Multifaceted Intervention to Prevent Venous Thromboembolism in Patients Hospitalized for Acute Medical Illness: A Multicenter Cluster-Randomized Trial

Abstract

Background

Misuse of thromboprophylaxis may increase preventable complications for hospitalized medical patients.

Objectives

To assess the net clinical benefit of a multifaceted intervention in emergency wards (educational lectures, posters, pocket cards, computerized clinical decision support systems and, where feasible, electronic reminders) for the prevention of venous thromboembolism.

Patients/Methods

Prospective cluster-randomized trial in 27 hospitals. After a pre-intervention period, centers were randomized as either intervention (n = 13) or control (n = 14). All patients over 40 years old, admitted to the emergency room, and hospitalized in a medical ward were included, totaling 1,402 (712 intervention and 690 control) and 15,351 (8,359 intervention and 6,992 control) in the pre-intervention and intervention periods, respectively.

Results

Symptomatic venous thromboembolism or major bleeding (primary outcome) occurred at 3 months in 3.1% and 3.2% of patients in the intervention and control groups, respectively (adjusted odds ratio: 1.02 [95% confidence interval: 0.78–1.34]). The rates of thromboembolism (1.9% vs. 1.9%), major bleedings (1.2% vs. 1.3%), and mortality (11.3% vs. 11.1%) did not differ between the groups. Between the pre-intervention and intervention periods, the proportion of patients who received prophylactic anticoagulant treatment more steeply increased in the intervention group (from 35.0% to 48.2%: +13.2%) than the control (40.7% to 44.1%: +3.4%), while the rate of adequate thromboprophylaxis remained stable in both groups (52.4% to 50.9%: -1.5%; 49.1% to 48.8%: -0.3%).

Conclusions

Our intervention neither improved adequate prophylaxis nor reduced the rates of clinical events. New strategies are required to improve thromboembolism prevention for hospitalized medical patients.

Trial Registration

ClinicalTrials.gov NCT01212393

About the Authors:

Pierre-Marie Roy

* E-mail: [email protected]

Affiliation: Département de Médecine d’Urgence, Centre Vasculaire et de la Coagulation, CHU Angers, Institut MITOVASC, EA3860, Université d’Angers, Angers, France

Antoine Rachas

Current address: Santé Publique-Epidémiologie, Hôpital Bicêtre, Assistance Publique Hôpitaux de Paris, Le Kremlin-Bicêtre, France

Affiliation: Unité d’Épidémiologie et de Recherche Clinique, Assistance Publique Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Centre d’investigation Épidémiologique 4, INSERM, Paris, France

Guy Meyer

Affiliation: Service de Pneumologie et Soins Intensifs, Hôpital Européen Georges Pompidou, Assistance Publique Hôpitaux de Paris, Université Paris Descartes, Sorbonne Paris Cité, INSERM U 970, CIC 1418, Paris, France

Grégoire Le Gal

Current address: Department of Medicine, Division of Haematology—Thrombosis Program, Ottawa Hospital Research Institute and University of Ottawa, Ottawa, Canada

Affiliation: Département de Médecine Interne et Pneumologie, CHU de la Cavale Blanche, Université de Bretagne Occidentale, EA3878 (GETBO), CIC INSERM 1412, Brest, France

Pierre Durieux

Affiliation: Département de Santé Publique et Informatique Médicale, Hôpital Européen Georges Pompidou, Assistance Publique Hôpitaux de Paris, Université Paris Descartes, Sorbonne Paris Cité, INSERM UMRS 872, Centre de recherche des Cordeliers, Paris, France

Dominique El Kouri

Affiliation: Service des Urgences, Médecine Polyvalente, CHU Hôtel Dieu, Nantes, France

Didier Honnart

Affiliation: Département de Médecine d’Urgence, CHU Dijon, Hôpital du Bocage, Dijon, France

Jeannot Schmidt

Affiliation: Service d’Accueil des Urgences, CHU Gabriel Montpied, Université de Clermont-Ferrand, Clermont-Ferrand, France

Catherine Legall

Affiliation: Service des Urgences, CH Argenteuil Victor Dupouy, Argenteuil, France

Pierre Hausfater

Affiliation: Service des Urgences, CHU La Pitié Salpétrière, Assistance Publique Hôpitaux de Paris, Université Paris 06 UPMC, Paris, France

Jean-Marie Chrétien

Affiliation: Délégation à la recherche clinique et innovation, cellule de gestion des données et évaluation, CHU Angers, Angers, France

Dominique Mottier

Affiliation: Département de Médecine Interne et Pneumologie, CHU de la Cavale Blanche, Université de Bretagne Occidentale, EA3878 (GETBO), CIC INSERM 1412, Brest, France

PREVENU study group

¶Membership of the PREVENU study group is provided in the Acknowledgments.

