It appears that everything else I've been looking at is quoting the same articles as these two sites.
From the first article:
Are there any empirical studies of anniversary
There are few empirical studies of anniversary reactions. In one study, 92 widows and widowers were interviewed on the first anniversary of their spouse's death. Four of the participants reported clinically significant depression that they connected to the anniversary date (Borstein & Clayton, 1972). In a series of studies, Morgan and colleagues examined anniversary reactions in Gulf War veterans two and six years after the end of the Gulf War (Morgan, Hill, Fox, Kingham, & Southwick, 1999; Morgan, Kingham, Nicolaou, & Southwick, 1998). The researchers asked the veterans and their wives to identify the veteran’s worst month of functioning in the past year. When the researchers compared the worst month identified to previously identified dates of traumatic events that occurred during the Gulf War, they found that 38% of participants reported that their worst month coincided with the month in which their trauma occurred (Morgan et al., 1999). Veterans with these anniversary reactions had significantly more PTSD symptoms than veterans who did not have anniversary reactions, and all of the veterans who met criteria for a diagnosis of PTSD had anniversary reactions (Morgan et al, 1999). Finally, one study was done examining patterns of hospital admissions in patients with seasonal mood disorders (Beratis, Gourzis, & Gabriel, 1996). Based on chart reviews, 4 out of 41 patients with multiple hospital admissions over a seven-year period exhibited depressive or manic episodes that coincided with the time of a past traumatic event.
Now here are more unique links:
http://www.trauma-pages.com/vanderk4.htm ==>Actually, this page is no long available, so I am going to use the archived version. Someone shared this page with me about four years ago and it is far too good to lose track of.