention across disciplines

This section describes the emphases of seven key disciplines and gives illustrative examples of concepts and tools used in each one. These descriptions are necessarily brief and not comprehensive. Table 1 summarizes highlights from each discipline, focusing on measures that are consistently associated with physical activity, diet, and obesity.

Table 1

Disciplines׳ emphasis and examples of built environment concepts and measures.

Discipline	Importance or emphasis	Examples of key concepts and measures/tools
Urban Planning	Focuses on the technical and social-political processes that shape land-use patterns and community design.	Use of geographic information systems (GIS) to extract measures of density and land-use mix from existing data sources.
		Measures such as walkability draw on urban planning concepts
Transportation	Planning and design of physical infrastructure of roads, sidewalks, bike paths, railroad tracks, bridges, etc.; understanding of daily travel choices	Use of GIS to extract measures of features of transportation systems. Measures of connectivity of street and pedestrian/bike systems.
Nutritional Science, Public Health Nutrition	Focus on how food choices are affected by neighborhood, store/restaurant and home food environments	Local/setting availability may influence what people eat.
		Measures: NEMS-S, NEMS-R
Exercise Science and Physical Activity (PA) Research	Attention to physical activity environments in neighborhoods and organizational settings where PA occurs	Self-report surveys, systematic observations, and secondary data analysis of walkability, bikeability, off-road walking/biking trails, parks and other physical activity settings.
		Measures: MAPS, SOPARC, SOPLAY
Epidemiology and Public Health	Study of the patterns, causes, and effects of health and disease conditions. Informs policy decisions and evidence-based practice. Often emphasizes health disparities	Contributes to study design, collection and statistical analysis of data, and interpretation and dissemination of results.
		May involve linking population-based behavioral or biological data to environment assessments
Behavioral and Social Sciences	Examines behavior and processes and social context, societal-level variables, and relationships within community. Emphasizes the importance of place to health.	Strong expertise in assessment, measurement development and psychometrics, experimental design, multi-level analysis and complex modeling.
		Measures: Wellness Child Care Assessment Tool, NEMS-P
Economics	Examines the importance of environmental factors including pricing, taxation and marketing on food consumption, PA behaviors and health-related outcomes. Often emphasizes health disparities.	Linkage of individual-level and contextual data sets with emphasis on fixed effects and longitudinal models. Focus on how enactment of policies affect the BE through changes such as food pricing, availability, and advertising.
		Tools/measures: Price elasticity of demand, BTG-COMP, secondary data analysis

Urban planning and transportation

Built environment assessment is rooted in the fields of urban planning and transportation. Historically urban planners have been concerned with a wide variety of effects of the built environment, including economic, social, and environmental, and have focused on health effects only recently (Berke et al., 2014, Handy et al., 2002, Saelens et al., 2003). Planners are interested in measures to define assets and needs, as well as to evaluate changes over time.

The most common built environment measures in the planning field focus on function. “Land use” typically refers to the distribution of activities across space, including the location and density of different activities, where activities are grouped into relatively coarse categories, such as residential, commercial, office, or industrial (Ewing and Cervero, 2010, Handy et al., 2002). Land-use mix, an important indicator for understanding the potential for walking as well as access to food, is measured in a variety of ways, such as the share of land area for different uses or using an “entropy” index (Cervero & Kockelman, 1997). Accessibility measures, such as the distance to the nearest destination of a given type or the number of job opportunities within a specified travel time, also reflect land-use mix (Iacono, Krizek, & El-Geneidy, 2010). These measures are usually developed from existing data sources, such as national population censuses and local tax assessor databases, with the help of geographic information systems (GIS) (Forsyth, Schmitz, Oakes, Zimmerman, & Koepp, 2006). Geographic information systems (GIS) are widely used to measure the built environment in the planning field. Planners are also concerned with aesthetic qualities of the built environment (Ewing and Handy, 2009, Ewing et al., 2006). Measures of such qualities generally depend on extensive field work to collect data on design details using one of many audit instruments developed for this purpose (Day et al., 2006, Hoehner et al., 2007).

