Environmental health perceptions of urban youth from low‐income communities: A qualitative photovoice study and framework

Abstract Background Children are amongst the most susceptible groups to environmental exposures, for both immediate and life‐course health outcomes. Despite their increased susceptibility, children's knowledge, experiences and voices are understudied. A deeper understanding of children's environmental health perceptions has the potential to better inform policy, develop targeted interventions and improve public health outcomes. Methods In this study, our community–academic partnership used the Photovoice research method to examine how urban children from low‐income communities perceive environmental influences on their health. Twenty children, ages 10–12, took photographs and participated in focus group interviews regarding their perspectives on how the environment influences their health. Results Qualitative analyses revealed five major thematic categories: environmental exposures, environmental health sentiments, environmental health outcomes, interest in environmental health and environmental health solutions. We used the findings to develop an environmental health perspective theoretical framework that can inform future work designed to promote the environmental health and well‐being of children from low‐income communities in urban communities. Conclusion Photovoice enabled children from low‐income communities to capture and communicate their environmental health perceptions. These findings have the potential to inform and identify potential targets and opportunities for environmental health interventions and promotion in their communities. Patient or Public Contribution Partnerships with community‐based organizations were central to the present study. By design, these community‐based partners were involved in the conduct and procedures of the study.


| INTRODUCTION
Children are amongst the most susceptible groups to environmental exposures. 1 Childhood environmental exposures have significant implications for immediate health outcomes as well as health outcomes throughout the life course. 2 In the United States (US), there are striking racial, ethnic, and socioeconomic disparities in childhood environmental exposures. 3 Specifically, compared to their White, non-Hispanic, or high-income counterparts, Black, Hispanic, or children from low-income communities consistently experience higher exposure to environmental hazards and lower exposures to health-promotive environmental facets. 4,5 Such environmental hazards include exposure to lead-based paint, tobacco smoke, ambient and traffic-related air pollution, and living in closer proximity to hazardous waste. 6 Examples of promotive environmental facets include nature contact, tree shade, clean neighbourhoods, and access to healthy food. 7,8 Furthermore, individual and community-level perceptions of the environment and associated health impacts play an important role in how individuals engage with their surroundings. [9][10][11] Environmental health perceptions may influence the diet, physical activity, and physical safety of community members, especially children, and thus impact their development and behaviours later in life. 6,12 Currently, there is a substantial knowledge gap in the literature as children's environmental health perceptions are critically understudied. Researchers must engage, examine, and incorporate children's environmental health perceptions into public health and public policy discourse to better improve health outcomes, especially in urban and low-income communities.
Photovoice is a community-based participatory action research methodology in which researchers can understand and incorporate childhood environmental health perceptions. [13][14][15] Briefly, Photovoice engages research participants by asking them to take photos on a given topic and later asking them to discuss these photos during a focus group. [13][14][15] The Photovoice research method has been traditionally viewed as an avenue through which marginalized communities can participate in academic research and ultimately have influence over policy decisions. 13,16,17 Photovoice allows for collaboration between community members and researchers by empowering community members to capture and communicate their perceptions and knowledge. 18 Photovoice has been shown to be an effective tool for empowering children and youth. 15,19,20 For example, Photovoice enabled minority New York City youth to reflect on food justice issues and engage in promoting positive community changes. 16 Photovoice has also been used as a participatory process for research and social change allowing children to feel 'seen' by adults. 19 Given its participatory focus, Photovoice may be a valuable tool for allowing children to express their environmental health concerns and perceptions. Therefore, the purpose of the current study was to identify and characterize the environmental health perceptions of urban, low-income US children. We utilized Photovoice methods to engage a sample of children from St a student body that is 95% or more on the free and reduced lunch programme. 15,21,22 This partnership was established as St. Louis Public School students face disproportionate amounts of environmental and behavioural health disparities compared to students in neighbouring school districts. 23 The environmental education intervention run by this community-academic partnership has been described in detail in prior publications. 15,21 Briefly, the intervention consists of (1) weekly interactive, in-class science, technology, engineering, art and math lessons focused on environmental health and (2) monthly nature-based field trips.

