- Email: [email protected]
Clinical and economic outcomes with appropriate or partial prophylaxis
Thrombosis Research 125 (2010) 513–517
Contents lists available at ScienceDirect
Thrombosis Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t h r o m r e s
Regular Article
Clinical and economic outcomes with appropriate or partial prophylaxis A.N. Amin a,⁎, J. Lin b, B.H. Johnson c, K.L. Schulman c a b c
University of California – Irvine, Orange, CA, USA sanofi-aventis, Bridgewater, NJ, USA Healthcare, Thomson Reuters, Cambridge, MA, USA
a r t i c l e
i n f o
Article history: Received 22 June 2009 Received in revised form 12 October 2009 Accepted 23 October 2009 Keywords: guideline adherence prophylaxis venous thromboembolism
a b s t r a c t Introduction: Despite the existence of evidence-based guidelines for venous thromboembolism (VTE) prevention, prophylaxis is often inappropriately prescribed. This study compared the efficacy, safety, and cost of appropriate (ACCP-recommended) prophylaxis with partial prophylaxis (not completely conforming to ACCP guidelines) in patients at-risk of VTE receiving enoxaparin or unfractionated heparin. Methods: The MarketScan® Hospital Drug Database from Thomson Reuters (January 2004–March 2007), was queried for medical and surgical patients at high risk of VTE, aged ≥ 40 years, and with a hospital stay ≥ 6 days. Univariate and multivariate analyses compared hospital-acquired VTE events, adverse events, and hospital costs between appropriate or partial prophylaxis discharges. Results: Of the 21,001 discharge records included, appropriate prophylaxis was received by 5136 (24.5%) patients. Compared with partial prophylaxis, appropriate prophylaxis was associated with significantly lower incidences of hospital-acquired pulmonary embolism (0.9% vs 0.5%; adjusted odds ratio [OR] 0.55, 95% confidence intervals [CI] 0.35–0.87, P = 0.010), and bleeding events (10.7% vs 5.1%; adjusted OR 0.57, 95% CI 0.50–0.66, P b 0.001). Total costs per discharge were lower for appropriate prophylaxis ($17,386 ± 12,004) than partial prophylaxis ($23,823 ± 19,783) with an adjusted mean difference of $6370 in favor of appropriate prophylaxis (P b 0.001). Conclusion: This retrospective study suggests that ACCP-guideline recommended appropriate prophylaxis reduces hospital-acquired pulmonary embolism and bleeding events in patients at-risk of VTE and is costsaving when total direct medical costs are considered. The substantial US clinical and economic VTE burden may, therefore, be reduced by improving prophylaxis adherence with guideline recommendations. © 2009 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism (VTE), which includes deep-vein thrombosis (DVT) and pulmonary embolism (PE), is a common complication during and after hospitalization for acute medical illness and surgery [1–3]. Almost a third of hospitalized patients are thought to be at risk of VTE annually and, given the considerable medical costs of treating VTE and its sequelae, VTE is associated with both substantial clinical and economic burdens [1,3–6]. Efficacious agents exist to prevent VTE, and these have been proven to be both well tolerated and cost effective [2,7–12]. Recommendations regarding the use of prophylactic agents are provided in best-practice guidelines, such as those from the American College of Chest Physicians (ACCP) [2]. Guideline-recommended appropriate prophylaxis is achieved only by providing the patient with the appropriate drug (or device), at the appropriate dose, and for the ⁎ Corresponding author. University of California – Irvine, 101 The City Drive South, Building 58, Room 110, ZC-4076H, Orange, CA 92868, USA. Tel.: +1 714 456 3785; fax: +1 714 456 3875. E-mail address: [email protected] (A.N. Amin). 0049-3848/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2009.10.018
appropriate duration. Despite the existence of best-practice guidelines, there is growing evidence in the literature that patients at-risk of VTE do not consistently receive prophylaxis, with only a minority receiving full and appropriate prophylaxis [3,13–16]. When VTE prophylaxis practices were investigated in 358 hospitals in 32 countries worldwide (n = 68,183), guideline-recommended prophylaxis was only received by 59% of surgical patients and 40% of medical patients [3]. Similarly, in 196,104 medical patients and 85,970 surgical patients at high-risk for VTE in 227 acute-care hospitals across the US, only 34% of medical patients and 32% of surgical patients received appropriate prophylaxis [13,14]. To date, few studies have evaluated the clinical and economic implications of failing to meet guideline recommendations for appropriate prophylaxis in everyday hospital practice. Use of a hospital administrative database provides information about a large, realworld population of medical and surgical patients at risk of VTE. Furthermore, such a database enables the study of symptomatic events alone, providing additional clinical relevance compared with trials, which assess a combination of symptomatic and asymptomatic events. The current study, conducted using a large US database, compared efficacy, safety, and costs of best-practice appropriate prophylaxis
514
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
with that of partial prophylaxis (prophylaxis not meeting bestpractice criteria) in medical and surgical patients at high risk of VTE. Methods Patient selection This study used a subset of the data from the MarketScan® Hospital Drug Database from Thomson Reuters. This database contains data from approximately 40 million discharge records from 550 acute care hospitals across the US. In 2006, the database contained information on one of every five discharges in the US. Cost data from hospital billing systems are available on more than a third of these records. Included patient discharges were aged ≥40 years and were admitted on or after January 1, 2004 and discharged on or before March 31, 2007. Eligible hospitalizations were included if they had a complete hospital stay of at least 6 days (to reflect a relatively high-degree of disease severity and potentially limited mobility) [17] and had a procedure and/or a diagnosis that deemed them at risk of VTE according to ACCP guidelines (medical risk factors were congestive heart failure, severe respiratory disease, active cancer, and stroke; surgical risk factors were general surgery, vascular surgery, gynecologic surgery, urologic surgery, laparoscopic surgery, major orthopedic surgery, and neurosurgery) [18]. In addition, patients were excluded if they were transferred from, or discharged to, another acute-care facility or if they had a potential contraindication to, or disease requiring, modification of ACCP-recommended anticoagulant therapy. Only discharges receiving either enoxaparin (with or without warfarin) or unfractionated heparin (UFH; with or without warfarin) were included. Enoxaparin was selected as the predominant lowmolecular-weight heparin provided in the database. Although ACCP guidelines suggest dosing requirements based on package inserts of the recommended anticoagulants, we assumed that best practice prophylaxis dosage per day was at least 40 mg for enoxaparin and at least 10,000 units for UFH. Prophylaxis duration was considered appropriate if it was received each day of the admission except 2 days for patients at medical risk (to allow for partial days of stay at admission and discharge), and each day of the admission after surgery except for 2 days in patients with surgical risk only. Outcomes The proportion of PE, DVT, and any hospital-acquired VTE event was compared between groups. Patients were classified as having VTE if they had a diagnosis of PE or DVT, and received at least 1 day of enoxaparin (≥100 mg per day) or UFH (≥25,000 units per day) during the qualifying hospitalization. The incidence of adverse events, including the proportion of major bleeding, non-major bleeding, and thrombocytopenia were also compared. Hospital utilization, charges (hospital billing), and costs (hospitalbased costs) were measured over the duration of the qualifying hospital stay to assess the economic implications of appropriate versus partial prophylaxis. Length of hospital stay for the qualifying admission and length of stay including readmission were calculated. Data are expressed as mean ± standard deviation. Analysis In addition to univariate analysis of clinical and economic outcomes, multivariate models were used for each outcome measure to adjust for differences in admission characteristics. Explanatory variables included length of stay, primary payer, gender, age, region, hospital teaching status, hospital bed size, risk categories (medical risk only, surgical risk only, both medical and surgical risk), use of antiplatelet therapy or mechanical prophylaxis, length of anticoagulation, and the number of personal risk factors.
