|Year : 2020 | Volume
| Issue : 4 | Page : 121-125
The race for readmission reduction: Primary care follow-ups reduce debility readmissions after acute inpatient rehabilitation
David Sherwood1, Benjamin Gill2, Derek Schirmer3, Alexandra Arickx1, Cheng Shu1, Anthony Jackson4, Sarah Eickmeyer1
1 Department of Rehabilitation Medicine, University of Kansas Medical Center, Kansas City, KS, USA
2 Department of Physical Medicine and Rehabilitation, University of Missouri-Columbia, Columbia, MO, USA
3 Department of Internal Medicine, University of Missouri Kansas-City, Kansas City, MO, USA
4 Department of Internal Medicine, University of Kansas Medical Center, Kansas City, KS, USA
|Date of Submission||02-Mar-2020|
|Date of Decision||11-Jun-2020|
|Date of Acceptance||15-Jun-2020|
|Date of Web Publication||19-Oct-2020|
Dr. David Sherwood
Department of Rehabilitation Medicine, University of Kansas Medical Center, Kansas City, KS
Source of Support: None, Conflict of Interest: None
Background: In 2012, the Centers for Medicare and Medicaid Services began to reduce payments to qualifying hospitals for 30-day readmission rates that were higher than predicted for specific diagnoses. The process was broadened to include skilled nursing facilities in 2018. It is reasonable to expect future expansion will include acute inpatient rehabilitation facilities. A pre-intervention quality improvement project from 2017 identified that patients admitted to an acute inpatient rehabilitation facility (IRF) for the primary diagnosis of debility were readmitted within 30 days of discharge at a rate of 38%, which was nearly three times higher than the next most readmitted diagnosis. A literature review identified rapid primary care provider (PCP) outpatient follow up as a worthy intervention to reduce readmissions. Objectives: Over a six-month intervention period, we attempted to achieve a reduction in 30-day readmission rates in the debility population of an IRF by scheduling PCP follow-ups within seven business days after discharge. Results: Of those that received the intervention, 7% were readmitted (P=0.018). Of those who did not receive the intervention, 56% were readmitted. Conclusion: The adoption of PCP follow-up within seven business days of discharge may lower the 30-day readmission rate for patients admitted to IRF with a primary diagnosis of debility.
Keywords: Debility, quality improvement, readmission, rehabilitation
|How to cite this article:|
Sherwood D, Gill B, Schirmer D, Arickx A, Shu C, Jackson A, Eickmeyer S. The race for readmission reduction: Primary care follow-ups reduce debility readmissions after acute inpatient rehabilitation. J Int Soc Phys Rehabil Med 2020;3:121-5
|How to cite this URL:|
Sherwood D, Gill B, Schirmer D, Arickx A, Shu C, Jackson A, Eickmeyer S. The race for readmission reduction: Primary care follow-ups reduce debility readmissions after acute inpatient rehabilitation. J Int Soc Phys Rehabil Med [serial online] 2020 [cited 2020 Nov 28];3:121-5. Available from: https://www.jisprm.org/text.asp?2020/3/4/121/298609
| Introduction|| |
In 2012, by way of the Affordable Care Act's Hospital Readmission Reduction Program, the Centers for Medicare and Medicaid Services began to reduce payments to qualifying hospitals for 30-day readmission rates that were higher than predicted for a few specific diagnoses. The policy was broadened to include skilled nursing facilities in 2018. It is reasonable to expect future expansion will include acute inpatient rehabilitation facilities (IRF).
A subsequent literature review for articles published between 2007 and 2017 revealed a paucity of outcome studies related to IRF 30-day readmission rates. However, there was robust discussion on variables that correlated with higher 30-day readmission rates at acute care hospitals. Common variables included: socioeconomic demographics, comorbidities, diagnosis, readmission reason, insurance, specific hospital course events, laboratory values, length of stay, discharge location, and physician follow-up postdischarge, among others.,,,,,,,,, Many of these variables (socioeconomics, insurance, demographics, comorbidities, and employment) cannot be changed in the scope of a quality improvement project and are subject to the influence of factors outside of the control of a physician team. Moreover, monitoring discharge laboratories and vital signs for acute abnormalities at discharge seemed to be in line with the current standard of practice, so they were not selected as our intervention point.
