Sage Journals: Discover world-class research

Abstract

Background:

Mismatch in admission and discharge rates with poor coordination of bed availability to prepare for peak admission time do not favor patient flow in any healthcare situation. In order to meet the admissions demand from the emergency department, early discharge is advocated.

Aim:

The purpose of this study was to increase the 11:30 discharge rate in the acute medical ward to 20% within 5 months in order to facilitate patient transfers from the emergency department.

Method and intervention:

A nurse-led multidisciplinary team discussion at 10:00 was formed empowering nurses and aims to prioritize early discharges or transfer patients who require a longer stay in the hospital. The indication of expected discharge date was determined by team doctors on either day 1 or 2 of admission, to facilitate and prioritize allied health intervention or even prescription processing. Physical transfer to the discharge lounge was to occur earlier, awaiting preparation of prescription and documentation. Dedicated nursing staff who manage the discharge lounge processes, were to ensure the transfers and care arrangements to be smooth and coordinated with enhanced communication to next of kin.

Results:

The overall median rate for discharges before 11:30 in the acute medical ward was at 17%, compared to baseline median discharge rate of 12% (p=0.05) and evening discharges that remained. With the implementation of a framework for multidisciplinary team discussion, the mean time spent on multidisciplinary team reduced to 40 min from the pre-intervention stage, ranging approximately 57–68 min.

Conclusion:

The discharge process is complex and requires multidisciplinary team collaboration and coordination. This project also created an opportunity for more relevant project targeting evening discharges that addresses another emergency department peak admission time.

Keywords

Multidisciplinary team early discharge discharge lounge

Introduction

Patient flow in hospital has been well studied to facilitate early discharges. Acute medical units (AMUs) were designed for early patient care and the success has been measured in length of stay and readmissions.¹ Evidence supports that various models of AMU designed to fit to individual hospital setting able to achieve desirable outcome.

Early discharges are often seen as a mean of ensuring accessibility to treatment by enhancing bed turnovers. In the Acute Medical Ward (AMW) in Singapore General Hospital (SGH), we are faced with inadequate morning discharges due to many inherent workflows that involve various roles of healthcare professionals.²

The AMW was first established for rapid turnover of acute medical admissions with infection-related conditions and an expected length of stay of 72 h. The 67-bedded ward is managed under the Department of Internal Medicine, the largest department in the hospital managing about 300 inflight patients at the wards at any one time. This ward manages over 5000 cases a year and consists of multi-disciplinary teams (MDTs) to provide comprehensive assessment and management of both medical illness and functional disability. Providing early and rapid medical assessments, timely diagnostic evaluations and treatments so that efficient and effective treatments are given to all patients through appropriate care are the key features of AMW. It serves the busy emergency department (ED) with high patient load for the preselected cases based on admission diagnosis.^3,4

Patients who require a longer stay will be transferred to various other wards with the help of the bed management unit. While early discharges can be achieved for elective admissions, acute medical admission is consistently not meeting the expected target due to varied reasons. Data collected from November–December 2017 showed that the median 11:30 discharge rate was at 12%, whereas the national benchmark was at 30%.⁵ This project was aimed to increase 11:30 discharge rate at the AMW to 20% within 5 months in order to facilitate patient transfers from the ED. The aim was set at 20% as the team felt it was a reasonable target to achieve and all changes could only be focused for the weekday as a start. Holistic intervention for the entire discharge process is beyond the scope of this project.

This article is prepared based on the Standards for QUality Improvement Reporting Excellence (SQUIRE) 2.0 guidelines which is the standard guide for quality improvement reporting excellence.⁶

Methods

Baseline measurement

To understand the problem better, the patient flow from the ED to the AMW and within the AMW was analyzed to identify the causes for the relatively low noon (12:00) discharge rate to develop the proposed interventions (Table 1). The analysis showed a delay in care coordination by various healthcare workers was the crucial and rate limiting step for early discharge. Nursing is the key player in this process of patient disposition and prioritizing decision-making either for discharge from the patient bed or discharge lounge (DL) considering patient ambulation status. They also ensured that prescriptions were prepared in a timely manner for pharmacists to process and dispense medication, delivered education of feeding or wound care, assumed the role of the key communicator to next of kin and liaised with other allied healthcare workers for necessary support needed. Such coordination, delivery of care and treatment need to be completed timely to meet the rapid turnover at AMW, to achieve the targeted discharge time.

