Article Text
PDF
Abstract
Aim To evaluate the current provision of medical same day emergency care (SDEC) services within the UK, and the current utilisation of these pathways in the assessment of unplanned medical attendances.
Design Survey data was used from the Society for Acute Medicine Benchmarking Audit (SAMBA), including anonymised patient-level data collected annually using a day of care survey.
Setting Hospitals accepting unplanned medical attendances within the UK, 2019–2023.
Participants 34 948 unplanned and 4342 planned attendances across 188 hospital sites.
Results 29.8% of unplanned medical attendances received their initial medical assessment within SDEC services (2403 patients in SAMBA23), with the proportion increasing over time. 82.4% of patients assessed in SDEC services were discharged without overnight admission. Assessment in SDEC services was less likely in male patients, patients with frailty and older adults (all p<0.005).
Selected operational standards for SDEC delivery, set by the Society for Acute Medicine, were met in 64%–91% of hospitals. Most hospitals (82%) accepted referrals from emergency department triage and 63% accepted referrals directly from the paramedic team. 38% of hospitals did not use a recognised selection criteria to identify suitable patients for SDEC and only 8% used a criteria designed to identify patients suitable for discharge. Overall, 34.7% of medical attendances discharged without overnight admission received their medical assessment in locations other than SDEC.
Conclusions Medical SDEC provides assessment for one-third of patients seen through acute medicine services. Although the proportion of patients assessed within SDEC is increasing, further innovation and improvements are needed to ensure appropriate patients access this service.
- INTERNAL MEDICINE
- Organisation of health services
- GENERAL MEDICINE (see Internal Medicine)
- Health Services
Data availability statement
Data are available upon reasonable request. Data from this study are available from PIONEER, the Health Data Hub in Acute care, in accordance with Hub processes. See www.pioneerdatahub.co.uk and contact PIONEER@uhb.nhs.uk for more details.
This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.
Statistics from Altmetric.com
- INTERNAL MEDICINE
- Organisation of health services
- GENERAL MEDICINE (see Internal Medicine)
- Health Services
Strengths and limitations of this study
Medical same day emergency care (SDEC) has been widely adopted in the UK to deliver care to patients without overnight hospital admission; however, there is limited evidence guiding the development of this service.
This study compares hospital-level data describing SDEC service structure and processes, and patient-level data for over 35 000 patient attendances at 188 hospitals in the UK.
This is the largest evaluation of medical SDEC to date and demonstrates an increase in the use of SDEC for medical patients nationally.
The participation rate was higher among hospitals in England compared with the other three UK nations, which may limit generalisability.
Introduction
Acute hospital services continue to experience increasing pressure, with more than 500 000 unplanned emergency admissions each month in England alone.1 The majority of emergency admissions are due to medical conditions, which are assessed and managed through multiple pathways within acute services. The standard model of care within the UK is for patients requiring admission to an inpatient medical bed to be initially assessed and managed within an acute medical unit (AMU). However, many patients can be assessed, treated and discharged without an overnight stay, through medical same day emergency care (SDEC) services.2 3
SDEC services and their UK counterparts, previously known as ambulatory emergency care (AEC), aim to provide timely assessment, investigation and treatment of emergency conditions without the need for overnight admission.2 4 By reducing pressure on inpatient services and minimising the risks associated with hospital admission, SDEC should benefit individual patients and the healthcare service as a whole.5 6 As a result, the increased utilisation of SDEC in all acute hospitals is a key ambition within the NHS Long Term Plan, which suggests that one in three patients should be discharged same day, with improvements in this metric driven by increased utilisation of SDEC services.7 This increase may be facilitated by appropriate identification of patients suitable for treatment within SDEC services, and by condition-specific ambulatory management pathways.8
Despite the widespread adoption of SDEC as a model of care within acute services, there remains considerable variation in the provision of SDEC services between hospitals, from the specific services offered to the multi-professional workforce used to deliver care, whereby advanced practitioners and consultant practitioners often work in partnership with medical practitioners in those services. There is currently limited evidence evaluating the impact of SDEC provision on patient flow, pathways through acute services or patient care itself.
We aimed to describe the current structure of medical SDEC services within the UK, using data from the Society for Acute Medicine Benchmarking Audit (SAMBA), a day-of-care survey that evaluates the structure and performance of acute medicine services within the UK. We also aimed to assess variation in the utilisation of the SDEC pathway for planned and unplanned medical attendances, evaluating patient factors associated with management through same day services.