| Photovoice environmental education intervention
The students participated in a Photovoice community-based participatory research project (Photovoice activity). During the first class of the intervention, the students received disposable cameras and then participated in a 60-min lesson on how to use said cameras as well as how to obtain consent when photographing other humans. After the 60-min lesson, the students were instructed to take photographs of how the environment impacts their health. The students returned their cameras on week 2 of the intervention. On week 3, the developed photographs were brought to the classroom and the students were individually interviewed about the photographs that they took by trained research assistants.
We conducted individual interviews instead of focus groups because children may have been less likely to disclose their perspectives with their peers present. The interviews were semistructured interviews to explore children's perspectives in depth.
Sixty minutes were allotted for the interviews during week 3 of the intervention. Interview questions focused on what the student captured in the photograph, how the photograph related to environmental health and how the photograph made the child feel (an interview guide is available upon request from the senior author).
The interviews, which ranged from 20 to 52 min long, were recorded and later transcribed.

| Data analysis
The study used a pragmatism paradigm. 24 As such, we undertook a mixed thematic analysis approach that aimed to identify and characterize environmental health perceptions through both codebook and coding reliability approaches. Our thematic analysis approach was adapted from Castleberry and Nolen 25 and Braun and Clarke. 26 Briefly, Castleberry and Nolen's approach, widely used in the health sciences, consists of five steps to thematic analysis: compiling, disassembling, reassembling, interpreting and concluding. 25 Our approach also utilized what Braun and Clarke term 'coding reliability' and 'codebook' thematic analysis. 26 These approaches allowed for both a detailed accounting of participants' environmental health perceptions (coding reliability) and inductive theme development (codebook).
Four trained researchers familiarized themselves with the data through a preliminary analysis of the transcripts. Then, the four researchers independently brainstormed potential thematic codes based on the transcripts. After, the four researchers met and discussed their potential thematic codes. From that meeting, the researchers developed an official list of themes that were derived from the manuscript. A fifth researcher, who was purposefully excluded from the initial meetings, revised and approved the final list of themes. Once the final list of themes was approved, two researchers independently coded every transcript for every theme. Research checklist in reporting this study (Supporting Information: Appendix 1). 29 Lastly, using the subjectivist inductive approach to research, results from this study were used to develop a youth-informed environmental health perspectives theoretical framework. 30 The framework is presented in the discussion section of this paper.

| RESULTS
A total of 20 students participated in the current study. Of the 20 participants, 8 (40.0%) students were Black, 5 (25.0%) were White, 1 (5.0%) was Asian, 4 (20%) were two or more races and 2 (10.0%) were self-identified as other. The median age of the study participants was 12.
The thematic categories from the qualitative data analysis are presented in Table 1

| Built environment
Participants also described aspects of their environment that were structural or manmade, as opposed to exposures to nature. Examples included participants' schools, homes and other aspects of their built urban environment. One student described enjoying travelling on the highway, saying, 'I feel kind of happy, 'cause it's sunny, and whenever we're driving to, like, Walmart you can just smell the fresh air as we're driving on the highway, and it just always gives me a good vibe for some reason. I just feel happy knowing we're on the highway' (participant 3). This quote demonstrates the feelings of joy the student receives when engaging with features of the built environment (a highway and car).

| Climate change
Climate change and its influence on health were also discussed. Some students discussed their personal experiences with climate change in their own environment, while others discussed climate change on a global level. For example, one student voiced concerns about the health impacts of forest fires, as she stated 'Australia is on fire! People are losing their homes and it can't be good for breathing' (participant 4). This quote shows that students are aware of the global effects of climate change on human health and wellbeing.

| Pollution and waste management
Pollution and waste management were identified as environmental exposures that impact health. Specifically, students focused on exposure to air pollution, litter and noise pollution. Students discussed being exposed to air pollution. For example, one participant took a photo of a car's exhaust pipe (participant 5). Another student took a picture of his local Walmart and described the air as 'polluted'

| Food environment
Participants' mentions of nutrition and diet, and the effects of nutrition on their health, were categorized in the nutrition subtheme.
One student took a picture of water and stated that 'water is actually T A B L E 1 Themes and subthemes from Photovoice focus groups.

| Interest in environmental health topics
Participants expressed differing levels of engagement and curiosity toward their environment and its effect on their health. We coded these emotions into two subthemes: inquisitive and apathetic.