The transportation system is another critical component of the built environment. The transportation system includes the physical infrastructure of roads, sidewalks, bike paths, railroad tracks, bridges, and so on (Handy et al., 2002). In the transportation field, rooted in engineering, measures focus mainly on the capacity of systems to move vehicles or, less commonly, people. In studies of the links between the built environment and active modes of travel, measures of the connectivity of street systems are widely used (Ewing and Cervero, 2010, Frank et al., 2005). Most such measures are derived using GIS from street networks built for vehicle travel, though communities with extensive pedestrian or bicycle networks provide considerably higher connectivity for active travel (Chin et al., 2008, Tal and Handy, 2012). Unfortunately, few public agencies maintain databases on the detailed design characteristics of streets, such as the presence of sidewalks and bike lanes, their widths, pavement conditions, crosswalks, signals, etc. In the absence of such data, audits of the street environment on a block-by-block basis have been undertaken (Day et al., 2006, Hoehner et al., 2007).

Nutritional science, public health nutrition

Human nutrition is defined as the science of food, the nutrients and other substances contained therein, their action, interaction, and balance in relation to health and disease (Anonymous, 2004). Public health nutrition focuses on the promotion of good health through nutrition and primary prevention of nutrition-related illness in the population (Anonymous, 2014). Contemporary nutrition science operates within a broad, integrated conceptual framework, such that it is a social and environmental science concerned with personal and population health (Beauman et al., 2005).

Research about food and nutrition environments emerged from a blend of biologically-based nutritional science and public health nutrition (Glanz et al., 2005, Glanz et al., 2007, Ohri-Vachaspati and Leviton, 2010, Saelens et al., 2007). To improve access to healthy foods in urban areas, it is essential to conceptualize, measure, and ultimately address the opportunities and barriers that people face in purchasing and eating healthy foods. A growing body of research indicates that local availability of healthy foods influences what people eat (Caspi et al., 2012, Kamphuis et al., 2006) and may help explain racial and income disparities in healthy food consumption, obesity rates, and conditions such as diabetes and heart disease (Glanz et al., 2005).

Nutrition experts have developed tools to measure the nutrition, or food environment (Ohri-Vachaspati and Leviton, 2010, Saelens and Glanz, 2009). One of the most widely used, and adapted, tools is the Nutrition Environment Measures Survey (NEMS), an observational audit tool that assesses availability of healthful foods, prices, and promotion. NEMS tools are available to assess food environments in stores (Andreyeva et al., 2008, Cavanaugh et al., 2013, Franco et al., 2009, Glanz et al., 2007), restaurants (Saelens et al., 2007), vending machines (Voss, Klein, Glanz, & Clawson, 2012), and other settings (Honeycutt, Davis, Clawson, & Glanz, 2010). NEMS is based on established dietary guidelines and nutrition epidemiology data, and uses observational methods drawn from the field of sociology. New and more ambitious food environment assessments that aim to measure the caloric and nutrient equivalents of all foods in a store are in development (Gortmaker et al., 2013, Slining et al., 2013).

Exercise science and physical activity research

The field of exercise science includes an array of disciplines, e.g., kinesiology, biomechanics, exercise psychology, and athletic training. The role of physical activity in combating obesity has increasingly gained acceptance among researchers and health professionals worldwide (Heath et al., 2012, Kohl et al., 2012, Tremblay et al., 2007). When conceptualizing the built environment, physical activity research draws heavily from other fields, including urban planning and transportation, to create active living – supportive measures of environments, such as walkability indices, that include measures such as land mix, street connectivity, and accessibility (Raudenbush and Sampson, 1999, Schaefer-McDaniel et al., 2010).