Introduction

Hospitalization for acute medical illness is a significant risk factor of venous thromboembolism (VTE), accounting for up to 20% of all VTEs and 80% of in-hospital fatal cases of pulmonary embolism.[1–3] Pharmacological regimens have been shown to reduce thromboembolic events[4, 5] and for over a decade, consensus guidelines have recommended thromboprophylaxis for high-risk medical patients.

Yet prophylaxis appears to be used inappropriately and often underused for hospitalized medical patients.[6, 7] As system-wide standardized interventions may be more effective than relying on individual physicians’ routine practices, the American College of Chest Physicians has recommended “for every hospital, that a formal active strategy addressing the prevention of VTE be developed”.[5] A recent meta-analysis has suggested that alerts or multifaceted interventions increase prophylaxis prescription, although how this finding applies to a VTE or bleeding setting remains unknown, given that most studies are underpowered to assess these outcomes.[8]

We hypothesized that a multifaceted intervention for VTE guidelines implementation in emergency departments should improve prophylaxis use for patients hospitalized in medical wards and decrease the rate of VTE or major bleedings.

Methods

Design

We performed a multicenter prospective controlled cluster-randomized trial in community and academic hospitals.

Centers

Hospitals were eligible for participation if the annual number of visits in their emergency department was >30,000.

Patients

We prospectively enrolled all consecutive patients over 40 years old presenting to the emergency department of the participating sites for acute medical illness and requiring hospitalization in a medical ward.

Patients were excluded from analysis if they were hospitalized for less than 48 hours, if venous thrombosis or pulmonary embolism was diagnosed within 48 hours following admission (in order to rule out any VTE that had occurred before hospital admission), and if they received anticoagulant treatment at a therapeutic dosage for another reason than acute VTE at the admission and/or during hospitalization.

Pre-intervention period

All included emergency departments participated in a 1-week observational pre-intervention period. During this period, while information was collected on thromboprophylaxis use, patients were not followed-up for 3 months. This pre-intervention period was aimed at obtaining baseline values for comparison with the intervention period.

Intervention period

A random number table was used to assign hospitals to either the intervention or standard practice group (control group). Randomization was stratified in order to include the same number of academic hospitals and centers using a computerized medical file in each group.

Implementation of recommendations in the intervention group

The intervention was multifaceted, based on educational lectures, posters, and pocket cards, computerized clinical decision support systems, and computerized reminders (S1 and S2 Posters).

Emergency physicians and residents attended a lecture that presented the guidelines for thromboprophylaxis use in acutely ill medical patients.[5, 9] In summary, thromboprophylactic treatment was recommended for high-risk patients, i.e. patients confined to bed presenting with an acute medical condition associated with a high risk of VTE, and patients presenting with an acute medical condition associated with an intermediate risk of VTE and, at least, one VTE risk factor (Table 1). In patients for whom prophylaxis was recommended but with contraindication to antithrombotic treatment, a mechanical prophylaxis method was recommended, namely by means of compression stockings or intermittent pneumatic compression devices. The key objectives were to systematically evaluate the risk of thrombosis and to prescribe and initiate venous thromboembolism prophylaxis as soon as possible in those at high risk (Table 1).

[Figure omitted. See PDF.]

Table 1. Definition of high-risk patients and recommendations of thromboprophylactic treatment.

https://doi.org/10.1371/journal.pone.0154832.t001

The indications for thromboprophylaxis in hospitalized medical patients were summarized on posters and pocket cards. The list of antithrombotic treatments approved for this indication and their recommended doses were also provided.[5, 9]

A computer-based clinical decision support system was developed and installed on the emergency departments’ computers in the intervention group. According to patient diagnosis and comorbidities, this system provided information on whether or not thromboprophylaxis was recommended, and, should this be the case, indicated the adequate treatment and doses.

In the emergency wards with access to computerized medical files, we planned to implement a software program which would systematically remind the consulting physician to assess thrombosis risk and start prophylaxis in patients hospitalized in a medical setting.

The intervention period lasted for 8 to 10 weeks.

Control group

No intervention was performed in the centers allocated to the control group. The principal investigators were instructed to continue their practice as usual.

Clinical outcomes

The primary outcome was defined at the patient level as a composite endpoint comprising of symptomatic VTE events and major bleedings during the three months following hospital admission. VTE events, major bleedings, and all-cause mortality during the 3 months following admission and during hospitalization were secondary outcomes.

