RESUMO
The purpose of this paper is to estimate the volume and composition of referrals to home care on the basis of applicant characteristics. The relationships between the background and care needs of applicant groups on the one hand and the referral of home care packages on the other, are studied by means of a multinomial logit model. The model is estimated on the basis of more than 7000 requests for home care in the northern part of the Netherlands. Home health care institutions have to deal with clients who arrive from many different branches of the health care system. As a result the services or products provided by home health care institutions are characterized by wide variation. In the modeling emphasis has, therefore, been placed on the differentiation of clients and products. We find for instance that elderly chronically ill applicants have a greater chance of being referred for domestic help only, while applicants with psychosocial disorders are more liable to be offered packages that include social support. Patients discharged from hospital have a greater chance of a referral to domestic help only when they are slightly disabled, and are more likely to be offered packages including physical care when they are more disabled. The model has a range of policy applications in assessing the impact of changes in the health care system on the volume and structure of the demand for home care services. Examples are presented of the consequences of the ageing population and earlier discharge from hospitals on demand for home care packages.