Physical activity researchers have collaborated with urban and transportation planners to develop built environment surveys (Brownson et al., 2009, Schaefer-McDaniel et al., 2010, Zenk et al., 2016). These assessments include measures related to leisure physical activity and active transport, such as (1) neighborhood walkability; (2) neighborhood bikeability (presence of bike lanes, bike signage); (3) off-road walking/biking trails; and, (4) parks and playgrounds. Data collection methods for these assessments have included self-report surveys (Saelens et al., 2003, Sallis et al., 2010, Slater et al., 2013), systematic observations (Brownson et al., 2009, Raudenbush and Sampson, 1999, Schaefer-McDaniel et al., 2010), secondary data analysis using GIS (Brownson et al., 2009, Ding et al., 2011, Forsyth et al., 2006, Leslie et al., 2007), and more recently the use of omnidirectional imagery, such as Google Street View and Google Walkability (Badland et al., 2010, Clarke et al., 2010, Griew et al., 2013, Kelly et al., 2013, Rundle et al., 2011, Vargo et al., 2012).

Evidence showing associations between physical activity-related environmental measures, such as presence of sidewalks, public transit, and mixed land use, and increased walking and less obesity continues to grow (Boarnet et al., 2011, Ding et al., 2012, Pikora et al., 2006, Saelens et al., 2003, Slater et al., 2013). There remains a need to develop consistent measures for this research, through joint efforts with fields such as urban planning and transportation.

Epidemiology and public health

Epidemiology studies the patterns, causes, and effects of health and disease conditions in defined populations (Anonymous, 2014). Social epidemiology, a growing branch of epidemiology, focuses on the effects of social-structural factors, including socioeconomic advantage and disadvantage, on health (Berkman & Kawachi, 2000). Epidemiologists help with study design, conceptualization of determinants of behavior and health, collection and statistical analysis of data, and interpretation and dissemination of results (Porta, 2014). In studies of obesity and the built environment, environment measures are often considered “exposures” and studies are designed to examine population effects such as food intake, obesity, and chronic diseases and their risk factors. Studies of the epidemiology of risk factors for obesity and metabolic syndrome have been key to describing the lower access to supermarkets (i.e., “food deserts”) in minority and lower-income neighborhoods (Moore and Diez Roux, 2006, Morland et al., 2006).

Psychology

Psychologists bring expertise in assessment, measurement development, psychometrics, and experimental design. They have strong quantitative analytic skills, including multi-level analysis and complex modeling. They study psychosocial and socioeconomic variables, perceptions and attitudes, and behaviors and the environmental cues that impact them. Some of these variables need to be considered as they often moderate or mediate effects of observed built environment measures (Ding et al., 2012, Sallis et al., 2010) on health (e.g., perception of safety may be more critical to physical activity than actual safety markers). Psychologists have played a key role in the development of measures of both the food and physical activity environment (Henderson et al., 2011, Rosenberg et al., 2010, Saelens and Glanz, 2009, Sallis et al., 2010), often blending content expertise (nutrition, physical activity) with psychological constructs and methods. Other work includes evaluation of natural experiments and environmental manipulations across many settings including schools (Long et al., 2010, Long et al., 2013), workplaces (French et al., 2010), communities, and homes (Fiese et al., 2012); and studies of food access (Drewnowski et al., 2012, Epstein et al., 2012, Jiao et al., 2012).

Economics

Economists examine the importance of environmental factors in food consumption and physical activity behaviors and health-related outcomes such as obesity based on an economic framework wherein individuals are assumed to maximize utility, subject to a number of constraints. Within this framework individuals produce and demand health and weight, among other goods, including the consumption of food and beverages that both directly and indirectly (through changes in weight and health) affects utility. Utility maximization can also explain physical activity behaviors, and the transportation field has traditionally used this theory to explain choices about daily travel, including the use of active modes. In making decisions, individuals face several constraints, including budget constraints, time constraints, and biological constraints (Cawley, 2004).

