Atrial fibrillation: the risk of bleeding and stroke were most important in choosing antithrombotic therapy for elderly patients at risk of falling

Clinical bottom line (level 1b)

  1. In this decision analysis, warfarin was the optimal therapy for preventing stroke in elderly patients with atrial fibrillation at risk for falling.
  2. Aspirin was more favourable in patients at high risk for intracerebral haemorrhage or non-CNS bleeding, or in patients at low risk for stroke.
  3. Aspirin was more favourable in patients who particularly dislike taking warfarin.
  4. The probability of falling had no effect on the preferred outcome.
Man-Son-Hing et al: Archives of Internal Medicine 1999; 159: 677-685
Expires November 2003

The study

Markov decision analysis comparing warfarin, switching to aspirin if a major extracranial bleed occurred; aspirin, switching to warfarin if a TIA or resolving stroke occurred; no treatment, switching to aspirin if a TIA or resolving stroke occurred.
Setting: health-care provider

Data taken from a systematic literature review using Medline (1966 to 1996) (search strategy: accidental falls, anticoagulants, cerebral haemorrhage, subdural hematoma, aspirin, warfarin, cerebral hematoma, atrial fibrillation, outcome assessment (health care), treatment outcome, prognosis and risk factors) and hand-searching bibliographies of relevant articles and contacting experts to identify missed articles. 49 articles were found - details of selection criteria given in text.

  • Viewpoint: patients aged 65 or more with atrial fibrillation, who are at risk of falling
  • Benefit assessment: stroke, death, subdural haematoma or intracerebral haematoma, major non-CNS bleeding, falls
  • Resources and costs: Utilities based on a study of 69 elderly patients on the disability of minor, moderate and major strokes
  • Sensitivity analysis: varying probability of stroke, subdural or intracerebral haematomas, non-CNS bleeding with each of the medication, falls and the disutility associated with longterm medication use

    • The evidence

    intervention cost
    warfarin 12.9 QALYs
    aspirin 11.2 QALYs
    no treatment 10.2 QALYS

    Effect of sensitivity analysis: The analysis was most sensitive to changes in the probability of intracerebal haemorrhage, non-CNS bleeding, and risk of bleeding. The probability of falling had no effect on the preferred outcome.
    • Aspirin was favoured for patients who ranked the utility of long-term warfarin as less than 0.82. No patient in the utility study ranked warfarin this low.

    Comments

    1. This was a very robust, sound decision analysis, with no changes in utilities occurring with the use of sensitivity analysis.

    Citation

    1. Man-Son-Hing M, Nichol G, Lau A, et al: choosing antithrombotic therapy for elderly patients with atrial fibrillation who are at risk for falls. Archives of Internal Medicine 1999; 159: 677-685
    Search Terms: hand-search
    Contributor: Chris Ball and Clare Wotton, November 1999
    Reviewer: Daniel Sontheimer

    Clinical Question.
    Patient elderly with AF
    Intervention or Exposure aspirin
    Comparison warfarin
    Outcome utility