A Partnered Approach for Structured Observation to Assess the Environment of a Neighborhood With High Diabetes Rates

Abstract

Background

The Communities IMPACT Diabetes Center uses partnered methods to address diabetes-related conditions among African Americans and Latinos in East Harlem, New York.

Objectives

To describe a novel, partnered approach that integrates simultaneous structured observation by community and academic partners with “on-the-spot” resolution of differences to collect baseline data regarding the built and food environments in a two census tract area of East Harlem and present select findings.

Methods

We designed an environmental assessment to explore characteristics of the environment related to walking and eating. We paired community and academic partners to assess each block, resolve any differences, and report results. Nearly one year later, we surveyed the data collectors and analyzed responses using standard qualitative methods.

Results

Key themes included connection to and characteristics of the community; interactions with partners; surprises and learning, and aspects of data collection. All but the first were common to academic and community partners. Relationships between partners were generally amiable. Both community—“I think it was very helpful, we made sure neither of us made mistakes, and helped each other when we could”—and academic–“I really enjoyed it … I learned a lot about the areas I surveyed”—partners were complimentary. Community partners’ strengths included local knowledge of the community, whereas academic partners’ focus on adherence to the specifications was critical. Structured observation identified many sidewalks in disrepair or obstructed, few benches, and highly variable times allocated for pedestrians to cross at cross walks.

Conclusions

The partnered data collection was both successful and formative, building additional relationships and further capacity for ongoing partnership. Community partners saw their community in a new way, seeing, “little things that are important but people don’t pay attention to.” Structured observations added to our understanding of how an environment may contribute to diabetes.

Communities IMPACT (Inspired and Motivated to Prevent And ConTrol) Diabetes Center is a Centers for Disease Control and Prevention Racial and Ethnic Approaches to Community Health (REACH U.S.) Center of Excellence for the Elimination of Disparities, established to use community-partnered methods to reduce disparities caused by high rates of diabetes-related conditions among the Black and Latino communities, initially in East Harlem, and ultimately throughout New York and neighboring states.

East Harlem (Figure 1A), known as Spanish Harlem or El Barrio, is a community is of rich traditions, talented individuals and dedicated community organizations. In New York City, East Harlem has the highest burden of obesity, diabetes, diabetes complications, and diabetes mortality.¹ In striking contrast with the Upper East Side, East Harlem’s healthier, wealthier, and whiter neighbor to its south, East Harlem residents are predominantly Black and Latino, low income, and undereducated with 4-fold higher rates for obesity, and even greater differences in diabetes (7-fold), diabetes mortality (5-fold), and diabetes related amputations in the elderly (16-fold).^1,2 Blacks and Latinos are more likely to suffer with and die from diabetes than Whites: Current trends suggest that one in three Latino children could eventually develop diabetes.³

Figure 1.

(A, left) Map of Manhattan depicting location of East Harlem (in the northeastern corner). To provide a sense of scale, Central Park is approximately 2.5 miles from south to north. (B, right) The SEED within East Harlem.

The IMPACT Diabetes Center builds upon a decade-long effort to engage partners—community leaders and organizations in East Harlem—around problems related to diabetes.^4–11 In 2007, we convened three meetings of two to three dozen academic and community partners to discuss the application that funded the center. These meetings defined a dual focus on preventing development of diabetes and preventing complications in individuals with diabetes. They also expanded our focus to include the lifespan from before birth to older adults and evolved the idea of developing a smaller geographical area within East Harlem, a Sector of Excellence for the Elimination of Disparities (SEED). We began our work with local assessments and interventions, using lessons learned to initiate larger policy and systems change activities. This paper describes a subset of our initial evaluation work and some lessons learned.

