Research in Practice: Lesson 2
“Research in Practice: Lesson 1—Testing Same-Handed and Inpatient Rooms with Canted Headwalls Using a Master Database” appeared in the May 2012 issue of Healthcare Design. In that article, we discussed the benefits of developing a design research master database to investigate hard-hitting issues with inpatient room design.
However, design research does not actually begin with a master database. In this article, we explain the challenges of identifying the right problem and techniques for prioritizing problems.
Seeking out a problem is part science and part art. Individuals new to design research or who desire textbook solutions will find their first efforts especially challenging. Seasoned PhD design researchers in practice recognize that seeking out problems requires a careful balance of rigor, flexibility, and discovery. It is, assuredly, the most important step in obtaining results that are both relevant and enduring to the design process.
In this article, we’ll introduce readers to problem identification and will explain a little about pilot testing, an often-overlooked yet effective technique to seek out and identify problems. Well-structured pilot testing can be enormously helpful for a project team because it can be used to discern whether the problems originally identified are the right ones to target.
Which old problem? The wicked problem
The healthcare field is rife with multifaceted and interrelated problems. It is very common to find that the solution to one challenge reveals (or is related to) another. It is some consolation that these tangled challenges have been the subject of inquiry for several decades. They are formally known as “wicked problems” (Rittel & Webber, 1973) and have been recently characterized in design research as “… complex, persistent, and contradictory problems relevant to broad groups of stakeholders, each having a different potentially diametrically opposed perspective on an issue” (Hamilton, 2008).
Simply put, a wicked problem is, among other things, complex, ongoing, and enmeshed with other problems. Often, resolution of a wicked problem requires a tradeoff based on what is most important to the organization.
Examples of the wicked problems that routinely confront healthcare facilities might include staff inefficiencies, patient falls, hospital-acquired infections, and medication errors. For example, decentralization of medications can cut back on the number of medication errors due to distractions.
On the other hand, it also may lead to fewer opportunities for collaboration, shared awareness, and informal learning opportunities among staff. How you understand your problem tradeoffs can lead to different solutions.
Refining problems: An example where every step counts
Nurse walking distances can be a multifaceted problem related to other problems. How do you optimize nurse travel distances when each facility has specific challenges, multiple drivers that contribute to nurses’ walking distances, and there are several possible solutions? One popular solution is decentralized nursing.
However, there is a veritable menu of decentralized nursing options including room-side supply cabinets, workstations-on-wheels, bedside computer terminals, nurses’ stations outside of each room, and nurses’ stations every few rooms. Which you choose ultimately depends on a good understanding of your problem.
Examples from our project work illustrate how something as seemingly simple as nurse walking distances can be complex and idiosyncratic. Statistics from one hospital with distinct and small inpatient units showed us that nurses walked beyond the industry average largely because they were continually untangling cords around cluttered headwalls and outlets.
At another hospital with several decentralized nursing strategies, statistics showed that increased walking was attributed to documentation and medication retrieval at the main nurses’ station. Understanding the problem helped to refine it and enable potential solutions.
Pilot testing: A twist on an old technique
How can you tell that nurse walking distances is a major issue for you? How can you find the causes in your organization?
A twist on an old technique, pilot testing, will help.
Traditionally, pilot testing involves a dry run of all the protocols and tools of a research study. If you are new to design research, you can think of pilot testing as analogous to a mock-up of a key room for a new hospital. Testing your design research process down to the smallest detail provides insights for the mechanics of the research.
For example, pilot testing can reveal the extent to which staff can be involved, how a study may interrupt the day-to-day flow of existing operations, and how the location of study materials may affect workflow and infection control protocols. Even a relatively simple study can seem highly complex when the details are worked out.
Most important, pilot testing can provide insight in crucial areas beyond fine-tuning details of your protocol. Added advantages to pilot testing include the following:
- It can help filter down to a definitive scope and schedule what seemed initially like a daunting and broad design research project.
- It allows the research team to seek out both small-scale problems and to test larger assumptions to hone in on what matters most.
- It affords a training opportunity for individuals assisting with data collection.
It is easy to underestimate the extent to which pilot testing can reveal and refine problems, even for existing design research hypotheses that may be too narrow or underwhelming. Many times the pilot testing will help researchers develop new, exciting, and previously unconsidered hypotheses to test.
Putting the park back into Parkland
Design research work at Parkland Hospital in Dallas, offers a perfect example of the potential of pilot testing to reveal problems that matter most. Here, pilot testing not only helped prep for the main study but also defined the challenges unique to the organization.
An urban hospital that primarily provides care for the underserved communities of Dallas, Parkland Hospital is in the process of creating a new $1.27 billion, 1.97-million-square-foot facility. The goal for the new hospital is, aptly, to “put the park back into Parkland,” by creating an urban oasis that encourages healing with calming spaces and soothing environments.
Parkland’s presence at the heart of the Dallas community will affect healing and well-being within its walls and the urban fabric by reinforcing accessibility to care and sustainable communities. Figure 1 illustrates one of several healing landscapes that connect the existing and new Parkland Hospital to a new transit development and the Dallas urban fabric. The site will be accessible through a link with the Dallas Area Rapid Transit (DART) light rail.
Like the design of the new hospital, study development challenges were ambitious:
- Measure and explore links between patient-centered care and the patient’s journey (i.e., continuum of care) in and outside the hospital, using experiences at Parkland’s existing facility as the baseline.
- Measure and assess relationships among patient, patient visitor, and staff experiences to understand the FF&E needs of patients along their journeys.
- Use results from the i
nvestigation to inform the new Parkland Hospital’s FF&E selection and planning.
Where to begin? An initial screening of patient satisfaction scores from the hospital’s departments and an extensive on-site assessment of these departments narrowed the study’s scope to two key areas along the patient’s continuum of care: inpatient units and imaging.
