gms | German Medical Science

24th Annual Meeting of the German Drug Utilisation Research Group (GAA)

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie

30.11. - 01.12.2017, Erfurt

Prescription of oral anticoagulant drugs in the German federal state of Schleswig- Holstein; is there a regional pattern?

Meeting Abstract

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie e.V. (GAA). 24. Jahrestagung der Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie. Erfurt, 30.11.-01.12.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. Doc17gaa107

doi: 10.3205/17gaa107, urn:nbn:de:0183-17gaa1076

Published: December 5, 2017

© 2017 von Arnstedt et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Background: Since the launch of the direct oral anticoagulants (DOAC) rivaroxaban, dabigatran, apixaban and edoxaban in Germany there is an ongoing debate about the cost effectiveness of anticoagulant treatment. Looking at the GAMSI data (prescription data information service of the German statutory health insurance) 2010 to 2017 the increase in costs becomes visible. Whereas 2010 the anticoagulants accounted for 1,62 % of total drug expenses, in 2017 they will increase to 5,7%. Regarding Schleswig-Holstein this increase is even more distinct rising up to 6,8%. 2013 rivaroxaban (the most often prescribed DOAC) entered the list of the 20 top selling agents according to gross turnover in Germany. Starting at rank 10 in Schleswig Holstein it reached rank 3 in 2014 and since 2015 is the agent with the second most costs, with 2,68 % of the total expenditures on drugs expected in 2017. A similar trend has been observed by Schwabe et al [1].

As known the costs for the DOAC exceed those of the vitamin K antagonists (VKA) by about factor 15. This leads us to have a closer look at the ratio between VKA and DOAC prescription. We analysed the rise and fall of these two groups in Schleswig-Holstein and furthermore examined if the distribution varies in different regions of Schleswig-Holstein, giving hint to urban-rural differences or other patterns.

Materials and Methods: We analysed drug prescription data from Schleswig-Holstein from 2014 to 2016, grouping them by administrative districts, by applying the location of the prescribing physician. Only prescriptions assigned to an administrative district were included. The oral anticoagulant drugs included in this study are the ATC (Anatomical Therapeutic Chemical classification system) B01AA resembling VKA and B01AE07 and B01AF resembling DOAC. For comparison of the districts we used the relative indicators costs per 1,000 inhabitants and packages prescribed per 1,000 inhabitants.

Concerning the regional structure we applied public Data provided by GFK for 2015. Physicians were assigned by the postal code of their office.

Results: The costs of anticoagulant agents in Schleswig-Holstein from 2014 to 2016 nearly doubled. In detail, costs of the VKA decreased barely while expenses for DOAC more than doubled. Looking at the different DOAC the expenses for rivaroxaban increased only by factor 1,4, while substances, that gained market access more recently as apixaban and edoxaban displayed an immense growth. Overall number of packages prescribed raised only about 1,3 times. Looking at the subgroups, prescription of VKA in 2016 decreased to 88% of the numbers in 2014. In contrast DOAC prescription increased by 170%, again with a focus on apixaban.

Looking at the 15 administrative districts of Schleswig-Holstein, costs increased in all of them from 2014 to 2016. Factors of increase varied between 1,7 and 2. While twelve of the fifteen districts show expenses of about 15 000 € per 1000 inhabitants in 2016, there are three districts which are significantly higher: above 20 000 € per 1000 inhabitants. These are not as expected the urban districts, but two rural areas and one town, while the district with lowest costs is a town as well. This pattern is constant the three considered years.

Numbers of prescription are rising in all districts, as expected, with the mentioned three districts as expected at the top, while the order of the rest shifts over time.

These three top districts are quite different. Lübeck is a university town with about one physician per 350 inhabitants. Ostholstein and Dithmarschen are rural areas not neighbouring each other, with one physician per 522 and 726 inhabitants, respectively. The population density differs in the districts from 1010 inhabitants per m² in Lübeck to 143 in Ostholstein and 93 in Dithmarschen.

The percentage of DOAC in total anticoagulants rises in all areas and in 2016 levels from 55% to 70% in the different districts, with the ratios roughly staying in the pattern of 2014, but drawing nearer. Our three districts of interest are three of the top five in DOAC ratios.

Conclusion: The rise of the costs and doses of anticoagulant treatment in Schleswig -Holstein is due to the application of the DOAC especially of the newer drugs apixaban and edoxaban. While the expenses for the VKA and Dabigatran stay at a certain level and rivaroxaban increases only slightly in 2016 the newly launched drugs rise more distinctly.

The increase in DOAC ratios in anticoagulant treatment is likewise found in all administrative districts of Schleswig-Holstein. Nevertheless, the ratios differ widely. Differences in DOAC ratios cannot be explained by regional patterns like urbanity or e.g. spillover effects.

Even more evident is the increase in costs in all districts, but again the levels differ. The distinctly higher costs in some areas are due to higher numbers of anticoagulant prescription combined with a higher ratio of DOAC. However, the prescription numbers play the more crucial role, as the highest expenses per prescription are found in other districts. The problem of cost development in the field of anticoagulant treatment is primarily a problem of quantity. Similar results can be found in the literature [2].

The variation in the number of prescriptions per 1000 inhabitants can also not be explained by regional patterns. Neither population density, physician to population ratio nor localisation of the districts are crucial. Social factors might play a role in here. If available in the dataset, an implementation of patient-related information like age and insurance status would be eligible. One should also take in account that the evaluation to 1000 inhabitants might be problematic since we don´t know the percentage of people insured by public or private health insurances in each region and our data only covers statutory health insurance.


References

1.
Schwabe U, Paffrath D (Hrsg). Arzneiverordnungsreport 2016 S.23. Springer Verlag Berlin Heidelberg 2016
2.
Grant D, Schubert I. Barmer Arzneimittelreport 2016. Siegburg; 2016. S. 39.