Over my career, I have never seen epidemiological evidence of tritium exposure leading cancer clusters or other abnormalities. With significant human case study populations at DOE sites like SRS, Mound, LANL, and all the CANDU facilities (not to mention the emergency exit sign accidents); you would expect to see more correlation. Maybe I'm looking in the wrong places or not hard enough.
My personal experience with tritium exposure is that the people I know who have accrued significant lifetime doses from tritium are living happy and healthy retirements. Many of them joke of the tritium radiohormesis effect. I personally need to research the topic a bit more.