Economists seek to understand how various factors and policies related to the built environment can be expected to change behaviors and health-related outcomes. These include pricing (i.e., taxes/subsidies), products, placement, and promotion (i.e., advertising restrictions) (Chandon & Wansink, 2012). Internationally, studies have contributed to the evidence base on the price sensitivity of food and beverage consumption (Andreyeva et al., 2010, Powell et al., 2013, Thow et al., 2010), measured by a common metric called the price elasticity of demand. For example, based on U.S. data, the price elasticity of demand for sugar-sweetened beverages is estimated to be −1.2 suggesting that a 20% price increase would reduce demand by 24% (Powell et al., 2013). Economists also assess how behavior responses to economic and environmental factors differ across sociodemographic characteristics (Finkelstein et al., 2010, Powell and Han, 2011) and how environments contribute to health disparities (Powell, Wada, Krauss, & Wang, 2012).

To empirically estimate the impact of economic and contextual factors related to the built environment, researchers often rely on national commercial or government secondary contextual data sources that include, for example, C2ER (formally called ACCRA) and Nielsen Homescan data for prices, Dun & Bradstreet, InfoUSA, or Census Bureau outlet density data, and Nielsen Media Research or Competitive Media Reporting advertising data. These data are often linked with individual-level data using geographic identifiers. Although available with national geographic coverage, economists and other researchers have shown that these data have limitations including their cost, level of available geographic proximity, validity, and comprehensiveness (Powell and Chaloupka, 2009, Powell et al., 2011).

Other disciplines

Several other established and emerging disciplines have been important in measurement and research on built environments related to obesity. Within the field of education, school contexts are important because they provides a context where environmental factors can be assessed and influenced, as well a structured setting for assessment (Long et al., 2010, Long et al., 2013). Public policy and policy studies scholars help to frame and define the strategies and proposed outcomes of environmental change strategies (Chriqui et al., 2014, Masse et al., 2013). The fields of marketing (Glanz et al., 2012) and law (Gostin, 2007) also intersect with psychology, urban planning and policy studies. Geography contributes methods to built environment measurement, both in relation to healthy food access and active travel opportunities (Feng, Glass, Curriero, Stewart, & Schwartz, 2010).

Other disciplines often bring cross-disciplinary foci to the challenge of assessing built environments – among them, environmental psychology and behavioral geography (Sobal & Wansink, 2007). Health disparities researchers are increasingly examining obesity through a built environment lens, as theses scientists have primary backgrounds in nutrition (Ford & Dzewaltowski, 2008); physical activity (Gordon-Larsen, Nelson, Page, & Popkin, 2006); epidemiology (Lovasi, Hutson, Guerra, & Neckerman, 2009); and a blend of economic, geographic, and sociological expertise (Singh, Siahpush, & Kogan, 2010).

Examples of successful collaborations and areas for improvement

The co-authors identified several “success stories” in their research that illustrate effective interdisciplinary collaborations in developing and deploying assessments of the built environment. These successes often resulted from challenging discussions, compromises and co-learning. Here we present selected examples across the spectrum of disciplines. We also note that there are numerous areas for improvement, indicating that challenges remain in working across disciplines in built environment assessment and research, and provide recent illustrative examples of these.

Examples of successful collaborations

Measurement tools for food and activity environments that have been adapted for different audiences and contexts grew out of collaborative efforts. The Microscale Audit of Pedestrian Streetscapes (MAPS) is an approach to studying street design and pedestrian environments that draws on methods from physical activity research, urban planning, and transportation (Millstein et al., 2013). The associations between MAPS׳s well delineated characteristics and different types of physical activity (e.g., transport, leisure) have been demonstrated using data from children, adolescents, adults and older adults (Cain et al., 2014), and an validated abbreviated version of MAPS has also been created (Sallis et al., 2015). The Nutrition Environment Measures Survey (NEMS), initially developed to assess healthy food environments in stores and restaurants (Glanz et al., 2007, Saelens et al., 2007), came from a conceptual framework that blends nutrition, urban planning, social sciences and marketing (Glanz et al., 2005). It has subsequently been used to link environments with data from a study of cardiovascular epidemiology (Franco et al., 2009, Moore et al., 2012) and used for food environment assessment in a study of urban form, travel behavior and food destinations (Kerr et al., 2012).

