Factors associated with patients’ and companions’ satisfaction with a hospital emergency department: A descriptive, cross‐sectional study

Abstract Aim The aim was to study the level of satisfaction with an emergency department and to identify the factors associated with satisfaction. Design This research consisted of a descriptive, cross‐sectional study. Methods The study population was composed of patients and companions who visited the emergency department during a three‐month period. The patients were selected randomly from the register of visits to the emergency department during the period of study. Sociodemographic variables and variables related to the attention received were studied through questionnaires (response rate 33%). Results Older age was associated with greater satisfaction (p = 0.002), as was female sex (p = 0.013) and greater optimism (p < 0.001). Greater control of pain was a factor associated with satisfaction (p = 0.033), as was the perception of a shorter waiting time before the medical visit (p < 0.001).

Patient satisfaction refers to the subjective experience of the user of the health system (Boxer & Boxer Goldfarb, 2009). A satisfied patient better adheres to treatment, makes fewer malpractice complaints and is more willing to return to the service than otherwise (Boxer & Boxer Goldfarb, 2009). When a person goes to an emergency department, he or she is normally accompanied by a family member, friend or carer, who in many cases will give emotional support and will be able to give relevant information about the functional status of the patient (Ekwall, Gerdtz, & Manias, 2008;Nikki, Lepistö, & Paavilainen, 2012). Friends and family members play an important role in the care and well-being of the patient in the emergency department (Gordon, Sheppard, & Anaf, 2010), and the experience in the department affects companions' perception of the quality of care (Cooke, Watt, Wertzler, & Quan, 2006;Parra Hidalgo, Calle Urra, Ramón Esparza, Peiró Moreno, & Meneu de Guillerna, 2012).
In a review of the literature about patient and companion satisfaction with emergency departments, where studies of patient characteristics and perceived waiting times particularly stand out,
Based on the above, we can affirm that the satisfaction of patients and companions with emergency departments is a measure of quality of care and a factor that influences both the choice of the service by the user and adherence to the treatment prescribed.
However, we were unaware of the profile of the most satisfied users and companions in our setting and so we proposed the following objectives: • To study the level of satisfaction of patients and companions with an emergency department.
• To identify the factors associated with the satisfaction level of patients and companions with an emergency department.
• To determine the profile of the most satisfied users of the emergency department.

| Design
The research consisted of a descriptive, cross-sectional study of the satisfaction of patients and companions. The participants voluntarily accepted to participate in this study, which was conducted in the emergency department of a general hospital. The field was undertaken using questionnaires from October to December 2012.

| Study settings
This study was performed at a general hospital that attended 65,071 emergencies, a mean average of 178 visits per day.
The study population was composed of patients and companions who visited the emergency department of a general hospital from October to December 2012. The sampling technique was probabilistic, involving a random selection of patients from four tables of random numbers from the daily register of visits to the emergency department during the 3-month period.

| Participants
All patients over 18 years of age were included. Patients residing outside the Spanish territory were excluded from the study, as were those who left the service without being visited, those who needed only nursing care and hospital workers who were seen at the service.
When a patient who was selected died during the medical event, a letter was sent to family members expressing condolences and explaining the aim of the study.
The sample size was calculated considering a 10-point difference with regard to the results obtained in the study of González et al. (2008); thus to obtain a power of 80% with a significance level of 0.05, 194 patients and companions would be needed, calculating response losses of 40%.

| Measures
A satisfaction questionnaire was sent by ordinary post to the selected patients. At the beginning of the questionnaire, there was an item to identify whether the person responding was a patient or a companion. When the questionnaire was completed together by both a patient and a companion, the answers were treated as though they were of the patient.
A letter was sent to the home address of all the patients selected within the 10 days following the visit; it contained an information sheet about the study, a letter of informed consent to be signed, the questionnaire and a pre-paid envelope to return the questionnaire to the hospital. One month after the first mailing, a reminder letter was sent to all those who had not yet replied to the questionnaire.
Due to the low response rate, selected patients were contacted via telephone to determine if a third mailing would be necessary. All of those who agreed to reply were again sent a questionnaire with a pre-paid envelope.
The first variables studied were sociodemographic: age, sex, civil status, educational level and occupation. Questions on the following were asked: overall satisfaction with the visit to the emergency department (on a scale of 0-10, where 0 was the worst possible score and 10 the best possible score); pain (were you in pain when you visited the emergency department? with two possible responses: yes or no) and whether this pain was adequately controlled (answered via a Likert scale: no, not at all; slightly but not completely; quite a lot but not completely; or yes, completely); the waiting time to be seen for the first time by a nurse (in minutes); the perception of this time (answered via a Likert scale: very short, short, acceptable, long or very long); the waiting time to be seen by a doctor (in minutes); the perception of the waiting time for the medical visit (answered via a Likert scale: very short, short, acceptable, long or extremely long); information on the approximate waiting time (whether or not information was received); and whether or not they would recommend the department to a family member or friend (yes or no). Additionally, the perceived level of optimism was studied (on a scale from 0-10, where 0 was the worst score possible and 10 the best).
To study the factors associated with satisfaction, the numerical global satisfaction variable was recoded into two categories: "less satisfied," which included those users who gave a score of between 0-7 and "more satisfied," which included those who gave a score of between 8-10 points. The pain control variable was also recoded as "pain controlled," for those who had referred to their pain being sufficiently or completely controlled and "pain not controlled" for those referred to their pain having been little or not controlled.
The satisfaction questionnaire validated by

