We propose that voluntary work, characterized by social, physical and cognitive activity in later life is associated with fewer cognitive problems and lower dementia rates. We test these assumptions using 3-wave, self-reported, and registry data from the 2010, 2012, and 2014 Swedish National Prescribed Drug Register. We had three groups of seniors in our data: 1) no volunteering (N = 531), 2) discontinuous volunteering (N = 220), and 3) continuous volunteering (N = 250). We conducted a path analysis in Mplus to investigate the effect of voluntary work (discontinuously and continuously) on self-reported cognitive complaints and the likelihood of being prescribed an anti-dementia treatment after controlling for baseline and relevant background variables. Our results indicated that seniors, who continuously volunteered, reported a decrease in their cognitive complaints over time, whereas no such associations were found for the other groups. In addition, they were 2.44 (95%CI [1.86; 3.21]) and 2.46 (95%CI [1,89; 3.24]) times less likely to be prescribed an anti-dementia treatment in 2012 and 2014, respectively. Our results largely support the assumptions that voluntary work in later life is associated with lower self-reported cognitive complaints and a lower risk for dementia, relative to those who do not engage, or only engage episodically in voluntary work.