In school and child care settings, notable collaborations have supported the development of measures of environments and assessments of the impact of new policies. For example, to evaluate the impact of policy changes on the U.S. school food environment and United States Department of Agriculture school meal participation, a team with experts in nutrition, psychology, sociology, education and public health drew on methodologies from each discipline. With a shared commitment and common goals, they developed rigorous yet practical measures and data collection, and tackled the challenges of interpreting the findings (Long et al., 2010, Long et al., 2013).

Policy studies of food and activity-related environments have crossed disciplinary boundaries to use geographic, policy/taxation and media market data to estimate the impact of economic and contextual factors (Powell and Chaloupka, 2009, Powell et al., 2011, Smith et al., 2010). These analyses usually rely on available data sources, though one large, multidisciplinary study – Bridging the Gap Community Obesity Measures Project (BTG-COMP) – was undertaken from 2010 to 2012 in a U.S. national sample of neighborhoods where middle- and high-school students live and for which there are corresponding student cross-sectional survey data from the Monitoring the Future Study (Bridging the Gap, 2012). This ambitious study collected and coded local ordinances and master plans, and school district wellness policies; and undertook on-site observation of local retail food stores and restaurants and physical activity settings – thus blending methods from several disciplines (e.g., public health, economics, psychology, political science) with content expertise from nutrition and physical activity (〈http://www.bridgingthegapresearch.org/research/〉).

Examples of areas for improvement in crossing disciplinary boundaries

Although we presented our collective, broad definition of the built environment related to obesity at the beginning of this paper, we observe a lack of consensus among researchers on the definition and measurement of “built environment” across disciplines. For example, experts in planning, transportation, and physical activity tend to question and not embrace definitions and conceptual and empirical models used by those in the nutrition, economics and policy fields, and vice versa. This is both a cause of the continued existence of disciplinary silos and a consequence of the persistent patterns of publication in one or another area. Further, most research groups still work primarily in the activity or the nutrition area, rather than addressing energy balance as a combined concern (Economos, Hatfield, King, Ayala, & Pentz, 2015). Peer review panels for research grant proposals whose members often operate in silos perpetuate this division, as do research project budget limiations. The length limitations imposed by many scientific journals push researchers to analyze and write up their data to fit a specific content area, even when they have data that cut across disciplines.

Early experience with a new journal that was established in 2014, the Journal of Transport and Health, exemplifies the tendency of researchers to stay within the disciplinary lines in which they were trained. This journal aims to establish a forum to span the boundaries of transportation and public health fields. However, the editorial board has seen that the reference lists in submissions to the journal often strongly favor one discipline over another. These fields use different search engines for literature searches. The editorial board is considering a policy to require authors to deliberately use search engines in multiple fields in preparing their manuscripts.

An emerging need to better integrate data from new technologies into built environment research and interventions, for both activity and food environments, expands the disciplines needed. The increased availability of devices and media (e.g., activity trackers, mobile apps such as Eat Local and MapMyRun) that can link behaviors to environments calls for bringing engineers, computer scientists, database management experts and spatial statisticians to the table. The collaborative involvement of urban planners, engineers, activity and nutrition experts remains rare (King, Glanz, & Patrick, 2015) and is an area in need of development.

Another example of a disconnect between disciplines can be seen in the lack of comprehensive conceptual and empirical models of access, for example, to healthy food. We know that access consists of both availability and price dimensions, but many studies model only one of these concepts. To close this gap, public health and planning researchers need to work with economists as well as content-area experts in nutrition and activity.

