Editor's note: John G. Reiling, PhD, is president and CEO of Safe by Design. Dr. Reiling consults with hospitals and health systems nationwide regarding facility and patient care designs that emphasize safety, error reduction, and quality.

Dr. Robert Wachter, Editor, AHRQ WebM&M: Tell us what the National Learning Lab was.

Dr. John Reiling: It was conducted in April 2002. The Learning Lab participants were a collection of the leadership in the country focused on safety. Fifteen external organizations were present including AHA (American Heart Association), AMA (American Medical Association), JCAHO (Joint Commission on the Accreditation of Healthcare Organizations), and IHI (Institute for Healthcare Improvement). We also had our design and construction team present, architects, contractors, etc. The purpose was to discuss the underlying question: does the physical environment impact patient safety? If the answer was yes, then we would ask the question: "How would you design a facility with its equipment and technology focused on safety?"

RW: What would you say were the core principles that came out of that exercise?

JR: There were four core principles as a result of the National Learning Lab. The first principle is to design around the adverse events. For example, design around medication errors. The second principle is to design around latent conditions. For example, design for standardization, fatigue, and visibility. The third principle is to change the facility design process to focus on safety. For example, conduct a matrix exercise at the beginning of design and do mock-ups day one. The fourth principle is to change processes to focus on safety and to enhance patient safety culture. For example, develop common beliefs and implement Lean processes.

RW: Let me try to pin down how that plays out. Let's start with fatigue. You want to design around fatigue. What do you actually do?

JR: There was a recommendation to create what I call a matrix, a quality functional deployment tool, where we take an issue like fatigue, form a group that represents the people doing the work, and ask, "What causes fatigue?" And they had a whole list of things: walking around too much, not having supplies when the caregivers need them, standing up all day, noise, and hard surfaces to walk on. They talked about what causes them stress, some of the caregivers talked about the noise, which was a latent condition we were supposed to design around also. A lot of those ideas don't have anything to do with the physical facility, but a lot of them do. So we said we could make the place quiet. We could have soft floors for people to walk on. When they got into the room, we could have a place for them to sit down while charting. We could make sure that supplies are available at or near the bedside to decentralize pharmacies, so they're not running back to a pharmacy room. We tried to make up ideas around the process and around the location of things that would decrease steps and make things quieter.

Then we made a chart of ideas around all latent conditions and adverse events: medication errors, noise, standardization, and infections. We ultimately chose 9 active failures and about 9 latent conditions. We included these listings in our chart to improve safety and understand how much they would cost. For example, we thought putting medications at the bedside would lower medication errors, and it would also create less fatigue. We would crosscheck all these suggestions and measure the impact of the design feature against the cost and its perceived impact on safety. If something cost a million dollars and we didn't think it would have much impact, we ruled it out. We used this matrix exercise to try to get a series of design features at the front end. This is before we designed anything about what we wanted the building to look like, so when we designed we had the desired impact on safety and quality.

RW: I'm struck with one tension that you must feel in this kind of work, which is that you want quiet environments where people can do their work without distraction, yet you probably want to facilitate contact between clinicians, patients, and teamwork. How do you create that balance?

JR: I'm not sure they're mutually exclusive; I think that they can be inclusive. Last year a national group had a conference sponsored by AHRQ (Agency for Healthcare Research and Quality) and the Center for Health Design. The conclusion that the group came to was that it is important to do a lot of design work and discussion on the front end. If you get too deep in the design process, it's much more difficult to then say, "What can we do to make it safe?" You have to talk about these issues at the front end to get your processes focused on safety. For example, this will have the organization focused on adverse events and creating culture change to eliminate or minimize medication errors.

The first point is that having a quiet environment is a good idea irrespective of how people are organized within that environment. I see no downside and a lot of upsides to having a quiet environment. The only complaint we've heard about having very quiet environments is that if staff doesn't have enough contact with patients, the patients get a little nervous because it's so quiet that they wonder if anybody's around. This should put pressure on the organization to make sure that there is good contact with patients. We did studies with an AHRQ grant measuring how often people were in the room, and we found that 7 out of 8 hours during the day there was no contact with the patient. We wanted to change that.

