Results Summary
What was the research about?
Strokes occur when blood vessels to the brain leak or become blocked, cutting off the supply of oxygen to the brain. People who have had a stroke may be at risk of having another one. High cholesterol or an irregular heartbeat, called atrial fibrillation, can increase risk for another stroke. Different types of medicine may help reduce this risk.
The research team looked at how well different types of medicine helped people avoid returning to the hospital or prevented major heart problems or another stroke. The team looked at
- Blood thinners. Warfarin is a common blood thinner that keeps blood clots from forming. The team looked at how well warfarin works compared with no medicine in people with atrial fibrillation. The team also compared warfarin with newer blood thinners called NOACs.
- Statins. Statins are medicines that help lower cholesterol levels in the blood. Statins can reduce the risk of stroke and heart attack. They can also cause side effects, such as muscle aches. The research team compared patients who received different levels of statins with patients who received no statins.
What were the results?
Blood thinners. Patients with atrial fibrillation who received warfarin had a lower risk for major heart problems and spent more days at home than those who received no medicine. Compared with patients who received warfarin, patients who received NOACs had a lower risk for major heart problems and spent more days at home.
Statins. Patients who received statins had a lower risk for major heart problems and spent more days at home than those who received no statins. Patients receiving either a high-dose or low-dose statin had similar risk for major heart problems and a similar number of days spent a home.
Who was in the study?
The research team reviewed health and Medicare records for 12,552 patients in the blood thinner studies and 77,468 patients in the statin studies. All patients had a stroke and were ages 65 and older. Patients in the blood thinner studies had atrial fibrillation. Patients in the statin studies were not taking statins before they had a stroke.
What did the research team do?
The research team looked at patients’ health and Medicare records for two years after patients were treated for a stroke.
Patients, caregivers, healthcare providers, and agencies provided input on the study’s design. Patients also provided input on enrollment, analysis, and sharing results.
What were the limits of the study?
Medicare records may not capture all medicines and treatments that patients receive. Also, the research team couldn’t tell if patients took their medicines. If patients weren’t taking their medicines, the study’s results might be different. The team only looked at two years of records. Differences in long-term effects of medicines may not show up until after two years.
Future research could look at patients’ health records for longer than two years.
How can people use the results?
Patients and their doctors can use these results when considering treatments after a stroke.
Professional Abstract
Objective
To compare the safety and effectiveness of different types of postdischarge stroke care, including blood thinners and statins
Study Design
| Design Element | Description |
|---|---|
| Design | Observational: cohort study |
| Population | Blood thinners: Medical records for 12,552 patients treated for atrial fibrillation who did not take blood thinners before their hospitalization Cholesterol medicines: Medical records for 77,468 patients who did not take statins before their hospitalization |
| Interventions/ Comparators |
Blood thinners:
Statins:
|
| Outcomes | Home time, risk of major adverse cardiovascular events |
| Timeframe | 2-year follow-up for primary outcome |
This retrospective observational cohort study used patient medical records to compare the safety and effectiveness of postdischarge stroke care on improving home time and reducing major adverse cardiovascular events. One analysis compared different types of blood thinners for people being treated for atrial fibrillation, and another compared different types of statins for people who were not previously taking statins. Home time was the total number of days alive and out of a hospital or skilled nursing facility following discharge after a stroke hospitalization.
Researchers analyzed records from a stroke registry that linked its data with Medicare claims for patients ages 65 and older who were discharged from the hospital after having a stroke. The blood thinner analysis included medical records for 12,552 patients treated for atrial fibrillation between 2009 and 2014. The statin analysis included medical records for 77,468 patients between 2007 and 2011.
Patients, caregivers, healthcare providers, and representatives from healthcare systems, government agencies, and professional societies provided input on the study’s design. Patients also provided input on enrollment, analysis, and sharing results.
Results
Blood thinners
- Warfarin versus no blood thinner. Patients with atrial fibrillation treated with warfarin had a lower risk for major cardiovascular events (p<0.001) and spent an average of 48 more days at home after discharge (p<0.001) than patients with atrial fibrillation who received no blood thinner.