Though we did our homework, pilot testing would prove useful. There was little precedent for research of these topics within the design research field. Consequently, the problems were ill-formed for the field and the organization.
MRI anxiety and compliance
The first example relates to our research of waiting areas for magnetic resonance imaging (MRI) procedures. Initial literature reviews and brainstorming led our team to assume that the most anxiety-producing window of time for patients undergoing an MRI would be the period just prior to the start of the procedure. Logically, the appropriate area of focus in this case would have been on creation of a design to minimize anxiety in reception and waiting areas.
The eye-opening results from two-day pilot testing, however, revealed implications for design interventions that spanned the entirety of a patient’s MRI journey. What we learned was that patients at Parkland experienced the most anxiety not prior to the test but in the window of time following the test.
This shifted the design focus to the changing area used by patients after the test concluded, prompting our teams to explore a “decompression and patient education zone” for the main study. These decompression zones are like those proposed for counseling settings and used by survivors of trauma who experience memorials (Watkins & Anthony, 2007; Watkins, 2008).
Future tests will focus on high-impact and affordable art and furnishing interventions throughout the entire MRI experience. These interventions are argued to lower anxiety both before and after the procedure. The lower anxiety may lead to reductions in patient movement and sedative use during the actual MRI procedure. This, in turn, may improve post-procedural patient compliance and attendance for future MRIs (as measured by a subsequent mail-in questionnaire).
Musical chairs (on inpatient units)
Pilot testing on three inpatient units suggested some unanticipated relationships between operational efficiencies and family-centered care. Before the pilot testing, there was a lot of interest in learning more about where and how long patients’ visitors were waiting on the unit.
The concern was that excessive waiting in waiting rooms, corridors, and elevator banks was detracting from family-centered and patient-centered care. How could unit layout and furnishings prevent this? And which solution would matter more?
During the pilot testing, our team observed that patients with large families were moving chairs from the waiting areas into patient rooms. This created a challenge for the nurses who had to dedicate time to clearing rooms of chairs following visits.
The chairs were also creating clutter inside inpatient rooms, which the statistical results suggested may contribute to patient slips and falls. This game of musical chairs on an inpatient unit is hardly fun. It strains staff and may well contribute to patient safety risks.
Enter the wicked problem. Statistics also suggested that when patients were visited by more family members and these family members assisted nurses and patients by managing several smaller tasks (e.g., assisting with patient bathroom transfers; opening window blinds), patients were more satisfied with their care, felt more in control of their surroundings, and made fewer demands on staff.
The musical chairs were contributing to inefficiencies but perhaps indirectly contributing to increased satisfaction.
The problem to explore next in the main study was not how to limit chairs in patient rooms. Rather, we needed to seek ways to measure the impact of alternative seating arrangements within and outside the inpatient rooms. This might include examining the value of a seated respite area separate from the waiting areas and patient rooms and/or fold-out chairs inside patient rooms. Figure 2 illustrates the universal patient room deployed at Parkland.
An ounce of prevention
If you do engage in pilot testing, be prepared to allocate time, staffing, and budget accordingly. Ben Franklin is famously quoted for saying, “An ounce of prevention is worth a pound of cure.” His axiom holds for pilot testing. Pilot testing is a small fraction of the time and cost of a full study. If you invest fully in your pilot testing, you will avoid several headaches and unnecessary costs during your main study.
After pilot testing, be sure to account for the time needed to clean data, perform the data analyses, revise the research methods, and revisit your initial hypotheses. It is a good idea to let the data inspire you to focus on problems you had not originally intended to study within the main study and to clarify why you received a few unanticipated results.
Depending on your sample size, you may be limited to basic descriptive statistics for your analyses. Pay particular attention to participants’ open-ended responses as they will help explain surprising results and refine hypotheses.
To dig deeper, schedule reviews of the findings with the client team. Remain realistic and open-minded during these reviews: clearly convey to the client that you are sharing results from pilot testing to inspire ideas for the main study. The results are not conclusive.
Do not lose sight of the many journeys ahead—including the transition to the main study. The main study should come soon after the pilot testing to keep information fresh in the minds of research assistants and to minimize re-training based on the changes made to the study procedure. HCD
The authors would like to thank the staff of Parkland Hospital who took part in the pilot testing. Also, the authors would like to thank contributing team members and colleagues, including Robin Bajema, Kathy Harper, Gay Chabot, Elizabeth Jones, Sheeba Kuriakose, Bilu Mohid, Leslie Echols, Jodi Donovan, and Chris Korsh.
Nicholas Watkins, PhD, is HOK’s Director of Research & Innovation. The authors would like to thank the clinicians, patients, and visitors who have taken part in the research and findings shared in this article. Correspondence concerning this article should be addressed to Nicholas Watkins, HOK, 1065 Avenue of the Americas, 6th Floor, New York, NY 10018. He can also be reached by phone at 646.385.7624, by fax at 212.633.1163, or by email at firstname.lastname@example.org.
Hamilton, K. (2008). Wicked problems, black swans, and healthcare. Health Environments Research & Design Journal, 2(1):44-47.
Pena, W.M., & Parshall, S.A. (2001). Problem seeking: An architectural programming primer (4th Edition). New York, NY: John Wiley & Sons, Inc.
Rittel H. W. J., & Webber M. M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4, 155-169.
Watkins, N. (2007). Connecting research and design in counseling settings. Research*Design Connections, 3.
Watkins, N. (2007). Mirror mirror on the wall: Ambiguous place attachment and the Vietnam Veterans Memorial. In J. Bissell (Ed.) Proceedings of the Thirty-Eighth Annual Conference of the Environmental Design Research Association. Oklahoma City, OK: EDRA.