PREVENTING CHRONIC DISEASE PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY Volume 17, E126 OCTOBER 2020 ORIGINAL RESEARCH Quality of Life Benefits of Urban Rooftop Gardening for People With Intellectual Disabilities or Mental Health Disorders Margarita Triguero-Mas, PhD1,2,3,4; Isabelle Anguelovski, PhD1,2,3,4,5; Judith Cirac-Claveras, MSc1,2,3,4; James Connolly, PhD1,2,3,4; Ana Vazquez, BSc1; Ferran Urgell-Plaza, MSc6; Núria Cardona-Giralt, BSc6; Esther Sanyé-Mengual, PhD7; Jordi Alonso, PhD3,8,9; Helen Cole, PhD1,2,3,4 Accessible Version: www.cdc.gov/pcd/issues/2020/20_0087.htm Methods Suggested citation for this article: Triguero-Mas M, We collected quality of life information with a preliminary ver- Anguelovski I, Cirac-Claveras J, Connolly J, Vazquez A, Urgell- sion of the INTEGRAL Scale questionnaire from all gardeners (n Plaza F, et al. Quality of Life Benefits of Urban Rooftop = 54) and among a comparison group of nongardeners (n = 43). Gardening for People With Intellectual Disabilities or Mental We also conducted semi-structured interviews with participants Health Disorders. Prev Chronic Dis 2020;17:200087. DOI: https:// and technicians, and made field observations. doi.org/10.5888/pcd17.200087. Results Our results indicated that urban rooftop gardening was associated PEER REVIEWED with better personal development and suggested enhanced physic- al and emotional well-being, sense of purpose, social inclusion, in- Summary terpersonal relations (including new perspectives on the urban en- What is already known on this topic? vironment and the changes in social roles), and general quality of Community gardens can improve gardeners’ health but no research has life. focused on quality of life, rooftop gardens, or people with intellectual dis- ability or mental health disorders. Conclusion What is added by this study? Our study extends the evidence on the potential benefits of urban We evaluated 2 rooftop gardens in Barcelona and collected quantitative rooftop gardening in general, and specifically for those with intel- and qualitative data from January through June 2018. Gardening was as- sociated with personal development and may be related to emotional well- lectual disabilities and mental health disorders. being, physical well-being, social inclusion, interpersonal relations, sense of purpose, and overall quality of life. Introduction What are the implications for public health practice? Rooftop gardening can be used to promote quality of life among people The growing inclusion of community gardens in urban planning with intellectual disabilities or mental health disorders. has extended beyond vacant land. Buildings are increasingly seen as sites of cultivation (1–3) through indoor planting, balcony and terrace beds, urban rooftop gardens, green walls and vertical farm- Abstract ing (4), zero-acreage farming (5), building integrated agriculture (6), and skyfarming (7). Background The number of urban community gardens, including those on Community gardens can improve gardeners’ general health (8), rooftops, is increasing. However, few studies have explored the lower their obesity rates (9,10) or depression (11), and improve benefits of these gardens for people with intellectual disabilities or mood (8) or satisfaction with life (12,13). These health benefits mental health disorders. We evaluated the association between may derive from the restorative capacity of nature (14), aesthetic urban rooftop gardening and quality of life of individuals with experiences (15–18), improved social relationships, emotional moderate to very marked disability. connections, political engagement (16–18), provision of sense of The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 1 This publication is in the public domain and is therefore without copyright. All text from this work may be reprinted freely. Use of these materials should be properly cited. PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 freedom (19), encouragement of physical activity (12,16), or re- ilities from 3 different occupational centers. The other rooftop shaping the nutritional environment (15,16,18). garden (the “recent garden”) began operating in November 2017 and served individuals from 2 occupational centers, one that sup- Few studies have explored the potential health benefits of garden- ported people with intellectual disabilities and one that promoted ing for people with intellectual disabilities or mental health dis- the social integration of people with mental health disorders. orders, despite the fact that this population experiences additional limitations to participation and inclusion in urban projects and is Each occupational center was assigned a different morning of the more likely than others to have poor health outcomes such as dia- week to work in the garden. The participants engaged in planting, betes and obesity (20,21). For example, studies indicate that learn- harvesting, maintaining plants, and distributing produce with the ing and working around plants improved depressive symptoms, help of educators from the occupational centers and sometimes anxiety, sociability, and stress among people with depression or rooftop technicians. children with intellectual disabilities (11,22–25). One study found that gardening improved mood and was associated with higher en- Design and measures gagement in activities between people with dementia (26). Anoth- We used a mixed-methods design, incorporating both quantitative er study documented the potential of community gardens to foster and qualitative data collected from January through June 2018. All learning, occupational participation, and social inclusion among participants and their legal tutors were informed about the study individuals recovering from mental illness (27). according to standard procedures, and each participant or tutor However, most of these studies use quantitative methods and men- gave written informed consent before participation in the study. In tion qualitative findings as anecdotal evidence without incorporat- the case of individuals with moderate to very marked disability, ing them formally (23,26). One exception is the study of Kam and only those able to walk independently and to communicate Siu, which focused on people with psychological illness and incor- through spoken language (able to communicate preferences, porated semi-structured interviews (28). wants, and needs) were included. The study was approved by the plenary board of the Barcelona Town Council (according to art- In this study, we used a mixed-methods design to address these icle 201.1.d). gaps