Design and Health World Health Design

Gardens in healthcare facilities: steps towards evaluation and certification

Restorative outdoor spaces are desirable in healthcare facilities, but there is no consistent approach to their definition or evaluation. This paper charts the move towards formalisation and certification

Clare Cooper Marcus, Hon ASLA, MA, MCP, University of California, Berkeley, and Naomi Sachs, MLA, ASLA, EDAC, Texas A&M University

Landscapes that promote health and wellbeing, often referred to as ‘healing gardens’, are increasingly being incorporated into healthcare facilities throughout the world as an essential design component. This is due in no small part to a growing body of evidence on the health benefits to patients, visitors and staff, as well as the financial benefits to the organisation. Along with empirical research on the efficacy of such spaces, there is also sufficient evidence to inform decisions about design specifics, such as what elements, configurations and programming in outdoor spaces can elicit the best possible outcomes.

Unfortunately, this information often does not make it from research to practice. While acceptance of the need for restorative gardens and other ways to connect people with nature shows great progress, the implementation – how gardens are designed, constructed and maintained – has been accomplished with varying degrees of success. In order to bring about the best outcomes, decision makers need tools that can help them translate existing research into good design. A set of design guidelines in a forthcoming publication by Cooper Marcus and Sachs1 will help to educate and inform practitioners. The guidelines will also become the basis for: 1) development of a set of essential components to be incorporated into a standardised evaluation, in the form of an audit tool, of existing spaces, and 2) a certification programme comparable to the US Green Building Council’s (USGBC) LEED for Healthcare.2 The objective of this paper is to present a set of necessary steps that will lead to a potential evaluation process and certification of restorative gardens, and to discuss why at this time such a process should be considered.

Some definitions
Professionals and lay people use many different terms for outdoor spaces that promote health and wellbeing. Though ‘healing gardens’ is the most commonly used, others include landscapes for health, restorative landscapes, therapeutic landscapes, restorative gardens, therapeutic gardens, wellness gardens and rehabilitation gardens. For this paper, the authors use the term restorative garden(s) and healing garden(s) interchangeably, which they define as ‘an outdoor (and sometimes indoor) space designed for a specific population, a specific site and a specific outcome’. The authors also use the term ‘access to nature’, which they define as any type of sensory contact with nature. Access to nature includes both visual (being able to see elements of nature from indoor and outdoors) and/or physical (touch, smell, sound, taste) contact. Use of the term ‘access to nature’ asserts the need for all members of a design team, and not just the landscape architect, to design for nature contact throughout the entire project.
Acceptance of restorative gardens
Restorative gardens and other means of connecting people with nature – via outdoor gardens, views of nature, indoor plants and gardens, natural materials and other forms of biophilic design – have long been considered ‘the icing on the cake’: amenities that often were either not included at all, or were the first thing to be eliminated from a project when the budget began to tighten. In a short period of approximately 20 years, people’s opinions of gardens in healthcare facilities have evolved from a general, intuitive sense that ‘nature is good for us’ to increased acceptance, based on both quantitative and qualitative evidence, that the inclusion of opportunities for nature contact is an essential design element.
Decision makers at myriad facilities recognise that connection with nature increases not only positive health outcomes for consumers (patients and their families) and staff but also for the facility’s bottom line. In surveys conducted by the Center for Health Design, one-third of respondents reported that they ‘always’ implemented healing gardens in their designs.3,4

Beyond self-regulation
In the US, provision of access to nature and restorative gardens has now evolved beyond action taken by individual stakeholders. The Green Guide for Health Care (GGHC) was the sector’s first sustainable design rating system, a toolkit that integrated health principles and practices into the planning, design, construction, operations and maintenance of healthcare facilities. The 96-credit GGHC system includes credits such as SS-9.1, ‘Connection to the Natural World – Outdoor Places of Respite’, and SS-9.2, ‘Exterior Access for Patients’. The GGHC was accepted by the USGBC, reformatted as a LEED product, and launched in 2011 as LEED for Healthcare.2 Like the GGHC, the Sustainable Sites Initiative (SITES) was modelled on the LEED rating and accreditation system. SITES focuses on general outdoor environments. The programme creates voluntary national guidelines and performance benchmarks for sustainable land design, construction and maintenance practices. From its inception in the early 2000s, SITES recognised the salutary value of landscapes through its Human Health and Well-being credits. The USGBC is now a stakeholder in the initiative and anticipates incorporating SITES guidelines into the LEED Green Building Rating System in the near future.5

