SEARCH

SEARCH BY CITATION

Keywords:

  • electronic alerts;
  • health economics;
  • low molecular weight heparin;
  • prophylaxis;
  • venous thromboembolism

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

Summary. Objectives: The prevention of venous thromboembolism (VTE) is a priority for improved safety in hospitalised patients. Worldwide, there is growing concern over the undersuse of appropriate thromboprophylaxis. Computerised decision support improves the implementation of thromboprophylaxis and reduces inpatient VTE. However, an economic assessment of this approach has not yet been performed. Objectives: To evaluate the economic impact of an electronic alert (e-alert) system to prevent VTE in hospitalised patients over a 4 year period. Patients/methods: All hospitalised patients at a single institution during the first semesters of 2005–2009 (n = 32 280) were included. All cases of VTE developed during hospitalisation were followed and direct costs of diagnosis and management collected. Results: E-alerts achieved a sustained reduction of the incidence of in-hospital VTE, OR 0.50 (95% CI, 0.29–0.84), the impact being especially significant in medical patients, OR 0.44 (95% CI, 0.22–0.86). No increase in prophylaxis-related bleeding was observed. In our setting, the mean direct cost (during hospitalisation and after discharge) of an in-hospital VTE episode is €7058. Direct costs per single hospitalised patient were reduced after e-alerts from €21.6 to €11.8, while the increased use of thromboprophylaxis and the development of e-alerts meant €3 and €0.35 per patient, respectively. Thus, the implementation of e-alerts led to a net cost saving of €6.5 per hospitalised patient. Should all hospitalised patients in Spain be considered, total yearly savings would approach €30 million. Conclusions: E-alerts are useful and cost-effective tools for thromboprophylaxis strategy in hospitalised patients. Fewer thromboembolic complications and lower costs are achieved by its implementation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

Venous thromboembolism (VTE), encompassing both deep vein thrombosis (DVT) and pulmonary embolism (PE), remains a frequent but preventable condition in hospitalised patients and is associated with significant morbidity, mortality and consumption of health resources. In Western countries, the estimated annual incidence of VTE in general population is 1–2 cases per 1000 inhabitants [1]. Despite early diagnosis and adequate therapy, 30-day mortality after a VTE event reaches 5%–10% [2,3], and up to 30% of patients will develop a post-thrombotic syndrome and 4% chronic thromboembolic pulmonary hypertension [4–6].

A recent epidemiological study aimed to assess the incidence and distribution of VTE in Spain, analysing hospital discharges codified by the Spanish National Health System showed that VTE represented 0.82% of all discharges, 53% of the episodes corresponding to PE and 47% DVT [7,8]. PE-associated 30 day-mortality was 6.4% compared to 3.2% in patients with DVT. VTE occurred during admission in 4‰ (3.0‰–4.7‰) of patients hospitalised for any cause, 74% of them having been admitted due to medical illnesses [7]. The magnitude of the risk for acquiring VTE in US hospitals is staggering. About 8 million Medical Service and 4 million Surgical Service patients are at moderate or high risk for developing VTE each year [9].

Limited data exist regarding the economic burden of VTE. Published estimates suggest that the direct costs of VTE approach to $3–4 billion (€2400–3200 million) annually in the US [10]. In Spain, the estimated VTE-related direct hospital costs for year 2005 was €66 million [7]. However, these estimates do not take into account additional indirect costs such as lost working days and productivity due to VTE diagnosis.

Despite current scientific evidence regarding the efficacy and safety of thromboprophylaxis in both surgical and medical inpatients, several recent studies have shown an unacceptable underuse of thromboprophylaxis in at-risk patients worldwide, which requires an urgent call for action [11–13]. The latest American College of Chest Physicians (ACCP) guidelines recommend that every institution should adopt measures aimed to improve the appropriate use of thromboprophylaxis [14]. Healthcare agencies also have developed education programs and implemented policies to improve guideline adherence and increase the use of prophylactic regimens [15,16]. Failure to prevent in-hospital VTE may no longer be accepted. For example, Medicare and Medicaid will cease reimbursing hospitals for the incremental care needed to treat postoperative total hip or knee replacement patients who develop VTE [17].

Electronic alert (e-alert) systems can help clinicians assess the thrombotic risk of hospitalised patients and have been shown to be associated with improvement in the use of thromboprophylaxis and reduction in the incidence of VTE during hospitalisation and after discharge [18–22]. However to date, an economic evaluation of the impact of implementation of an e-alert system has not been performed. We report herein the direct economic impact due to the improvement in the incidence of VTE in hospitalised patients after the implementation of an e-alert system in a single institution.

