Abstract
Objectives:
Dabigatran etexilate is a new oral direct thrombin inhibitor for prophylaxis of venous thromboembolism (VTE) in patients who have elective surgery for total hip replacement (THR) or total knee replacement (TKR). Among the advantages of dabigatran etexilate over subcutaneous prophylaxis with Low Molecular Weight Heparin (LMWH) are reduced resource uses for (i) teaching patients to self-inject; (ii) home-care visits for subcutaneous administration; and (iii) absence of heparin-induced thrombocytopenia (HIT). Based on the demonstrated non-inferiority, the aim of this study was to conduct a cost-minimization analysis of oral dabigatran etexilate vs subcutaneous low-molecular weight heparin (LMWH) and fondaparinux from the Dutch healthcare perspective.
Methods:
A retrospective cohort study was conducted to measure resource use associated with subcutaneous prophylaxis. Results of this study were used in the model to elucidate specific advantages of dabigatran etexilate, next to reduced needs for self-inject teaching and lack of Heparin-Induced Thrombocytopenia. Drug and other resource utilization data were combined with local unit costs. Probabilistic sensitivity analysis was performed to account for uncertainty around relevant parameters included.
Results:
Home-care visits for subcutaneous administration problems were needed in 9.9% (95% CI = 6.4–13.4) and 9.6% (95% CI = 5.8–13.4) of THR and TKR patients, respectively. Based on costs for 1000 patients treated with dabigatran etexilate vs LMWHs, per patient cost-savings with dabigatran etexilate were estimated at €30.68 (95% CI = 2.01–65.52) and €23.19 (95% CI = 0.69–48.48) for THR and TKR, respectively. The probability that dabigatran etexilate would be cost-saving was estimated at 98.3% and 97.9% for THR and TKR, respectively. These cost-savings were even higher when including fondaparinux in the analysis, with per patient cost-savings of €69.87 (43.42–106.10) and €18.33 (1.63–41.26) for THR and TKR, respectively. Separate calculations for dabigatran etexilate vs nadroparin and dalteparin in THR resulted in probabilities of achieving cost-savings with dabigatran etexilate of 36.2% and 100%, respectively. For TKR these probabilities were estimated at 54.3% and 100%, respectively.
Conclusions:
Thromboprophylaxis with dabigatran etexilate is cost-saving in patients undergoing THR and TKR from the Dutch healthcare perspective, compared to subcutaneous LMWHs.
Introduction
Patients undergoing major orthopedic surgery are at elevated risk of venous thromboembolism (VTE), including both deep vein thrombosis (DVT) and pulmonary embolism (PE). Due to a period of immobilization, the risk of VTE persists for up to 3 months after orthopedic surgery. The incidence of thromboembolic complications among patients not receiving thromboprophylaxis after orthopedic surgery was previously estimated at 40–60% (DVT), 2–5% (non-fatal PE) and 0.1% (fatal PE)Citation1,Citation2. Primary prevention with thromboprophylactic treatment around orthopedic surgery reduces the risk of non-symptomatic and symptomatic VTE considerablyCitation1,Citation3,Citation4. International guidelines recommend thromboprophylaxis for patients undergoing total hip replacement (THR) or total knee replacement (TKR) with low molecular weight heparins (LMWHs), fondaparinux, or vitamin K antagonists (VKAs)Citation1,Citation5,Citation6. From national and international surveys it seems that most orthopedic patients receive thromboprophylactic therapy with LMWHsCitation7–10. In particular, data from an observational study showed that, of adults who have undergone THR and TKR surgery, 65–80% receive LMWHs in the NetherlandsCitation9,Citation10. It is recommended in Dutch guidelines to administer LMWHs for a period of 4–6 weeks after THR and TKR surgeryCitation6. The therapeutic complexity related to the subcutaneous route of administration of anticoagulants may lead to reluctance of patients to self-administer in the home-setting following discharge from hospital or even it is impossible for them to do so. This may lead to inaccurate dosing, low therapy compliance, and, therefore, lower effectiveness of subcutaneous thromboprophylaxis to reduce the risk of VTE in daily practiceCitation11–15.
