I'm not impressed. As far as I can tell, once you remove the "filler", they just assume a density and size of "sand" grains which then magically accounts for the acceleration and then they claim their assumptions as the result!

The "study" makes no reference to changes in acceleration over different time frames. If this idea was correct, there should be no anomalous acceleration except while traversing the Kuniper belt. This is not what has been observed.

The level of analysis here is pretty laughable when compared to the 2002 study (54 pages worth) by Anderson, et. al. And note that the authors still think there is a systematic reason for the acceleration, so they are not arguing for "farfetched hypotheses". http://xxx.lanl.gov/abs/gr-qc/0104064