Past research has suggested that caring for an elder with cognitive impairment presents unique problems that cannot be generalized to other chronic illnesses. The behavioral manifestations place strain on the caregiver-elder relationship and the ability of the informal caregiving system to provide care. Limited research is available examining the influence of cognitive status on the pattern of home care use. The purpose of this study was to determine whether an association existed between cognitive status and inhome service utilization by elders and their identified informal caregiver. Elders with three types of cognitive status were studied--those with dementia, those with cognitive loss without dementia and a third group of elders with no cognitive impairment. Four different inhome services were examined--homemaking, inhome nursing, personal care and home delivered meals. Data were obtained from the Manitoba Study of Health and Aging-1 (MSHA-1) using selected variables from the screening, clinical and community-caregiver phases. Caregiver and elder data were linked to facilitate analysis at the caregiver-elder dyad level (N = 380). A modification of the Andersen-Newman health service utilization model was constructed to facilitate analysis and comparison with other studies. Multivariable analysis included a hierarchical logistic regression modelling approach to determine the unique association of each of the variables and categories of variables on the use of inhome services. The study revealed a weak association between cognitive status and inhome service use. Dyads of an elder with dementia were more likely to use personal care services than dyads with a noncognitively impaired elder. Among inhome service users, dyads with an elder who had dementia were more likely to use two or more different types of inhome services than dyads comprising an elder with no cognitive impairment. Cognitive status was not significantly associated with the use of homemaking services, nursing, home delivered meals or the use of one or more inhome services. The presence of one or more functional limitations in the elder was significantly associated with the use of each of the inhome services. Study results indicated that cognitive status was not an important determinant of use of the services under study. Functional status was a powerful determinant of home service use. The findings point to the need to reexamine the operationalization of cognitive status to ensure the characteristics and behaviors associated with cognitive impairment are measured. Further investigation is needed to identify other services that may focus upon the unique needs of the cognitively impaired and their caregivers.