Empirical derivation and validation of a wandering typology.

OBJECTIVES To develop and validate a wandering typology. DESIGN Cross-sectional, correlational descriptive design. SETTING Twenty-two nursing homes and six assisted living facilities. PARTICIPANTS One hundred forty-two residents with dementia who spoke English, met Diagnostic and Statistical Manual for Mental Disorders, Fourth Edition, criteria for dementia, scored less than 24 on the Mini-Mental State Examination (MMSE), were ambulatory (with or without assistive device), and maintained a stable regime of psychotropic medications were studied. MEASUREMENTS Data on wandering were collected using direct observations, plotted serially according to rate and duration to yield 21 parameters, and reduced through factor analysis to four components: high rate, high duration, low to moderate rate and duration, and time of day. Other measures included the MMSE, Minimum Data Set 2.0 mobility items, Cumulative Illness Rating Scale-Geriatric, and tympanic body temperature readings. RESULTS Three groups of wanderers were identified through cluster analysis: classic, moderate, and subclinical. MMSE, mobility, and cardiac and upper and lower gastrointestinal problems differed between groups of wanderers and in comparison with nonwanderers. CONCLUSION Results have implications for improving identification of wanderers and treatment of possible contributing factors.