3 Screen Electronic Alert

Background and Significance:

Scope of the Problem: VTE is the third most common cardiovascular disease (heart attack and stroke are more common) but is the most preventable medical problem among patients hospitalized with other illnesses. DVT often causes chronic venous insufficiency, a potentially disabling and painful condition that lowers the quality of life. Acute PE may cause death or disabling chronic thromboembolic pulmonary hypertension. After discharge from a short-term hospital stay, one study reported that the incidence of DVT was 48 per 100,000 and the incidence of PE (with or without DVT) was 23 per 100,000.9 In addition, the incidence of VTE increases dramatically with age.10-11 Clinical manifestations of DVT often include swelling, pain, and extremity discoloration and may lead to PE. Manifestations of PE may include dyspnea, pleuritic pain, cough, and hemoptysis or may be asymptomatic.12-13 The 28-day mortality rate for a first VTE event is alarmingly high, approximately 11%.14 In addition, 3-month mortality rates for PE have been reported as high as 17%.15 However, appropriate prophylaxis-using anticoagulants, mechanical measures, or both anticoagulants plus mechanical measures-can prevent most episodes of VTE16. For example, mechanical measures as simple as "low tech" and inexpensive graduated compression stockings can reduce VTE rates.16a, 16b, 16c

Burden to Patients: VTE is associated with complications including recurrence of DVT or PE, chronic thromboembolic pulmonary hypertension, death, or the subsequent development of post-thrombotic syndrome. The risk of VTE recurrence is greatest within 6-12 months after the first event and continues for at least 10 years,17 and is a risk factor for death after PE.18 The post-thrombotic syndrome can have devastating effects on the patients' quality of life and is associated with venous hypertension that may cause pain, edema, skin changes, varicose veins, and skin ulcers. Valve dysfunction, which can lead to venous hypertension, is most often caused by a previous DVT.19 The cumulative incidence of the development of postthrombotic syndrome after a VTE event is high: 17% at one year, 23% at 2 years, and 29% at 8 years, with 30% of these cases classified as severe.20

Burden to Society: Direct medical costs to the United States health care system related to VTE events and complications approximate at least $600 million annually. Bullano and colleagues21 estimate that managed care plans pay a median of $3,131 per incident DVT, $6,424 per incident PE, and $12,200 for a DVT plus PE combination event. A 2004 registry of 5,451 patients with ultrasound-confirmed DVT at 183 institutions in the United States revealed that DVT led to hospitalization in 80% of those diagnosed as outpatients. Those initially diagnosed with DVT as inpatients remained hospitalized for a median of 12 days.7

Role of computerized healthcare delivery systems:

The use of computerized physician order entry (CPOE) is growing and is currently utilized in about 10% of U.S. Hospitals.22 When linked with clinical decision support, it provides a unique and important tool to improve ordering decisions, thereby reducing medication errors, transcribing errors, decreasing the frequency of adverse drug events, and increasing compliance with evidence-based clinical guidelines.23 This leads to overall improved patient care. There is growing recognition of the importance and utility of CPOE technology, and currently, there is a concerted effort nation-wide to implement it in all hospitals and private practices. CPOE programs provide an excellent platform to increase the implementation of clinical practice guidelines, thereby changing physician behavior.

At Brigham and Women's Hospital, we have developed an electronic alert system, which is integrated with the CPOE system but functions independently. We have created risk score criteria by which the system determines the risk of DVT or PE for a particular patient. All the information needed to accomplish this risk assessment is stored in the computer system, allowing the process to run in an automated fashion. The system according to present criteria determines whether to generate an alert. To do this, the system checks for the existence of current orders for prophylaxis, using the Brigham and Women's Hospital CPOE system. The presence of medication class orders for anticoagulants or general care orders for mechanical prophylaxis prevents the computer from issuing an alert. We will enhance this computerized system to produce a series of alerts that provide more information related to prescription or omission of thromboprophylaxis. We aim to test this new system in our proposed study.

