Genetic monitoring is important in small, fragmented populations that rely on gene flow to maintain genetic diversity. The Selkirk, Yaak, and Cabinet grizzly bear (Ursus arctos) populations are among the smallest in North America and are near the southernmost extent of the species’ range. These populations received little to no effective migration for generations but have recently experienced increased gene flow through natural migration and a population augmentation program. A long-term dataset of grizzly bear microsatellite genotypes from 1973 to 2021 presented a unique opportunity to examine genetic trends in these populations over time. We used this dataset of 464 bears to evaluate if gene flow affected observed heterozygosity (H_O), expected heterozygosity (H_E), allelic richness (A_R), and average pairwise relatedness (r) in each of these populations. We also estimated effective population size (N_e) using the temporal and linkage disequilibrium (LD) methods. Post gene flow, A_R increased in the Selkirk and Cabinet populations and r decreased in all three populations. We did not observe any significant changes in H_E or H_O, but H_E values in our populations were significantly higher than those estimated using a model without gene flow. Our N_e estimates were consistent between the temporal and LD methods and ranged from 15.2 to 15.8, 15.4–17.5, and 5.6–8.9 for the Selkirk, Yaak, and Cabinet populations, respectively. Overall, our findings indicate that gene flow is increasing or maintaining genetic diversity in these populations. However, N_e remains low and additional connectivity or augmentation may be needed, particularly in the Cabinet population.