Methods
Data were collected through SAMBA. Participation in SAMBA is voluntary and open to all hospitals accepting unplanned admissions to acute and/or general internal medicine; community hospitals were excluded. Multiple hospital sites can register from each Trust/Health Board or equivalent, using the Society for Acute Medicine (SAM) website. The full protocol for SAMBA is available online.9 Study data were collected and managed using the CaseCapture database (NetSolving) for 2019–2021 and REDCap electronic data capture tools hosted at the University of Birmingham for 2022–2023.10 There was no specific patient and/or public involvement in the design, conduct, or reporting of SAMBA.
Each round of SAMBA includes patient-level data for all medical admissions over a single 24-hour period on the penultimate Thursday of June (with the exception of winter SAMBA in January 2020), in conjunction with a unit-level organisational survey describing acute service structure. Comparison of performance against clinical quality indicators for acute medicine using patient-level data is published elsewhere.11–14
The organisational survey collected data detailing hospital size (including number of beds on the AMU and total number of inpatient beds); in addition, SAMBA22 contained more detailed questions regarding medical SDEC services, designed to assess adherence to national recommendations from SAM and the Royal College of Physicians of Edinburgh (RCPE), and describe service availability (online supplemental appendix 1).15 Conditions included as possible condition-specific ambulatory pathways were selected from guidance regarding ambulatory care sensitive conditions,16 the AEC directory4 and service evaluations presented at SAM conferences over the preceding 12 months.
Supplemental material
The terminology used to refer to SDEC changed across the time period covered by this dataset. AEC and SDEC services, as they relate to medical patients, are assumed to be equivalent, and the terms have been used interchangeably across the data collection period.
Patient-level data was combined from SAMBA19 (20 June 2019), Winter SAMBA20 (23 January 2020), SAMBA21 (17 June 2021), SAMBA22 (23 June 2022) and SAMBA23 (22 June 2023). Data was aggregated to allow comparison where levels of measurement had changed over time, for example, increased granularity recorded within the ‘time to consultant review’ measure in more recent data collection.
All patient-level data submitted to SAMBA is anonymised. Patient-level data was divided into planned and unplanned attendances. Age (in bands) and gender are recorded for both planned and unplanned attendances. Reason for attendance was available for planned attendances in 2022/2023. For unplanned attendances, the location of initial clinical assessment and first assessment by the medical team was recorded, categorised as emergency department (ED), AMU, SDEC or other locations. Records where neither the location of initial assessment nor medical assessment was documented were excluded from analysis (63 patients, 0.2% of unplanned admissions in the dataset). Patients presenting to non-standard units (frailty units (seven services), specialist cancer centres (one service) and stand-alone SDEC services (one service)) were excluded from analysis as patient pathways into and through these services are likely to be different from standard acute medicine services.17
Data was analysed using Microsoft Excel and STATA 16/18 (StataCorp, College Station, Texas, USA). Descriptive statistics were used to summarise results, with group comparisons made using χ2 test, or Fisher’s exact test where expected cell counts were <5. For group comparisons of continuous variables, Kruskal-Wallis and Mann-Whitney U tests were used for data that was not normally distributed. For analysis of survey data describing SDEC services included only in SAMBA22, participating units were stratified based on hospital size, into three groups: smaller (<400 inpatient beds, 41 hospitals), medium (400–599 inpatient beds, 47 hospitals) and larger (≥600 inpatient beds, 52 hospitals). Where questions regarding organisational structure were asked in only selected rounds of data collection, any associated comparison to patient-level data is limited to those years. Correlation between variables that were not normally distributed was assessed using Spearman’s rank correlation. Comparison between attendance numbers and SDEC and AMU size was performed using data from SAMBA23 only. Logistic regression using backwards selection was used to assess factors affecting the likelihood of meeting the same-day discharge target (one-third of admissions discharged without inpatient admission, outlined in NHS Long Term Plan)7 using data from hospitals that participated in all rounds of data collection 2021–2023; ORs and CIs are reported. A p value of <0.05 was considered statistically significant throughout.
