Comprehensive Post-Acute Stroke Services (COMPASS)

The purpose of this pragmatic study is to investigate whether implementation of a comprehensive post-acute stroke service model that integrates Early Supported Discharge (ESD) and Transitional Care Management (TCM) for stroke survivors discharged home improves functional outcomes post-stroke, reduces caregiver stress, and reduces readmission rates.

Stroke mortality is 20-40% higher in North Carolina (NC) than in the overall United States. After discharge, stroke patients are at high risk for complications. Although a model of stroke post-care (early supported discharge) exists in Europe and Canada, it has not been adapted for and tested in the US, although patients and stakeholders attest that post-acute care does not meet their needs. Transitional care services from hospital to home are now reimbursed by Centers for Medicaid and Medicare Services (CMS), but only for 30 days after discharge. The study team proposed a pragmatic, cluster randomized trial of 41 NC hospitals to determine the effectiveness of COMprehensive Post-Acute Stroke Services (COMPASS), a patient-centered intervention uniting transitional care management services and elements of early supported discharge in stroke patients discharged directly home. The study team will build on the successful North Carolina Stroke Care Collaborative (NCSCC) registry, a prospective stroke database in which 51 (of 113) hospitals in NC enroll patients. In preparation for COMPASS, the study team engaged these hospitals via webinars. Over 80% of NCSCC hospitals demonstrated an interest in participation and provided letters of support. The main question of this pragmatic trial is: Does implementation of COMPASS for all stroke patients discharged directly home improve functional outcomes as measured by the Stroke Impact Scale-16 (SIS-16) at 90 days post-stroke? The primary aim is to: compare the COMPASS model versus usual care on stroke survivors' self-reported functional status at 90 days post-stroke. The secondary aims are to determine if the COMPASS model affects: (1) caregiver strain (Modified Caregiver Strain Index); (2) self-reported general health; (3) disability (Modified Rankin Score); (4) self-reported physical activity; (5) depression (PHQ-2); (6) cognition (MoCA 5-min protocol); (7) medication adherence (Morisky Green Levine Scale-4); (8) self-reported falls; (9) self-reported fatigue (PROMIS Fatigue Instrument); (10) satisfaction with care; (11) secondary prevention - home blood pressure monitoring; (12) self-reported blood pressure; (13-15) all-cause hospital readmissions at 30-days, at 90-days and at 1 year after index discharge; (16-17) mortality at 90-days and at 1 year after index discharge; (18-20) healthcare utilization (emergency department visits, admissions to skilled nursing facilities/inpatient rehabilitation facilities); and (21) use of transitional care management billing codes. This study will also evaluate the effectiveness of the COMPASS Intervention in key patient subgroups based on race, sex, age, diagnosis (stroke versus TIA), stroke severity, and type of health insurance. English and Spanish-speaking patients ages 18 and older who are admitted to a participating hospital with a diagnosis of ischemic or hemorrhagic stroke or transient ischemic attack and discharged from acute care hospitalization to home will be included (about 6,000 patients/year). Participating hospitals will be randomized (stratified by stroke volume and primary stroke center status) to receive COMPASS or usual care (control group) in Phase 1. In Phase 2, usual care hospitals will cross over to COMPASS, while the early intervention hospitals sustain the intervention using hospital-based resources. The trial has three integrated intervention components: (1) COMPASS, which combines transitional care services provided by advanced practice providers (APPs) and early supported discharge services coordinated by the Post-Acute Coordinators (PAC); (2) COMPASS-funded post-acute care coordinators who will engage patient and stakeholder communities to improve post-acute stroke comprehensive stroke services; and (3) development of a stroke metrics scorecard for participating sites. Well-trained APPs and coordinators will have access to online learning and ongoing support/consultation from WFBH personnel and board-certified vascular neurologists. The study team will assess 90-day and 1-year outcomes. Outcomes at 90-days will be assessed by telephone surveyors blinded to patient's group assignment. Patients will be informed about COMPASS in the hospital and can opt out of 90-day phone follow-up at any time. Those who agree to be surveyed will be asked to provide informed consent at the 90-day phone call to collect outcomes data. This proposal is led by three highly experienced researchers as co-principal investigators. The team includes expertise in stroke care, large clinical trials, biostatistics, managing clinical registries, survey and acquisition of patient or proxy-reported outcomes, community-based practice improvement, building community coalitions to reduce readmissions, claims analyses, registry management, translating evidence into practice with large multi-site collaboratives, and engaging patients and stakeholders in research. The planning phase of this project has been guided by patients and stakeholders. Each community will form a community resource network to advise and support the implementation of COMPASS, provide feedback to the team, and help create sustainability. If the COMPASS model shows effectiveness, engaged patients and stakeholders will be key partners to disseminate and implement COMPASS throughout the state and beyond.

