Article Text

Download PDFPDF

Older women with depression were at increased risk for falls and fractures
  1. Roger Bullock, MBBS, MRCPsych
  1. Victoria Hospital Swindon, UK

Statistics from

OpenUrlCrossRefPubMedWeb of Science

Question Are older women with depression at increased risk of falls and fractures?


Cohort study with a mean follow up of 3.7 years.


Recruitment was from population based listings from 4 states in the US.


7414 ambulatory white women who were ≥67 years of age (mean age 74 y) who were enrolled in the Study of Osteoporotic Fractures, and during a second visit completed ≥10 of 15 items on the Geriatric Depression Scale (GDS) and had a measurement of bone density. Follow up for falls was available in 7317 women, for non-vertebral fractures in 7069 women, and for vertebral fractures in 6281 women.

Assessment of risk factors

A cutoff point of ≥6 symptoms on the GDS indicated depression (mild to moderate 6–10 or severe ≥11). Other potential risk factors measured were age; marital status; education; medical history; smoking; alcohol use; perceived health status; caffeine intake; use of calcium, oestrogen, thiazide, non-thiazide diuretics, steroids, and antidepressants; lifetime and current physical activity; body mass index; blood pressure; cognitive function; neuromuscular function; and functional status. Radiographs of fractures were viewed blinded to scores on the GDS.

Main outcome measures

Rates of falls and fractures.

Main results

Mean bone mineral density did not differ between women with and without depression. Women with depression had an increased risk of falls compared with women without depression (70% v 59%, p<0.001). The association was retained after adjustment for other risk factors (p=0.004) (table). Non-vertebral fractures were also increased in women with depression (28% v 21%, p<0.001) and had a 30% increased rate after adjustment for other risk factors (table). Rib fracture had the highest incidence in women with depression (adjusted hazard ratio 2.0, 95% CI 1.2 to 3.4, p=0.01) whereas wrist and humerus fracture were not associated with depression. Vertebral fractures occurred more frequently in women with depression than in those without depression (11% v 5%, p<0.001), and the association remained significant after adjustment for other risk factors (p<0.001) (table).

Falls and fractures in older women with depression v no depression at mean follow up of 3.7 years


Older women with depression were at increased risk of falls and fractures compared with older women without depression.


The study by Whooley et al lends proof to what we already suspect—depression increases morbidity. The data show high GDS scores (especially hopelessness, worthlessness, and dissatisfaction) correlating with increased risk of fracture in these patients.

Depression pervades all aspects of life, including recovery from injury. In combination, these 2 conditions consume substantial healthcare resources. The study is correlating a symptom scale with fractures, not diagnosing depression by clinical interview. The GDS, however, is a valid predictor, as shown by the typical prevalence found for depression (6.3%). The study highlights that depression is often a missed diagnosis in the elderly, and closer attention may benefit this population, reducing illness rates. One could predict similar findings with other illnesses. What is not clear is the part depression plays in falls and fractures. This study shows that depression increases the risk of fracture, but the antidepressant medication given has no effect on the outcome. Liu et al showed that both tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors were equally associated with an increase in falls, despite the broadly held view that TCAs were more strongly associated because of postural hypotension.1 Perhaps this finding is a disease effect. Another study suggests depression causes reduced bone density because of raised cortisol concentrations.2

If depression increases the risk of falls and reduces bone density, this is a high risk situation. Treatment that uses drugs only may add to that risk. We need new research to clarify these issues and perhaps point the way to non-drug interventions. Current research into corticotropin releasing factor blockers as a treatment for depression will add an interesting dimension to this question. Hopefully, new trial designs will reflect the need for psychiatric and medical covariables to be measured, giving us far more information than the current offerings that compare drugs effects using rating scales (“me too” graphs) often seen in depression today.



  • Source of funding: US Public Health Service.

  • For correspondence: Dr Mary A Whooley, General Internal Medicine Section, San Francisco Veterans Affairs Medical Center, 4150 Clement Street, 111A1, San Francisco, CA 94121, USA. Fax +1 415 386 4044.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.