Bone. 2026 Jul 14:118014. doi: 10.1016/j.bone.2026.118014. Online ahead of print.
ABSTRACT
BACKGROUND: Fractures in older adults arise from complex interactions between bone fragility, falls, and comorbid conditions. This study aimed to map the comorbidity environment of osteoporosis (M80/M81) by sex and age, and assess whether osteoporosis is systematically coded at the time of fracture-related hospitalization across fracture types.
METHODS: We conducted a nationwide retrospective analysis of Austrian inpatient data collected for insurance claims between 2003 and 2014, including 1,705,684 hospitalized patients aged 50-79 years. Comorbidity network analysis was used to characterize sex- and age-specific diagnoses co-occurring with osteoporosis (ICD-10: osteoporosis with pathological fracture M80 and osteoporosis without pathological fracture M81). In paralell, fracture-centered networks were constructed for major fracture types (S22, S32, S42, S52, S62, S72) to assess whether osteoporosis was systematically coded during fracture related hospitalizations. Sex-specific odds ratios (ORs) were calculated to compare comorbidity associations between males and females across three age groups (50-59, 60-69, 70-79 years).
RESULTS: Across all age groups, females exhibited substantially broader M81-associated comorbidity networks than males, with the number of linked diagnoses increasing from 142 to 245 to 330 in females and from 76 to 137 to 164 in males between ages 50-59, 60-69, and 70-79 years, respectively. Fracture-centered analyses showed that across age groups, fractures were embedded within multimorbidity networks rather than forming isolated diagnostic links with osteoporosis. Hip fractures (S72) showed the strongest co-occurrence with osteoporosis diagnoses, particularly in females aged ≥60 years, whereas upper limb fractures - forearm (S52), wrist and hand (S62), and humerus (S42) - showed no osteoporosis diagnoses among significant co-diagnoses in any age group despite these fractures being considered early fragility fractures. In patients aged 50-69 years, male fracture networks showed stronger co-occurrence with trauma-related and alcohol-related diagnoses, while in older age groups networks in both sexes increasingly included neurodegenerative, cardiovascular, respiratory, and renal diseases. In sex-stratified analyses, the co-occurrence between osteoporosis without pathological fracture (M81) and osteoporosis with pathological fracture (M80) was consistently stronger in males than in females across all ages (e.g. OR 63.74 vs 18.53 at 50-59 years).
CONCLUSIONS: Osteoporosis is not systematically coded across all fracture types in routine hospital data. Upper limb and thoracic fractures were rarely recorded alongside osteoporosis diagnoses despite being recognized as potential fragility fractures. These patterns suggest under-recognition of osteoporosis at the time of fracture, particularly in younger patients and in males.
PMID:42448012 | DOI:10.1016/j.bone.2026.118014