In the 2014 Guidelines for Design and Construction of Health Care Facilities, a new key element in the physical component of the Environment of Care section will be ‘access to nature’. This is momentous, because access to nature has previously appeared only in the Appendix as a ‘should’ rather than in the body of the text as a ‘shall’. The intent of the guidelines, adopted by 42 of the 50 US states, is to identify the minimum requirements for the design of new or renovated healthcare facilities. The authorities having jurisdiction (AHJs) use the guidelines as a basis for design approval. Drawings are only approved, and the project can only go forward, if the guidelines’ minimum standards are met. Other regulations are following suit. For example, in the latest Recommended Standards for Newborn ICU Design, ‘Access to Nature and Other Positive Distractions’ is standard number 25: “When possible, views of nature shall be provided in at least one space that is accessible to all families and one space that is accessible to all staff. Other forms of positive distraction shall be provided for families in infant and family spaces, and for staff in staff spaces.”6

But what about quality?
Clearly the message is getting through about the restorative benefits of nature. This is an important and exciting step. But now new issues and question arise: What kind of nature? What kind of restorative gardens? What sorts of spaces, and elements within those spaces, will be the most beneficial for patients, visitors, and staff? Sadly, designers and clients are often unaware of the research available suggesting what is essential if a garden is to be truly restorative. Outdoor spaces designated as ‘healing’ often lack such basic necessities as shade, comfortable seating, places for privacy or enough greenery to even be perceived as a garden. Components that have become popular, such as labyrinths, are incorporated without consideration for their appropriateness to the site, understanding of their meaning, or potential users’ ability and energy levels.

One of the most frequently recurring reports by garden users about why they visit a garden, and how they feel about it, is that it offers an opportunity to ‘get away’. It serves as a much-needed contrast to an indoor environment that is often sterile, alienating and frightening. And yet all too often, the outdoor spaces provided do not offer the escape that people so desperately need.

One assumes it would not occur to a designer of a cancer care clinic or a new emergency department not to consult the latest evidence-based design literature, or at least the minimum standards. In fact, in most cases it would not be awarded the contract unless it had done so. In the case of healing or restorative gardens, this is generally not so. A designer may be hired because she or he has previously created successful parks, office plazas or corporate headquarters – and it is assumed that a healthcare garden is not very different, despite that fact that it may be serving a user group with very specific needs, such as those with cancer, post-traumatic stress disorder or Alzheimer’s disease. Some professionals are continually hired to design restorative gardens because their resume states that they have done several already, regardless of whether or not the existing gardens have been particularly successful.

In most cases, problems arise when designers and clients are not sufficiently familiar with the essential requirements for restorative gardens. In some cases, the research is known but ignored due to cost concerns. Sometimes the space is designed by a professional not trained in garden design. Landscape architects are the only professionals with enough training and experience in the planning of outdoor space, planting design, detailing and so on to design a restorative garden. A landscape architect must be on the design team from the project’s inception. Even within the landscape architect profession, many are not equipped to design spaces for healthcare facilities. Roger Ulrich,7 whose research spearheaded the movement towards incorporation of nature and evidence-based design in healthcare facilities, put it succinctly when he wrote: “Regardless of whether a garden might garner praise in professional design journals as ‘good design’, the environment will qualify as bad or failed design in healthcare terms if it is found to produce negative reactions. These points imply that the use of the term ‘healing’ in the context of healthcare gardens ethically obligates the garden designer to subordinate or align his or her personal tastes to the paramount objective of creating a user-centered, supportive environment.”

