Emergency departments in the UK are facing increasing demands and economic pressures (Care Quality Commission (CQC), 2017a; NHS England and NHS Digital, 2018). These pressures along with staff shortages can lead to substandard care. In some instances, aspects of fundamental patient care, such as emotional support and provision of nutrition, have been reported by patients to be poor in emergency departments (CQC, 2017b).
The psychological and emotional aspects of patient care are essential parts of holistic care and the patient experience (Gerteis et al, 2002; Committee on Quality of Healthcare in America, 2001; National Clinical Guidelines Centre (NCGC), 2012; NHS National Quality Board, 2012; Paparella 2016). In the 2018 CQC survey of adult inpatients 28% of patients stated that they were unable to find a staff member to talk about their worries and fears (CQC, 2019). A Royal College of Nursing (2012) report found that 78% of nurses stated that on their last shift, they were unable to adequately comfort patients due to a lack of time. Furthermore, the CQC's survey of emergency department patients in 2016 found that of those who were feeling distressed only 48% felt that a member of staff definitely reassured them and 57% said that a doctor or nurse discussed their fears and anxieties about treatment and their condition (CQC, 2017b). There is therefore a need to improve this aspect of care. This study investigated whether volunteers can help improve patient experience by providing emotional support to patients in the emergency department.
The nutritional aspect of patient care is a vital part of patient-centred care and the patient experience. Patients should be supported in meeting their individual nutrition and hydration needs. Furthermore, the quality standard from the National Institute for Health and Care Excellence (NICE) (2012) reiterates the importance of nutritional support in ensuring the best health outcomes because it is a cause and effect of ill-health. With rising numbers of hospital emergency admissions and a lack of hospital capacity to meet these demands, patients remain in emergency departments for greater lengths of time (NHS England and NHS Digital, 2018). It is therefore essential that there is adequate provision for nutrition and hydration in emergency departments. Yet the CQC survey (2017b) found that, when patients were asked if they had access to food and drink, only 56% said they could access suitable nutrition and 24% stated that they were unable to get any food or drink. It is therefore necessary that this part of patient care is improved in emergency departments. This study explored whether volunteers can contribute to providing better nutritional support for patients.
Hospital volunteers are generally received well by staff and patients. Although evidence is limited, volunteers have been shown to improve patient hospital experience in relation to psychosocial care (Mundle et al, 2013; Koivula and Karttunen, 2014; Steunenberg et al, 2016; Babudu et al, 2016; O'Sullivan et al, 2016). There is also evidence indicating improved patient satisfaction with the introduction of volunteer schemes in emergency departments. However, supporting evidence is either outdated or of poor quality (Johnson et al, 1993; Wolford, 1995; Munn and Takeno, 1996; Quinn, 2009; Sanon et al, 2014; Samaddar et al, 2018).
Hospital volunteers have been shown to help in meeting patients' nutritional needs by assisting patients at mealtimes (Green et al, 2011; Tassone et al, 2015; Howson et al, 2017). Studies have shown volunteers to be helpful in providing dietary intake for patients; this relates to protein and energy intake (Walton et al, 2008, Wong et al, 2008; Wright et al, 2008; Manning et al, 2012; Roberts et al, 2014; Huang et al, 2015) and also the amount of food consumed (Robinson et al, 2002; Gilbert et al, 2013). Staff were shown to value volunteers because they felt that patients' psychosocial and nutritional needs were met and nursing time could be dedicated to clinical tasks (Wolford, 1995; Munn and Takeno, 1996; Prabhu et al, 2008; Walton et al, 2008; Brown and Jones, 2009; Quinn, 2009; Sneddon and Best, 2011; Manning et al, 2012; Buys et al, 2013; Roberts et al, 2014; Robison et al, 2015; Babudu et al, 2016; O'Sullivan et al, 2016; Steunenberg et al, 2016). Due to the lack of literature concerning volunteers in emergency departments, this research was undertaken to assess the effectiveness of hospital volunteers in the emergency setting.
