In our current era, where private organizations are assuming ever greater responsibility for public service provision, social and health service providers also face increased pressure to engage client and patient perspectives in the service production process. This dissertation investigates how providers co-produce services with vulnerable users, and how those endeavors are associated with service offerings and service utilization in the field of substance use disorder (SUD) treatment units, which serve stigmatized client populations whose voices have traditionally been marginalized. Bringing together the literature on human service organizations from social work, deliberative democracy and co-production from public administration, and patient-centered care from the field of medicine, I theorize that SUD clinics adopt two co-production mechanisms: (1) patient-centered care (i.e., directly collaborating with patients in clinical decision-making), and (2) peer co-production (i.e., hiring staff members with lived experience of addiction as proxies of patients’ voice). To test my theory empirically, I use the 2017 National Drug Abuse Treatment System Survey—a nationally representative longitudinal split-panel survey of approximately 700 alcohol and drug use disorder treatment facilities in the United States.,The results show that more than a half of treatment centers across the U.S. implemented either patient-centered care or peer co-production mechanisms to incorporate patient’s perspectives into care processes. Residential units were more likely to implement patient-centered care and peer co-production methods compared to outpatient units, possibly due to sufficient time for patient-clinician interactions and the tradition for leveraging peers in these settings, respectively. For-profit units, compared to nonprofit and public clinics, were more likely to adopt both co-production mechanisms, not only to reducing staffing costs, but also to customize services for individual patients in competitive markets. Besides, clinics were much more likely to practice peer co-production when managers believe in peer co-production potentials among staff with lived experience. Lastly, clinics serving more patients with opioid use disorder patients tend to not practice patient-centered care, and clinics serving a greater proportion of prescription opioid use disorder were less likely to implement peer co-production method—signaling differentiated co-production efforts in the SUD treatment field. ,This dissertation provides meaningful but limited evidence of associations (and lack of associations) between co-production efforts and service output patterns. Availabilities of various services offered at clinics were not strongly associated with patient-centered care or peer co-production efforts, but with various organizational factors (such as service modality, unit type, and staff and patient compositions, and revenue sources). In terms of service utilization, practicing patient-centered care was correlated with patients’ greater utilizations of various harm reduction and supportive services that support patients’ long-term recovery, while peer co-production had only few associations with service utilization patterns. The overall lack of associations may signal that co-production efforts in a service delivery phase might not have strong relationships with service offering patterns. Peer co-production mechanism’s negative associations with opioid maintenance therapy availability and lack of associations with utilization of various services signal a potentially critical limitation of peer co-production that staff with lived experience may not serve as good representatives of patients’ best interests. ,Using a strong research design and a nationally representative dataset, this study provides multiple implications for theory, policy, and practice. The dissertation proposes a framework to conceptualize various co-production models and differentiate them from other service production modes. By testing the framework with the nationally representative data, the dissertation also suggests possible operations of multiple co-production mechanisms in a service field and encourages future research on conditions for and impacts of various co-production efforts in many health and social service fields. In terms of policy implications, the current dissertation provides important suggestive evidence that collaborative process can be an important way to address the opioid crisis, encouraging policy makers and government officials to emphasize and incentivize co-production efforts at SUD clinics. Regarding implications for practice, the dissertation encourages health and social service organizational managers to think about various ways to co-produce with service users. By uncovering the non-clinical functions of staff with lived experience at SUD clinics, this study encourages managers not only to re-consider the functions of staff members with different backgrounds and experience, but also to nurture collaborative and democratic processes among staff members to recognize each other’s unique contributions to the care process.