Dabigatran etexilate (Pradaxa®, Boehringer Ingelheim, Ingelheim, Germany) is a novel oral direct thrombin inhibitor for prophylaxis of VTE for adult patients who have undergone elective THR or TKR surgery. Phase III clinical studies—RE-NOVATE and RE-MODEL—have demonstrated non-inferiority for dabigatran etexilate vs enoxaparin, a low-molecular weight heparin (LMWH) in patients who have undergone THR or TKR surgeryCitation16–18. Oral dabigatran etexilate was found to be as effective as the subcutaneous LMWH enoxaparin in reducing the risk of VTE and VTE-related mortalityCitation18. Also, safety of dabigatran etexilate and LMWHs was comparable with a similar bleeding profile for both these types of agentsCitation18.
Oral anticoagulants are recognized as more patient-friendly in daily practice. In particular, the new oral thromboprophylactic agents like dabigatran etexilate, rivaroxaban, and apixapan have potential advantages over parenteral prophylaxis with LMWHs or fondaparinux, given the absence of possible problems with subcutaneous administration for the former. However, drug costs for, for example, dabigatran exceed those of the LMWHs. Yet, downstream savings might be expected related to the advantages of the new thromboprophylactic agents. In the Netherlands, advantages of these oral direct thrombin inhibitors for VTE prophylaxis include (but may not be limited to) reduced resource use for (i) teaching patients to self-inject, (ii) home-care visits for parenteral administration by nurses due to self-administration problems, and (iii) absence of heparin-induced thrombocytopenia (HIT)Citation19,Citation20.
Yet, only scarce information is available on the occurrence of problems with self-administration of subcutaneous antithrombotics at home. In particular, on the involvement of homecare visits by a nurse to solve these problems, no hard data exist. Several studies indicate that such problems occur in 7.5–44% of the patients administering LMWHs or fondaparinux at homeCitation9,Citation10,Citation12–15.
This study was designed to quantify the potential administration problems with subcutaneous thromboprophylaxis by measuring the prevalence of such problems among patients that have undergone THR or TKR. Furthermore, based the demonstrated non-inferiority and data on the prevalence of administration problems, we conducted a cost-minimization analysis of oral dabigatran etexilate vs parenteral LMWHs and fondaparinux from the Dutch healthcare perspective.
Methods
Patient interview on administration problems
The occurrence of administration problems with subcutaneous administration of thromboprophylaxis was evaluated among 687 Dutch patients who had undergone THR or TKR surgery by a retrospective telephone interview. The research was conducted in two regional Dutch hospitals (De Tjongerschans, Heerenveen; Antonius Ziekenhuis, Sneek) and one teaching hospital (Medisch Centrum Leeuwarden) in 2008. Approval was received from the Medical Ethical Review Committee ‘RTPO’. Patients who had undergone THR or TKR surgery in 2008 (February–December in Heerenveen, May and June in Sneek, and September–December in Leeuwarden) received an informative letter in which the research was introduced. After oral informed consent, patients were interviewed using a structured questionnaire. Next to patient characteristics, several questions were asked to measure healthcare resource use (e.g. type of surgery, hospital stay, instruction for administration), potential problems related to self-administration of subcutaneous administration in the home-setting, and patients’ preferences for subcutaneous and/or oral thromboprophylaxis.
Cost-minimization analysis
A cost-minimization analysis was conducted to compare oral dabigatran etexilate with subcutaneous thromboprophylaxis formulations. Dutch drug-utilization patterns and healthcare resource data were combined with local unit costs to calculate the healthcare cost of thromboprophylaxis with dabigatran etexilate, LMWH, and fondaparinux formulations. In particular, potential advantages of dabigatran etexilate over LMWHs and fondaparinux could be quantified relating to: (i) resource use incurred giving patients and/or relatives instructions for self-administration of LMWHs or fondaparinux; (ii) homecare for the administration of LMWHs or fondaparinux following discharge from hospital; and (iii) heparin-induced thrombocytopenia (HIT)Citation19,Citation20.
