The four levels of evidence-based practice
It has been said that evidence-based design is the natural parallel and analog to evidence-based medicine. It is the deliberate attempt to base design decisions on the best available research evidence. Healthcare architecture can be powerfully influenced by the efforts of practitioners who seek to understand the effects of the environment on patients and staff. Interior designers, urban planners, landscape architects, engineers, industrial designers, graphic designers, art consultants, and wayfinding and other consultants may also practice or contribute to evidence-based design. But in what ways does this occur? In fact, there are four levels of involvement, each progressively more profound.
Evidence-Based: Meaning and Scope
Evidence-based healthcare designs are used to create environments that are therapeutic, supportive of family involvement, efficient for staff performance, and restorative for workers under stress. An evidence-based designer, together with an informed client, makes decisions based on the best information available from research and project evaluations. Critical thinking is required to develop an appropriate solution to the design problem; the pool of information will rarely offer a precise fit with a client's unique situation. In the last analysis, though, an evidence-based healthcare design should result in demonstrated improvements in the organization's clinical outcomes, economic performance, productivity, customer satisfaction, and cultural measures.
Researchers in the area of the physical environment and healthcare have been reporting their results for some time. Roger Ulrich, PhD, conducted a pioneering 1984 study disclosing that surgery patients with a view of nature suffered fewer complications, used less pain medication, and were discharged sooner than those with a view of a brick wall. Research by others shows that heart patients with strong social support systems survive longer postdischarge than those without. There are studies on the antimicrobial characteristics of new carpet fibers, appropriate lighting for neonates, the negative effects of noise on critical care patients, and many more on a wide range of subjects that impact the design of health facilities (for a sampling, see Suggested Reading). In all, approximately 120 credible evidence-based design studies have been identified as having specific environmental relevance. Authors currently preparing a book on critical care design are using more than 300 research citations from multiple literature sources.
Most healthcare architects have developed a strong functional perspective in their work. It should therefore be a comfort to them to know that important design concepts have, in fact, been tested and that there are data available to inform their medical designs. With serious issues at stake, including sickness and health—even life and death—many architects welcome the emergence of data-based design.
On the other hand, serious designers might be concerned that evidence-based design represents a challenge that could limit their creativity or freedom of choice. To fear loss of creativity, however, is to overlook the exciting challenges of continuously inventing responses to emerging results and interesting new facts. This calls for an exceptionally creative and ever-changing interpretation of new data. Research can offer complex and sometimes contradictory insights into the world of architectural design, encouraging designers to test new and interesting ideas. The result is not reduced creative opportunities, but a commitment to observe the results of each design, and to use what might have been learned on future projects.
There is also an understandable fear that evidence-based design can lead to “cookbook” architecture, suggesting a pattern of dull and repetitious buildings stamped from the same mold of a bureaucratic prescription. The world of evidence-based design, however, lives closer to the real world of scientific research, with its continuous search for the verifiable. Evidence-based design is not static, and does not easily conform to fixed regulations that may soon be rendered ineffective by the steady stream of newly reported results.
The dean of Harvard's medical school reportedly tells each entering class that 50% of what they will learn is wrong—but we don't yet know which 50%. Conscientious evidence-based practitioners should experience fewer such doubts as they make an increasing percentage of their decisions on the basis of research. Rather than conforming to tightly limiting facility regulations, using loosely framed performance guidelines should encourage designers to become increasingly aware of environmental research in healthcare.
Not all design decisions—perhaps not even most decisions on a project—will be evidence-based. Just as the art of architecture has always been paired with the science of physics and engineering, art and creativity are integral features of the best healthcare designs based on rigorous research evidence. Good healthcare projects feature exceptional architecture that serves their purpose well, and it contains the magic of the human spirit, infused with the sacred, the inspired, the grand, the intimate, the full richness of life.
Evidence-Based Practice: Four Levels
Four “levels” can be used as a means to identify different commitments and practices as stages in which architects use this material on behalf of their clients (figure 1). To explain:
Four levels of evidence-based practice
Level-one practitioners: These architects and design professionals make a careful effort to design based on the available evidence. They make an effort to stay current with literature in the field. The designer interprets the meaning of the evidence as it relates to his or her project, and makes a judgment as to the best design for that specific circumstance. These designers are producing work that advances the state of the art because they are learning from others and developing new examples for others, while delivering better designs for their clients.