The second point is, how do you create an environment that isn't distracting but also promotes teamwork? My view is that you have to create spaces in the patient room. Let's assume you're in an academic medical center where people can get into the room and be there as a group and talk. Then you have to create a quiet, non-distracting space where they can debrief. I was involved with a design in Florida where we had some glass conference rooms so you could see patients and a place where teams could get together and react. There were also alcoves where people can chart immediately about the patients' conditions. If you're in a traditional non-academic health center where you know caregivers are rounding, they should have a place where they can chart and order right away. We think a lot of error happens because clinicians try to batch these rounding processes, go back to a nursing station, and download all of what they've been thinking about. We took a Lean view of this process and created a continuous flow.

RW: So does that mean computer terminals in all the patients' rooms?

JR: Exactly, in addition to in all the rooms, we had computer terminals in alcoves next to all the rooms. If people wanted privacy, they could leave the room, sit down, look back into the room, and then call up and do all their ordering and communications. Only then would they go to the next room. This is a classic Lean principle (continuous flow), but our physicians were very concerned about how much time it would take to do a continuous flow and look back at the patient in the room. However, once implemented, they found out that they cut their rounding time materially by changing the process.

RW: What are some of the principles of designing a health care facility to try to minimize the risk of infections?

JR: There's a lot of data around infections and their causes. HGRC (Health Guidelines Revision Committee) is always trying to figure out ways of reducing infections. When you look at the facility around how these bacteria continue to live, the first observation is that people are walking into the room without washing their hands. In my judgment, it is the most serious problem. It is important to design environments so that it's convenient and easy for workers to wash their hands in patient care spaces. Technology is being developed to monitor whether or not they did it. In other words, there can be reminder systems to do it or there can be recording about whether or not it was done so there can be some accountability. The fundamental thing is to create a process where people wash their hands. I think it's one of the most interesting cultural, process, and cognitive issues because we have such poor performance around washing in an environment where everyone knows it's important.

There are other issues too. In a Midwest setting with hot summers and cold winters, we found evidence that condensation would form on the windows and drip down to the bottom, and bacteria would form in that moisture. We developed a fan that blew at the bottom to minimize the condensation. We didn't have drapes on the windows that bacteria could build up on; our blinds were in the window. We made sure all the surfaces were hard surfaces with no seams so that bacteria wouldn't grow. We talked about making sure the airflow systems were designed so that the vents and the grills were stainless steel and easy to clean. We put HEPA filters for the entire hospital and ultraviolet technology in most of the critical spaces in the hospital. We received an AHRQ grant to study adverse events and latent conditions and their impact on safety. Through that study, we found that we reduced hospital infections by 80%.

RW: When people talk about design safety, one issue that always comes up is standardizing the room so every room looks exactly the same and all the things are stored in the same places. How important is that?

JR: I think it's very important, but it's interesting how controversial this is. It is important to try to understand cognitive functioning, why people make mistakes, and what are the conditions in which they won't make as many mistakes. People talk about smart thinking, which is more aimed at how to create habits that will make you become smarter in terms of looking at situations, but habits are an example of creating something that becomes subconscious. For example, you don't think about brushing your teeth, you just do it. The quality of a habit that works is to do it over and over again in a standardized way. The evidence is that if you do these subconscious activities and the circumstances are the same, these habits become more accurate than thinking.

Another example is you have a standardized way of driving home, and you just drive it all the time—it's amazing how many times you don't remember driving home because your brain has figured how to get you home. All that is to say is that standardization is an important thing. I don't see how people can argue that trying to standardize is not a good idea. In addition, from a cost standpoint, I don't see the downside. The evidence in the projects I have been involved with where we have standardized as much as possible is that the construction costs are less. I think the field should standardize until there's evidence that you shouldn't.

St. Joseph's was a small hospital, only 80 beds. We were told at first that it would cost us a quarter of a million dollars more to standardize because by standardizing the rooms you would have to bring up through every wall line, the gases, the water, etc. But our construction company said it actually cost them less, they were able to prefab a lot of those rooms' components offsite because they were all the same. They also felt that the construction time would be faster because the workers wouldn't be walking into the room saying, "In this room I have to put the soap dispenser on the left side, and in another room I have to put it on the right side. And I forgot which room I'm in now." My reaction is you should start out every project and standardize everything that you can think of. Then if there are mitigating circumstances you compromise standardization, not the reverse.