Brooke et al. found that early follow-up, mostly defined as within 2 weeks of discharge, with a primary care provider significantly reduced the risk of readmission among select surgical patients who experienced perioperative complications, from 35.0% (without follow-up) to 20.4% (with follow-up) (P < 0.001). They produced a simple and intuitive hypothesis which suggests that for patients undergoing high-risk surgery, rapid follow-up with a PCP is a means by which complications, in discharge planning can be recognized before they escalate into a problem warranting readmission. Shen et al. reviewed the efficacy of reducing readmission with dedicated PCP posthospital discharge follow-up. They found that any follow-up visit with a primary care clinician within 7 days of discharge was associated with a lower risk for 30-day readmission for patients on medicine services. Given this growing literature base supporting the efficacy of PCP follow-up at the acute care hospital level, we elected to pursue rapid (within seven business days) PCP follow-up as our intervention point in the IRF setting.,,,,
The aim of the quality improvement project was to reduce the 30-day readmission rate for patients admitted to an IRF for debility with rapid PCP follow-up.
| Methods|| |
Our quality improvement project took place at a large academic medical center's IRF. The study was structured as a prospective cohort design. The project was approved by the institutional IRB as a quality improvement project.
Using the Plan-Do-Study-Act (PDSA) cycle, our initial “Plan” phase set out to identify high-risk populations for 30-day readmission at our IRF in 2017. Preintervention, discharge data from May 1, 2017, to October 31, 2017, was collected from 344 discharges from our facility. Four physicians separately reviewed the 344 charts and identified 86/344 discharges as readmitted within 30 days of discharge. A subsequent review, also carried out separately by four physicians, excluded inappropriate readmissions, namely (1) those that were planned readmissions for chemotherapy or planned surgery, or (2) readmissions that were discharged directly back to the acute care hospital from inpatient rehabilitation. This left 37/86 readmissions meeting our criteria for 30-day readmissions for inclusion in the planning and data collection phase. Each of the included charts was then manually reviewed by four physicians for the following variables: the presence of PCP follow-up at discharge, age, gender, race, congestive heart failure, diabetes mellitus, hyperlipidemia, cancer, obesity, vital signs at discharge, the need for oxygen at discharge, length of IRF stay, length of total hospital course, LACE + discharge score, insurance type, need for blood transfusion during IRF stay, sepsis diagnosis during IRF stay, urinary tract infection during IRF stay, disposition from IRF, bladder management functional independence measure (FIM) at discharge, bowel management FIM at discharge, ambulating FIM at discharge, employment status, prehospital housing status, anticoagulation status at discharge, number of days between IRF discharge and any physician follow-up, sodium level at discharge, creatinine level at discharge, potassium level at discharge, deep vein thrombosis (DVT) detection during IRF, blood glucose on day of discharge, number of consultants during IRF stay, and readmission diagnosis. These variables were chosen based on previously studied variables identified in our literature review or variables the data collecting team felt anecdotally may correlate with a higher readmission risk. The data were reviewed by two physicians for trends, and the targeted intervention of rapid PCP follow-up within 7 days of IRF discharge was chosen based on emerging evidence to support the practice.,,
During the intervention, or “Do,” phase, July 1, 2018, and December 31, 2018, a protected ledger was created to manually record the MRN, Discharge Date, PCP follow-up date, whether the PCP was within the hospital system, and if the patient was readmitted within 30 days of discharge for all patients admitted to IRF during that time frame with a primary diagnosis of debility. The primary resident physician caring for each patient in our cohort was responsible for scheduling a rapid PCP follow-up within 7 business days of discharge for each patient admitted with a primary admission diagnosis of debility. The residents were instructed to make the appointment before discharge by either electronic request or by phone for patients with PCPs within the hospital system, or by phone for those which existed outside the hospital system as no system existed for electronic follow-ups outside the health system during the intervention phase. Thirty days following the date that the last eligible debility patient could be discharged (January 30, 2019), the readmission data were compiled. Medical record numbers for our cohort were reviewed 30 days' post discharge. Those with an inpatient encounter created, which was not scheduled (i.e., chemotherapy and planned surgical intervention), within 30 days of discharge were designated to be included as readmitted within 30 days. A subsequent review process performed by two physicians confirmed the accuracy of the 30-day readmission data.