Methods
Study design
This was a single-centre, cross-sectional study at a major 24-hour emergency department in England. In this department, volunteers were placed in clinical areas, ‘Majors—Chairs’ and ‘Majors—Trolleys’. Patients in ‘Majors–Trolleys’ are non-ambulatory patients and patients in ‘Majors–Chairs’ are ambulatory patients requiring blood investigations or enhanced observation.
Hospital volunteers undertook a training programme. The volunteer role was clearly outlined to both volunteers and clinical staff. Responsibilities include offering drinks, ordering meals, reassuring patients and further helping make patients comfortable by fetching blankets and providing pillows. However, volunteers do not assist in tasks such as toileting or mobilising patients.
Patient questionnaires were distributed and data collected on days when the volunteer scheme was operating. Comparative data were collected when volunteers were not in the department. Days when the volunteer scheme was running and not running were interspersed throughout the months from November 2018 to February 2019.
The study received ethical approval from IRAS (Integrated Research Application System, Project ID 241680), and ERGO II (Ethics and Research Governance Online, Submission ID 40148) and trust research governance committees. All participants were informed of the purpose of the questionnaire and required to give informed consent. Participating respondents were assured of confidentiality and anonymity of all data collected.
Participants
Participants attending the emergency department responded to a questionnaire between November 2018 and February 2019. Data were collected on 26 separate days.
Individuals who had been in the department for less than 30 minutes, anyone under the age of 16 years, patients in high-acuity areas such as resuscitation, and those who lacked capacity (due to issues in gaining informed consent) were excluded. Patients approaching the end of life and their relatives were excluded because this may have been inappropriate.
The patient admission screen was monitored. All patients arriving in the departmental areas, Majors-Trolleys and Majors-Chairs, while the volunteer scheme was running, were given a questionnaire (Table 1). Patients were requested to complete this once discharged from the department or just before being moved to a ward, ensuring that responses were based on their’ full experience in the department. Patients were asked to return questionnaires to a staff member or place them in the return box. This same procedure was followed for collecting comparative data when volunteers were not in the department.
| Question | Responses |
|---|---|
| Q5. If you were feeling distressed in the emergency department did a member of staff comfort or reassure you? | Yes, definitely |
| Q6. If yes, who provided comfort or reassurance? | Nurse/healthcare assistant |
| Q7. Were you able to get food and drink while you were in the emergency department? | Yes |
| Q8. Did a member of staff offer you food or drink? | Yes |
| Q9. If yes, who provided the food or drink? Please tick all that are relevant | Nurse/healthcare assistant |
| Q10. If you had contact with a hospital volunteer, did they impact positively on your experience in the emergency department? | Yes, definitely |
Outcomes
Patient questionnaires were based on the CQC Emergency Department Survey (CQC, 2017b) and included the Friends & Family Test (FFT) (NHS England, 2015). As summarised in the development report for the emergency department survey (Picker Institute Europe, 2016) the CQC questionnaire followed best practice guidelines and question changes from the previous survey were cognitively tested. Patients were given the option to add comments. The following demographic factors were collected: age, sex, presence of mental health condition and area within department.
Participants were asked to answer questions in relation to emotional distress, access to food and drink and experience of volunteers. At the end, patients were asked to complete the FFT (NHS England, 2015) for assessing overall satisfaction with the emergency department service.
Sample size
Sample size calculations were based on the key questions, Q5 and Q7 (Table 1). Since the scores for the trust in which the data were collected were banded as ‘about the same’ compared with most other trusts, national data estimates from the 2016 Emergency Department Survey (CQC, 2017b) were used: 48% of those who felt distressed that said a member of staff had definitely helped to reassure them; 56% of those who wanted something to eat or drink said they were able to get something suitable. Calculations were based on improving the satisfaction rates of 48% and 56% by 20–25% respectively, sample sizes of between 82 and 52 and between 76 and 48 patients in each group would be required for a one-tailed significance level of 0.05 and power of 0.80.
Statistical analysis
All data, including demographics and questionnaire responses, were entered into a database. Data were cleaned, checking for valid responses such as more than one box checked when only one is required, and missing data recorded as appropriate. All data were analysed using SPSSv24 statistical package.
