Overview
After hip or knee replacement surgery, patients are commonly prescribed blood-thinning medication, or anticoagulation, to lower the risk of blood clots. These clots can form in the leg veins and travel to the lungs, causing a pulmonary embolism, which can be life-threatening. Choosing the best anticoagulant after joint replacement is not only a clinical decision; it is also a preference-sensitive decision, because the available options differ in their benefits, risks, convenience, and cost.
This study examined how patients who had recently undergone hip or knee replacement valued the trade-offs among three commonly used anticoagulation strategies: aspirin, warfarin, and rivaroxaban. The research was designed as an adjunct to the Pulmonary Embolism Prevention after Hip and Knee Replacement (PEPPER) trial, a large pragmatic study comparing these medications.
Why patient preferences matter
In routine care, clinicians often focus on effectiveness and safety. However, in real-world decision-making, patients may weigh outcomes differently. One person may be most concerned about avoiding a pulmonary embolism at almost any cost. Another may strongly prefer a medicine with a lower chance of bleeding. A third may care most about avoiding frequent monitoring or out-of-pocket expenses.
This matters because anticoagulants are not interchangeable in how they affect daily life. Warfarin is effective but requires regular blood tests and has many food and drug interactions. Rivaroxaban is convenient because it does not usually require routine monitoring, but it may be more expensive and is still associated with bleeding risk. Aspirin is inexpensive and familiar, but it may be less potent than some alternatives in certain settings. Understanding these trade-offs helps clinicians individualize care rather than assuming one treatment fits all.
Study design
The investigators created a multimedia conjoint analysis survey. Conjoint analysis is a method that asks people to compare hypothetical treatment options described by different features, such as the chance of bleeding, the risk of blood clots, the risk of pulmonary embolism, and cost. By analyzing the choices patients make, researchers can estimate how much weight people place on each feature.
The survey used benefit and harm probabilities that matched those being studied in the PEPPER trial. A total of 192 patients who had undergone hip or knee replacement at the Medical University of South Carolina participated. They were surveyed 1 to 7 months after surgery and were eligible for the PEPPER trial. To identify distinct patterns in preferences, the researchers used k-means clustering, a statistical technique that groups people with similar response patterns into subgroups.
Main findings
Across the entire study group, the average ranking of outcomes showed that major adverse effects were considered important in a fairly balanced way. Patients cared about bleeding, venous thrombosis, and pulmonary embolism, with a somewhat stronger emphasis on avoiding pulmonary embolism. In other words, many patients recognized that preventing clots is critical after joint replacement surgery.
But the most important finding was that the average did not describe most individuals very well. Few patients had preferences close to the population average. Instead, the study found three distinct preference phenotypes, meaning three recurring patterns of how patients valued risks and benefits.
The three phenotypes were:
1. Thrombosis-focused values
These patients placed the greatest importance on avoiding blood clots, especially pulmonary embolism. Their preference pattern most closely aligned with rivaroxaban, a medication that is often viewed as a stronger anticoagulant option.
2. Balanced values
These patients gave substantial weight to both bleeding risk and clot prevention. Their values most closely aligned with aspirin, suggesting a preference for a treatment that balances safety, simplicity, and effectiveness.
3. Out-of-pocket-cost-focused values
These patients placed the greatest importance on treatment cost and practical affordability. Their preferences aligned most closely with aspirin or warfarin, both of which are generally less expensive than rivaroxaban.
The groups overlapped very little, suggesting that patients did not simply differ by a small degree. Rather, they clustered into qualitatively different types of decision-making.
What the results mean clinically
This study shows that patients recovering from hip or knee replacement do not all think about anticoagulation the same way. Some are willing to accept higher cost or more complexity if they believe the treatment offers stronger clot prevention. Others are more worried about bleeding, convenience, or finances.
For clinicians, this supports a shared decision-making approach. Instead of presenting anticoagulation as a purely technical prescription, the decision can be framed around a short discussion of what matters most to the patient. That conversation may improve adherence, satisfaction, and trust, because the selected medication is more likely to match the patient’s own values.
The findings also suggest that clinical trial results should be interpreted with individual preferences in mind. A medication may be statistically favorable on average, but the best option for any one patient may depend on how that person values bleeding risk, clot prevention, monitoring burden, and cost.
About the medications studied
Aspirin is widely available and low-cost. In orthopedic practice, it is often used in selected patients at lower thrombotic risk or as part of a broader prophylaxis strategy, depending on institutional protocols.
Warfarin has been used for decades and is effective, but it requires careful dose adjustment and regular INR testing to ensure the blood is not too thin or too thick.
Rivaroxaban is a direct oral anticoagulant that is taken orally and generally does not require routine laboratory monitoring. It is convenient for many patients, but affordability and bleeding concerns may influence whether it is preferred.
The study did not claim that one medication is best for every patient. Instead, it highlighted that the “best” choice may be different depending on patient priorities and the care context.
Strengths and limitations
A strength of this study is that it examined real patients after surgery, rather than asking healthy volunteers to imagine decisions in the abstract. The use of multimedia conjoint analysis also made the survey more realistic and easier to understand.
The study has limitations, however. It was conducted at a single academic medical center, which may limit how broadly the results apply to other populations. Preferences were measured after surgery, so patients’ views may have been influenced by their recovery experience. Also, while conjoint analysis is useful for revealing trade-offs, it is still a survey method and does not capture every factor that may affect actual treatment decisions in practice.
Bottom line
Among patients who had hip or knee replacement surgery, preferences for postoperative anticoagulation were not uniform. Instead, researchers identified three distinct preference phenotypes: thrombosis-focused, balanced, and cost-focused. These patterns suggest that anticoagulation choice should be individualized through shared decision-making, taking into account not only medical risk but also patient values, convenience, and financial considerations.
As joint replacement surgery becomes more common and anticoagulation choices remain varied, tools that better match treatment to patient preference may help improve both care quality and patient satisfaction.