Success of restorative gardens is critical for many reasons. First, a well-designed, well-constructed and well-maintained garden promotes the best health outcomes. Second, such a garden will bring a positive image to the facility and all of the stakeholders involved. Third, every successful restorative garden is powerful testimony to restorative gardens, and access to nature, as a whole. On the other hand, an unsuccessful restorative garden – whether it was not well designed or has fallen into disrepair – conveys the message that provision of access to nature is not something worth spending precious budget dollars on.
The next step: guidelines
The world is in the midst of a giant healthcare design and construction boom. There is a critical need for evidence-based, or research-informed, guidelines to aid in the design of outdoor spaces that facilitate health. Design guidelines for restorative gardens appeared first in the late 1990s with the publication of Clare Cooper Marcus and Marni Barnes’ Healing Gardens: Therapeutic benefits and design recommendations8 and Martha Tyson’s The Healing Landscape: Therapeutic outdoor environments.9 The culmination of recent work by Cooper Marcus and Sachs1 will appear in a co-authored book, Therapeutic Landscapes: An evidence-based approach to designing healing gardens and restorative outdoor spaces. The heart of the book is a set of design guidelines. While there is not a vast amount of research on restorative gardens based on post-occupancy evaluations (POEs), there are sufficient repeated findings from these – as well as from less formal evaluations, plus existing audits and best practice – to suggest what is essential in the design of a successful outdoor space in a healthcare facility. In Therapeutic Landscapes, one chapter features a comprehensive set of design guidelines applicable to all types of healthcare facilities. This is supplemented by guidelines in the eight following chapters on facilities for particular patient groups.

Guidelines are an important step in aiding designers and clients to make good decisions. However, are they enough? There is a need for designers and clients to be more accountable for what they provide. Two important and related tools are needed now, and both can easily grow out of the guidelines already created and the research upon which they were based: first, a tool for systematic evaluation, and second, a certification programme.

Evaluating restorative gardens
It is important to consider why and how some healing gardens succeed and others fail. Landscape architecture, like other design professions, has been lax in evaluating built work. Unless this happens in a consistent way, designers cannot learn from past mistakes and the profession cannot move forward. It is true that a systematic evaluation costs money. This suggests the need for a line item in the budget for a garden that covers an evaluation – and fine-tuning – one to two years after implementation. Evaluation is an opportunity to learn, to improve an existing space and to add to the fund of knowledge about design.

Post-occupancy evaluation
There are several ways of evaluating the success (from the users’ perspective) of a therapeutic healthcare outdoor space, regardless of whether or not it is specifically labelled a healing or restorative garden. The most comprehensive method is a diagnostic POE, where the goals of the original design and how they were translated into physical form are compared with how the space is now used (or not used). Use (by whom and for what) is recorded by an objective measure such as activity mapping, stop-frame filming and the like. The users’ feelings about the space are recorded via subjective measures such as interviews or surveys to find out what they like or don’t like, how often they visit, impediments to use, recommendations for change and so on. The use of multiple methods to provide reliable findings is essential. A diagnostic POE – the most in-depth form of POE – is ideally carried out by a team consisting of one or more social scientists familiar with the methods and one or more healthcare garden designers, but not the designers of the garden being studied or anyone responsible for it in any way. This avoids possible bias. Together these two basic categories of data – behaviour-mapping providing objective information about use, and interviews providing subjective information about motivation and feeling – offer a good overview of the success and/or shortcomings of the garden being studied.

An unbiased evaluation of a restorative garden two or three years after construction can document how well the intent of the garden and the needs of its users were understood, how well the original goals of the design were translated into physical form, how well the garden serves the users it was intended for (as well as those it was not planned for), how well the planting is doing, how well the space is being maintained, and what changes in physical design, maintenance, or policy need to be implemented. Ideally, the results of any kind of POE are compared with the designer’s and client’s original intentions, maintenance or construction changes since the garden was opened, changes to the use of adjacent buildings, and changes of policy regarding who may use the garden, hours the garden is open or similar issues. Particularly important would be the evaluation of a garden based on evidence-based design with specific desired outcomes. A series of POEs of similar types of gardens can provide valuable information on exemplary gardens that can serve as models of user-oriented design, as well as outdoor spaces intended to be therapeutic that have failed to live up to their promise.

