Multimorbidity, the presence of 2 or more chronic diseases, poses a major challenge to public health though its contribution to post-fracture mortality remains unclear. We sought to identify the pattern of multimorbidity and its relationship to post-fracture excess mortality.
This nationwide population-based cohort study involved 307,870 adults in Denmark born on or before 1/1/1950 with an incident low-trauma fracture between 2001 and 2014 who were followed through 2016. Fracture and 32 predefined chronic diseases recorded within 5 years prior to the index fracture were identified using ICD-10 codes from the Danish National Hospital Discharge Register. Because most diseases are correlated, we conducted to identify clusters of comorbidities, and relative survival analysis to quantify excess mortality attributable to the combination of multimorbidity and specific fracture sites.
There were 95,372 men (age at fracture: 72±11 years) and 212,498 women (75±11) with incident fractures followed by 41,017 and 81,727 deaths, respectively. Almost half of fracture patients had multimorbidity. Comorbidities at fracture grouped into a low-multimorbidity (60.5% in men, 66.5% in women), cardiovascular (23.7%, 23.5%), diabetic (5.6%, 5.0%), malignant (5.1%, 5.0%) and mixed hepatic/inflammatory clusters (5.1%, men only). These clusters distinguished individuals with advanced, complex, or late-stage disease from those having earlier stage disease. Multimorbidity and proximal or lower leg fractures were associated with significantly increased mortality risk, with the highest excess mortality found in hip fracture patients in the malignant cluster. Importantly, the combination of multimorbidity and fracture compounded mortality, conferring much greater risk than either alone (Figure).
Concomitant illnesses were common and clustered into distinct multimorbidity clusters that imparted excess mortality post-fracture. The compound contribution of multimorbidity to post-fracture excess mortality highlights the need for more comprehensive approaches in these high-risk patients. The analytical approach we have applied to fracture could be also used to examine other sentinel health events.