The past 15 years have seen an explosion in the emergency medicine literature about patient satisfaction. From the smaller rural ED to the inner city Level I trauma center, practitioners and administrators have had to come to terms with the concept of customer satisfaction in this clinical arena. Attitudes like, “They get great medical care, what more do they (the patients) want?” are no longer possible in today's competitive world of health care.
What do all of these studies tell the health care provider in the trenches and on the front line? Are there themes to guide the emergency practitioner trying to improve this aspect of his practice? Are there small changes easily and inexpensively implemented, which might enhance the overall experience of the emergency department patient? Are there any quick fixes or simple innovations to enhance patient satisfaction? What are the practical applications of this growing body of research?
Patient Satisfaction in Context
The concept of customer satisfaction (generally speaking, how well a customer's expectations are met) and more specifically patient satisfaction have to be placed in the context of overall quality improvement. Louis Graff, MD, and others describe the triad of quality that includes clinical quality, cost efficiency, and service quality. (Acad Emerg Med 2002;9[II]:1091.) All three elements must be evident to have a robust and successful quality improvement program, and it is this service quality that is the essence of the study of patient satisfaction in the ED.
Though attempts have been made to follow the course of other service industries, emergency medicine has unique factors to the customer service model. First, patients can report great patient satisfaction even though poor clinical care was rendered and vice versa. Patients are not necessarily reliable assessors of clinical quality. A second challenge involves the time frame for satisfaction measures. Patients frequently view their health care in terms of illness episodes. The heart patient will recall the ED, the cardiac catheterization lab, the operating room, and the CCU as a continuum of health care without clearly distinguishing the differing elements or venues of care. (JAMA 1997;278:1608.) Measuring patient satisfaction has proved a formidable task. While an eating establishment may count patrons or profits to evaluate customer satisfaction, in health care there are no such easy measures.
A working definition of patient satisfaction includes overall satisfaction (usually by survey), likelihood to recommend, and willingness to return. Indeed, these three overall measures as practical indicators of patient satisfaction abound in the literature. Early patient satisfaction surveys were seldom validated instruments, and had built-in bias with very low response rates. The past 15 years have seen improvement in this area with the development of survey instruments specifically for emergency department patients. (Ann Emerg Med 2001;38:527.)
Other quantifiable measures have been developed to clarify the elusive patient satisfaction picture including door-to-doctor times (J Health Care Mark 1993;14:26), which correlate well with satisfaction or the ultimate indicator of patient dissatisfaction, leaving without being seen. (J Emerg Med 1997;15:397.) Researchers are regularly using CQI tools to measure the response to particular process improvement changes all in an effort to enhance and improve the patient experience in the ED. (J Emerg Med 1997;15:889.)
Why Pursue Satisfaction?
From a clinical perspective, patient satisfaction makes sense. Patients who are satisfied with their care are more likely to be compliant and respond better to their treatment. (Psychosom Med 1995:57:234; Br J Clin Psychol 1982;21:241.) From a risk management perspective, patient satisfaction also makes sense. Caregivers who participate in a system of good customer satisfaction experience fewer malpractice suits than their counterparts. (The Quality Connection in Health Care: Integrating Patient Satisfaction and Risk Management. San Francisco: Jossey-Bass, 1991; Lancet 1994;343:1609.) Additionally, there is a connection between patient satisfaction and staff satisfaction. Results of Press Ganey surveys in which patient satisfaction and staff satisfaction were measured show a clear relationship between the two, and at one hospital while customer satisfaction increased, employee turnover decreased by 57 percent! What is good for the patients does in fact appear to be good for the caregivers! (Patient Satisfaction: Defining, Measuring and Improving the Experience of Care. Health Administration Press; 2002.)
Finally and of primary importance to an institution's operating executives, good patient satisfaction translates into fiscal improvement. “An ER visit is a significant encounter between patient and hospital, and one that affects ‘repurchase’ decisions for future healthcare,” notes J.V. Mack in an analysis of emergency department choices among Medicare patients. (J Amb Care Mark 1995;6:45.) Despite the elderly being disproportionate users of health care, surprisingly about half have no regular physician and so choose emergency department care. Ninety-seven percent in this study had a choice of ED, and more than half had been referred by the advice of others. This verbal networking and utilization of services by the elderly has huge implications in terms of the focus of patient satisfaction efforts. Elder services including home health aides, equipment (walkers and bedside commodes), and geriatric consultants should be available to ED patients to improve services given to seniors and to enhance their experience of care in the ED.