Early Supported Discharge, Transitional Care, Post-acute Stroke Care, Pragmatic Trial, Recovery and Rehabilitation, Secondary Prevention

Participating hospitals randomized to the usual care group will provide their usual, post-acute stroke care to their patients.

Participating hospitals randomized to the intervention will change the structure and process for delivery of post-acute stroke care.

A Post-Acute Coordinator (PAC) will visit each patient prior to discharge from the hospital. Patient will receive a follow-up telephone call two days after having been discharged. 7-14 days after discharge, the patient will attend post-acute stroke clinic visit and receive an assessment from an Advanced Practice Provider (APP), a brief patient-reported functional assessment to generate an individualized Care Plan, and referrals from an APP. The patient's primary caregiver will be assessed to ensure caregiver availability and ability to support the patient and the caregiver's ability to cope with the new challenges of caregiving. Patient will receive a call at 30 and 60 days post-discharge for follow-up of functional status, recovery, risk factor management and their access or utilization of recommended services.

16-item survey to assess the difficulty level of performing basic physical activities; scores range from 0-100; higher scores correspond to more favorable outcomes

13-item survey to measure strain that caregivers may experience; scores range from 0-100; higher scores indicate more caregiver burden

Self-reported general health is a single question to rate their general health. Responses on a 5-point Likert Scale (Excellent, Very Good, Good, Fair, or Poor) will be analyzed as a continuous variable. Scores range from 95-15 with a higher score indicating better health.

to measure the degree of disability or dependence; scores range from 0-6; higher scores correspond to less favorable outcomes

Participants are asked whether they walked continuously for at least 10 minutes on any of the last seven days, how many of those days they walked continuously for at least 10 minutes and how many minutes they walked, on average, each day. The physical activity endpoint will be self-reported total number of minutes walked during the past seven days.

Based on answers to Patient Health Questionnaire 2-Item (PHQ-2) which is a 2-item questionnaire to determine the frequency of depressed mood; scores range from 0-6; higher scores correspond to less favorable outcomes

4-item questionnaire to determine vascular cognitive impairment; scores range from 0-30; higher scores are more favorable

4 items with yes/no response options; scores range from 0-4; higher scores correspond to less medication adherence

Participants are asked 4 questions to determine whether they have fallen (yes versus no) since hospital discharge, whether or not the fall resulted in a doctor/emergency room visit, whether they have fallen multiple times since discharge, and how many times they have fallen since discharge. Analysis of falls will be based on incidence of any fall since hospital discharge (no falls versus at least one fall).

4-question instrument to determine level of fatigue; higher scores correspond to less favorable outcomes; The total raw score is obtained by summing individual question scores and has a range of 4-20. For analysis, raw scores are translated into T-scores which range from 33.7 - 75.8. The T-score rescales the raw score into a standardized score with a mean of 50 and a SD of 10.

6 questions to determine satisfaction with care; scores range from 0-100; higher scores correspond to higher satisfaction of care

Participants are asked 2 questions to determine whether they monitor their blood pressure at home (yes or no) and, if they answer in the affirmative, how frequently (daily, weekly, and monthly). Home blood pressure monitoring was analyzed as a dichotomous endpoint (monitoring with any frequency versus no monitoring).

1 question to determine self-reported blood pressure. Self-reported systolic and diastolic BP will each be analyzed as a continuous endpoint. In addition, self-reported systolic and diastolic BP will be used to create a dichotomous hypertension endpoint (systolic BP >= 140 versus systolic BP < 140).

Deaths within 90 days of index discharge were ascertained from the North Carolina State Death Index as well as insurance claims beneficiary summary files (i.e. FFS Medicare). A death identified in either database is considered a death.

Number of Subjects With All-cause Mortality Using NC State Death Index & Fee-For-Service (FFS) Medicare

Deaths within 1 year of index discharge were ascertained from the North Carolina State Death Index as well as insurance claims beneficiary summary files (i.e. FFS Medicare). A death identified in either database is considered a death.

Number of Subjects With Claims-based Admissions to Skilled Nursing Facilities (SNF) and Inpatient Rehabilitation Facilities (IRF)

Inclusion Criteria: English and Spanish speaking stroke patients with diagnosis of ischemic stroke, hemorrhagic stroke or TIA who are discharged home from participating hospitals Must be 18 years of age and older at the time of the stroke Exclusion Criteria: Excludes subdural or aneurysmal subarachnoid hemorrhage

De-identified data will be made available to researchers through the PCORI-Designated Repository in accordance with their Data Sharing Policy.