The base-case cost-minimization analysis reflects the comparative costs and potential savings for oral vs subcutaneous thromboprophylaxis using a mix of LMWHs reflecting actual use. Within the healthcare perspective used, Dutch cost data for drugs and healthcare resource use were derived from a previous Dutch observational studyCitation9,Citation10, and the patient interview designed specifically for this analysis (see above). All analyses were conducted separately for both THR and TKR surgery populations. All costs were expressed in 2010 values, without discounting, as all costs are incurred within a maximum period of 10 weeks within the same financial yearCitation21,Citation22.
Scenario analysis, deterministic and probabilistic sensitivity analysis
Various scenario analyses were conducted based on the different possible comparator treatment options or combinations of these comparator therapies. In the absence of data on various aspects, in probabilistic sensitivity analysis distributions were generally chosen to follow the symmetric triangular distribution, with plausible minima and maxima (see below and Appendix). The probability of administration problems for both the THR and TKR analyses were assumed to follow a Beta distributionCitation23. For the occurrence of HIT, a specific study was available, enabling the specification of a Beta distribution with observed mean and estimated standard errorCitation20. Next to varying all relevant parameters at once (multivariate), a univariate probabilistic approach was elaborated to identify the importance of each individual parameter, separately. For this purpose, subsequently each relevant parameter was allowed to vary over its specified distribution, with all other parameters kept at their respective expected values (2000 iterations per parameter). All analyses were done in Excel (MicroSoft), using @Risk add-in (Palisade, Athica, NY, USA) to perform probabilistic analyses.
Dutch population data
Dutch treatment guidelines recommend extended prophylaxis after discharge up to 6 and 4 weeks in THR and TKR, respectivelyCitation6. Based on an average hospital length of stay of 8 days for THR as well as for TKRCitation9,Citation10, total treatment periods were assumed to be 50 and 36 days for THR and TKR, respectivelyCitation24,Citation25. In probabilistic sensitivity analysis, symmetric triangular distributions were used (THR: minimum at 2 weeks, maximum at 10 weeks; TKR: minimum at 2 week, maximum at 6 weeks)Citation15.
Data on inpatient days and drug-utilization from July 2005–July 2006 were available for further analysis (source: PHARMO Utrecht, a specific report on the data is available on request). To adequately reflect the patient population who had undergone elective THR or TKR, all admissions for hip and knee fractures were excludedCitation9,Citation10. From these data, the percentage distribution of specific LMWHs and fondaparinux used in Dutch current practice was estimated (enoxaparin is not routinely used in the Netherlands). We assumed that the clinical trial results for dabigatran etexilate vs enoxaparin are also valid vs other types of LMWHs (and fondaparinux)Citation26. In particular, we analyzed two options for the comparator: the LMWH mix and the LMWH/fondaparinux mix from the daily practice observations in the PHARMO-data (for exact percentage distributions see below). The latter LMWH/fondaparinux comparator for dabigatran etexilate may be considered to most closely reflect the Dutch real-life situation, and the LMWH mix may be considered the preferred comparator in most guidelines.
Costing
Prior to discharge from hospital, THR and TKR patients are instructed on self-administration of subcutaneous VTE prophylaxis to ensure that patients use the correct injection technique, to prevent needle-stick injuries, and to early identify patients who are unable, unwilling, or highly reluctant to inject themselves. Costs for such an instruction were based on an assumed 30 minutes of nursing time in the hospital, according to instruction protocols (in probabilistic sensitivity analysis: symmetric triangular distribution with minimum 20 and maximum 40 minutes). Given the hourly costsCitation22, the costs for half an hour of instruction were estimated at €15.52 per patient.