VTE was defined as one of the following events: i) deep vein thrombosis, ii) pulmonary embolism, and iii) sudden death with no obvious cause (possible fatal pulmonary embolism). The VTE events had to be symptomatic and confirmed by objective tests.[10, 11] Major bleeding was defined according to the International Society on Thrombosis and Haemostasis’ criteria (any bleeding resulting in death, in a critical organ or resulting in the transfusion of at least two packs of blood red cells).[12] For patients who experienced several events, i.e. of venous thromboembolism and/or major bleedings, only the first was taken into account.

An independent adjudication committee, blinded to the group allocation, performed a chart review and analyzed all suspected outcome events.

Prophylaxis adequacy

An adequacy assessment was performed, which was blinded to group assignment and mostly automated, based on several criteria:

* Patient’s risk of VTE (high-risk or low-risk) according to the main reason for admission and VTE risk factors (Table 1)

* Delay between admission and treatment initiation (< 5 days)

* Duration of anticoagulation (at least 5 days or until discharge)

* Eventual contraindication for anticoagulant treatment, especially bleeding

* Dosage of anticoagulant treatment according to recommendations.

Thromboprophylaxis was considered as adequate if: i) for high-risk patients with thromboprophylaxis recommended as per the guidelines, antithrombotic treatment was initiated before Day 5 and administered at the correct dosage for at least 5 days or ii) for low-risk patients with no indication or those with contra-indication for anticoagulant treatment, antithrombotic treatment was not administered [5, 9]. Mechanical thromboprophylaxis was not assessed.

As thromboembolism could occur very early after admission [3] and in order to assess the rate of thromboprophylactic treatment prescribed by the emergency physicians, we performed a sensitivity analysis considered as adequate for high-risk patients, antithrombotic treatment initiated within the first 2 days following admission.

Sample size

With the study design assuming 30 participating centers divided into two groups of 15, an intra-cluster correlation of 0.01, a 5% combined rate of VTE or major bleedings in the control group, a total of 16,170 patients with 8,085 in each group were required to detect a 1.5 percentage absolute difference between the two groups, with 3.5% in the intervention group, at a power of 80% and significance level of 5%.[13]

Considering that approximately 15% of patients would be hospitalized for less than 48 hours and 5% would be lost to follow-up, we planned to enroll a total of 20,000 patients.

Statistical analysis

The statistical analysis plan was defined following closure of the database, prior to any statistical analysis (S1 Statistical analysis plan). The statistician was blinded to the randomization group and center names. We estimated the odds ratio in mixed-effects logistic regression, adjusting for significant patient- and center-level confounders and taking into account the dependence between patients from the same hospital, also known as the clustering effect.[14]

Subgroup analyses were conducted for age (≤ or >75 years old) and according to whether or not a treatment was recommended. Sensitivity analyses were conducted using other definitions of the primary endpoint, not considering unexplained sudden deaths, such as venous thromboembolism events, and considering all unexplained deaths as venous thromboembolism events.

The change in practice adequacy was compared between the pre-intervention and intervention periods by using a mixed-effects logistic regression model, including study group, study period, and a term for the group-by-period interaction, adjusting for significant confounders. Adjusted absolute differences from the marginal predictions of probabilities in the two groups were derived from the logistic equation. P-values and 95% confidence intervals (95% CIs) were computed by using the standard errors estimated with the delta method. In order to compare our results to others, we performed a post-hoc analysis of the rate of thromboprophylactic treatment regardless of adequacy to recommendations.

We estimated intra-class correlation by using the Murray formula.[15] All statistical analyses were performed using the STATA software (release 11; Stata Corp., College Station, Texas, USA).

Ethics

The Ethics Committee of Angers University Hospital approved this study for all centers. The study was registered and approved by the French competent authorities on June 04, 2009 prior to perform the first inclusion (n°: 09–256). French regulations consider randomized cluster trials aiming to improve the implementation of good practice recommendations as non-interventional trials and do not require written consent from participants. However, we sought oral consent from patients for follow-up and informed all patients of their right to request the withdrawal of their personal data. The study was also registered at the international registry of clinical trials on September 29, 2010 (ClinicalTrials.gov Identifier: NCT01212393).

Results

Clusters

In total, 40 hospitals were screened for eligibility, six refused participation, and seven could not participate for logistical reasons (Fig 1). The 27 participating centers were all located throughout France and included 20 academic and seven community hospitals. All completed the study between November 2009 for the first center and November 2010 for the last center.

[Figure omitted. See PDF.]

Fig 1. Flow chart.

https://doi.org/10.1371/journal.pone.0154832.g001

At the end of the pre-intervention period, the centers were randomized into the intervention group (n = 13) or control group (n = 14). Ten centers were academic hospitals and seven centers used computerized medical files, in each group.