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

Study design

An e-alert software was developed with the aim to identify hospitalised patients at increased risk of VTE, linked to the computerised patients’ database of the institution, to acquire essential data to stratify the patients’ thrombotic risk. The new software was first installed in September 2005, and fully operative from January 2006. All hospitalised adult patients at the University Clinic of Navarra, institution accredited by the Joint Commission International since 2004, during the periods January–June 2006, 2007, 2008 and 2009, were included in this economic evaluation. The control group included hospitalised patients between January and June 2005, before the implementation of e-alert.

The e-alert software mechanism has been described previously [19]. Briefly, the risk of VTE in medical patients was determined according to the validated PRETEMED scale [23]. In this point scale major risk factors such as active cancer, previous VTE, acute myocardial infarction, ischaemic stroke with limb paralysis, decompensated chronic obstructive pulmonary disease and thrombophilia were assigned a score of 3; congestive heart failure, chronic renal insufficiency/nephrotic syndrome, severe acute infection, lower limb cast or prolonged bed rest were assigned a score of 2; pregnancy/postpartum period, recent prolonged flight, lower limb paresis, estrogen therapy, thalidomide/lenalidomide administration, use of central vein catheter, obesity, age > 60 years or smoking were assigned a score of 1. High risk of VTE was defined as a cumulative risk score of at least four points. In surgical patients, the thrombotic risk was estimated following the stratification recommended by the ACCP guidelines [14].

The software is linked to the computerised patient database of our institution acquiring automatically essential data to stratify the patients’ thrombotic risk from a variety of sources, such as medical orders, nursery daily reports, surgery registries or laboratory results. However, the reason for admission or the previous history of VTE had to be entered by the physician by fulfilling a simple electronic questionnaire that appeared in the computer screen at the time of admission and also when completing the first prescription form for the Hospital Pharmacy. Yearly audits were performed to evaluate the completeness of data entry.

Daily screening of the risk of thrombosis of all hospitalised patients was performed using the computerised software, sending an alert to those at high risk (score ≥ 4). A symbol ‘VTE!’ appeared on the screen of the responsible physicians, who were free to order or withhold prophylaxis. Moreover, the program is linked to the hospital VTE prevention guidelines allowing clinicians to review the indications of different thromboprophylaxis measures. Low molecular weight heparin (LMWH) was the prophylaxis recommended in at-risk hospitalised patients, with the exception of high bleeding risk subjects (thrombocytopenia < 50 000 platelets per mm3, known bleeding diathesis or active bleeding), who are encouraged to receive mechanical prophylaxis, i.e., elastic stockings or pneumatic compression devices.

Study endpoints and data collection

The primary end point of the study was to determine if objectively confirmed VTE incidence during hospitalisation was reduced after implementation of e-alerts. In the present study, incremental costs for objectively confirmed VTE diagnosis and therapy and the economic outcomes of the complications associated with VTE management were used to evaluate if, from an economic point of view, the implementation of e-alerts was a reasonable cost-effective measure.

Data from all patients hospitalised during the study periods and the incidence of VTE during hospitalisation were obtained from the registries of the Documentation Service of our institution. The incidence of DVT and PE among hospitalised patients who were at least 18 years old was determined by evaluation of the Hospital Discharge Minimum Basic Data Set (MBDS), which includes clinical and administrative data on each hospital discharge. To minimise the risk of missing any in-hospital VTE episodes, the data provided by the Documentation Service were compared with the records of the Hematology Service (responsible of the initial treatment of VTE in our institution). The clinical information was coded using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), as stated by the Spanish Ministry of Health. The corresponding diagnosis-related groups were incorporated to identify medical or surgical discharge. MBDS was checked to record age, active neoplasm and average length of hospital stay. The corresponding diagnosis-related groups were incorporated to identify medical or surgical discharge. Every VTE case developed during hospitalisation was evaluated individually and costs (in €) calculated according to tests performed for diagnosis, treatment received, follow-up visits and management of complications. Costs were calculated according to our institutional fares for year 2009.

As the analysis was performed from an institutional perspective, only direct medical costs associated with VTE prophylaxis and management of untoward events were included. We also include the cost of medical treatment and further management or readmission beyond the initial hospitalisation period. Thromboprophylaxis drug costs were obtained from the inpatient pharmacy.