The statistical analysis was carried out using Fisher's exact test given the small sample size in the analysis. The test is under a null hypothesis of independence to a hypergeometric distribution of the numbers in the cells of the table.
| Results|| |
On internal review of the 30-day readmission data from the preintervention phase, it became clear that the debility population was the most at risk for 30-day readmission. As shown in [Figure 1], these patients accounted for 38% (14/37) of our total 30-day readmissions. The next most likely population for readmission was only readmitted at a rate of 14% (ischemic stroke, 5/37).
|Figure 1: Primary rehab admission diagnosis of those readmitted to an acute care hospital within 30 days of inpatient rehabilitation facilities discharge. Preintervention|
Click here to view
Total debility admissions in the preintervention period were 35 patients, meaning that 40% (14/35) of patients admitted for debility were readmitted within 30 days. There were 26 total debility admissions during the intervention phase. Three were excluded from the cohort on the grounds that they were urgently transferred to the acute hospital setting without returning to the rehabilitation facility, had planned readmissions for surgical or chemotherapeutic intervention, or died before 30 total days' post-IRF discharge had passed. Therefore, 23 patients qualified for inclusion in the intervention phase.
As shown in [Figure 2], rapid PCP follow-up was scheduled for 61% of patients (14/23). Of these, 7% (1/14, P = 0.01831) were readmitted within 30 days. 39% (9/23) either had no PCP follow-up or had PCP follow-up later than 7 business days' post-discharge. 56% (5/9) of those not receiving the intervention were readmitted within 30 days. There was a reduction from a preintervention 30-day readmission rate of 40% (14/35) for debility patients to a 26% (6/23) postintervention 30-day readmission rate (P < 0.05, odds ratio 0.07). Among our cohort of 14 patients who scheduled rapid follow-up, three patients were scheduled within our health system. 2/3 of those patients attended their follow-up appointment through electronic medical record review. None of the three patients were readmitted within 30 days of discharge.
Postintervention, readmissions for a primary diagnosis of debility were collected by a hospital administrator from January 1, 2019, to June 1, 2019. A single physician individually reviewed discharge summaries and charts. Data collected included whether the patient was readmitted within 30 days, if the patient had a follow-up listed in their chart, and if the follow-up was within 7 business days of discharge. This 5-month poststudy period experienced a 17% (5/30) rate of readmission within 30 days of discharge in patients with diagnoses of debility. On closer examination, only 3/30 (10%) had a rapid PCP follow-up documented at discharge. Despite the removal of our study intervention, the readmission rate in the debility population actually declined from 26% to 17%.
Readmission data were limited to those within our health system. Any readmissions that occurred to other health systems were not tracked. We were unable to verify patient attendance to follow-ups scheduled with PCPs external to our hospital system. While we can confirm that appointments were made through discussion with PCP, PCP scheduler, or confirmed through electronic record, we are unable to fully account for attendance. Furthermore, the small sample size of this quality improvement project may limit generalizability to the larger IRF population.
| Discussion|| |
On review of the “Study” phase of our cycle, rapid PCP follow-up from IRF reduced 30-day readmissions at a statistically significant rate for the debility population. The debility population at IRF is typically composed of medically complex geriatric patients who have recently experienced a prolonged hospital course which have led to their debility. It is reasonable to expect that rapid PCP follow-up reduces readmissions by serving as a resource to review medications, coordinate care, address symptomatic concerns, and serve as an intermittent support system to patients and their caregivers. In essence, rapid PCP follow-up functions as a means to proactively address the oft-cited dangers associated with transitions of care, in this case between acute care hospitals to IRF to home.
We did not expect that nine patients in our cohort would not receive, or would refuse, the intervention. Given the spectrum of illness in the debility population, some patients believed the specialists involved in their care were acting as their primary care providers. Many of these patients were resistant to the idea of scheduling with a PCP or following up with their PCP within 7 business days as they already had follow-up scheduled with specialists. In certain populations frequently seen at IRF (i.e., oncology, transplant, and advanced heart failure), the specialty clinics often serve as their medical home. However, our study would suggest that patients who also have PCP follow-up have a reduced risk of readmission. In the postintervention phase, the rate of 7-day follow-ups went from 61% to 10%, which suggests that this intervention is not sustainable with the current systems of care. The majority of rapid PCP follow-ups were secured through phone. Many of these PCP follows up were not acquired easily and required multiple follow-up phone calls across multiple days, lengthy hold times, and an inability of schedulers to adequately have patients seen within 7 business days. If rapid PCP follow-up is an intervention worth pursuing, then the process would require significant streamlining, or the adoption of a member of the care team who owned the responsibility of scheduling PCP follow-ups.
