The coronavirus disease 2019 (COVID-19) pandemic has been associated with important
changes in the use of anticoagulants, including (1) the recommendation of low-molecular-weight
heparin (LMWH) or fondaparinux to prevent thrombotic complications associated with
COVID [1–3], and the increased use of direct oral anticoagulants (DOACs) in atrial
fibrillation to prevent exposure to healthcare facilities for blood monitoring required
with warfarin.
We quantified the change in global sales of anticoagulant agents after the COVID-19
pandemic declaration in March 2020.
MIDAS® data for August 2014–August 2020 were obtained from IQVIA [4], which contains
retail and hospital drug purchase data for 39 countries. We extracted records for
anticoagulants, including injectable anticoagulants (heparins and LMWH), vitamin K
antagonists, and DOACs.
For each month and country, we calculated the number of units sold per 1000 population,
and we conducted interrupted time-series analyses to examine changes in the trend
of sales after the pandemic declaration in March 2020. Regression models included
a continuous variable for month, an indicator variable for the period after March
2020, the interaction between them, and an indicator variable for developing countries.
To quantify country-level variation, we calculated the proportion change in the number
of units sold in March–August 2020 versus March–August 2019. Due to the instability
of estimates for smaller countries, we constrained these analyses of variation in
units for March–August 2020 versus March–August 2019 to countries with a population
of at least 20 million (n = 17 countries).
Before 2020, sales of anticoagulants increased by an annual average of 6%, driven
by an annual average increase in sales of DOACs of 26%. Sales of vitamin K antagonists
decreased by an annual average of 8%, and sales of injectable anticoagulants remained
constant (Fig. 1). After the pandemic declaration in March 2020, there was a significant
increase in sales of all anticoagulant products (p-value for level change = 0.006)
and DOACs (p-value for level change = 0.019). Global sales of injectable products
remained constant, while sales of vitamin K antagonists increased after the pandemic
declaration, but the change was not significant (p = 0.133).
Fig. 1
Trends in anticoagulant sales, August 2014–August 2020
On average, countries purchased 11% more anticoagulant units in March–August 2020
than in March–August 2019 (Fig. 2). Russia and Mexico had the largest increases in
sales of anticoagulant products, with increases of 39% and 40%, respectively. The
UK was the only country with a decrease in sales of anticoagulant products (− 2%).
Country-level variation in changes in anticoagulant sales was particularly large for
injectable products. While global sales of injectable anticoagulants were lower in
March–August 2020 than in March–August 2019, five countries increased their demand,
most remarkably Australia (55%), Russia (42%), and Mexico (34%). Russia and Mexico
also presented the greatest increases in purchases of DOACs (73% and 43%, respectively),
while the US had the largest decrease in purchases of injectable anticoagulants (− 22%).
Fig. 2
Country-level variation in change in units March–August 2020 versus March–August 2019.
The figure represents the proportion change between the number of units sold per 1000
in March–August 2020 and March–August 2019 at the country level. The lines represent
median and interquartile range. Analyses were performed for the subset of countries
with a population of at least 20 million (n = 17), including China, India, US, Brazil,
Russia, Japan, Mexico, the Philippines, Germany, Turkey, the UK, France, Italy, Spain,
Poland, Australia, and Canada. Countries excluded, with a population of < 20 million
included Austria, Belarus, Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Finland,
Hungary, Kazakhstan, Lithuania, The Netherlands, New Zealand, Norway, Portugal, Romania,
Slovakia, Sweden, Switzerland and Tunisia. DOACs direct oral anticoagulants
Our paper is subject to four main limitations. First, we did not include fondaparinux
in analyses because of missing data for some countries. Second, because we did not
have access to drug utilization data, it is possible that changes in purchases of
anticoagulants did not translate into changes in their clinical use. Additionally,
it is unclear what indications anticoagulants were used for. Third, we used March
2020 as the start of the pandemic period, even though the pandemic peak did not happen
simultaneously for all countries. Fourth, our analyses included countries with widely
different healthcare systems, where changes in purchases of anticoagulants may not
similarly translate into changes in their use.
Nevertheless, our study is an important contribution to the literature on the impact
of COVID on the demand for medications. Sales of DOACs presented a significant increase
in March 2020, which could reflect patients’ tendency to hoard medications in the
early months of the COVID-19 pandemic for chronic indications of anticoagulants or
switches from warfarin to DOACs to avoid exposing patients to healthcare facilities
for blood testing. Although injectable anticoagulants have been recommended to prevent
thrombotic complications in hospitalized COVID-19 patients since May 2020, global
sales remained constant through August 2020. Only a few countries increased demand
for these agents, most remarkably Australia, Russia, and Mexico. Interestingly, the
US presented the strongest decrease in purchases of injectable anticoagulants. This
could reflect the marked decrease in non-COVID-19 hospital admissions observed in
the US following the pandemic outbreak [5], or the long periods with postponed elective
procedures, for which injectable anticoagulants are often recommended. Additionally,
country-level variation in purchases of injectable anticoagulants could be due to
variability in national and institutional stockpile reserves and in the implementation
of policies in response to COVID. For example, in January 2020, Russia released special
rules for the import, pricing, and distribution of pharmaceutical products [6], which
could partially explain the high increase in anticoagulant purchases observed. Finally,
international variability in purchases of anticoagulants could represent differential
speed in the uptake of new scientific evidence and guideline recommendations on COVID-19
treatments. Future research using patient-level data should examine to what extent
the decrease in purchases of injectable anticoagulants observed in the US was due
to the decrease in demand for inpatient care for non-COVID-19 conditions versus a
delay in the use of oral anticoagulants to prevent thrombotic complications associated
with COVID-19.