Process improvements boost patient satisfaction and quality 
at Stanford University Hospital

by Sharon Flynn Hollander, Laurel Gunderson & Jody Mechanic

Like many institutions, we have initiated many quality improvement efforts in the past decade-some more effective and long-lasting than others. Those that were good sparked dramatic service improvements in some parts of the hospital. Others were little more than "big talk" campaigns that roused a lot of emotion among employees, but really did very little to produce the changes we needed. What our older quality efforts have had in common was a lack of breadth and continuity. One program might develop an idea that would take care of a problem in one specific area, but it didn't fit in anywhere else. We didn't know how to integrate or coordinate these efforts.

To solve this problem we decided to use a process-management approach to coordinate all our quality initiatives around the goal of moving the organization toward "patient-centered care." Significantly, although Stanford University Hospital has cut $50 million from the budget over the past four years, our patient satisfaction ratings have never been higher. We attribute this accomplishment directly to the TQM and process-management efforts, and specifically to four aspects that changed the way we previously thought of quality management. This process:

-Challenged us to think through the identity of the hospital's "customers," both internal and external.
-Increased the amount of feedback to caregivers on concerns expressed by our patients.
-Forced us to use quantitative measures to control and evaluate the effectiveness of routine sub-processes.
-Kept us accountable and focused on our mission with regularly scheduled review sessions.

A "walk through"
With support from top management, a Patient-Centered Care Team (PCCT) brought together managers from all areas of the hospital. Under the direction of our consultants, Shaw Resources, the team decided to limit its work initially to those processes that directly involve patient care, eliminating such behind-the-scenes functions as data processing. We also narrowed our focus to inpatient hospital care; outpatient processes will be tackled at a later date.

The Patient-Centered Care Team then did something that we had never considered before-we "walked through" a typical hospital experience, from admission to discharge, from the patient's point of view. We supplemented our biweekly meetings with outside reading assignments and information gathered in videotaped focus groups with former patients. What our quality efforts had in common was a lack of breadth and continuity. It was a startling experience. The Customer-Supplier Process Model (see chart, page 3) used by Shaw Resources as a template for our work forced us to look at our organization from a totally new perspective. We began analyzing processes through the eyes of the patient, rather than from our own experience or knowledge of what was efficient or cost-effective.

The next step was to identify specific areas to address through separate "process management teams" of the Patient-Centered Care Team. We chose to concentrate in this first phase on the following processes and we launched one team per month until all seven were up and running:

• Admit Patient
• Provide Laboratory Support
• Provide Hotel Services
• Provide Patient/Family Support
• Manage Patient Care
• Provide Diagnostic & Treatment Support
• Management Complaints

A "process owner" was identified for each team (based on who had the most resources and was most accountable for that process) and that individual selected appropriate people to serve. Team members set the agenda and frequency of meetings.

After a team has been in operation for a while, members report back on their work and results to the Patient-Centered Care Team at six-month intervals. The accountability of these "report-back" sessions is integral to the forward movement of each team. In addition, these report sessions have offered an opportunity for collaborative problem solving by top administrators. Any early fears we had about team members becoming too focused on boxes and circles of a process flow-chart were dismantled when the report-back sessions started. The Patient-Centered Care Team keeps its focus on the big picture and the individual teams concentrate on the details of measurement, change, and improvement.

All of the teams have experienced "breakthrough" insights in identifying customers and customer expectations, and using measurements that quantify change. Described here are examples from the Manage Patient Care Team and the Provide Diagnostic & Treatment Support Team.

Understanding patient expectations
"Patients have high expectations when they come to Stanford University Hospital," comments Joann Zimmerman, assistant director of nursing for medical/surgical. "They expect expert care and take for granted the technology or experimental drugs that might save their lives. But if their meal is left on a table just outside their reach from the bed, or if it takes ten minutes before someone responds to their call bell, they are highly critical. From our point of view, we were providing quality care. But we quickly learned that patients have different expectations and definitions of 'quality' than we do as healthcare providers."

The Manage Patient Care Team is different from all others because its members are homogenous-all are RNs with similar education, professional experience, job responsibilities, problems, and the same manager, Zimmerman, who is also the "process owner" of this team. In many ways, this homogeneity allowed the group to coalesce and move into action quickly.