Results
Organisational survey
Of the 149 UK hospitals that participated in SAMBA22, 140 hospitals responded to the organisational questions regarding SDEC; this included 122 hospitals from England (87% of participating hospitals), 8 from Scotland (5.7%), 6 from Wales (4.3%) and 4 from Northern Ireland (2.9%) (online supplemental table 1). Hospital size ranged from 0 to 1700 inpatient beds (median 520, IQR 369–688), with one stand-alone SDEC service at a site with no inpatient beds. Only one hospital that responded did not have a medical SDEC service. SDEC units were physically separate from the AMU in 89% (120 units).
Supplemental material
SDEC units were open for a median of 12 hours (IQR 11–14 hours, range 4–24 hours) although nine SDEC units (6.5%) were open 24 hours a day.
Recommended standards
Comparison to recommended SAM/RCPE standards for SDEC is shown in online supplemental table 2. Overall, a consultant was physically available throughout SDEC opening hours in only 64% (89 units). There was no difference in the proportion of unplanned attendances assessed in SDEC comparing those with a consultant available to those without (Mann-Whitney U, p=0.867); however, when comparing SDEC discharge rates, more units that had a consultant available discharged over 80% of patients assessed in SDEC the same day (72% vs 45%, χ2 p=0.007).
A nominated clinician had overall leadership of SDEC in 85% (118 units); of these, 87% (103) were consultant physicians, 3.4% (4) were nurses, 5.9% (7) were advanced clinical practitioners (ACPs) and 3.4% (4) were specialist or specialty doctors. Compliance with standards for SDEC did not vary with hospital size, including availability of private consultation areas, patient feedback collection and contact with non-attenders.
Overall, 113 SDEC units (81%) had a Standard Operating Procedure (SOP), 17 (12%) did not and 9 (6.5%) were unsure; this did not vary with hospital size (p=0.51).
Pathways into SDEC services
Triage
85 hospitals (61%) used a specific scoring system to assist in identifying patients suitable for treatment in SDEC; 42% (58 hospitals) used NEWS2,18 11% (15 hospitals) used centre-specific criteria, 5.8% (8 hospitals) used the Amb score,19 2.2% (3 hospitals) used the Glasgow Admission Prediction Score (GAPS)20 and 0.7% (one hospital) used the Clinical Frailty Scale (CFS).21
There was no difference in the proportion of unplanned attendances assessed in SDEC comparing those that used a scoring system and those that did not (Mann-Whitney U, p=0.075), and no difference in the proportion of patients assessed in SDEC who were discharged the same day when comparing hospitals that did not use a criteria to those using NEWS, or an Amb or GAPS score (Kruskal-Wallis, p=0.162).
34 SDEC units (24%) exclusively saw patients attending with specific conditions (eg, deep vein thrombosis (DVT)) or on protocolised pathways; of these, 28 (82%) had an SOP. Comparing these units to those that were not limited to specific conditions/pathways, there was no difference in the proportion of unplanned admissions seen in SDEC (Mann-Whitney U, p=0.247), the proportion of all unplanned attendances discharged without overnight admission (p=0.593), or the proportion of patients seen in SDEC discharged without overnight admission (p=0.157).
32 SDEC units (23%) did not accept patients that required assistance with mobility, 16 units (12%) did not accept patients confined to a chair and 104 units (75%) did not accept patients confined to a bed.
Referral
Accepted referral sources are shown in table 1. Most units (82%) accepted patients to SDEC without requiring prior full clinical review; of these, 98 (86%) had an SOP. Most units (51%, 71 units) accepted referrals from the ED without discussion with the medical team. Of the 34 units accepting only selected conditions/pathways, 82% (24 units) accepted these patients from ED without assessment by an emergency medicine (EM) clinician.
Form of referral accepted by same day emergency care units
SDEC services available
Specific conditions
Of the prespecified conditions included, condition-specific ambulatory pathways were most common for pulmonary embolism (PE) and DVT, both overall and when stratified by hospital size (table 2). Nine SDEC units (6.5%) had no condition-specific ambulatory pathways available, and an additional three units (2.1%) did not report provision of any ambulatory pathways for the conditions listed. Provision of condition-specific pathways was similar when stratifying by hospital size; larger hospitals were more likely to have guidelines for ambulatory management of papilloedema.
The provision of condition-specific ambulatory pathways in same day emergency care units
Planned reattendance
99% of units (136/138) enabled booked patients to return to SDEC, used by acute medicine (98.5%, 134 units), EM (79%, 108 units) and inpatient medical wards (63%, 86 units). Returning patients were booked to timeslots in 72% of units (97/135). 82% of those offering booked return (111/136) had an SDEC SOP.