Older patients are more likely to express patient satisfaction than those who are younger (Am J Emerg Med 2000;18:394), and this correlates with other data cited by Press Ganey. Additionally, young and black patients are less satisfied with care (Ann Emerg Med 2000;35:426), and this is consistent with data across from outpatient and in-hospital settings. In one study, patients who were insured were more likely to recommend the ED to others, while the uninsured and indigent were less likely to do so. In another study, this correlation was not seen. (Acad Emerg Med 2004;11:51.) Another variable which correlates with good patient satisfaction is acuity: Higher acuity patients are more satisfied, and patients receiving multiple treatments also have higher patient satisfaction scores. (Acad Emerg Med 2004;11:162.)
Characteristics which did not influence patient satisfaction in the ED included gender (though other health surveys have shown men to be more difficult to please), weekday versus weekend, time of day, and disposition. Patient volume did not affect satisfaction (Hosp Health Serv Admin 1996;41:515), though typically teaching hospitals and trauma centers perform score lower on patient satisfaction surveys, perhaps because of longer wait times. (Ann Emerg Med 2003;41:35.) Other weak predictors of patient satisfaction included satisfaction with registration process, mode of arrival, admission status. (Acad Emerg Med 2004;11:51.) A large review also showed weak correlation with marital status, diagnosis, daily census, satisfaction with tests, presence of chronic illness, number of previous visits, and type of treatment. (J Emerg Med 2004;26:13.)
A growing body of research has demonstrated repeated themes associated with high ED patient satisfaction including the empathy and attitude of health care staff (bedside manner), acceptable waiting times (specifically perceived times versus actual wait times), technical issues (both technical skills and available technology), pain management, and information dispensation. (Acad Emerg Med 2004;11:51; Soc Sci Med 1988;27:935; Acad Emerg Med 2002;9:1379; Acad Emerg Med 2003;10:261; Ann Emerg Med 1993;22:586.)
It is becoming increasingly apparent that the art of caring for patients correlates with satisfaction. An uncaring attitude is cited in seven percent to 13 percent of emergency department complaints. (Ann Emerg Med 1991;20:1014; Ann Emerg Med 1987;16:857.) Caring physicians and nurses are variables that show up repeatedly in satisfaction data, and at times even override waiting times as predictors of patient satisfaction. (Acad Emerg Med 2004;11:51.) Put simply, even speed cannot compensate for rudeness, disrespect, or an uncaring attitude.
Some institutions have embarked on customer service training programs to improve interactions between health care providers and patients with great results. (J Healthcare Management 1998;43:421.) The most successful and sustained programs involve an institutional commitment to the principles of customer satisfaction and service. (Leadership for Great Customer Service, Health Administration Press. Chicago. 2004.) Another new area showing promise in improving the communication and interpersonal interaction of the encounter is scripting, in which staff are provided with positive dialogue for specific situations in the ED. (Emerg Med Clin North Am 2004;22:87.) These scripts can be helpful in many situations, including during registration, during phone call requests, and when dealing with distressed family members, complaints, drug seekers, patients enduring long waits, and angry physicians.
Other service variables which correlate with patient satisfaction and the perception of caring include organized staff, staff introducing self, and satisfactory discharge instruction. (Ann Emerg Med 1993;22:586.) Overall, when physicians display more affect, give more information, and encourage dialogue with their patients, the result is higher satisfaction. (Soc Sci Med 1991;32:541.)
Lastly, call-back systems also are being used as a patient satisfaction intervention. Patients who have left before treatment is complete are called to determine why they left and to check on their clinical course. Patients who may have had a poor ED experience due to delays or unmet expectations can be called back to the ED. This is a chance to salvage the ED encounter, it is an effective risk management tool, moving the patient satisfaction program forward. (Emerg Med Clin North Am 2004;22:87.)
Language barriers also pose a problem in terms of patient satisfaction scores. (J Gen Intern Med 1999;14:82.) By establishing an interpreter program or training physicians directly in language, some of this lost patient satisfaction ground can be made up. (Med Care 1998;36:1461; Arch Pediatr Adolesc Med 2002;156:693.)