- Warfarin versus non-vitamin K antagonist oral anticoagulants (NOACs). Patients with atrial fibrillation treated with NOACs had a lower risk for major cardiovascular events (p<0.001) and spent an average of 16 more days at home after discharge (p<0.001) than patients with atrial fibrillation who received warfarin.
Statins
- Statins versus no statins. Patients receiving statins had a lower risk for major cardiovascular events (p<0.001) and spent an average of 28 more days at home (p<0.001) than patients who received no statin.
- High-dose statin versus low-dose statin. The two groups did not differ significantly in risk of major cardiovascular events and days at home.
Limitations
Medical records may not fully capture all treatments and medicines that patients receive, or whether patients take medicines after hospital discharge. If patients discontinued medicines after having a stroke, that may have influenced the findings. Study follow-up was limited to two years; researchers may need a longer time period to understand long-term stroke outcomes.
Conclusions and Relevance
In this study, warfarin and NOACs helped patients with atrial fibrillation remain at home and lowered their risks for major cardiovascular events after a stroke. Patients with atrial fibrillation taking NOACs spent more days at home and had lower risk of cardiovascular events than patients taking warfarin. Patients taking statins had lower risk of cardiovascular events and more home time than patients not taking statins.
Future Research Needs
Researchers could follow participants for longer than two years to explore the long-term implications of blood thinners and statins for patients who have had a stroke.
Final Research Report
View this project's final research report.
More to Explore...
Audio Interview
JAMA Neurology Author Interviews
In an interview as part of the journal's "Author Interviews" series, Adrian F. Hernandez, MD, MHS, discusses what prompted him to launch the study, and what’s new and additive about this study. He notes that it’s one of the largest studies with its focus, and that it enrolled many older patients who have not traditionally been enrolled in studies with this focus.
Journal Citations
Article Highlight: This study compared outcomes of older adults who have atrial fibrillation and have had a stroke. In a new paper published in JAMA Neurology, the research team reports that patients who were prescribed direct oral anticoagulants when discharged from the hospital following a stroke had relatively better outcomes—including more days at home and fewer adverse events—than patients prescribed the blood thinner warfarin. This reinforces findings from prior research on the subject.
Results of This Project
Related Journal Citations
Stories and Videos
PCORI Stories
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot 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 descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented and the researchers made changes or provided responses. Those comments and responses included the following:
- The reviewers noted that aim 3 of the study, involving the development and testing of a novel decision aid for stroke survivors, originally was not reported in the draft report. The researchers added information about the decision aid they created in the methods, results, and discussion sections of the report.
- The reviewers were unclear on how the retrospective and prospective study cohorts aligned with the aims, and how the two populations differed in characteristics and in what outcomes the researchers measured. The researchers explained that they designed the prospective cohort to evaluate the patient-reported outcomes that are typically not available in registry data. They noted that depression and antidepressant use are likely to affect these outcomes, so these factors were specifically examined in the prospective group. Similarly, they investigated statin and oral anticoagulant use in the retrospective cohort because subsets of the stroke survivor population used those medications.. Therefore, they needed the larger sample size that the registry-based population provided in order to detect a meaningful effect.
- The reviewers wondered how clinically meaningful some of the observed associations were, since the study’s large sample size allowed for differences of relatively small magnitude to be statistically significant. The researchers agreed this was a good point and explained that they used methods to help minimize false positives and overly strong conclusions based on multiple comparisons. They added that they felt that differences they described, like a 9-13 percent reduction in major adverse cardiovascular events, were clinically meaningful.
- The reviewers asked for more information on the background of patient coinvestigators and how the researchers selected them. The researchers said they used American Heart Association resources to identify stroke survivors within their network of volunteers. All patient coinvestigators turned out to be women who had strokes between their 20s and their 50s. Despite the homogeneity in this group, these patient coinvestigators, as active members of stroke support groups, tried to speak from the perspective of a wide range of stroke survivors rather than only from personal experience.
- The reviewers requested more detail on the study’s use of focus groups, their demographic makeup, and the methods used to survey them. The researchers said that the focus groups were a small part of the project and did not contribute to the project’s aims, so the researchers did not formally record the information the reviewers requested. The researchers explained that their goals for the focus groups were to confirm that the project’s aims were meaningful to a wide range of stroke patients and to inform subgroup analyses.