An example of a level-one project is an early design for Trinity Medical Center in Moline, Illinois. Watkins Hamilton Ross Architects (WHR) worked with Trinity to develop a Planetree demonstration unit that provided family support amenities, including a family kitchen, den, resource library, and foldout sleeping chair in the patient room (figure 2). Barriers to family involvement were also reduced in the nontraditional nursing station design. The design concepts were based on benchmark reviews of other Planetree projects and interpretations of published research.
This Trinity Medical Center patient room design, the result of patient preference research, includes a foldout sleeping chair for overnight visitors
Level-two practitioners: Evidence-based practitioners at level two take another important step. Based on readings, they hypothesize the expected outcomes of design interventions and subsequently measure the results. This level of attention makes design less subjective and requires attention to new ways to design. At this level, the designer must understand the research, interpret its implications, and be able to build a chain of logic to connect the design decision to a measurable outcome. This discipline reduces the number of arbitrary design decisions in healthcare projects and delivers solutions linked to outcomes. It also raises the challenge of preventing bias from reducing objectivity in the gathering and reporting of results. The evidence-based practitioner must strive for the truth and resist the temptation to tilt the reporting of findings to emphasize success or downplay failure.
A level-two project of note was the replacement hospital WHR designed for St. Michael Health Care Center in Texarkana, Texas. The design team hypothesized reductions in travel distance in the patient units. Travel distance was measured, once the design was constructed, and demonstrated clear improvement. But the research also told a valuable story. In developing a pattern of decentralized nursing intended to have the caregivers closer to bedside during rounds, it was expected that nurses would work in alcoves near clusters of beds on the busy morning shift, spend more time at the two traditional nurse stations in the afternoon, and collapse their activity to a single station at night. This pattern did not hold. Nurses found that proximity to the patient influenced their preferred work location on all shifts, and they used the decentralized features on each shift.
Level-three practitioners: In addition to following the literature, hypothesizing the intended outcome of their design interventions, and measuring the results, these designers report their results in the public arena. Publishing in the popular press or speaking at conferences makes an important contribution to the field and advances the state of the art. It also subjects the designers' methods and results to the scrutiny of others who may or may not agree with them. These practitioners soon learn they need to better understand research methods and often seek some level of advanced education to promote greater rigor in their documentation of results.
I would submit that the speaking and publishing on healthcare design topics by members of our firm fits the level-three model. In my own case, I gained insight into qualitative and quantitative research methods while earning an advanced degree in organization development (see “Relating Facility Design to Organization Design,” p. 26, HealthCare Design, September 2003). My thesis was an exploratory case study that investigated the relationship between the design of patient units at two hospitals and the relevant organizational performance. I used quantitative data from several economic, clinical, and satisfaction indicators to measure the organizational performance, as well as qualitative data from surveys and interviews recording the opinions of nurses, physicians, and executive leadership. I have since begun to use methods developed for the thesis to evaluate other projects by our firm.
Level-four practitioners: Practitioners/scholars at this level perform the same tasks as those at the other levels—following the literature, hypothesizing the outcomes of design interventions, measuring the results, and reporting in the public arena. These practitioners take the next step by publishing in quality journals that require review by qualified peers. They may also collaborate with social scientists in academic settings who contribute to the formal literature. These level-four evidence-based practitioners are working directly in the field, designing and building operating healthcare facilities, but they are also subjecting their work to the highest levels of rigorous review and formally advancing the useful evidence in the field.
For example, working for The Institute for Rehabilitation and Research in Houston, WHR developed an innovative concept in their recent master plan. The hypothesis was to include greenery and live plants within the physical therapy area, based on implications from research about the role of nature in the healing process. Prior to developing a funded project, a Fellow in WHR's office performed a study of staff and patient opinion about the type and amount of greenery that might be used (figure 3). In essence, the study sought staff/patient reactions to a set of overlays of increasingly progressive vegetation on a typical gymnasium scene, as well as to a scene of natural vegetation for comparison. The study was done with review by academic researchers from Texas A&M University who had participated in the original research. The overlay comparison method was formally approved by the university's and the hospital's institutional review board (IRB). This process fits the profile of a level-four project, and it resulted in heightened credibility of the research findings.
