There are various well-validated taxonomic classifiers for profiling shotgun metagenomics sequencing data, with two popular methods, MetaPhlAn and Kraken, at the forefront of many studies. Despite substantial differences between classification approaches and calls for consensus-based methods, most metagenomic studies rely on a single taxonomic classifier. To compare inferences from multiple taxonomic classifiers, we performed an analysis using MetaPhlAn4 and Kraken2 in parallel and examined diversity trends and species relative abundance associations with age in two studies of extreme human longevity. We used a consensus-based and a novel meta-analytic approach to compare and integrate findings from both taxonomic classifiers. While many results were consistent across the two classifiers, we found classifier-specific inferences that would be lost when using one classifier alone. Alpha diversity analyses using both classifiers captured consistent age associations at higher taxonomic levels in the two cohorts (e.g., increasing trend in phylum-level alpha diversity with age), though trends differed between classifiers at the species level. Beta diversity analyses using both classifiers captured similar community-level changes with age for most taxonomic levels in both cohorts, with Procrustes analysis showing high correspondence between taxonomic profiles based on the different classifiers. A novel correlated meta-analysis approach for differential abundance analysis across classifiers captured more age associated taxa than a consensus approach, including 17 taxa robustly associated with age across cohorts. Several species with previously documented age associations (e.g., Anaerostipes hadrus, Faecalibacterium prausnitzii, and Akkermansia muciniphila) would not have been detected using one classifier alone. This case study emphasizes the value of employing multiple classifiers and recommends novel approaches that can help integrate results from multiple methodologies and provide comprehensive analysis of the data.