in the literature and to assess the extent to which participat- ing in an urban rooftop gardening program may be associated with Quantitative data collection better quality of life for individuals with moderate to very marked We purposively selected individuals, and used the pilot rooftop disability derived from intellectual disabilities and mental health gardens as the intervention group. For the comparison group, we disorders. invited all those individuals that were attending the same occupa- tional centers as the rooftop garden users and who had similar Methods demographic characteristics (ie, sex, age, and degree of disability) but who did not garden. In total, 54 individuals from the interven- Rooftop gardens program tion group (32 in the initial and 22 in the recent garden) and 43 in the comparison group (21 in the centers using the initial garden We evaluated 2 urban rooftop gardens from a Barcelona city coun- and 22 in the centers using the recent garden) agreed to participate cil pilot program designed for Barcelona citizens with moderate to in the study. very marked disability, defined as those with 33% or greater de- gree of disability. The disability degree was based on the actual We collected information on quality of life (Chronbach’s α = .90) disability (ie, having difficulties in any of the following areas: and 8 of its dimensions: material well-being (α = .66), physical body functions and structures, activities, and involvement in cer- well-being (α = .78), emotional well-being (α = .69), self- tain areas of life [29]) and the factors that may affect that person’s determination (α = .69), personal development (α = .60), interper- social integration, such as with regard to family, working situ- sonal relationships (α = .62), social inclusion (α = .61), and rights ation, education, and other cultural factors. Disability degree was (α = .70). To do so, we administered a preliminary version of the established according to the technical guidelines defined by the subjective subscale of the INTEGRAL Scale questionnaire (31). Spanish government (30). The evaluated gardens were located on More details can be found in the Appendix. For logistic reasons, municipal administrative buildings in 2 different neighborhoods of we collected quality of life information once in the initial garden Barcelona, Spain. (at 22 months) and twice in the recent garden (at 3 and 7 months of exposure to the intervention). One of the rooftop gardens opened in April 2016 (the “initial garden”) and brought together participants with intellectual disab- The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. 2 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2020/20_0087.htm PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 For each participant, we also collected information on age, sex, Then we grouped the findings into 3 dimensions of quality of life: disability degree (defined as moderate for those with 33% to 64% physical well-being, interpersonal relations and social inclusion, degree of disability, marked for those between 65% and 74%, and and emotional well-being (including personal development and very marked for those between 75% and 100%), and involvement having a sense of purpose in life). in nongardening indoor or outdoor activities organized by the oc- cupational centers. Results Qualitative data collection Quantitative results Apart from 12 field visits to the evaluated gardens, we conducted 56 semi-structured interviews with all key stakeholders involved Most participants in the intervention group were male and had a in the gardens who wanted to participate in this part of the evalu- marked or very marked degree of disability, and characteristics of ation study: 48 interviews with individuals from the intervention participants also differed between the initial and recent garden group and 8 with rooftop technicians and educators from the occu- groups (Table 1). Participants in the intervention and comparison pational centers. Semi-structured interviews were guided by pre- groups had similar characteristics but significantly differed ac- pared open-ended questions that included all aspects of the garden cording to the nongardening indoor and outdoor activities in which activity (Appendix). those assigned to the initial garden participated. Analysis In the adjusted models for individuals from centers that used the initial garden, we found higher scores of emotional well-being for For our quantitative data, we used bivariate analyses to explore the intervention group than for the comparison group (P = .05) possible differences in characteristics between rooftop gardens and (Table 2). For individuals from centers assigned to the recent between intervention and comparison groups. We developed mul- garden, we found that those who were exposed to the rooftop tivariate linear regression models to estimate the association garden for 7 months had personal development scores 1.30 (95% between each quality of life dimension and exposure to the inter- confidence interval, 0.27–2.33) points higher than the comparison vention and fitted separate models for each garden. We hypothes- group after adjusting for sex, age, disability degree, and personal ized that the effect of the pilot rooftop gardens on quality of life development score at 3 months. We also found higher quality of would be strong at the beginning of the exposure and stable after life after 7 months for the intervention group compared to the that. Consequently, we expected the effects to be different comparison group in the recent garden (P = .07). between the 2 gardens and our data to be better modeled separ- ately, by garden, because of differences in exposure times (the ini- Qualitative results tial garden was initiated 19 months before the start of the evalu- Physical well-being of rooftop gardeners. Gardening was directly ation and the recent garden immediately before the start of the linked to mental and physical energy, and participants looked for- evaluation) and differences in data availability (we had data on ward to the activity every week. According to results of the semi- quality of life at 22 months in the initial garden and at 3 and 7 structured interviews, gardening helped participants overcome months for the recent garden). We also fitted separate models for some of their limitations. Rather than being physically inactive or total score and each of the 8 dimension-specific scores. Initial passive, gardening gave the individuals an incentive to leave their models included all variables (occupational center, sex, age, disab- homes to garden (Table 3). ility degree, indoor activities, outdoor activities, being exposed to the intervention, one of the quality of life scores in each model Interpersonal relations and social inclusion for rooftop gardeners. and, in the recent garden, the corresponding quality of life score at Social educators reported that, compared with people not exposed 3 months) and possible interaction terms. From these, we fol- to the intervention, people participating in the rooftop project lowed a stepwise process to determine the final simplified models managed to build more relationships through cooperation and with the only restriction being that exposure to the intervention friendship and socialize more with one another outside of typical and sex, age, and disability degree should always remain in the fi- short or formal interactions. This was true even among people nal models. We used R statistical package version 3.6 (R Founda- with difficult relationships and those who were most isolated or tion), and significance was set at P < .05. lonely (Table 3). We analyzed our qualitative data by using a thematic analysis that This finding seemed to be due to the nature of gardening work it- identified the quality of life aspects and associated mechanisms. self (ie, including collective interdependent tasks) and particularly when gardeners met participants from other occupational centers on the rooftop site. In addition, gardeners seemed to be particu- The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 3 PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 larly enthusiastic about the new contacts with the rooftop techni- to plan and accomplish, so they often ended up being the ones who cians, and many also appreciated the casual conversations with the tended to propose new tasks and next steps in the production and staff members of the municipal building on which the gardens harvesting cycle and who suggested new activities. This level of were located (Table 3). engagement empowered them and gave them a greater sense of ownership of and attachment to the activity (Table 3). Gardeners witnessed that their work contributed to fresh food they could bring to other residents (including their family, municipal Educators also reported that gardening helped the participants workers, and other vulnerable people). Moreover, their gardening learn about frustration, the importance of flexibility, the ability to activity was of interest to others (eg, schools visit the gardens and overcome disappointments, and emotional control. For example, learn about them via the gardeners themselves), so gardeners be- the transfer of knowledge and skills about gardening translated in- came providers of educational information. Both aspects made to participants’ learning about the life cycle of plants, and about gardeners realize that they were able to volunteer their time, in- natural events such as disease or death, which they then were bet- form and give to others, and become “contributors to society and ter able to accept and integrate into their daily lives. Over the its needs” rather than recipients of social aid and support (Table course of a production cycle, gardeners became more patient, less 3). stressed-out, and more relaxed. Such learning processes and dy- namics were particularly powerful because they were achieved in- The rooftop project also transformed gardeners’ day-to-day dependently, with little direction. Several gardeners were also able routines and activities by exposing them to new modes of urban to transfer their new personal management skills to environments transportation (ie, arriving at the garden by public transit rather outside the gardens, as educators highlighted. Those benefits ac- than private transport from their families or from the social crued over time, highlighting the importance of a longer-term and center), to new neighborhoods and elements of the city, and to continuous involvement of participants. new views and experiences of their city. Such experience contrib- uted to new spatial links and approximations for them within the Gardening tasks also involved the ability to become creative and urban space (Table 3). find new ways to resolve emerging problems during the growing cycle. As participants became more familiar with the different Moreover, tactile engagement with plants, insects, and other biota stages of growing food and the different responsibilities involved allowed gardeners to explore new senses and feelings and connect and became able to explain them using their own words, they differently with their environment. These contacts were valued as gained feelings of security (Table 3). a way to disconnect from city disturbances (Table 3). Gardening was a gratifying and rewarding activity that contrib- Emotional well-being, personal development, and sense of uted to happiness and a healthy lifestyle. Respondents explained purpose for rooftop gardeners. Lastly, gardening was associated that gardeners wanted to come back and were motivated to contin- with enhanced individual discovery and freedom; self-confidence, ue taking part in the activity and be fully engaged in it through the autonomy, emotional strength, and problem resolution; and hav- months. Many gardeners looked forward to gardening every week ing a purpose in life. and seeing how plants had grown and what work was needed in Respondents indicated that gardens offered participants the possib- the garden. Gardening emerged as a meaningful activity and as a ility of exploring unknown outdoor environments in a much freer way for participants to find themselves (Table 3). and less formal manner than they usually would as participants at the occupational center, where they are constrained by norms, Discussion rules, and schedules (Table 3). According to respondents, gardens We evaluated whether participating in an urban rooftop gardening were a space where the users were not constantly controlled and program was linked to better quality of life for individuals with directed by social workers, therapists, or family members and moderate to very marked disability derived from intellectual disab- where they did not feel judged or looked down on for their atti- ilities and mental health disorders. Both our qualitative and quant- tudes, choices, or simply the way in which their disability mani- itative results indicated that gardening was associated with person- fested. al development and emotional well-being (eg, feelings of freedom; Engaging in the different steps of the gardening activity helped enhanced autonomy, capacity to deal with frustration and disap- gardeners gain overall autonomy and self-esteem over time. These pointment). Moreover, our qualitative findings suggested that benefits were partly associated with gardening being a space for gardening was linked to physical well-being (increased energy and trial and error (Table 3). Over time, gardeners gained initiative. getting around one’s own limitations), social inclusion and inter- They became aware of the care and production tasks they needed personal relations (new and improved relationships with others, re- The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. 