In each of the intervention group centers, educational lectures were organized and posters, pocket cards, and a downloadable computerized clinical decision support system were provided. However, for technical and administrative reasons, computerized reminders could be implemented in only two centers.

Patients

A total of 20,377 patients were enrolled, 1,686 in the pre-intervention period and 18,691 in the intervention period. Of these, 284 (16.8%) and 3,340 patients (17.9%) were secondarily excluded primarily due to early discharge, in the pre-intervention and intervention periods, respectively, leaving 1,402 (712 in the intervention group and 690 in the control group) and 15,351 (8,359 intervention and 6,992 control) patients in each period, respectively. The rate of patients lost to follow-up was 3.9%, resulting in a total of 14,760 patients analyzed for the primary outcome, composed of 8,068 in the intervention group and 6,692 in the control group (Fig 1).

All data regarding demographic characteristics, past medical history and co-morbidities, medication on admission, main acute medical condition (principal reason for hospitalization), procedures performed during hospitalization, and length of hospital stay has been provided in Table 2 and S1 Table.

[Figure omitted. See PDF.]

Table 2. Characteristics of the study population.

https://doi.org/10.1371/journal.pone.0154832.t002

Primary and secondary clinical outcomes

The primary outcome occurred in 250 of 8,359 patients (3.1%) in the intervention group and in 214 of 6,992 (3.2%) in the control group (adjusted odds ratio: 1.02; 95% CI: 0.78–1.34; p = 0.87) (Table 3). Thromboembolic events (1.9% vs. 1.9%), major bleedings (1.2% vs. 1.3%), and all-cause mortality (11.3% vs. 11.1%) did not differ between the groups at three months.

[Figure omitted. See PDF.]

Table 3. Clinical events.

https://doi.org/10.1371/journal.pone.0154832.t003

There was no difference regarding thromboembolic events or major bleedings between the two arms in patients aged ≤75 years or >75 years, nor between those with or without an indication for prophylaxis. The use of different definitions of VTE did not change the results. The in-hospital event rates did not differ between the groups (S2, S3 and S4 Tables).

Prophylactic practice outcomes

During the pre-intervention period, the rate of adequate thromboprophylaxis practice was 52.4% in the intervention group and 49.1% in the control group. During the intervention period, the rate of adequate practice did not change significantly in either group (-1.5% in the intervention group and -0.3% in the control group) (Table 4). Similar results were produced if the thromboprophylaxis was considered adequate only if the treatment was initiated within the first 2 days, as opposed to 5, or, in the subgroup analysis in terms of whether an antithrombotic treatment was indicated (S5 and S6 Tables).

[Figure omitted. See PDF.]

Table 4. Thromboprophylaxis practice adequacy.

https://doi.org/10.1371/journal.pone.0154832.t004

Indeed, in the intervention group, the rate of high-risk patients for whom prophylaxis was indicated (Table 1) and who received adequate antithrombotic treatment increased during the intervention period (+6.8%), though this was accompanied by an increase in the rate of high-risk patients who received inadequate treatment because of inadequate dosage, delay, or duration (+2.3%) and in the rate of patients who received treatment when prophylaxis was not indicated or antithrombotic treatment was contraindicated (+3.4%). In the control group, the rate of patients for whom prophylaxis was indicated and who received adequate antithrombotic treatment increased (+5.0%), yet the rate of treatment remained stable for the other patients (Table 4).

The overall proportion of patients who received prophylactic anticoagulant treatment, regardless of the adequacy of the prescription, increased significantly more in the intervention group (+13.2% [35.0% to 48.2%]) than in the control group (+3.4% [40.7% to 44.1%]) (adjusted between-group difference in the change: 6.6% [1.6 to 11.6]; post-hoc analysis; S7 Table).

Discussion

In this cluster-randomized trial, a multifaceted intervention aimed at implementing venous thromboembolism prophylaxis recommendations did not reduce the rate of symptomatic thromboembolic events or major bleedings within 3 months following hospitalization for acute medical illness. Although it led to an increase in the rate of antithrombotic treatment use, the intervention did not increase the rate of adequate prophylaxis.