A number of physician attributes and behaviors have been studied and deserve mention. Does the gender of the physician affect patient satisfaction? The simple answer is yes and no. Female physicians are positively associated with women's satisfaction in the ED. (J Gen Intern Med 2001;16:218.) This may be factored into physician assignments, especially where more invasive examinations and procedures will be required on female patients, particularly adolescents or those whose religious background makes interacting with male physicians difficult.
Although a physician's appearance is important to patients, they were more tolerant of casual dress by emergency physicians than other physicians. Half of patients prefer their emergency physicians in white lab coats, and 18 percent disliked scrubs. Both physicians and patients did not like jeans or sandals, but neither did they like frills such as ribbons and ruffles, excessive jewelry, and long fingernails. (Ann Emerg Med 1989;18:145.)
In another study, patients had the highest confidence in images of physicians dressed in scrubs with a white lab coat, and the least confidence in images of physicians dressed casually. (Am J Obstet Gynecol 2002;190:1484.)
While patients report incorrectly 30 percent of the time whether their physician was wearing a neck tie, they correlate the tie with a positive impression of the physician. Although this did not affect their impressions of the care they received, they preferred the appearance of the physician in a tie. (J Emerg Med 1998;16:541.) Perhaps less objectionable to some physicians than the neck tie is the physician business card. This simple and inexpensive item has been shown to correlate with enhanced patient satisfaction. (Am J Emerg Med 1994;12:125.) Published data also show how patients prefer to be addressed by their physicians in the ED. While most physicians addressed patients by surname, almost half of patients preferred being called by their first names. (Ann Emerg Med 1989;18:145.) Another study, however, concluded that one cannot reliably predict how the patient wants to be addressed. (J Am Board Fam Prac 1992;5:517.)
Acceptable Waiting Times
Patients presenting to the ED frequently overestimate the urgency of their need for health care. They typically do not understand the triage system and interpret patient flow in the department as somehow being unfair. (Hosp Health Serv Admin 1996;41:515.) These factors set up the patient to perceive his wait as too long. Bursch et al has shown that perceived waiting time is the most important variable contributing to patient satisfaction as opposed to actual wait time. (Ann Emerg Med 1993;22:586.) This finding has been replicated by others. (Ann Emerg Med 1996;28:657.)
Similarly, higher patient satisfaction has been shown to correlate with waits which are less than expected. (Acad Emerg Med 2002;9:15.) The authors suggest that a focus on appropriate expectations regarding wait times should have a positive effect on patient satisfaction as well.
By moving patients quickly to a care area and having the physician evaluate them in a timely fashion (under 30 minutes is the accepted service quality goal), patients perceive that wait times are acceptable. (Ann Emerg Med 1993;22:586; Health Care Manager 2002;21:46.) When the time interval from triage to physician evaluation increases, the rate of patients leaving without being seen goes up linearly. (Ann Emerg Med 1999;34:3.) Innovations such as bedside registration, which can be implemented with a registration clerk and a clipboard where bedside computers are lacking, and medication treatment teams improve satisfaction because they effectively get the physician to the bedside sooner, and this is the most critical time interval from the patient's point of view. (Am J Emerg Med 2002;20:267.) Also, by tracking door-to-doctor times and overall length of stay and sharing data with practitioners through a comprehensive CQI program, time intervals can be reduced. (Ann Emerg Med 2001:38:533.)
Understanding the utilization patterns in the ED can help each facility meet the demands of its patients. There are predictable patterns of arrival to emergency departments regardless of size, location, and type of hospital. Almost 50 percent of visits arrive during the day, although they often overflow into the 35 percent seen on the evening shift or the 15 percent seen on the night shift. Nationwide 16 percent are admitted to the hospital, and the volume and acuity are slightly higher on the weekend. Acuity also is higher on the night shift. Staffing accordingly is a key to managing wait times. (Health Care Manager 2002;21:46.)
Occupied time feels shorter than unoccupied time, and televisions, magazines, and videos are promising diversions. Though bedside televisions did not statistically improve patient satisfaction in one study (Am J Emerg Med 2000; 18:119), the numbers were quite small and the average length of stay for the study population was 270 minutes, perhaps too long even for a TV watcher to be satisfied. On the other hand, a videotape with information regarding the ED operations resulted in a more favorable perception of delays and overall perceptions of the ED experience. Many office practices are now providing phones, beepers, and computer jacks for laptops in their waiting rooms. Perhaps some of these innovations would be useful in patient rooms where prolonged stays are anticipated.