shows the unit costs for the different types of home care after hospital discharge, using estimated time investments and guideline unit pricesCitation22. We considered two options for those patients with parenteral self-administration problems at home. In particular: (i) we conservatively assumed that these patients would require standard domiciliary care anyhow for other health reasons, and we inserted only the extra costs for a visit by a trained nurse (for subcutaneous administration) over a visit not requiring a trained nurse as savings for dabigatran etexilate (€9.95 per day; difference between standard price for domicillary care (€6.24) and domiciliary care by trained nurse (€16.19)); and (ii) we assumed a maximal approach considering that these patients would have no reason for follow-up visits for reasons other than the administration of thromboprophylaxis and the full €16.19 per day (standard price for domiciliary car by trained nurse) was included as savings for patients receiving dabigatran etexilate. In the threshold analysis, both the conservative and maximal approach were elaborated. In the base case, it was assumed that the two options occurred 50–50 in practice. In probabilistic sensitivity analysis a symmetric triangular distribution was used for this distribution (minimum: 25%; maximum: 75%).
Table 1. Unit costs for different types of homecareCitation22, based on normative durations of visits and inclusive travel costs, difference shown for whether a (trained) nurse is involved or not (homekeeping help only).
In the Netherlands, dalteparin and nadroparin are the LMWHs that are mostly used following elective orthopedic surgery (). Additionally, 28.2% and 12.5% of patients undergoing THR and TKR use fondaparinux. Based on the Z-index (‘Taxe’ reflecting the official Dutch pricelist), daily costs for dabigatran etexilate were assumed at €4.41 (both 150 mg and 220 mg), €3.45 for dalteparin (5000 units), €2.30 for nadroparin (3800 units), and €7.66 for fondaparinux (2.5 mg)Citation27. These daily costs and mentioned percentages translate into cost estimates for 50 days use after THR and 36 days use after TKR, as shown in Citation9,Citation10,Citation27.
Table 2. Percentage distribution for LMWH used in the NetherlandsCitation9,Citation10, used in the LMWH mix comparator with and without fondaparinux.
Table 3. Per-person costs for prophylaxis following THR (50 days) or TKR (36 days)Citation6,Citation9,Citation10.
Costs related to HIT were based on an 0.8% incidence of HIT among LMWH users in the Netherlands, as reported by ten Berg et al.Citation28, which is in line with data found in international studiesCitation20,Citation29,Citation30. In probabilistic sensitivity analysis, a Beta distribution was assumed with parameters α and β at 14 and 1640, respectivelyCitation30. Costs for HIT were assumed to amount to €3110 in 2010 and were not assumed for fondaparinux (i.e. the LMWH/fondaparinux mix)Citation31.
Results
Administration problems
Six hundred and eighty-seven patients (response rate 87.4%) were interviewed. Patient characteristics are presented in . A total of 511 (74.4%) of these patients used subcutaneous thromboprophylaxis at home, with the remainder primarily using vitamin K antagonists. Of the interviewed patients, 48.8% reported administration problems varying from local pain, bruises, and itches. Almost 60% of all THR and TKR patients would prefer oral over subcutaneous thromboprophylaxis, if efficacy and safety would be comparable. Home-care visits for subcutaneous administration problems were required by 9.9% (95% CI = 6.4–13.4) and 9.6% (95% CI = 5.8–13.4) of THR and TKR patients, respectivelyCitation30. As mentioned above, homecare visits for subcutaneous administration problems can be sub-divided in patients that solely require homecare for administration and patients that require homecare for administration next to other help ().
Table 4. Patient characteristics, homecare needs, and average duration of hospitalization.
Table 5. Proportion of patients requiring homecare for parenteral administration problems.
Cost-minimization analysis
reports the costs for separate cost components and different comparator scenarios in the base case. For both THR and TKR, costs for treatment with dabigatran etexilate exceed costs of treatment in other comparator treatment scenarios. Relevant cost offsets are estimated for all three other components: instruction, homecare, and HIT. In general, total costs for the different treatment scenarios were lower for dabigatran etexilate, which means that future dabigatran etexilate use could result in cost-savings. Dabigatran etexilate does not appear to offer cost-savings vs nadroparin, however the additional costs are low.