Several explanations might account for these negative and unexpected results. First, we planned to implement a multifaceted intervention including a computer-based clinical decision support system and computerized reminders in the intervention group because such interventions appear to be the most effective system-wide measures for improving the quality of care in hospitals.[8, 16] Although all other components of the intervention were implemented in all sites in the intervention group, we were only able to implement plugin reminders in two centers. Two sites had no computerized medical file, four had a computerized file incompatible with the plugin and in five centers, hospital policy did not allow plugin implementation. Moreover, the plugin that we implemented in these two centers could not select high-risk patients and provided reminder for all hospitalized medical patients. This result emphasizes that, complex interventions based on information technology systems and equipment remain difficult to generalize.[17]

Second, the intervention did not increase the rate of adequate prophylaxis. Although an increase in the rate of prophylaxis was observed in the intervention group, this was partially explained by an increase in over-treatment in patients who were not at risk, along with an increase in inappropriate treatment in others. Of note, these outcomes had hardly ever previously been studied. Most previous trials were focused on high-risk patients and did not analyze change versus baseline prescription rates.[17–22] Only one previous study demonstrated a reduction in over-treatment using a specific anticoagulant prescription form, [23] yet other reports echo our own in their apparent increases in over-treatment.[24] No prior study has assessed the rate of inadequate treatment, even though inappropriate dosages appear inefficient in the prevention of VTE and over-treatment may expose patients to an unnecessary risk of bleeding.[25]

Third, assessing the thromboembolic risk in medical patients is not always straightforward. Current recommendations are mostly based on the main reason for admission, but a clear definition of most diseases known to be at-risk of VTE is lacking. This is at least the case for acute respiratory failure and rheumatic disorders. The diagnosis may also be uncertain at the time of admission and the definition of bedrest may vary across physicians. A simpler and more reliable risk assessment model may be more effective for a system-wide intervention.[26]

To our knowledge, this study is the first randomized trial planned to assess the clinical benefit-risk ratio of such a multifaceted intervention in unselected hospitalized patients. Among the ten prior randomized studies [17–24, 27], only one demonstrated a decrease in thromboembolic events.[20] This single-center study was based on a complex strategy based on the analysis of the computerized medical files and using electronic alerts in a selected group of untreated high-risk medical and surgical patients (80% patients with underlying cancer). The other studies reported so far were weakened by their study design (primarily before-and-after cycle), size, and outcome criteria (most often not clinical events).[8] A recent meta-analysis of randomized control trials evaluating different interventions to improve thromboprophylaxis has reported an increase in the proportion of high-risk patients receiving treatment, though has failed to demonstrate an improvement in clinical events.[8] Our study, including a large number of patients providing a high power to detect a clinically significant improvement in outcomes, confirms these results.

The main strengths of our study are its size, design, and primary outcome, assessing the clinical benefit of the intervention rather than processes of care. Although the number of participating centers was 27 instead of 30, the statistical power and the cluster-randomized design, allowing us to check for time effect and avoid contamination bias, give us confidence in our results. The study also has some limitations. Firstly, some of our criteria for adequate prophylaxis based on 2008 guidelines may today be debatable.[26, 28] Secondly, we focused our analysis on antithrombotic treatment and did not assess mechanical thromboprophylaxis. Thirdly, due to the lack of stratification by the centers’ activities, there were some differences in the number of patients and patients’ characteristics between the two groups. Fouthly, the rate of thromboembolic events in the control group was lower than expected. Physicians were aware of an ongoing study of thromboprophylaxis and may have improved their baseline practice (Hawthorne effect). Finally, ICU admission and invasive treatments were not recorded.

In conclusion, our multifaceted intervention, aiming to implement current recommendations on thromboprophylaxis for unselected medical patients, failed to demonstrate any clinical benefit. The intervention was associated with an increase in the use of antithrombotic treatment but no increase in adequate thromboprophylaxis was observed. New strategies are required to address thromboprophylaxis in hospitalized medical patients, including simpler assessment of the risk of VTE, at least in hospitals where complex strategies based on computerized reminders and alerts are not implementable.

Statistical analysis

The statistical analysis was performed by Dr Antoine Rachas and supervised by Pr Gilles Chatellier (Centre d’investigation Épidémiologique 4, INSERM, Paris, France; Unité d’Épidémiologie et de Recherche Clinique, Assistance Publique Hopitaux de Paris, Hôpital Européen Georges Pompidou, Paris, France).

Supporting Information

[Figure omitted. See PDF.]

S1 CONSORT Checklist. CONSORT extension for cluster trial checklist.

https://doi.org/10.1371/journal.pone.0154832.s001

(DOCX)

S1 Poster. Poster A used in the intervention group (in French).

https://doi.org/10.1371/journal.pone.0154832.s002

(PDF)

S2 Poster. Poster B used in the intervention group (in French).

https://doi.org/10.1371/journal.pone.0154832.s003

(PDF)

S1 Protocol. Study protocol (English version).

https://doi.org/10.1371/journal.pone.0154832.s004

(DOCX)

S1 Statistical Analysis Plan.

https://doi.org/10.1371/journal.pone.0154832.s005

(PDF)

S1 Table. Baseline characteristics of patients included during the intervention period by lost to follow-up status.

