writing

JNCQ-D-20-00317.pdf

Standardized Discharge Planning Tool Leads to Earlier Discharges and Fewer Readmissions

Colleen Kucharczuk, DNP, RN, AGACNP-BC; Erin Lightheart, MBA, CSSBB; Allison Kodan, MSN, RN, OCN, AGACNP-BC; Cassandra Haynes, MPH, PA-C; Susan Rabatin, MSN, RN, AGACNP-BC; Joy Burke, MSN, RN, AGACNP-BC; Jamie Senger, MSN, RN, AGACNP-BC; Lindsay Lee, MSN, RN, AGACNP-BC; Sarah Brinley, PharmD, BCOP; Mary Angela Decena, MSN, RN, AOCNP, AGACNP-BC; Juanita M. Cruz, MSN, RN; Rebecca Hirsh, MD; Kathy McCauley, PhD, RN

ABSTRACT Background: In an inpatient setting, aspects of discharge planning are often left to the provider’s memory, leading to errors, inefficiencies, and avoidable costs. Methods: A multidisciplinary team of oncology practitioners used process improvement methodologies to redesign the discharge planning process. Interventions: The primary intervention was an evidence-based discharge planning tool, called the discharge navigator, used from admission through discharge. Results: Thirty-day unplanned readmission rates decreased by 29.0% from preimplementation (March 2017 through August 2017) to postimplementation (September 2017 through March 2020). The percentage of patients discharged before noon increased 76.2%. A comparable service not utilizing the intervention saw lesser or no improvement in these measures. Conclusion: The tool provided a systematic approach to discharge planning. Key design elements included a centralized location within the electronic health record and an electronic shortcut to populate the tool. Although developed for a specialized population, most elements are applicable to any hospitalized patient. Keywords: discharge planning, patient discharge, quality improvement, readmissions, transitional care

Facilitating timely, safe, and efficient tran- sitional care is paramount to optimizing

clinical outcomes, the patient experience, and organizational throughput. Poor discharge plan-

Author Affiliations: Departments of Advanced Practice (Dr Kucharczuk and Mss Kodan, Haynes, Rabatin, Senger, Lee, and Decena), Quality and Safety (Ms Lightheart), Inpatient Pharmacy (Dr Brinley), Clinical Resource Management (Ms Cruz), and Medicine (Dr Hirsh), Hospital of University of Pennsylvania, Philadelphia; Division of Hospital Medicine, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania (Ms Burke); and University of Pennsylvania School of Nursing, Philadelphia (Dr McCauley).

The authors declare no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.jncqjournal.com).

Correspondence: Colleen Kucharczuk, DNP, RN, AGACNP-BC, Department of Advanced Practice, Hospital of University of Pennsylvania, 3400 Spruce St, Rhoads Pavilion, 3rd Floor, Philadelphia, PA 19104 ([email protected]).

Accepted for publication: January 16, 2021

Published ahead of print: March 10, 2021

DOI: 10.1097/NCQ.0000000000000558

ning can lead to devastating adverse events for patients, including unplanned hospital read- missions and death. Oncology patients who experience an unplanned readmission within 30 days of a hospitalization have a significantly higher risk of mortality.1-3 In addition to affecting clinical outcomes, unplanned readmissions neg- atively affect hospital spending4 and reputation.5

While the discharge process is of utmost im- portance to a patient’s clinical outcome and suc- cessful transition, critical components are often left to the provider’s memory,6 which may lead to errors, omissions, and inefficiencies. Specif- ically, patients with hematologic malignancies have complex medical needs during care tran- sitions, including frequent laboratory test result monitoring, central catheter care, transfusions, growth factor administration, and comprehen- sive coordinated follow-up visits.7

AVAILABLE KNOWLEDGE A well-organized discharge plan with multiple checkpoints can improve the follow-up care of

January–March 2022 • Volume 37 • Number 1 www.jncqjournal.com 55

patients with complex care needs, and many discharge checklists have been developed to ad- dress this need.6-12 Specific efforts focused on provider education and timely postdischarge care have reduced readmission rates.13 Ensuring a well-coordinated and timely discharge has the potential to improve patient throughput while mitigating poor patient outcomes.