Table 6. Costs for 1000 patients treated with respective medications for THR and TKR (in €).
Sensitivity analysis
shows the per-patient cost-savings for dabigatran etexilate vs the other therapeutic strategies including probabilistic results on 95% confidence intervals and the probabilities of achieving cost-savings with dabigatran etexilate. Per patient cost-savings with dabigatran etexilate vs the observed mix for LMWHs are estimated at €30.68 (95% CI = 2.01–65.52) and €23.19 (95% CI = 0.69–48.48) for THR and TKR patients, respectively. Furthermore, the probability that dabigatran etexilate would be cost-saving is 98.3% and 97.9% for THR and TKR, respectively. If costs of HIT are excluded from the analysis, these cost-savings are lower and turn in low additional costs for dabigatran etexilate for TKR patients (€3.14; 95% CI = − 20.92–19.17). Cost-savings are higher when including fondaparinux alongside the observed mix of LMWHs, with per patient cost-savings varying between €18.33 (95% CI = 1.63–41.26) for TKR patients and €69.87 (95% CI = 43.42–106.10) for THR patients, if using dabigatran etexilate. The probabilities of achieving cost-savings with dabigatran etexilate vs the LMWH/fondaparinux mix strategy are close to or even 100%.
Table 7. Per-patient cost-savings of dabigatran etexilate vs LMWHs, LMWH mix, and LMWH/fondaparinux mix.
Separate calculations for dabigatran etexilate vs nadroparin and dalteparin in THR result in probabilities of achieving cost-savings with dabigatran etexilate of 36.2% and 100%, respectively. These figures are essentially similar for TKR. Exclusion of costs for HIT results in overall slightly less favorable findings for dabigatran etexilate compared to the different LMWHs (dalteparin and nadroparin). Yet, the probability of achieving cost-savings reduces dramatically with dabigatan etexilate vs nadroparin, approaching 0%.
Results presented in and show limited variation around the mean estimate for dabigatran etexilate vs the LMWH mix. In particular, the percentage requiring homecare, occurrence of HIT, and costs related to these aspects resulted in the largest confidence intervals for the per-patient cost-savings. However, all variations in individual parameter assumptions did not result in incremental costs.
Figure 1. Tornado-diagram for THR showing the mean and 95% confidence limits following from univariate probabilistic sensitivity analysis.
Figure 2. Tornado-diagram for TKR showing the mean and 95% confidence limits following from univariate probabilistic sensitivity analysis.
Discussion
The current analysis highlights the potential for cost-savings with dabigatran etexilate as VTE prophylaxis following orthopedic surgery vs subcutaneous thromboprophylxis with either LMWH or fondaparinux. The higher acquisiton cost of dabigatran etexilate is generally offset by the costs for homecare, HIT, and related healthcare resource use associated with LMWHs. Probabilistic sensitivity analysis resulted in high probabilities of achieving cost-savings with dabigatran etexilate. Results were estimated to be most sensitive to the percentage of patients experiencing administration problems, occurrence of HIT, and to the costs related to these aspects. Results were, for example, less sensitive to exact durations of treatment and training.
In the presence of non-inferiority studies for oral dabigatran etexilate, we designed a cost-minimization framework to analyze the health economics of dabigatran etexilate as compared to its subcutaneous comparators. We note that our cost-minimization framework was limited and conservative in the sense that some potential further benefits of the oral formulation over the subcutaneous administration were not included. For example, needle-stick injuries and sharps disposal will be absent for the oral formulation, whereas it does occur in real-life settings with subcutaneous administrationsCitation14. Additionally, the hospital nursing staff will need less time for administering oral dabigatran etexilate than for parenteral formulations. Besides, platelet monitoring for HIT and potential differences in the occurrence of VTE assuming poor adherence to thromboprophylaxis following discharge from the hospital could be other potential cost drivers for the subcutaneous LMWH or oral dabigatran as well. Finally, we note that some other economic evaluations have inserted superior point estimates of dabigatran etexilate over LMWHs into their models, as suggested in the clinical trialsCitation32. The presented approach has been successful for achieving reimbursement in the Netherlands and has been acknowledged in a quality assessment of these type of studies in the context of evidence-based medicine and reimbursement in the NetherlandsCitation33,Citation34.
