Abstract
Homeless people experience elevated rates of risk factors for cognitive impairment. We reviewed available peer-reviewed studies reporting data from objective measures of cognition in samples identified as homeless. Pooled sample-weighted estimates of global cognitive screening measures, full-scale intelligence quotient (IQ), and premorbid IQ were calculated, in addition to pooled sample characteristics, to understand the representativeness of available studies. A total of 24 unique studies were identified, with 2969 subjects. The pooled estimate for the frequency of cognitive impairment was 25%, and the mean full-scale IQ score was 85, 1 standard deviation below the mean of the normal population. Cognitive impairment was found to be common among homeless adults and may be a transdiagnostic problem that impedes rehabilitative efforts in this population. Comparatively little data are available about cognition in homeless women and unsheltered persons.