Abstract
Purpose
The prevention of a subsequent, contralateral hip fracture is targeted as an avoidable event in the elderly. Fall prevention and bone strengthening measures have met with limited success and the urgency of their effect is undetermined. Our objective was to evaluate the time to second hip fracture (the time between a first and a subsequent, contralateral fracture) in elderly patients, using a population-based administrative health data set.
Method
The 58,286 records of persons older than 60 yrs and hospitalized for a hip fracture between 1985 and 2005 were obtained from a Provincial administrative health database. We excluded non-traumatic cases and identified the care episodes related to a subsequent hip fracture for each patient using unique identifiers. We used a 5 year “wash-out period” to avoid counting a second fracture as a first one.
We calculated the proportion of first and second fractures and sex distribution over time (fiscal years) and quantified the time between first and second fracture, while correlating it to age, sex and fracture type.
Results
Overall, 3,866 patients sustained a second hip fracture between 1990 and 2005; 3,119 (81%) were women, in contrast to 73% for primary fractures (chi-square =137.8, df=1, p<0.001).
In 33% cases, the type of a subsequent fracture (transcervical vs pertrochanteric) was different from the first.
The median time from first fracture was 3 years, 90% occurred by 9yrs. The age at the first fracture most influenced the time to second fracture. The median time (90th percentile in parentheses) between fractures decreased as patients got older and was 5 (13), 4 (10), 3 (7), 2 (5) years for patients who were correspondingly 60–69, 70–79, 80–89 and 90+ years old at first fracture.
Conclusion
Among survivors of an initial hip fracture, the occurrence of a second hip fracture appears to affect a greater proportion of women than primary fractures. Our results identify the time frame which preventative interventions should target when aiming at reducing second hip fractures, that target being increasingly small (from 5 to 2 years) as patients age. This information identifies a time frame researchers must target as they seek new fracture prevention methods. In the shorter term however, these data could influence health administrators and policy makers as they decide to support one hip fracture prevention method over another.
