Wastewater-based epidemiology (WBE), which involves using biological indicators in sewage to provide information on the overall health of a community, is a powerful tool to monitor public health. WBE offers several advantages that make it complementary to conventional clinical surveillance: it is rapid and resource-efficient, enables broad monitoring of large populations, is able to detect symptomatic and asymptomatic infections, and is not biased by health seeking behavior or access to healthcare resources. Recent studies have shown that WBE is a powerful tool for estimating community-level prevalence of COVID-19 by measuring levels of SARS-CoV-2 RNA in wastewater, and for predicting the prevalence of clinical antibiotic resistance by screening wastewater for antibiotic resistance genes. Furthermore, WBE has enabled global collaboration through national (e.g., National Wastewater Surveillance System (NWSS) on COVID-19) and international (e.g., the Enhanced Gonococcal Antimicrobial Surveillance Program) programs to advance the integration of WBE into public health response. Despite the surge of interest in applying WBE, there are currently no standardized methods for wastewater disease monitoring, including how and when to collect samples, what methods to use for analysis, and how to interpret the data to inform action. Without a more complete understanding of the methodological challenges involved in characterizing target indicators in wastewater samples, our ability to leverage WBE for routine monitoring and international collaboration is limited. This dissertation aims to evaluate the strengths and weaknesses of several current methods used for wastewater monitoring of antimicrobial resistance (AMR) and SARS-CoV-2 and discuss implications of method selection for future WBE work. The research focuses on four objectives, corresponding to the four chapters presented in this dissertation: (1) characterize the impact of wastewater sampling designs (i.e., grab and composite sampling) on the ARG removal rates achieved by a wastewater treatment plant (WWTP), (2) elucidate the fate of different forms of cell-associated and cell-free ARGs in an emerging wastewater treatment process, (3) compare two targeted methods (i.e., RT-ddPCR and targeted amplicon sequencing) for monitoring SARS-CoV-2 mutations in wastewater, and (4) evaluate short- and long-read metagenomics and a targeted method (epicPCR) for tracking ARG host range across a WWTP. Sampling design is critical to the collection of representative samples for WBE and for estimating removal rates of genes across wastewater treatment processes. We compared grab and composite sampling in terms of their effects on removal rates for a suite of genes, including several clinically-relevant ARGs (blaNDM-1, blaOXA-1, MCR-1, MCR-5, MCR-10, and qnrA). We find that the diurnal variation of ARG loading in the WWTP influent and effluent created significantly different instantaneous ARG removal rates among all grab samples collected throughout a day, indicating grab sampling can introduce bias to ARG removal calculations. Overall, using composite samples are more representative for WBE and for assessing removal of ARGs across wastewater treatment processes as compared to grab sampling which may overestimate ARG removal rates. The form of the ARG, specifically whether it is cell-free or cell-associated, is critical to understanding ARG removal across wastewater treatment processes. We found that the fraction of cell-associated ARGs decreased whereas the fraction of cell-free ARGs increased in the treated effluent as the influent organic loading rate was gradually increased. The results indicate that the ARGs in treated effluent can transit between cell-associated and cell-free DNA in response to changing operational conditions, which should be considered to better evaluate the total ARGs in the wastewater treatment system. WBE has been widely applied to track SARS-CoV-2 infections in communities and in some cases to identify circulating variants of concern. There are several different methods that have been applied to screen for variants of concern in wastewater. We compared targeted methods for screening for SARS-CoV-2 variants of concern in wastewater samples. The results demonstrated that RT-ddPCR is more sensitive and should be applied for mutation quantification or variant confirmation in wastewater, whereas detection via targeted amplicon sequencing was influenced by the depth of sequencing, viral load, and mutation concentration. These findings caution the use of quantitative measurements of SARS-CoV-2 variants in wastewater samples determined solely based on targeted amplicon sequencing. We compared targeted and untargeted methods for ARG detection in wastewater. The results demonstrate that despite its significantly lower sequencing depth, long-read sequencing outperforms short-read sequencing with higher sensitivity for detecting ARGs, especially for ARGs associated with mobile genetic elements (MGEs). In addition, long-read sequencing consistently revealed a wider range of ARG hosts compared to short-read sequencing. Nonetheless, the host range detected by long-read sequencing represented only a subset of the host range detected by a targeted method, epicPCR (Emulsion, Paired Isolation, and Concatenation PCR). Taken together, the results have implications for future WBE, particularly in terms of method selection: 1) collect composite samples rather than grab samples to acquire a representative view of the monitoring targets in a population; 2) include different forms of DNA (cell-associated and cell-free) to analyze ARGs because effluent ARGs are present in both forms and can transition between these forms in response to environmental conditions; 3) apply RT-ddPCR for quantitative analysis and early variant detection if targets are known; and 4) use long-read sequencing for routine wastewater AMR surveillance and use epicPCR to obtain a high-resolution host range of clinically relevant ARGs. The findings provided by this research contribute to establishing a scientific consensus on method selection for WBE, thus advancing it as a routine tool for public health surveillance.