RW: You've mentioned cost a few times so let's turn to that. For a 200-bed hospital, what would the difference in cost be between designing it the way that we might have 20 years ago and designing it with all of the safety principles you've mentioned? Infection prevention, standardization, and adequate spaces for both privacy and to facilitate teamwork—are we talking about a 5% difference, a 20% difference? What's the magnitude?

JR: I've heard a lot of discussion that if you are going to improve safety and quality in design you have to add a 10% to 20% cost premium. I don't buy that. My view is to understand the amount of capital available and you live within that limit. Back to my point about the front-end matrix. You need to decide what the best features are that will match the safety needs for the amount of capital available. You have to make tradeoffs: I can either afford to put HEPA filters in the entire hospital, or I can afford to have a luxurious lobby. If we do the front-end matrix properly within the budgeted capital, I think you will receive outstanding value; we saw this at St. Joseph's.

RW: I'm going to turn to the role of regulators and accreditors. I'm guessing that the reason you paid a lot of attention to fire safety 25 years ago but less attention to other kinds of safety is there were strict regulations around fire safety from both the Fire Marshall and The Joint Commission, and relatively few regulations around some of these other kinds of patient safety. First of all is that true? And second of all, do you think that regulators and accreditors are acting appropriately in terms of the standards they set around these other safety issues?

JR: Your observation I think is right on. The reason why there's so much work done with fire alarms and everything else in institutions is because there are regulations around it. To get approval to build your facility you must meet code. This is how the regulators affect safety. However, the complexity is that regulators want to be black and white about the effect of a design feature. You need a study to show the feature improves safety. I'm on the committee creating the new national guidelines on adverse events. For example, data for medication errors has shown that reducing interruptions and having a quiet "pharmacy zone" or "medication zone" reduces errors. I believe the regulators will adopt these changes because they want to create conditions where people are safe and there is data to support this.

By creating a facility that has less worker fatigue, there should be less worker error. Creating standardized environments should create conditions where people can standardize their processes. But, there are no studies that can directly link a standardized environment to less error because of intervening variables. Therefore, it's difficult for regulators to adopt a specific design element and say we're going to make people do this, because these facilities are designed without having clear data to support this.

RW: If I were a student in architecture school but my intention was to go into health care facility work, would I be learning any of this stuff?

JR: A few programs have people come in and guest lecture around these topics. Some schools have a specialty architectural program. They are starting to have people think about these issues in the design. Some architectural schools have received federal funding to create the hospital room of the future, in which safety is a component. I think we're at an important transition. But in general I wish architectural schools in this country would focus more on safety by design.

RW: Thinking 10 to 20 years from now, what does the hospital room of the future look like?

JR: If you take emerging technology, culture management issues, idealized process, and payment reform—if you take all these variables and a deep understanding of cognitive behavior, would you operate the way you are now? I think the answer is no. The hospital room of the future is going to be very technologically advanced, highly digitized, and digitized in a way where the information is automatically going into the medical records. We're also heading for a system with Watson [IBM's computer] and decision support technology. Having said all that, you're going to continue to see single-patient rooms, which will look more like intensive care. More procedures are going to be done at the bedside so that you're not moving patients because it's risky to move patients. Spaces will have to be large enough so teams of people can come in those rooms and help around that. We're going to have enhanced technology so patients can see what's going on with their care process.

RW: Anything you wanted to talk about that we didn't get to?

JR: I'd like to talk about the importance of the architectural process—that the design process itself needs to be modified to create safety. We talk about design elements like how to get rid of infections, etc. But the way that the architect interfaces with the organization and goes about the design process can have an impact on the outcomes. So architects and organizations—and organizations may have to drive this more than architects—need to sit down at the front end and describe their objectives for high levels of safety and the outcomes that they want. The organizations need to say to the architects, "We want to have fewer adverse events. We want you to facilitate a matrix exercise on the front end so that we can, as a group, think through all the safety issues. Given the capital we have to spend, we want to produce the highest level of safety. Finally, we want to go through a process where we are actively involved in the design in all aspects including mock-ups." The way you design and the role of the architect and the organization in the design process needs to be modified to have a lasting impact on improved design for safety and quality.