Dose-effects functions for carboxyhemoglobin and behavior.

Literature on dose-effects functions of carboxyhemoglobin (COHb) and human behavior is internally inconsistent. A recent study was performed to describe such a function for tracking behavior in humans (7). No significant effects were found, however, even for COHb levels up to ca. 17%. The present review and analysis of published data was undertaken to attempt to reduce uncertainty about COHb dose-effects functions. Dose-effects data from six laboratory animal operant behavior studies were reviewed. Maximum COHb levels ranged from 33 to 59%. Data were converted to a common metric and dose-effects functions were fitted. The preferred function was nonlinear, with a shallow slope, indicating small effects (if any) for COHb levels less than ca. 20%. As COHb increased further, the slope of the function became progressively steeper. Review of the human dose-effects function publications for COHb and behavior revealed that no significant findings had survived the test of direct or partial replication. Other studies found no significant effects. A dose-effects curve was fitted to the human data from one study (7) and compared to the laboratory animal curves. Extrapolation of the human curve was used to suggest that the dose-effects curves for humans and laboratory animals were of similar form and that effects of COHb on behavior in man are small or absent below ca. 20%. It is also possible that some uncontrolled variables are affecting the results so that COHb produces effects only some of the time. No conclusion can be drawn from the body of literature due to the divergence of results.