| Analysis
IBM ® SPSS Statistics ® version 19 was used to analyse the data.
The numerical variables are described by mean and standard deviation or by median and interquartile range. To study the association between the variables that met the criteria of normality, Student's t test was used to compare quantitative and categorical variables, the Pearson's correlation was used to compare quantitative variables and the chi-squared test was used for categorical variables. For variables that did not meet the criteria of normality, the Mann-Whitney U test and the Kruskal-Wallis non-parametric test were used to compare quantitative and categorical variables and the Spearman's rho test was used for quantitative variables. A binary regression model was run from factors associated with patients' and companions' satisfaction with the emergency department adjusting for age, sex and level of optimism. The level of significance was taken as p < 0.05.

| Ethics
All the participants who responded to the questionnaire also returned the signed informed consent. This project respects the Helsinki Declaration of the World Medical Association. The study was approved on 1 August 2012 by the ethical committee of the hospital before beginning the study.

| RE SULTS
A total of 15,273 patients were attended to, from whom, 1,526 patients were randomly selected (864 patients were included and 662 were excluded). A total of 285 responses were received (response rate 33.0%); of which, 221 (77.5%) corresponded to patients, 62 (21.8%) to companions and 2 (0.7%) to respondents who failed to identify themselves as either a patient or companion. The mean age of the respondents was 54.6 years (SD = 18.3), and 53.6% were women. The sociodemographic data of the patients and companions are presented in Table 1. Analysing the association between subjective overall satisfaction and the age of the participants shows that older patients scored higher than younger patients with respect to overall satisfaction with the visit (r = 0.236; p < 0.001) (Pearson's correlation).
The presence of pain was compared with overall satisfaction.
Patients with pain gave a mean satisfaction score of 7.5 (SD = 2.3) [median = 8; IQR = 2], whereas those who were not in pain gave a mean score of 7.8 (SD = 1.6) [median = 8; IQR = 2] (Mann-Whitney U test, p = 0.898). On the other hand, studying the association of control of pain and overall satisfaction shows that people whose pain was better controlled gave higher scores for overall satisfaction with the visit (Kruskal-Wallis test, p < 0.001).
The perceived waiting time for nursing triage was compared with subjective overall satisfaction. It was found that patients who perceived that they had waited for a very short or short period of time gave higher scores for subjective satisfaction than did those who perceived that they had waited a long or extremely long time (Kruskal-Wallis test, p < 0.001). The relationship between the perceived waiting time for the medical visit and subjective overall satisfaction was also studied; patients who had the perception of having waited for a short or for a very short period of time gave higher scores for subjective satisfaction than did those who had the perception of having waited for a long time or for an extremely long time (Kruskal-Wallis test, p < 0.001).     Control of pain was a factor associated with satisfaction and greater control of pain was associated with greater satisfaction (OR = 0.418; CI 95% [0.19-0.93]; p = 0.033). The perceived waiting time was also associated with satisfaction, with those who perceived that the waiting time for the medical visit was very short being among the most satisfied (OR = 5.456; CI 95% [2.14-13.94];

TA B L E 3 Factors associated with the satisfaction of patients and companions a
p < 0.001).