Generalized linear models with gamma distributed error and log link were used to predict differences in hospital costs and charges between the groups. Generalized linear models with negative binomial distributed error and log link were used to predict differences in length of stay. Similar explanatory variables used in the clinical models were included in the economic multivariate analyses, with the exception of length of stay. A P-value of b0.05 was considered statistically significant. Results From an approximate 15 million records in the database, a total of 21,001 discharge records were included in the current analysis. Of these, only 5136 (24.5%) discharges received appropriate prophylaxis, with the majority of patients (15,865 [75.5%]) receiving partial prophylaxis. Patient demographics and clinical characteristics are shown in Table 1. A significantly greater proportion of females than males received appropriate prophylaxis (P b 0.001). The age of patients in the partial and appropriate prophylaxis groups was similar (65.6 ± 12.5 years and 65.2 ± 12.3 years, respectively), with no significant differences observed by age group. In general, the type of payer did not appear to influence the appropriateness of therapy, except that appropriate prophylaxis was more common than partial prophylaxis in Medicaid discharges (6.4% vs 5.3%, P = 0.003). Use of appropriate prophylaxis was significantly higher in medical discharges (58.8% vs 50.8%, P b 0.001); however, partial prophylaxis was observed in significantly more surgical discharges (29.0% vs 21.7%, P b 0.001). The remaining discharges in each group (20.2% of surgical and 19.5% of medical) received no prophylaxis. Clinical outcomes When hospital-acquired VTE events were compared, a VTE event occurred in 1.9% of the partial prophylaxis group compared with 1.4% in the appropriate prophylaxis group (Table 2). Similarly, the absolute proportions of both DVT and PE appeared higher in the partial prophylaxis group than the appropriate prophylaxis group (DVT: 1.2% vs 1.0%; and PE: 0.9% vs 0.5%, respectively). In the multivariate analysis, significantly fewer discharges in the appropriate prophylaxis group were likely to experience PE than in the partial prophylaxis group (odds ratio [OR] 0.55, 95% confidence interval [CI] 0.35–0.87, P = 0.010; Table 3). However, the likelihood of experiencing any hospital-acquired VTE event was not significantly different after adjustment. In the safety analysis, the probability of experiencing any adverse event (major bleeding, non-major bleeding, or thrombocytopenia) was significantly lower in the appropriate prophylaxis group (OR 0.57, 95% CI 0.50–0.66, P b 0.001; Table 3). Significant ORs were also observed for non-major bleeding and thrombocytopenia, but not for major bleeding (Table 3). It should be noted however that bleeding risk was not controlled for in the multivariate analysis, and patients who experienced bleeding were likely to have prophylaxis stopped, thus placing them in the partial prophylaxis group. Economic outcomes In a measure of hospital utilization, length of hospital stay was compared between the two groups. The mean total length of stay for the qualifying admission was shorter for patients receiving appropriate prophylaxis compared with partial prophylaxis (9.3 ± 4.8 days vs 11.2 ± 7.2 days, P b 0.001). In addition, a highly significant difference was observed with the mean total length of stay, including readmission, among appropriate prophylaxis and partial prophylaxis groups (10.3 ± 6.1 days vs 12.2 ± 8.3 days, respectively; adjusted difference − 2.95 days, P b 0.0001). It is also important to note that there was a variance in the proportion of time spent in the intensive
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
515
Table 1 Demographics and clinical characteristics of the partial and appropriate prophylaxis groups. Partial prophylaxis (n = 15,865)
Discharge status
Discharged to home Died Unknown Private insurance Medicare Medicaid Other government Workers’ compensation Other Female Male 40–44 45–54 55–64 65–74 75–84 85+ Northeast South Midwest West Yes No 1–199 200–299 300–499 500+
Primary payer
Gender Age, years
Region
Teaching status Licensed bed numbers
N
%
N
%
P-value
13,974 1062 829 5513 8338 844 126 61 983 7732 8133 712 2687 4004 4191 3312 959 140 12,836 2683 206 3747 12,118 1575 2955 6182 5153
88.1 6.7 5.2 34.8 52.6 5.3 0.8 0.4 6.2 48.7 51.3 4.5 16.9 25.2 26.4 20.9 6.0 0.9 80.9 16.9 1.3 23.6 76.4 9.9 18.6 39.0 32.5
4651 253 232 1800 2665 330 46 12 283 2768 2368 260 864 1342 1384 1014 272 60 4046 1007 23 1150 3986 655 892 1931 1658
90.6 4.9 4.5 35.1 51.9 6.4 0.9 0.2 5.5 53.9 46.1 5.1 16.8 26.1 27.0 19.7 5.3 1.2 78.8 19.6 0.5 22.4 77.6 12.8 17.4 37.6 32.3
b0.001 b0.001 0.044 0.698 0.405 0.003 0.483 0.110 0.073 b0.001 b0.001 0.089 0.849 0.202 0.454 0.081 0.047 0.067 b0.001 b0.001 b0.001 0.071 0.071 b0.001 0.043 0.080 0.792
care unit between groups, and this will likely have impacted on cost. For almost all charges and costs compared, including room and board, medical supplies and total pharmacy costs, partial prophylaxis was
Table 2 Clinical outcomes with partial prophylaxis versus appropriate prophylaxis. Partial prophylaxis (n = 15,865)
DVT PE Both DVT and PE Any hospital-acquired VTE Major bleeding Non-major bleeding Thrombocytopenia Any adverse event
Appropriate prophylaxis (n = 5136)
Number of events
%
Number of events
%
189 145 36 298 63 1,568 95 1,702
1.2 0.9 0.2 1.9 0.4 9.9 0.6 10.7
51 24 2 73 5 248 9 262
1.0 0.5 b 0.1 1.4 0.1 4.8 0.2 5.1
DVT, deep-vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.
Table 3 Clinical outcomes with appropriate prophylaxis compared with partial prophylaxis.
DVT PE Any hospital-acquired VTE Major bleeding Non-major bleeding Thrombocytopenia Any adverse event
Adjusted OR (95% CIs)⁎
P-value
1.09 0.55 0.90 0.42 0.58 0.46 0.57
0.625 0.010 0.464 0.072 b 0.001 0.032 b 0.001
(0.78–1.52) (0.35–0.87) (0.68–1.19) (0.17–1.08) (0.50–0.67) (0.22–0.93) (0.50–0.66)
Appropriate prophylaxis (n = 5136)
⁎Multivariate logistic regression analysis incorporating length of stay, primary payer, gender, age, region, hospital teaching status, hospital bed size, risk categories (medical risk only, surgical risk only, both medical and surgical risk), use of antiplatelet therapy or mechanical prophylaxis, length of anticoagulation, and the number of personal risk factors as explanatory variables. CI, confidence interval; DVT, deep-vein thrombosis; OR, odds ratio; PE, pulmonary embolism; VTE, venous thromboembolism.