https://doi.org/10.1371/journal.pone.0154832.s006

(DOC)

S2 Table. Subgroups analysis of clinical events.

https://doi.org/10.1371/journal.pone.0154832.s007

(DOC)

S3 Table. Sensitivity analyses of the main outcome.

https://doi.org/10.1371/journal.pone.0154832.s008

(DOC)

S4 Table. In-hospital outcomes.

https://doi.org/10.1371/journal.pone.0154832.s009

(DOC)

S5 Table. Thromboprophylaxis practices adequacy according to whether a preventive anticoagulant treatment was recommended.

https://doi.org/10.1371/journal.pone.0154832.s010

(DOC)

S6 Table. Thromboprophylaxis practices adequacy and delay of prescription.

https://doi.org/10.1371/journal.pone.0154832.s011

(DOC)

S7 Table. Prescription of prophylactic anticoagulant treatment.

https://doi.org/10.1371/journal.pone.0154832.s012

(DOC)

S8 Table. Three-months outcomes in the 2 hospitals for which a computerized reminder was implemented.

https://doi.org/10.1371/journal.pone.0154832.s013

(DOC)

S9 Table. Thromboprophylaxis practices adequacy in the 2 hospitals for which a computerized reminder was implemented.

https://doi.org/10.1371/journal.pone.0154832.s014

(DOC)

S10 Table. Intracluster correlation coefficients.

https://doi.org/10.1371/journal.pone.0154832.s015

(DOC)

Acknowledgments

The study was supported by the French investigation network on VTE (INNOVTE). We would like to thank all the members of the PREVENU study group and the research assistants, secretaries, physicians, and residents of the emergency departments for their invaluable collaboration. We also wish to thank Babak Khoshnood for his statistical support and Guillaume de Charette for his technical support.

PREVENU study group

The members of the PREVENU study group were as follows:

Steering Committee—Pierre-Marie Roy (Study Chair and PREVENU study group leader; CHU Angers, France, [email protected]), Dominique Mottier (CHU, Brest, France), Jeannot Schmidt (CHU, Clermont Ferrand, France), Catherine Le Gall (CH, Argenteuil, France), Pierre Hausfater (APHP, Pitiè-Salpétrière, Paris, France), Grégoire Le Gal (CHU, Brest, France), and Pierre Durieux (APHP, Hôpital Européen Georges Pompidou, Paris, France);

Critical Events Adjudication Committee—Guy Meyer (APHP, Hôpital Européen Georges Pompidou, Paris, France), Philippe Girard (Institut Montsouris, Paris, France), Françoise Parent (APHP, Hôpital de Bicêtre, le Kremlin Bicêtre, France), Olivier Sanchez (APHP, Hôpital Européen Georges Pompidou, Paris, France), and Andrea Penaloza (Cliniques Universitaires Saint Luc, Bruxelles, Belgium);

Statistical analysis—Antoine Rachas (APHP, Hôpital de Bicêtre, le Kremlin Bicêtre, France) and Gilles Chatellier (Hôpital Européen Georges Pompidou, Paris, France);

Data management: Jean-Marie Chrétien (CHU, Angers, France);

Investigators (by city)—Albert Trinh-Duc and Albane Buan (CH, Agen, France); Aurore Armand, Thibault Schotte and Sophie Dambrine (CHU, Angers, France); Catherine Le Gall and Pascal Peudepièce (CH, Argenteuil, France); Mohamed Hachelaf and Thibault Desmesttre (CHU, Besançon, France); Alain-Eric Dubart (CH, Béthune, France); Frédéric Adnet and Sabine Guinemer (APHP, Bobigny); David Elkharrat (CHU Boulogne); Morgan Jaffrelot and Erwann L’Her (CHU, Brest, France); Guillem Bouilleau and Louis Soulat (CH, Chateauroux, France); Jeannot Schmidt and Farès Moustafa (CHU, Clermont Ferrand, France); Patrick Miroux, Géraldine Layani, and Yan Normand (CH, Compiègne, France); Bertrand Renaud, Aline Santin, and Alfred N’Gako (APHP, Hôpital Henri Mondor, Créteil, France); Didier Honnart and Cindy Tissier (CHU, Dijon, France); Françoise Carpentier and Claire Ara Somohano (CHU, Grenoble, France), Eric Wiel and Pierre Williatte (CHU, Lilles, France), Marc-Louis Alazia and Patrick Gerbeaux (APHM Marseille, France); François Braun (CHG, Metz, France); Dominique Elkouri and Nicolas Goffinet (CHU, Nantes, France); Lionel Nace and Anne Mercuri (CHU, Nancy, France); Jean-Emmanuel De La Coussaye and Jean-Yves Lefrant (CH, Nîmes, France); Enrique Casalino (APHP, Hôpital Bichat, Paris, France); Yann-Erick Claessens and Jean-Christophe Allo (APHP, Hôpital Cochin, Paris, France); Florence Dumas and Emmanuel Bloch-Laine (APHP, Hôtel-Dieu, Paris, France); Pierre Hausfater and Samuel Delerme (APHP, La Pitié Salpétrière, Paris, France); Fatima Rayeh-Pelardy, Michel Scepi, and Jean-Yves Lardeur (CHU, Poitiers, France); Jacques Bouget (CHU, Rennes, France); Frédéric Staiskowsky (CHG, Saint Pierre, La Réunion);