Lonely waits also feel longer than time spent with people. Though visitors are often viewed warily by staff in the ED, family members and friends need to be with loved ones in the ED for the comfort and satisfaction of the patients.
Perceived technical skill correlates well with positive perceptions of staff (Hosp Health Serv Admin 1996;41:515), and perceived good technical skills have been found in two studies to be the best predictor of global satisfaction (Health Care Manager Rev 1995;20:7; J Emerg Med 1996;14:679), even more correlative than bedside manner. This may be a troublesome realization for emergency department practitioners at our nation's teaching hospitals. If technical skill is highly correlated with patient satisfaction and an enhanced patient experience, ought we to have the least experienced among us learning their technical skills on ED patients?
The data on this topic are somewhat mixed. One study found that ED patients would allow medical students to perform very simple noninvasive procedures (IVs, splints, and suturing). (Acad Emerg Med 2002;9:495.) On the other hand, Graber et al concluded (Acad Emerg Med 2002;10:1329) that patients are reluctant to have medical students perform a first procedure on them, and many would not allow medical students to perform some procedures at all. This presents dilemmas in medical education and informed consent. It may be an area where scripting could help, and guidelines for effectively enrolling patients in such learning encounters may need to be set for teaching hospitals.
Another technical area which has had its effect on ED patient satisfaction is ultrasound in the ED. In one study, patients receiving ED ultrasonography versus radiology ultrasonography viewed the physician as having a more caring attitude, rated the physician as having better skills and ability, and rated the overall satisfaction with the ED visit higher! (Am J Emerg Med 1999;17:642.) The scores improved as the physician's skills progressed. There would appear to be no downside in terms of patient satisfaction with the introduction of bedside ultrasonography in the emergency department.
Pain is one of the major symptoms that cause patients to seek medical care on an emergent basis, and the complexities of pain management in the ED are only beginning to be unraveled. (Acad Emerg Med 1998;5:851; ) Ann Emerg Med 1996;27:438.) Though there is a general correlation between pain relief and satisfaction, cultural factors, the intensity of the original pain experience, and differing pain scales all work to confound an understanding of this area of medical research. Though this area is worthy of a literature review in and of itself, a few themes relating to pain management and satisfaction in the ED are emerging and are worth noting.
Pain management in children (Acad Emerg Med. 2002;9:1379) correlates highly with patient satisfaction, and ought to be a focus for departments seeing significant pediatric volumes. Patients appear to have preferences and expectations regarding pain management in the ED, and these could be easily met. (J Emerg Med 2004;26:7.) In general, patients under 54 years of age prefer oral analgesia, while the more senior the citizen, the more likely he will prefer IV analgesia, though oral medication is still preferred. The more severe the pain intensity reported, the more likely the patient will prefer parenteral medication. Another study found a slight preference for parenteral analgesics over oral, but the study population was exclusively orthopedic fractures in older patients. (Am J Emerg Med 2000:18:376.) In both studies, intramuscular analgesia was the least preferred route of administration for analgesia in the emergency department. Earlier studies suggested differences in management of pain in the ED with undertreatment of ethnic minorities and women. (JAMA 1993;269:1537; Ann Emerg Med 1996;27:1421.) Timely alleviation of adverse symptoms has also been shown to deter patients from leaving before being seen by a doctor. (Ann Emerg Med 2003;42:3.)
In short, an ED should have a well stocked selection of oral analgesics on hand, and be very liberal with their dispensation. Further, it should be easy to assess in triage which patients may need parenteral medication and therefore expedite IV placement.
Some studies have shown that a lack of explanation has a greater effect on patient satisfaction than perceived wait times (Patient Satisfaction: Defining, Measuring and Improving the Experience of Care. Health Administration Press; 2002), and that staff overestimate the amount of information they give patients. (Through the Patient's Eyes, San Francisco. Jossey-Bass. 1993.) This has given rise to a new position in the emergency department: the patient advocate. The patient advocate can be an LPN, a social worker, or a volunteer. The patient advocate makes frequent contact with the patient and family members, keeping them informed of delays and progress, and may also be trained to assist with noninvasive comfort measures such as getting blankets, telephones, or ice chips. An additional benefit of having a patient advocate in the department is to free the professional staff for the more pressing technical tasks.
