 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
Critical Care Specialist, Department of Pharmacy Practice; Clinical Assistant Professor, College of Pharmacy and Health Sciences Center, Mercer University, Atlanta, GA
Medication Use Evaluation/Adverse Drug Reaction Coordinator, Pharmacy CARE Team, University of Colorado Hospital; Clinical Assistant Professor and Drug Information Externship Co-Preceptor, School of Pharmacy, University of Colorado–Denver, Aurora, CO
Critical Care Clinical Pharmacy Specialist, University of Colorado Hospital; Associate Professor, School of Pharmacy, University of Colorado–Denver
Medication Policy/Business Coordinator, Pharmacy CARE Team; Co-Chair, Pharmacy and Therapeutics Committee, University of Colorado Hospital, Aurora
Chairman, Pharmacy and Therapeutics Committee, University of Colorado Hospital; Professor of Pulmonary Sciences, School of Medicine, University of Colorado–Denver
Reprints: Dr. Golightly, Health Sciences Library/Center for Drug Information, Education, and Evaluation–Campus Box A-003, 12950 E. Montview Blvd., PO Box 6508, Aurora, CO 80045, fax 720/848-4190, larry.golightly{at}uch.edu
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
OBJECTIVE: To comparatively describe the utilization of rVIIa upon implementation of an evidence-based guideline at a university hospital.
METHODS: With advisory direction from a multidisciplinary task force, an evidence-based guideline for use of rVIIa was developed, approved, and fully implemented. Assessment of appropriateness of use and retrospective review were required for all cases. Effects of these actions were evaluated by auditing and comparing rVIIa use in patients treated in two 6-month observation periods before and after guideline implementation. Outcomes assessed were proportions of patients deemed appropriate to receive rVIIa, compliance with dosing recommendations, and acquisition costs.
RESULTS: Twenty-two and 29 patients were treated in the periods before and after guideline implementation, respectively. Patient characteristics were similar, except more cardiothoracic surgeries were performed in patients treated before implementation of the guideline. Indications for rVIIa use were judged appropriate in 21 (95.5%) before-cases and in all (100%) after-cases. The dose was compliant in 1 (4.6%) before-case and 27 (93.1%) after-cases (p < 0.001). Mean dosages of rVIIa administered were 81.8 µg/kg and 45.3 µg/kg in before- and after-cases, respectively (p < 0.001). During the respective periods of observation, amounts of rVIIa purchased monthly averaged 42.6 mg and 21.8 mg, a 49% difference. Semiannual expenditures for rVIIa decreased approximately $110,000 following guideline implementation. Patient outcomes were similar.
CONCLUSIONS: A guideline based on currently available evidence can serve to sustain the clinical appropriateness of rVIIa therapy and substantially decrease costs.
Key Words: formulary management, hemophilia, recombinant human coagulation factor VIIa
Published Online, May 13, 2008. www.theannals.com, DOI 10.1345/aph.1L047
An issue currently burdening some US healthcare systems is the ever expanding use of drugs for noncompendial or off-label purposes. Current advancements in recombinant technology in the field of hematology, in particular, have provided clinicians with hemostatic agents that offer patients safe and effective treatment for life-threatening hemorrhagic situations.5
Recombinant human coagulation factor VIIa (rVIIa) is a vitamin K–dependent glycoprotein that is structurally similar to human plasma–derived factor VIIa. It is indicated for the prevention and treatment of bleeding episodes in patients with hemophilia A or B with inhibitors of factor VIII or factor IX, respectively.6 Because this agent rapidly promotes hemostasis and reverses coagulopathies, off-label use for prevention and treatment of hemorrhage unrelated to hemophilia is common. Reports of off-label use of rVIIa include treatment of hemorrhage associated with warfarin toxicity,7,8 upper gastrointestinal varices,9,10 liver disease,11,12 various solid organ transplants,13,14 hematopoietic stem cell transplantation,15 trauma surgery,16 cardiothoracic surgery,17 various general surgery populations,18 and intracerebral hemorrhage.19
For the treatment of bleeding episodes in hemophiliac patients with inhibitors of factor VIII or IX, the recommended single dose of rVIIa is 90 µg/kg intravenously.6 Despite minimal data suggesting dose-dependent efficacy, this dose is frequently reported as being applied in off-label use of the drug. However, higher cumulative doses of rVIIa have been associated with adverse effects including myocardial infarction, cerebrovascular infarct, venous thrombosis, and disseminated intravascular coagulation.20-22 In addition, the high cost of rVIIa limits routine use.