associated with greater expenditure than appropriate prophylaxis (Table 4). The main exception to this was in enoxaparin costs, with these being higher in those receiving appropriate prophylaxis compared with partial prophylaxis ($560 ± 600 and $279 ± 539, respectively). When all associated costs were considered together, the mean total hospital costs per discharge were lower in the appropriate prophylaxis group ($17,386 ± 12,004) compared with the partial prophylaxis group ($23,823 ± 19,783). The cost difference per discharge was highly significant, in favor of appropriate prophylaxis, with similar considerable savings of $6437 (95% CI $5865–7010, P b 0.001) in the univariate analysis and $6370 (95% CI $6067–6670, P b 0.001) following adjustment in the multivariate analysis. Discussion This study demonstrates that among patients receiving enoxaparin or UFH, only 25% of medical and surgical patients at high risk of VTE received appropriate prophylaxis as recommended by the ACCP guidelines. Despite this, appropriate prophylaxis significantly reduced the probability of PE by almost a half. Moreover, in the safety analyses a significant reduction in the likelihood of non-major bleeding and thrombocytopenia was observed with appropriate versus partial prophylaxis. In addition to the reduction in clinical and adverse events with appropriate prophylaxis, substantial cost-savings were also observed, highlighting both the clinical and economic benefits of fully adhering to best-practice ACCP guidelines. Results of the present study are consistent with our previous investigations, which have also demonstrated the suboptimal use of appropriate prophylaxis [13,14]. When the proportion of appropriate versus partial prophylaxis was assessed in patients with acute medical conditions (n = 196,104), appropriate prophylaxis was received by 34% of medical patients, with the lowest proportions in trauma discharges without surgery (20.3%) and acute spinal cord injury discharges without surgery (20.8%) [13]. Similarly, in a recent study in 85,970 surgical discharges, appropriate VTE prophylaxis was only received by 32% of discharges [14]. The absence of any prophylaxis,
516
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
Table 4 Hospital charges and costs in the partial and appropriate prophylaxis groups.
Hospital charges All room and board Operating and recovery room Radiology Laboratory tests Medical supplies Respiratory therapy Mechanical prophylaxis Pharmacy Total anticoagulants Other inpatient pharmacy Total pharmacy Total hospital charges Hospital costs All room and board Operating and recovery room Radiology Laboratory tests Medical supplies Respiratory therapy Mechanical prophylaxis Pharmacy Total anticoagulants Other inpatient pharmacy Total pharmacy Total hospital costs
Partial prophylaxis (n = 15,865)
Appropriate prophylaxis (n = 5136)
$, mean ± SD
$, mean ± SD
10,022 ± 15,523 9811 ± 12,653 5566 ± 6528 7351 ± 8025 10,024 ± 15,557 4659 ± 9240 21 ± 124
5304 ± 9656 5734 ± 9203 4230 ± 4888 5255 ± 4313 5300 ± 9653 3907 ± 6327 15 ± 89
884 ± 1639 8883 ± 13,440
1203 ± 1146 6464 ± 7530
13,160 ± 18,956 69,997 ± 58,905
9878 ± 10,991 47,981 ± 33,743
2868 ± 4558 1887 ± 2435 675 ± 1025 944 ± 1371 2869 ± 4572 678 ± 1324 5 ± 29
1540 ± 2822 1088 ± 1566 499 ± 1063 605 ± 611 1540 ± 2822 636 ± 1103 4 ± 21
308 ± 546 1357 ± 1954
577 ± 591 1012 ± 1112
2181 ± 2879 23,823 ± 19,783
1935 ± 1843 17,386 ± 12,004
SD, standard deviation.
mechanical prophylaxis only and insufficient prophylaxis duration were common reasons for not receiving appropriate VTE prevention in these two studies [13,14]. The current study furthers the results of previous investigations [3,13–15], as both the clinical and economic impact of failing to adhere to best-practice guidelines for appropriate prophylaxis were also reported. Furthermore, assessment of efficacy and safety was performed not only across the large diverse sample, but also using multivariate analysis to adjust for different patient clinical characteristics and demographics. Randomized trials provide valuable information about the efficacy and safety of specific thromboprophylactic agents, usually in subsets of medical or surgical patients. However, a key strength of the current study is that the impact of thromboprophylaxis was investigated in a large real-world population, rather than a highly selected trial population. Furthermore, the current study assessed symptomatic VTE events alone, rather than both symptomatic and asymptomatic events which are generally investigated in clinical studies due to the need for statistical power. A previous retrospective study found that at-risk medical patients with a diagnosis of acute myocardial infarction, cancer, heart failure, or severe lung disease had a significantly lower risk-adjusted mortality rate if they had received prophylaxis compared with those patients receiving no prophylaxis [19]. However, the quality of prophylaxis (appropriate or partial) was not assessed in this study. In the current study, we were unable to accurately measure mortality as the data set could not capture mortality that occurred as an outpatient or within a different hospital setting. However, this remains an interesting question that should be studied in a prospective clinical trial. There is a currently ongoing trial that may help to answer the question of mortality as it investigates 30-day allcause mortality in medical patients receiving enoxaparin and
graduated compression stockings when compared with patients given placebo plus graduated compression stockings [20]. It is also interesting that in the current study, appropriate prophylaxis was associated with lower bleeding proportions than partial prophylaxis. Although we are unable to determine the reasons for this, it should be noted that patients who experienced a bleeding event had prophylaxis temporarily halted and therefore automatically fell into the partial prophylaxis group. As bleeding risk was not adjusted for in the analysis, the above findings should be interpreted with caution and a separate study that is designed to compare bleeding risk is needed. As this analysis was based on a US database, we examined the outcomes and costs only in patients receiving UFH or enoxaparin (with or without warfarin). This was due to the vast majority of patients receiving these anticoagulants, although it should be noted that other anticoagulants, such as dalteparin and fondaparinux are available for certain patient populations. Interestingly, a comparison of only those patients that received appropriate prophylaxis with enoxaparin or UFH found that enoxaparin was associated with lower costs and improved clinical outcomes than UFH [21]. The economic burden associated with VTE is substantial and cost considerations relating to anticoagulant therapy are of increasing importance to hospitals [6]. In a retrospective observational cohort study of reimbursed costs, the adjusted annualized mean total healthcare costs for DVT, PE, or DVT plus PE were $33,200, $31,300, and $38,300, respectively, compared with $2800 for control patients [4]. Furthermore, the adjusted annualized mean total healthcare cost was $47,596 for post-thrombotic syndrome, indicating that the longterm sequelae of VTE are associated with even greater costs. In the current study which looked at total hospital costs rather than reimbursed costs, $6370 per discharge was saved with appropriate prophylaxis. Given the large numbers of patients at-risk of VTE in US hospitals, improved adherence with ACCP guidelines may help to reduce the substantial healthcare costs associated with the condition. Fortunately, the need to improve VTE prevention is being increasingly recognized in the US, and a number of recent initiatives exist which may promote the wider use of appropriate prophylaxis. The Surgical Care Improvement Project has developed VTE-related performance measures for surgical patients [22]. Furthermore, the Joint Commission, in collaboration with the National Quality Forum, is developing performance measures for the prevention and care of VTE, which specifically acknowledge the risks faced by hospitalized patients [23]. In order to improve guideline