Demographics of the patients were described using means and standard deviations or counts and percentages, as appropriate. Questionnaire data have been reported as counts and percentages for each response categorisation.
Comparisons between the characteristics of the two groups of patients, those attending when the volunteer scheme was running and those when it was not, were made using Fisher's exact test and t-test as appropriate. For patient experience responses percentage rate differences (%RD) were calculated as the percentage improvement in satisfaction rates in patients when the volunteer scheme was operating compared with responses when there were no volunteers; 95% confidence intervals (95% CI) and P values were computed. The statistical level of significance was set at P<0.05.
For the key questions, Q5 and Q7, the scoring method in the CQC survey statistical release (CQC, 2017b) was implemented. For Q5 the responses were dichotomised as ‘Yes definitely’ versus the remaining responses. Patients who were not distressed were excluded. For Q7, responses were scored as ‘Yes’ versus the rest; patients who did not want to eat or drink were excluded. For Q10 responses were dichotomised as ‘Yes definitely’ versus the remaining responses (Table 1).
For the question relating to recommending the department to friends and family if they needed similar care or treatment, the scale was dichotomised as ‘Extremely likely’ or ‘Likely’ vs the remaining responses (‘Neither likely nor unlikely’, ‘Unlikely’, ‘Extremely unlikely’).
Subgroup analyses were performed for characteristic factors that differed significantly between the patient groups. Any significant results for patient experiences were analysed further in separate subgroups stratified by the differing factor.
Results
Of the 241 questionnaires distributed, 203 were returned giving a response rate of 84.2%. Of the returned questionnaires, 151 patients were in the group in which the volunteer scheme was operating and 52 patients in the group attending the department when it was not; 89 (43.8%) felt ‘distressed’ 163 (80.3%) wanted food or drink. Of the patients present while the volunteer scheme was running 77/138 (55.8%) reported having contact with a volunteer. Data reporting contact with a volunteer were missing in 13 cases.
Patient demographics and characteristics were similar between the two groups (Table 2). The only significant difference was for arrival time: when the volunteer scheme was running, patients arrived after noon less often than when there were no volunteers (69/142 vs 36/52, P=0.014).
| All participants | Volunteer scheme | No volunteer scheme | |||||
|---|---|---|---|---|---|---|---|
| Total | n (%)* | Total | n (%)* | Total | n (%)* | P† | |
| Age (years) | 199 | 59.79 (23.5) | 148 | 59.53 (23.7) | 51 | 60.53 (23.1) | 0.79 |
| Male | 203 | 94 (46.3) | 151 | 67 (44.4) | 52 | 27 (51.9) | 0.42 |
| Area (Majors—Trolleys) | 202 | 101 (50.0) | 150 | 72 (48.0) | 52 | 29 (55.8) | 0.42 |
| Treated mental health | 201 | 25 (12.4) | 149 | 16 (10.7) | 52 | 9 (17.3) | 0.23 |
| Day of week (weekend) | 203 | 23 (11.3) | 151 | 15 (9.9) | 52 | 8 (15.4) | 0.31 |
| Arrival time>12:00 | 194 | 105 (54.1) | 142 | 69 (48.6) | 52 | 36 (69.2) | 0.014 |
| Distressed | 203 | 89 (43.8) | 151 | 68 (45.0) | 52 | 21 (40.4) | 0.63 |
| Wanted food or drink | 203 | 163 (80.3) | 151 | 124 (82.1) | 52 | 39 (75.0) | 0.31 |
Number of cases with missing data: age (4); area (1); treated mental health (2); arrival time (9)
Patients attending the department when the volunteer scheme was running reported getting something to eat and drink significantly more often (Table 3) compared with when there were no volunteers (96/124 vs 20/39, %RD 26, 95% CI 10–42, P=0.002), with a member of staff offering patients something to eat and drink significantly more frequently compared with when the scheme was not operating (96/146 vs 19/52, %RD 29, 95% CI 14–45, P<0.001. When the volunteer scheme was running, nurses or healthcare assistants offered patients food and drink less often than when a volunteer was not in the department (48/96 vs 14/19, %RD -24, 95% CI -48 to +1, P=0.058), although this difference did not quite reach statistical significance.
