"In Search of Evidence - Uncovering Information on Hospital Design", (c) D. Kirk Hamilton,  Health Facilities Management (8/07)

Evidence-based design is a growing long-term trend in the field of health care architecture. Design informed by research has been used to reduce patient and staff stress, reduce infection and preventable errors, promote patient safety, deliver greater patient throughput and a host of other quality and economic measures. But, what can designers do when there appears to be no evidence to guide them toward achieving a particular outcome, or if they don’t know where to start looking?

The most probable answer is that the designer may not have looked hard enough or they may not have looked in the right places. They may need to choose different key words in their search for information. Or they may need to try looking at similar conditions outside of health care. Designers should keep trying until they are absolutely certain there is nothing that they can critically interpret for its influence on their design assumptions.

There will, of course, be instances where the design issue to be addressed has not been the subject of credible research. An absence of academic material is often the case. In many instances, the topic also will not have been the subject of opinion pieces or speculative and informal writing. Designers may feel they are covering new ground or perhaps the old ground is so well trod that custom has governed without systematic intellectual analysis or questioning. However, even if there is no credible research, there is still relevant information available.

Where to begin

I was recently meeting with some folks at a hospital in the Northwest that is widely recognized for its quality of care. Their architects had been told that the hospital intended to implement a neutropenic unit. The architects had no idea what this odd word meant. Neither did I, to be absolutely honest, but I ventured a guess that it meant a unit for immunocompromised patients. The oncology nurse confirmed my guess, but we didn’t have time to elaborate. This brings up one of the first rules of looking for design evidence: Begin with clear definitions.

Here is a summary of the Wikipedia (www.wikipedia.org) definition: Neutropenia (or neutropaenia, adjective: neutrop(a)enic) is a hematological disorder characterized by an abnormally low number of neutrophil granulocytes (a type of white blood cell). Neutrophils usually make up 50 percent to 70 percent of circulating white blood cells and serve as the primary defense against infections by destroying bacteria in the blood. Hence, patients with neutropenia are more susceptible to bacterial infections and without prompt medical attention, the condition may become life-threatening. Neutropenia can be acute or chronic depending on the duration of the illness. A patient has chronic neutropenia if the condition lasts for greater than three months. Neutropenia is sometimes used interchangeably with the term leukopenia. However, neutropenia is more properly considered a subset of leukopenia as a whole.

Many of my academic colleagues scoff at the use of Wikipedia as a source because anyone can enter the system and edit it. On the other hand, if designers are watchful for dubious material and confirm what they find in other sources—like a good encyclopedia or dictionary—it can be a useful starting place in a search for information.

Definitions of a topic or its key terms are mandatory for a proper search. The Wikipedia entry suggests I should search for “neutropenia” in addition to “neutropenic.” The American Heritage Dictionary gives much less information about neutropenia: n. An abnormal decrease in the number of neutrophils in the blood.

The definitions alone suggest that a unit for this type of patient would be extremely specialized and likely to have a small census. I wonder if the client was throwing around a big word for effect, when a less precise word, or a description of an isolation unit for patients whose immunity was compromised by chemotherapy drugs as well as patients whose immunity has been suppressed due to other reasons, would have sufficed? Perhaps the architects will let me know when they find the answer.

Expanding the search

After the definitions of topics and key words comes a more thorough search. Designers should already know how to perform a search using Google (www.google.com), Yahoo! (www.yahoo.com), AltaVista (www.altavista.com) or one of the other widely available search engines. If not, their teenagers should be able to show them how. In any case, designers should resist the temptation to stop their search after having only explored a few sites.

It is also important to specifically search the available literature. I find it interesting that most design practitioners do not automatically use a formal academic literature search when exploring a topic. I suspect this is because most of us weren’t taught how to make such a search and we lack confidence in our ability to extract useful information from highly academic writing. This is unfortunate because the information found in scholarly and peer-reviewed literature may offer important insight into a design.

PubMed (www.pubmed.com) and MEDLINE (www.medline.com) are online sources for articles from clinical authors. The local university library can also help in the literature search.

