Long term time trends in use of medications associated with risk of developing osteoporosis: Nationwide data for Denmark from 1999 to 2016

Highlights

• In Denmark, the use of most medications associated with an increased risk of osteoporosis has increased from 1999 to 2016

• Total sale of proton pump inhibitors (PPI) per 1,000 capita per day increased from 11.9 to 66.8 DDD in 2016

• During 2016, 9.0% of men and 11.4% of women collected ≥1 prescription of PPI (3.0% and 3.4%, respectively, in 1999)

• Total sale of selective serotonin reuptake inhibitors doubled from 1999 to 2016, driven by primary sector prescriptions

• Total sale of venlafaxine increased by 613% from 1999 to 2016, primarily driven by the increase in the female subgroup

Abstract

Purpose

To evaluate the development in the use of medications associated with an increased risk of developing osteoporosis over the time period from 1999 to 2016.

Methods

We extracted data on total sale, sales rate and usage rate for the medications of interest from www.medstat.dk, which is an online, open-source database reporting the monthly sale of both over-the-counter and prescription-based medications in Denmark. The dataset covers both the primary and secondary health sectors.

Results

Most medications exhibited an increasing use from 1999 to 2016, though some had stable (e.g. glucocorticoids) or declining use. Notably, some medications showed widespread and increasing use, including proton pump inhibitors (PPI), selective serotonine reuptake inhibitors (SSRI) and venlafaxine. For PPI, sales rates increased by 461% from 1999 to 2016, with 9% of men and 11.4% of women filling at least one prescription in 2016. The use of SSRI and venlafaxine increased by 114% and 613%, respectively. This was more pronounced in women and for SSRI also in the elderly (80+ years). The sale of aromatase inhibitors was moderate (1–10 DDD per 1000 capita per day) in 2016, yet grew by 2400% from 1999, almost exclusively in women aged 80 years or older.

Conclusion

We found a trend of increasing use from 1999 to 2016 of most medications with a potential for causing osteoporosis, often most pronounced in fracture risk groups (postmenopausal women and/or in the elderly). This may play a clinically relevant role in both current and future causality of osteoporosis.

Introduction

Osteoporosis is characterized by low bone mass and microarchitectural changes. Clinically it manifests itself mainly with femoral, pelvic, humeral, antebrachial, and vertebral low energy fractures. However, fractures at any other skeletal sites may also be sustained and patients with osteoporosis are also at increased risk of high energy fractures [1].

In recent years, hip fracture incidence has shown a declining trend across Europe, initially in women and later in men [2]. However, in Sweden and Denmark some of the decrease in hip fracture rates appears to be a temporary respite due to a particularly low rate of hip fractures in persons born in the 1930’s with high overall incidence rates in subsequent generations [3].

In addition to certain chemicals and different systemic diseases - many of which exhibit significant time trends – some medications can be a potential cause of secondary osteoporosis and of injurious falls and fractures [4]. For example, one study demonstrated that 30–50% of patients treated with glucocorticoids developed bone fractures on treatment [5]. Likewise, a meta-analysis recently demonstrated a significantly increased risk of hip and vertebral fractures with proton pump inhibitor therapy, with some studies suggesting a dose-dependent relationship [6,7]. Similarly, selective serotonin reuptake inhibitor (SSRI) therapy is associated with accelerated bone loss and increased fracture risk [[8], [9], [10]].

In this study, we aimed to assess temporal trends in the use of medications associated with an increased risk of osteoporosis and osteoporotic fractures for the time period from 1999 to 2016. Specifically, we wanted to identify risk medications with an inclining use in the age groups subject to the largest osteoporotic fracture burden. Hence, it would be a concern to find widespread and fast increasing use of risk medications in the oldest age groups while rare and fast disappearing drug exposures in the youngest patients would be of low clinical concern.