This quality improvement (QI) project oc- curred at an 807-bed hospital within an academic medical center in the northeast United States. The Oncology Advanced Practice Provider (APP) service at this hospital manages approximately 1600 hematologic malignancy inpatient visits per year, averaging 140 dis- charges per month. Nurse practitioners and physician assistants work in collaboration with attending physicians to deliver comprehensive care on this service, 24 hours per day, 7 days per week. Patients with hematologic malignancy at the institution are also cared for by a resident- or house staff–based service. Inpatient oncology services at the hospital typically operate at or above 100% capacity. Furthermore, the APP service expanded its model of care from day- time only to 24-hour coverage in 2017. This required about a 50% increase in the number of providers. The need for standardization and error proofing of the discharge process became readily apparent.

SPECIFIC AIMS The purpose of this initiative was to redesign the discharge planning process to facilitate effective care transitions by utilizing process improve- ment tools and methodologies. The primary aim was to improve 30-day unplanned readmission rates from 15.59% at baseline (March 2017 through August 2017) to 14.81% (a 5% re- duction) in the following year after intervention (September 2017 through August 2018). The secondary aim was to improve the rate of dis- charge by noon from 21% at baseline to 25%, an organization-wide goal established by execu- tive leadership.

METHODS Context The project team consisted of APPs, a QI advisor, nurses, pharmacists, physicians, social workers, and discharge planners. The intervention was limited to the APP-based service. The service utilizing house staff was not included in the

intervention but was utilized as a comparison group. For both services, the models of inpatient clinical care delivery and the patient population remained stable in the pre- and postimplemen- tation periods, represented by a stable case-mix index. This project was exempt from needing in- stitutional review board approval.

Intervention The project team used various process improve- ment tools. A multidisciplinary brainstorming session identified the barriers to an effective discharge process, and an affinity diagram cat- egorized the issues. The team developed possi- ble interventions, and an impact/effort matrix helped prioritize those options. The team used Plan-Do-Study-Act (PDSA) cycles to continually review and improve the intervention.

The intervention was a discharge planning tool called the discharge navigator, first imple- mented on the APP service in August 2017. The navigator was designed internally by the project team and exists in a table format em- bedded in the electronic health record (EHR). The tool in its entirety can be referenced in the Supplemental Digital Content, Table 1 (available at: http://links.lww.com/JNCQ/A843). The nav- igator was initiated by APPs on admission or transfer to the service, using an electronic short- cut, which enabled them to populate the table into an electronic note within seconds. The ta- ble was formatted to allow users to seamlessly move between elements using the tab key. The navigator was updated daily by APPs and dis- charge planners and finalized at least 24 hours prior to anticipated discharge. The navigator was used for all patients on the APP service without exclusion. Adherence to utilizing the navigator served as a process metric and was monitored in real time by the service’s administrative assistant.

Each element of the navigator was carefully considered on the basis of expert opinion and evidence. The project team considered and incor- porated several elements of the Re-Engineered Discharge (RED) Toolkit from the Agency for Health Research and Quality (AHRQ)14 and Project BOOST from the Society of Hospital Medicine15 to build the navigator and redesign the discharge process. Examples include the early identification of a patient’s language preference and translation of written discharge materials, a proactive plan to obtain correct medications, and a discharge medication reconciliation by

56 Standardized Discharge Planning Tool Journal of Nursing Care Quality

a pharmacist. Other examples are scheduling follow-up appointments prior to discharge, co- ordinating outpatient laboratory test results, securing durable medical equipment and services such as physical therapy, and a 72-hour postdis- charge phone call to reinforce the discharge plan.

After the introduction of the navigator in August 2017, regular feedback was elicited from users to guide iterative improvements to the tool. For the first PDSA cycle, the project team added a new element to prompt the user to con- sider whether the patient was a candidate for a rideshare program. In June 2018, the service began utilizing this program to facilitate timely and convenient transportation home. This effort was a separate QI project implemented dur- ing the study period. The rideshare option was available to both the APP (intervention) team and the house staff (comparison) team. Another change to the navigator during PDSA cycle 1 was a prompt for the provider to place an or- der for a pharmacist’s reconciliation of discharge medications.

