Folks interested in this topic might be interested in reading results posted on astro-ph last night:
…which looks at ASAS-3 data for relatively bright red variables to identify those with long-term secondary periods. This references recent work based on AAVSO data alone.
\Brian