To date, some of the most useful POEs of hospital gardens have been conducted by landscape architecture or psychology students as research at masters or PhD level10-12 or by academics.13,14 Very few have been conducted or commissioned by the designers or clients. The usual reasons are that the POE is time-consuming, there is no line item in the budget to cover the cost, or the designer does not have the skills to carry out such a study. A less comprehensive and less costly POE, referred to as an indicative POE, can be conducted in a short time span (a few hours to a few days) and can include interviews with the staff and/or the designers and a walk-through evaluation.15 Of all of the types of POEs, the indicative POE is the closest to an audit.

Audits: an alternative to POE

Audits are, in general, less expensive and less time-consuming than POEs. An audit tool is a scored checklist of elements and qualities that should, ideally, be incorporated into a restorative garden. Unlike a POE, an audit is a process for evaluating only the finished built work and whether it meets minimum standards. This form of evaluation can reveal a great deal about how well design details have been implemented – including suggesting areas in need of change – but reveals nothing about who uses the space or their motivations or the initial goals of the client, designer, and other stakeholders. An audit is usually conducted by three or four individuals who are knowledgeable about therapeutic garden design, but who were not involved in the original programming, design, and construction process. They evaluate the garden separately and then their scores are averaged to avoid any bias.16

Cooper Marcus first developed a simple audit tool in 2006 for students on field trips to hospital outdoor spaces. Elements that were considered potentially important were organised and worded so that students could focus on one feature after another as a way to make their visit more instructive than a random wandering through and taking pictures. This initial attempt was followed in 2008 by a more systematic, research-based Alzheimer’s Garden Audit Tool (AGAT) for gardens at dementia care facilities,16 and later by audits for evaluating gardens for the frail elderly, gardens at children’s hospitals and at general acute care hospitals. In the latter case, Cooper Marcus and Barnes were hired by a Chicago area landscape architecture firm (Hitchcock Design Group) that specialises in healthcare design. The firm wanted to know how well its designs were functioning. Existing research for each of these types of outdoor space was drawn upon to create a series of qualities and features that, theoretically, should be present. These elements and qualities were scored in each of six gardens being evaluated on a 5-point scale: 0 – Not applicable; 1 – Feature not present or quality missing; 2 – Feature or quality poorly provided; 3 – Feature or quality moderately well provided; and 4 – Feature or quality very successfully provided. Four people evaluated each garden separately and the results were averaged to avoid bias. The 2008 consultant report on the post-occupancy evaluation of six Hitchcock Design healing gardens resulted in some elements being eliminated from the audit tool since they were too difficult to evaluate (for example, ‘The garden is culturally appropriate’). In 2012–13 the AGAT tool was adapted and improved by Alzheimer’s Australia, and its Dementia Therapeutic Garden Audit Tool will be tested for reliability (

Potential audit tool: acute hospitals
The authors are now working with colleagues to refine the audit tool developed in 2008. It is organised under two broad headings, similar to the guidelines in Therapeutic Gardens: What is essential (required), and what is desirable but not essential (recommended).1 At this point it seems appropriate to create such a tool for evaluating a healthcare outdoor space that serves the broadest possible variety of users – an acute care hospital with a range of patient types, staff and visitors. Once this tool is tested, others could be developed for more specific populations and spaces.
The added benefit of a standardised audit tool is that, as audits are performed and information is gathered, we begin to build a database and a collection of case studies of existing built works, something that is sorely lacking in the scholarship of healthcare design. The more that healthcare gardens can be evaluated and documented, the more examples will be available for clients and designers to use as models.

The audit tool is at a preliminary stage of development. It will be developed around a scoring system consisting of two parts: benchmarks (essential features or qualities supported by research and/or best practice) and credits (elements that are desirable but optional, supported by extensive observation, best practice and common sense, but not yet by research evidence).