Statistical analysis

Categorical variables were expressed as frequencies and percentages, and quantitative variables as either mean (± standard deviation) or median and interquartile range (IQR), depending on the distribution. Differences in baseline characteristics for patients in the pre-intervention and postintervention groups were compared using mid-P exact value for categorical variables and the Wilcoxon rank sum test as a nonparametric approach for quantitative variables. The odds ratio (OR) for VTE (post-intervention phase vs. pre-intervention phase) and its corresponding 95% confidence interval (95% CI) were estimated. An OR lower than 1 meant that patients in the post-intervention groups were at decreased risk for any VTE. Type I error was set at 0.05. A sensitivity analysis was used to identify important model uncertainties. For clinical outcomes, ranges for variables were based on 95% CI calculated from the data. A sensitivity analysis was also displayed in the cost-effectiveness plane as an analysis of extremes. The worst and best case therapeutic scenarios were determined by considering the upper and lower cost estimates (real present cost ± 25%) as well as the lower and upper limits of effectiveness of the pre- and post-intervention period. Analyses were performed using spss 15.0 (Chicago, IL, USA) software.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

During the first semester of years 2006, 2007, 2008 and 2009, 25 839 adult patients were hospitalised at our institution (> 6000 per semester). All groups had similar baseline characteristics (Table 1). Mean age was 55 years, while mean length of stay was 5.8 days. Overall, 12 161 individuals (47.1%) were hospitalised due to medical illnesses, while 13 678 (52.9%) were surgical patients. Nearly 30% of patients suffered from active cancer.

Table 1.   Baseline characteristics of patients throughout the studied periods
 2005* (n = 6441)2006* (n = 6312)2007* (n = 6585)2008* (n = 6676)2009* (n = 6266)
  1. IQR, interquartile range; *first semester.

Age – years, mean (SD)54.8 (16.1)55.2 (16.3)55.3 (16.7)55.5 (16.9)55.9 (16.5)
Age > 70 years, n (%)1198 (18.6)1238 (19.6)1356 (20.6)1393 (20.9)1311 (20.9)
Male sex, n (%)3528 (54.8)3403 (53.9)3500 (53.2)3511 (52.6)3327 (53.1)
Length of stay – days, median (IQR)4 (2–7)4 (2–7)4 (2–7)4 (2–7)4 (2–7)
Surgical patients, n (%)3051 (47.4)3175 (50.3)3262 (49.5)3654 (54.7)3587 (57.3)
Knee/hip replacement, n (%)80 (1.2)139 (2.2)164 (2.5)179 (2.7)161 (2.6)
Cardiovascular/thorax surgery, n (%)319 (5.0)370 (5.9)350 (5.3)438 (6.6)423 (6.8)
Abdominal surgery, n (%)678 (10.5)653 (10.3)732 (11.1)683 (10.2)666 (10.6)
Urological surgery, n (%)189 (2.9)183 (2.9)194 (2.9)209 (3.1)231 (3.7)
Medical patients, n (%)3390 (52.6)3137 (49.7)3323 (50.5)3022 (45.3)2679 (42.8)
Cancer, n (%)2382 (37.0)2016 (31.9)1955 (29.7)1857 (27.8)1756 (28)
Chronic lung disease, n (%)273 (4.2)294 (4.7)387 (5.9)379 (5.7)4127 (6.6)
Congestive heart failure/ischemic cardiopathy, n (%)449 (7.0)444 (7.0)587 (8.9)688 (10.3)618 (9.9)
Pregnancy/delivery, n (%)204 (3.2)226 (3.6)242 (3.7)271 (4.1)280 (4.4)
Stroke, n (%)21 (0.3)19 (0.3)33 (0.5)26 (0.4)25 (0.4)
Diabetes mellitus, n (%)537 (8.3)590 (9.3)637 (9.7)702 (10.5)634 (10.1)

E-alert, prophylaxis use and incidence of VTE during the study periods

The number of e-alerts sent and appropriate thromboprophylaxis use during hospitalisation are shown in Table 2. In medical patients, the use of appropriate prophylaxis against VTE in at-risk patients increased from 27% (data obtained from an internal survey of clinical histories performed in 2005) to over 60% during 2007–2009. In surgical patients, the use of adequate thromboprophylaxis remained over 80% throughout the studied periods.

Table 2.   E-alerts sent and use of prophylaxis
 2006*2007*2008*2009*
  1. *First semester.

Overall population
Alerts sent2073/6312 (32.8%)2121/6585 (32.2%)2540/6676 (38.1%)2376/6266 (38.0%)
Appropriate prophylaxis1737/2073 (83.8%)1684/2121 (84.1%)2032/2540 (80.4%)1824/2376 (76.8%)
Medical patients
Alerts sent303/3137 (9.7%)385/3323 (11.6%)707/3022 (23.4%)729/2679 (27.2%)
Appropriate prophylaxis149/303 (49.2%)248/385 (64.4%)458/707 (64.8%)449/729 (61.6%)
Surgical patients
Alerts sent1770/3175 (55.7%)1736/3262 (53.2%)1823/3659 (50.0%)1647/3587 (45.9%)
Appropriate prophylaxis1588/1770 (89.7%)1536/1736 (88.5%)1574/1823 (86.3%)1375/1647 (83.5%)