We extended a conceptual model previously developed by our partners that suggests that disparities regarding diabetes are influenced by a environmental context with the food and built environment being prominent influences. Using Frieden’s terms, we considered context to promote default behaviors.¹¹ Resources from the Centers for Disease Control and Prevention have allowed us to prioritize the food environment and the built environment as it relates to physical activity for children or adults. Such characteristics include population density, public transportation systems, lighting, sidewalks, greening, and safety. As foundational for our efforts, we sought to systematically observe the built environment within our two-census tract SEED.^13–15

This paper describes lessons learned from a series of systematic observations made in the East Harlem SEED as part of the baseline data collection to support planning for and evaluation of interventions aimed at eliminating diabetes health disparities.^16,17

METHODS

Coalition Involvement in Project Development

Coalition members have played leadership roles from the outset. Our planning meetings brought several dozen organizations together, including public agencies, such as the city and state health departments and the New York City Housing authority; local health care stakeholders, including community health centers, local public and private hospitals, and the East Harlem Community Health Committee; social services agencies, such as Union Settlement Association, Children’s Aid Society, Yorkville Common Pantry, Little Sisters of the Assumption Family Health Service, and Linkage House; grassroots organizations, such as public housing tenants associations, the Neighborhood Open Space Coalition, and the East Harlem HIV Network; local business leaders including from La Marqueta (a well-known, historic market place), and farmers’ market and bodega (convenience food store) associations; and community residents and natural leaders. Academic partners from a variety of Departments at the Mount Sinai Medical School and Mount Sinai Hospital also attended at least one of our three planning meetings. Although community partnerships evolve all of the time, with new partners coming in to the coalition and others leaving it, many of these organizations remain a part of our ongoing activities. Structurally, our coalition has a variety of key committees and subcommittees that serve as working groups and a task force that serves as a leadership group. All are chaired by community partners, who typically receive compensation for taking on leadership roles and/or for participation in time-consuming tasks, such as the data collection we describe. Four authors of this paper (DL, PB, HRN, and MAR) are community partners; each played an active role in at least the conception and editing of this paper.

The SEED is a two-census tract area within East Harlem (Figure 1B) and is proximate to Mount Sinai. A community partnered process selected the SEED using local knowledge, community-developed selection criteria, and maps, populated using geographical information systems. Academic partners served as conveners and technical consultants to the community-led selection process, which is described elsewhere in this issue.¹⁸

Evaluation Overview

The Evaluation Committee, chaired by a community partner, developed an assessment tool to serve both as baseline data for the evaluation and as key input for planning interventions. All human subjects’ research is approved by the Mount Sinai Institutional Review Board. Community partners who chaired committees and those who served as data collectors were paid modest stipends.

Our multifaceted approach to studying our SEED is depicted in Figure 2. We conducted two field surveys of adult residents at outdoor sites chosen to produce a representative sample and a survey of parents about their children. The adult survey asked about health, health knowledge, eating, exercise, shopping behaviors, perceptions regarding the community and measured body mass index. The parent survey asked about their child’s eating and exercise habits, health status, use of television, video games, and computers, and measured body mass index as well. Community health data were made available by the New York City Department of Health and Mental Hygiene.³ A video assessment captured images of the SEED during the work day using previously described methods.¹⁵ A photovoice project complemented the video assessment, providing striking images of how a group of women of childbearing age experienced facilitators and barriers to healthy behavior for them and for their families. We are preparing separate manuscripts to describe these findings, and although not the focus of this paper, they describe context for our current report regarding the block-by-block walking assessment survey of the built and food environment. This was carried out by pairs of academic and community partners. We present initial findings from these structured observations and describe the process using interviews of the data collectors for the walking assessment.

Figure 2.

Structured Observation: The Walking Assessment

Community and academic partners selected constructs for data collection. Academic partners staffed development of paper instruments for recording observations, including specifications and a structured data collection form (Figure 3). Pairs of community–academic partners served as observers on each block. Observers were trained using an 8-page instruction guide. Partners were trained to independently record data from one side of the block (a face block), while their partner did the opposite side, cross the street and independently record data for the opposite face block that their partner had just completed, meet, compare notes, and resolve differences, often by returning to reexamine the area in question. These differences typically involved a discrepancy in a count, such as trees. One community partner who was experienced in research and an investigator for the overall project served as an academic partner during the assessment of several face blocks and as a community partner during the assessment of others. Otherwise, all data collectors were either academic or community partners.

Figure 3. Page 1 of Data Collection Form.

Abbreviations: Bus = Business; CS = construction site; L = streetlamp; P = pothole; RB = residential building; T = tree.

