Wright, Greg MSN, RN; Causey, Sherry MSN, RN; Dienemann, Jacqueline PhD, RN; Guiton, Paula BSN, RN; Coleman, Frankie Sue RN; Nussbaum, Marcy MS
Emergency departments (EDs) are frequently viewed as the front door to the hospital, which has peaked administrators’ interest in patient satisfaction scores. Patient satisfaction with care in all areas is important because of the increasingly competitive hospital market.1 Patient satisfaction is also important for its relationship to future utilization of services by patients.2,3 Patient satisfaction with nursing care is an important indicator of quality of care in both inpatient and outpatient settings.2 Nurses practicing in the ED report a feeling of conflict of priorities between increasing patient satisfaction and providing quality of nursing care in balancing expectations and effectiveness.4
Reduced perceived waiting time, provision of engaging activities while waiting, technical skills, personal interactions with staff, feelings of involvement in care, and provision of information have been noted to directly increase ED patient satisfaction scores.5-9 Distance to travel and expected wait times were found to be the primary factors influencing choice of ED.10 Providing information on estimated wait time increased willingness to wait,11 but total wait time was the most important determinant of satisfaction.10,11 One successful strategy to reduce perception of wait time was for an employee to provide periodic personal interaction and clinical information on the ED visit progress.12 One frequently cited goal for EDs is to reduce wait time. Patients are increasingly viewed as consumers who demand timely service. In addition, dissatisfied patients will negatively influence others, also decreasing future revenues.11 Strategies to manage capacity, reduce wait time, and act to meet the psychological needs and expectations of patients during their wait increase satisfaction in EDs.7,8 Patient satisfaction is affected by several factors including affective support, health information, decisional control, and professional/technical competence of nurses and other providers.5 No clear association of patient satisfaction and patient characteristics has been found. Studies of gender13 and age10,14 have yielded varying results. Higher educational attainment raises expectations.14 Language barriers and cultural differences may result in lower patient satisfaction outcomes.15 This study, conducted at 2 sites, investigates the patient perception of targeted interventions designed to decrease dissatisfaction with wait times.
1. Will perceived wait time in the ED treatment area be reduced by the systematic provision of comfort items, information on the visit, and engaging activities?
2. Will systematic offering of comfort measures, information on the ED visit process, and engaging activities be helpful for patients to cope with waiting in the treatment area?
3. Will systematic offering of comfort measures, information on the ED visit process, and engaging activities increase patient satisfaction with nursing care in the ED?
4. What differences are there between urban and suburban ED patient populations in patient expectations of wait time, helpfulness of comfort measures, information on ED visit process and engaging activities, and satisfaction with nursing care?
This controlled quasi-experimental design utilized a convenience sample of patients recruited at 2 EDs. To reduce contamination, usual care participants were studied 1st and then intervention participants. The research team was composed of a doctorally prepared nurse and 3 BSN-prepared nurses who completed institutional review board (IRB) training and participated in all aspects of the study. After IRB approval, 1 nurse was assigned to be the site coordinator for each site. Information was posted in the physician and RN lounges about the study and implementation plans. A 30-minute training class in the protocol was given at each site by the site coordinators and doctorally prepared RN to nurses interested in participating. All participating nurses completed Good Clinical Practice (provided by Carolinas Healthcare System) and IRB training by Collaborative Institutional Training Initiative.20 The protocol included inclusion and exclusion criteria, conducting the study sequentially with phase 1 being the usual care group followed by phase 2 for the intervention group and detailed instructions for recruitment, data collection, and protection of confidentiality. A script was used to recruit patients. Confidentiality was protected by coding and sealing data envelopes that patients and nurses put in a research box that was emptied daily and kept in a secure location. The doctorally prepared nurse collected the data monthly from each site and met briefly with the research nurses to answer questions. Data were stored in a secure location at the university. Before commencing the data collection for the intervention group, a celebration and a repeat of the training session were held with emphasis on details of the intervention and the importance of following the protocol. Posters with information on the typical ED visit process were posted in each treatment room for the intervention period. Inclusion criteria were adult 21 years or older, able to give consent, and able to read English or Spanish. Consent was assumed by completion of the questionnaire. The incentive for participation was a grip-contoured ballpoint pen with the hospital logo.