| Volunteer scheme n/total (%) | No volunteer scheme n/total (%) | Rate difference (%)* %RD (95% CI) | P | |
|---|---|---|---|---|
| Emotional support | ||||
| Definitely comforted | 49/68 (72.1%) | 14/21 (66.7%) | 5 (-17–28) | 0.63 |
| Definitely comforted by nurse/healthcare assistant | 43/49 (87.8%) | 9/14 (64.3%) | 23 (1–46) | 0.041 |
| Definitely comforted by doctor | 23/49 (46.9%) | 6/14 (42.9%) | 4 (-26–34) | 0.79 |
| Definitely comforted by other | 1/49 (2.0%) | 1/14 (7.1%) | -5 (-16–5) | 0.34 |
| Access to food and drink | ||||
| Able to get food or drink† | 96/124 (77.4%) | 20/39 (51.3%) | 26 (10–42) | 0.002 |
| Offered food or drink by staff | 96/146 (65.8%) | 19/52 (36.5%) | 29 (14–45) | <0.001 |
| Offered by nurse/healthcare assistant | 48/96 (50.0%) | 14/19 (73.7%) | -24 (-48–1) | 0.058 |
| Offered by doctor | 4/96 (4.2%) | 2/19 (10.5%) | -6 (-17–5) | 0.25 |
| Offered by other | 6/96 (6.3%) | 3/19 (15.8%) | -10 (-23–4) | 0.16 |
Number of cases with missing data: definitely comforted (4); were offered food or drink (11)
For emotional support of those who felt distressed, more patients reported being definitely comforted by a staff member in the group of patients in the department while the volunteer scheme was running (Table 3), however, this difference was not significant (49/68 vs 14/21, %RD 5, 95% CI -17 to +28, P=0.63). Patients were primarily comforted by nurses or healthcare assistants and doctors; 43 (87.8%) reported being definitely comforted by a nurse or healthcare assistant and 23 (46.9%) by a doctor as opposed to only 8 (16.3%) by a hospital volunteer. Significantly more patients reported definitely being comforted by a nurse or healthcare assistant when the volunteer scheme was running than when the scheme was not running (43/49 vs 9/14, %RD 23, 95% CI 1–46, P=0.041).
Because there was a significant difference in arrival times between patients when the volunteer scheme was running and patients when it was not, additional subgroup analyses, stratified into subgroups before and after the arrival time of 12:00, were carried out to investigate whether patient arrival time had influenced the significant results for patient experiences of nutritional support (Table 4). Arrival time was dichotomised around 12:00 because the scheme only ran during the daytime and it was similar to the median arrival time of 12:15. The subgroup analyses found that the significant effect of the volunteer scheme was in patients arriving after 12:00 with no significance difference between the two patient groups before 12:00. In the subgroup with arrival times after 12:00, patients present while the volunteer scheme was running, compared with patients when the scheme was not, reported being able to get food and drink more often (37/53 vs 8/26, %RD 39, 95% CI 16–62, P=0.001) and also being offered food and drink by a member of staff more often (41/67 vs 6/36, %RD 45, 95% CI 24–65, P<0.001).
| Volunteer scheme n/total (%) | No volunteer scheme n/total (%) | Rate difference (%)*%RD (95% CI) | P | |
|---|---|---|---|---|
| Arrival time ≤12:00 | ||||
| Able to get food or drink† | 54/64 (84.4%) | 12/13 (92.3%) | -8 (-29–13) | 0.46 |
| Offered food or drink by staff | 52/70 (74.3%) | 13/16 (81.3%) | -7 (-30–16) | 0.56 |
| Arrival time >12:00 | ||||
| Able to get food or drink† | 37/53 (69.8%) | 8/26 (30.8%) | 39 (16–62) | 0.001 |
| Offered food or drink by staff | 41/67 (61.2%) | 6/36 (16.7%) | 45 (24–65) | <0.001 |
Number of cases with missing arrival time 9
The majority of patients had a positive experience of volunteers; 57/77 (74.0%) reported that volunteers definitely impacted positively on their experience in the department. In response to the FFT, more than 95% of patients responded that they would be ‘extremely likely’ to recommend the emergency department to friends and family if they needed similar care or treatment in both groups, when the volunteer scheme was running and not running (137/143 vs 47/49, %RD 0, 95% CI-7 to +6, P=0.97).