Research operation team: Béatrice Gable (Chair), Aurore Hamard, Gaëlle Durand, Marie-Lyne Pinot, Nabahat Ibrir, and Sybille Quentin-Georget (CHU Angers, France)

Author Contributions

Conceived and designed the experiments: PMR PD GM GLG JS DM. Performed the experiments: PMR JS PH DEK DH CL. Analyzed the data: PMR AR GM GLG JS PH PD JMC DM. Contributed reagents/materials/analysis tools: PMR AR GM PD JMC. Wrote the paper: PMR AR PD GM GLG JMC.

Citation: Roy P-M, Rachas A, Meyer G, Le Gal G, Durieux P, El Kouri D, et al. (2016) Multifaceted Intervention to Prevent Venous Thromboembolism in Patients Hospitalized for Acute Medical Illness: A Multicenter Cluster-Randomized Trial. PLoS ONE 11(5): e0154832. https://doi.org/10.1371/journal.pone.0154832

References

1. Alikhan R, Peters F, Wilmott R, Cohen AT. Fatal pulmonary embolism in hospitalised patients: a necropsy review. J Clin Pathol. 2004;57(12):1254–7. pmid:15563663; PubMed Central PMCID: PMC1770519.

2. Cohen AT, Agnelli G, Anderson FA, Arcelus JI, Bergqvist D, Brecht JG, et al. Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality. Thromb Haemost. 2007;98(4):756–64. Epub 2007/10/17. 07100756 [pii]. pmid:17938798.

3. Oger E, Bressollette L, Nonent M, Lacut K, Guias B, Couturaud F, et al. High prevalence of asymptomatic deep vein thrombosis on admission in a medical unit among elderly patients. The TADEUS Project. Thromb Haemost. 2002;88:592–7. pmid:12362229

4. Dentali F, Douketis JD, Gianni M, Lim W, Crowther MA. Meta-analysis: anticoagulant prophylaxis to prevent symptomatic venous thromboembolism in hospitalized medical patients. Ann Intern Med. 2007;146(4):278–88. pmid:17310052.

5. Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6 Suppl):381S–453S. Epub 2008/07/24. 133/6_suppl/381S [pii] pmid:18574271.

6. Tapson VF, Decousus H, Pini M, Chong BH, Froehlich JB, Monreal M, et al. Venous thromboembolism prophylaxis in acutely ill hospitalized medical patients: findings from the International Medical Prevention Registry on Venous Thromboembolism. Chest. 2007;132(3):936–45. Epub 2007/06/19. pmid:17573514.

7. Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, et al. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. Lancet. 2008;371(9610):387–94. Epub 2008/02/05. S0140-6736(08)60202-0 [pii] pmid:18242412.

8. Kahn SR, Morrison DR, Cohen JM, Emed J, Tagalakis V, Roussin A, et al. Interventions for implementation of thromboprophylaxis in hospitalized medical and surgical patients at risk for venous thromboembolism. The Cochrane database of systematic reviews. 2013;7:Cd008201. Epub 2013/07/19. pmid:23861035.

9. AFSSAPS. Prévention et traitement de la maladie thrombo-embolique veineuse en médecine. Agence Française de sécurité sanitaire des produits de santé: Recommandations de bonne pratique [Internet]. 2009 2014-07-15:[1–13 pp.]. Available: http://ansm.sante.fr/var/ansm_site/storage/original/application/ae4209ebc36d7164d4b7c876ddeaabab.pdf.

10. Guyatt GH, Norris SL, Schulman S, Hirsh J, Eckman MH, Akl EA, et al. Methodology for the development of antithrombotic therapy and prevention of thrombosis guidelines: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):53S–70S. Epub 2012/02/15. pmid:22315256; PubMed Central PMCID: PMC3278053.

11. Torbicki A, Perrier A, Konstantinides S, Agnelli G, Galie N, Pruszczyk P, et al. Guidelines on the diagnosis and management of acute pulmonary embolism: the Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). Eur Heart J. 2008;29(18):2276–315. Epub 2008/09/02. ehn310 [pii] pmid:18757870.