Services offered through planned SDEC reattendance varied (table 3). Specialty review was the least common service, offered in 72% (100 units).
Services available for planned patients returning to same day emergency care
Patient-level data
Patient-level data was available for 39 722 patients, of which 4342 (11%) were planned reattendances to SDEC services. Excluding patients presenting to non-standard units and those without assessment location data available, data for 34 948 unplanned attendances from 188 hospitals were available for analysis (online supplemental figure 1). Distribution by UK nation is shown in online supplemental table 1.
Unplanned attendances
The proportion of unplanned admissions receiving their initial medical team assessment in SDEC has increased over time (figure 1), with 30% of unplanned attendances in SAMBA23 (2403 patients), assessed in SDEC services (online supplemental table 3, χ2 p<0.005).
Comparison of unplanned attendances and same day discharges by year. SDEC, same day emergency care.
The proportion of patients receiving medical assessment in SDEC services varied between hospitals (figure 2a, median 22%, IQR 12%–32%, range 0%–71%).
Comparison of performance by site. (a) Proportion of unplanned attendances receiving their medical assessment within same day emergency care (SDEC) services by site. Units ordered along the x-axis by proportion assessed within SDEC. Median unit performance: 21.8%. (b) Proportion of patients receiving medical assessment in SDEC who were discharged same day, by unit. Units ranked along x-axis by performance. Median unit performance: 85.7%.
A higher proportion of patients in England (26%) were assessed within SDEC (or their counterpart) services than in other UK nations (Scotland 10%, Wales 16%, Northern Ireland 6.7%, p<0.005). Among English sites, units assessed a median of 25% of patients in SDEC (IQR 15%–33%, range 0%–71%).
Overall, 18% of patients receiving their medical assessment in SDEC had been assessed in another location prior to this; this was lower in 2023 (15%) compared with previous years (2019: 19%, 2020: 17%, 2021: 19%, 2022: 19%, χ2 p<0.001).
Patient characteristics by assessment location
Patient demographics and acuity (assessed by NEWS score) were compared for patients receiving medical assessment in SDEC compared with those assessed in the ED, AMU and other locations (table 4). A higher proportion of female patients received medical assessment in SDEC compared with male patients (26% vs 22%, p<0.005). A lower proportion of those assessed in SDEC were aged ≥70 years compared with those assessed in other locations (30% of patients assessed in SDEC, 54% average across other locations, χ2 p<0.005). This also equates to a lower proportion of patients aged over 70 receiving their initial assessment in SDEC compared with those aged under 70 (15% vs 33%).
Patient characteristics by location of first assessment by the medical team
Care home residents were less likely to have received initial medical assessment in SDEC than patients living at home or in other settings (4.1% vs 26%, p<0.005). In those aged over 70 years, 7.4% of patients with CFS ≥5 were seen in SDEC, compared with 30% of those with a CFS 1–4 (p<0.005). Patients who had been in hospital in the preceding 30 days were also less likely to have been assessed in SDEC (17% vs 26%, p<0.005).
Only 7.6% of patients seen within SDEC services had a NEWS score of ≥3 on arrival to hospital, compared with 38% of those assessed in the ED and 31% of those assessed on AMU (χ2 p<0.005).
Patient characteristics over time
Patient demographics and acuity in patients receiving medical assessment within SDEC services were compared across data collection periods (online supplemental table 4); there was no significant change in the proportion of patients seen within SDEC services that were aged over 70 years, female or care home residents.
There was no significant difference in the NEWS scores of patients seen within SDEC services when comparing the summer data collection periods—a higher proportion of patients assessed in SDEC services in the January 2020 cohort had a NEWS score ≥3 compared with the summer cohorts (p=0.01).
Adjusting for time of arrival and source of referral, patients who were care home residents, male or had been recently discharged from the hospital were less likely to have been assessed within SDEC services (table 5). Odds of assessment within SDEC services reduced with increasing age (from the age of 50 onwards) and increasing NEWS score. The likelihood of assessment in SDEC was higher in recent time periods in comparison to 2019.
Logistic regression model for likelihood of medical assessment within same day emergency care services
Outcomes
Outcome data within 7 days was available for 34 621 patients. Six patients (0.1%) receiving medical assessment in SDEC died within 7 days.