Discussion
This study is the first to assess the effectiveness of hospital volunteers in an emergency department through the evaluation of the patient experience. Hospital volunteers were shown to be significantly effective in meeting patients' nutritional needs. Patients in the department when the volunteer scheme was running reported obtaining food and drink and being offered something to eat and drink by a staff member more often, compared with patients when the volunteer scheme was not running. It should be noted that some patients were able to obtain food and drink through other means such as vending machines and therefore may not have been offered food and drink by a staff member. Of the clinical staff, it was mainly nurses and healthcare assistants who offered patients food and drink. Although there was some overlap with nurses or healthcare assistants and hospital volunteers offering nutritional support, volunteers often provided nutrition to patients independently of other staff members. The results for emotional support are less conclusive, and volunteers were not shown to be significantly effective when it came to providing emotional support. Of patients who felt distressed, more reported being definitely comforted by a member of staff while the volunteer scheme was running, however, this difference was not significant. Moreover, patients were primarily comforted by nurses or healthcare assistants and doctors as opposed to hospital volunteers.
In common with existing research, which is mainly based in inpatient wards, this study found that volunteers provide a valuable contribution in meeting the nutritional needs of patients (Green et al, 2011; Tassone et al, 2015; Howson et al, 2017). The role of the volunteers in existing studies was primarily as ‘mealtime assistants’, supporting patients with their nutritional intake by helping cut up food or feed patients. Volunteers have been shown to be helpful in providing dietary intake for patients (Robinson et al, 2002; Walton et al, 2008; Wong et al, 2008; Wright et al, 2008; Manning et al, 2012; Gilbert et al, 2013; Roberts et al, 2014; Huang et al, 2015). Although it was beyond the scope of this study to investigate whether patients' dietary intake was improved, volunteers were shown to be effective in sourcing and providing nutrition for patients. This is an aspect of nutritional support not investigated in any other study and especially relevant to emergency departments where food and drink is not routinely provided at structured mealtimes.
The volunteer scheme in providing nutritional support was shown to be significantly effective for patients arriving after 12:00 with no significance difference between the two patient groups before 12:00. Without additional research into the number of patient attendances, the acuity of patients and staffing levels, it is difficult to ascertain the true reason for this finding. However, a potential reason could be that the emergency department may have been busier in the afternoon, compared with the morning. This could mean that there is a greater need for volunteers in helping nursing staff to provide nutritional support at busier times. It would be useful to undertake further research in exploring the association between time of day and demands in the department so that volunteer schemes can be optimised.
Studies have shown how volunteers were seen to be beneficial in relieving nursing workload by helping patients with nutrition. In some studies nurses reported having more time to support patients at mealtimes and assist patients with complex needs, such as dysphagia, when volunteers were helping (Walton et al, 2008; Brown and Jones, 2009; Buys et al, 2013; Roberts et al, 2014; Robison et al, 2015). Mealtime assistants were found to be helpful in preparing patients for meals, by opening packaging and providing cutlery (Walton et al, 2008; Brown and Jones, 2009; Sneddon and Best, 2011; Manning et al, 2012, Roberts et al, 2014, Robison et al, 2015). These findings are supported by this emergency department study, which showed that when the volunteer scheme was running nurses or healthcare assistants offered patients food and drink less often than when there was no volunteer in the department. Moreover, volunteers contributed to patients' nutritional needs, independently of other staff members. This would imply that hospital volunteers not only add to staff capacity in providing nutritional support for patients but also partially relieve nursing staff from this aspect of care.