The incidence of VTE during hospitalisation is shown in Table 3. Compared to the first semester of 2005 (pre-intervention period) the implementation of the e-alert software was associated with a maintained reduction in the incidence of VTE among hospitalised patients, OR 0.50 (95% CI, 0.29–0.84), the impact being especially significant in medical patients, OR 0.44 (95% CI, 0.22–0.86) (Table 3). Interestingly among high-risk patients, VTE occurred despite adequate prophylaxis in 7/18 (39%) patients who developed VTE during the first semester of 2005, 3/9 (33%) patients in the first semester of 2006, 5/10 (50%) patients in the first semester of 2007, 7/9 (78%) patients in the first semester of 2008, and 4/7 (57%) patients in the first semester of 2009 (Table 4).

Table 3.   Incidence of VTE during hospitalisation
 2005*2006*2007*2008*2009*Pre- vs. post-intervention period OR
  1. *First semester.

Overall population
n644163126585667662660.50 (CI 95% 0.29–0.84)
VTE events211111119
Incidence (per 1000 patients)3.261.741.671.651.40
Medical patients
n339031373323302226790.44 (CI 95% 0.22–0.86)
VTE events147564
Incidence (per 1000 patients)4.132.231.501.991.49
Surgical patients
n305131753262365435870.80 (CI 95% 0.34–1.90)
VTE events74655
Incidence (per 1000 patients)2.291.261.841.371.40
Table 4.   VTE during hospitalisation according to risk stratification and prophylaxis use
 2005*2006*2007*2008*2009*
  1. *First semester; the number of patients included in the high-risk and low-risk categories was estimated by the e-alert software, except during the 2005 period, estimated by applying the mean percentage of high-risk patients of the following years 2006–2009.

High-risk patients (n)22702073212125402376
VTE events18 (7.9‰)9 (4.3‰)10 (4.7‰)9 (3.5‰)7 (2.9‰)
VTE despite prophylaxis7/18 (39%)3/9 (33%)5/10 (50%)7/9 (78%)4/7 (57%)
Low risk patients (n)41714239446441363890
VTE events3 (0.7‰)2 (0.4‰)1 (0.2‰)2 (0.5‰)2 (0.5‰)

No increase in the incidence of massive bleeding (defined by the use of at least six blood units in < 12 h) or intracranial bleeding in hospitalised patients was observed. The percentage of patients requiring at least two red cell concentrates was 8.5% before the use of e-alert and between 7.7% and 8.0% in the following years after e-alert implementation. The rate of surgical re-interventions due to post-operative bleeding remained similar during the study periods. Overall mortality in hospitalised patients ranged between 1.24% and 1.56% during the different study periods, despite a slight increase in the complexity score of in-patients in the last 2 years (data not shown).

Assessment of prophylaxis and VTE-associated costs

During the first semester of 2005, the standardised prophylaxis cost in hospitalised patients, including LMWH and compression stockings, amounted to €58 193 (about €9 per patient). After the implementation of e-alert, the standardised prophylaxis cost amounted to €64 045 during the first semester of 2006; €77 268 in 2007; €86 899 in 2008 and €80 685 in 2009. Although total in-patient prophylaxis costs increased during the first 3 years after implementation of e-alert (€6000–€13 000 per analysed period), the cost seems to have stabilised during the last year, leading to a mean incremental cost of the prophylaxis administered during hospitalisation of €3 per patient (Fig. 1).

image

Figure 1.  Cost per patient of prophylaxis, e-alert software design and maintenance and VTE developed during hospitalisation along the study periods (the overall costs are divided by all the patients that were hospitalised in our institution during the study periods).

Download figure to PowerPoint

Development of the e-alert software required an initial expenditure of €6000, followed by a maintenance expenditure of €750 per year. Initial cost and maintenance expenditure have been considered professedly amortised during the last 4 years.

Table 5 summarises the costs of procedures for diagnosis, treatment, prolongation of hospital stay and management of complications of all the patients, half of them cancer patients, who developed a VTE episode during hospitalisation before and after the implementation of the e-alert software. Median follow-up of all patients was 30 days (IQR 13–180). Overall mortality, mostly due to neoplastic underlying disease, in patients diagnosed with VTE during hospitalisation was 25/63 (39.7%) with a median follow up of 30 days (IQR 20–150). The total cost per DVT or PE event was variable, depending particularly on the kind of anticoagulant used for long-term treatment, LMWH vs. vitamin K-antagonists, and its duration. In our setting, the mean direct cost (during hospitalisation and after discharge) of an in-hospital VTE episode is €7058: €6050 for DVT and €8492 for PE, respectively.