The Manage Patient Care Team asked nurse managers to proactively collect patient complaints based on defined quality measurements. "This was a difficult transition for many of us," Zimmerman says. "No one wants to hear a complaint and it took a great deal of effort to convince units that there would be no shame attached to having a high number of complaints." In a two-month period, managers of ten general medical-surgical units gathered 178 documented complaints. "We didn't expect that patients would be dissatisfied with nursing care. We were Wrong!" "There were more complaints than we anticipated, and we didn't expect that patients would be dissatisfied with nursing care," Zimmerman says. "We were wrong. The number one complaint was related to nursing care, followed by an uncaring attitude from caregivers, call light response time, and response time to requests for pain medication."

Responding to complaints
In addition to collecting patient complaints, the members selected as a starting point two processes that are very important to the patients-pain medication delivery time and call button response time.

The team discovered that, typically, patients were being supplied with pain medication within 4.5 minutes of their request, but that this respectable number was averaged from a wide range of response times. More study showed that the delay almost universally was caused when no pain medication appeared on orders after surgery. The nurse had to locate the physician and obtain a pain medication protocol before any drug could be obtained or administered. This could be a time-consuming process.

As a result of these findings, doctors' orders on all patients admitted to general medical-surgical units are now scanned when the patient arrives to see if a pain medication protocol is indicated. If none is noted, the physician is contacted at this point-before the patient is in pain and requesting medication.

Reducing call button response time is a lengthier, more involved challenge. According to our previous research, we knew we answered 95 percent of all call lights within one minute. What we didn't realize was that "answer" had a different definition to us than it did to the patient. We were pleased to be able to say that within 60 seconds, a patient pushing a call light could expect to hear a clerk respond with a reassuring "The nurse will be right with you." But from the patient's point of view, the call light was not "answered" until someone appeared in the room, which could be as much as ten to 15 minutes later.

The team decided that the best way to adapt this process to fit the patients' expectations is to train clerks to categorize call lights (elimination, pain med, comfort, etc.) and then give patients a more realistic time estimate of when their need or request will be met. For example, a patient with a low-priority request in the comfort category will be told "The nurse will be there in five minutes" so that he or she won't be expecting someone to pop in the door within seconds. Soon, the nurses' beepers will be coded so that they know immediately what patients' needs are by category and, therefore, how quickly they need to respond.

"We've always measured our quality by the standards of the healthcare professionals, not by the measures of patients," adds Zimmerman. "It was the caregiver's agenda, not the patient's agenda. Through process management, we got a different viewpoint on what we have taken for granted." Nurse managers now make rounds three to five times per week to interview patients about the quality of care. Constant and consistent communication is nipping off minor annoyances before they have time to bud into major problems, and the patient "inputs" become valued information for the Manage Patient Care Team to study.

Unexpected results
Meanwhile, members of the Provide Diagnostic & Treatment Support Team were getting acquainted. This group brings together managers of several nursing units and departments that provide diagnostic and treatment services-the cath-angio lab, the cardiology lab, dialysis, endoscopy, nuclear medicine, the pharmacy, the pulmonary/blood gas lab, the neuro lab, radiation oncology, radiology, rehabilitation services, and respiratory therapy.

"Many of these people didn't know each other and we had to spend a great deal of time identifying the common threads in our respective areas," says Judy Lanigan, process owner of the Provide Diagnostic & Treatment Support Team and assistant director of nursing for ambulatory/psychiatry. "Once we used the template (see chart, below) to create flow-charts of the processes in each department, we could see the similarities. We were all surprised at the parallels and at how interdependent we are."

I think one reason that our previous quality efforts have not been the solutions we hoped for is that they tried to solve problems within the confines of one department," she adds. "Once we saw the flow-charts it was clear to us that this would not work. Patients move from one area to another and they don't care that they may have crossed an organizational boundary. They judge their total experience in the hospital. This was even more dramatic when we took the additional step of completing a flow-chart from the patient's perspective."

The team has identified numerous areas for further improvement and measurement. Some of these are: (1) patient transport; (2) clarification of physicians' orders; and (3) turnaround time of reports to physicians.

Already, minor process improvements are showing results. Reminder calls to outpatients about their appointments in the cardiology lab have limited late arrivals to one since March of 1993. Routine surveys of patients before and after respiratory therapy treatments sparked a new policy of re-orientation for professional staff that has been gone on leave. "Patients can spot when a caregiver is not at ease working with new equipment," Lanigan explains. "This discomfort is jarring to the patient's expectations of what the treatment experience should be like. By interviewing respiratory therapy patients we discovered this had occurred when someone had been on leave for a while and new equipment had arrived during that period. The respiratory therapy department subsequently changed its process to 'check off' returning staff on familiarity with new equipment or procedures."