Google’s “Scholar” feature provides another search engine that combs journals and literature. In a search for “neutropenia,” for instance, Google Scholar turns up “about 145,000” citations. A quick scan of those results tells me that the neutropenia patients can include bone marrow transplant patients, and that: The accepted standard for treatment of neutropenic fever remains inpatient therapy with IV broad-spectrum antibiotics (Moores, 2007).

The above quote comes from a promising article turned up by Google Scholar called “Safe and effective outpatient treatment of adults with chemotherapy-induced neutropenic fever” by Kevin G. Moores, PharmD. It is from his abstract. The paper appeared in the April 2007 American Journal of Health-System Pharmacy, and Moores is the director of the Division of Drug Information Service at the University of Iowa’s College of Pharmacy. That seems to make him and his current conclusion credible.

More to the point, the quote also begins to tell me what is likely to be done in such a patient unit. There will be a regimen of antibiotics given while patients receiving chemotherapy or bone marrow transplants are watched for signs of infection and protected from airborne sources of infection by filtered positive-pressure isolation. There will also be serious attention paid to contact transmission and hand hygiene.

The range of sources for relevant evidence is virtually infinite. There are many domains of knowledge and architects are usually educated in only a few. In this case, the search could be about neutropenic diseases and their treatment, facilities designed to offer such treatment, design of isolation units, high-efficiency particulate air-filtered air handling systems, design of patient rooms, artwork appropriate for cancer patients, social support for patients in isolation, medication errors, decentralized caregiving or staffing efficiency.

This would only scratch the surface of possible topics, so the search must be narrowed based on the client’s goals for such a project. For some clients, for instance, the issue of information systems and electronic records would be among the key design issues.

In fact, because the designer is doing the work on the client’s behalf, the client’s information and data will be vital to the success of the project. If it were my project, after spending more time searching the literature and exploring new directions suggested by my reading, I might be ready to share what I learned with the client.

These meetings should include the project’s champions. That way, the designer can find out whether the information matches what the project champions were intending. The designer should ask: “What are the examples of this type of patient unit that caused you to want to develop one at your hospital? Do you know anyone we can talk to at such a unit?”

These questions could lead to a list of promising physical sites to visit. Benchmarking comparable sites for their lessons learned and, especially, what they wouldn’t do again, is another potentially valuable way to gather relevant design information.

As health care designers proceed through this process, they must always keep current in their specialty. They should read what their clients read and keep up with the constant advances in the health care field. They should make it a point to maintain awareness of regulatory changes. For instance, important information relevant to a project may appear in today’s news.

They should also keep up to date by looking to the lessons learned in their own practice. Field observations and measurements are important sources of relevant data for subsequent projects.

Designers should not discount the importance of their experience in the information equation. A designer’s experience is most credible, of course, if it is in the form of carefully documented and measured material rather than casual anecdotes. Designers should also seriously consider publishing what they’ve learned in their own practice.

Ultimately, the real problem with information is how to decide when there is enough to make a good decision. Designers need to feel confident that they have carefully scanned the field for relevant information. They need to have actively sought differing opinions. They must also discover conflicting findings and explore why one potential direction should be chosen over another.

They will have built up a body of information from a variety of sources, and will have used critical thinking to interpret its implications for the specific situation of the unique project and the particular client.

They can follow the path indicated by the majority of what they have found, or they may deliberately choose to stray from the preponderance of the evidence and base their design decisions on a hypothesis about the design’s likely impact on a desired outcome. In either case, they should measure the outcome and report their findings so those who follow can learn from the path chosen.

Good luck!

The search for best current evidence from research and practice can sometimes be a daunting task. On the other hand, a successful search can be very satisfying.

The most difficult step is to get started on the right foot. Hospital designers should be sure to explore a few different entry points to the knowledge base they seek, and be sure to explore domains of knowledge that are not their own areas of expertise. They should also seek advice and counsel from others around them, including experts in the field.

I am convinced that once designers get the hang of these new design tools, they will find the process and the results to be fully rewarding. Good luck with that next important search!