Existing research has also shown that volunteers have an impact on the amount of time nurses spend on other non-clinical tasks, such as talking with and comforting patients. One hospital's volunteer service evaluation found that, when volunteers were present, nurses spent less time on tasks that did not require a qualified nurse (Babudu et al, 2016). This allowed nurses to focus time on clinical duties. In other literature nurses reported that volunteers were able to provide patients with extra attention that they were unable to give due to clinical duties (Halford and Fraser, 2013; Charalambous, 2014; Fitzsimons et al, 2014; Steunenberg et al, 2016; Ross et al, 2018). Although it was beyond the scope of this study to explore the impact of the volunteer scheme on clinical nursing duties, it was found that more patients in the department while the volunteer scheme was running reported definitely being comforted by a nurse or healthcare assistant than when the volunteer scheme was not running. A possible explanation, to be explored in future research, could be related to volunteer impact on staff capacity. As volunteers were providing a significant contribution to nutritional support for patients, nurses may have had more time to dedicate to providing comfort to patients. Investigating the impact of hospital volunteers on nursing workload could be further examined by a staff questionnaire to gain insight into nursing perspectives.
In this study, the majority of patients reported that volunteers definitely impacted positively on their experience in the emergency department. Previous studies have demonstrated how volunteers can positively impact on patient experience by providing emotional support and offering company to patients (Babudu et al, 2016; Steunenberg et al, 2016; Ross et al, 2018). An evaluation report of hospital volunteer schemes in 10 hospital trusts in England reported a significant improvement in patient mood, in relation to social engagement, when the volunteer schemes were running (Babudu et al, 2016). There was, however, no significant improvement in terms of levels of patient distress. The lack of improvement to patient distress was similar to the findings in this study. Although more patients reported being definitely comforted by a member of staff in the group of patients in the department while the volunteer scheme was running, this difference was not significant. Furthermore patients were primarily comforted by nurses or healthcare assistants and doctors as opposed to hospital volunteers. It is difficult to ascertain the reason for this finding; one explanation could be difficulty in volunteers establishing rapport with patients. Often patients had contact with volunteers in the ‘Majors–Chairs’ waiting room, which is a public area where patients might not want to have personal conversations. This may have hindered volunteers' abilities to establish rapport with patients. Patients reported receiving emotional support more from clinical staff. This could have been because patients built more meaningful rapport with these staff members, as opposed to volunteers, due to engaging in more personal conversations in private. Additional research would need to be undertaken in exploring this further.
The majority of patients responded that they would recommend the emergency department to friends and family if they needed similar care or treatment, both when the volunteer scheme was running and when it was not. There was no difference in the overall satisfaction between the group of patients present at the same time as the volunteer scheme and not. However, questions have been raised in relation to the validity of the FFT (Cornwell, 2012; NHS England, 2014). Patients are likely to go to their closest emergency department and recommend their family also do the same in an emergency. It is most likely to be the location and services of the department that determine patient attendances at emergency departments (NHS England, 2014). Prior negative experience would be unlikely to take precedence over receiving critical, urgent medical care. This would suggest that this question does not add much in assessing the impact of volunteers on patient satisfaction.
Limitations
The study was observational and participants were not randomised into the two patient groups, which could be a potential source of bias. Therefore one cannot reliably attribute the differences found between patient groups directly to the presence of the volunteer scheme. However, the characteristics of the two patient groups appear similar and the statistical association, especially in relation to nutritional support, is substantial and highly significant.
Patients who had dementia or delirium were excluded from the study due to issues of consent. There is some evidence to support the effectiveness of hospital volunteers on improving mood and distress for patients with dementia (van der Ploeg et al, 2012; McDonnell et al, 2014; Babudu et al, 2016). These patients could have been emotionally distressed and comforted by volunteers but were not included in this study. In addition, it is likely that this group of patients may have needed assistance with eating and drinking and might have received assistance from volunteers.
Conclusion
Volunteers have been shown to provide a significant contribution in meeting the nutritional needs of patients in emergency departments by sourcing and providing food and drink for patients. Hospital volunteers added to staff capacity in providing nutritional support for patients, partially relieving this aspect of patient care from nursing staff. This study did not show volunteers to be significantly effective in providing comfort to patients who reported being distressed. It was primarily nurses, healthcare assistants and doctors who provided emotional support to these patients. Additional research would need to be undertaken to fully understand the reasons for these findings.