Ward’s hierarchical clustering method clustered all specimens extracted using the Norgen kit distinctly from those extracted using the other two kits, with considerable overlap observed between the Qiagen and MagMAX methods. Within the extraction method clusters, specimens did not clearly segregate by the number of centrifugation steps, collection tube type, or plasma/serum aliquot volume. The main genus driving the separation based on extraction method were Gordonia, Micrococcus, Streptococcus, Delftia, and Pseudomonas. Simarly, principal component analysis also separated specimens extracted using the Norgen kit from those extracted using the other two kits, with no separation based on the number of centrifugation steps, collection tube type, or plasma/serum aliquot volume observed. In 52 of 54 cases (96%), PERMANOVA identified significant differences in microbial sequence abundance based on extraction method, regardless of number of centrifugation steps, collection tube type, or plasma/serum aliquot volume. As expected, differences were significant between Norgen and the other two methods under all conditions, but significant differences between Qiagen and MagMAX methods were found with two exceptions (using 100 or 500 µL single-spun serum or 200 µL single spun sodium citrate plasma). Only 9% of specimens displayed differences based on aliquot volume and 4% based on tube type. DNA extraction method had the largest effect on the bacterial composition, with more similarity observed between the Qiagen and MagMAX extraction methods at both the phyla and genus level than between either kit and the Norgen kit; 16 of the 20 most abundant phyla were common among the Qiagen and MagMAX methods, but only 5 of these were found among the top 20 when extraction was with the Norgen kit. Compared to specimens extracted using the Qiagen or MagMAX kit, specimens extracted using the Norgen kit had significant differences in levels of 548 and 533 bacteria, respectively (30 bacteria per group). Only 25 bacterial abundances were found to be significantly different between Qiagen and MagMAX kits. By comparison tube type, the number of centrifugation steps, and plasma/serum aliquot volume had an impact on the β-diversity of 28, 23, and 27 bacteria, respectively, and many of these were only observed when the Norgen method was used for extraction. Limited effects of collection tube type, input volume, and extraction method on α-diversity were observed. Collection tube type affected Pielou’s evenness but only when serum/plasma was separated by single centrifugation and extraction was using 500 µL of Serum/Plasma along with the MagMAX kit (P=0.042). The volume of input did not affect bacterial diversity, but bacterial richness was lower when extraction was from 100 µL EDTA plasma using MagMAX kit or 100 µL sodium citrate plasma using the Qiagen or Norgen kits compared to at least one larger volume (200 or 500 µL, P<0.05). Bacterial richness was higher when plasma/serum was obtained by double rather than single centrifugation and when extraction was with the Norgen kit, regardless of collection tube. Bacterial richness was generally higher when extraction was with the Norgen kit compared to the other kits evaluated, with significant differences found in the double centrifugation group (regardless of volume or tube type), and when extraction was from 100 or 500 µL serum.