The unprecedented widespread closing of buildings due to the COVID-19 pandemic has allowed water to stagnate in premise plumbing systems, creating conditions that may facilitate the growth of opportunistic pathogens. In this study, we flushed and collected samples from showers in buildings that had been unoccupied for approximately two months and quantified Legionella pneumophila using a commercial cultivation-based assay. In addition, all bacteria, Legionella spp., L. pneumophila, L. pneumophila serogroup 1, non-tuberculous mycobacteria (NTM), and Mycobacterium avium complex (MAC) were analyzed using quantitative PCR (qPCR). Despite low or negligible total chlorine in the stagnant pre-flush water samples, L. pneumophila were not detected by either method; Legionella spp., NTM, and MAC, however, were widespread. Using quantitative microbial risk assessment (QMRA), estimated risks of clinical illness from exposure to legionella and MAC via showering were generally low, but the risk of subclinical infection via Legionella spp. could exceed a 10-7 daily risk threshold if just a small fraction (≥0.1 %) of those legionellae detected by qPCR are highly infectious. Flushing cold and hot water lines rapidly restored a total chlorine (as chloramine) residual and decreased all bacterial gene targets to building inlet water levels within 30 min. Following flushing, the chlorine residual rapidly dissipated and bacterial gene targets rebounded, approaching pre-flush concentrations after 6 to 7 days of stagnation. These results suggest that stagnant water in premise plumbing may contain elevated levels of opportunistic pathogens; flushing, however, can rapidly improve water quality and reduce the health risk but the improvement will be short-lived if building disuse persists.