Table 5.   Procedures performed and cost in patients diagnosed of VTE during hospitalisation in the first semester of 2005–2009
 20052006200720082009
Cost (€) DVT (n = 13)Cost (€) PE (n = 8)Cost (€) DVT (n = 5)Cost (€) PE (n = 6)Cost (€) DVT (n = 7)Cost (€) PE (n = 4)Cost (€) DVT (n = 6)Cost (€) PE (n = 5)Cost (€) DVT (n = 6)Cost (€) PE (n = 3)
Pharmacy (LMWH, UFH, AVK, Vena cava filter, compression stockings)13 50511 584.91037.86977.615 229.210 981.34966.66616.16297.92716
Laboratory and radiology tests12 005.515 933619811 845.517 3528329.514 296.58552.54891.55109
Professional fees30365863726111817161056132033020461914
Prolongation of Hospital stay (room and board)21 00034 25012 25019 75014 00015 00015 75018 50016 25011 000
Management of complication (bleeding, VTE recurrences)15 1307028447610 5424359351478737873
Total (excluding complications)49 546.567 630.920 211.839 691.148 297.235 366.836 333.133 998.629 485.420 379
Total × event (excluding complications)3811.38453.94042.46615.26899.68841.76055.16799.74914.36913
Total (including complications)64 676.574 658.920 211.844 167.161 759.239 725.839 847.141 871.637 358.420 379
Total × event (including complications)4975.19332.44042.47361.28405.69931.56641.28374.36226.46913

Counting on amortisation of the e-alert system divided by elapsed years (€0.35 per patient), together with the costs of increased use of thromboprophylaxis (from €9 to €12 per patient) the marked decrease of the incidence of VTE suggests that the software development has been a morbidity- and cost-saving policy. VTE incidence during hospitalisation decreased by 50%. Thus, comparing overall costs, the implementation of the e-alerts led to a saving of €6.54 per every hospitalised patient (Fig. 1).

A sensitivity analysis was displayed in the cost-effectiveness plane as an analysis of extremes. The worst and best case therapeutic scenarios were determined by considering the upper and lower cost estimates (obtained by adding or subtracting 25% from the baseline estimate) as well as the upper and lower limits of effectiveness (upper and lower 95% CIs of VTE rates) for both options (e-alert vs. no intervention). In both cases, the e-alert system is dominant over no intervention, providing cost savings and avoiding additional VTE events (Table 6).

Table 6.   Results of the incremental cost-effectiveness analysis for the e-alert system vs. no intervention for VTE prophylaxis in hospitalised patients. Sensitivity analysis was performed by worst-case/best-case scenario analysis
 Total cost (€) per 1000 patientsTotal VTE events per 1000 patientsICER
  1. ICER, incremental cost-effectiveness ratio; *a negative difference favours e-alert system, representing cost savings and/or events avoided.

Base case
E-alert24 1161.6 
No intervention30 6633.3 
Difference*−6543−1.7Dominant
Worst case
E-alert34 7582.1 
No intervention49 7994.7 
Difference*−15 041−2.6Dominant
Best case
E-alert system15 3161.1 
No intervention16 1131.9 
Difference*−797−0.8Dominant

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

This is the first health economic analysis showing the cost-effectiveness of an e-alert system to prevent VTE in hospitalised patients. Prevention of in-hospital VTE is identified internationally as a priority to improve patient safety [9,24]. However, there is a gap between clinical trial evidence favouring prophylaxis and physicians’ ordering preventive measures among hospitalised patients at risk for VTE. The use of e-alerts can help clinicians to identify hospitalised patients at risk of VTE and improve the use of thromboprohylaxis [18–21]. In our experience, the implementation of e-alert software was associated with an increase in the appropriate use of thromboprophylaxis and a marked reduction of VTE episodes during hospitalisation, particularly in medical patients who suffer PE more often than non-medical patients [25]. More importantly, the reduction in the incidence of VTE during hospitalisation has maintained or even improved over time. So far, the ‘alert fatigue’ phenomenon described in other studies using e-alerts for the prevention of VTE has not been clearly observed at our institution [21]. On the contrary, acceptance of the alert program by clinicians has increased over time as they are becoming more familiar with it. This can explain the higher number of alerts sent in medical patients, due to a higher rate of fulfillment of the simple electronic questionnaire at admission by physicians (50% in an audit performed in 2007 and over 80% in 2008 and 2009), together with an increased proportion of at-risk patients receiving appropriate thromboprophylaxis. Thus, even though the thrombotic risk of some patients might have been infraestimated, leading to a shortcoming effect on the impact of the e-alerts, our results show a significant reduction in the incidence of in-hospital VTE.