Same day discharge
Overall, 31% of all patients referred to AIM were discharged without overnight admission; 82% of patients receiving medical assessment in SDEC were discharged without overnight admission (online supplemental table 5). The proportion of patients discharged without overnight admission was highest in 2023 (χ2 p<0.001). Likelihood of same-day discharge was associated with patient factors similar to those influencing likelihood of assessment within SDEC services (online supplemental table 6).
The likelihood of meeting the NHS target for one-third of patients being discharged same day was assessed for hospitals participating in all rounds of 2021–2023 (115 hospitals). The odds of a hospital meeting the target of one third increased with increasing number of unplanned admissions and an increased percentage of patients that were daytime arrivals or were general practice referrals, and decreased with increasing percentage of patients that were aged over 70 or had a NEWS2 ≥3 (table 6).
Logistic regression assessing likelihood of hospital meeting target for same day discharges
3672 patients discharged same day received their medical assessment in non-SDEC locations; this was 35% of those discharged without overnight admission, and 11% of patients overall. This has decreased over time (figure 1; 2019: 39%, 2020: 37%, 2021: 40%, 2022: 33%, 2023: 27%, χ2 p<0.001). Those discharged same day from other locations received their medical assessment in ED (54%), AMU (40%) and other locations (6.1%).
SDEC discharge rates
The proportion of unplanned attendances assessed in SDEC who were discharged same day varied between units (figure 2b, median 86%, IQR 75%–93%, range 0%–100%). Comparison by time period is shown in figure 1 and online supplemental table 7. In the most recent data collection, the conversion rate to inpatient admission was <20% in most units (69%, 97 units), 20%–30% in 12% (17 units) and >30% in 19% (26 units).
Data regarding unplanned re-attendances within 7 days was available for 6625 unplanned attendances who received medical assessment in SDEC and were discharged same day; 4.7% (309 patients) had an unplanned reattendance. This was to ED (148 patients), AMU (23 patients), SDEC (137 patients) and/or other locations (18 patients).
Planned reattendances
Of the 4342 patients included as a planned reattendance, reason for this reattendance was recorded for 2450 (SAMBA22 and SAMBA23 only); 36% were ≥70 years, and 56% were female. Clinical review was the most common reason for reattendance (1163 patients, 48%), followed by repeat blood tests (642, 26%), DVT investigation/treatment (456, 19%), imaging (432, 18%), intravenous antibiotics (209, 8.5%), non-antibiotic intravenous medication (201, 8.2%) and ambulatory PE diagnosis/treatment (143, 5.8%). 90% were assessed by a clinician during this reattendance (2201/2445), with initial assessment by doctor more junior than registrar in 32% (686/2182), registrar in 26% (566), ANP/ACP in 24% (515), consultant in 13% (280) and PA in 2.5% (56).
92.6% of planned reattendances were discharged home without overnight admission (2229/2407), 3.6% were discharged within the next 7 days, 1.0% (25) discharged against medical advice and 0.04% (one patient) died in hospital within 7 days.
Attendances and SDEC size
Number of planned and unplanned attendances seen within SDEC services was compared with hospital size, AMU size and number of assessment spaces on AMU using data from SAMBA23. For 143 hospitals with this data available, the number of unplanned attendances receiving their medical assessment in SDEC ranged from 0 to 61 (median 14, IQR 6–21), with 16 units (11%) reporting that no patients received their medical assessment in SDEC on SAMBA day. The number of planned reattendances seen in SDEC ranged from 0 to 52 (median 7, IQR 3–12), with 18 units (13%) reporting no planned reattenders on the day of data collection. The proportion of patients seen within SDEC that were planned returners ranged from 0% to 100% (median 35%, IQR 20%–51%).
There was a moderate positive correlation between the number of planned reattenders and the number of unplanned attendances receiving medical review in SDEC (r=0.46, p<0.005), and between the number of unplanned admissions and the proportion of unplanned admissions assessed within SDEC (r=0.42, p<0.005).
SDEC services had a median number of ten assessment spaces (IQR 6–13). There was no significant correlation between the number of SDEC assessment spaces and size of hospital (assessed by number of inpatient beds, r=0.12, p=0.16) or number of beds on the AMU (r=0.07, p=0.38); however, there was a moderate correlation between number of AMU beds and hospital size (r=0.51, p<0.005).