Organisation of the audit tool will likely be in the following categories:
• Visual and physical accessibility: for example, garden is visible from well-used indoor spaces such as lobby, waiting room; doors and thresholds to garden are easily navigable
• Safety, security, and privacy: for example, clear boundaries or sense of enclosure; places for people to retreat on their own or with others; adequate lighting
• Emotional and physical comfort: for example, covered seating area at garden entry; comfortable seating throughout the garden; mitigation of extreme weather; quiet location
• Nature distraction/engagement: for example, high ratio of greenery to landscape; rich sensory details; seasonal interest; sight and sound of water, wildlife
• Social connection and support: for example, semi-private seating clusters; close proximity to nursing units, waiting rooms, staff break rooms
• Physical movement and exercise: for example, level, non-glare pathways; appropriate traction; destination points
• Sense of control: for example, moveable seating; variety of walking loops; places to sit in sun or shade
• Adequate maintenance: for example, garden approach to be regularly well maintained.

The essential next steps will be to refine the audit tool by testing for validity and reliability among people using it, as well as across various different sites and to propose the weighting of various elements. Then, compare this with other audit tools developed for different kinds of indoor and outdoor spaces. These will include the Seniors Outdoor Survey for Staff (SOS-1) and Seniors Outdoor Survey for Research (SOS-2) developed by Susan Rodiek17; the US Department of Transportation Walkability Checklist;18 and the three audit tools described by Ian Forbes in the April 2013 issue of this publication:19 the Achieving Excellence Design Evaluation Toolkit (AEDET), A Staff and Patient Environment Calibration Toolkit (ASPECT) and the dementia Evaluation Audit Tool (EAT). Sachs is conducting a literature review to locate other existing audit tools that could potentially be used as models.

The third step will be to consider how the proposed audit process can become the basis for the certification of restorative gardens. Testing the audit tool will likely reinform the original Therapeutic Landscapes guidelines. In other words, this is not a one-way process, but rather an iterative series of steps that will lead to the best method(s) to both guide and assess healing gardens. At the same time that the audit is being developed and tested and the guidelines are being refined, a certification programme will be developed. While guidelines, evaluations and audits can help designers and clients understand what they should do (or should have done), only certification can hold designers and healthcare facilities to certain (minimum) standards.

Steps to certification
The logical procedure culminating in a viable restorative garden certification programme should follow a sequence of: research • evidence-based design guidelines and standards • audit tool • refine guidelines as needed • certification. Ideally, research will continue to be conducted throughout the entire process.
A secondary benefit with certification would be the validation of the initial guidelines. As certification carries more weight than recommendations, it has the potential to reinforce design guidelines. One informs and strengthens the other. Furthermore, once guidelines are tested through an audit process and formalised into a certification programme, they can be incorporated as details for other certifying and regulatory bodies such as LEED, SITES and the Joint Commission.
At this point, the issue of certification, which will most likely be voluntary, raises many questions: Does the same group of people develop both the audit and the certification programme? How will both be developed, tested and implemented? Who will administer the certification process? What is the incentive for a client or designer of a restorative garden to have it certified? Will it encourage a design-to-a-minimum or checklist mentality? Should certification be a one-time event or should it be renewed?

Despite all these challenging issues, it is clear that healthcare facilities are used to being credentialed and may welcome a method for evaluating their gardens. Staff and patients are attracted to facilities with high-quality environments, including outdoor spaces, and a credential document will give designers a common language as to what are healing or restorative gardens.

A useful first step may be to institute a healing or restorative garden award category to be added to those administered by the International Academy for Design & Health (IADH) or by the American Society for Landscape Architects (ASLA). This would help to focus attention on what is good design in this important component of the healthcare environment. While we certainly do not have all the answers regarding these steps towards a consistent approach to evaluation and certification, it is time for us to carry the hard work that has been done thus far to the next level to ensure the best possible outcomes for all involved.