However, the impact in economy terms of any e-alert system had not been evaluated yet. The use of e-alerts led to an increase in the use of thromboprophylaxis in hospitalised patients with an extra cost of between €6000 and €13 000 per year, representing a mean incremental cost of about 3€ per hospitalised patient. Every avoided VTE episode saved a mean expenditure of €7058, without differences related to decrease in the incidence of the most severe and expensive process since DVT/PE ratio has remained similar along the study periods. The direct cost of a VTE episode in this report is relatively similar to the estimations of other Spanish and foreign health institutions [7,26,27], taking into account that we considered not only in-hospital direct costs, but also long-term treatment, follow-up visits and management of complications during anticoagulant treatment. A recent review performed in US concluded that the cost of managing an initial episode of DVT was estimated at $7712–10 804 and for an initial PE event $9566–16 644 [28]. Of note in the present study, almost half of the patients developing VTE during hospitalisation were cancer patients, a population with worse outcome of thromboembolic episodes in terms of recurrence, bleeding and mortality [29], in whom the management of acute VTE can cost more than $20 000 [28]. After adding the costs of the design, implementation and maintenance of the software itself, the result is markedly cost-effective, with an overall saving of €6.5 per hospitalised patient. Should all hospitalised patients in one single year in Spain be considered, (more than 4 500 000) total yearly savings would approach €30 million and approximately 7650 additional VTE events could be avoided. Furthermore, the cost-effectiveness sensitivity analysis showed cost savings of e-alerts in both worst and best case scenarios, supporting the robustness of the results,

One of the main differences of our study compared to other e-alerts reports is that our approach was to apply the software and the evaluation of results to the whole hospital inpatient population, not just a selected high-risk population. Despite a potential attenuating effect of this generalised approach on the impact of the e-alert, the results observed are largely positive, both in terms of clinical benefit and economic impact.

The present health economic analysis has some limitations. First, we assessed the cost-effectiveness from an institutional perspective, without taking into account quality-adjusted life years, making it difficult to compare the analysis with other health technologies. However, we believe such analysis is still very relevant as it reveals that adopting a cost-effective institutional strategy can also have a great impact from a societal perspective. Second, the economic evaluation of the e-alert was performed retrospectively, relying on discharge records and, therefore, was subject to limitations. Although the clinical characteristics of hospitalised patients are not exactly the same in all the different periods analysed, the complexity score of in-patients was similar during the study periods, even with a slight increase in the last 2 years. Importantly, in-hospital stay was also similar in all groups and the same thromboprophylaxis and VTE-treatment protocols were active. Third, follow-up of all patients after discharge could not be performed, except of those with in-hospital VTE, and the evaluation of the use of thromboprophylaxis and the incidence of VTE was limited to the hospitalisation period. Thus, the possibility that the benefit of e-alerts was not maintained after hospital discharge cannot be ruled out. However, there is a link between out-patient and in-patient cases of VTE, and it seems reasonable to believe that a more appropriate use of thromboprophylaxis during hospitalisation should be associated with better indications at discharge, both in high-risk medical and surgical patients [30]. In addition, if the appropriateness of prophylaxis during hospitalisation increases after e-alerts implementation, probably the rates of early after-discharge-VTE will be reduced accordingly. Should this hypothesis be confirmed, the economic impact of e-alerts would be even greater.

Fourth, data regarding the incidence of major bleeding or clinically relevant (as defined in clinical trials of anticoagulant drugs) complications secondary to thromboprophylaxis before and after implementation of the e-alerts are lacking, although the incidence of major bleeding associated with prophylactic doses of LMWH has been shown to be low in clinical trials in both surgical and medical patients [14,31]. On the other hand, we have observed no significant increase in the number of patients requiring at least two red cell concentrates during hospitalisation, re-operation due to bleeding, nor in the incidence of massive bleeding or intracranial bleeding in hospitalised patients during the study periods.

No other intervention, neither local nor national campaign for VTE prevention, was started at the time this study was conducted, so the possibility that the results may come from a general increase in awareness of VTE rather than an effect of the e-alert is unlikely.

Of note in the present study, LMWHs were the drug of choice for thromboprophylaxis, which is a cost-effective alternative for thromboprophylaxis [32], whereas LMWH and vitamin-K antagonists were used for short and long-term VTE treatment. With the near use of new oral anticoagulants for prophylaxis and treatment of VTE the results obtained could be different. Also, the implementation of second generation softwares such as multi-screen e-alerts can further increase the use of thromboprophylaxis among physicians who had declined an initial single screen alert and could improve the impact [33].

Finally, only descriptive statistics have been performed. The size of our hospital and the recent implementation of the e-alert software do not allow exhaustive statistical analyses other than objectivation of a significant reduction of annual incidence of VTE and associated costs. However, a strength of this analysis is that it provides data in a real-world setting.