The data collection form prompted for detailed information, such as state of repair of the sidewalk, fixed (e.g., construction) or temporary (e.g., crowds of people, garbage bags) obstructions to passing on the sidewalk, presence or absence of ramps at the end of the block, pedestrian crossing lights, time allowed to cross, and, when present, health-related advertisements on the street, observed physical activity of individuals, number of trees, community gardens, graffiti, or murals, types of food related businesses, and so on. These instruments included an opportunity to draw maps of each block to record the presence of benches, trees, businesses, and vacant lots or abandoned buildings (Figure 3). A second round of data collection included visits to restaurants and food stores. Observers were given disposable cameras and encouraged to take photos.

Qualitative Survey of Data Collectors

We report findings of a qualitative survey of observers that served as our main process assessment. Specifically, respondents were asked to describe strengths and weakness of the SEED, their experience working with partners, lessons learned, and surprises about the experience or their findings. Surveys were conducted nearly 1 year after the walking assessment. Respondents had the option to type in answers to open ended-questions or to answer the questions verbally and review and revise responses that were typed by the interviewer. Analysis incorporated grounded theory with the coding of themes by one investigator using the method of constant comparatives.^19–24

RESULTS

Data were collected for 74 face blocks by six academic and seven community partners. Data collection took a mean of 13 minutes 6 seconds per face block (median, 12.0), with no systematic differences between academic and community partners (p = .70). Three quarters of assessments were completed within 16 minutes, with 3 of 148 taking 30 minutes or longer.

Structured Assessment

Tables 1 and 2 highlights results from the structured assessment. Of note, only one block face had any benches, the condition of the sidewalk was characterized as good on only 53.4% of block faces, and there was nearly 4-fold variability in the length of time pedestrians are allotted to cross streets at marked crosswalks. There was one park or playground for every seven block faces in the SEED. Figure 4 illustrates some findings.

Table 1.

Characteristics of SEED Physical Environment

Characteristic, No. Per Block Face	Mean	Total per SEED	Range	25th–75th Percentile
Benches	0.04	3	0–3	0–0
Broken Windows	0.30	21	0–12	0–0
Community Garden	0.09	6	0–2	0–0
Parks/Playgrounds	0.07	5	0–2	0–0
Potholes	0.31	22	0–4	0–0
Trees	3.60	253	0–18	0–5
Time to Cross Street (seconds)	36.70	NA	14.0–51.4	29.0–48.4

Table 2.

Sidewalk Conditions in SEED

Characteristic	%
Overall Condition
Good	53.4
Fair	38.9
Poor	7.6
Any Sidewalk Obstruction	30.1
People, n	7
Parked cars, n	5
Garbage cans/bags, n	6
Walkable for Full Length	95.8

Figure 4. Conditions.

(A, top) Sidewalk conditions. (B, middle) Community garden under a fast food billboard, with data collector. (C, bottom) Construction site makes sidewalk impassable.

Among the restaurants in our survey, one in five was a national fast food chain, and one in three either a pizza restaurant or served Chinese food for takeout. Two thirds served at least one nominally healthy item (typically a salad), and the median cost of the food that was self-described as most popular was $5.99. We surveyed 37 stores that served food: 43% sold apples, 38% tomatoes and 19% fat-free milk. Displays of beverages reflected our findings from the adult and children’s surveys that diet drinks were not popular, with both water and sugar-sweetened beverages favored. Community partners inform us that there is one green grocer in the SEED and two just outside of the SEED boundaries that try to meet local dietary preferences, including those specific to various cultures, including Mexican, Latino, and Caribbean. They also describe “shop hopping,” or seeking bargains and opportunities to meet their dietary preferences at a variety of stores in and around the SEED.

Qualitative Survey

Respondents included three of six academic partners and six community partners. One community partner, who also is an IMPACT co-investigator, served both as an academic partner and as a community partner and completed the survey as an academic partner. The three missing academic partners were no longer available by the time of the survey. The Community respondents were mostly women (77%), ranged from their 20s to 50s in age, and included individuals with college education; 50% were Hispanic, 17% Black, and 33% non-Hispanic White. Two academic respondents were Hispanics, two women, and the age range was similar.