4 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2020/20_0087.htm PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 formed relationships with the city, and acquiring a new role in the healthier nutritional habits (18,35,36), or decreased physical con- society), and sense of purpose (such as finding themselves). Our straints, acute health complaints including sleep, chronic illnesses, quantitative results also indicated that gardening may be associ- and consultations with a general practitioner (12) for the general ated with quality of life overall. population or those participating in a cardiac rehabilitation pro- gram. Our finding of an association between gardening and better emo- tional well-being, personal development, and having a purpose in This study has several strengths. First, it is the first empirical eval- life are in line with previous research. One study found that self- uation (to our knowledge) of a rooftop gardening program that is control in children with intellectual disorders who were exposed to farmed by urban residents with disabilities. It also provides novel gardening activities was better than children who were not ex- results about the beneficial effects of urban rooftop gardening on posed, which is in agreement with our findings on gardening and quality of life (mainly the new perspectives on the urban environ- emotional control (25). Our findings about the links between ment and the changes in social role). To our knowledge, it is only gardening and self-confidence, multi-layered learning, and con- the second study on gardening and quality of life or health using a nection with nature also coincide with those of previous studies mixed-methods design and triangulated data. that focused on similar population groups (23,27,28). Gonzalez et al also suggested that therapeutic horticulture could help people Our study also has limitations. Our quantitative data sample size with depression or bipolar disorder find a life purpose, with was small (despite including all the users), which limits our statist- gardening perceived as a meaningful activity that contributes to ical power and, consequently, the strength of our conclusions, par- changing participants’ view of life (22). Moreover, our findings ticularly in relation to potential interactions. We were also unable that gardening is associated with increased freedom, autonomy, to include pregardening scores of quality of life in our models, be- initiative, and creativity build on previous research, highlighting cause our quantitative data collection started after gardening had the flexibility of community gardens, offering choices, and re- begun in all pilot urban rooftops. It was not possible to control specting participants with mental illnesses’ autonomy and pace participants’ level of engagement with the activity, because we (27). could not control the number of participant visits to the gardens. However, we know that individuals were going to the garden Our observed results of better social inclusion and interpersonal every week whenever possible since it started operating, so we ex- relations possibly being linked with gardening are in line with pre- pect that measurement error, if any, is low. Our qualitative data vious findings on social skills, particularly cooperative and col- also did not include information from relatives of the gardeners lective skills, and extension of social networks for children with because of logistic limitations. intellectual disabilities (25) and adults with severe mental ill- nesses (27,28). Improved social well-being has also been linked to Future studies should try to replicate our findings with similar the capacity of gardening to bring people together and create con- mixed-methods approaches. They should also evaluate the poten- nections with the wider community (27). tial differential effect of diverse exposure lengths, repeated expos- ures, frequency, and how long benefits are sustained. However, our results indicating that higher general quality of life may be associated with participating in urban rooftop gardening We explored the range of quality of life benefits associated with contrasts with findings of a previous study (28), which found no urban rooftop gardening by individuals with intellectual disabilit- association between quality of life changes and horticulture ther- ies and mental health disorders. Our findings indicated better per- apy among individuals with mental health or intellectual disorders. sonal development and suggested enhanced physical and emotion- The study based its quality of life measurement on only 7 ques- al well-being, sense of purpose, social inclusion, interpersonal re- tions, compared with our 47-item scale, which could explain the lations, and general quality of life associated with gardening. This difference in our findings. Other studies that used an assessment study extends the existing evidence on the potential benefits of tool similar to ours found results similar to our study for the asso- gardening. It also suggests that interventions such as ours are op- ciations between access to a garden and improved quality of life portunities to promote green, healthy, and equitable cities for a (32). variety of residents. Rooftop gardening may enhance health equity in a broad sense, providing a space where those with mental and We are not aware of comparable results regarding the possibility intellectual disabilities can improve their quality of life. that enhanced physical well-being is related to gardening for indi- viduals with mental health disorders or intellectual disabilities. However, we can draw some comparisons with recent studies finding associations between gardening and vigor/energy (33,34), The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 5 PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 Acknowledgments 5. Specht K, Siebert R, Hartmann I, Freisinger UB, Sawicka M, Werner A, et al. Urban agriculture of the future: an overview We thank all the study participants. J.C. and M.T.M. are funded by of sustainability aspects of food production in and on Juan de la Cierva fellowships (no. IJCI-2016-31100, FJCI-2017- buildings. Agric Human Values 2014;31(1):33–51. 