Hip fracture in men-survival and subsequent fractures: a cohort study with 22-year follow-up.

Research output: Contribution to journalArticlepeer-review


To evaluate the influence of age on survival and risk of subsequent fracture in men with hip fracture, applying a residual lifetime perspective.

Retrospective cohort study with 22-year follow-up.

Skåne University Hospital, Malmö, Sweden.

Men aged 60 and older (N=226) with an index hip fracture during 1984/85.

Twenty-two-year survival (mortality) and risk of new fractures evaluated in 5-year age bands and age groups (<75, 75-84, ≥85).

Mean age was 78±9. Mortality at 22 years was 98%. Survivors were all younger than 75 at inclusion. Mortality was dependent on age at all time points (18%, 38%, 69% at 1 year, increasing to 71%, 93%, 100% by 10 years in <75, 75-84, ≥85, respectively). Median survival was 5.4 years, 2.0 years, and 3 months, respectively, in these age groups, and 33%, 27%, and 13% of each age-group sustained subsequent fractures, generally within 5 years. Overall 10-year fracture risk was 29% (95% confidence interval (CI)=19-38%), increasing to 44% (95% CI=30-58%) when adjusted for mortality. Residual lifetime risk of new fracture was 33% (95% CI=23-43%), and mortality-adjusted risk was 63% (95% CI=45-81%). Participants younger than 75 at index hip fracture were at greatest risk of new fracture (hazard ratio=2.7, 95% CI=1.1-6.4, P=.03).

Almost one-third of men with hip fracture have subsequent fractures during their remaining lifetime. Time at risk is highly dependent on age. Most new fractures occur in relatively younger men and within 5 years, whereas most aged 75 and older die before experiencing a new fracture.
Original languageEnglish
Pages (from-to)806-813
JournalJournal of the American Geriatrics Society
Issue number5
Publication statusPublished - 2011

Subject classification (UKÄ)

  • Geriatrics


Dive into the research topics of 'Hip fracture in men-survival and subsequent fractures: a cohort study with 22-year follow-up.'. Together they form a unique fingerprint.

Cite this