adherence and meet the new performance measures, hospitals will need to develop, implement, and audit their own VTE prevention policies ensuring that all healthcare professionals within their organization are aware of, and comply with, these initiatives. Regarding limitations, as the current analysis was based on discharge records, the individual circumstances that may have led physicians not to adopt practices in-line with best-practice guidelines cannot be fully evaluated. Furthermore, although the time-frame of our study is concurrent to the availability of the Seventh ACCP guidelines, it may be that there was a lag in implementing the guideline recommendations that impacted the study results. Another possible limitation relates to the use of the MarketScan® Hospital Drug Database from Thomson Reuters. Although the database contains a wealth of information from different payers and different types of hospitals in diverse geographical areas across the US, it is not a random sample and may not be representative of the population of US hospitalized patients. For inclusion in the present study, patients had to have stayed in hospital for at least 6 days. This inclusion criterion was based on the Prophylaxis in Medical patients with Enoxaparin (MEDENOX) trial, where at least 6 days represented the typical duration of hospitalization of medical patients [17]. It is possible that the proportion of VTE was underestimated in the database population as a whole, as VTE events may have occurred in patients with shorter hospitalization. Furthermore, stipulating a hospital stay of at least 6 days may have selected high-risk
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
patients, especially in the surgical setting. Thus, the results in this study should be considered as a conservative estimate of the real-world population. An interesting follow-up study would be to consider the results in individual diagnosis categories within the medical and surgical populations, as this could show different outcomes depending on the patients’ level of risk. Further follow-up studies could also include a comparison of inhospital mortality between the two groups and a comparison of outcomes in hospitals with VTE prophylaxis protocols against those without protocols. In conclusion, this study highlights the current underuse of appropriate prophylaxis in a large population of hospitalized medical and surgical patients at high risk of VTE. Despite its suboptimal use, adherence with ACCP-recommended guidelines was more effective for the prevention of hospital-acquired PE (VTE) and bleeding (adverse) events, and was associated with reduced hospital costs compared with partial prophylaxis. Improving adherence with best-practice guidelines by providing prophylaxis of the recommended type, dose, and duration is urgently required to effectively reduce the substantial healthcare burden posed by VTE, both for individual patients and for healthcare organizations worldwide. Disclosure of Conflict of Interests This study was funded by sanofi-aventis U.S. Inc. The authors received editorial and writing support in the preparation of this manuscript, funded by sanofi-aventis U.S., Inc. The authors, however, were fully responsible for all content and editorial decisions. Acknowledgements This study was funded by sanofi-aventis U.S., Inc. The authors received editorial and writing support in the preparation of this manuscript, funded by sanofi-aventis U.S., Inc. The authors, however, were fully responsible for all content and editorial decisions. References [1] Anderson Jr FA, Zayaruzny M, Heit JA, Fidan D, Cohen AT. Estimated annual numbers of US acute-care hospital patients at risk for venous thromboembolism. Am J Hematol 2007;82:777–82. [2] Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. American College of Chest Physicians. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133(6 Suppl):381S–453S. [3] Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, et al. ENDORSE Investigators. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. Lancet 2008;371:387–94. [4] MacDougall DA, Feliu AL, Boccuzzi SJ, Lin J. Economic burden of deep-vein thrombosis, pulmonary embolism, and post-thrombotic syndrome. Am J Health Syst Pharm 2006;63(20 Suppl 6):S5–S15.
517
[5] Spyropoulos AC, Lin J. Direct medical costs of venous thromboembolism and subsequent hospital readmission rates: an administrative claims analysis from 30 managed care organizations. J Manag Care Pharm 2007;13:475–86. [6] Merli G, Ferrufino C, Lin J, Hussein M, Battleman D. Hospital-based costs associated with venous thromboembolism treatment regimens. J Thromb Haemost 2008;6:1077–86. [7] de Lissovoy G, Subedi P. Economic evaluation of enoxaparin as prophylaxis against venous thromboembolism in seriously ill medical patients: a US perspective. Am J Manag Care 2002;8:1082–8. [8] Oster G, Tuden RL, Colditz GA. A cost-effectiveness analysis of prophylaxis against deep-vein thrombosis in major orthopedic surgery. JAMA 1987;257:203–8. [9] Mismetti P, Quenet S, Levine M, Merli G, Decousus H, Derobert E, et al. Enoxaparin in the treatment of deep vein thrombosis with or without pulmonary embolism: an individual patient data meta-analysis. Chest 2005;128:2203–10. [10] Kleber FX, Witt C, Vogel G, Koppenhagen K, Schomaker U, Flosbach CW, et al. Randomized comparison of enoxaparin with unfractionated heparin for the prevention of venous thromboembolism in medical patients with heart failure or severe respiratory disease. Am Heart J 2003;145:614–21. [11] Kanaan AO, Silva MA, Donovan JL, Roy T, Al-Homsi AS. Meta-analysis of venous thromboembolism prophylaxis in medically ill patients. Clin Ther 2007;29:2395–405. [12] Wein L, Wein S, Haas SJ, Shaw J, Krum H. Pharmacological venous thromboembolism prophylaxis in hospitalized medical patients: a meta-analysis of randomized controlled trials. Arch Intern Med 2007;167:1476–86. [13] Amin A, Stemkowski S, Lin J, Yang G. Thromboprophylaxis rates in US medical centers: success or failure? J Thromb Haemost 2007;5:1610–6. [14] Amin AN, Stemkowski S, Lin J, Yang G. Preventing venous thromboembolism in US hospitals: are surgical patients receiving appropriate prophylaxis? Thromb Haemost 2008;99:796–7. [15] Kahn SR, Panju A, Geerts W, Pineo GF, Desjardins L, Turpie AG, et al. Sebaldt RJ; CURVE study investigators. Multicenter evaluation of the use of venous thromboembolism prophylaxis in acutely ill medical patients in Canada. Thromb Res 2007;119:145–55. [16] Tapson VF, Decousus H, Pini M, Chong BH, Froehlich JB, Monreal M, et al. IMPROVE Investigators. Venous thromboembolism prophylaxis in acutely ill hospitalized medical patients: findings from the International Medical Prevention Registry on Venous Thromboembolism. Chest 2007;132:936–45. [17] Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med 1999;341:793–800. [18] Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126(3 Suppl):338S–400S. [19] Burleigh E, Wang C, Foster D, Heller S, Dunn D, Safavi K, et al. Thromboprophylaxis in medically ill patients at risk for venous thromboembolism. Am J Health Syst Pharm 2006;63(20 Suppl 6):S23–9. [20] sanofi-aventis, February 2008. Study to evaluate the mortality reduction of enoxaparin in hospitalized acutely ill medical receiving enoxaparin (LIFENOX). Available: http:// clinicaltrials.gov/ct2/show/NCT00622648?term=enoxaparin&recr=Open&rank=1. Accessed February 16, 2009. [21] Amin AN, Lin J, Lenhart G, Schulman KL. Clinical and economic outcomes in patients at risk of venous thromboembolism receiving appropriate enoxaparin or unfractionated heparin prophylaxis. Thromb Haemost 2009;102:321–6. [22] QualityNet, 2009. Surgical Care Improvement Project (SCIP) Measure: venous thromboembolism. Available: http://www.medqic.org/dcs/ContentServer? cid=1137448680860&pagename=Medqic%2FMeasure%2FMeasuresHome& parentName=Topic&level3=Measures&c=MQParents. Accessed February 16, 2009. [23] The Joint Commission, October 2008. Performance Measurement Initiatives: National Consensus Standards for Prevention and Care of Venous Thromboembolism (VTE). Available: http://www.jointcommission.org/PerformanceMeasurement/ PerformanceMeasurement/VTE.htm. Accessed February 16, 2009.