We generally aimed at following a conservative approach in our analysis, for example, excluding the risk for HIT in the full LMWH/fondaparinux mix. Yet, in the absence of information on discounts we applied list prices for LMWHs and fondaparinux. In practice, discount may be provided to hospitals and could worsen our estimated cost-effectiveness. Notably, in addition to the advantages modeled, an oral administration route can be expected to result in further inpatient (as well as outpatient) cost-savings due to reduced nursing-time for administration of oral medication, and also the absence of risk for needle-stick injuries and sharps disposal. Furthermore, recent evidence might support limited adherence next to limited convenience with self-administration of LMWHsCitation12,Citation13,Citation15,Citation35,Citation36. Observational studies among THR and TKR patients have shown that ∼35–40% of patients have problems with self-administration of subcutaneous prophylaxis, which ultimately leads to a situation where 7.5–37.5% of patients are completely or partially non-compliant, thereby reducing the effectiveness of VTE prophylaxis and posing risks for DVTsCitation12,Citation13,Citation15. The UK’s National Institute for Health and Clinical Excellence (NICE) estimated the risk of developing DVT following orthopedic surgery in the absence of (effective) thromboprophylaxis to be as high as 44% for patients having THR and 27% for patients having TKR, highlighting the importance of compliance to anti-thrombotic therapy in generalCitation36. Also, we note that our findings from the patient interviews that almost 60% of all THR and TKR patients would prefer oral over subcutaneous thromboprophylaxis, if efficacy and safety would be comparable, might still under-estimate the real preference. In particular, coping mechanisms with the current administration form might play a role here. Finally, we limited our scope on those costs related to HITs actually occurring and preventable with dabigatran, whereas also just suspected HITs already generate costs. Additionally, averted HIT in general not only provides cost savings, but also reduces (though relatively rare) devastating complicationsCitation37. For the specific Dutch setting, our cost-minimization approach was relevant and appropriate. Whether the same applies to other countries depends on various factors inclusive of the relative costs of the different drugs involved and the specific levels of costs for home in individual countries. Transferability to other countries of our results may therefore be limited.
Conclusion
We conclude that—based on our conservatively designed cost-minimization analysis—thromboprophylaxis with dabigatran etexilate is likely to be cost-saving in patients undergoing THR and TKR compared to its subcutaneous comparators if considered from the Dutch healthcare perspective.
Transparency
Declaration of funding
This research was supported by a grant from Boehringer Ingelheim (Alkmaar, Netherlands). All stages of the research were discussed with employees of Boehringer Ingelheim. One of the co-authors is from Boehringer Ingelheim.
Declaration of financial/other relationships
MJP, MvH, and JRBJB have received grants, honoraria, and travel stipends from various pharmaceutical companies, inclusive those developing and marketing the new oral anti-coagulants. BSK is an employee of Boehringer Ingelheim. The peer reviewers on this manuscript have disclosed that they have no relevant financial relationships.
Acknowledgments
The authors acknowledge Eric van Roon (Medisch Centrum Leeuwarden, Leeuwarden, Netherlands) and Michiel Duyvendak (Antonius Ziekenhuis, Sneek, Netherlands) for their role in the patient recruitment process. Furthermore, Cornelis Boersma (University of Groningen, Netherlands) is acknowledged for commenting on a draft version of the paper.
MJP and MvH performed this study in the employment of HECTA BV Ltd. HECTA ceased its activities January 1st 2011, and all running projects and its follow-ups were integrated into the University of Groningen.
