Nonadherence in Outpatient Thromboprophylaxis after Major Orthopedic Surgery: A Systematic Review

After Major Orthopedic Surgery: A Systematic Review

Thomas Wilke; Sabrina Müller


Expert Rev Pharmacoeconomics Outcomes Res. 2010;10(6):691-700. 

In This Article

Discussion & Conclusion

At the moment, there exists a plethora of scientific investigations into medication-based adherence by patients. The majority of these studies concern themselves with medication for the most common chronic indications. In such studies, the NA rates of patients, which are commonly derived, are between 30 and 50%, depending on the medication concerned, the sample and the measurement method used.[31,101]

Acute and short-term medication is far less frequently the subject of NA research than is chronic medication. There are two possible reasons for this: first, at least for the most-common chronic diseases, patient-based NA only gains in relevance a few months after the start of the initial treatment.[32] On this basis, one could assume the NA of patients in acute or short-term treatment to be of little importance. Second, adherence analysis based on administrative data is based on differences between the time of the handing in of several prescriptions and the profile of the length of time for which the medicine needs to be taken.[30] As only one or a very few sequential prescriptions are needed in acute or short-term cases, the NA of such patients cannot be measured by administrative data. Consequently, it can only be estimated by the other measures listed in Figure 2. This is exactly the case for studies that concern themselves with outpatient thromboprophylaxis following hip- or knee-replacement surgery.

The clinical importance of extended thromboprophylaxis after major orthopedic surgery is well documented. It is important that medical practitioners consider this when issuing prescriptions ('guideline adherence').[33,34] Nevertheless, the fact that doctors prescribe extended thromboprophylaxis is not a guarantee for the implementation of this prophylaxis. Possible NA by the patients represents a danger to the goals of the treatment.

The main findings of our systematic review can be summarized as follows: ambulant thrombosis prophylaxis in the form of self-administrated prophylaxis will increase in importance; based on the six studies summarized in this review, approximately 13–37% of patients are nonadherent during the ambulant parenteral thrombosis prophylaxis; based on the results of one large German survey, NA patients do not use between 37 and 43% of their outpatient prophylaxis injections; two available studies show that NA appears to be a phenomenon influenced by subjective patient-based factors as well as objective care-provision factors.

Current international trends in care-provision structures, in particular, the trend towards shorter stationary treatment times and the associated lengthening of the ambulant care phase during the first 35 days following surgery, are likely to lead to an increase in the numbers of NA patients and the length of the NA periods. It is, ceteris paribus, not to be dismissed that thrombosis rates will increase following major orthopedic surgery. To prevent this trend from developing any further, it is necessary to influence patients' adherence levels by means of adherence interventions. Detailed knowledge about subjective/objective causes of NA is a necessary prerequisit for the development of effective adherence intervention strategies. The research results reviewed here represent the first attempts to provide the required knowledge basis, and make clear the need for more detailed information concerning patients,[35] possible simplifications in the application of medication and particular support for the patients in the early post-acute treatment phase.

We acknowledge limitations to this article. First, we only identified six papers dealing with NA in medication-based outpatient thromboprophylaxis after major orthopedic surgery. That is certainly a very limited amount of evidence. Second, most of the studies do not measure patients' NA with multi-indicator-validated measures. Mostly, simple self-report questions were used. Third, comparisons of NA quotas demonstrated in the studies are difficult to make because the thromboprophylaxis environment (duration of inpatient prophylaxis, duration of outpatient prophylaxis or patient support during outpatient prophylaxis) differs between the studies or is unknown (especially length of prophylaxis and length of observation period). Fourth, the studies included in this article relate to the use of parenteral LMWH or fondaparinux. This is especially important because the choice of an anticoagulant depends not only on the NA of patients but also on other issues surrounding the thromboprophylaxis, for example, clinical effectiveness/safety, different care models, prophylaxis costs and monitoring costs. Fifth, only two studies identified reasons explaining the NA, so the evidence for the development of effective adherence interventions is limited. Nevertheless, despite the methodological limitations, this article presents the most comprehensive systematic overview of what is currently known regarding the NA of patients in outpatient thrombosis prophylaxis after orthopedic surgery.

Three particular areas are likely to determine the future direction of postoperative thrombosis prophylaxis adherence research. First, there needs to be more clarity regarding the clinical relevance of discontinuing thrombosis prophylaxis only a few days after the acute treatment has been completed. Second, it will be important to document the factors causing NA in ambulant thrombosis prophylaxis in considerably more detail than presently exists and, in this regard, to evaluate aspects such as different patient groups, health literacy effects[35–37] and other psychological factors. Third, oral anticoagulants (Vitamin K antagonists, but especially dabigatran/rivaraoxaban) can be expected to play a much more important role in future thrombosis prophylaxis after major orthopedic surgery; owing to their easier oral application, they have the potential for increasing medication adherence.[17,19] How this potential is translated into better real-life thromboprophylaxis adherence and whether additional adherence programs are needed are matters for future research.

The clinical development of anticoagulants over the last decade has greatly improved the safety and effectiveness of thrombosis prophylaxis medications. In current clinical studies, the displayed clinically manifested thrombosis rates lie between approximately 2.2 and 2.7%, accompanied by serious bleeding in approximately 2.0% of cases.[1–5,38] In real-life settings, these medications only perform to maximum effect when a high rate of patient adherence exists. However, the results described in NA research and the current trends in care provision show that reaching a high level of adherence requires explicit effort.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.