What was the research about?
After a stroke, patients can go from the hospital to either a rehabilitation, or rehab, center or a skilled nursing center until they are ready to go home. Rehab centers provide more hours of daily therapy than skilled nursing centers. Stroke patients may also go straight home from the hospital. Patients and their doctors need to choose the right place for patients to recover.
In this study, the research team wanted to learn if stroke patients recovered better at rehab centers or skilled nursing centers. The team also compared patients who went straight home from the hospital and saw a doctor within a week with those who waited longer to see a doctor.
What were the results?
After leaving the hospital, 22 percent went to a rehab center, 25 percent went to a skilled nursing center, and 44 percent of patients went straight home.
In the first year after a stroke, patients who went to a rehab center were more likely to recover better than patients who went to a skilled nursing center. Compared with patients who went to a skilled nursing center, those who went to a rehab center
- Were more likely to live for at least one year
- Spent more days at home
- Were less likely to go back to the hospital
- Were less likely to move into a nursing home
Among patients who went straight home, the team didn’t find differences between those who saw a doctor within a week and those who had a later follow-up visit.
Who was in the study?
The research team looked at medical records and insurance claims for 162,432 people who had a stroke between 2006 and 2008. Of these, 55 percent of rehab center patients and 62 percent of skilled nursing center patients had a stroke of at least moderate severity. All patients had Medicare. They received care at one of 1,192 hospitals taking part in a large national stroke registry.
What did the research team do?
The research team looked at
- How severe the patients’ strokes were
- If they had other health problems
- Where patients went and what care they had after they left the hospital
- How many patients died in the year after their stroke
- If patients went back to the hospital
An advisory group of patients and health professionals helped design the study and analyze the data.
What were the limits of the study?
The study included people ages 65 and older who went to hospitals that were part of a national stroke registry. Results may be different for people younger than 65 or who receive care at other hospitals.
Future research could look at how getting care at more than one type of center after a stroke affects patients’ health.
How can people use the results?
Patients and their doctors could use the results of this study to help decide where to recover from a stroke when patients leave the hospital but aren’t ready for home care.
(1) To describe the services used by stroke patients after being in the hospital and identify factors associated with receiving rehabilitation care following hospital discharge; (2) To compare the effectiveness of posthospital care provided by high-intensity inpatient rehabilitation facilities (IRFs) versus low-intensity skilled nursing facilities (SNFs) on 1-year outcomes among stroke patients who continued with short-term inpatient care after hospital discharge; (3) To examine if outcomes for stroke patients who receive follow-up with community-based providers within seven days of being discharged home from the hospital differ from those who receive later follow-up
|Design||Observational: cohort study|
|Population||162,432 acute ischemic stroke patients who were 65 years or older with Medicare coverage|
1-year survival, home time (days alive and out of inpatient care), rehospitalization, institutionalization into nursing home
|Timeframe||1-year follow-up for study outcomes|
This study examined data from 1,192 hospitals enrolled in the American Heart Association’s Get With The Guidelines® (GWTG)-Stroke registry to help stroke patients and clinicians decide where to continue recovery after hospital discharge. The research team linked the GWTG data with longitudinal Medicare claims and standardized patient-reported 1-year outcomes from the Adherence eValuation After Ischemic Stroke Longitudinal (AVAIL) study.
The research team analyzed data from 162,432 acute ischemic stroke patients who were ages 65 years or older, had Medicare coverage, and had received acute stroke care. Among patients who continued with short-term inpatient care immediately after hospital discharge, the team compared recovery outcomes of those who went to an IRF with those who went to a SNF. For patients who were discharged home, the team compared recovery outcomes of those who had follow-up with community-based medical providers within seven days of discharge and those who received later follow-up. The analysis used instrumental variables and inverse probability weighted estimation propensity score-based methods.
An advisory group including patients and health professionals helped design the study and analyze the data.
Among patients included in the analyses, 44% were discharged directly home, 22% were discharged to IRFs, and 25% were discharged to SNFs. Stroke patients discharged home had experienced minor strokes whereas about 55% of IRF patients and 62% of SNF patients had a National Institutes of Health Stroke Scale score greater than or equal to six, indicating a stroke of at least moderate severity.
IRF patients had a higher likelihood of better outcomes when compared with SNF patients.
- 1-year survival: 82% for IRF patients versus 61% for SNF patients (Risk ratio=0.92; 95% confidence interval [CI]: 0.86, 0.98)
- Home-time: an average of 271 days for IRF patients at 1 year versus an average of 196 days for SNF patients (Risk ratio=1.06; 95% CI: 1.02, 1.09)
- Rehospitalization: 54% at 1 year for IRF patients versus 68% for SNF patients (Risk ratio=0.94; 95% CI: 0.91, 0.98)
- Institutionalization into nursing home: 19% at 1 year for IRF patients versus 32% for SNF patients (Odds ratio=0.54; 95% CI: 0.33, 0.88)
Among patients discharged directly home, the research team did not find differences between those who received follow-up with community-based providers within seven days after being discharged and those who received later follow-up.
The study included people ages 65 and older who had Medicare and received care at hospitals that were part of the GWTG hospital-based registry. Findings may not be generalizable to non-Medicare populations or patients treated at other hospitals.
Conclusions and Relevance
Healthcare professionals need evidence to make informed decisions about where to discharge patients for postacute stroke care. The study found that patients who went to an IRF after hospitalization for an acute ischemic stroke had better outcomes than patients who went to an SNF.
Future Research Needs
Future research could examine the effect of complex posthospital care patterns with multiple care transitions on health outcomes.
Final Research Report
View this project's final research report.
Related Journal Citations
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also confirms that the research has followed PCORI’s Methodology Standards. During peer review, experts who were not members of the research team read a draft report of the research. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. Reviewers do not have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve how the research team analyzed its results or reported its conclusions. Learn more about PCORI’s peer-review process here.
In response to peer review, the PI made changes including
- Providing a detailed description of stakeholder engagement in the development and implementation of the research study, but also noting that the description might look like less engagement than typically seen in prospective studies because of the retrospective design of the research (which means that the data exist before the study is designed, which provides fewer opportunities for stakeholder input and participation).
- Responding to reviewer requests to identify which of their three analyses were considered primary, since the results of the three analyses differed. The researchers explained that although the report describes the rationale for using instrumental variables in the analyses, the study stakeholders and advisory committee were unable to reach agreement about which instrumental variable approach to use. The researchers added additional discussion of the strengths and limitations of the two instrumental variables used in these analyses, to provide a clearer understanding of the findings.
- Confirming that the authors decided which subgroup analyses and multivariate risk models to use during the course of the study rather than pre-specifying them.
- Providing a rationale, in response to reviewers’ requests, for the use of simple imputation to model missing data rather than multiple imputation.
- Adding a paragraph and table of results to describe the adjusted comparative findings of Aim 3, in response to reviewers’ comments that the findings were not adequately reported in the Results section. The researchers also added comments about the Aim 3 results to the Limitations section.
Conflict of Interest Disclosures
Study Registration Information
- Has Results