Although there was a moderate correlation between the number of unplanned attendances and both hospital size (r=0.59, p<0.005) and number of beds on AMU (r=0.45, p<0.005), the number of SDEC assessment spaces available was very weakly correlated with the total number of unplanned attendances (r=0.17, p=0.046), and weakly correlated with the number of unplanned attendances receiving medical assessment in SDEC (r=0.21, p=0.013). There was no significant association between the number of planned reattenders and number of SDEC assessment spaces (r=0.13, p=0.14), AMU beds (r=0.19, p=0.02) or inpatient beds (r=0.18, p=0.03).
Opening hours and number of assessment spaces were used to calculate the total space available daily (136 units with available data). The number of patients that would require assessment within SDEC to achieve the recommended discharge rate of one third was calculated from the total number of unplanned admissions seen, assuming a 20% conversion rate from SDEC assessment. Each patient was assumed to require 1 hour of assessment space. Two units (1.5%) did not have enough space available to assess the required patient numbers.
Discussion
Medical SDEC services are a key component of assessment pathways for acute medical patients within the UK; however, there is considerable variation in the provision of these services, in terms of both operational factors and clinical delivery.
The proportion of medical patients assessed through SDEC services has increased over the last 5 years, but considerable variation remains between hospitals and between UK nations. Although NHS England has consistently advocated for increased use of SDEC,6 7 22 23 and NHS Scotland has supported its counterpart Rapid Assessment and Discharge, there has arguably been slower adoption in Wales and Northern Ireland.24 This may have influenced the higher proportion of patients seen in SDEC services within English hospitals in this study.
Whether an individual hospital met the ‘Long Term Plan’ target7 that one third of patients be discharged without overnight admission was influenced by the patient population, including the proportion of patients aged over 70 and with high acuity (assessed by NEWS2 score). Hospitals with more unplanned medical attendances were more likely to achieve the target; increased attendances have prompted higher SDEC activity to maintain patient flow and mitigate service pressures. The factors identified here as relating to the achievement of target discharge rates are not amenable to alteration by hospitals, but may provide an understanding of why individual services may not achieve the target, and support approaches targeting improved delivery of SDEC and provision of community services for older adults.
Most SDEC services (69%) had a conversion rate to inpatient admission of less than 20%; only 12% of units had a conversion rate of 20%–30% as recommended in the SAMEDAY strategy.23 There is little evidence available assessing the optimum SDEC conversion rate; accepting patients into SDEC services with a higher chance of admission may impact the ability to deliver assessment to low-risk patients due to the limited space available, while some patients discharged without admission may have been suitable for assessment through other care pathways, such as primary care, community or outpatient services. Challenges in access to these services, perceived or genuine, may contribute to the increase in patients seen through SDEC services; greater understanding of these factors is likely to be beneficial to UEC services but is beyond the scope of this study.
Our results suggest there is low mortality in patients currently assessed through SDEC services and discharged without overnight admission (0.1% within 7 days), with 5% of patients discharged after SDEC assessment having an unplanned reattendance within 7 days. The methodology used does not allow further exploration of mortality cases, and data could not be linked to external sources to confirm mortality or 30-day readmission rates.
Services had an average of 10 spaces available for patient assessment within SDEC; although the size of a hospital’s AMU had some correlation to hospital size and the number of unplanned admissions seen daily, there was only weak association between these factors and the space available for SDEC suggesting the SDEC footprint was not mapped against potential demand. Only 1.5% of units were estimated to lack the necessary space to assess the patient numbers required to meet the current discharge targets; however, this assumes availability of staffing, and uniform flow of patients into the service. There is currently no guidance describing how to calculate the space needed to deliver SDEC effectively based on expected or desired patient flow, and greater understanding of optimal physical set-up, including size, is needed. Our results suggest greater physical space alone currently does not equate to increased clinical activity. There was some association between the number of planned and unplanned attendances seen within SDEC, suggesting services seeing more unplanned attendances also provide more scheduled care delivery. While this may reflect greater performance driving increases in both, it may be that increased levels of scheduled care are required to facilitate the delivery of unscheduled care through SDEC.25
It is likely that some patients suitable for SDEC are still receiving medical assessment in other locations: almost 30% of patients discharged without overnight admission were assessed in non-SDEC locations. Many of these may have been suitable for SDEC; however, a zero-day length of stay should not be assumed to equate to suitability for SDEC. Correct identification of suitable patients is vital for effective delivery of SDEC.18 Over 80% of SDECs accept patients from ED triage without full clinician review, necessitating a robust process to ensure that only patients likely to be discharged after medical team intervention are directed through SDEC services. Inappropriate identification of patients could result in delayed delivery of care to those requiring inpatient care or more appropriately managed through other pathways, as well as reducing the ability to deliver SDEC to those who are suitable. An understanding of how these processes can be operationalised and delivered effectively and the impact on patient outcomes is needed.