Table 1.

Patient flow interventions.

Solutions	Validation
Team doctors to identify EDD and communicate to MDT	Ensuring all members of the healthcare team are "on the same page" and working towards the identified discharge goal.
Junior doctors to submit electronic prescriptions during rounds	To expedite discharge process and facilitating pharmacist to perform medication packing early and reduce delay in discharging patients.
Therapist to review patients early	To allow early intervention for home safety for patients to discharge in a timely manner.
Identify information necessary for the MDT	To reduce time taken to conduct the MDT and allow the team to allocate the remaining time for expediting discharge process. The discussion during the MDT aims to identify any potential discharge issues.
Communicate discharge plans to all members of team	To increase awareness on issues to be addressed by various healthcare personel to facilitate discharge e.g. caregiver training for insulin injections.
Tailor operational hours of the DL to the peak discharge hours	To encourage patients to use the DL whilst waiting for their next of kin to arrive without being charged for an extra stay.

DL: discharge lounge; EDD: estimated discharge date; MDT: multidisciplinary team.

Rates of discharges completed before 11:30 were measured from November–December 2017. The median discharge rate was at 12%, against the national benchmark of 30%. The team performed a fishbone analysis and also value-stream mapping to understand the processes focusing on a day before and the day of discharge. As the AMW had a short-stay functionality, the expected discharge date (EDD) was focused to achieve the outcome. As part of analysis of the interventions DL utilization was also tracked. The time taken for MDT discussion was also monitored as part of improvement to measure the success of the framework for discussion as it occurs at a crucial time for patient transfer to start the discharge process and shorter MDT discussion time allows more time to complete other processes.

Design

The AMW received acute medical admissions from the ED only, with infection-related diagnosis as the admitting diagnosis. This ward was originally launched as a pilot ward for rapid turnover with only infection-related conditions being admitted. As a tertiary hospital, patients with multiple co-existing medical conditions were expected and the majority of patients were above 75 years old.

This case study was conducted in Singapore General Hospital. Newly admitted patients in the AMW were identified by nurses and tracked until the time the patient left the ward for discharge during the period from December 2017–April 2018. The DL was created by modifying the day room during the initial launch of the AMW in 2015.

Variants such as discharge planning and discharge orders in the electronic progress notes were monitored for all newly admitted patients until their discharges. Electronic prescription (Erx) submissions were also tracked and aimed to have been completed by 10:00. Other factors, such as referrals to allied health professionals (e.g. physiotherapists and occupational therapists) to certify patients to be fit for discharge, were also evaluated on the effects of actual discharge time. These variants were monitored against the daily discharge rates.

The study was considered a quality improvement activity in which personal identifiable data were not revealed and participant anonymity was maintained. A convenience sample of patients who were admitted to a large tertiary teaching hospital in Singapore were included in this study.

Data analysis

The team members brainstormed possible causes of patients discharging after 11:30 using the cause and effect diagram (Figure 1). Team members analyzed the root causes and voted to select the vital causes. In addition, a survey was conducted to solicit more reasons from stakeholders on factors affecting discharge time. The vital few root causes were identified through multi-voting on the fine details in the cause and effect diagram.

Figure 1.

Cause and effect diagram showing delayed discharges after 11:30.

Using the 80/20 principle, the following root causes were discovered to be responsible for 80% of the problems. The top four root causes were (a) multi-disciplinary rounds can be disorganized and time-consuming; (b) electronic prescriptions were only submitted in the afternoon which caused a bottleneck; (c) expected date of discharges were not well established until actual date of discharge; and (d) patients’ preference to rest in bed until it was time to receive discharge documents and medications.