Common themes among community partners included connection to and characteristics of their community, interactions with academic partners, surprises and learning, and aspects of data collection. Table 3 shows exemplar quotes. Table 4 shows that academic partners’ responses reflected similar themes, other than connection to the community. In the body of the text, we refer to the first quote in Table 3 as Q2.1, the fourth quote in Table 3 as Q3.4, and so on.

Table 3.

Findings from Community Partners

Connection to Own Community
1. I decided to participate in the walking assessment in order to learn more about the environment in my own community.
2. I liked the entire enterprise and staff, the idea of walking around the neighborhood and volunteering to do something that seemed like an anthropological study. Plus I was unemployed and the stipend helped feed us.
3. I decided to do this assessment because my input was going to make a difference in my community, knowing that this will be a change for the future.
The SEED Community
4. The biggest asset or strength of and in the SEED is the determination of its members to survive and enjoy life. People also allow each other to have imperfections and refrain from being overly judgmental. The biggest weakness in the seed is the lack of collective community activism on issues which have serious ramifications for its members…. I also observe lack of care for the environment exhibited by unnecessary garbage and dog poop…. The major negative change taking place in the SEED is gentrification because people who live in the neighborhood may eventually not be able to afford to live there.
5. Its culture is irreplaceable. [Deficits include] poverty, ignorance of outside opportunities, [and] gentrification.
6. There are new developments, housing and nice parks. There were no supermarkets and no healthy restaurants. Also, it’s hard to be Hispanic and see that our food is not healthy. There were fast food restaurants that turn into obesity and diabetes. [A negative thing is] New co-op buildings that are not built for people in the community.
7. [Biggest SEED asset is] meeting people that care… [the biggest positive change is] New buildings being built and seeing blocks maintaining their upkeep, [with the biggest weakness being] Certain blocks appear not to show any significant progress. [Negative changes include] Certain corners do not appear that safe. [The biggest surprise was] that people were positive … I was surprised the way the neighborhood is progressing and changing.
8. There are many good changes, like the trees that were put in our community, the charter schools for better education, and the parent leaders that make a difference in a child’s or parents life…. The bad changes are the new condos that are being built. The culture and roots of east and Spanish Harlem are becoming just a memory. Also, the neighborhood is now the site of Odyssey House and methadone programs.
Learning and Surprises
9. In general I enjoyed working with an academic partner because it gave me the opportunity to share some of my ideas about my community and learn about some of their impressions. Sometimes a person on the outside looking in can be insightful and capture things that a person misses when they are on the inside so to speak…. I learned to pay more attention to the physical environment in my community as it relates to health and safety.
10. [I was surprised by] how much is actually going on in my environment: Construction, traffic, and commerce…. I don’t think I learned anything I didn’t already know; I already knew there weren’t enough places to exercise, and that the food available for my budget wasn’t very healthy.
11. What surprised me the most was to see little things that are important but people don’t pay attention to these. For example, the light changing too fast is like an accident waiting to happen; so many little things that I never even thought about…. The most important thing was the experience I gained. After doing the evaluation I started to take everything a lot more serious.
12. What surprised me the most was the minute things I saw. For example, I was surprised how many empty building there are, how many cracked sidewalks we had, how on every corner there was no trash cans, and how quickly the crossing light change.
Partner Interactions
13. The positive experiences associated with my partnership were someone to share the work in a relaxed atmosphere. Other than a few minor disagreements on the status of stores in the neighborhood there were no negative experiences.
14. I think it was very helpful, we made sure neither of us made mistakes, and helped each other when we could. I don’t feel like there were any negative experiences.
15. [The partnership] was very successful and interesting. One of the observations was about a crosswalk missing and a week later they had put a crosswalk there.
16. We were in a team, so I felt safe. We compared notes and corrected each other. I could give more of an insider info since I live in East Harlem.
17. One day I was with a new person who was really bossy and felt superior and almost disrespectful of me. I could not make her listen to my POV. She was aggravating. I complained about her and was told she won’t be on the team anymore anyway…. As I said, I enjoyed all the teamsters apart from that one arrogant youth.
Data Collection
18. The major differences between the data I collected and that of one of my academic partners was based on the knowledge that I had about the social patterns in the areas assessed. For example, I had to inform one academic partner that a restaurant which she thought was no longer in business was closed because it opens later on for dinner.
19. [Differences in our findings were] Minor details; some things were missed that should’ve been noticed. Nothing too major though.
20. It is amazing how little things are the most important.
21. [How did the partnership affect data collection?] In a positive way. I got to have my input and they helped me be more professional, though at times it seemed too hair splitting, like timing the same stop light several times…. The staff helped, but I think without us the data would have been less accurate. I also feel good about being part of my neighborhood thru giving something back.