Over a third of SDEC units did not use a screening tool to identify suitable patients. This may make patient selection more subjective; however, use of a screening tool was not associated with higher rates of SDEC assessment or same day discharge in our results. NEWS2, which identifies patients at high risk of impending deterioration, was the most common tool used to identify suitable patients, although cut-offs used were not recorded.18 26 NHS Improvement suggests only those with NEWS <4 be considered for SDEC due to potential clinical instability, but not all deterioration is preceded by a raised NEWS, and use without consideration of other factors may not prevent inappropriate referrals.2 27 Scores incorporating additional features have been suggested,28 including the Amb score and GAPS19 20; however, their discriminatory ability appears to be lower when applied outside the original setting.29–31
Patient factors such as age and recent hospital attendance, that feature in these scoring systems, were associated with decreased likelihood of assessment within SDEC services, and of same day discharge, within our analysis. This suggests these features are linked to suitability for SDEC, although this may reflect the availability of services that facilitate discharge in selected patient groups. Although the likelihood of receiving assessment within SDEC was influenced by the patient factors discussed here, other features not included within the patient-level data may play a role. There are likely to be significant barriers to medical SDEC services for those with reduced mobility; a quarter of units did not accept patients requiring assistance with mobility and 1 in 10 did not accept patients confined to a chair. Our results suggest that older patients were less likely to be assessed through SDEC services, and less likely to be discharged without inpatient admission; previous analysis suggested assessment in SDEC was less common in those with frailty or presenting with a geriatric syndrome.32 Although this may disproportionately impact older adults and those with frailty, these patients may now be supported by the increased emphasis on frailty SDEC services.17 33 How these services are delivered and interact with medical SDEC, including the clinical conditions amenable to management through these services, requires further evaluation.
Condition-specific pathways can improve patient outcomes and reduce cost, with multiple acute medical conditions suggested as suitable for management through SDEC.34 35 Despite this, aside from DVT and PE, many SDEC units lacked condition-specific ambulatory pathways. While this risks inconsistent service provision within and between SDEC units, there is limited evidence regarding how condition-specific SDEC pathways may impact quality of care, delays in management, resource utilisation and patient experience.
Adherence to SDEC standards recommended by SAM and RCPE is variable.15 More than a third of participating services did not have a consultant physically available throughout operational hours. This may impact care delivery by introducing delays or inefficiencies when junior clinical staff require input in complex cases; our results suggest units with consultant presence were more likely to discharge a high proportion (>80%) of the patients assessed in SDEC. However, the SAMEDAY strategy now recommends a more lenient target, suggesting ‘access to an appropriate consultant’ as a minimum requirement23; how the recommended standards can be delivered in practice, and barriers such as workforce limitations, should be explored. Further understanding of how consultant physician availability may affect operational factors within SDEC services would be helpful to understand the relationship between senior workforce availability and discharge rates.
Although there are recommended standards, the operational protocols and pathways used within SDEC units remain the responsibility of local teams, allowing services to be tailored to the requirements of local pressures and populations. Standard operational policies are recommended to ensure SDEC is not used for patients that would be more appropriately managed through alternative pathways or inpatient care36; 12% of units did not have an SOP, and an additional 7% were unsure. Similar figures were seen in services accepting ED triage referrals without full clinician review, and services allowing booked patient reattendance, where variation may increase risk and a more structured approach may be beneficial. There is concern nationally that SDEC services may be used to house patients that have spent prolonged periods in the ED prior to transfer, including those awaiting diagnostic test results and where medical physician input is not required25 36; however, evaluation of this was beyond the scope of this study.
Across the 48 hours of SAMBA22 and SAMBA23, almost 2500 patients were seen in SDEC as a planned reattendance, most commonly for clinical review. Despite the large numbers seen through this route, there is limited guidance describing how planned reattendance to SDEC services should be used to facilitate discharge outside of specific conditions, such as PE,37 and little evidence evaluating how these attendances impact patient care, acute medicine resource use and service pressures.
