The aim of the investigation was to assess the relations between pairs of personal, indoor, and outdoor levels of fine particles and their components with respect to effects for older subjects with cardiovascular disease. In the framework of a study funded by the European Union (Exposure and Risk Assessment for Fine and Ultrafine Particles in Ambient Air; referred to as ULTRA)*, panel studies were conducted in Amsterdam (The Netherlands) and Helsinki (Finland). Concentrations of outdoor particulate matter 2.5 pm or smaller in aerodynamic diameter (PM2.5) were measured at a fixed site in each location. With HEI funding, each subject's personal and indoor PM2.5 exposure was measured every other week for 6 months during the 24-hour period preceding intensive health measurements. Particle reflectance was measured as a marker for diesel exhaust. Elemental content of more than 50% of the personal and indoor samples and all corresponding outdoor samples was measured using x-ray fluorescence (XRF). Ion content (sulfate, nitrate) was measured using chromatography. For Amsterdam, 337 personal and 409 indoor measurements were collected from 37 subjects; for Helsinki, 336 personal and 503 indoor measurements were collected from 47 subjects. Median personal, indoor, and outdoor PM2.5 concentrations were 13.6, 13.6, and 16.5 microg/m3 in Amsterdam and 9.2, 9.2, and 11.1 microg/m3 in Helsinki. In both cities, personal and indoor PM2.5 concentrations were lower than and highly correlated with outdoor concentrations (median correlation coefficient [R] 0.7-0.8). For most elements, personal and indoor concentrations were also highly correlated with outdoor concentrations. The highest correlations (median R > 0.9) were found for sulfur (S), sulfate, and particle reflectance (reported as the absorption coefficient). Reflectance was a useful proxy for elemental carbon (EC), but site-specific calibration with EC data is necessary. The findings of this study support using fixed-site measurements as a measure of exposure to PM in time-series studies linking the day-to-day variations in PM to the day-to-day variations in health endpoints, especially for components of PM that are generally associated with fine particles and have few indoor sources.