In order to quantify the impact of parasites on host population dynamics, experimental manipulations that perturb the parasite-host relationship are needed but, logistically, this is difficult for wild hosts. Here, we describe the use of a delayed-release anthelmintic delivery system that can be administered when the hosts can be captured and its activity delayed until a more appropriate period in the host-parasite cycle. Our model system is Svalbard reindeer infected with a nematode parasite, Marshallagia marshalli, which appears to accumulate during the Arctic winter. To determine the extent to which this occurs and the effect on host fitness, reindeer need to be treated with anthelmintics in late autumn but they can only be caught and handled in April. To solve this problem, we devised an intra-ruminal capsule that releases the anthelmintic from up to 6 months after being administered. The capsule was trialed in cannulated sheep and red deer to determine optimum capsule orifice size and release rates. Capsules were estimated to release placebo for 100–153 days followed by abamectin for 22–34 days. To test the efficacy of treatment in reindeer, capsules were administered in April and retrieved in October. All capsules had fully released the anthelmintic and treated reindeer had significantly lower worm burdens than controls. Thus, success of this system allows repeated treatment over several years to test the effect of winter parasitism on host fitness.