Condon C, Lycan S, Duncan P, Bushnell C. Reducing Readmissions After Stroke With a Structured Nurse Practitioner/Registered Nurse Transitional Stroke Program. Stroke. 2016 Jun;47(6):1599-604. doi: 10.1161/STROKEAHA.115.012524. Epub 2016 Apr 28.

Gesell SB, Klein KP, Halladay J, Bettger JP, Freburger J, Cummings DM, Lutz BJ, Coleman S, Bushnell C, Rosamond W, Duncan PW. Methods guiding stakeholder engagement in planning a pragmatic study on changing stroke systems of care. J Clin Transl Sci. 2017 Apr;1(2):121-128. doi: 10.1017/cts.2016.26. Epub 2017 Feb 27.

Duncan PW, Bushnell CD, Rosamond WD, Jones Berkeley SB, Gesell SB, D'Agostino RB Jr, Ambrosius WT, Barton-Percival B, Bettger JP, Coleman SW, Cummings DM, Freburger JK, Halladay J, Johnson AM, Kucharska-Newton AM, Lundy-Lamm G, Lutz BJ, Mettam LH, Pastva AM, Sissine ME, Vetter B. The Comprehensive Post-Acute Stroke Services (COMPASS) study: design and methods for a cluster-randomized pragmatic trial. BMC Neurol. 2017 Jul 17;17(1):133. doi: 10.1186/s12883-017-0907-1.

Johnson AM, Jones SB, Duncan PW, Bushnell CD, Coleman SW, Mettam LH, Kucharska-Newton AM, Sissine ME, Rosamond WD. Hospital recruitment for a pragmatic cluster-randomized clinical trial: Lessons learned from the COMPASS study. Trials. 2018 Jan 26;19(1):74. doi: 10.1186/s13063-017-2434-1.

Bushnell CD, Duncan PW, Lycan SL, Condon CN, Pastva AM, Lutz BJ, Halladay JR, Cummings DM, Arnan MK, Jones SB, Sissine ME, Coleman SW, Johnson AM, Gesell SB, Mettam LH, Freburger JK, Barton-Percival B, Taylor KM, Prvu-Bettger J, Lundy-Lamm G, Rosamond WD; COMPASS Trial.. A Person-Centered Approach to Poststroke Care: The COMprehensive Post-Acute Stroke Services Model. J Am Geriatr Soc. 2018 May;66(5):1025-1030. doi: 10.1111/jgs.15322. Epub 2018 Mar 23.

Andrews JE, Moore JB, Weinberg RB, Sissine M, Gesell S, Halladay J, Rosamond W, Bushnell C, Jones S, Means P, King NMP, Omoyeni D, Duncan PW; COMPASS investigators and stakeholders. Ensuring respect for persons in COMPASS: a cluster randomised pragmatic clinical trial. J Med Ethics. 2018 Aug;44(8):560-566. doi: 10.1136/medethics-2017-104478. Epub 2018 May 2.

Duncan PW, Abbott RM, Rushing S, Johnson AM, Condon CN, Lycan SL, Lutz BJ, Cummings DM, Pastva AM, D'Agostino RB Jr, Stafford JM, Amoroso RM, Jones SB, Psioda MA, Gesell SB, Rosamond WD, Prvu-Bettger J, Sissine ME, Boynton MD, Bushnell CD; COMPASS Investigative Team. COMPASS-CP: An Electronic Application to Capture Patient-Reported Outcomes to Develop Actionable Stroke and Transient Ischemic Attack Care Plans. Circ Cardiovasc Qual Outcomes. 2018 Aug;11(8):e004444. doi: 10.1161/CIRCOUTCOMES.117.004444.

Bettger JP, Jones SB, Kucharska-Newton AM, Freburger JK, Coleman SW, Mettam LH, Sissine ME, Gesell SB, Bushnell CD, Duncan PW, Rosamond WD. Meeting Medicare requirements for transitional care: Do stroke care and policy align? Neurology. 2019 Feb 26;92(9):427-434. doi: 10.1212/WNL.0000000000006921. Epub 2019 Jan 11.

Halladay J, Bushnell C, Psioda M, Jones S, Lycan S, Condon C, Xenakis J, Prvu-Bettger J; COMPASS Investigative Team. Patient Factors Associated With Attendance at a Comprehensive Postacute Stroke Visit: Insight From the Vanguard Site. Arch Rehabil Res Clin Transl. 2019 Dec 21;2(1):100037. doi: 10.1016/j.arrct.2019.100037. eCollection 2020 Mar.

Gesell SB, Halladay JR, Mettam LH, Sissine ME, Staplefoote-Boynton BL, Duncan PW. Using REDCap to track stakeholder engagement: A time-saving tool for PCORI-funded studies. J Clin Transl Sci. 2020 Feb 6;4(2):108-114. doi: 10.1017/cts.2019.444. eCollection 2020 Apr.