In November 2018, PDSA cycle 2 incorpo- rated a prompt to request a follow-up phone call for a particular subset of patients at high risk for readmissions and other adverse clinical out- comes. The implementation of follow-up calls required a separate QI project and will not be discussed in this article. This illustrates how the navigator tool can be used in conjunction with and to support other processes.

In September 2019, PDSA cycle 3 incorpo- rated a prompt to obtain final discharge recom- mendations from every consultant involved in the patient’s care. Also during this PDSA cycle, providers were equipped by their manager with a monthly individualized scorecard of their dis- charge by noon performance compared with the team average.

Measures and analysis The readmission rate was defined as the total number of patients readmitted to the facil- ity for any reason within 30 days divided by the total number of discharges. Patients who expired and patients with a scheduled readmis- sion were excluded. Monthly readmission rates were obtained through the organization’s in- ternal database. The monthly discharge before noon rate was defined as discharges occurring between 12:00 am and 11:59 am divided by total discharges for the month, without exception. A

report of discharge times was automatically gen- erated from the EHR.

Values for the outcome measures were cal- culated and plotted by the QI advisor monthly during the study time period and ongoing. The preimplementation period was defined as March 2017 through August 2017. This time frame was limited to 6 months because a new EHR was implemented in March 2017. Prior to that point, readmissions data were not able to be segmented by service. Thus, data for the APP (in- tervention) and house staff (comparison) services could not be distinguished from one another. The postimplementation period was September 2017 through March 2020. Run charts were employed for data analysis, tracking, and visualization.

To calculate estimated cost savings, the before and after readmission rates were multiplied by the average number of annual admissions to the intervention service. The difference in the num- ber of readmissions was then multiplied by the average cost of an oncology readmission as de- fined by the AHRQ.4

It was important that the tool improved out- comes while not extending patient stays or hindering staff workflow. To ensure this was the case, average observed-to-expected length of stay (LOS) and measures of staff satisfaction with the navigator served as balancing metrics. LOS data were pulled from the organization’s internal database and were charted and tracked monthly by the QI advisor. An anonymous sur- vey of the multidisciplinary team was conducted 4 months after implementation, surveying those who documented within the tool (APPs and dis- charge planners) and those who viewed it but did not document within it (bedside nurses). Sur- vey questions measured ease of use and effect on workflow using a 3-point Likert scale. Beyond this survey, continual dialogue occurred among the team about how to enhance the tool’s ease of use. Changes to format, wording, and the order of elements were made accordingly.

RESULTS Adherence to utilizing the navigator was 100% in the 6 months following intervention. Every 3 months afterward, a random sample of 20 charts was reviewed and demonstrated ongoing adherence.

Thirty-day unplanned readmission rates for the APP service initially decreased from 15.59% (March 2017 through August 2017) to a

January–March 2022 • Volume 37 • Number 1 www.jncqjournal.com 57

Figure 1. Thirty-day unplanned readmission rates. APP indicates advanced practice provider; PDSA, Plan-Do-Study-Act.

sustained 14.02% (September 2017 through August 2019), surpassing the initial goal. A sub- sequent reduction to 11.07% (September 2019 through March 2020) occurred following PDSA cycle 3. This represents an overall 29.0% reduc- tion from pre- to postintervention. The house staff service saw a more modest overall reduc- tion of 14.5% (Figure 1), suggesting there were changes external to the navigator that influenced readmissions for both services.

The annual cost reduction due to improved readmission rates on the APP service based on a readmission rate of 11.07% is $1 294 228. This assumes an average avoidable cost of $17 900 per readmission.4 The percentage of patients dis- charged before noon increased from a baseline average of 21% to 27% after the initial inter- vention. There were subsequent improvements to 32% and 37% following PDSA cycle 1 and PDSA cycle 3, respectively. This represents an overall increase by 76.2%. Comparatively, the discharge before noon rate for the house staff ser- vice remained stable at 6% (Figure 2).

Observed-to-expected LOS remained stable throughout the baseline and intervention peri- ods (Figure 3). This demonstrates that providers were not extending the patient’s hospitalization an additional day in order to meet the discharge before noon goal.