The 2 community hospitals belonged to different health systems in the same metropolitan area and were similar in size (125 and 117 beds). Both used the same 5-level triage system, with higher levels indicating less urgency. The units were primarily staffed by RNs and a contracting physician group that included physician assistants and nurse practitioners. Both sites assigned RNs geographically by treatment room, and responsibilities included providing assessments, treatments, education, and discharge instructions. The sites were chosen to provide diversity in the overall sample. The EDs differed in size, treatment room specialization, and patient population (Table 1). The nursing research consultant served on the nursing research council at both sites. The urban site saw an average of 200 patients per day, and the treatment rooms are designated as minor or major urgency based on triage. Each treatment room had little décor and no reading material available in the main waiting area. The population served was ethnically/culturally diverse and included 42.1% uninsured patients. An on-site full-time interpreter was available for Spanish-speaking patients.
The suburban site saw an average of 85 patients per day. The rooms were decorated with pastel colors; a glass door, a television, and reading materials were provided in the main waiting area. The population served was primarily white, and 17.9% were uninsured patients.
Those assigned to the intervention received from the RN (1) an offering of a warmed blanket, nonskid socks and lowered lights; (2) an informational brochure on the ED process of a visit and RN rounding for updates every 30 minutes or less; (3) offer of a puzzle and coloring picture and crayons for accompanying children, a reminder that family and friends may accompany them, and they may use books, smart phones, or other items brought from home and television (only provided at the suburban hospital). These interventions were performed from admission to the treatment area until discharged. Those assigned to usual care did not systematically receive comfort items, structured information, or engaging activity items or reminders; however, usual ED care was rendered.
The patient questionnaire included demographic information, expected and perceived wait time, and an investigator-developed measure of utilization and perceptions of the intervention including number and nature of comfort, information, and engaging activities received and their perceived helpfulness to cope with waiting (measured using a Likert scale with 1 = very helpful and 4 = not helpful). Before the study, this was pretested for clarity by review by all members of the research team. Patient satisfaction was measured using the Consumer Emergency Care Satisfaction Scale (CECSS).16 This is a 19-item questionnaire that has 2 subscales for teaching and caring. Items are measured using a Likert scale with 5 = completely agree to 1 = completely disagree. There are 3 inversely worded questions that are not scored. If the respondent rates all items the same, including the inversely worded items, the questionnaire is deemed invalid and deleted from analysis. The content and construct validity for subscales have been confirmed.17 The reliability for the subscales was 0.85 to 0.92 for teaching and 0.87 to 0.88 for caring.17 The CECSS was translated into Spanish using the translation-retranslation method18; subscale validity was confirmed, and reliability remained adequate with Cronbach’s α of .81. Author permission was obtained. Nurse data collected from charts were the same for all participants and included presenting complaint, triage score, age, race, and payer. The information sheet on the study for potential participants was also translated into Spanish using the same methods.
Descriptive statistics were done using SAS version 9 (Cary, North Carolina) for all variables followed by testing for similarity of usual care and intervention groups using χ2. Missing and incomplete data were excluded for each calculation. The chosen level of significance was .05. χ2 Tests for comparing usual care and intervention were calculated; the 2 sites were then compared. The differences in means for caring subscales and overall by control group and intervention group followed by site were calculated and compared using analysis of variance (ANOVA).
The total number of completed, acceptable questionnaires was 573 (Table 2). There were 246 usual care and 327 intervention participants. The 2 samples did not differ significantly. The typical patient was a white woman with private insurance aged 25 to 39 years with a triage level 3 or 4 complaint. Pain not in chest ranked as the no. 1 complaint in both the usual care (36%) and in the intervention group (43.7%).
The mean for expected wait time was 114.5 minutes for usual care and 113.9 minutes for the intervention, approximately a little less than 2 hours. When comparing those whose expectations were met and those who were not, no significant difference (P = .43) by intervention or usual care was found. When examining how close the perceived wait time was to the actual, no significant difference was found (P = .14). Expected versus actual wait time was not significantly different (P = .15).
Findings from the effects of the intervention activities are summarized in Table 3. In looking at comfort items, only small differences were found in usual and intervention care, except for lowering lights (P < .001) and use of comfort items from home (P < .01), both of which increased with the intervention. Patients also reported that the helpfulness of the comfort items increased significantly (P < .001) with the intervention.
Perceptions of nurses providing information increased only 1.5% with the intervention, and there was no difference by providing a poster. A significant difference was found regarding the brochure (P < .001) being handed to the patient in the intervention. Perceived helpfulness of provision of information increased significantly (P < .001) with the intervention.