Table 4.

Findings From Academic Partners

The SEED Community
1. I really didn’t know what to expect. I assumed we would find that there were a lot of fast food restaurants and not a lot of healthy food options. Other than that, I thought that we may find there was a lot of barriers to a healthy lifestyle, although I wasn’t entirely sure what some of those barriers may be.
2. I think that the sense of community in the SEED was its strongest asset. Nowhere else in Manhattan does it feel as though the people in the community actually know and care about each other. Additionally, many people living in the SEED have lived there their whole lives, which means that they feel a strong tie to the area.
3. The biggest weakness is that people living in the area tend to be of low SES…. Money is the biggest challenge to engaging in healthy habits since being healthy often costs more.
4. By far, the biggest complaint I heard from members in the community was in regards to gentrification. Many people felt that, instead of building expensive luxury apartments, low-cost housing should be built. Others also felt that gentrification was ruining the community. That is, these apartments were attracting outsiders, and these outsiders did not have the same connection to Harlem as those who had lived there their whole lives. Since I saw how important the ties to the community were, I also understood why gentrification posed such a threat.
Learning and Surprises
5. Thought I’d learn more specifically the composition of the community—what was located on which blocks…. I didn’t really see the entire SEED but I noticed many sidewalks were not smooth and flat, had many dangerous holes and not enough time to cross the streets before the lights changed
6. I think what I found most surprising to learn was that the residents of East Harlem truly have a sense of community. I live in a suburb on Long Island, and I don’t feel the same sense of community as the people in East Harlem. I think that this plays a strong role in the decisions individuals make, and it is something I hadn’t considered.
7. I also did not realize the limitations that urban life has on one’s health behaviors. For example, it is difficult to go for a run on the sidewalks if the sidewalks are in poor shape, there is garbage lining the sidewalks, or, quite simply, there are just too many people walking around. Working on the assessment made me look at all physical environments in a different way (even though it has been a year since I worked on the project, I still take note of how walkable an area is or how supermarkets organize the food on their shelves). Once again, I had not considered how much the physical environment can impact one’s daily choices and this assessment definitely proved otherwise.
8. I was surprised by how much I learned about the vibrant life of the community. The streets were new and unfamiliar to me and I viewed them with the impartial eyes of a researcher, nothing connecting me personally to what I was mapping. As a longtime community member, [my community partner] has seen these streets change and evolve with time.
9. Through this environmental survey, I learned about the life of the community and discovered some of the powerful ways that community partners are affecting challenged in the SEED area. My greatest discovery was [my partner], who inspired me and challenged me see the SEED community in a totally different and more realistic way.
Interactions With Community Partners
10. My partners had a lot of additional knowledge about the community, what had existed at empty locations for example, how the community façade has changed over time, historical perspective…. I think we balanced each other and it made the experience richer. In terms of data collection, it seemed pretty standardized so I don’t think it affected the data collection per se…. [Community partners] really are the experts on their community.
11. At first, I was not looking forward to it. The academic partners and I had figured out how to do the assessment quickly and I knew that it would take longer with a community partner. I also was apprehensive about working with someone I did not know. However, these were only minor concerns. I actually found I liked working with them because they provided insight I would not have gotten otherwise…. To me, the best part about conducting the assessment with a member of the community was that this person knew a lot more about the community than I did. Often times, there were store fronts that seemed to be closed or out of business. While I may have assumed that the store was out of business, my partner would be able to tell me that, in fact, it was still in business but didn’t open until later in the day. Additionally, the community partners knew members in the community. Thus, other individuals were more receptive to our work because they saw a familiar face was involved with the project.
12. Sometimes the community partners would get a little defensive if we differed. I think they assumed that if we got different counts, that they must be wrong. However, this was not always the case—I made mistakes sometimes too. For subjective ratings, we usually gave reasons for why we chose a certain category, and then decided based on the strength of each person’s arguments
13. I found working with [my partner] to be an extremely rewarding experience. She taught me so much about her history and culture and the history of her neighborhood and shared a lot of personal stories about her experiences living in New York. She inspired me with her commitment to her community and making positive changes for herself.
14. I worked with two different partners, and the experiences were quite different. One of them knew the area very well, and I was able to learn a lot about the area (e.g., one area had a lot of churches and she talked about each church). This partner was very easy going. The other partner I worked with disagreed with what we measured in the assessment, and I felt like I had to constantly justify why everything was important and why we recorded the data the way we did.
Data Collection
15. I really enjoyed it and didn’t feel it was a burden. I learned a lot about the areas I surveyed.
16. The only problem was that sometimes, they did not record information as precisely as the researchers did, or they would miss certain things (e.g., they would forget to also count trees, or they would only count some of them). I wondered if maybe they did not understand how important it is to be precise. However, I usually worked with the community partners on their first or second time out, so I think that they just needed more practice…. I think that there were often more discrepancies between academic and community partners’ data than there was between two academic partners’ data. I think that this was the result of either (1) the community partners just needed more practice or (2) sometimes there were buildings that the academic partner believed to be vacant/closed but the community partner knew was not. I expected that, perhaps, the community partners would be more biased in their judgments (on the more subjective measures, such as the condition of the sidewalks). However, this was not the case; the community partners provided accurate subjective measures. Overall, I think the data was more accurate having collaborated with the community partners…. There tended to be differences when we had to count things (e.g., number of residences). I think that this was largely because the academic partners had had more practice. We knew that if you drew a really good map, you could just count everything up at the end. When I worked with the community partners, they were still learning. Like I mentioned above, I expected the community partners to have difficulty providing objective measures of subjective things, but I found that this was not the case.
17. When we compared our maps, they looked the same. We counted the same number of trees, noted the same abandoned buildings, and recorded the same amount of time it takes to cross the sidewalk. At the same time, our maps were different. [my partner’s] research maps were alive with stories, insight, hopes, and disappointments. While we talked, I learned things about [her]—that she had arranged the planting of an impressive number of trees on one particular street in the hope that her children could walk in the shade of many trees and breathe cleaner air.