Strategy

Plan-Do-Study-Act (PDSA) 1: Implementation of identification of estimated discharge date (EDD) on admission⁷ and organized MDT discussion

Too many details were needed for handover, ranging from patient’s demographics, care provider to every investigation, were discussed during the MDT discussion that occurred at 10:00. This discussion was attended by doctors, nurses, allied health, and patient navigators. During initial assessment and clerking, doctors would not indicate the EDD. The column on EDD on the MDT board was only filled on the day of discharge. Until the day that the patient was scheduled for discharge, the care team was unaware of planned discharge date and might not have a distinct direction on when to engage family for caregiver training or for the occupational therapist to assess home safety, unless referred by the team’s doctor. This process gap was potentially contributed by the senior doctor who would make decisions on discharges.

Interventions in PDSA 1

A column to indicate the EDD was created on the multidisciplinary board to communicate to all members of the healthcare team about the planned discharge date and resolve any home issues to expedite discharging patients. Doctors in charge were also encouraged to write down the EDD as soon as possible upon patient’s admission.

In this PDSA, nurses were advocated to conduct the MDT discussion based on an agreed framework. The team members discussed and reached a consensus on what should be discussed during the MDT, to create the biggest impact on patient care within the shortest time possible. Patients on comfort measures or on the dangerously ill list would not be discussed as active discharge plans would be unlikely and it would be difficult to determine the EDD at the point of discussion, as well as patients identified for home. This exclusion method allowed more time to expedite the discharge processes across all stakeholders: patient, carers, nurses, doctors, and management.

At pre-intervention, the time spent on MDT ranged approximately 57–68 min while for post-intervention stage the MDT meeting was completed within 40 min.

PDSA 2: Early referral for allied health physiotherapist and occupational therapist to facilitate assessment and timely discharge and modification of operational hours of the DL to facilitate discharges and bed turnover.⁸

Patients requiring assessment safety for home or discharge had to be screened by the physiotherapist and occupational therapist. They were only seen after an electronic referral had been made. This could be delayed at times, as doctors might only initiate the referral right before patient’s discharge, at times prolonging another day’s stay. The team noticed that there was an increasing number of afternoon discharges as there were many patients awaiting blood results and therapists’ assessments.

Interventions in PDSA 2

Doctors were to put up an electronic referral to physiotherapist and occupational therapist on admission once they had identified that patient had functionally declined. While there were additional allied health services in the AMW, the referral time played an important role in discharge timing. Nurses were also empowered to initiate referrals for cases deemed to require therapy assessment during admission orientation or during the inpatient care. The nurse in-charge of DL would actively identify patients who were suitable for DL during MDT discussion and the decision would be agreed with rest of the staff involved in the care of the patient.^9,10

In this PDSA, contact numbers of next-of-kin (NOK) were confirmed during admission and compared against the patient electronic records to ensure notifications of appointment through short message service (SMS). The NOK were able to collect discharge documents any time after discharge at ward level, if discharge documents were completed after 11:30. In addition to this, three clinical nurses who were trained to obtain appointments for the Early Ambulatory Clinic that operated as a discharge-assist clinic, were instructed to do so, especially over weekends when no ward clerk was available to make appointments. One registered nurse and one enrolled nurse were in-charge of managing the DL from 08:00–20:00 on weekdays to facilitate the discharge processes.

PDSA 3: Submission of Erx by 10:00 and early screening by dedicated AMW pharmacists to facilitate packing of medications

Before this initiative, patients were expected to wait for their medications to be delivered, which occurred during peak hours (12:00–15:00). When observed, prescriptions were often submitted after completing entire ward rounds at around 11:30, causing a bottleneck.

Interventions in PDSA 3

Doctors were required to submit the Erx during morning rounds before 10:00 to expedite the process of preparing discharge medications. Senior doctors communicated to junior doctors on the project implementation where the doctors prioritized by submitting Erx early to facilitate bed turnover in AMW. Once junior doctors had submitted the Erx, the nurses would highlight to the pharmacist for them to screen the prescription and clarify any discrepancies with submitting doctors to expedite the discharging medication process. Along with this initiative, prescription submission 1 day before the discharge was also encouraged. Selected patients were allowed to go home with NOK obtaining medication. NOK would collect medications in the evening after work. Free medication delivery was also offered to patients with repeated prescriptions, when they had no medications.