Unexplained and uncertain waits feel longer, and have a negative effect on patients' perceptions of the wait. Frequent updates regarding a patient's progress and delays correlate with patient satisfacti on and satisfactory length of stay. (Am J Emerg Med 2002;20:506; J Emerg Nurs 2004;30:336.) An informational sheet handed to patients had equivocal results based on contradictory studies. (Ann Emerg Med 1993;22:568; Ann Emerg Med 2004;44:378.) Some centers have set service goals for staff to give informational updates at specified time intervals to patients and their families with excellent results. (Am J Emerg Med 2002;20:506; Ann Emerg Med 1996;28:657.) Other informational dispensation techniques are currently being investigated, including videotapes, closed circuit television, and pamphlets, though nothing may exceed the benefit to patient satisfaction of human interaction and verbally delivered updates.
Other issues which sometimes correlate with patient satisfaction include privacy (especially in triage), cleanliness, and safety. (Acad Emerg Med 2003;10:261; Am J Emerg Med 2000;18:394.) Particularly at high-volume EDs that care for prisoners, psychiatric patients, and the homeless, patients who feel threatened or unsafe are more likely to report being generally unsatisfied. Highly visible security guards and police officers can have a positive effect on the patient encounter.
Patients frequently complain about noise pollution in the department. (Emerg Med Clin N Am 1992;10:551.) The term “acoustic isolation” has been coined to indicate the placement of a noisy patient in a soundproof room. This benefits other patients, limits excessive stimulation of the out-of-control patient, and benefits the staff as well. It is becoming a standard of care in newer departments.
Patients also frequently complain about misdiagnosis or mismanagement in the ED, though such complaints are more frequently communication failures than true medical mistakes. Patients may not feel satisfied with care when certain diagnostics are not ordered, particularly skull films and ankle films. Some authors are suggesting somewhat controversially ordering “reassurance diagnostics or treatments.” With flexibility and grace, if the emergency physician can provide tests or treatments of negligible risk, the reassurance factor can enhance the patient experience. (Emerg Med Clin N Am 1992;10:551.) Variables such as ease of parking, simplicity of financial matters, and food, which have been often predictive of customer satisfaction in other settings, were by and large trivial in ED patient satisfaction data over the past 15 years.
The body of literature on patient satisfaction in the emergency department has grown at a breakneck pace over the past 15 years. With it has come an understanding of patient satisfaction as an entity analogous to but distinct from customer satisfaction, and its context in the realm of quality improvement is better understood. There is still much work to be done in this area. In particular the patient satisfaction needs of different subpopulations of patients still need to be defined. In particular, age-specific and disease-specific data are notably lacking. New tools for measuring this ubiquitous commodity known as patient satisfaction are required. With information technology at our fingertips and process improvement concepts at the ready, these research opportunities are waiting to be tackled.
With massive structural and process redesign grand improvements in patient satisfaction can be realized. (Ann Emerg Med 2002;31:169.) There also are gains to be made through small innovations and attitudinal changes. By knowing the demographics and utilization characteristics of its patient population and staffing accordingly, an emergency department can set the stage for success. By focusing on five main areas — perceived wait time, empathy, technical issues, pain management, and information dispensation — and implementing modest changes in infrastructure and operations, an ED can expect to see gains in its patient satisfaction measures. Most of the innovations suggested by the literature cited in this review involve very little capital investment. These innovations involve more changes in culture than in the physical plant.
Where patient satisfaction in the emergency department is concerned, are there any quick fixes and simple innovations? The answer is a resounding “yes,” and they start with the staff rather than the physical plant. The investment in interpersonal assets may yield far more in this setting than previously imagined.
Reasons for Pursuing Patient Satisfaction
▪ Patient compliance
▪ Malpractice and risk management
▪ Staff satisfaction and retention
▪ Fiscal improvement
Improved Perception of Wait Times
▪ Know your utilization data
▪ Perform bedside registration
▪ Track door-to-doctor times
▪ Set patient expectations appropriately
▪ Provide televisions and magazines as diversions
▪ Play instructional videos explaining the ED process
▪ Allow family members at the bedside
▪ Provide phones, beepers, and modem ports