toms, diaries of symptoms and medication use have been effective in identifying air pollution as a risk factor for respiratory illness in recent years. Most of these studies also demonstrate a lack of association between air pollution and nonrespiratory illness, giving credence to the primary findings. Monitoring of peak flow rates and symptoms was also useful in the analysis of occupational asthma in electronic workers employed in soldering (Burge et al., 1979).
Schwartz and Zeger (1990) re-examined an earlier diary study of student nurses in Los Angeles, California (Hammer et al., 1974), and found significant associations between exposure to ozone and the incidence of coughing, between exposure to nitrogen dioxide and daily incidence rates of sore throats, and between exposure to carbon monoxide and headaches. Analysis of the diary entries also proved a sensitive instrument for detecting the connection between chronic exposure to passive smoke and coughing with phlegm production.
A study in Utah of respiratory symptoms, medication use, and daily Peak Expiratory Flow (PEF) measurements of children and adults with asthma (Pope, 1991) finds a clear association between increased symptoms and decreased PEF values and levels of PM10 particulate pollution, which originated mostly from a nearby steel mill. This study was not complicated by the presence of other pollutants, such as ozone, sulfur dioxide, or sulfuric acid aerosol. Ozone has been associated with symptom reporting in persons with asthma (Whittemore and Korn, 1980). Ostro et al. (1991) report that symptoms worsened in a group of asthma patients studied in Denver, Colorado, in relation to daily aerosol hydrogen ion levels.
Studies reported by Lippman (1989), Raizenne et al. (1989), and Kinney et al. (1988) of children attending summer camps are a special case, because daily symptom reporting and daily measurements of lung function can be correlated to pollution levels measured continuously at the same site. The researchers found it difficult to attribute symptoms of function loss to specific pollutants, but they did show that combinations of pollutants, such as ozone, aerosol sulfates, and sulfuric acid, are associated with a variety of adverse respiratory effects. In these studies, daily measurements of lung function have generally been used as the outcome measurement.
Diary data might be useful in assessing the occurrence of symptoms when exposure to a single-point source (such as a factory) is intermittent because of changes in wind direction, and they would of course be much easier to interpret where the individuals have not been aware of fluctuations in the concentrations of the pollutants.