In the staff satisfaction survey, of the 10 doc- umenting users, 0 reported it was difficult to use and all 10 reported the navigator had a very positive or somewhat positive effect on their workflow. Of the 22 viewing users, 1 person re- ported it was somewhat difficult to use and had

a somewhat negative effect on workflow. The re- maining reported neutral to positive associations with the navigator. The survey was limited in its response rate, capturing approximately 40% of documenting users and 10% of viewing users. Survey details can be found in Supplemental Dig- ital Content, Table 2 (available at: http://links. lww.com/JNCQ/A844).

DISCUSSION Summary This QI project implemented an inpatient dis- charge planning tool embedded in the EHR that contributed to reduced readmissions and ear- lier discharge times without negatively affecting LOS or staff workflow. This tool has become a standard of care on the service, fostering collab- oration among members of the multidisciplinary team and enabling transparency and seamless handoffs. At the time of this writing, outcomes have been sustained for 32 months after imple- mentation. Ongoing measurement continues.

Prioritizing discharge before noon as a goal metric and providing quantitative individualized feedback to providers were pivotal to facilitating earlier discharges. Also central to this outcome were utilizing the navigator from admission through discharge and incorporating elements specifically focused on throughput. We believe that the incorporation of elements shown to enhance patient safety led to the observed reduc- tion in readmissions, by allowing providers to meet the needs of a complex patient population without relying on memory. Key design elements that contributed to the tool’s ease of use included

58 Standardized Discharge Planning Tool Journal of Nursing Care Quality

Figure 2. Percentage of patients discharged before noon. APP indicates advanced practice provider; PDSA, Plan-Do-Study-Act.

a centralized location within the EHR and an electronic shortcut to populate the navigator.

Interpretation Checklists have been developed to facilitate efficient discharge planning in emergency departments.11 Others have developed checklists to enhance patient safety during the inpatient discharge process6,8,10,12 and operating room procedures.16 In addition, checklists have aimed to improve specific outcomes for patients with hematologic malignancy, including the number of appointments secured prior to discharge9 and securing patients’ complex medications needs.7

While these have all been important to improv- ing various metrics associated with the discharge

process, to our knowledge, this is the first dis- charge tool developed that has both improved efficiency and been associated with a reduction in 30-day unplanned readmissions for this complex patient population with significant cost savings. We believe that the discharge navigator tool’s multifaceted approach to discharge planning, its standardization of processes, its promotion of interdisciplinary communication, and its unique ability to support other QI initiatives have all contributed to its success. To our knowledge, the navigator is the first tool of its type.

The navigator tool is distinct from a check- list. A checklist is a list of items to be completed or considered and serves as a reminder to the user. The navigator fulfills that purpose and also

Figure 3. Discharge before noon compared with length of stay over time.

January–March 2022 • Volume 37 • Number 1 www.jncqjournal.com 59

functions as a dynamic communication tool to address individualized patient needs. For exam- ple, when editing the navigator, the discharge planner does not simply check a box to indi- cate that home care has been arranged. Instead, the discharge planner can add explanatory infor- mation, such as the patient’s lack of insurance coverage for a central catheter at home that would necessitate line removal before discharge. There are many sections of the navigator where this type of nuanced communication occurs.

The navigator has become the first line of intervention for addressing systems issues and newly identified gaps in the discharge process, increasing its comprehensiveness. For instance, when the rideshare program was implemented, a new element was added to the navigator to prompt providers to consider this resource. For virtually any QI project taking place on our ser- vice, we incorporate relevant prompts into the navigator.

The navigator was designed to be used from admission through discharge, and its elements are organized along a timeline. For instance, on admission, the navigator prompts the provider to identify any home medications that will need to be refilled at discharge. During the middle of the inpatient stay, the provider is prompted to obtain prior authorization for medications, which can take up to 72 hours. At least 24 hours before anticipated discharge, the provider is trig- gered to request a pharmacy reconciliation of discharge medications. Identifying and mitigat- ing these common barriers early on, rather than in the 24 hours before the patient leaves, allow for a smooth and timely discharge.

Transparency, both in adherence to the new process and in performance metrics, was essen- tial to our success. Before the navigator became a standard of care, we monitored closely each provider’s adherence to using the tool and pro- vided real-time feedback. We targeted a specific team goal for discharge before noon rate and shared the data regularly. Monthly individual scorecards enabled providers to track their per- formance against the team average.

