Cohen, a member of the staff of the College of Agriculture at Texas A&M University, and his team compared characteristics between three groups of mares: 121 mares who lost their pregnancies early last spring (ETA), 120 mares stabled alongside them who did not lose their foals (control), and 83 mares whose did not abort and whose circumstances made them less susceptible to symptoms of MRLS (low-impact). It is important to note that the median last breeding date of mares in the ETA group was Feb. 28, while the control groups had a medians of March 14 and 15, respectively.

Of the more than 100 characteristics researchers compared, five were more prevalent among the mares who lost their foals versus those in the low-impact group. The study said these five factors were "found to increase the risk of MRLS-associated early-term abortion" in 2001:

- feeding hay in pasture during the four-week period prior to abortion
- a greater than usual amount of white clover in pasture during the four-week period prior to abortion
- a heavier burden of caterpillars in pastures
- prior abortion during the previous five years
- elk or deer observed at the premises during the preceding 12 months

This report did not identify cherry trees or hemlock as risk factors for early-term abortions.

Of the five factors above, the report said "a commonality for those three factors most strongly associated with early-term abortion (feeding hay in pasture, rapid growth of white clover in pasture prior to abortion, and heavier burden of caterpillars in pastures) is that these factors relate to exposure to pasture."

However, the study, which is ongoing, could not pinpoint a single factor as the probable or suspected cause of MRLS.

In fact, many of the mares in both groups came in contact with white clover and caterpillars. The study reported 64% of the mares in the low-impact group were in contact with pastures where white clover was considered in excess in April and May of 2001: of the group that lost their foals, 96% of those mares were in such pastures.

"Although it is possible that there was a toxic principle in the white clover, it is also possible that the rapid growth of white clover was merely a marker for the toxic agent(s) or environmental conditions favoring elaboration of the toxic agent(s) to which affected mares were exposed," the report stated.

"Feeding hay in pasture was considered to be an indicator of increased exposure to pasture...It is possible that horses fed in pasture (presumably on the ground) had greater contact with the causal agent(s) found in pasture."

A "heavy burden" of caterpillars was evident 63% of the time the low-impact group, but of the ETA mares, 91% of them were exposed.

When looking at farms in terms of total losses, the report said "the most important factors at farms that were severely affected were pasture exposure among mares at a particular time of year and at a particular time of gestation." Further reports on that subject, as well as results of studies of pericarditis and late-term abortions are expected in the near future.