33842) awarded by the Spanish Ministry of Economy and Com- 6. Caplow T. Building integrated agriculture: philosophy and petitiveness. The research presented in this article received fund- practice. In: Urban futures 2030: Visionen künftigen ing from the Town Council of Barcelona and from the European Städtebaus und urbaner Lebensweisen. [Urban futures 2030: Research Council under the European Union’s Horizon 2020 re- urban development and urban Lifestyles of the future.] Berlin search and innovation program (grant agreement no. 678034). (DE): Heinrich-Böll-Stiftung; 2009. These sponsors had no role in the design or analysis of this study. 7. Germer J, Sauerborn J, Asch F, de Boer J, Schreiber J, Weber No copyrighted material was used in this manuscript. G, et al. Skyfarming an ecological innovation to enhance global food security. J Für Verbraucherschutz Leb 2011; Author Information 6(2):237–51. 8. Wood CJ, Pretty J, Griffin M. A case-control study of the Corresponding Author: Margarita Triguero-Mas, PhD, Universitat health and well-being benefits of allotment gardening. J Public Autònoma de Barcelona, C/ Doctor Aiguader 88, 08003, Health (Oxf) 2016;38(3):e336–44. Barcelona, Spain. Telephone: 011-34-93-316-0711. Email: 9. George DR, Rovniak LS, Kraschnewski JL, Hanson R, mtrigueromas@gmail.com. Sciamanna CN. A growing opportunity: community gardens affiliated with US hospitals and academic health centers. Prev Author Affiliations: 1 Universitat Autònoma de Barcelona, Med Rep 2015;2:35–9. Barcelona, Spain. 2 Institute for Environmental Science and 10. Zick CD, Smith KR, Kowaleski-Jones L, Uno C, Merrill BJ. Technology, Barcelona, Spain. 3IMIM (Hospital del Mar Medical Harvesting more than vegetables: the potential weight control Research Institute), Barcelona, Spain. 4Barcelona Lab for Urban benefits of community gardening. Am J Public Health 2013; Environmental Justice and Sustainability, Barcelona, Spain. 103(6):1110–5. 5 ICREA (Institució Catalana de Recerca i Estudis Avançats), 11. Ghanbari S, Jafari F, Bagheri N, Neamtolahi S, Shayanpour R. Barcelona, Spain. 6Municipal Institute for People with Disabilities, Study of the effect of using purposeful activity (gardening) on Barcelona City Council, Barcelona, Spain. 7Joint Research Centre, depression of female resident in Golestan Dormitory of Ahvaz European Commission, Ispra, Italy. 8Universitat Pompeu Fabra Jundishapur University of Medical Sciences. J Rehabil Sci Res (UPF), Barcelona, Spain. 9CIBER Epidemiología y Salud Pública 2015;2(1):8–11. (CIBERESP), Barcelona, Spain. 12. van den Berg AE, van Winsum-Westra M, de Vries S, van Dillen SM. Allotment gardening and health: a comparative References survey among allotment gardeners and their neighbors without an allotment. Environ Health 2010;9(1):74. 1. Cohen N, Reynolds K, Sanghvi R. Five Borough Farm: 13. Min S, Ha Y-J, Kang J-H, Kang H-Y. The effects of seeding the future of urban agriculture in New York City. New horticultural therapy on the well-being and hope of women in York (NY): Design Trust for Public Space; 2012. rural Korea. J Nurs Care 2013;03(06). 2. Orsini F, Dubbeling M, de Zeeuw H, Prosdocimi Gianquinto 14. Soga M, Gaston KJ, Yamaura Y. Gardening is beneficial for G., editors. Rooftop urban agriculture. New York (NY): health: a meta-analysis. Prev Med Rep 2016;5:92–9. Springer International Publishing; 2017. 15. Alaimo K, Beavers AW, Crawford C, Snyder EH, Litt JS. 3. Sanyé-Mengual E, Anguelovski I, Oliver-Solà J, Montero JI, Amplifying health through community gardens: a framework Rieradevall J. Resolving differing stakeholder perceptions of for advancing multicomponent, behaviorally based urban rooftop farming in Mediterranean cities: promoting food neighborhood interventions. Curr Environ Health Rep 2016; production as a driver for innovative forms of urban 3(3):302–12. agriculture. Agric Human Values 2016;33(1):101–20. 16. Egli V, Oliver M, Tautolo S. The development of a model of 4. Despommier D. The vertical farm: controlled environment community garden benefits to well-being. Prev Med Rep 2016; agriculture carried out in tall buildings would create greater 3:348–52. food safety and security for large urban populations. J Für Verbraucherschutz Leb 2011;6(2):233–6. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. 6 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2020/20_0087.htm PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 17. Hale J, Knapp C, Bardwell L, Buchenau M, Marshall J, Sancar 30. Agencia Estatal Boletin Oficial del Estado. Royal decree 1971/ F, et al. Connecting food environments and health through the 1999, of 23 December, of procedure for the recognition, relational nature of aesthetics: gaining insight through the declaration and classification of the degree of disability; 2012. community gardening experience. Soc Sci Med 2011; https://www.boe.es/buscar/act.php?id=BOE-A-2000-1546. 72(11):1853–63. Accessed August 20, 2020. 18. Litt JS, Schmiege SJ, Hale JW, Buchenau M, Sancar F. 31. Gómez LE, Arias B, Verdugo MÁ, Navas P. Application of the Exploring ecological, emotional and social levers of self-rated Rasch rating scale model to the assessment of quality of life of health for urban gardeners and non-gardeners: a path analysis. persons with intellectual disability. J Intellect Dev Disabil Soc Sci Med 2015;144:1–8. 2012;37(2):141–50. 19. Pitt H. Therapeutic experiences of community gardens: putting 32. Korn A, Bolton SM, Spencer B, Alarcon JA, Andrews L, Voss flow in its place. Health Place 2014;27:84–91. JG. Physical and mental health impacts of household gardens 20. Melville CA, Oppewal A, Schäfer Elinder L, Freiberger E, in an urban slum in Lima, Peru. Int J Environ Res Public Guerra-Balic M, Hilgenkamp TIM, et al. Definitions, Health 2018;15(8):1751. measurement and prevalence of sedentary behaviour in adults 33. Hayashi N, Wada T, Hirai H, Miyake T, Matsuura Y, Shimizu with intellectual disabilities: a systematic review. Prev Med N, et al. The effects of horticultural activity in a community 2017;97:62–71. garden on mood changes. Environ Control Biol 2008; 21. Oppewal A, Hilgenkamp TIM, Schäfer Elinder L, Freiberger 46(4):233–40. E, Rintala P, Guerra-Balic M, et al. Correlates of sedentary 34. Wichrowski M, Whiteson J, Haas F, Mola A, Rey MJ. Effects behaviour in adults with intellectual disabilities: a systematic of horticultural therapy on mood and heart rate