Contents lists available at ScienceDirect
Thrombosis Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t h r o m r e s
Regular Article
Clinical and economic outcomes with appropriate or partial prophylaxis A.N. Amin a,⁎, J. Lin b, B.H. Johnson c, K.L. Schulman c a b c
University of California – Irvine, Orange, CA, USA sanofi-aventis, Bridgewater, NJ, USA Healthcare, Thomson Reuters, Cambridge, MA, USA
a r t i c l e
i n f o
Article history: Received 22 June 2009 Received in revised form 12 October 2009 Accepted 23 October 2009 Keywords: guideline adherence prophylaxis venous thromboembolism
a b s t r a c t Introduction: Despite the existence of evidence-based guidelines for venous thromboembolism (VTE) prevention, prophylaxis is often inappropriately prescribed. This study compared the efficacy, safety, and cost of appropriate (ACCP-recommended) prophylaxis with partial prophylaxis (not completely conforming to ACCP guidelines) in patients at-risk of VTE receiving enoxaparin or unfractionated heparin. Methods: The MarketScan® Hospital Drug Database from Thomson Reuters (January 2004–March 2007), was queried for medical and surgical patients at high risk of VTE, aged ≥ 40 years, and with a hospital stay ≥ 6 days. Univariate and multivariate analyses compared hospital-acquired VTE events, adverse events, and hospital costs between appropriate or partial prophylaxis discharges. Results: Of the 21,001 discharge records included, appropriate prophylaxis was received by 5136 (24.5%) patients. Compared with partial prophylaxis, appropriate prophylaxis was associated with significantly lower incidences of hospital-acquired pulmonary embolism (0.9% vs 0.5%; adjusted odds ratio [OR] 0.55, 95% confidence intervals [CI] 0.35–0.87, P = 0.010), and bleeding events (10.7% vs 5.1%; adjusted OR 0.57, 95% CI 0.50–0.66, P b 0.001). Total costs per discharge were lower for appropriate prophylaxis ($17,386 ± 12,004) than partial prophylaxis ($23,823 ± 19,783) with an adjusted mean difference of $6370 in favor of appropriate prophylaxis (P b 0.001). Conclusion: This retrospective study suggests that ACCP-guideline recommended appropriate prophylaxis reduces hospital-acquired pulmonary embolism and bleeding events in patients at-risk of VTE and is costsaving when total direct medical costs are considered. The substantial US clinical and economic VTE burden may, therefore, be reduced by improving prophylaxis adherence with guideline recommendations. © 2009 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism (VTE), which includes deep-vein thrombosis (DVT) and pulmonary embolism (PE), is a common complication during and after hospitalization for acute medical illness and surgery [1–3]. Almost a third of hospitalized patients are thought to be at risk of VTE annually and, given the considerable medical costs of treating VTE and its sequelae, VTE is associated with both substantial clinical and economic burdens [1,3–6]. Efficacious agents exist to prevent VTE, and these have been proven to be both well tolerated and cost effective [2,7–12]. Recommendations regarding the use of prophylactic agents are provided in best-practice guidelines, such as those from the American College of Chest Physicians (ACCP) [2]. Guideline-recommended appropriate prophylaxis is achieved only by providing the patient with the appropriate drug (or device), at the appropriate dose, and for the ⁎ Corresponding author. University of California – Irvine, 101 The City Drive South, Building 58, Room 110, ZC-4076H, Orange, CA 92868, USA. Tel.: +1 714 456 3785; fax: +1 714 456 3875. E-mail address: [email protected] (A.N. Amin). 0049-3848/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2009.10.018
appropriate duration. Despite the existence of best-practice guidelines, there is growing evidence in the literature that patients at-risk of VTE do not consistently receive prophylaxis, with only a minority receiving full and appropriate prophylaxis [3,13–16]. When VTE prophylaxis practices were investigated in 358 hospitals in 32 countries worldwide (n = 68,183), guideline-recommended prophylaxis was only received by 59% of surgical patients and 40% of medical patients [3]. Similarly, in 196,104 medical patients and 85,970 surgical patients at high-risk for VTE in 227 acute-care hospitals across the US, only 34% of medical patients and 32% of surgical patients received appropriate prophylaxis [13,14]. To date, few studies have evaluated the clinical and economic implications of failing to meet guideline recommendations for appropriate prophylaxis in everyday hospital practice. Use of a hospital administrative database provides information about a large, realworld population of medical and surgical patients at risk of VTE. Furthermore, such a database enables the study of symptomatic events alone, providing additional clinical relevance compared with trials, which assess a combination of symptomatic and asymptomatic events. The current study, conducted using a large US database, compared efficacy, safety, and costs of best-practice appropriate prophylaxis
514
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
with that of partial prophylaxis (prophylaxis not meeting bestpractice criteria) in medical and surgical patients at high risk of VTE. Methods Patient selection This study used a subset of the data from the MarketScan® Hospital Drug Database from Thomson Reuters. This database contains data from approximately 40 million discharge records from 550 acute care hospitals across the US. In 2006, the database contained information on one of every five discharges in the US. Cost data from hospital billing systems are available on more than a third of these records. Included patient discharges were aged ≥40 years and were admitted on or after January 1, 2004 and discharged on or before March 31, 2007. Eligible hospitalizations were included if they had a complete hospital stay of at least 6 days (to reflect a relatively high-degree of disease severity and potentially limited mobility) [17] and had a procedure and/or a diagnosis that deemed them at risk of VTE according to ACCP guidelines (medical risk factors were congestive heart failure, severe respiratory disease, active cancer, and stroke; surgical risk factors were general surgery, vascular surgery, gynecologic surgery, urologic surgery, laparoscopic surgery, major orthopedic surgery, and neurosurgery) [18]. In addition, patients were excluded if they were transferred from, or discharged to, another acute-care facility or if they had a potential contraindication to, or disease requiring, modification of ACCP-recommended anticoagulant therapy. Only discharges receiving either enoxaparin (with or without warfarin) or unfractionated heparin (UFH; with or without warfarin) were included. Enoxaparin was selected as the predominant lowmolecular-weight heparin provided in the database. Although ACCP guidelines suggest dosing requirements based on package inserts of the recommended anticoagulants, we assumed that best practice prophylaxis dosage per day was at least 40 mg for enoxaparin and at least 10,000 units for UFH. Prophylaxis duration was considered appropriate if it was received each day of the admission except 2 days for patients at medical risk (to allow for partial days of stay at admission and discharge), and each day of the admission after surgery except for 2 days in patients with surgical risk only. Outcomes The proportion of PE, DVT, and any hospital-acquired VTE event was compared between groups. Patients were classified as having VTE if they had a diagnosis of PE or DVT, and received at least 1 day of enoxaparin (≥100 mg per day) or UFH (≥25,000 units per day) during the qualifying hospitalization. The incidence of adverse events, including the proportion of major bleeding, non-major bleeding, and thrombocytopenia were also compared. Hospital utilization, charges (hospital billing), and costs (hospitalbased costs) were measured over the duration of the qualifying hospital stay to assess the economic implications of appropriate versus partial prophylaxis. Length of hospital stay for the qualifying admission and length of stay including readmission were calculated. Data are expressed as mean ± standard deviation. Analysis In addition to univariate analysis of clinical and economic outcomes, multivariate models were used for each outcome measure to adjust for differences in admission characteristics. Explanatory variables included length of stay, primary payer, gender, age, region, hospital teaching status, hospital bed size, risk categories (medical risk only, surgical risk only, both medical and surgical risk), use of antiplatelet therapy or mechanical prophylaxis, length of anticoagulation, and the number of personal risk factors.