Moreover, in the postintervention phase, as shown in [Figure 3], the 30 day readmission rate in the debility population declined from 26% to 17%. A modified Hawthorne effect may explain this change. Residents, attendings, and staff at our facility spent 6 consecutive months aware of the study, which emphasized readmissions in high-risk debility patients. Therefore, subconscious changes to the way care was delivered and coordinated with a focus on transition of care may have led to such an improvement in 30-day readmissions despite the presence of the intervention.
| Conclusions|| |
There was a statistically significant reduction in 30-day readmissions with our intervention of rapid PCP follow-up post-IRF discharge for patients admitted with a primary diagnosis of debility. This lends support to our shared theory with Brooke et al. regarding rapid PCP follow-up serving as a way-point to recognize issues and errors before they escalate into a problem warranting readmission. For the “Act” phase of our PDSA cycle, it may be beneficial to institute a rapid PCP follow-up for all discharges to reduce 30-day readmissions and improve patient care although generalizability to all diagnoses was not specifically addressed in this project. In addition, this would require emphasis on future sustainability of the intervention as there was decreased adherence to PCP follow-up scheduling observed after the intervention phase formally ended. Further research should seek how to streamline the process so that rapid PCP follow-up becomes routine within the discharge process for each patient.
If 30-day readmissions are to become a quality care metric for IRFs as we expect, it may be in the best interest for the health of patients admitted with a debility diagnosis and the financial interest of IRFs to mandate a policy requiring rapid PCP follow-up for debility patients.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Ali AM, Gibbons CE. Predictors of 30-day hospital readmission after hip fracture: A systematic review. Injury 2017;48:243-52.
Benbassat J, Taragin M. Hospital readmissions as a measure of quality of health care. Arch Internal Med 2000;160:1074.
Bradley EH, Curry L, Horwitz LI, Sipsma H, Wang Y, Walsh MN, Krumholz HM. Hospital strategies associated with 30-day readmission rates for patients with heart failure. Circulation 2013;6:444-50.
Greenberg JK, Washington CW, Guniganti R, Dacey RG, Derdeyn CP, Zipfel GJ. Causes of 30-day readmission after aneurysmal subarachnoid hemorrhage. J Neurosurg 2016;124:743-9.
Guterman EL, Douglas VC, Shah MP, Parsons T, Barba J, Josephson SA. National characteristics and predictors of neurologic 30-day readmissions. Neurology 2016;86:669-75.
Joynt KE. Thirty-day readmission rates for medicare beneficiaries by race and site of care. JAMA 2011;305:675.
Kassin MT, Owen RM, Perez SD, Leeds I, Cox JC, Schnier K, et al
. Risk factors for 30-day hospital readmission among general surgery patients. J Am Coll Surg 2012;215:322-30.
Nguyen OK, Makam AN, Clark C, Zhang S, Xie B, Velasco F, et al
. Predicting all-cause readmissions using electronic health record data from the entire hospitalization: Model development and comparison. J Hosp Med 2016;11:473-80.
Nguyen OK, Makam AN, Clark C, Zhang S, Xie B, Velasco F, et al
. Vital signs are still vital: Instability on discharge and the risk of post-discharge adverse outcomes. J Gen Internal Med 2017;32:42-8.
Ottenbacher KJ, Karmarkar A, Graham JE, Kuo YF, Deutsch A, Reistetter TA, et al
. Thirty-day hospital readmission following discharge from postacute rehabilitation in fee-for-service medicare patients. JAMA 2014;311:604.
Brooke BS, Stone DH, Cronenwett JL, Nolan B, DeMartino RR, MacKenzie TA, et al
. Early primary care provider follow-up and readmission after high-risk surgery. JAMA Surg 2014;149:821.
Shen E, Koyama SY, Huynh DN, Watson HL, Mittman B, Kanter MH, et al
. Association of a dedicated post–hospital discharge follow-up visit and 30-day readmission risk in a medicare advantage population. JAMA Internal Med 2017;177:132.
Jack BW, Chetty VK, Anthony D, Greenwald JL, Sanchez GM, Johnson AE, et al
. A reengineered hospital discharge program to decrease rehospitalization: a randomized trial. Ann Internal Med 2009;150:178-87.
Misky GJ, Wald HL, Coleman EA. Post-hospitalization transitions: Examining the effects of timing of primary care provider follow-up. J Hosp Med 2010;5:392-7.
Wiest D, Yang Q, Wilson C, Dravid N. Outcomes of a citywide campaign to reduce medicaid hospital readmissions with connection to primary care within 7 days of hospital discharge. JAMA Network Open 2019;2:e187369.
Fisher RA. The logic of inductive inference. J R Stat Soc Series A 1935;98:39-54.
[Figure 1], [Figure 2], [Figure 3]