While bedside nursing feedback was essential to the development of our process and tool, the navigator was not embedded into the nurse’s daily process, and nurses were not document- ing users in the tool. We believe it is within the scope of bedside nursing to complete and col- laborate on several elements of the navigator,

such as how the patient plans to get home from the hospital. Another possible point of collabo- ration could be a “day of discharge pause,” in which the discharging RN in collaboration with the discharging APP ensures that all elements of the navigator have been addressed. This type of pause has been effective in preventing errors for others who developed discharge checklists.12

Enhancing the electronic interface between the patient’s medical record and the naviga- tor would further increase its ease of use and efficiency. For instance, it would be ideal for con- sulting services’ recommendations for discharge to be automatically populated into the naviga- tor, rather than the discharging provider having to search the EHR to identify the consultants’ recommendations. Prolonged and complicated hospital stays increase the likelihood that this el- ement of care would be overlooked, and better electronic integration into the navigator would enhance its impact.

We also believe the effects of the naviga- tor could be enhanced if the tool were further extended into the postdischarge setting. For ex- ample, improved communication across sites of care, along with a transitions coach, resulted in a significant reduction in readmissions.17 Similarly, APPs leading a comprehensive 3-month transi- tional care intervention significantly lengthened time to first readmission.18 Our navigator tool could easily be coupled with existing transitional care programs to provide seamless availability of information to postdischarge care teams.

Limitations Although the discharge navigator was developed for a highly specialized patient population, the majority of its elements are applicable to any hospitalized patient. It is recommended that or- ganizations implement the general elements for all patients and enable services to add specialized components as needed. Further investigation may provide insight into which elements were most influential on the outcomes. For instance, medication reconciliation by a pharmacist at dis- charge is known to enhance patient safety,19 but its unique contribution has not been quantified in the context of this project. It is also possible, and we believe likely, that it was the combined influence of the navigator’s various elements that resulted in the improvements we observed.

Organization-wide implementation and a tar- geted training process are recommended. It

60 Standardized Discharge Planning Tool Journal of Nursing Care Quality

should be noted that having a single care team following the patient is not required to real- ize the benefits of the tool. The APP service comprises 30 providers who share patients in ro- tating shifts. Thus, the navigator is designed to be used by multiple providers in the course of a single patient’s hospitalization.

A confounding factor to our measurement is the health system’s broad and continual focus on improving readmissions. There are many ini- tiatives, both in the inpatient and outpatient settings, that could have played a role in the observed reduction in readmission rates. This makes it difficult to say definitively that the nav- igator is what caused the improvement in this measure. The comparison with a similar service allowed us to isolate the unique effect of the nav- igator intervention.

CONCLUSIONS The navigator tool has helped reduce un- planned hospital admissions and improve hospi- tal throughput, outcomes that are important for patients and facilities alike. It has proven to be a sustainable intervention with continued staff approval and prolonged results. However, other potential benefits were not objectively measured in our study, such as improvements in trans- parency, communication, and patient experience. These could be considered in the future to help guide enhancement of the navigator and make a case for wider implementation.

Nurses and APPs are well situated and respon- sible for coordinating comprehensive discharge planning. The navigator helps fulfill this charge by eliminating reliance on memory and bringing transparency to the discharge planning process, resulting in safe, efficient, cost-effective, and patient-centered care.

REFERENCES 1. Hembree TN, Thirlwell S, Reich RR, Pabbathi S, Extermann

M, Ramsakal A. Predicting survival in cancer patients with and without 30-day readmission of an unplanned hospital- ization using a deficit accumulation approach. Cancer Med. 2019;8(15):6503-6518. doi:10.1002/cam4.2472

2. Spring L, Li S, Soiffer RJ, Antin JH, Alyea EP, Glotzbecker B. Risk factors for readmission after allogeneic hematopoi- etic stem cell transplantation and impact on overall sur- vival. Biol Blood Marrow Transplant. 2015;21(3):509-516. doi:10.1016/j.bbmt.2014.11.682

3. Bejanyan N, Bolwell BJ, Lazaryan A, et al. Risk fac- tors for 30-day hospital readmission following myeloab- lative allogeneic hematopoietic cell transplantation (allo- HCT). Biol Blood Marrow Transplant. 2012;18(6):874-880. doi:10.1016/j.bbmt.2011.10.032