Generalized linear models with gamma distributed error and log link were used to predict differences in hospital costs and charges between the groups. Generalized linear models with negative binomial distributed error and log link were used to predict differences in length of stay. Similar explanatory variables used in the clinical models were included in the economic multivariate analyses, with the exception of length of stay. A P-value of b0.05 was considered statistically significant. Results From an approximate 15 million records in the database, a total of 21,001 discharge records were included in the current analysis. Of these, only 5136 (24.5%) discharges received appropriate prophylaxis, with the majority of patients (15,865 [75.5%]) receiving partial prophylaxis. Patient demographics and clinical characteristics are shown in Table 1. A significantly greater proportion of females than males received appropriate prophylaxis (P b 0.001). The age of patients in the partial and appropriate prophylaxis groups was similar (65.6 ± 12.5 years and 65.2 ± 12.3 years, respectively), with no significant differences observed by age group. In general, the type of payer did not appear to influence the appropriateness of therapy, except that appropriate prophylaxis was more common than partial prophylaxis in Medicaid discharges (6.4% vs 5.3%, P = 0.003). Use of appropriate prophylaxis was significantly higher in medical discharges (58.8% vs 50.8%, P b 0.001); however, partial prophylaxis was observed in significantly more surgical discharges (29.0% vs 21.7%, P b 0.001). The remaining discharges in each group (20.2% of surgical and 19.5% of medical) received no prophylaxis. Clinical outcomes When hospital-acquired VTE events were compared, a VTE event occurred in 1.9% of the partial prophylaxis group compared with 1.4% in the appropriate prophylaxis group (Table 2). Similarly, the absolute proportions of both DVT and PE appeared higher in the partial prophylaxis group than the appropriate prophylaxis group (DVT: 1.2% vs 1.0%; and PE: 0.9% vs 0.5%, respectively). In the multivariate analysis, significantly fewer discharges in the appropriate prophylaxis group were likely to experience PE than in the partial prophylaxis group (odds ratio [OR] 0.55, 95% confidence interval [CI] 0.35–0.87, P = 0.010; Table 3). However, the likelihood of experiencing any hospital-acquired VTE event was not significantly different after adjustment. In the safety analysis, the probability of experiencing any adverse event (major bleeding, non-major bleeding, or thrombocytopenia) was significantly lower in the appropriate prophylaxis group (OR 0.57, 95% CI 0.50–0.66, P b 0.001; Table 3). Significant ORs were also observed for non-major bleeding and thrombocytopenia, but not for major bleeding (Table 3). It should be noted however that bleeding risk was not controlled for in the multivariate analysis, and patients who experienced bleeding were likely to have prophylaxis stopped, thus placing them in the partial prophylaxis group. Economic outcomes In a measure of hospital utilization, length of hospital stay was compared between the two groups. The mean total length of stay for the qualifying admission was shorter for patients receiving appropriate prophylaxis compared with partial prophylaxis (9.3 ± 4.8 days vs 11.2 ± 7.2 days, P b 0.001). In addition, a highly significant difference was observed with the mean total length of stay, including readmission, among appropriate prophylaxis and partial prophylaxis groups (10.3 ± 6.1 days vs 12.2 ± 8.3 days, respectively; adjusted difference − 2.95 days, P b 0.0001). It is also important to note that there was a variance in the proportion of time spent in the intensive
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
515
Table 1 Demographics and clinical characteristics of the partial and appropriate prophylaxis groups. Partial prophylaxis (n = 15,865)
Discharge status
Discharged to home Died Unknown Private insurance Medicare Medicaid Other government Workers’ compensation Other Female Male 40–44 45–54 55–64 65–74 75–84 85+ Northeast South Midwest West Yes No 1–199 200–299 300–499 500+
Primary payer
Gender Age, years
Region
Teaching status Licensed bed numbers
N
%
N
%
P-value
13,974 1062 829 5513 8338 844 126 61 983 7732 8133 712 2687 4004 4191 3312 959 140 12,836 2683 206 3747 12,118 1575 2955 6182 5153
88.1 6.7 5.2 34.8 52.6 5.3 0.8 0.4 6.2 48.7 51.3 4.5 16.9 25.2 26.4 20.9 6.0 0.9 80.9 16.9 1.3 23.6 76.4 9.9 18.6 39.0 32.5
4651 253 232 1800 2665 330 46 12 283 2768 2368 260 864 1342 1384 1014 272 60 4046 1007 23 1150 3986 655 892 1931 1658
90.6 4.9 4.5 35.1 51.9 6.4 0.9 0.2 5.5 53.9 46.1 5.1 16.8 26.1 27.0 19.7 5.3 1.2 78.8 19.6 0.5 22.4 77.6 12.8 17.4 37.6 32.3
b0.001 b0.001 0.044 0.698 0.405 0.003 0.483 0.110 0.073 b0.001 b0.001 0.089 0.849 0.202 0.454 0.081 0.047 0.067 b0.001 b0.001 b0.001 0.071 0.071 b0.001 0.043 0.080 0.792
care unit between groups, and this will likely have impacted on cost. For almost all charges and costs compared, including room and board, medical supplies and total pharmacy costs, partial prophylaxis was
Table 2 Clinical outcomes with partial prophylaxis versus appropriate prophylaxis. Partial prophylaxis (n = 15,865)
DVT PE Both DVT and PE Any hospital-acquired VTE Major bleeding Non-major bleeding Thrombocytopenia Any adverse event
Appropriate prophylaxis (n = 5136)
Number of events
%
Number of events
%
189 145 36 298 63 1,568 95 1,702
1.2 0.9 0.2 1.9 0.4 9.9 0.6 10.7
51 24 2 73 5 248 9 262
1.0 0.5 b 0.1 1.4 0.1 4.8 0.2 5.1
DVT, deep-vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.
Table 3 Clinical outcomes with appropriate prophylaxis compared with partial prophylaxis.
DVT PE Any hospital-acquired VTE Major bleeding Non-major bleeding Thrombocytopenia Any adverse event
Adjusted OR (95% CIs)⁎
P-value
1.09 0.55 0.90 0.42 0.58 0.46 0.57
0.625 0.010 0.464 0.072 b 0.001 0.032 b 0.001
(0.78–1.52) (0.35–0.87) (0.68–1.19) (0.17–1.08) (0.50–0.67) (0.22–0.93) (0.50–0.66)
Appropriate prophylaxis (n = 5136)
⁎Multivariate logistic regression analysis incorporating length of stay, primary payer, gender, age, region, hospital teaching status, hospital bed size, risk categories (medical risk only, surgical risk only, both medical and surgical risk), use of antiplatelet therapy or mechanical prophylaxis, length of anticoagulation, and the number of personal risk factors as explanatory variables. CI, confidence interval; DVT, deep-vein thrombosis; OR, odds ratio; PE, pulmonary embolism; VTE, venous thromboembolism.
