Trypanosomatids are a diverse family of protozoan parasites, some of which cause devastating human and livestock diseases. There are two distinct infection life-cycles in trypanosomatids; some species complete their entire life-cycle in a single host (monoxenous) while others infect two hosts (dixenous). Dixenous trypanosomatids are mostly vectored by insects, and the human trypanosomatid diseases are transmitted mainly by vectored parasites. While infection prevalence has been described for subsets of hosts and trypanosomatids, little is known about whether monoxenous and dixenous trypanosomatids differ in infection prevalence. Here, we use meta-analyses to synthesise all published evidence of trypanosomatid infection prevalence for the last two decades. We employed a semi-automated screening protocol, using machine learning algorithms and natural language processing, resulting in the qualitative inclusion of 569 citations and quantitative inclusion of 261 citations. We find striking difference in infection prevalence with monoxenous species having twice the infection prevalence (19.8%) than dixenous species (8.68%) in both insect and non-insect hosts and are more than three-fold more prevalent in insects (20.9%) compared to their dixenous kins (6.61%). We also find that dixenous trypanosomatids have lower infection prevalence among insects compared to their definitive hosts, which is consistent across dixenous genera. Within the monoxenous trypanosomatids, genera infecting bees are characterised with the highest prevalence, which does not vary between wild and managed bees. To our knowledge, these results reveal for the first time, a fundamental difference in infection prevalence according to host specificity where vectored species suffer from lower infection prevalence as a result of a ‘jack of all trades, master of none’ style trade-off between the vector and definitive host.