Motivated by concerns specifically regarding the efficacy, safety, and cost associated with rVIIa use, the University of Colorado Hospital (UCH) implemented an EBM guideline for use of the drug. The purpose of this study was to assess compliance with use of the guideline, compliance with dosing recommendations, and the clinical and economic impact of the guideline.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
UCH is a 411-bed tertiary level academic medical center located in Aurora, CO. UCH operates with an academic group practice medical staff. At the time of this project, the Pharmacy and Therapeutics (P&T) Committee of the Medical Board consisted of 23 voting members: 12 physicians and 11 nonphysicians with hospital-wide, multidisciplinary representation.
Before implementation of a guideline, rVIIa (NovoSeven, NovoNordisk Pharmaceuticals Inc., Princeton, NJ) was distributed from pharmacy after approval from blood banking services. However, concerns regarding the frequent off-label use of rVIIa and related safety, efficacy, and cost issues prompted the P&T committee to assemble a multidisciplinary task force to develop a guideline for appropriate use. This task force was composed of representatives from hepatology, cardiothoracic surgery, hematology, neurosurgery, anesthesiology, critical care/pulmonary sciences, and clinical laboratory/blood banking services. Nonphysician membership included pharmacists representing administrative, clinical, and academic areas.
Prior to the first meeting of the task force, an English-language search of MEDLINE (1967–September 2006) was conducted using paired MeSH terms for key and text word identification of hemorrhage, hemostasis, coagulopathy, surgery, critical care, transfusion, factor VII, and protocol. All citations involving humans were retrieved and the bibliographies of those articles were reviewed to obtain pertinent material not identified in the original search. Abstracts of recent critical care, emergency medicine, surgery, transfusion, and pharmacy conferences were reviewed to obtain unpublished data. After reviewing the data pertinent to their practice, each medical discipline developed recommendations for use of rVIIa specific to their patient populations. These drafts were reviewed by the task force as a whole and revised by a subgroup of 3 pharmacists and the P&T physician-chair. The guideline was redistributed electronically to the task force members for constructive feedback. This version received consensus approval from all involved clinical service directors prior to implementation (January 2007). The guideline consisted of documented clinical uses with corresponding dosing information as well as contraindications, precautions, and other considerations prior to use (Appendix I, www.hwbooks.com/pdf/appendices/L047.pdf).
The P&T committee recommended that the pharmacist processing the rVIIa order be responsible for ensuring that the indication and prescribed dose were compliant with the guideline. If differences were noted, the pharmacist was instructed to contact the prescribing physician to ascertain the appropriateness of the indication and dosage regimen. Noncompliance with the guideline was immediately reported to the prescribing physician. In discrepant cases, the P&T physician-chair, who was ultimately responsible for ensuring guideline compliance, was notified immediately. The pharmacist was required to document all cases of rVIIa use in terms of indication and dose. On a weekly basis, a report is generated from the pharmacy medication management software detailing rVIIa use and sent to the medication use evaluation coordinator for ongoing monitoring and subsequent hospital committee reporting.
Guideline education occurred over 2 weeks prior to implementation and consisted of in-service programs for pharmacists, verbal notification at committee and educational meetings, mail distribution of the guideline, and publication of the guideline in hospital newsletters and on the hospital intranet site. The guideline was implemented February 1, 2007.
The primary outcome of this study was compliance with the recommended guideline defined by the percentage of patients who received rVIIa for indications deemed appropriate after guideline implementation. Pharmacist involvement in promoting compliance with the guideline is also described here. Secondary endpoints were compliance with dosing recommendations after implementation, rVIIa acquisition costs before and after implementation, and patient outcomes. Data were collected for 6-month periods before (July 1 to December 31, 2006) and after (February 1 to July 31, 2007) guideline implementation. These periods were chosen to ensure that enough cases were available to adequately assess use and prescribing adjustments over time. Use data for January 2007 were excluded because guideline education and promotion occurred during this month. The following data, if available, were collected on every patient: demographics (age, sex, weight, primary diagnosis, secondary diagnoses, past medical history), location (eg, surgical floor), length of hospital and intensive care unit stay, mortality, bleeding cessation within 12 hours of rVIIa administration, and adverse events (catheter/line clotting, arterial or venous thromboembolic event) within 24 hours after rVIIa administration. There was no institutional mandate for documentation of hemostatic success or adverse events associated with rVIIa; this was left to the preference and professional judgment of the prescriber. Pharmacy purchasing expenditures were audited to obtain acquisition cost.
A power analysis was not conducted, as this evaluation was designed
primarily to assess guideline compliance after implementation. Data presented
as proportions were compared using the 
2 test unless
otherwise noted. Means were compared using Student's t-test where
appropriate. All tests were 2-tailed. A p value less than 0.05 was considered
significant. Data are presented as mean ± SD.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
|
During the 6 months before and after guideline implementation, prescriber compliance with guideline recommendations according to indication was observed in 21 (95.5%) and 29 (100%) cases, respectively. Doses of rVIIa were compliant with the guideline (40 µg/kg for all documented treatment indications in this patient population) in 1 (4.6%) before-case and 27 (93.1%) after-cases (p < 0.001). Documentation of pharmacist interventional effort to promote compliance with guideline dosage recommendations at the time of dispensing was not forthcoming in either of the 2 deviant after-cases, presumably due to either lack of education on the requirement or practice error. In both deviant after-cases, the prescribing physician was notified afterward of guideline noncompliance, with no subsequent deviations.