associated with greater expenditure than appropriate prophylaxis (Table 4). The main exception to this was in enoxaparin costs, with these being higher in those receiving appropriate prophylaxis compared with partial prophylaxis ($560 ± 600 and $279 ± 539, respectively). When all associated costs were considered together, the mean total hospital costs per discharge were lower in the appropriate prophylaxis group ($17,386 ± 12,004) compared with the partial prophylaxis group ($23,823 ± 19,783). The cost difference per discharge was highly significant, in favor of appropriate prophylaxis, with similar considerable savings of $6437 (95% CI $5865–7010, P b 0.001) in the univariate analysis and $6370 (95% CI $6067–6670, P b 0.001) following adjustment in the multivariate analysis. Discussion This study demonstrates that among patients receiving enoxaparin or UFH, only 25% of medical and surgical patients at high risk of VTE received appropriate prophylaxis as recommended by the ACCP guidelines. Despite this, appropriate prophylaxis significantly reduced the probability of PE by almost a half. Moreover, in the safety analyses a significant reduction in the likelihood of non-major bleeding and thrombocytopenia was observed with appropriate versus partial prophylaxis. In addition to the reduction in clinical and adverse events with appropriate prophylaxis, substantial cost-savings were also observed, highlighting both the clinical and economic benefits of fully adhering to best-practice ACCP guidelines. Results of the present study are consistent with our previous investigations, which have also demonstrated the suboptimal use of appropriate prophylaxis [13,14]. When the proportion of appropriate versus partial prophylaxis was assessed in patients with acute medical conditions (n = 196,104), appropriate prophylaxis was received by 34% of medical patients, with the lowest proportions in trauma discharges without surgery (20.3%) and acute spinal cord injury discharges without surgery (20.8%) [13]. Similarly, in a recent study in 85,970 surgical discharges, appropriate VTE prophylaxis was only received by 32% of discharges [14]. The absence of any prophylaxis,
516
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
Table 4 Hospital charges and costs in the partial and appropriate prophylaxis groups.
Hospital charges All room and board Operating and recovery room Radiology Laboratory tests Medical supplies Respiratory therapy Mechanical prophylaxis Pharmacy Total anticoagulants Other inpatient pharmacy Total pharmacy Total hospital charges Hospital costs All room and board Operating and recovery room Radiology Laboratory tests Medical supplies Respiratory therapy Mechanical prophylaxis Pharmacy Total anticoagulants Other inpatient pharmacy Total pharmacy Total hospital costs
Partial prophylaxis (n = 15,865)
Appropriate prophylaxis (n = 5136)
$, mean ± SD
$, mean ± SD
10,022 ± 15,523 9811 ± 12,653 5566 ± 6528 7351 ± 8025 10,024 ± 15,557 4659 ± 9240 21 ± 124
5304 ± 9656 5734 ± 9203 4230 ± 4888 5255 ± 4313 5300 ± 9653 3907 ± 6327 15 ± 89
884 ± 1639 8883 ± 13,440
1203 ± 1146 6464 ± 7530
13,160 ± 18,956 69,997 ± 58,905
9878 ± 10,991 47,981 ± 33,743
2868 ± 4558 1887 ± 2435 675 ± 1025 944 ± 1371 2869 ± 4572 678 ± 1324 5 ± 29
1540 ± 2822 1088 ± 1566 499 ± 1063 605 ± 611 1540 ± 2822 636 ± 1103 4 ± 21
308 ± 546 1357 ± 1954
577 ± 591 1012 ± 1112
2181 ± 2879 23,823 ± 19,783
1935 ± 1843 17,386 ± 12,004
SD, standard deviation.
mechanical prophylaxis only and insufficient prophylaxis duration were common reasons for not receiving appropriate VTE prevention in these two studies [13,14]. The current study furthers the results of previous investigations [3,13–15], as both the clinical and economic impact of failing to adhere to best-practice guidelines for appropriate prophylaxis were also reported. Furthermore, assessment of efficacy and safety was performed not only across the large diverse sample, but also using multivariate analysis to adjust for different patient clinical characteristics and demographics. Randomized trials provide valuable information about the efficacy and safety of specific thromboprophylactic agents, usually in subsets of medical or surgical patients. However, a key strength of the current study is that the impact of thromboprophylaxis was investigated in a large real-world population, rather than a highly selected trial population. Furthermore, the current study assessed symptomatic VTE events alone, rather than both symptomatic and asymptomatic events which are generally investigated in clinical studies due to the need for statistical power. A previous retrospective study found that at-risk medical patients with a diagnosis of acute myocardial infarction, cancer, heart failure, or severe lung disease had a significantly lower risk-adjusted mortality rate if they had received prophylaxis compared with those patients receiving no prophylaxis [19]. However, the quality of prophylaxis (appropriate or partial) was not assessed in this study. In the current study, we were unable to accurately measure mortality as the data set could not capture mortality that occurred as an outpatient or within a different hospital setting. However, this remains an interesting question that should be studied in a prospective clinical trial. There is a currently ongoing trial that may help to answer the question of mortality as it investigates 30-day allcause mortality in medical patients receiving enoxaparin and
graduated compression stockings when compared with patients given placebo plus graduated compression stockings [20]. It is also interesting that in the current study, appropriate prophylaxis was associated with lower bleeding proportions than partial prophylaxis. Although we are unable to determine the reasons for this, it should be noted that patients who experienced a bleeding event had prophylaxis temporarily halted and therefore automatically fell into the partial prophylaxis group. As bleeding risk was not adjusted for in the analysis, the above findings should be interpreted with caution and a separate study that is designed to compare bleeding risk is needed. As this analysis was based on a US database, we examined the outcomes and costs only in patients receiving UFH or enoxaparin (with or without warfarin). This was due to the vast majority of patients receiving these anticoagulants, although it should be noted that other anticoagulants, such as dalteparin and fondaparinux are available for certain patient populations. Interestingly, a comparison of only those patients that received appropriate prophylaxis with enoxaparin or UFH found that enoxaparin was associated with lower costs and improved clinical outcomes than UFH [21]. The economic burden associated with VTE is substantial and cost considerations relating to anticoagulant therapy are of increasing importance to hospitals [6]. In a retrospective observational cohort study of reimbursed costs, the adjusted annualized mean total healthcare costs for DVT, PE, or DVT plus PE were $33,200, $31,300, and $38,300, respectively, compared with $2800 for control patients [4]. Furthermore, the adjusted annualized mean total healthcare cost was $47,596 for post-thrombotic syndrome, indicating that the longterm sequelae of VTE are associated with even greater costs. In the current study which looked at total hospital costs rather than reimbursed costs, $6370 per discharge was saved with appropriate prophylaxis. Given the large numbers of patients at-risk of VTE in US hospitals, improved adherence with ACCP guidelines may help to reduce the substantial healthcare costs associated with the condition. Fortunately, the need to improve VTE prevention is being increasingly recognized in the US, and a number of recent initiatives exist which may promote the wider use of appropriate prophylaxis. The Surgical Care Improvement Project has developed VTE-related performance measures for surgical patients [22]. Furthermore, the Joint Commission, in collaboration with the National Quality Forum, is developing performance measures for the prevention and care of VTE, which specifically acknowledge the risks faced by hospitalized patients [23]. In order to improve guideline