over the period; the highest percentage (18 percent) appeared in 1983, and the lowest (12 percent) appeared in 1985. When the articles were grouped into three categories suggested by Zielhuis (1984, 1985) it was shown that 69 percent of the studies measured a toxin, 19 percent of the studies measured a metabolite, and 12 percent measured an agent-specific nonadverse effect (a reversible biologic change that is related to a specific exposure).
Then the articles were grouped into six categories, the three previously mentioned and three others. The most frequent topics were measurements of biologic substances that were not agent-specific but that caused nonadverse biological effects. This group, exemplified by changes in routine clinical chemistries, could be termed ambiguous effects, and it accounted for 26 percent of studies. Another group of measures of health effects, such as abnormal urine cytology, made up 14 percent of the studies. A third group of studies constituted 31 papers (5.3 percent) involving genetic monitoring and 4 papers (0.6 percent) involving genetic screening. Overall, it was found that 41 percent of the studies measured a toxic substance, 11 percent measured a metabolite, and 7 percent measured an agent-specific nonadverse effect.
Approximately half of the studies involved sampling blood for biologic monitoring. There appears to be an increasing trend in the use of blood specimens for monitoring. Urine was monitored in 28.3 percent, lung tissue in 4.3 percent, expired air in 2.3 percent, hair in 2.2 percent, adipose tissue in 1.2 percent, and saliva in 1.1 percent. Other media, such as breast milk, teeth, semen, bone, feces, liver, conjunctival fluid, and skin, were 14.9 percent of the media evaluated. Each of these other media was examined in 1-2 percent of the studies. Except for blood, the trends in all other media were fairly constant.
Bernard and Lauwerys (1986) observe that most biologic monitoring studies focus on the relationships between internal dose and external exposure rather than between internal dose and adverse effect. Schulte et al. (1987) confirmed this observation—74 percent of the studies reviewed in their paper evaluate the relationship between the level of environmental exposure and the biologic level of the toxicant or metabolite. Typical among this type of study was the correlation of lead measured in air and the concentration of lead in blood. Approximately 21 percent of studies attempt to link the results of biologic monitoring done concurrently with the assessment of a health outcome. Similarly, but to a lesser degree, approximately 3 percent