in patients review. Int J Environ Res Public Health 2018;15(10):2274. participating in an inpatient cardiopulmonary rehabilitation 22. Gonzalez MT, Hartig T, Patil GG, Martinsen EW, Kirkevold program. J Cardiopulm Rehabil 2005;25(5):270–4. M. Therapeutic horticulture in clinical depression: a 35. Litt JS, Soobader M-J, Turbin MS, Hale JW, Buchenau M, prospective study of active components. J Adv Nurs 2010; Marshall JA. The influence of social involvement, 66(9):2002–13. neighborhood aesthetics, and community garden participation 23. Gonzalez MT, Hartig T, Patil GG, Martinsen EW, Kirkevold on fruit and vegetable consumption. Am J Public Health 2011; M. A prospective study of existential issues in therapeutic 101(8):1466–73. horticulture for clinical depression. Issues Ment Health Nurs 36. Sommerfeld AJ, McFarland AL, Waliczek TM, Zajicek JM. 2011;32(1):73–81. Growing minds: evaluating the relationship between gardening 24. Gonzalez MT, Hartig T, Patil GG, Martinsen EW, Kirkevold and fruit and vegetable consumption in older adults. M. A prospective study of group cohesiveness in therapeutic Horttechnology 2010;20(4):711–7. horticulture for clinical depression. Int J Ment Health Nurs 2011;20(2):119–29. 25. Kim B-Y, Park S-A, Song J-E, Son K-C. Horticultural therapy program for the improvement of attention and sociality in children with intellectual disabilities. Horttechnology 2012; 22(3):320–4. 26. Gigliotti CM, Jarrott SE. Effects of horticulture therapy on engagement and affect. Can J Aging 2005;24(4):367–77. 27. Whatley E, Fortune T, Williams AE. Enabling occupational participation and social inclusion for people recovering from mental ill-health through community gardening. Aust Occup Ther J 2015;62(6):428–37. 28. Kam MCY, Siu AMH. Evaluation of a horticultural activity programme for persons with psychiatric illness. Hong Kong J Occup Ther 2010;20(2):80–6. 29. The World Health Organization. World report on disability; 2011. https://www.who.int/publications-detail/world-report- on-disability. Accessed August 7, 2020. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 7 PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 Tables Table 1. Demographic Characteristics and Quality of Life Outcomes of Participants With Moderate to Very Marked Disability, by Rooftop Garden and Intervention Group, Pilot Rooftop Garden Study, Barcelona, Spain, January–June, 2018a Characteristic Intervention Nonintervention P Valueb P Valuec d Initial Garden Female sex, n (%) 12 (38) 7 (33) .99 NA Age ≥45 y, n (%) 15 (47) 9 (43) >.99 NA Disability degree, n (%) Moderate 0 3 (14) Marked 14 (44) 8 (38) .12 NA Very marked 18 (56) 10 (48) Other indoor activities, yes, n (%) 32 (100) 17 (81) .02 NA Other outdoor activities, yes, n (%) 20 (62) 5 (24) .01 NA e Quality of life dimensions, median (IQR) Emotional well-being 15 (13–16) 15 (14–16) .73 NA Material well-being 23 (21.5–27.5) 24 (22–27) .94 NA Physical well-being 23 (22–26) 25 (22–27) .42 NA Self-determination 26 (21–18) 23 (19–31) .55 NA Personal development 8 (6–9) 8.5 (7–10) .18 NA Interpersonal relations 23 (21–25) 22 (21–24) .70 NA Social inclusion 18 (14–23) 18 (16–21) .78 NA Rights 11 (9–14) 12 (9–13.5) .81 NA Quality of life, median (IQR)e 148 (136–166) 149 (125–162) .80 NA f Recent Garden Female sex, n (%) 4 (18) 3 (14) >.99 .05 Age ≥45 y, n (%) 13 (59) 13 (59) >.99 .25 Disability degree, n (%) Moderate 2 (9) 1 (5) Marked 16 (73) 15 (68) .78 .01 Very marked 4 (18) 6 (27) Other indoor activities, yes, n (%) 22 (100) 20 (10) NA .13 Other outdoor activities, yes, n (%) 15 (68) 11 (55) .58 .22 Quality of life dimensions, median (IQR)e Emotional well-being 13 (12–15) 13 (10–15.2) .45 <.01 Abbreviations: IQR, interquartile range; NA, not applicable. a At 22 months for the initial garden and at 7 months for the recent garden. Fisher exact text or χ2 test used for categorical variables, and Mann–Whitney-Wilcoxon text use for variables that report IQR. b Comparison between participants in intervention vs nonintervention groups. c Comparison between participants in initial garden vs recent garden groups. d Intervention, n = 32; nonintervention, n = 21. e Assessed by using the INTEGRAL Scale questionnaire (31). f Intervention and nonintervention, n = 22. (continued on next page) The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. 8 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2020/20_0087.htm PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 (continued) Table 1. Demographic Characteristics and Quality of Life Outcomes of Participants With Moderate to Very Marked Disability, by Rooftop Garden and Intervention Group, Pilot Rooftop Garden Study, Barcelona, Spain, January–June, 2018a Characteristic Intervention Nonintervention P Valueb P Valuec Material well-being 23 (21–24) 23 (20–25) .67 .08 Physical well-being 26 (21–27) 24.5 (19.8–27) .4 .77 Self-determination 25 (22–29) 26 (22–29) .83 .58 Personal development 9 (8–11) 9 (7.75–10) .25 .12 Interpersonal relations 24 (20–24) 20.5 (17–25) .54 .24 Social inclusion 21 (20–23) 20 (16.8–22) .19 .05 Rights 13 (11–15) 13.5 (12.8–15.2) .71 <.01 Quality of life, median (IQR)e 153 (146–161) 143 (137–151) .11 .98 Abbreviations: IQR, interquartile range; NA, not applicable. a At 22 months for the initial garden and at 7 months for the recent garden. Fisher exact text or χ2 test used for categorical variables, and Mann–Whitney-Wilcoxon text use for variables that report IQR. b Comparison between participants in intervention vs nonintervention groups. c Comparison between participants in initial garden vs recent garden groups. d Intervention, n = 32; nonintervention, n = 21. e Assessed by using the INTEGRAL Scale questionnaire (31). f Intervention and nonintervention, n = 22. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 9 PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 Table 2. Adjusted Models for Exposure to Rooftop Garden Interventions and Quality of Life Outcomes of Participants With Moderate to Very Marked Disability, Pilot Rooftop Garden Study, Barcelona, Spain, January–June 2018a Initial Garden Recent Garden Outcomes Coefficient (95% CI) P Value Coefficient (95% CI) P Value Physical well-being 0.85 (–0.65 to 2.35) .26 2.02 (–0.72 to 4.75) .14 Social inclusion 0.13 (–2.31 to 2.58) .91 0.32 (–1.67 to 2.32) .74 b,c Interpersonal relations –0.53 (–2.52 to 1.45) .59 0.87 (–1.19 to 2.93) .39 Emotional well-being 0.85 (–0.01 to 1.70) .05d 0.45 (–0.85 to 1.74) .48 b Personal development 0.34 (–0.85 to 1.53) .56 1.30 (0.27 to 2.33) .02 Material well-being –1.21 (–2.61 to 0.19) .09b 0.23 (–2.74 to 2.27) .85 c Self-determination –1.34 (–4.80 to 2.13) .44 1.93 (–0.99 to 4.85) .19 Rights –1.29 (–2.74 to 0.16) .08b,c 0.75 (–1.10 to 2.59) .41 b,c Quality of life –5.70 (–16.88 to 5.49) .31 7.81 (–0.63 to 16.25) .07 a All models adjusted by sex, age, and degree of disability. Models of the recent garden also adjusted by corresponding quality of life total score or dimension- specific score at 3 months. b Models additionally adjusted by occupational center. c Models additionally adjusted by indoor activities. d Models additionally adjusted by outdoor activities. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. 10 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2020/20_0087.htm PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 Table 3. Selected Quotes of Participants With Moderate to Very Marked Disability, Pilot Rooftop Garden Study, Barcelona, Spain, January–June 2018 Quality of Life Dimension/ Quotes Aspects Physical Well-being for Rooftop Gardeners Increased energy “X, who is a person who is in the phase of aging and we think perhaps has some mental illness that we are not very clear about yet, but that for example here [in the workshop] does not participate in any activity and sits on the couch and as she is a great help to the cafeteria and prepares coffee, but she stays on the sofa and at 3 in the afternoon she is already waiting at the door to leave. Well, on the day of the garden, she wants to participate. In fact, she has memory problems, and it is unclear for her whether she has done it the week before or not and tries to sneak in every week. And she does not want to do any other activity, none at all.” [Educator from occupational center assigned to the initial garden] Overcoming own limitations “It’s a way for me to be by myself. Find myself in a moment . . . to have the obligation to get up on a Friday, to know that I have to go to a community garden, I know that I will meet people, I know we are going to . . . I don’t know. . . . It makes a group more human and also to have an end goal that is a cultivated garden. That’s right, because it helps you to find yourself, with yourself.” [Gardener from the recent garden] Interpersonal Relations and Social Inclusion for Rooftop Gardeners New social relations and “[I like it] because you learn . . . to plant and meet other guys from other workshops.” [Gardener from the initial garden] contacts “One day we went to have a drink [after the gardening activity with the gardeners from the other occupational center] and we were talking while having the drink. And I really liked it.” [Gardener from the recent garden] “There are many activities in which there are relationships, but this is one of the most, because for example there are tasks like in a chain: pick this up and give it to someone else who takes it.” [Educator from occupational center assigned to the initial garden] “X is a person who does not relate well with anyone. She really only talks with the 4 people that interest her. And, for example, she went to the garden and greeted Mr. Y, she hung out with Z, she hung out with M, she talked with N, with whoever was around her.” [Educator from occupational center assigned to the initial garden] “Opportunities to get together with so many other people and share and work . . . they don’t have that. And relate to the people in the building where they work . . . all that. I find that it is very good.” [Educator from occupational center assigned to the initial garden] From aid recipients to providers “Last time I went, when we went to bring chard to the Food Bank . . being able to give them something you have cultivated, of social benefits for others something that you have done, something that you have seen grow . . . and that people [from the Food Bank] take advantage of it and use it . . . you get excited, you feel good.” [Gardener from the recent garden] “In fact, it is one of the things that they like the most I believe, to give to others. And now that we are donating to the Food Bank, when we go there and tell them ‘those who have no food can eat thanks to what you all have brought,’ so . . . to me it is exciting, because they also feel satisfied. In fact they already ask, ‘When are we going to give food?’, ‘When are we going to distribute?’ So it’s something they value.” [Educator from the initial garden] New perspectives on the city and “I like the fact that is an activity we do outside” [Gardener from the initial garden] surrounding environment “It is something that gets them away from the normality of the center and the workshop, and they take public transport, which they do not usually do, then we go for breakfast, in the garden they know. Well yesterday, for example we were 4 occupational centers together doing the same thing. They hardly ever have this.” [Educator from occupational center assigned to the initial garden] Emotional Well-being, Personal Development, and Having a Sense of Purpose in Life for Rooftop Gardeners Spaces of individual discovery When they leave the workshop you can see that they feel free because there [in the workshop] they are in an enclosed place and and freedom there are other activities and other ways of doing things. Here [in the garden] they have . . . their freedom. They are outdoors, doing activities that do not censure them at all.” [Rooftop technician] Loss of fear and improved self- “When there is no gardening activity, we could do some training. We could, for example, talk about when tomato plants have to confidence and autonomy be planted, what types of tomato plants exist…and things like this. Well, to learn a bit about the garden. I would also like to know about the moons, why depending when we plant the vegetables will grow more or less.” [Gardener from the recent garden] “Before, they were waiting for you to tell them to “look at this” and now some come and [say] ‘It is being watered or is not being watered,’ or ‘Is this what I can start?’ Before if you did not tell them ‘This is what you have to touch’ and now they already know ‘This is it’ and ‘Today we are going to look at the zucchini, right?’ or ‘Today we will be able to collect this, right?’” [Educator from occupational center assigned to the initial garden] “Logically, if I ask one of them [participant of the initial garden] and one of the others [participant of the recent garden] to plant chard, you notice that one does so with more fear. and the other is more relaxed, because they already have confidence and if something breaks you already know that nothing will happen because it is not important.” [Rooftop technician] Improved emotional strength and “There are no tools. . . . Everything is manual. So everything is more creative, right? With more sensitivity. Having to pick, pull a problem resolution chard out instead of cutting it with a knife, right? I was used to always cutting it with a knife and now it’s with the hand, ‘Klak.’ (continued on next page) The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 11 PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 (continued) Table 3. Selected Quotes of Participants With Moderate to Very Marked Disability, Pilot Rooftop Garden Study, Barcelona, Spain, January–June 2018 Quality of Life Dimension/ Quotes Aspects These are trivialities but I do not know. . . . It seems . . .” [Gardener from the recent garden] “They talk about this activity with their fellow workshop attendees and there are fellow workshop attendees who ask them what is this about the Garden of the District Office and what is this about planting in sacks. They have the ability to explain in their own way, with their words.” [Gardener from the initial garden] Having a purpose in life “Most, many, think of it as a responsibility, as something they want to take care of. Not so much ‘Without me this won’t survive,’ but ‘It is something I have taken responsibility of and that I am interested in and I want to be aware of it.’” [Educator from occupational center assigned to the initial garden] “X, who is a person who is in the phase of aging and we think perhaps has some mental illness that we are not very clear about yet, but that for example here [in the workshop] does not participate in any activity and sits on the couch and as she is a great help to the cafeteria and prepares coffee, but she stays on the sofa and at 3 in the afternoon she is already waiting at the door to leave. Well, on the day of the garden, she wants to participate. In fact, she has memory problems and it is unclear for her whether she has done it the week before or not and tries to sneak in every week. And she does not want to do any other activity, none at all.” [Educator from occupational center assigned to the initial garden] “It’s a way for me to be by myself. Find myself in a moment. . . . To have the obligation to get up on a Friday, to know that I have to go to a community garden, I know that I will meet people, I know we are going to . . . I don’t know. . . . It makes a group more human and also to have an end goal that is a cultivated garden. That’s right, because it helps you to find yourself, with yourself.” [Gardener from the recent garden] “There are people who are waiting for the whole week for Wednesday, so they can pick up their [transport] card and go there [to the garden] and spend some time in the garden and everything that that activity represents.” [Educator from the recent garden] The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. 12 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2020/20_0087.htm PREVENTING CHRONIC DISEASE VOLUME 17, E126 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY OCTOBER 2020 Appendix. Evaluation of Quality of Life Benefits of Urban Rooftops Program for Gardeners With Intellectual Disability or Mental Health Disorders Quantitative quality of life measures methodology Most of our data were collected face to face, despite the use of a self-reported scale, to guarantee the understanding of instructions, items, and the response scales. To administer the questionnaire, one person from the research team and one worker from the corresponding occupational center were present. In general, sessions to answer the questionnaires were conducted in groups, as advised by the occupational center workers, who considered the group format to be the best approach given the needs and abilities of the participants. We organized 9 group sessions with respondents from the intervention group and the nonintervention group separately in order to answer the questionnaires. For logistical reasons, 2 of these sessions included people from both intervention and nonintervention group of the same occupational center. However, as one of the occupational centers assigned to the recent garden did consider that their users were sufficiently autonomous to read, understand, and answer the questions on their own, the INTEGRAL Scale was self-administered by those individuals (7 from the intervention group and 7 from the nonintervention group, all from the occupational center serving people with mental health disorders). In specific cases in which someone could not come to the scheduled date with the group, a worker from the occupational center administered the questionnaire without the research team member present. Each of the quality of life data collection sessions had the same structure and lasted for approximately 2 hours. The INTEGRAL Scale of quality of life is a tool developed to collect information on the quality of life of people with intellectual disabilities. The scale includes 47 items to be answered using 4 possible Likert responses (completely agree, agree, disagree, completely disagree) used to build a quality of life total score ranging from 47 to 188, with higher scores indicating higher quality of life. The included items are from 8 dimensions (with higher scores indicating greater well-being on each dimension): material well-being (score range, 7–28), physical well-being (range, 7–28), emotional well-being (range, 4–16), self-determination (range, 9–36), personal development (range, 3–12), interpersonal relationships (range, 7–28), social inclusion (range, 6–24) and rights (range, 4–16). Qualitative data collection details The prepared questions were shared for feedback with the designers of the rooftop gardens pilot program and tested and refined before the first full round of interviews. Finally, the questions included logistic and organizational issues around the activity, aspects related to the implementation of the activity itself and the different stages and aspects of gardening, questions about the broader implications and dimensions beyond the activity itself, and suggestions for change or improvements. Most of the interviews were conducted individually except for a few interviews with gardeners, for whom the presence of a social educator helped ease the conversation or clarify some responses or our own questions. We repeated some interviews with 9 of the individuals from the intervention group from the recently implemented garden after 3 or 4 months of the first interview to identify possible changes. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. www.cdc.gov/pcd/issues/2020/20_0087.htm • Centers for Disease Control and Prevention 13