The provision of puzzles, such as sudoku or word search, and a pencil was significantly different at the P < .001 level. Using items from home (P < .05) and talking to family/friends (P < .01) also changed significantly with the intervention. The suburban site had televisions in treatment rooms, and approximately half of the patients reported using it. The perceived helpfulness of engaging activities increased significantly (P < .001) with the intervention.
Patient Satisfaction With Nursing Care
Patient satisfaction with nursing care overall and for caring and teaching subscales was significant (P < .001) with the intervention. Patients under usual care had good satisfaction. With the intervention, the variance decreased for caring (SD, 7.0-5.9) and overall (SD, 8.9-7.9), and the total level increased significantly. Using the CECSS-recommended cutoffs for caring, 94.7% of the usual care participants were satisfied, increasing to 96.8% of the intervention participants; for teaching, 82.7% of usual care participants were satisfied, increasing to 91.6% of the intervention participants.
The intervention samples at the urban and suburban sites were then compared (Table 4). The suburban site had a significantly (P < .001) higher proportion of white patients and fewer Latino participants (P < .05). This site also had significantly (P < .001) more private pay participants. Suburban participants were also significantly older (P < .001). The suburban triage levels were significantly more urgent (P < .001).
No significant differences in usual care and intervention wait-time expectations were found by site. When breaking down the expected wait time, using only the intervention data, the suburban patients expected a significantly (P = .003) shorter wait (93.3-minute average compared with 114.3 minutes at the urban hospital). Their actual wait times did not differ significantly (125.6 minutes at the urban ED vs 122.6 minutes at the suburban ED, P = .091). Suburban patients were significantly (P < .001) more accurate in their estimates of actual wait time (50.6% vs 36.6%).
Regarding intervention helpfulness, all 3 intervention actions were viewed as less helpful by the suburban participants compared with the urban participants. Comfort helpfulness difference was significant (P = .004). Information and engaging activities were significantly different at the P ≤ .001 level (Table 3). Despite comfort being received more often, for example, pillow 62.5% versus 43.1% by suburban patients, they expressed that they were less helpful in coping with wait time.
Regarding perceptions of satisfaction with nursing care, means were compared using ANOVA, resulting in satisfaction reported as higher at the urban hospital (P ≤ .0001). Breaking down the patient satisfaction of caring and teaching, the same significant difference was found for each (P ≤ .0001). This is also illustrated by the recommended cutoff scores; urban participants were 99.0% satisfied with caring, and 96.9% were satisfied with teaching, but the suburban participants were 93.5% satisfied with caring and 82.5% satisfied with teaching.
By definition, coming to an ED means that patients are uncomfortable and have a health complaint they view as needing immediate attention. Perhaps the most interesting findings were that suburban patients reported less satisfaction with nursing care, expected a shorter wait time, and were more accurate in perceiving wait time. This is consistent with other studies that found that wait time was the primary basis for patient satisfaction.7,10 Despite receiving more comfort items, having television in the treatment area, and receiving prompt information, they were less likely to view the nurses as caring, less satisfied with nurse teaching, and less likely to view comfort, information and engaging activities as very helpful. The differences in expectations at a site with a higher percentage of insured patients may reflect higher expectations for both shorter wait times, nurse teaching, and amenities.
Emergency departments have changed their priorities to include shorter wait time before seeing a physician. To some degree, this reflects a shift in priorities creating a conflict between efficiency and what many ED nurses view as effectiveness.4 Systematically providing comfort items including handing a brochure out, rounding every 30 minutes, and providing engaging activities and reminders of options did not significantly increase wait time but did increase satisfaction with nursing care. Lowering lights was a comfort item rarely offered to usual care but valued by patients. Perhaps the increase in perceived helpfulness of comfort measures was related to nurses systematically offering them (comfort measures), thus increasing patient’s perceptions of comfort and caring.
When the brochures with information about the ED visit was at the admission desk and given by the nurse in the treatment room coincidental with the intervention, the numbers of patients reading the brochure increased dramatically. This possibly reduced need for information on the process of ED care. A high percentage of participants at both sites reported receiving updates from the nurse with usual and intervention care. We believe the systematic rounding contributed to their perceptions that information was helpful. Another study on rounding in the ED found a resultant rise in patient satisfaction and reduction in falls.12 Both modes of information reduced the patient’s feelings of not knowing why they were waiting, a major dissatisfier.1
Waiting for treatment or results of tests may be boring. Engaging activities reduce boredom. The most utilized engaging activity in this study was talking with a family member or friend while in the ED treatment room. This has been found to also provide social support, increasing patient satisfaction.19 Many patients in the urban hospital did not have family or friends accompany them and anecdotally reported that the puzzle as a big help to help time pass faster.