A composite study for an indoor landscaping project at The Institute for Rehabilitation and Research, Houston. Overlays on a typical gymnasium scene showed the effect of increasing greenery, and staff were asked to indicate preferences. Outdoor photo allowed choice of an outdoor setting as the preferred venue for access to nature
It Isn't as Easy as It Looks
Like motherhood and apple pie, evidence-based design should be widely popular. It seems none but the most jaded and ego-driven, would-be artistic geniuses could object to design based on the knowledge that specific concepts can be used to achieve predictable positive results in healthcare settings. The dark side of this trend is the appearance of practitioners who would like to be associated with evidence-based design but who have not been able to do the hard work required to become current. There are almost endless potential sources of information, and there is a need to reach speculative conclusions about the design implications of highly specialized and narrow studies. The role of the architect in translating the research into useful designs in the field is crucial. It is necessary to determine whether these translations deliver the intended outcomes.
Inexperienced practitioners will find it difficult to make the leap from data about clinical conditions such as heart rates or mortality to the successful design of a patient room or visitor facility. The vast number of confounding variables in any healthcare setting renders prescriptive rule making or single-minded solutions suspect.
Then there are the “level-zero practitioners,” individuals who grasp the concept that the environment has an effect on those who are in it, and that there is evidence to support various conclusions about those effects. These people, while they might mean well, often take isolated comments from an article or a conference presentation, make a personal interpretation that fits their design bias, and claim the subsequent design is evidence-based. Level-zero practitioners have rarely read the original research, do not understand how to draw valid, broad inferences from narrow and precise studies and, as a result, misapply important principles.
Without hypothesis and measurement, these practitioners complete a project and search for any observable success. While their successes may be trumpeted as “evidence,” the absence of a prospective statement of design intent breaks any link of planned causality. Such a design may have observed outcomes, but it was never evidence-based.
Another aspect of level-zero practice is that it might promote a project with carpeting and indirect lighting as a “healing environment,” although the project features carpeting that supports bacterial growth, wallcoverings that harbor pathologic organisms, poorly located hand-washing sinks that discourage good infection control, inappropriate abstract art, and inefficient support space for staff.
Architects often tend to focus on visual media, three-dimensional modeling, and kinesthetic experience to understand space as they concentrate their efforts on designing buildings. This partially explains why few architects write articles and why there has been only a modest volume of literature specific to healthcare architecture. There is, however, a small but growing body of literature from architects, interior designers, landscape architects, healthcare professionals, and environmental scientists regarding healthcare design.
Numerous sources of information that are potentially helpful to the designer are available. Delving into this body of material and taking time to draw the proper design inferences from such diverse sources are challenging. Among the most directly relevant sources are the writings of authors from the field of environmental psychology. While literature from the fields of medicine, nursing, management, engineering, industrial design, and technology are helpful, the literature of psychology, sociology, anthropology, and economics are also relevant sources.
The popular press includes science journals, newspapers, and magazines, as well as documentary films and television programs that sometimes delve into healthcare design. Other available sources of information include industry data guides, guidelines from specialty boards, quality review data, infection-control data, manufacturers' testing information, association reports, and documents of accreditation agencies and code authorities. Practitioners can gather new information from conference presentations, workshops, continuing education programs, and benchmarking tours of exemplary facilities. The Internet has also become a rich source of information for designers.
In my experience, the best way to filter the nearly overwhelming pool of information sources is to use an orderly design process. Establishing the project goals at the outset helps the team identify the significant design issues and problems to be addressed. Opportunities for evidence-based design can be identified in the programming and predesign phase. The available research is then combed for material relevant to the narrower list of those design issues. Judgment is then required to sort potentially conflicting implications of the research.
Critical thinking helps identify the most credible research and the most important implications for design. Designers can add this information to the factors they use to test alternative design concepts, just as on any conventional project. I have always been inclined toward process in design, and the use of research findings has always been a part of my attempt to understand the problem posed by a project.
Architects are rarely taught about research methods and, as practitioners, most feel they do not have the training to fully understand, much less perform, serious research. With a focus on tangible projects and standard contracts that offer no fees for research or postoccu-pancy evaluations, few architects can see a way to make time for such unfunded efforts, no matter how worthwhile. If the client isn't willing to pay for the evaluation, how can the architect afford to expend this effort? Perhaps the promise of better projects, each demonstrating measurably better results for happy clients, makes the marketing case.
Evidence-based practitioners can feel good about proven results associated with their work and can be differentiated in the marketplace by clients who seek higher performance from their costly projects. WHR has at times been chosen for a project in recognition of its research-based design methods.
