Hip fractures in Sweden and Denmark 1987-2010 - period and cohort effects

Bo Abrahamsen; Cyrus Cooper; Jonas Bjork; Bjorn Rosengren

doi:10.1530/boneabs.3.OC2.3

OC2.3

Prev Next

Section Contents Cite

Bone Abstracts (2014) 3 OC2.3 | DOI: 10.1530/boneabs.3.OC2.3

ECTS2014 Oral Communications Osteoporosis epidemiology (6 abstracts)

Hip fractures in Sweden and Denmark 1987–2010 – period and cohort effects

Bo Abrahamsen ¹ , Cyrus Cooper ² , Jonas Bjork ³ & Björn Rosengren ⁴

Views

Author affiliations

¹Research Centre for Ageing and Osteoporosis, Department of Medicine M, Glostrup Hospital, Copenhagen, Denmark; ²MRC Lifecourse Epidemiology Unit, Southampton General Hospital, Southampton, UK; ³Unit for Medical Statistics and Epidemiology at R and D Centre Skane, Skane University Hospital Lund, Skane, Sweden; ⁴Clinical and Molecular Osteoporosis Research Unit, Department of Orthopedics and Clinical Sciences, Skane University Hospital Malmo, Malmo, Sweden.

Objective: The reasons for the recent decrease in hip fracture rates remain unclear. While current antiosteoporosis efforts are important also factors earlier in life seem essential and we examined age-period-cohort (APC) effects in hip fracture incidence in the Sweden (SE) and Denmark (DK).

Material and Methods: We studied the entire populations aged ≧50 years from 1987 to 2010 in SE and DK and ascertained acute hip fractures in nationwide discharge registers using diagnosis and surgical procedure codes for proximal femoral fracture. APC effects were evaluated country specific by log-likelihood estimates in Poisson regression models (with adjustment for sex and a scale parameter included to account for overdispersion). Results are presented as Incidence Rate Ratios (IRR) compared to the most recent 3-year period (2008–2010) or 6 year birth cohort (1953–60).

Results: During the examined years there were 399 596 hip fractures in SE and 207 304 in DK. The combined period and cohort effects were generally stronger in SE than DK and in women than men. IRR ranged from 1.05–1.30 in SE women, 1.04–1.18 in SE men, 1.21 –1.11 in DK women and 0.95–1.11 in DK men per period. The corresponding IRR per birth cohort ranged from 1.15–3.13 in SE women, 1.07–1.78 in SE men, 1.07–1.67 in DK women and 0.85–1.14 in DK men. Relative period effects increased with successive period for men and women in SE and described a convex curve for both men and women in Denmark with higher than expected risk in the periods in the middle of the examination years. Relative cohort effects were increasing with successive birth cohort for both genders in both countries but with markedly lower risks for DK women born 1925–44 and DK men born 1929–52 and a lower risk for SE women born 1933–44.

Conclusion: Cohort and period effects were different in SE and DK. This may in part be referred to differences in general health as evident in differences in life expectancy and to differences in exposure to war and famine as well as differences in use of osteoporosis drugs.