12. Schulman S, Kearon C, Subcommittee on Control of Anticoagulation of the S, Standardization Committee of the International Society on T, Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost. 2005;3(4):692–4. Epub 2005/04/22. pmid:15842354.

13. Kerry SM, Bland JM. Sample size in cluster randomisation. Bmj. 1998;316(7130):549. pmid:9501723.

14. Donner A. Some Aspects of the Design and Analysis of Cluster Randomization Trials. Applied Statistics. 1998;47:95–113.

15. Murray DM, Varnell SP, Blitstein JL. Design and analysis of group-randomized trials: a review of recent methodological developments. American journal of public health. 2004;94(3):423–32. pmid:14998806; PubMed Central PMCID: PMC1448268.

16. Grimshaw J, Eccles M, Thomas R, MacLennan G, Ramsay C, Fraser C, et al. Toward evidence-based quality improvement. Evidence (and its limitations) of the effectiveness of guideline dissemination and implementation strategies 1966–1998. J Gen Intern Med. 2006;21 Suppl 2:S14–20. pmid:16637955; PubMed Central PMCID: PMCPMC2557130.

17. Piazza G, Rosenbaum EJ, Pendergast W, Jacobson JO, Pendleton RC, McLaren GD, et al. Physician alerts to prevent symptomatic venous thromboembolism in hospitalized patients. Circulation. 2009;119(16):2196–201. Epub 2009/04/15. pmid:19364975; PubMed Central PMCID: PMCPmc2901546.

18. Garcia DA, Highfill J, Finnerty K, Varoz E, McConkey S, Hutchinson K, et al. A prospective, controlled trial of a pharmacy-driven alert system to increase thromboprophylaxis rates in medical inpatients. Blood Coagulation & Fibrinolysis. 2009;20(7):541–5

19. Labarere J, Bosson JL, Sevestre MA, Sellier E, Richaud C, Legagneux A. Intervention targeted at nurses to improve venous thromboprophylaxis. Int J Qual Health Care. 2007;19(5):301–8. pmid:17726037.

20. Kucher N, Koo S, Quiroz R, Cooper JM, Paterno MD, Soukonnikov B, et al. Electronic alerts to prevent venous thromboembolism among hospitalized patients. N Engl J Med. 2005;352(10):969–77. Epub 2005/03/11. pmid:15758007.

21. Dexter PR, Perkins S, Overhage JM, Maharry K, Kohler RB, McDonald CJ. A computerized reminder system to increase the use of preventive care for hospitalized patients. N Engl J Med. 2001;345(13):965–70. Epub 2001/09/29. pmid:11575289.

22. Anderson FA Jr., Wheeler HB, Goldberg RJ, Hosmer DW, Forcier A, Patwardhan NA. Changing clinical practice. Prospective study of the impact of continuing medical education and quality assurance programs on use of prophylaxis for venous thromboembolism. Arch Intern Med. 1994;154(6):669–77. 8129501. pmid:8129501

23. Fontaine A, MahÉ I, Bergmann JF, Fiessinger JN, Dhote R, Cohen P, et al. Effectiveness of written guidelines on the appropriateness of thromboprophylaxis prescriptions for medical patients: a prospective randomized study. Journal of Internal Medicine. 2006;260(4):369–76. pmid:16961674

24. Nendaz MR, Chopard P, Lovis C, Kucher N, Asmis LM, DÖRffler J, et al. Adequacy of venous thromboprophylaxis in acutely ill medical patients (IMPART): multisite comparison of different clinical decision support systems. Journal of Thrombosis and Haemostasis. 2010;8(6):1230–4. pmid:20175871

25. Khanna R, Vittinghoff E, Maselli J, Auerbach A. Unintended consequences of a standard admission order set on venous thromboembolism prophylaxis and patient outcomes. J Gen Intern Med. 2012;27(3):318–24. pmid:21948203; PubMed Central PMCID: PMC3286563.

26. Maynard G, Jenkins IH, Merli GJ. Venous thromboembolism prevention guidelines for medical inpatients: mind the (implementation) gap. J Hosp Med. 2013;8(10):582–8. pmid:23983041.

27. Overhage JM, Tierney WM, McDonald CJ. Computer reminders to implement preventive care guidelines for hospitalized patients. Arch Intern Med. 1996;156(14):1551–6. Epub 1996/07/22. pmid:8687263.

28. Kahn SR, Lim W, Dunn AS, Cushman M, Dentali F, Akl EA, et al. Prevention of VTE in nonsurgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):e195S–226S. Epub 2012/02/15. pmid:22315261; PubMed Central PMCID: PMC3278052.