Objectives This study evaluated multiple blood lead measures collected over time and assessed differences arising from exposure and testing variability.

Methods Blood lead data was used to compare individuals from manufacturing and construction occupational cohorts. Trends of blood lead levels (BLLs) over time were analysed using mixed model analysis. Random selection of BLL values was used to determine the improvement in the precision of mean BLL estimates as the number of tests increased.

Results From 2003–2007, there were 619 manufacturing and 657 construction workers with more than one blood lead test reported. Construction workers had much more variability in their blood lead trends. They also tended to have less frequent follow-up blood tests compared with manufacturing workers. Both occupational cohorts had persistent BLLs that resulted in many workers with chronically high blood lead values (>25 µg/dL). Approximately 11.2% of construction workers and 34.8% of manufacturing workers with an initial blood lead test above 25 µg/dL remained above this blood level through the study period. The precision in the mean BLL estimates increased more substantially for construction workers when compared with manufacturing workers as the number of blood lead tests per worker increased.

Conclusions This study confirmed differences in the pattern of blood lead tests and the resulting trends for manufacturing compared with construction workers. It also suggested that the number of blood lead tests performed on a worker is an important consideration in the assessment of a worker's mean blood lead estimate, and this is particularly true for workers with highly variable exposures.