In order to deliver such changes, the operational hours of the DL were modified to start at 08:00 until 20:00 from the original operating hours from 09:00–17:00, to accommodate the evening discharges and address the bottleneck discharge time, thus allowing patients to rest in the DL while freeing up beds.

Results

Descriptive statistics and t-test were performed on the control and intervention group. The overall median of discharges before 11:30 in the AMW was at 17%, compared to the baseline median discharge of 12% (p=0.05) (Figure 2). The median of 15:30 discharges was at 68% compared to baseline 65%. Due to the varied case mix we were unable to show an improvement in the overall length of stay for cases discharged from this ward. General feedback was positive which indicated better nursing empowerment and better communication between doctors and allied health staff through MDT discussion. Before the identified framework, the time spent on the MDT ranged approximately 57–68 min. Post-intervention, MDT discussion was completed within 40 min. Patients who were discharged home or to the DL by 11:30 had potential cost savings of SGD $35–123.08 per patient for extra half-day charge. Solutions that were implemented were constantly validated in order to achieve the target as described (Table 1)

Figure 2.

Run chart demonstrating various Plan-Do-Study-Act (PDSA) interventions and change in median level.

Discussion

The discharge process is complex, and only with good teamwork and coordinated care can the process be executed efficiently while maintaining the safety of the patient. An appropriate platform for such an interface with all healthcare workers is crucial when time is of the essence.¹¹ Governed by the nursing team who play a key role in the discharge process and care coordination, strict rules are inevitable such as attendance at the MDT discussion at a fixed time in order to save time. The MDT discussion takes place at a prime time and potentially interferes with doctors’ rounds, involving bedside education in this teaching hospital, which poses a challenge. Clinical care and education need to be arranged so that there will not be any interference and compromise on the quality of both.

Learning the data from each PDSA was crucial to determine if the appropriate intervention was effective. As this project entailed various changes done concomitantly in order to achieve target, surrogate measures were useful. For example, a 15:30 discharge could potentially be due to a failed 11:30 discharge and thus needs tweaking of workflow which sometimes has to be case-specific. This would encourage the team to look into the granularity and make a specific intervention which might not apply universally. Daily reporting of both 11:30 and 15:30 discharges during morning handovers ensured targets were maintained and acted as reminders in a busy ward with many other quality improvement initiatives running concurrently.

Although this study was conducted in a selected ward with specific diagnoses which was the original design of the ward for rapid turnover, the interventions could be generalized to most acute medical admissions which have similar processes involved as we have described. The set target of 20% or higher could potentially be achieved if all interventions continued even during the weekend with fixed junior staffing and rapid turnover could be maintained with involvement of bed management units emphasized. The cost of running the DL for 12 h with dedicated staff is expected to balance the overall discharge for the day though we were unable to conclude this as the outcome of this particular study.

The biggest challenge was roster planning, especially for junior doctors, which was prioritized based on their level of training and meeting their training requirements. The junior doctors would undergo regular rotations out of the AMW, which posed disruption to the interventions. Their rotation times varied for different levels of trainees and new batches that came in for the foundation year. Constant communication and reminders were needed in order to adhere to timeliness of the MDT and Erx preparation, while being aware of initiatives to enhance communication among the stakeholders. Significant delays were also noted when the changeover dates and also at the beginning of the month when the junior doctors change. Constant reinforcement was required to overcome this.

Weekends were limited by reduced manpower with the same amount of ward work needing to be completed. This posed a challenge to keep the initiatives running smoothly.¹² For example, the MDT is of limited value during the weekend with even lower staff coverage especially from allied health staff that only operate as per need basis. While the ED admission peaks remained constant, variations in ED manpower, physician admitting behaviors and portering services which were thin on weekends would affect patient flow.^13–15 The problem is further compounded during long weekends with lesser services available which would potentially lead to more delayed investigations and appropriate care. The after-effect was noticed on the day after the long weekend. Furthermore, if the ward had considerable numbers of elderly patients that required assistance and assessments by therapists early, this caused them to be unsuitable for transfer to the DL.¹⁶ This is indeed one major impediment to the 11:30 discharge. Vigilant planning for such occurrences is crucial to intervention and requires close collaboration with the ED.

