Discharge Today: the Efficacy of a Multi-disciplinary Discharge Team

INTRODUCTION

Hospitals frequently face hospital capacity strain.^1,2 Unfortunately, hospital discharges frequently occur in the afternoon or evening hours.³

While some of the delays in discharges result, appropriately, from the caring of other patients, our previous work also identified that providers may be able to prioritize their work in a different way in order to facilitate this throughput.⁴ We created a Multidisciplinary Discharge Team (MDT) to expedite discharges while the care team proceeded with other patient care.

METHODS

The prospective, single-center pilot was conducted from June 5, 2019, to July 25, 2019, Monday through Friday at a 650-bed University hospital and was approved by the Colorado Multiple Institutions Review Board.

The MDT consisted of a Hospitalist APP, a pharmacist, and a discharge nurse (RN). During the first week of the pilot, less complex patients were prioritized for the service; however, that quickly evolved to referrals of any patients on the Hospitalist services. The discharge RN was floor based and the discharge pharmacist was reallocated from existing staff. There was some variability in the availability of the discharge RN. The team coordinated the selected discharge(s) including placing discharge orders, conducting discharge medication reconciliation, preparing and reviewing discharge instructions, and providing patient and family education.

Primary outcomes were (1) time of day the clinician enters the discharge order, (2) time of day the patient leaves the hospital, and (3) average length of stay. Open-ended, unstructured feedback was also obtained.

Statistical Analysis

We estimated means and standard deviations (SDs) for continuous variables when approximately normally distributed, medians and interquartile ranges (IQRs) when not, and frequencies for categorical variables. A Student’s t-test was used to evaluate continuous variables and categorical variables were tested using chi-square tests. Multivariable regression modeling was used to analyze primary and secondary outcomes, adjusting for Charlson Comorbidity Index (CCI).

RESULTS

During a 6-week pilot test, 87 patients were seen by the MDT and of those patients, 77 were discharged on the same day. Comparing patient discharges coordinated by the MDT team to those that were not, more patients were discharged before noon and discharge times were on average 2 h earlier (Table 1). After adjusting for patient acuity using the CCI, we observed a significant reduction in discharge order time and the time of the day the patient actually left the hospital (p <.0001 for both). In addition after adjusting for CCI, we observed a significant increase in the odds of being discharged before 11 am and before noon (p <.0001 for both). While we observed a lower length of stay in hours, after adjusting for CCI, this finding was not significant (Table 1). When an APP, RN, and pharmacist all staffed the discharge team (N = 26), the delay between discharge order and discharge from the hospital was reduced to an hour and a half. Fifty-six faculty members were sent the survey and we had a total of 39 responses. Seventy-nine percent of clinical staff that worked with the team and completed a survey would recommend the MDT to other providers. Clinical staff reflected that working with the team did save time and was most effective when an APP, pharmacist, and nurse were all involved (Table 2).

Table 1.

Multidisciplinary Discharge Team (MDT) Patients Compared to Concurrent Patients

	Coordinated by MDT N = 77	Not coordinated by MDT N = 945	p value
Patient demographics
Age, mean ± SD	52 ± 17	55 ± 16	0.1781
Gender, N (%)			0.2359
Male	44 (57.1%)	468 (49.5%)
Female	33 (42.9%)	477 (50.5%)
Non-English speaking, N (%)			0.5470
Yes	9 (11.7%)	90 (9.5%)
No	68 (88.3%)	852 (90.2%)
Missing	0 (0%)	3 (0.3%)
Charlson Comorbidity Index, median (IQR)	3 (2, 5)	5 (2, 7)	0.0008
Discharge disposition, N (%)			0.6043
Home	71 (92.2%)	819 (86.7%)
Post-acute care	5 (6.5%)	91 (9.6%)
Left against medical advice	0 (0%)	19 (2.0%)
Other	1 (1.3%)	16 (1.7%)
Patient flow
Discharge before 11 am, N (%)	20 (26.0%)	86 (9.1%)	<.0001
Discharge before noon, N (%)	42 (54.6%)	149 (15.8%)	<.0001
Discharge order time, mean ± SD	10:29 ± 1:30	12:51 ± 2:38	<.0001
Discharge time, mean ± SD	12:28 ± 1:53	14:49 ± 2:54	<.0001
Length of stay in days, mean ± SD	4.3 ± 3.0	5.5 ± 9.3	0.2609
Quality metrics
All cause readmission within 7 days, N (%)	3 (3.9%)	79 (8.4%)	0.1954
All cause readmission within 30 days, N (%)	12 (15.6%)	197 (20.9%)	0.3064
Models comparing patients with and without coordination by Multidisciplinary Discharge Team, adjusted for the Charlson Comorbidity Index (CCI)
		Parameter estimate	Standard error	p value
Model 1. Discharge before 11 am	Odds Ratio (95% CI)
Multidisciplinary Discharge Team	1.8 (1.4, 2.4)	0.5883	0.1434	<.0001
CCI		−0.0844	0.0337	0.0123
Model 2. Discharge before Noon	Odds Ratio (95% CI)
Multidisciplinary Discharge Team	2.5 (1.9, 3.1)	0.9038	0.1247	<.0001
CCI		−0.0752	0.0263	0.0042
Model 3. Discharge Order Time	Mean difference (95% CI)
Multidisciplinary Discharge Team	−2.3 h (−3.0, −1.7)	−2.34486	0.30848	<.0001
CCI		0.01673	0.02395	0.4849
Model 4. Discharge Time	Mean difference (95% CI)
Multidisciplinary Discharge Team	−2.3 h (−3.0, −1.6)	−2.29312	0.34139	<.0001
CCI		0.04232	0.02616	0.1061
Model 5. Length of stay in hours	Mean % change
Multidisciplinary Discharge Team	−55% reduction (−95%, 273%)	−0.79732	1.07769	0.4596
CCI		0.27446	0.08259	0.0009

Table 2.

