Venous thromboembolism (VTE) comprises deep vein thrombosis and pulmonary embolism (PE) and affects up to 600 000 people in the United States annually, with more than 100 000 dying from fatal PE. Approximately half of these VTE events are hospital associated. Optimal prevention of this largely preventable condition requires completion of 3 steps for hospitalized patients: accurate risk assessment (balancing the risks of thrombosis and bleeding); prescription of risk-appropriate VTE prophylaxis; and administration of all risk-appropriate VTE prophylaxis.1

Formed in 2005, our multidisciplinary Johns Hopkins VTE Collaborative initially focused on prescription of VTE prophylaxis-with a goal of ensuring that all hospitalized patients are prescribed the optimal VTE prophylaxis regimen appropriate for their individual risk profile. To achieve this goal, we developed a mandatory VTE risk assessment tool that an admitting clinician (physician, nurse practitioner, or physician assistant) would be required to complete during the admission process for every hospitalized patient. Coupled with robust, computerized clinical decision support, the risk assessment tool would recommend the appropriate VTE prophylaxis regimen for each patient based on their calculated risk profile.2 After implementation, risk-appropriate VTE prophylaxis prescription improved significantly among medically ill patients from 65.6% to 90.1%3 and among trauma patients from 66.2% to 84.4%.4

Despite significant improvement in risk-appropriate VTE prophylaxis prescription, many patients were still not being prescribed optimal care. Recognizing that trainees (interns and residents) complete the majority of admission orders and are primarily responsible for VTE prophylaxis prescriptions,5 we implemented an intervention to provide individualized feedback to interns and residents on the appropriateness of their VTE prophylaxis prescription practices. After providing monthly feedback to residents, risk-appropriate VTE prophylaxis prescription increased from 89.4% to 95.4% among surgical residents.6

Risk assessment and prescription of VTE prophylaxis are critical first steps, but administration of these medications by a nurse to an accepting patient is the final crucial step. Prescription of VTE prophylaxis does not ensure administration. In a sample of more than 100 000 prescribed doses, nearly 12% were not administered to hospitalized patients. Sixty percent of these nonadministered doses were documented as due to patient refusal.7 Furthermore, we found the majority of patients who developed VTE during their hospital stay received suboptimal prophylaxis by having an inappropriate prescription and/or missing doses of prophylaxis.8

A mixed-methods study including a survey of nurses and direct observation of nurse interactions with patients discovered gaps in communication and misconceptions about the importance of VTE prophylaxis.9 A common misconception was the suggestion that ambulation was a substitute for pharmacologic VTE prophylaxis, despite no evidence that ambulation effectively prevents VTE among hospitalized patients.10 To improve communication between nurses and patients, we developed nurse education modules to teach nurses about the importance of VTE prophylaxis and approaches to best practice VTE prevention. After educating all medical and surgical nurses, we found a 13% reduction in missed doses of VTE prophylaxis.11 The online module is freely available at the Anticoagulation Forum website (https://acforum.org/VTE-C1/story.html).

To further improve administration, we partnered with patients and family members via the National Blood Clot Alliance (https://www.stoptheclot.org), the North American Thrombosis Forum (https://thrombosis.org), ClotCare (http://www.clotcare.com/), and the Johns Hopkins Patient and Family Advisory Council to develop patient-centered education for hospitalized patients. On the basis of their feedback, we developed a patient education bundle that includes an educational handout, a video, and a structured conversation with a clinician to convey the harms of VTE, its signs and symptoms, and the importance of appropriate prophylaxis.12 All these written materials are freely available on the Armstrong Institute webpage (http://bit.ly/bloodclots). The video has been viewed almost 300 000 times on YouTube (https://www.youtube.com/watch?v=0o3yadu4DFw&t=1s). We tested delivery of this intervention, facilitated by a real-time alert that notified a nurse educator when a dose of prescribed VTE prophylaxis medication was not administered to a hospitalized patient. During this intervention, the nurse educator responded to each alert first by having a conversation with the documenting nurse and then, if appropriate, with the patient to present the education bundle. After implementation of the intervention, missed doses of VTE prophylaxis decreased by 43%.13 To assess the generalizability of the intervention, we saw similar results when the same intervention was done at a community hospital14 and when we sent the real-time alert to charge nurses on each floor at the Johns Hopkins Hospital.15

Many quality measures exist to assess VTE prevention practice in hospitals, though none accurately measure each step in the care delivery process that represents defect-free VTE prevention.1 Successful VTE prevention requires hospitals to hold themselves accountable to stringent measures and critically assess opportunities to improve. We suggest that health system leaders implement a multifaceted approach to VTE prevention for all hospitalized patients. First, hospitals should adopt a patient population-specific, validated VTE risk assessment tool and ensure that every patient admitted has a standardized VTE risk assessment. Second, there must be a mechanism to ensure that patients are prescribed VTE prophylaxis that is evidence-based and appropriate for their individualized risk profile. For low-risk patients, who may experience more harm from anticoagulation than benefit, no prophylaxis is necessary. Third, it is essential to measure nonadministration of prescribed VTE prophylaxis. We have discovered the widespread misbelief that ambulatory patients do not need VTE prophylaxis, leading nurses to withhold VTE prophylaxis medication. Finally, VTE outcomes should be measured both in the hospital and after discharge, with keen attention to potential defects in prevention practices. While practice improvement efforts should be focused on the entire hospitalized patient population, it is critical to better understand factors associated with the development of hospital-associated VTE and optimize prevention practices.

REFERENCES