Most explanatory research into the utilization of home care for the elderly has been carried out in a cross-sectional design with multiple regression as the main method of analysis. For methodological reasons we chose another design in a project called 'Professional home care and informal help for the elderly', which has been conducted in the northern part of the Netherlands. Two types of causal variables were distinguished to detect influencing factors on professional home care utilization among the elderly, other than physical limitation: person-bound variables and social network variables. A Mokken Scale analysis for Polychotomous items (MSP) was used to measure the level of physical limitation and a matching procedure to compare 'users' and 'non-users' of professional home care. With regard to person-bound variables, sex, whether or not a person was living alone and the level of the elderly person's income appeared to play a role in the utilization of home care: the user group comprised significantly more women, more elderly living alone and more persons on a low income. Contrary to the findings in other Dutch research, depression and feelings of loneliness did not seem to discriminate between the two groups. With regard to social network variables, the size and structure of the social network was more or less identical in both groups. The non-users network lives slightly closer. In general, the small differences found between the groups were to the non-users' advantage. Moreover, the non-users received more informal and private care with ADL and IADL activities.