At Brigham and Women's Hospital, we also leverage the use of the multiple outpatient electronic medical records and inpatient CPOE systems to facilitate the process of medication reconciliation before, during, and after the hospital stay. This system aggregates medication data from electronic medical records (EMR) and CPOE systems so that clinicians can efficiently generate an accurate pre-admission medication list. Information collected with the use of this application subsequently supports the writing of admission and discharge orders by physicians, performance of admission assessment by nurses, and reconciliation of inpatient orders by pharmacists. Medication histories are thus seamlessly transferred during transition of care from inpatient admission to discharge to home, rehabilitation, or skilled nursing facilities. This efficient system will allow us to carefully document the use or omission of thromboprophylaxis orders. Furthermore, for 90 day follow-up, we will be able to track thromboprophylaxis prescriptions after hospital discharge by using a combination of the Brigham and Women's Hospital's pharmacy's electronic outpatient prescription reconciliation program and Partners' internally developed EMR, the Longitudinal Medical Record.

The Longitudinal Medical Record is used by our outpatient clinics to record clinical visit information such as history, notes, problems, medications, and allergies, and provides inpatient and outpatient test results, including imaging tests and laboratory blood tests, from 7 Partners-affiliated institutions: Brigham and Women's Hospital, Massachusetts General Hospital, Dana Farber Cancer Institute, Newton Wellesley Hospital, North Shore Medical Center, Faulkner Hospital, and Spaulding Rehabilitation Hospital. The Longitudinal Medical Record is especially useful for following patients who might undergo evaluation for a new problem, such as sudden onset of shortness of breath, at a suburban hospital such as Newton Wellesley Hospital, after a prior hospitalization at Brigham and Women's Hospital for treatment of a complicated medical illness with a high risk of VTE. In our 2005 randomized trial of electronic alerts for VTE prophylaxis,8 we relied upon the Longitudinal Medical Record to achieve 94% follow-up at 90 days of all patients enrolled at Brigham and Women's Hospital.

Between 2000 and 2004, we developed a computer generated alert system for VTE prophylaxis and tested it in a large randomized controlled trial at Brigham and Women's Hospital. The system was linked to the patient database, identifying patients at high risk for VTE (defined as ≥ 4 score points and shown in Table 1) who were not receiving prophylaxis. High-risk patients not receiving VTE prophylaxis were randomized to an intervention group in which their physicians received an electronic alert or a control group in which physicians were not alerted. Patients in the intervention group experienced a 41% reduction in the incidence of symptomatic VTE events within 90 days, without an increase in the risk of major or minor bleeding. However, prophylactic measures were prescribed for only 33.5% of the patients in the intervention group. For the remaining patients (66.5%), we were surprised that physicians declined to implement the suggested intervention, initiation of VTE prophylaxis.

There is still considerable scope for electronic methodology and technology to increase prophylaxis rates and reduce the rates of VTE. Therefore, we will focus these new study initiatives on improving our current methods and strategies to further reduce the incidence of VTE.

Significance of our proposed studies:

Our projects will address several gaps in our understanding of the issues related to implementation of VTE prophylaxis in hospitalized patients. We aim to identify subgroups in the hospitalized population that are at especially high risk for VTE. An improved and novel automated electronic alerting system will facilitate this identification. Our new alert system will be much more rigorous than our prior protocol. The new alert will have three sequential screens rather than the single alert screen that we had in the prior version.

This computerized system will identify hospitalized patients at high risk for VTE and will analyze their records to determine whether they have received thromboprophylaxis orders. If the system identifies any high-risk patient who does not have orders for thromboprophylaxis, then a series of prompts will be issued to the ordering physician. The physician whose patient is at risk for VTE will be requested to institute prophylaxis. If the automated request is not followed, a series of alerts will query reasons for withholding preventive measures. Further prompts will lead to centralized computer-generated ordering of graduated compression stockings, unless the physician opts out and unselects this order. Important elements in the alerts include an up-to-date selection of pharmacological agents, doses based on renal clearance, and the documentation of reasons for withholding prophylaxis by the ordering physician.

We will follow all high-risk patients whose physicians received an alert. We will determine whether within 90 days, the patient has developed symptomatic DVT or PE. The Longitudinal Medical Record will facilitate follow-up after hospital discharge.

Through the use of this novel alert system, we will identify patients at high risk for VTE, maximize the implementation of VTE prophylaxis, learn the rationale for physicians' withholding anticoagulant or mechanical prophylaxis in some high-risk patients, and test whether this novel electronic intervention reduces clinically important episodes of PE and DVT. The success of this program will provide a useful model for dissemination of VTE preventive strategies nationwide and worldwide.