References
- Geerts WH, Bergqvist D, Pineo GF, et al. American College of Chest Physicians. Prevention of venous thromboembolism. American college of chest physicians evidence-based clinical practice guidelines. 8th edn. Chest 2008;133(6 Suppl):381S-453S
- Douketis JD, Eikelboom JW, Quinlan DJ, et al. Short-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of prospective studies investigating symptomatic outcomes. Arch Intern Med 2002;162:1465-71
- Eikelboom JW, Quinlan DJ, Douketis JD. Extended-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of the randomised trials. Lancet 2001;358:9-15
- Ettema HB, Hoppener MR, Büller HR, et al. Tromboprofylaxe in de orthopedische chirurgie: inzichten en onzekerheden. Ned Tijdschr Geneeskd 2003;147:1842-7 [in Dutch]
- Hill J, Treasure T; on behalf of the National Clinical Guideline Center for Acute and Chronic Conditions. Reducing the risk of venous thromboembolism in patients admitted to hospital: summary of NICE guidance. BMJ 2010;340:259-260
- Biiller HR; Richtlijn diagnostiek, preventie en behandeling van veneuze tromboembolie en secundaire preventie van arteriële trombose. Kwaliteitsinstituut voor de Gezondheidszorg CBO/Nederlands Huisartsen Genootschap. Utrecht 2008. www.cbo.nl/Downloads/492/rl_stol_09.pdf. Accessed August 5, 2010
- Struijk-Mulder MC, Ettema HB, Verheyen CC, et al. Comparing consensus guidelines on thromboprophylaxis in orthopedic surgery. J Thromb Haemostas 2010;8:678-83
- National Joint Registry for England and Wales 5th Annual Report. 2008; http://www.njrcentre.org.uk/njrcentre/default.aspx. Accessed November 4, 2010
- Pharmo Institute. Prevention of thromboembolic disease following knee or hip replacement: ‘burden of illness’. Utrecht: Pharmo Institute, 2008. http://www.pharmo.nl/frames.php?htmlpage=navigatie/reports.htm. Accessed June 8, 2010
- Kappelhoff BS. Primaire preventie van veneuze trombo-embolieën na totale heup- of knievervangende operaties: ‘burden of illness’. Pharm Weekbl Wet Platform 2009;4:128-34 [in Dutch]
- Anderson FA, Jr. Hirsh J, White K, et al. Temporal trends in prevention of venous thromboembolism following primary total hip or knee arthroplasty 1996-2001: findings from the Hip and Knee registry. Chest 2003;124(Suppl):349S-56S
- Colwell CW, Jr. Pulido P, Hardwick ME, et al. Patient compliance with outpatient prophylaxis: an observational study. Orthopedics 2005;28:143-7
- Spahn G. Compliance with self-administration of heparin injections in outpatients. Eur J Trauma 2002;28:104-9
- Muntz JE, Hultburg R. Safety syringes can reduce the risk of needlestick injuries in venous thromboembolism prophylaxis. J Surg Orthop Adv 2004;13:15-19
- Ettema HB, Hoppener MR, Henny CP, et al. Compliance of Dutch orthopedic departments with national guidelines of thromboprophylaxis. A survey of Dutch orthopedic thromboprophylaxis. Acta Orthop 2005;76:99-103
- Eriksson BI, Dahl OE, Rosencher N, et al. Dabigatran etexilate versus enoxaparin for prevention of venous thromboembolism after total hip replacement: a randomised double-blind, non-inferiority trial. Lancet 2007;370:949-56
- Eriksson BI, Dalh OE, Rosencher N, et al. Oral dabigatran etexilate versus subcutaneous enoxaparin for the prevention of venous tromboembolism after total knee replacement: the RE-MODEL randomized trial. J Tromb Haemost 2007;5:2178-85
- Friedman RJ, Dahl OE, Rosencher N, et al. Dabigatran versus enoxaparin for prevention of venous thromboembolism after hip or knee arthoplasty: a pooled analysis of three trials. Thromb Res 2010;126:175-82