| D ISCUSS I ON
The present study has identified the factors associated with the patients' and companions' satisfaction with the emergency department of a Spanish general hospital.
The chosen method for the questionnaire, a mailing with a prepaid envelope, achieved a 33% response rate, which was similar to that of other studies (González-Valentín, Padín López, & de Ramón Garrido, 2005;Pujiula-Masó et al., 2006).
Subjective overall satisfaction with the visit to the emergency department was high, with a mean score of 7.6, which is in line with the scores obtained in similar studies (Gea, Hernán-García, Jiménez-Martín, & Cabrera, 2001;Parra Hidalgo et al., 2012). The dimension on the satisfaction scale that scored best was cleanliness, followed by information and medical care; in contrast, the worst scores were for comfort. A comparison with the results found in the literature shows that the dimensions of information and medical care and nursing care were better valued than in the study of González et al. (2008), where three of the four hospitals surveyed obtained lower scores in this dimension. On the other hand, with regard to privacy, the scores given by the patients and companions in our study were lower than those of these authors.
Older patients were found to be more satisfied, coinciding with other authors (Crow et al., 2002;Quintana et al., 2006). Additionally, women were found to be more likely to belong to the group of more satisfied users, unlike in previous studies, which have not found sex to have any relationship with satisfaction (Crow et al., 2002).
In our study, more than eight out of ten of the patients surveyed referred to pain when they visited the emergency department and in most cases, the pain was controlled completely or sufficiently. This result is similar to Kamali, Jain, Jain, and Schneider (2013), who found that more than half of all patients referred to having had some type of pain on arrival at the emergency department. In the present study, satisfaction with the service was not associated with the presence of pain but, rather, with the control of pain. Patients whose pain was controlled sufficiently or completely were among the group of most satisfied users, which is in line with previous studies (Byczkowski et al., 2013;Soleimanpour et al., 2011;Welch, 2010). Similarly, Marco, Davis, Chang, Mann, and Olson (2015) observed that the treatment of pain was a factor of dissatisfaction in a study carried out in a hospital emergency department. Muntlin et al. (2006) also observed that patients were not satisfied with the pain relief and that pain continues to be undertreated in emergency departments; thus, they recommend that pain should be treated individually. Furthermore, there are authors who consider that the implementation of a plan for pain management can be effective in improving patient satisfaction (Nairn et al., 2004;Welch, 2010).
In our study, the perception of the waiting time was one of the factors that most influenced the subjective overall satisfaction of users in terms of both nursing care and the medical visit. Similarly, in a study undertaken in an emergency department, patients who referred to having waited less time were more satisfied than those who had waited longer (Mercer, Hernandez-Boussard, Mahadevan, & Strehlow, 2014). A long or extremely long perceived waiting time has been shown to be negatively associated with satisfaction and users who referred to this aspect were the least satisfied, which is in line with earlier findings (Bos, Van Stel, Schrijvers, & Sturms, 2015;Brown et al., 2005;Nairn et al., 2004).
Being informed of the approximate waiting time also influences users' satisfaction. The present results show that users who were informed about the approximate waiting time gave better scores for satisfaction with the service. In an investigation in an emergency department, Kington and Short (2010) observed that those who were surveyed asked for greater information about the waiting time. Similarly, Burström, Starrin, Engström, and Thulesius (2013) observed that giving information about the approximate waiting time and the attitudes of the staff were important aspects with regard to patient satisfaction. Along the same lines, in a literature review, Innes, Jackson, Plummer, and Elliott (2015) highlighted the importance of communicating the waiting times in emergency departments.
Patient satisfaction must include, along with overall satisfaction with the service, the probability that the patient will recommend the service to others and the willingness to return to the service (Welch, 2010). As in earlier studies (Boudreaux & O'Hea, 2004;Parker & Marco, 2014;Soleimanpour et al., 2011), we observed that user satisfaction is a key component in the choice and recommendation of an emergency service. On the other hand, user dissatisfaction may affect the viability of health institutions (Boudreaux & O'Hea, 2004) and influence the perception that users have of a hospital (Broadwater-Hollifield et al., 2014;Wagner, 2014).
Regarding optimism, the most satisfied users scored higher for optimism. Costello et al. (2008) observed that pessimistic patients gave lower scores for their level of satisfaction with the care received.
Identifying the factors associated with user satisfaction makes it possible to define strategies and actions to improve the quality of the emergency department. It is especially important to improve the control of pain in the emergency department and to reduce the real and perceived waiting times as well as to improve the supply of information about these times.

| Limitations
The possible limitations of the present study are related to the methodology employed; the cross-sectional design does not allow causality to be established. Furthermore, we have no way of knowing the motivations of those who left the service without being visited and it is likely that these may include dissatisfaction with the service. One of the strengths of this investigation is the inclusion of companions in the patient satisfaction study.

| CON CLUS ION
The level of satisfaction with the emergency department was high, and the factors associated with the satisfaction of patients and companions were age, sex, optimism, control of pain and the perceived waiting time before the medical visit.
The most satisfied users were those who were older, women, patients whose pain was better controlled and those who perceived shorter waiting times and who were informed of an approximate waiting time.