4. Bailey MK, Weiss AJ, Barret ML, Jing J. Characteris- tics of 30-Day All-Cause Hospital Readmissions, 2010- 2016. AHRQ; 2019. Statistical Brief No. 248. Accessed October 20, 2020. https://www.ncbi.nlm.nih.gov/books/ NBK538941

5. FAQ: how and why we rank and rate hospitals. U.S. News & World Report. July 28, 2020. Accessed October 20, 2020. https://health.usnews.com/health-care/best- hospitals/articles/faq-how-and-why-we-rank-and-rate- hospitals#best-hospitals-overview

6. Garg T, Lee JY, Evans KH, Chen J, Shieh L. Development and evaluation of an electronic health record–based best- practice discharge checklist for hospital patients. Jt Comm J Qual Patient Saf. 2015;41(3):126-131. doi:10.1016/S1553- 7250(15)41017-7

7. Warsame R, Kasi PM, Villasboas-Bisneto J, et al. Tran- sition of care for inpatient hematology patients receiving chemotherapy: development of hospital discharge huddle process and effects of implementation. J Oncol Pract. 2016; 12(1):e88-e94. doi:10.1200/JOP.2015.005785

8. Halasyamani L, Kripalani S, Coleman E, et al. Transition of care for hospitalized elderly patients: development of a discharge checklist for hospitalists. J Hosp Med. 2006;1(6): 354-360. doi:10.1002/jhm.129

9. Prince M, Allen D, Chittenden S, Misuraca J, Hockenberry MJ. Improving transitional care: the role of handoffs and discharge checklists in hematologic malignancies. Clin J On- col Nurs. 2019;23(1):36-42. doi:10.1188/19.CJON.36-42

10. Soong C, Daub S, Lee J, et al. Development of a checklist of safe discharge practices for hospital patients. J Hosp Med. 2013;8(8):444-449. doi:10.1002/jhm.2032

11. Beck MJ, Okerblom D, Kumar A, Bandyopadhyay S, Scalzi LV. Lean intervention improves patient discharge times, improves emergency department throughput and reduces congestion. Hosp Pract. 2016;44(5):252-259. doi:10.1080/21548331.2016.1254559

12. Drake K, McBride M, Bergin J, Vandeweerd H, Higgins A. Ensuring safe discharge with a standardized check- list and discharge pause. Nursing. 2017;47(8):65-68. doi:10.1097/01.NURSE.0000521042.81195.86

13. Montero AJ, Stevenson J, Guthrie AE, et al. Reducing unplanned medical oncology readmissions by improving outpatient care transitions: a process improvement project at the Cleveland Clinic. J Oncol Pract. 2016;12(5):e594-e602. doi:10.1200/JOP.2015.007880

14. Jack B, Paasche-Orlow M, Mitchell S, et al. Re-Engineered Discharge (RED) Toolkit. AHRQ; 2013. Publication No. 12(13)-0084. Accessed October 20, 2020. http://www.ahrq. gov/patient-safety/settings/hospital/red/toolkit/index.html

15. Project BOOST® Implementation Guide: Second Edition. Society of Hospital Medicine; 2013. Accessed October 20, 2020. https://www.hospitalmedicine.org/globalassets/ professional-development/professional-dev-pdf/boost- guide-second-edition.pdf

16. Bergs J, Hellings J, Cleemput I, et al. Systematic review and meta-analysis of the effect of the world health organization surgical safety checklist on postoperative complications. Br J Surg. 2014;101(3):150-158. doi:10.1002/bjs.9381

17. Coleman EA, Parry C, Chalmers S, Min S. The care transitions intervention: results of a randomized con- trolled trial. Arch Intern Med. 2006;166(17):1822-1828. doi:10.1001/archinte.166.17.1822

18. Naylor M, Brooten D, Campbell R, et al. Transitional care of older adults hospitalized with heart failure: a random- ized controlled trial. J Am Geriatr Soc. 2004;52:675-682 doi:10.1111/j.1532-5415.2004.52202.x

19. Patel E, Pevnick JM, Kennelty KA. Pharmacists and med- ication reconciliation: a review of recent literature. Integr Pharm Res Pract. 2019;8:39-45. doi:10.2147/IPRP.S169727