The aim of this study was to describe the actual use of medicines in an elderly population and to relate that use to age, gender, type of housing, and type of cohabitation. We used data from the Kungsholmen project, an ongoing longitudinal study in Stockholm, Sweden. The Kungsholmen project focuses on the medical, psychological, and social problems of aging, with an emphasis on dementia. All inhabitants of the Kungsholmen parish aged 75 years or more were selected for the study. The participation rate was 76%. More than 80% of those living in their own homes1and more than 90% of those living in sheltered accommodation or in a healthcare institution used medicines. The mean number of medicines used was 3.2, ranging from 0 to 15. Women used a larger number of medicines than men, 3.4vs2.6. People in the age group 75 to 79 years used fewer medicines than those in the older age groups. People who were 85 years or older used significantly more medicines than those between 75 and 84 years. The number of medicines used by elderly persons in different age groups was different among women, but not among men. People living in sheltered accomodation used a larger number of medicines on average than those who lived in their own home or in nursing homes. The whole population, men and women, and those aged 80 to 84 years, living in their own homes, used fewer medicines than those living in sheltered accomodation. There was no difference between people living in different types of cohabitation with respect to the number of medicines used. The most commonly used groups of therapeutic medicines were psycholeptic agents, diuretic agents and cardiac drugs, accounting for 41% of the total medicines used. The therapeutic/pharmacological subgroups of medicines used by most people were hypnotic and sedative agents [Anatomical Classification System (ATC) group N05C], used by 26%; cardiac glycosides (ATC group C01A), used by 19%; and loop diuretics (ATC group C03C), used by 17%. Women used significantly more hypnotic, sedative and anxiolytic agents than men. In conclusion, our findings show that a higher proportion than has previously been shown of elderly people living in their own homes use medicines, that women use a higher number of medicines than men only in the age group 80 to 89 years, and that more women than men used psycholeptic agents. Furthermore, the number of medicines used increased only up to the age of 89 years. People living in sheltered accommodation used more medicines than those living in their own homes or nursing homes. Age and gender affect the number of medicines used when the person is living in his or her own home, but not when living in an institution.