Study eligibility criteria
Abstracts were screened and potentially relevant articles
obtained. Retrieved articles were independently evaluated by
two people to determine whether they met the following in-
clusion criteria for full review; any disagreement was
resolved by consensus:
• Sample participants ≥60 years.
• Prospective cohort design, at least 1 year duration, with
minimum 80% sample follow-up. Fall prevention trials
were excluded, as the intervention may have successfully
modified deficits.
• Samples comprised community or institution-dwelling
populations. Studies of acute care settings with short-term
length of stay were not included.
• ‘Falls’ was the primary outcome, including ‘any fall’, ‘re-
current falls’ and ‘injurious falls’. For the evaluation of ex-
tremity fractures, an explicit statement was needed that
fractures were the result of falling.
• Cognitive function was measured at baseline.
Measurement of cognitive function had to be explicitly
detailed in the methods or referenced for validity and/or
reliability.
• Inclusion and exclusion criteria and demographic infor-
mation were reported.
• Confounding factors were reported and used in multivari-
able regression analysis to generate adjusted risk estimates.
• Samples did not comprise a disease-defined population
(i.e. stroke, Parkinson’s disease and dementia).
For a study to be considered within the disease-specific
diagnosis category, the study needed to have analysed the
association between fall risk and dementia, a dementia
sub-type or mild cognitive impairment; a measure of global
cognitive status was defined as the utilisation of general
cognitive screening tests that evaluate several cognitive
domains in a single summary score and specific cognitive
domains were classified according to specific test demands
and the mental processes required to execute them (i.e. EF,
memory).