Conclusion

This study identified that communication amongst healthcare professionals and the role of a nurse as a discharge coordinator were instrumental in orchestrating the discharge process. Dedicated nursing staff governing the DL and taking over the discharge processes allowed quick turnover of beds, resulting in a higher rate of successful early discharge. Early discharge rate is aligned to the Ministry of Health’s goal of ensuring access to good and affordable healthcare for all Singaporeans, that is commensurate with their needs. Careful consideration will be necessary to balance the cost of the intervention versus the benefit that could be achieved. More studies and interventions would help to understand the process and details to attain the desired outcomes with continuous effort.

Footnotes

The authors would like to acknowledge the following: Tracy Carol Ayre,Magdalene Ng,Hartini Osman and Ms Nanthakumahrie from the Division of Nursing,Matthew Han from Process Transformation and Improvement (PTI),Grace Lim and Yong Yuen Yan from the Division of Allied Health.

Authors’ contributions

Siti Khadijah Binte Zainuddin: data collection,data analysis,interpretation and drafting of article;Tharmmambal Balakrishnan: data analysis,interpretation,conception,design of the work,drafting of article,critical revision.

Availability of data and materials

Data is kept by both authors and available on reasonable request. Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Conflict of interest

The author(s) declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Ethical approval

Singapore General Hospital does not require ethical approval for reporting individual cases or case series.

Funding

The author(s) received no financial support for the research,authorship,and/or publication of this article.

Informed consent

Not applicable.

ORCID iD

Tharmmambal Balakrishnan

References

Reid

Dinesen

Jones

, et al. The effectiveness and variation of acute medical units: A systematic review. Int J Qual Health Care 2016; 28: 433–446.

Wertheimer

Jacobs

Bailey

, et al. Discharge before noon: An achievable hospital goal. J Hosp Med 2014; 9: 210–214.

Balakrishnan

Teo

EPS

Lim

, et al. Acute medical ward for better care coordination – evidence from a tertiary hospital in Singapore. Acute Med 2017; 16: 169–173.

Epstein

NE.

Multidisciplinary in-hospital teams improve patient outcomes: A review. Surg Neurol Int 2014; 5: S295–S303.

Ministry of Health, Singapore. About us, https://www.moh.gov.sg/about-moh (accessed 28 May 2020).

Ogrinc

Davies

Goodman

, et al SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): Revised publication guidelines from a detailed consensus process. BMJ Qual Saf 2016; 25: 986–992.

De Grood

Blades

Pendharkar

. A review of discharge-prediction processes in acute care hospitals. Healthc Policy 2016; 12: 105–115.

O’Brien

Mitchell

Skinner

, et al. What makes weekend allied health services effective and cost-effective (or not) in acute medical and surgical wards? Perceptions of medical, nursing, and allied health workers. BMC Health Serv Res 2017; 17: 345.

Shaikh

Robinson

Cheet

, et al. Risks predicting prolonged hospital discharge boarding in a regional acute care hospital. BMC Health Serv Res 2018; 18: 59.

10.

Franklin

Vakili

Huckman

, et al. The inpatient discharge lounge as a potential mechanism to mitigate emergency department boarding and crowding. Ann Emerg Med 2020; 75: 704–714.

11.

Okoniewska

Santana

Groshaus

, et al. Barriers to discharge in an acute care medical teaching unit: A qualitative analysis of health providers’ perceptions. J Multidiscip Healthc 2015; 8: 83–89.

12.

McAlister

Youngson

Padwal

, et al. Similar outcomes from weekend discharge. J Hosp Med 2015; 2: 69–74.

13.

Abualenain

Frohna

Shesser

, et al. Emergency department physician-level and hospital-level variation in admission rates. Ann Emerg Med 2013; 61: 638–643.

14.

Shen

Lee

LH.

Improving the wait time to consultation at the emergency department. BMJ Open Qual 2018; 7: e000131.

15.

Paton

Mitra

Considine

Longer time to transfer from the emergency department after bed request is associated with worse outcomes. Emerg Med Australas 2019; 31: 211–215.

16.

Wright

Tan

Iliffe

, et al. The impact of a new emergency admission avoidance system for older people on length of stay and same-day discharges. Age Ageing 2014; 43: 116–121.