Feedback About and From the Multidisciplinary Discharge Team

Survey results	N = 39
Did you utilize the APP Discharge Team over the last month? N (%)
Yes	29 (74.4)
No	10 (25.6)
How did the team affect your workflow? N (%)
Saved time	20 (69.0)
Added time	2 (6.9)
Did not change	7 (24.1)
Would you recommend the discharge team to other providers? N (%)
Yes	23 (79.3)
No	6 (20.7)
What prevented you from using the discharge team? N (%)
Patients not appropriate for D/C team	6 (15.4)
Lack of time	0 (0)
Lack of knowledge/training on team functionality	4 (10.3)
Forgot/overlooked	2 (5.1)
Chose not to	2 (5.1)
Other	3 (7.7)
Primary care team positive feedback
“Eliminated the time I would have had to take to complete discharge paperwork, as well as communicate with the patient should any changes or questions arise.”
“They could really do the work and the counseling that was required and given that they had more time, I felt like there were less errors. Plus with secure chat it was really easy to communicate!”
“We were able to round on new patients first and ensure consults were called early while the discharge advanced practice provider went and did discharge education with the patient.”
“A straightforward discharge does not take much time to do. Honestly the time-savings for a straightforward discharge is not that much. Where I think it is helpful is with a complicated discharge (i.e., someone who has multiple needs -- the infectious disease (ID) team needs to be contacted to set up outpatient parenteral antibiotic therapy, they need home oxygen, they need home health to be set up, etc). Those patients were specifically not supposed to be included, but honestly I think that's where the team could be most helpful.”
Primary care team negative feedback
“Discharging is an essential part of the primary team's job. We know the patients the best and can provide the safest discharges. Also if a legal issue occurred around a discharge the primary team is the one at fault even if they did not hit the 'discharge patient order.' A discharge team is really only useful on very easy patients, which are rare these days and would only take 5 minutes for me to do myself. All of the patients need to be evaluated by the primary team anyway, it doesn't take that much.”
“I remain hesitant about bringing in even more complexity by having yet another person work on something already fraught with potential for error. That said, I've heard from others that so long as patients are well selected, this can actually save time and be done well. One area of concern ok my part is I find the discharge meds on the discharge summary and After Visit Summary are often screwy and misaligned, often for reasons unclear. One thing I'd need to ensure and be more willing to use this service is to make sure all the meds are fully aligned including so when I actually go about writing my discharge summary.”
“Patient safety must be our top priority and discharging a patient is the most important aspect of patient's hospitalization. Having someone who did not take care of the patient during their hospitalization discharge them could lead to unsafe outcomes.”
Multidisciplinary Discharge Team feedback
“We were able to coordinate with respiratory therapy (RT) to get her oxygen that she could afford, and have pharmacy review all of her prior to admission meds, as well as all of her discharge prescriptions (including an albuterol inhaler), and work with social work/outpatient pharmacy to ensure she had a voucher so these would all be paid for.”
“We as a discharge team caught things or were able to provide an added level of education and patient-centered care with each patient since we had more time to spend with the patient and we went over things in detail. For example, I did some smoking cessation when waiting for the primary team and even printed some materials for the patient (which I rarely do when we are busy). I was able to also sit with the floor nurse for one patient for a few minutes and go over questions that she had and clarify medications.”
“We helped with one complex discharge, which required coordination with multiple services (Transplant ID, renal transplant, renal, and RT). Because my load was light today, I was happy to help out and it went very smoothly (this can easily be done if communication with the primary team is fluid).”
“If this is a metric that the hospital cares about/benefits the system with turnover etc., I see the dedicated APP/pharmacy team as helpful in increasing efficiency as we can focus on a few targeted patients together/early in the day.”
“Work of a discharge team may not always be the most rewarding work and may in some senses be inefficient (new face seeing patient, often hear and address questions and concerns multiple times - probably benefits the patient's understanding/satisfaction in at least some cases), however does allow earlier and more focused execution of tasks. Involving another team with discharge presents some safety concerns but I found with appropriately selected patients, good communication with the primary team on plan, and support from pharmacy things seemed safe/supported (arguably a distracted/multitasking primary team may miss things on medication reconciliation too, for example)”.

DISCUSSION

The important findings of this study are as follows: (1) the MDT expedited earlier discharges, (2) was further augmented when a discharge nurse was involved, and (3) overall resulted in a positive response from providers who utilized the service.

Work by Khanna et al. suggested that moving discharges earlier in the day by even an hour reduced hospital overcrowding.⁵ There have been multiple studies highlighting efforts around discharges before noon. Typically, these interventions utilize mass communication, huddles, and incentives to do earlier discharges and often result in additional interruptions, are redundant, and variably successful.

This is a single-center study and patients were not randomized though results adjusted for CCI. We were able to reallocate staff during the time period; however, this type of team may require additional resources. While trends suggest a lower length of stay, additional studies are warranted given insufficient sample size.

CONCLUSIONS

The MDT may improve efficiency in discharges while allowing teams to provide ongoing care to other patients concurrently.

Funding

This work was funded by the University of Colorado Data Science to Patient Value Grant Program.

Declarations

Conflict of Interest

The authors declare that they do not have a conflict of interest.