Training in built environment assessment and intervention: progress, barriers and opportunities

Training and education is an important area to address in the search for solutions to the continuing gaps between disciplines in work to assess and improve the built environment (Botchwey et al., 2009, Pilkington et al., 2008). Here we take a closer look at formal training programs in institutions of higher education, and at conferences and short-courses. We ask the question of whether recommendations for multidisciplinary curricula on the built environment and public health have advanced since Botchwey and colleagues (Botchwey et al., 2009) examined this arena in 2007 and concluded that there is little evidence that a multidisciplinary approach is widespread in professional education. At the time of their review, they found 11 relevant graduate-level courses, all of which included urban planning, but only half of which addressed transportation and/or nutrition.

In 2015, we performed a Google search using terms like “built environment health” to find courses/classes and dual degree programs in public health, urban planning and related fields. We restricted our search to English-language accredited universities (in the US and other countries) and excluded courses focused on single areas such as housing which were not intended to be multidisciplinary. Like Botchwey et al., we found that most courses in the US were based in urban planning departments. One course includes nutrition, and one addresses physical activity. Many universities in the US and internationally now offer courses or certificate programs in GIS and Public Health. In the US we found 14 universities that offer dual degrees in Public Health and Planning. Most of these programs require students to fulfill requirements of each program separately, and only a few require integrative coursework, theses or capstone projects. Of the 16 schools of built environment in the UK, Australia, New Zealand, and Canada, none emphasized coursework or programs integrating public health aspects of built environment research.

A limitation of this review is that we may have missed other course offerings; however, our team׳s anecdotal experiences underscore structural (policy and financial) obstacles to encouraging students to study across departments and schools, and thus disciplines. The flow of tuition funds sometimes leads departments to advise against (and even forbid) taking courses in other fields. Class size caps may lead to refusal to accept students from other degree programs. Decentralized budgeting models – which have become the norm at universities in the past 10–15 years – can discourage truly multidisciplinary training.

To overcome these obstacles, structural changes in higher education are needed. One source of wisdom for fostering successful collaboration among experts from different disciplines is that of interprofessional education (IPE), which is garnering increasing support for training health professionals who need to work in teams. IPE facilitates learning a common vocabulary and its translation (Evans, Cashman, Page, & Garr, 2011) and cross-training in discipline-specific analytic approaches (Breitbach et al., 2013). Ideal IPE methods involve teams of instructors, and teams of students – working across disciplines – and training experiences and practica that take trainees into the field together (Choi and Pak, 2006, Dow et al., 2013, Evans et al., 2011).

Recommendations

The science of measuring and improving the built environment related to nutrition, activity and obesity is inherently multidisciplinary. Research to assess and change built environments related to obesity, activity and nutrition has increased many interdisciplinary collaborations. However, there is much that remains to be done to take these initial advances to the next level. Future progress depends on forging effective collaborations across disciplines, improving training and education, increasing the resources available across disciplines and provided by funding agencies (e.g., NIH Interdisciplinary Research program), and supporting and rewarding publications that cross traditional boundaries.

To forge more effective collaborations across disciplines, it is important to make explicit efforts to learn basic vocabulary and methods across disciplines; develop a shared vision; share recognition and credit; and to foster trust and handle conflicts constructively (Bennett, Gadlin, & Levine-Finley, 2010).

In order to advance broad training in the built environment and public health, it will be important to develop curricular policies and practices that reduce barriers to students taking courses in research and intervention methods outside their departments and schools, and to increase their exposures to different approaches, colleagues and relationships.

To increase research funding resources, federal agencies and foundations should assume leadership. Special set-aside funds for interdisciplinary studies, guidance to peer review panels, and allowing flexible funding can drive and encourage interdisciplinary research.

Journal editors should collaborate with professional organizations to support and reward outstanding interdisciplinary research publications. They should consider relaxing length limits on manuscripts and enable authors to publish related multiple articles that tell a broader story as a “package” in their journals. Finally, academic appointment and promotion committees should communicate the importance of meritorious cross-disciplinary publications to internal and external referees.