- National Collaborating Centre for Acute Care (NCC-AC). Venous thromboembolism: the prevention of venous thromboembolism (deep vein thrombosis and pulmonary embolism) in patients undergoing orthopaedic surgery and other high-risk surgical procedures. London: National Collaborating Centre for Acute Care, 2007. http://www.rcseng.ac.uk/publications/docs/venous_thrombo.html. Accessed November 4, 2010
- Arepally GM, Ortel TL. Heparin-induced thrombocytopenia. New Engl J Med 2006;355:809-17
- Health Care Insurance Board. Dutch pharmacoeconomic guidelines 2004 [in Dutch]. Amstelveen: Health Care Insurrance Board, 2003. http://www.cvz.nl/resources/FARM_richtlijnen_farmaco-economisch_onderzoek_2006_tcm28-19118.pdf. Accessed August 5, 2010
- Oostenbrink JB, Bouwmans CAM, Koopmanschap MA, et al. Guideline for costing research, methods and standardized prices for economic evaluations in health care [in Dutch]. Diemen: Health Care Insurance Board, 2004. http://www.cvz.nl/resources/feo_handleiding-kostenonderzoek2004_tcm28-17013.pdf. Accessed December 21, 2010
- Briggs A, Claxton K, Sculpher M. Decision modelling for health economic evaluation. Oxford: Oxford University Press, 2008
- http://www.fk.cvz.nl/. Accessed August 5, 2010
- Nederlandse Orthopaedische Vereniging. Diagnostiek en behandeling van heup- en knieartrose, richtlijn [in Dutch]. Nijmegen: Nederlandse Orthopaedische Vereniging, 2007
- National Collaborating Centre for Acute Care (NCC-AC). Venous thromboembolism: the prevention of venous thromboembolism (deep vein thrombosis and pulmonary embolism) in patients undergoing orthopaedic surgery and other high-risk surgical procedures. 2007
- http://www.medicijnkosten.nl/. Accessed December 20, 2010
- Rice L. Heparin-Induced Thrombocytopenia. Myths and misconceptions. Arch Intern Med 2004;164:1961-4
- Oliveira GBF, Crespo EM, Becker RC, et al. Incidence and prognostic significance of thrombocytopenia in patients treated with prolonged heparin therapy. Arch Intern Med 2008;168:94-102
- Ten Berg M, Huisman A, van den Bemt P, et al. Case-finding of possible heparin-induced thrombocytopenia in an academic teaching hospital by linking laboratory and medication data. Pharmacoepidemiol Drug Saf 2007;16:S51
- Institute for Medical Technology Assessment. Cost analysis of treatment of adverse events related to treatment of patients with CML. Rotterdam: Institute for Medical Technology Assessment, Erasmus MC, 2007. Report 07.99 (data on file)
- Wolowacz SE, Hess N, Brennan VK, et al. Cost-effectiveness of venous thromboembolism prophylaxis in total hip and knee replacement surgery. Current Med Res Opinion 2008;24:2993-3006
- Hoomans T, van der Roer N, Severens JL, et al. Kosteneffectiviteit van Nieuwe Geneesmiddelen; van belang bij geneesmiddelenvergoeding, maar voor verbetering vatbaa (in Dutch). Nederlands Tijdschrift voor Geneeskunde 2010;154:A958
- www.cvz.nl. Accessed December 21, 2010
- Boersma C, Broer ACM, Kappelhoff BS, et al. Problemen bij subcutane toediening van laagmoleculairgewichtheparines na totale heup- en knievervangende operaties [in Dutch]. Pharm Weekbl 2010;4:20-4
- Dranitsaris G, Jelincic V, Choe Y. Meta regression analysis to indirectly compare dalteparin to enoxaparin for the prevention of venous thromboembolic events following total hip replacement. Thromb J 2001;9:3
- Smythe MA, Koerber JM, Mattson JC. Identifying the incidence of recognized heparin induced thrombocytopenia in a large tertiary care teaching hospital. Chest 2007;131:1644-9
Appendix: Parameter distributions
The distributional assumptions for all parameters included are listed below. In the absence of any information on distributional properties, triangular distributions were specified with plausible minima and maxima.