There were very few Spanish-speaking participants, reducing representativeness of this minority group. Research nurses were in staffing and included their patients as participants. This may have skewed responses. In addition, research nurses worked only dayshift; thus, the patient population did not include those coming to the ED at night.
Implications for Emergency Nurses and Conclusion
Emergency nurses work in fast-paced field where time management is crucial. They have minimal control over how long a patient waits in the treatment area and often see satisfaction decrease as time passes. The nurses can control communication and caring behaviors they exhibit, both of which contribute to satisfaction.3 This study supports the value of communication and caring behaviors such as comfort measures, nursing rounds, and provision of engaging activities. Small gestures can make a difference; for example, we found that handing patients brochures led to more use and possibly lowered anxiety about the visit. We also found patients found simple interventions such as lowering lights or providing a puzzle to pass the time helpful and possibly built rapport. Making a difference with similar interventions as used in this study is under the nurse’s control and provides both job satisfaction and patient satisfaction.
1. Taylor D, Kennedy M, Virtue E, McDonald GA. Multifaceted intervention improves patient satisfaction and perceptions of emergency department care. Int J Qual Health Care. 2006; 18 (3): 238–245.
2. Laschinger HS, McGillis-Hall L, Almost J. A psychometric analysis of the patient satisfaction with nursing care quality questionnaire. J Nurs Care Qual. 2004; 20 (5): 220–230.
3. Liu SS, Franz D, Allen M, et al. ED services: the impact of caring behaviors on patient loyalty. J Emerg Nurs. 2010; 36 (5): 404–414.
4. Johansen ML. Conflicting priorities: emergency nurses’ perceived disconnect between patient satisfaction and the delivery of quality patient care [published online ahead of print]. J Emerg Nurs. 2012.
5. Wagner D, Bear M. Patient satisfaction with nursing care: a concept analysis within a nursing framework. J Adv Nurs. 2009; 65 (3): 692–701.
6. Boudreaux SN, O’Hea E. Patient satisfaction in the emergency department: a review of the literature and implications for practice. J Emerg Med. 2004; 26: 13–26.
7. Taylor C, Benger JR. Patient satisfaction with emergency medicine. Emerg Med J. 2004; 21 (5): 528–532.
8. Fottler MD, Ford RC. Managing patient waits in hospital emergency departments. Health Care Manag. 2002: 21 (1): 46–61.
9. Grafstein E, Wilson D, Stenstrom R, et al. A regional survey to determine factors influencing patient choices in selecting a particular emergency department for care. Acad Emerg Med. 2013; 20: 63–70.
10. Arendt KW, Sadosty AT, Weaver AL, Brent CR, Boie ET. The left-without-being-seen patients: what would keep them from leaving? Ann Emerg Med. 2003; 42 (3): 317–323.
11. Meade CM, Kennedy J, Kaplan J. The effects of emergency department staff rounding on patient safety and satisfaction. J Emerg Med. 2010; 38 (5): 666–674.
12. Soremekun OA, Yakayesu JK, Bohan SJ. Framework for analyzing wait times and other factors that impact patient satisfaction in the emergency department. J Emerg Med. 2011; 41 (6): 686–692.
13. Wilde-Larrson B, Larsson G, Starrin B. Patient’s views on quality of care: a comparison of men and women. J Nurs Manag. 1999; 7: 133–139.
14. Franzen C, Bjornstig U, Jansson L, Stenlund H, Brulin C. Injured road user’s experiences of care in the emergency department. J Clin Nurs. 2007; 17: 726–734.
15. Davis BA, Bush HA. Patient satisfaction of emergency nursing care in the United States, Slovenia, and Australia. J Nurs Care Qual. 2003; 16 (4): 267–273.
16. Davis BA, Duffy E. Patient satisfaction with nursing care in a rural and urban emergency department. Aust J Rural Health. 1999; 7: 97–103.
17. Davis B, Kiesel C, McFarland J, Collard A, Coston K, Keeton A. Evaluating instruments for quality. J Nurs Care Qual. 2005; 20 (4): 364–368.
18. Barrio AC, Garcia CB, Cereijo CR, Lopez FG. Spanish validation of an instrument to measure quality of nursing care in hospital emergency units. J Nurs Care Qual. 2002; 16 (3): 13–23.
19. Kristensson J, Ekwall A. Psychometric qualities of the consumer emergency care satisfaction scale. J Nurs Care Qual. 2008; 23 (3): 277–282.