Lutz BJ, Reimold AE, Coleman SW, Guzik AK, Russell LP, Radman MD, Johnson AM, Duncan PW, Bushnell CD, Rosamond WD, Gesell SB. Implementation of a Transitional Care Model for Stroke: Perspectives From Frontline Clinicians, Administrators, and COMPASS-TC Implementation Staff. Gerontologist. 2020 Aug 14;60(6):1071-1084. doi: 10.1093/geront/gnaa029.

Gesell SB, Coleman SW, Mettam LH, Johnson AM, Sissine ME, Duncan PW. How engagement of a diverse set of stakeholders shaped the design, implementation, and dissemination of a multicenter pragmatic trial of stroke transitional care: The COMPASS study. J Clin Transl Sci. 2020 Nov 5;5(1):e60. doi: 10.1017/cts.2020.552.

Pastva AM, Coyle PC, Coleman SW, Radman MD, Taylor KM, Jones SB, Bushnell CD, Rosamond WD, Johnson AM, Duncan PW, Freburger JK; COMPASS Investigative Team. Movement Matters, and So Does Context: Lessons Learned From Multisite Implementation of the Movement Matters Activity Program for Stroke in the Comprehensive Postacute Stroke Services Study. Arch Phys Med Rehabil. 2021 Mar;102(3):532-542. doi: 10.1016/j.apmr.2020.09.386. Epub 2020 Oct 22.

Duncan PW, Bushnell C, Sissine M, Coleman S, Lutz BJ, Johnson AM, Radman M, Pvru Bettger J, Zorowitz RD, Stein J. Comprehensive Stroke Care and Outcomes: Time for a Paradigm Shift. Stroke. 2021 Jan;52(1):385-393. doi: 10.1161/STROKEAHA.120.029678. Epub 2020 Dec 22.

Bayliss WS, Bushnell CD, Halladay JR, Duncan PW, Freburger JK, Kucharska-Newton AM, Trogdon JG. The Cost of Implementing and Sustaining the COMprehensive Post-Acute Stroke Services Model. Med Care. 2021 Feb 1;59(2):163-168. doi: 10.1097/MLR.0000000000001462.

Gesell SB, Bushnell CD, Jones SB, Coleman SW, Levy SM, Xenakis JG, Lutz BJ, Bettger JP, Freburger J, Halladay JR, Johnson AM, Kucharska-Newton AM, Mettam LH, Pastva AM, Psioda MA, Radman MD, Rosamond WD, Sissine ME, Halls J, Duncan PW. Implementation of a billable transitional care model for stroke patients: the COMPASS study. BMC Health Serv Res. 2019 Dec 19;19(1):978. doi: 10.1186/s12913-019-4771-0.

Duncan PW, Bushnell CD, Jones SB, Psioda MA, Gesell SB, D'Agostino RB Jr, Sissine ME, Coleman SW, Johnson AM, Barton-Percival BF, Prvu-Bettger J, Calhoun AG, Cummings DM, Freburger JK, Halladay JR, Kucharska-Newton AM, Lundy-Lamm G, Lutz BJ, Mettam LH, Pastva AM, Xenakis JG, Ambrosius WT, Radman MD, Vetter B, Rosamond WD; COMPASS Site Investigators and Teams.. Randomized Pragmatic Trial of Stroke Transitional Care: The COMPASS Study. Circ Cardiovasc Qual Outcomes. 2020 Jun;13(6):e006285. doi: 10.1161/CIRCOUTCOMES.119.006285. Epub 2020 Jun 1.

Bushnell CD, Kucharska-Newton AM, Jones SB, Psioda MA, Johnson AM, Daras LC, Halladay JR, Prvu Bettger J, Freburger JK, Gesell SB, Coleman SW, Sissine ME, Wen F, Hunt GP, Rosamond WD, Duncan PW. Hospital Readmissions and Mortality Among Fee-for-Service Medicare Patients With Minor Stroke or Transient Ischemic Attack: Findings From the COMPASS Cluster-Randomized Pragmatic Trial. J Am Heart Assoc. 2021 Dec 7;10(23):e023394. doi: 10.1161/JAHA.121.023394. Epub 2021 Nov 3.

Rosamond WD, Kucharska-Newton AM, Jones SB, Psioda MA, Lutz BJ, Johnson AM, Coleman SW, Schilsky SR, Patel MD, Duncan PW. Emergency department utilization after hospitalization discharge for acute stroke: The COMprehensive Post-Acute Stroke Services (COMPASS) study. Acad Emerg Med. 2022 Mar;29(3):369-371. doi: 10.1111/acem.14401. Epub 2021 Nov 1. No abstract available.