Interaction between partners was generally positive with only one less pleasant interaction described (Q2.17 and Q3.13). This same academic partner appeared to develop a strongly positive sense of the community (Q3.2). Further, some important perspectives of community partners came through to this individual, “By far, the biggest complaint I heard from members in the community was in regards to gentrification, … I also understood why gentrification posed such a threat” (Q3.4). These quotes suggest that the conflictual interaction did not prevent the academic partner form gaining empathy for the community perspective.

We observed complementary perspectives about the data collection between community and academic partners. From the community partners we heard some surprise about the highly detailed and meticulous nature of the data collection (Q2.19, 1.21, and 1.11) “at times it seemed too hair splitting.” Some academic partners were aware of this difference, “The only problem was that sometimes, they did not record information as precisely as the researchers did.” The academic who made the above comment nonetheless noted the similarity between the subjective judgments when rating, rather than counting, aspects of the environment, for example, the condition of the sidewalk (Q3.15).

DISCUSSION

We describe the approach of the Communities IMPACT Diabetes Center to bring together community and academic partners to undertake structured observations of a neighborhood as baseline data collection that was presented to the larger coalition for a community health improvement project.

In general, interactions were described as respectful, collegial, and fun. Academic partners learned about their community partners and the community. They appreciated the local knowledge of their partners and developed a sense of the richness and cohesiveness of the community. Local knowledge improved data collection (Q1.18). Community partners repeatedly commented about important details they had noticed that were new to them (Q2.10–12). Community partners appreciated the opportunity to learn how their community was seen through the eyes of someone who did not have specific ties within the community (Q2.9). Whereas community partners brought a longitudinal perspective rich in history, stories, and understanding, academic partners brought a focus on the details, specifications, and required mechanics of data collection, “they helped me be more professional, though at times it seemed too hair splitting” (Q2.21). Community partners appreciated the opportunity to see their community in fine relief, rather than solely as they typically experienced it.

