ECTS Abstracts (2015) 1 OC6.1

Secular change in fracture incidence is not associated with better post-fracture outcomes: a time-trend comparison between two birth cohorts

Dana Bliuc1,2, Tuan Nguyen1,2, John Eisman1,2 & Jacqueline Center1,2


1Garvan Institute of Medical Research, Sydney, New South Wales, Australia; 2UNSW Australia, Sydney, New South Wales, Australia; 3St Vincent’s Hospital, Sydney, New South Wales, Australia; 4University of Notre Dame, Sydney, New South Wales, Australia.


During the last decade, hip fracture incidence declined and life expectancy improved. However, it is unclear whether the outcomes following osteoporotic fracture have also changed. The aim of this study was to compare re-fracture risk and excess mortality following osteoporotic fracture between two birth cohorts and over two time intervals 1989–1999 and 2000–2010. Study participants comprised women and men 60+ participating into DOES1 (born before 1930) and DOES2 (born after 1930). All fractures excluding head, fingers and toes were recorded between 1989 and 2010. Age-standardised fracture incidence and mortality rates were calculated in two time intervals: 1989–1999 (for DOES1) and 2000–2010 (for DOES2). The difference in excess mortality between the two cohorts was assessed using standardised mortality ratios (SMR) calculated for each study cohort using time-specific population mortality rates. The prevalence of osteoporosis declined and the level of treatment increased significantly in DOES 2 compared to DOES 1. Fracture incidence declined by ~10% in both genders, however, not significantly. Interestingly, re-fracture risk was similar for DOES1 and DOES2 (women age-adjusted RR 2.0 (95% CI, 1.6–2.5) in DOES1 and 1.9 (95% CI, 1.7–2.3) in DOES2 and men, 3.5 (95% CI, 2.7–4.8) in DOES1 and 3.4 (95% CI, 2.7–4.5) in DOES2). Crude mortality rates decreased during study follow-up. However, after taking into account the difference in general population life expectancy during the two study periods, the excess mortality post-fracture was similar (women, SMR 2.1 (95% CI, 1.7–2.6) in DOES1 and 1.7 (95% CI, 1.2–2.4) in DOES2, and men, 1.9 (95% CI, 1.5–2.5) in DOES1 and 1.9 (95% CI, 1.3–2.7) in DOES2). Thus despite a reduction in the prevalence of osteoporosis and improvement in treatment uptake over the last two decades, re-fracture risk and fracture-associated excess mortality was similar. The reasons for this deserve urgent exploration.

Disclosure: The authors declared no competing interests. This work was supported by the National Health Medical Research Council Australia (NHMRC project ID; DB 1073430, JRC 1008219, TVN, JRC and JAE 1070187).

Article tools

My recent searches

No recent searches.