I believe that the use of research findings to improve design de-cisions comes naturally for many designers. Adding a level of rigor to what we already do is a major element of the shift to evidence-based practice. It is the data gathering and postoccupancy research that frequently are not funded. One useful approach is to enlist the client's available resources in the effort; i.e., the architect must learn to request needed data that are already being collected by the client. The architect should try to convince the client of the value of post-occupancy data for comparison. The best results will come from an unbiased, independent third-party evaluation.
Ultimately, architects may need to spend time and money collecting data for their own purposes. There is a clear business case for good design, and an even stronger case for design linked to evidence of positive economic, clinical, and satisfaction outcomes.
Architects have a moral obligation to use the best and most reliable information available in the design of healthcare facilities. This is a sacred public trust, granted with professional licensure. Like Pandora's box, which once opened could not be closed again, the moral obligation of a practitioner, once evidence-based design has been encountered and understood, cannot be avoided.
Evidence-based design appeals to the scientific minds of physicians and other clinicians who are trying to practice on the basis of medical evidence. It offers the prospect of improving clinical outcomes, and it gives patients and families the prospect of a higher-quality experience in their healthcare encounters. The public, consumer groups, and payers are pleased with anything leading to more effective and lower-cost healthcare.
Evidence-based design also appeals to the business-minded administrative leaders of hospitals. It offers them the prospect of re-duced costs and/or improved organizational performance and can provide justification for some of the costly decisions made on their building projects.
Exemplary evidence-based architecture comfortably blends the architect's rich experience, understanding of classic design principles, and creative inspiration with design decisions based on insightful interpretation of a broad range of research results. The trend toward research-informed designs is profoundly transforming the field for the better. Architects should embrace evidence-based design with excitement and begin to enthusiastically explore this fertile ground through serious study of clinical outcomes, economic performance, organizational effectiveness, satisfaction measures, and their relationships to the physical settings of healthcare. The physical environments they design will have a measurable positive impact in each of these areas. Evidence-based design signals the dawn of a promising and hopeful era in healthcare architecture.
D. Kirk Hamilton, FAIA, FACHA, is a founding principal with Watkins Hamilton Ross Architects in Houston, and leader of Q Group Advisors, the firm's consulting division. He is a past-president of the American College of Healthcare Architects and the AIA Academy of Architecture for Health. He is a member of the board of directors of The Center for Health Design and the Coalition for Health Environments Research. He has authored and edited three books on health facility design and is currently working on a new book about evidence-based design for critical care. He has recently completed a Master of Science in Organization Development at Pepperdine University. For further information, call (713) 665-5665, e-mail
firstname.lastname@example.org, or visit
To comment on this article, please send e-mail to email@example.com.
- Becker FD, Poe DB. The effects of user-generated design modifications in a general hospital. Journal of Nonverbal Behavior 1980; 4:195-218.
- Buerhaus PI, Staiger DO, Auerbach DI. Implications of an aging registered nurse workforce. Journal of the American Medical Association 2002; 283:2948-54.
- Burrington M. Can private rooms be justified in today's healthcare market? Houston:Center for Innovation in Health Facilities, 1999.
- Canter DV, Canter S (eds). Designing for Therapeutic Environments: A Review of Research. New York:John Wiley & Sons, 1979.
- Cassidy R. Positive prognosis. Building Design & Construction, Feb. 1, 2003.
- Copeland Y, Johnson LB, Orr R. Opening the gateway to change: Creating a human-centered medical center—strategies for competing in the healthcare marketplace. Journal of Healthcare Design 1997; 9:105-8.
- Coulter S. Redesign not downsize. Online Journal of Issues in Nursing, Jan. 6, 1997. (http://www.nursingworld.org/ojin/tpc2/tpc2_6.htm)
- Cresswell JW. Research Design: Qualitative and Quantitative Approaches. Thousand Oaks Calif.:SAGE Publications, 1994.
- Dilani A (ed). Design & Health: The Therapeutic Benefits of Design. Stockholm:Svensk Byggtjänst, 2001.
- Durham J, Hayward C. Future design trends: Facility design strategies for operational restructuring. Journal of Healthcare Design 1996; 8:85-9.
- Gallant D, Lanning K. Streamlining patient care processes through flexible room and equipment design. Critical Care Nursing Quarterly 2001; 24 (3): 59-76.