This study represents the largest analysis of SDEC services to date, providing evaluation at both patient and unit level that has not been previously reported. There are currently no other multicentre studies evaluating the delivery of SDEC, and the data reported here is not available through any routinely collected data.38 There are approximately 250 AMUs within the UK, although the reported number varies and fluctuates, in part due to frequent changes such as mergers between hospital sites39; 80% of services contributed data that has been included in this analysis, however, an estimated 57% of units provided detailed information describing unit structure in SAMBA22. There may be systematic differences in those hospitals that did not participate; the higher response rate from English services (online supplemental table 1) means the results may be less reflective of practice in the other nations, which have different policy approaches to SDEC.24 40 Our data represents a single day within each year, and variation in performance may be expected across time.
All data regarding the availability of services and unit structure was self-reported, and therefore at risk of bias. We assume that if the clinical team is unaware of pathways and standard procedures available, then they were not being used. Not all suggested standards were evaluated; additional metrics including clinician assessment of patients within an hour of arrival to SDEC, utilisation of validated risk stratification tools for specific conditions, and regular review of SDEC performance using predefined metrics should be evaluated in future audit. SDEC services are often delivered through a multi-professional workforce, including advanced practitioners, that is key to processes from patient selection to senior clinical review. Comparison between staffing models was not evaluated in this current study; however, further evaluation to support and guide workforce development and staffing recommendations would be beneficial.
Our analysis did not show any significant difference in the provision of hospital services when stratifying by hospital size. For the purposes of our analysis presented here, hospitals with less than 400 beds were grouped as ‘smaller’,41 due to the low number of small hospitals in this sample. There may be differences in the delivery of care at hospitals that are small or rural,42 and these hospitals have different demands, access to specialist services and pathways, and logistic differences, such as patient travel time, that may affect how SDEC is delivered. Further in-depth evaluation of how SDEC services currently function in these specific settings may provide helpful insights into effective operation when influenced by these factors, but it requires an alternative methodology to that reported here.
Further focused research is needed to ensure effective delivery of medical SDEC, and equity of access across hospitals and patient cohorts. This will necessitate robust studies of multiple aspects of service organisation, alongside prospective evaluation of outcomes for patients assessed in SDEC, building an evidence base for medical SDEC which can inform more comprehensive guidance and policy from key groups, including SAM and the NHS.
Conclusion
Medical SDEC services continue to be a key component of assessment pathways within acute medical services, with a third of unplanned medical attendances managed through these pathways. There is considerable variation in the provision of these services nationally, in both operational factors and clinical delivery. Further evaluation is needed to understand how SDEC services can be more effectively delivered across different patient populations and hospital settings, and to ensure patients can receive care within the most appropriate setting.
Data availability statement
Data are available upon reasonable request. Data from this study are available from PIONEER, the Health Data Hub in Acute care, in accordance with Hub processes. See www.pioneerdatahub.co.uk and contact PIONEER@uhb.nhs.uk for more details.
Ethics statements
Patient consent for publication
Ethics approval
Local approvals were obtained by individual sites, including Caldicott Guardian approval. Health Research Authority approval has been granted to allow secondary analysis on non-identifiable data (REC 21/HRA/4196).
Acknowledgments
The authors would like to acknowledge those that have assisted with the running of SAMBA, and those at the sites that participated in SAMBA.
Footnotes
X @catatkin
Contributors CA and MP designed and conducted data analysis and drafted the initial manuscript. CA, TK, TC, MH, CS, DSL and RV contributed to design of initial data collection. All authors contributed to and approved the final manuscript. CA is the guarantor.
Funding No specific funding was received for the study as reported here. The database used for SAMBA data collection is funded by the Society for Acute Medicine. CA reports funding from the NIHR. ES reports funding support from Health Data Research UK, MRC, Wellcome Trust, NIHR, Alpha 1 Foundation, EPSRC and British Lung Foundation, and is funded by the National Institute for Health and Care Research (NIHR) Midlands Patient Safety Research Collaboration (PSRC). This study was supported by the National Institute for Health Research (NIHR) Applied Research Collaboration (ARC) West Midlands, the NIHR Oxford Biomedical Research Centre (BRC) and NIHR HealthTech Research Centre (HRC) for Community Healthcare through salary support to DSL. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.