ATTACHMENT B
                              (TAB G)

                          CAUSALITY TABLE

             ***DRAFT*FOR DISCUSSION PURPOSES ONLY***
                          CAUSALITY TABLE

Status of                Causal Relationship in Population
Individual          Unlikely       Controversial       Accepted

Exposure and Disease
present; Other
explanations 
unlikely; PC high        N.R.           ??                  Yes

Exposure and disease
present; Other ex-
planations possible;
or PC low                No             ??                  ??

Exposure undocumented
or absent, or disease
undocumented or absent   No             No                  No

PC=Probability of Causation
N.R.=Not Relevant (PC could not be high if the population
relationship is nonexistent)

     This table summarizes the thought process involved in
considering the causality issue.  The relationship between
exposure and disease in the population must be considered first. 
(This does not necessarily need to be assessed in the particular
population of experimental subjects, however, only in properly
designed epidemiologic studies that would be generalizable to the
particular experiment.)  Remedies should only be considered in
situations where there is a relationship between exposure and
disease that is generally accepted as causal, or at least
reasonably plausible.  Obviously the individual must also have
both exposure and the particular disease under consideration for
a remedy to be contemplated.  The key question are how much
exposure is needed and how confounding and modifying factors are
to be allowed for.  This is where the presumptions fit in to
resolve the "??"s in the table.