Complicity with guideline dosage recommendations for rVIIa led to significant decreases in dose. Total per patient mean dosages of rVIIa administered were 81.8 ± 25.4 µg/kg and 45.3 ± 1.6 µg/kg in the before- and after-implementation groups, respectively (p < 0.001). This includes 2 patients in the before-group and 2 patients in the after-group who received 2 doses of rVIIa (the additional doses in the after-group were compliant with the guideline). Dosages remained stable throughout the after-implementation observation period, with no deviations toward either higher or lower dosages discernable among prescribers over time.
Despite more patients receiving rVIIa after guideline implementation, decreases in individual doses corresponded to an overall decline in utilization. During the before- and after-implementation observation periods, average monthly use amounts of rVIIa were 42.6 mg and 21.8 mg, respectively. This represents a 49% drop in the amounts administered after guideline implementation.
Decreases in rVIIa utilization were also manifested as lower costs. Audited pharmacy purchasing expenditures for the drug during the 6-month before- and after-implementation periods were $217,993 and $107,979, respectively (Figure 1). Thus, the semiannual costs for rVIIa decreased by $110,014 after guideline implementation.
|
Despite the lack of an institutional requirement, documentation practices proved feasible. During the first 6 months after the rVIIa guideline was implemented, pharmacists completed required tracking documents for 28 (96.6%) of 29 treated patients. Pharmacist intervention was not documented on any rVIIa order and involvement of the P&T physician-chair was not required.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
Our procedures and these results generally are similar to those described in other recent reports.23,24 Using defined indication and dosing criteria with required senior physician authorization, guideline compliance was documented in 17 of 19 (89%) patients requiring rVIIa during the first 6 months after adoption of institutional policy changes for this medication at another teaching hospital.24 Comprehensive professional education and detailed follow-up were cited as major contributors to this level of guideline compliance and overall program success. Findings from our assessment support these conclusions regarding the foremost importance of education and follow-up. The previous assessments did not evaluate cost issues, whereas we showed substantial cost-minimization associated with guideline compliance.
The processes of development and implementation and the subsequent economic impact of an rVIIa guideline at UCH were successful because of several additional factors. The UCH pharmacy department representatives to the P&T committee were instrumental in voicing concern for the need of an institutional guideline for the administration of off-label rVIIa, given the lack of nationally recognized consensus guidelines for use. Similarly variable usage patterns from other academic centers,25,26 recent reports of adverse events with accompanying litigation,27 and a documented lack of reimbursement28 gave the UCH pharmacy department firm justification that implementation of a guideline for rVIIa was warranted.
Critical to this practice change was the decision by the physician-chair and cochairman of the P&T committee to convene a multidisciplinary task force to address guideline implementation. This task force included physician and nonphysician representatives of the necessary stakeholder departments involved with use of rVIIa. The convening of such a group of practitioners was crucial, not only for generating discussion about the pharmacoeconomic concerns of off-label rVIIa use, but also for full acceptance of an institutional guideline.29
The implementation and assessment process at UCH was in accordance with fundamental medication use evaluation principles, which state that the key to successful change in medication-use patterns is the need for collaborative efforts and institutional support among physicians, pharmacists, and administrators.30 The efforts at UCH to address and implement practice-changing policy are reflective of these principles and represent a successful multifaceted approach to addressing drug expenditure issues.
|
Limitations |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
Other limiting considerations include the apparent difference in the distribution of surgical subtypes between the before- and after-groups; however, the dose recommended in the guideline was the same across all types of surgical procedures, so the subtype of surgery should not have influenced the results obtained after the guideline was implemented. Assessments of the impact of the guideline on cost included rVIIa but not other blood products, procoagulant medications, or ancillary treatment measures for acute hemorrhage. However, our cost comparison comprised 2 groups of patients, both of whom received rVIIa. Although our guideline includes prior administration of blood products as a consideration for rVIIa use, we have no reason to believe that differences exist in the techniques or agents used for management of acute hemorrhage in these groups. A previous survey, including patients at our institution, indicated that rVIIa administration was not associated with changes in use of blood products or other resources for the management of acute hemorrhage.26
|
Conclusions |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
|
Footnotes |
|---|
Dr. MacLaren serves as a NovoNordisk advisor.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionLimitationsConclusionsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