adherence and meet the new performance measures, hospitals will need to develop, implement, and audit their own VTE prevention policies ensuring that all healthcare professionals within their organization are aware of, and comply with, these initiatives. Regarding limitations, as the current analysis was based on discharge records, the individual circumstances that may have led physicians not to adopt practices in-line with best-practice guidelines cannot be fully evaluated. Furthermore, although the time-frame of our study is concurrent to the availability of the Seventh ACCP guidelines, it may be that there was a lag in implementing the guideline recommendations that impacted the study results. Another possible limitation relates to the use of the MarketScan® Hospital Drug Database from Thomson Reuters. Although the database contains a wealth of information from different payers and different types of hospitals in diverse geographical areas across the US, it is not a random sample and may not be representative of the population of US hospitalized patients. For inclusion in the present study, patients had to have stayed in hospital for at least 6 days. This inclusion criterion was based on the Prophylaxis in Medical patients with Enoxaparin (MEDENOX) trial, where at least 6 days represented the typical duration of hospitalization of medical patients [17]. It is possible that the proportion of VTE was underestimated in the database population as a whole, as VTE events may have occurred in patients with shorter hospitalization. Furthermore, stipulating a hospital stay of at least 6 days may have selected high-risk
A.N. Amin et al. / Thrombosis Research 125 (2010) 513–517
patients, especially in the surgical setting. Thus, the results in this study should be considered as a conservative estimate of the real-world population. An interesting follow-up study would be to consider the results in individual diagnosis categories within the medical and surgical populations, as this could show different outcomes depending on the patients’ level of risk. Further follow-up studies could also include a comparison of inhospital mortality between the two groups and a comparison of outcomes in hospitals with VTE prophylaxis protocols against those without protocols. In conclusion, this study highlights the current underuse of appropriate prophylaxis in a large population of hospitalized medical and surgical patients at high risk of VTE. Despite its suboptimal use, adherence with ACCP-recommended guidelines was more effective for the prevention of hospital-acquired PE (VTE) and bleeding (adverse) events, and was associated with reduced hospital costs compared with partial prophylaxis. Improving adherence with best-practice guidelines by providing prophylaxis of the recommended type, dose, and duration is urgently required to effectively reduce the substantial healthcare burden posed by VTE, both for individual patients and for healthcare organizations worldwide. Disclosure of Conflict of Interests This study was funded by sanofi-aventis U.S. Inc. The authors received editorial and writing support in the preparation of this manuscript, funded by sanofi-aventis U.S., Inc. The authors, however, were fully responsible for all content and editorial decisions. Acknowledgements This study was funded by sanofi-aventis U.S., Inc. The authors received editorial and writing support in the preparation of this manuscript, funded by sanofi-aventis U.S., Inc. The authors, however, were fully responsible for all content and editorial decisions. References [1] Anderson Jr FA, Zayaruzny M, Heit JA, Fidan D, Cohen AT. Estimated annual numbers of US acute-care hospital patients at risk for venous thromboembolism. Am J Hematol 2007;82:777–82. [2] Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. American College of Chest Physicians. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133(6 Suppl):381S–453S. [3] Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, et al. ENDORSE Investigators. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. Lancet 2008;371:387–94. [4] MacDougall DA, Feliu AL, Boccuzzi SJ, Lin J. Economic burden of deep-vein thrombosis, pulmonary embolism, and post-thrombotic syndrome. Am J Health Syst Pharm 2006;63(20 Suppl 6):S5–S15.
517
[5] Spyropoulos AC, Lin J. Direct medical costs of venous thromboembolism and subsequent hospital readmission rates: an administrative claims analysis from 30 managed care organizations. J Manag Care Pharm 2007;13:475–86. [6] Merli G, Ferrufino C, Lin J, Hussein M, Battleman D. Hospital-based costs associated with venous thromboembolism treatment regimens. J Thromb Haemost 2008;6:1077–86. [7] de Lissovoy G, Subedi P. Economic evaluation of enoxaparin as prophylaxis against venous thromboembolism in seriously ill medical patients: a US perspective. Am J Manag Care 2002;8:1082–8. [8] Oster G, Tuden RL, Colditz GA. A cost-effectiveness analysis of prophylaxis against deep-vein thrombosis in major orthopedic surgery. JAMA 1987;257:203–8. [9] Mismetti P, Quenet S, Levine M, Merli G, Decousus H, Derobert E, et al. Enoxaparin in the treatment of deep vein thrombosis with or without pulmonary embolism: an individual patient data meta-analysis. Chest 2005;128:2203–10. [10] Kleber FX, Witt C, Vogel G, Koppenhagen K, Schomaker U, Flosbach CW, et al. Randomized comparison of enoxaparin with unfractionated heparin for the prevention of venous thromboembolism in medical patients with heart failure or severe respiratory disease. Am Heart J 2003;145:614–21. [11] Kanaan AO, Silva MA, Donovan JL, Roy T, Al-Homsi AS. Meta-analysis of venous thromboembolism prophylaxis in medically ill patients. Clin Ther 2007;29:2395–405. [12] Wein L, Wein S, Haas SJ, Shaw J, Krum H. Pharmacological venous thromboembolism prophylaxis in hospitalized medical patients: a meta-analysis of randomized controlled trials. Arch Intern Med 2007;167:1476–86. [13] Amin A, Stemkowski S, Lin J, Yang G. Thromboprophylaxis rates in US medical centers: success or failure? J Thromb Haemost 2007;5:1610–6. [14] Amin AN, Stemkowski S, Lin J, Yang G. Preventing venous thromboembolism in US hospitals: are surgical patients receiving appropriate prophylaxis? Thromb Haemost 2008;99:796–7. [15] Kahn SR, Panju A, Geerts W, Pineo GF, Desjardins L, Turpie AG, et al. Sebaldt RJ; CURVE study investigators. Multicenter evaluation of the use of venous thromboembolism prophylaxis in acutely ill medical patients in Canada. Thromb Res 2007;119:145–55. [16] Tapson VF, Decousus H, Pini M, Chong BH, Froehlich JB, Monreal M, et al. IMPROVE Investigators. Venous thromboembolism prophylaxis in acutely ill hospitalized medical patients: findings from the International Medical Prevention Registry on Venous Thromboembolism. Chest 2007;132:936–45. [17] Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med 1999;341:793–800. [18] Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126(3 Suppl):338S–400S. [19] Burleigh E, Wang C, Foster D, Heller S, Dunn D, Safavi K, et al. Thromboprophylaxis in medically ill patients at risk for venous thromboembolism. Am J Health Syst Pharm 2006;63(20 Suppl 6):S23–9. [20] sanofi-aventis, February 2008. Study to evaluate the mortality reduction of enoxaparin in hospitalized acutely ill medical receiving enoxaparin (LIFENOX). Available: http:// clinicaltrials.gov/ct2/show/NCT00622648?term=enoxaparin&recr=Open&rank=1. Accessed February 16, 2009. [21] Amin AN, Lin J, Lenhart G, Schulman KL. Clinical and economic outcomes in patients at risk of venous thromboembolism receiving appropriate enoxaparin or unfractionated heparin prophylaxis. Thromb Haemost 2009;102:321–6. [22] QualityNet, 2009. Surgical Care Improvement Project (SCIP) Measure: venous thromboembolism. Available: http://www.medqic.org/dcs/ContentServer? cid=1137448680860&pagename=Medqic%2FMeasure%2FMeasuresHome& parentName=Topic&level3=Measures&c=MQParents. Accessed February 16, 2009. [23] The Joint Commission, October 2008. Performance Measurement Initiatives: National Consensus Standards for Prevention and Care of Venous Thromboembolism (VTE). Available: http://www.jointcommission.org/PerformanceMeasurement/ PerformanceMeasurement/VTE.htm. Accessed February 16, 2009.