Clare Cooper Marcus, Hon. ASLA, MA, MCP is professor emerita at the Departments of Architecture and Landscape Architecture, University of California, Berkeley. Naomi Sachs, MLA, ASLA, EDAC, is the founding director of the Therapeutic Landscapes Network and a PhD student in Architecture at Texas A&M University’s Center for Health Systems and Design.

1. Cooper Marcus C, Sachs NA. Therapeutic Landscapes: An evidence-based approach to designing healing gardens and restorative outdoor spaces. Hoboken (NJ): John Wiley and Sons; forthcoming 2013.
2. LEED. LEED for Healthcare. Washington, DC: US Green Building Council; c2013 [accessed 26 May 2013]. Available from:
3. Taylor EM. Survey of Design Research in Healthcare Settings: The use and impact of evidence-based design. Martinez, CA: Center for Health Design; 2009.
4. Taylor EM. Survey of Design Research in Healthcare Settings: The use and impact of evidence-based design. Martinez, CA: Center for Health Design; 2010.
5. Austin, TX: The Sustainable Sites Initiative; c2013 [accessed 26 May 2013]. Available from:
6. White RD, Smith JA, Shepley MM. Recommended standards for newborn ICU design. J Perinatol 2013; 33:S2-S16.
7. Ulrich R. Effects of gardens on health outcomes: Theory and research. In: Cooper Marcus C, Barnes M, (eds). Healing Gardens: Therapeutic benefits and design recommendations. New York: John Wiley and Sons; 1999. p27-86.
8. Cooper Marcus C, Barnes M (eds). Healing Gardens: Therapeutic benefits and design recommendations. New York: John Wiley and Sons; 1999.
9. Tyson M. The Healing Landscape: Therapeutic outdoor environments. New York: McGraw Hill; 1998.
10. Pasha S. Barriers to garden visitation in children’s hospitals. HERD. Forthcoming.
11. Sherman SA, Varni JW, Ulrich RS, Valcarne VL. Post occupancy evaluation of healing gardens in a pediatric cancer center. In: Wagenaar C (ed). The Architecture of Hospitals. Rotterdam, Netherlands: NAi Publishers; 2006. p330-42.
12. Sorensen KT. Effect of Time Spent in a Hospital Garden on Satisfaction with Hospital Care [dissertation]. Urbana, IL: University of Illinois at Urbana-Champaign; 2002.
13. Cooper Marcus C, Barnes M. Gardens in Health Care Facilities: Uses, therapeutic benefits, and design considerations. Martinez, CA: Center for Health Design; 1995.
14. Davis BE. Rooftop hospital gardens for physical therapy: a post-occupancy evaluation. HERD 2011; 4(2):14-43.
15. Anderzhon JW, Fraley IL, Green M. Design for Aging Post-occupancy Evaluations. Hoboken, NJ: John Wiley and Sons; 2007.
16. Cooper Marcus C. Alzheimer Garden Audit Tool. In: Rodiek S, Schwarz B (eds). Outdoor Environments for People with Dementia. New York: Haworth Press; 2008. p179-91.
17. Rodiek S. Seniors outdoor survey for providers (SOS-P); Seniors outdoor survey for researchers (SOS-R). College Station, TX: Center for Health Systems & Design; 2013. Available from
18. Federal Highway Administration, National Highway Traffic Safety Administration. Walkability Checklist. Washington, DC: US Department of Transportation; 2013 [accessed 26 May 2013]. Available from
19. Forbes I. From POE to design-in-use: benchmarking for health facility evaluation tools. World Health Design 2013; 6(2):50-57.

Further reading
Cooper Marcus C. Post occupancy evaluation. In: Hopper LJ (ed). Landscape Architectural Graphic Standards. New York: John Wiley and Sons; 2007. p. 57-63. Dallas: Facility Guidelines Institute; c2013 [cited 2013 May 26]. Available from:
Heath Y, Gifford R. Post-occupancy evaluation of therapeutic gardens in a multi-level care facility for the aged. Activities, Adaption and Aging 2001; 25(2):21-43.

©2018 All Rights Reserved. Website Design Graphic Evidence