In conclusion, e-alerts are very useful tools to improve the use of prophylaxis and reduce the incidence of VTE during hospitalisation, particularly in medical patients, thus representing a cost-effective thromboprophylaxis strategy in hospitalised patients. Fewer complications and lower costs will be achieved by implementing in-hospital electronic alerts.

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

R. Lecumberri, E. García-Quetglas and J.A. Páramo designed the research and wrote the manuscript. R. Lecumberri, E. Panizo, A. Gomez-Guiu, S. Varea, E. García-Quetglas, M. Serrano and M. Marqués obtained and analysed the patients’s data. R. Lecumberri, J.A. Páramo and A. García-Mouriz designed the e-alert software. R. Lecumberri, A. Gómez-Outes and J.A. Páramo substantially edited the manuscript. All authors revised and approved the final version of the manuscript.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

This study was partially funded by a grant from the Instituto Carlos III para la evaluación de tecnologías sanitarias (PI05/900094). The views and opinions expressed in this article are of the authors alone and do not necessarily represent the official views of their institutions or any other party.

Disclosure of Conflict of Interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References

The authors state that they have no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interest
  10. References
  • 1
    Spencer FA, Emery C, Lessar D, Anderson F, Emani S, Aragam J, Becker RC, Goldberg RJ. The Worcester Venous Thromboembolism study: a population-based study of the clinical epidemiology of venous thromboembolism. J Gen Intern Med 2006; 2: 7227.
  • 2
    Spencer FA, Gore JM, Lessard D, Douketis JD, Emery C, Goldberg RJ. Patient outcomes after deep vein thrombosis and pulmonary embolism: the Worcester Venous Thromboembolism Study. Arch Intern Med 2008; 168: 42530.
  • 3
    Lecumberri R, Soler S, Del Toro J, Barba R, Rosa V, Ciammaichella MM, Monreal M; the RIETE Investigators. Effect of the time of diagnosis on outcome in patients with acute venous thromboembolism. Findings from the RIETE Registry. Thromb Haemost 2011; 105: 4551.
  • 4
    Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, Cattelan AM, Polistena P, Bernardi E, Prins MH. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996; 125: 17.
  • 5
    Kahn SR, Shrier I, Julian JA, Ducruet T, Arsenault L, Miron MJ, Roussin A, Desmarais S, Joyal F, Kassis J, Solymoss S, Desjardins L, Lamping DL, Johri M, Ginsberg JS. Determinants and time course of the post-thrombotic syndrome after acute deep venous thrombosis. Ann Intern Med 2008; 149: 698707.
  • 6
    Pengo V, Lensing AW, Prins MH, Marchiori A, Davidson BL, Tiozzo F, Albanese P, Biasiolo A, Pegorano C, Lliceta S, Prandoni P. Incidence of chronic thromboembolic pulmonary hypertension after pulmonary embolism; Thromboembolic Pulmonary Hypertension Study Group. N Engl J Med 2004; 350: 225764.
  • 7
    Grupo Multidisciplinar para el Estudio de la Enfermedad Tromboembólica en España. Estudio sobre la Enfermedad Tromboembólica Venosa en España. Sociedad Española de Medicina Interna (SEMI). 2006. Depósito Legal: M-30133-2006.
  • 8
    Guijarro Merino R, Montes Santiago J, San Román Terán CM. Epidemiology of venous thromboembolic disease in Spain. Med Clin (Barc). 2008; 131(Suppl. 2): 29.
  • 9
    Goldhaber SZ. Eradication of hospital-acquired venous thromboembolism. Thromb Haemost 2010; 104: 108992.
  • 10
    de Lissovoy G. Economic issues in the treatment and prevention of deep vein thrombosis from a managed care perspective. Am J Manag Care 2001; 7(Suppl.): S5358.
  • 11
    Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, Huang W, Zayaruzny M, Emery L, Anderson FA Jr; ENDORSE Investigators. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. Lancet 2008; 71: 8794.
  • 12
    Goldhaber SZ. Preventing pulmonary embolism and deep vein thrombosis: a ‘call to action’ for vascular medicine specialists. J Thromb Haemost 2007; 5: 16079.
  • 13
    Galson SK. The Surgeon General’s Call to Action to Prevent Deep Vein Thrombosis and Pulmonary Embolism. http://www.surgeongeneral.gov/topics/deepvein/. Accessed 29 April 2010.
  • 14
    Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, Colwell CW; American College of Chest Physicians. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133(Suppl.): 381S453S.
  • 15
    American Public Health Association. Deep vein thrombosis: advancing awareness to protect patient lives-White paper. http://www.apha.org/ppt/DVT_White_paper.pdf. Accessed 29 April 2010.
  • 16
    Weller WE, Gallagher BK, Cen L, Hannan EL. Readmissions for venous thromboembolism: expanding the definition of patient safety indicators. Joint Commission Journal on Quality and Patient Safety 2004; 30: 497504.
  • 17
    Streiff MB, Haut ER. The CMS ruling on venous thromboembolism after total knee or hip arthroplasty: weighing risks and benefits. JAMA 2009; 301: 10635.
  • 18
    Kucher N, Koo S, Quiroz R, Cooper JM, Paterno MD, Soukonnikov B, Goldhaber SZ. Electronic alerts to prevent venous thromboembolism among hospitalized patients. N Engl J Med 2005; 352: 96977.
  • 19
    Lecumberri R, Marqués M, Díaz-Navarlaz MT, Panizo E, Toledo J, García-Mouriz A, Páramo JA. Maintained effectiveness of an electronic alert system to prevent venous thromboembolism among hospitalized patients. Thromb Haemost 2008; 100: 699704.
  • 20
    Baroletti S, Munz K, Sonis J, Fanikos J, Fiumara K, Paterno M, Goldhaber SZ. Electronic alerts for hospitalized high-VTE risk patients not receiving prophylaxis: a cohort study. J Thromb Thrombolysis 2008; 25: 14650.
  • 21
    Kucher N, Puck M, Blaser J, Bucklar G, Eschmann E, Lüscher TF. Physician compliance with advanced electronic alerts for preventing venous thromboembolism among hospitalized medical patients. J Thromb Haemost 2009; 7: 12916.
  • 22
    Piazza G, Goldhaber SZ. Computerized decision support for the cardiovascular clinician: applications for venous thromboembolism prevention and beyond. Circulation 2009; 120: 11337.
  • 23
    Alonso C, Medrano FJ, Romero A, Villar E, Calderón E, Marín I. Guía PRETEMED 2003. Guía de profilaxis de enfermedad tromboembólica en patología médica. Córdoba: SADEMI, 2003.
  • 24
    Baglin T. Defining the population in need of thromboprophylaxis – making hospitals safer. Br J Haematol 2010; 149: 80512.
  • 25
    Piazza G, Fanikos J, Zayaruzny M, Goldhaber SZ. Venous thromboembolic events in hospitalised medical patients. Thromb Haemost 2009; 102: 50510.
  • 26
    RESOLUCIÓN 882/2010, de 3 de mayo, del Director Gerente del Servicio Navarro de Salud-Osasunbidea, por la que se actualizan las tarifas por prestación de servicios en los centros y establecimientos asistenciales del Servicio Navarro de Salud-Osasunbidea. http://www.navarra.es/home_es/Actualidad/BON/Boletines/2010/71/Anuncio-10/. Accessed 31 October 2010.
  • 27
    Merli G, Ferrufino C, Lin J, Hussein M, Battleman D. Hospital-based costs associated with venous thromboembolism treatment regimens. J Thromb Haemost 2008; 6: 107786.
  • 28
    Dobesh PP. Economic burden of venous thromboembolism in hospitalized patients. Pharmacotherapy 2009; 29: 94353.
  • 29
    Monreal M, Falgá C, Valdés M, Suárez C, Gabriel F, Tolosa C, Montes J; RIETE Investigators. Fatal pulmonary embolism and fatal bleeding in cancer patients with venous thromboembolism: findings from the RIETE registry. J Thromb Haemost 2006; 4: 19506.
  • 30
    Goldhaber SZ. Outpatient venous thromboembolism: a common but often preventable public health threat. Arch Intern Med 2007; 167: 14512.
  • 31
    Hull RD, Schellong SM, Tapson VF, Monreal M, Samama MM, Nicol P, Vicaut E, Turpie AG, Yusen RD; EXCLAIM (Extended Prophylaxis for Venous ThromboEmbolism in Acutely Ill Medical Patients With Prolonged Immobilization) study. Extended-duration venous thromboembolism prophylaxis in acutely ill medical patients with recently reduced mobility: a randomized trial. Ann Intern Med 2010; 153: 818.
  • 32
    Wilbur K, Lynd L, Sadatsafavi M. Low-molecular-weight heparin versus unfractionated heparin for prophylaxis of venous thromboembolism in medicine patients: a pharmacoeconomic analysis. Clin Appl Thromb Hemost 2010; DOI 10.1177/1076029610376935.
  • 33
    Fiumara K, Piovella C, Hurwitz S, Piazza G, Niles CM, Fanikos J, Paterno M, Labreche M, Stevens LA, Baroletti S, Goldhaber SZ. Multi-screen electronic alerts to augment venous thromboembolism prophylaxis. Thromb Haemost 2010; 103: 3127.