- Gerteis M, Edgman-Levitan S, Daley J, Delbanco T (eds). Through the Patient's Eyes: Understanding and Promoting Patient-Centered Care. San Francisco:Jossey-Bass, 1993.
- Gifford R. Environmental Psychology: Principles and Practice (2nd ed.). Needham Heights Mass.:Allyn and Bacon, 1997.
- Gilpin L, Nelson K, Schweitzer M. A healing environment: The Planetree Hospital Project at San Jose Medical Center. Journal of Healthcare Design 1991; 3:139-48.
- Hall ET. The Hidden Dimension. Garden City N.Y.:Doubleday, 1966.
- Hamilton DK. First design the organization, then design the building! Interiors & Sources, Jan./Feb. 2002.
- Hamilton DK (ed). ICU 2010: ICU design for the future, a critical care design symposium. Houston:Center for Innovation in Health Facilities, 2000.
- Hamilton DK (ed). Unit 2000: Patient beds for the future: A nursing unit design symposium. Houston:Center for Innovation in Health Facilities, 1993.
- Heimsath C. Behavioral Architecture: Toward an Accountable Design Process. New York:McGraw-Hill, 1977.
- Horowitz SF. Design trends: Designing for advanced therapeutic outcomes—Acute care. Journal of Healthcare Design 1995; 7:103-8.
- Horsburgh CR Jr. Current issues: Hospital design qualities to facilitate healing. Journal of Healthcare Design 1997; 9:89-92.
- Institute of Medicine Committee on Quality of Health Care in America. Crossing the quality chasm: A new health system for the 21st century. Washington D.C.:The National Academies Press, 2001.
- Kelly K (ed). Health Care Work Redesign, Series on Nursing Administration ,volume 7. Thousand Oaks Calif.:SAGE Publications, 1995.
- Langley GJ, Nolan KM, Norman CL, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. San Francisco:Jossey-Bass, 1996.
- Larson JA. Healthcare Redesign Tools and Techniques. New York:Productivity, Inc., 1997.
- Lathrop JP. Restructuring Health Care: The Patient-Focused Paradigm. San Francisco:Jossey-Bass, 1993.
- Malkin J. The business case for creating a healing environment. Board Room Press, The Governance Institute ,October 2002.
- Malkin J. Hospital Interior Architecture: Creating Healing Environments for Special Patient Populations. New York:Van Nostrand Reinhold, 1992.
- Orr R. Healthcare environments for healing. Journal of Healthcare Design 1989; 1:71-6.
- Orr R. The Planetree philosophy. Journal of Healthcare Design 1992; 4:29-34.
- Press I. Patient Satisfaction: Defining, Measuring, and Improving the Experience of Care. Ann Arbor:Health Administration Press, 2002.
- Rich M. Healthy Hospital Designs. Wall Street Journal, Nov. 27, 2002, B1.
- Roesch A. Facilities solutions for reducing clinical costs. Facilities Planning News 1994; 13 (4): 3 ,20.
- Rubin HR, Owens AJ, Golden G. Status Report: An Investigation to Determine Whether the Built Environment Affects Patients' Medical Outcomes. Martinez Calif.:The Center for Health Design, 1998.
- Runy LA. The dynamics of satisfaction. Hospitals & Health Networks 2002;(Nov.):57-62.
- Shepley MM. A users' guide to healthcare design research. Journal of Healthcare Design 1997; 9:9-12.
- Sommer R. Personal Space: The Behavioral Basis of Design. Englewood Cliffs N.J.:Prentice-Hall, 1969.
- Sommer R. Social Design. Englewood Cliffs N.J.:Prentice-Hall, 1983.
- Steele FI. Physical Settings and Organization Development. Reading Mass.:Addison-Wesley, 1973.
- Ulrich RS. Effects of interior design on wellness: Theory and recent scientific research. Journal of Healthcare Design 1991; 3:97-109.
- Ulrich RS. A theory of supportive design for healthcare facilities. Journal of Healthcare Design 1997; 9:3-7.
- Ulrich RS. View through a window may influence recovery from surgery. Science 1984; 224:420-1.
- Williams M. The physical environment and patient care. Annual Review of Nursing Research 1988; 6:61-84.
- Zeisel J. Inquiry by Design. Monterey Calif.:Cole Publishing, 1981.
Healthcare Design 2003 November;3(4):18-26