We combined field and laboratory experimentation to evaluate the effects of nitrogen dioxide in a panel of Los Angeles area residents with chronic respiratory illness, 15 men and 11 women aged 47 to 69. All had heavy smoking history, chronic symptoms, and low FEV1; some also had low FVC. During the fall-winter high-NO2 season, they monitored themselves for 2-wk periods using spirometers in the home, passive NO2 sampling badges, and diaries to record time and activity patterns and clinical status. In the middle of each self-monitoring week they were exposed in a chamber, once to clean air and once to 0.3 ppm NO2. Chamber exposures were double blind, lasted 4 h, and included four 7-min exercise sessions with average ventilation rates near 25 L/min. Symptom reports and hourly forced expiratory function tests showed no statistically significant differences between clean air and NO2 chamber exposures, although peak flow showed a approximately 3% loss with NO2 relative to clean air during the first 2 h of exposure only (p = 0.056). No significant overall differences were found between field self-measurements and measurements of lung function in the chamber or between field measurements in clean air and NO2 exposure weeks. Field data showed that group average lung function and symptom levels were worse in the morning than later in the day (p < 0.005) but otherwise were stable over 2 wk. Even though most subjects smoked and stayed indoors 80 to 90% of the time, personal NO2 exposures correlated significantly with outdoor NO2 concentrations as reported by local monitoring stations.(ABSTRACT TRUNCATED AT 250 WORDS)