As described by both community and academic partners, data collection was enhanced through partnered data collection. Resolution of differences was generally easy, and there were no observations of bias evident in the data collection, including regarding subjective judgments—most differences related to easily resolved errors in counts, uncertainties regarding application of the specifications, or differences in local knowledge, such as knowing that a shuttered restaurant was open only at night. The interviews reinforce our understanding that use of a data collection manual and form with specifications and examples was helpful.

One striking finding was the complex reaction of community partners toward growth and building. On the one hand, it was seen as progress; on the other, as creating a risk for undesirable gentrification that would be unaffordable for the community residents (Q2.4–6,8). Q2.7 offers a complex view of progress and change. The concern underlying antipathy to gentrification included both a concern about the loss of cultural heritage (Q2.5,8) and affordability (Q2.4). Community partners also note with concern the ongoing selling of green space in the area for construction, whether it will be used by community members or for gentrification.²⁵

The structured observations suggest why East Harlem may be a diabetogenic environment. Whether food stores and restaurants are primarily responsible for or are shaped by community tastes, the limited availability of produce in food stores and the preponderance of restaurants serving inexpensive, calorie dense processed food is noteworthy. One community partner noted, “It’s hard to be Hispanic and see that our food is not healthy. There were fast food restaurants that turn into obesity and diabetes” (Q2.6). Suboptimal sidewalks, variability in the extent of greening, relatively few parks and playgrounds in a very densely populated residential community, and a nearly absolute lack of benches indicate that the SEED is not an inviting place to go for a walk. Short and variable crosswalk times pose a pedestrian danger, particularly for those who are frail or slow.

This paper describes a single instance of data collection; thus, our findings may be idiosyncratic. Nonetheless, we believe that we have valuable lessons to share.^26–28 We demonstrate the opportunity for partnered data collection to build relationships and to enhance the quality of data. The qualitative survey data suggest the value of discovery capacity when assessing novel partnerships.²⁹ The authors believe that the highly partnered design and execution of this baseline data collection both reinforced our overall approach to partnership and engagement, and built relationships that furthered the overall conduct of the partnership. Using rigorous evaluation as a means to build partnerships is unusual, if not novel. Although our findings are unlikely to be representative of all urban minority communities, they come from one with a well-established culture and hence may represent a more positive picture than typical for a low socioeconomic status, urban, minority environment.

Our qualitative survey was conducted nearly 1 year after the actual data collection. As a result, several academic partners had moved on and were not available to be surveyed. Although some detail was lost, what remains is likely to include those perceptions that were most meaningful; the timing of our data collection remains a limitation.

The structured observations are limited by our walking only on the block faces and not in the public housing developments. The Evaluation Committee grounded this decision in concerns of respect for the community and safety of the observers. Our findings did not catalog a large green space containing not well maintained playgrounds, plazas, and game courts. Food selection in restaurants and the stores describes only those establishments that would allow our observers inside. The impression of the observers was there was not any systematic exclusion from any particular type of establishment serving or selling food.

LESSONS LEARNED

We present a novel approach to community engaged data collection in the context of a baseline assessment for a community health improvement project. We learned that data collection may be improved by integrating community and academic partners; on-the-spot resolution of differences in observations is feasible; and both types of partners found added value. Qualitative evaluation of the experience proved valuable. Data collection both mirrored and reinforced our philosophy, enhancing the partnership moving forward. Our findings identified aspects of the community that may contribute to high rates of obesity and diabetes, including ready access to cheap calorically dense prepared foods and limited access to comfortable venues for walking or playing. Community partners offer critical insights into the interpretation of these data, and their participation in both data collection and program planning assures their ability to bring both the data and community perspectives into the health improvement work. Ultimately, translation of baseline assessment into interventions was served by the close relationship of both community and academic partners to the baseline data, its collection, and its interpretation.