Simultaneous with the patient-centered care efforts at Stanford was a move to decentralize support services to the point of care. This major organizational change, which was initiated earlier as part of our cost-reduction programs, enhanced the work of the process management teams and helped boost our patient satisfaction ratings. For example, studies showed that patient treatments were often delayed while supplies were delivered to the unit. Now the hospital has a mini-supply room on each floor, controlled by a state-of-the-art computerized inventory and tracking system. We can now guarantee that supplies will be available within five minutes.

Support service assistants
Another structural shift was the merger of housekeeping and transportation personnel and the decentralization of these services. Stanford Hospital now has "support service assistants" (SSAs) assigned to each unit to provide housekeeping and transport services. The same employee cleans a patient's room, takes him or her to X-ray, and delivers meals. The contributions to patient-centered care have been enormous:

• It has reduced the number of faces a patient sees in the course of a day. (Too many interactions from too many different people-a high probability at a teaching hospital-are a major source of patient complaints.)

• It personalizes the service patients do receive. It is not "housekeeping" that is making their rooms spotless - it's Rose, or Jose, or Mike.

• It has given the SSAs more feelings of connection and ownership for doing a good job. They see immediately how their jobs affect patient care. Hours of training, coaching, and preaching about the importance of prompt service cannot equal the impact of seeing for themselves how a patient feels to sit and wait for a wheelchair.

• It has saved money by eliminating a second layer of supervision. SSAs report to the nurse manager in the unit.

The introduction of SSAs and decentralization has not been without challenges. Nurse managers were unaccustomed to hiring and supervising this level of personnel and needed additional training. Previously, these support employees were distanced from the patients and had little direct contact with them. Now they deliver their services bedside and get to know patients by name and face-and they care about the outcomes.

"Before I never thought about why a room was empty. My job was to clean. Now I know that Room 304 is empty because Mrs. Smith has died." SSAs now receive training and counseling on death and dying.

Next steps
As the process management teams continue their work, the Patient-Centered Care Team is wrestling with how to attack our next set of challenges. We have identified three "next steps":

• Educate staff on the process management methodology, especially the value of complaints in improving patient-centered care. Currently, process management teams involve fewer than 100 of the approximately 4,700 employees of the hospital. Training needs to take a variety of forms and levels of depth to accommodate the diverse range of educational levels, cultural backgrounds, and licensed and non-licensed positions.
• Continue to effectively manage the overwhelming amount of data available to the process management teams. Unlike some organizations, a hospital already has a wealth of information recorded. The challenge is to analyze it productively and feed it back to employees in a way that is meaningful and produces change and improvement.
• Devise reward systems that are valued by our people. In the '90s a reward for a job well done no longer necessarily needs to be money. Many employees prefer a flexible schedule or an extra day off to a bonus in the next paycheck.
• Dovetail patient-centered care activities with process management.

There are no quick fixes in a healthcare organization-our problems are too complex, our issues too critical to the well being of our patients. But process management has pointed us all in the same direction with the same goal in mind. We are seeing real change and real results that will be long-lasting because they are altering the core infrastructure of our organization.

"Outstanding performance and patient-centered care are not based on how many resources you have but on the quality of your leadership," says Zimmerman. Process management has taken our leadership skills up a notch and everyone connected with the hospital-physicians, staff, and most of all, patients-is benefiting.

Sharon Flynn Hollander is associate hospital director for patient care services/director of nursing and chair of the Patient-Centered Care Team. Laurel Gunderson is assistant director of nursing for critical care and acting chair of the Patient-Centered Care Team while Flynn Hollander was on sabbatical. Jody Mechanic is a research associate and a facilitator for the Patient-Centered Care Team. For more information, contact Flynn Hollander at (415) 723-5181.

Reprinted in Cooperation with: Strategies for Healthcare Excellence (ISSN 1058- 7829) is published monthly by COR Healthcare Resources. Editor: Susan J. Anthony 415/824-8007. Managing Editor: Janet Glasheen. Publisher: Dean H. Anderson 805/564-2177. Send subscription inquiries to Strategies for Healthcare Excellence, P.O. Box 40959, Santa Barbara, CA 93140-0959; 805/564-2177; fax 805/ 564-2146,

Volume 7 January 1994 Number 1