| Inquisitive
Participants expressed curiosity toward the elements captured in their photos and how they might affect their health. When asked how the trees she captured affected her health, one participant (1) asked,

| Apathetic
Participants also expressed apathy towards their photos and/or the concept of environmental health in general. When asked why their photos were important to them, some participants could not name a reason. Others declared their photos unimportant altogether. When asked how the environment affects their health, seven participants responded with 'I don't know'. This response indicates that some participants had never previously considered the impact of their environment on their health and did not express interest in doing so during the interview. While the participants had feelings of apathy toward how their environment impacts their health, they were able to discuss their photographs in detail.

| Community-level solutions
Other participants offered ideas to improve their community, neighbourhood or city. These solutions would often require a group actor, such as a government or neighbourhood association, to offer more environmental amenities. For example, one participant stated that they would, '…get more trees and have all the houses more away from each other. I don't know. Just have more trees, more grass, and just more stuff where people can breathe better and just have more space to have fun and stuff, I guess, outside' (participant 1). In this quote, the student is demonstrating a community-level intervention to improve overall environmental health. health, well-being, attention, cardiovascular, pain, vision and respiratory). 36,37 Alternatively, exposure to environmental toxins may cause negative health outcomes. 38 For example, living in a neighbourhood with limited access to supermarkets and fresh produce has been associated with increased risks for obesity and related health outcomes. 39,40 Exposures to ambient and household air pollution have been associated with a number of adverse health outcomes (including asthma, respiratory disease, cardiovascular disease, lung cancer and mortality). 41,42 There is a strong existing body of evidence base supporting the causal association between environmental exposures and health outcomes.

| Global solutions
Research has also highlighted the significant impact of environmental health sentiments on environmental health outcomes. 32,33 One study investigated the relationship between environmental exposures, perceptions, and outcomes for indoor and outdoor air quality in Seoul, Korea by collecting data and surveying 396 elementary school students (average age = 11) and 64 parents. 32 This study found that environmental health sentiments significantly influenced environmental health outcomes, even in areas with comparable environmental exposures. 32 The current evidence base supports our child-inform theoretical framework, with both environmental exposures and perceptions influencing outcomes ( Figure 2).
Also shown in the theoretical framework (Figure 2) Michigan. 44 This crisis elevated the awareness and issue of lead poisoning and ignited a public outcry worldwide. 45,46 Studies suggest that these events caused an increase in the public interest. 46 In the United States, this elevated awareness (concern over lead water in drinking water) forced school districts to take action and create environmental health solutions (testing school drinking water and shutting down any drinking fountain with potential lead pollution), 46,47 leading to calls for more community-oriented, creative and intersectional solutions. For children, evidence suggests that engaging in environmental health education is an effective intervention for increasing inquisitive interest in young people, 48 and could thus lead to informed solutions.
Our theoretical framework indicates that environmental health solutions may lead to new, positive environmental exposures ( Figure 2). For example, the development of new greenspaces may promote improved mental and physical health outcomes. 49 One study investigated the changes in children's physical activity patterns before and after a large-scale playground greening intervention at a low-income Los Angeles Public School. 49 The study found a 10.0% decrease in sedentary activity and a 48% increase in vigorous activity participation after the playground greening intervention was complete. 49 This is an example of an environmental health solution (greening playgrounds) altering the environmental exposure (playgrounds and greenspace) and, in turn, altering environmental health outcomes (physical activity). 49 The main strengths of this study include its use of participatory research methods to examine children's perceptions and develop a children-informed environmental health perspective theoretical framework. Using a mixed thematic analysis approach to develop thematic codes allowed for the children's voices to lead the research, rather than deferring solely to a potentially biased predetermined framework. The limitations of this study should also be noted. First, this study was conducted during the COVID-19 pandemic, which may have limited children's outdoor activities and therefore influenced perceived environmental exposures and sentiments.
Another limitation includes the transferability of these study findings. should be considered as the first steps in a series of research strategies aimed at improving our understanding of child environmental health outcomes through improved measurements and assessments. As such, our findings have the potential to inform and identify potential targets and opportunities for interventions aimed to promote child health in an urban, low-income community.