Fractures cause significant mortality and morbidity and a decrease in quality of life. The risk of a subsequent fracture is higher after a recent fracture, when compared to persons without a recent fracture. Interestingly, this risk is highest in the first period after the fracture, compared to a few years later. This phenomenon is known as the imminent fracture risk and has gained much attention over the past decennium. Understanding risk factors for subsequent fractures is key for adequate prevention. The majority of all fractures are caused by a fall, and we hypothesize the imminent fracture risk might actually be caused by an imminent fall risk. The aim of this study is to assess an imminent fall and fracture risk in the CPRD population of 50 years or older, with and without a recent incident fracture. For both groups we will explore the cumulative incidence of general practitioner reported falls, with death as a competing risk. To adequately interpret these results we will first explore secular trends of falls for participants in both groups.
An imminent risk of subsequent fracture after an incident fracture is well established. Most fractures are caused by falls. However, more insight is needed about fall incidence and timing in patients with a recent fracture. We hypothesise that imminent fracture risk could be explained by imminent fall risk in the first years after an incident fracture. We will conduct a descriptive, retrospective study using routinely collected primary care data from the CPRD GOLD database. The study cohort will include all CPRD participants of 50 years with an incident fracture (index date) between 1990-2015, who will be matched one-to-one based on year of birth, sex and practice to a control subject without a fracture. Follow up for outcome of interest will run until the end of 2020. First, secular trends of falls during the inclusion period will be assessed and compared for both groups. One-year fall incidence rates will be calculated by counting all reported first falls divided by the total time at risk, for each calendar year. Incidence rate ratio’s will be compared between groups. Second, cumulative incidences of falls and fractures will be assessed for both groups by using the cumulative incidence competing risk (CICR) method and death will be considered a competing risk. Outcome of interest will be fall or fracture after the index date at multiple time intervals. Results will be stratified based on age, sex and index fracture.
Our outcome measures are:
• Secular trends of falls in CPRD participants of 50 years or older after an incident fracture compared to participants without fracture.
• The cumulative incidence of falls and the cumulative incidence of fractures in CPRD participants of 50 years or older after an incident fracture compared to participants without fracture.
Patrick Souverein - Chief Investigator - Utrecht University
Patrick Souverein - Corresponding Applicant - Utrecht University
Caroline Wyers - Collaborator - VieCuri Medical Center
Hanna Willems - Collaborator - University of Amsterdam
Johanna Driessen - Collaborator - Utrecht University
Joop van den Bergh - Collaborator - Maastricht University
Marle Gemmeke - Collaborator - Utrecht Institute for Pharmaceutical Sciences
Merle Schene - Collaborator - Maastricht University Medical Centre