Background: The opioid epidemic is a public health crisis that disproportionately affects our unsheltered neighbors. Because medication-assisted treatment (MAT) is effective for preventing deaths from drug overdose and retention is associated with better health outcomes, there is a clear need for more research on factors impacting retention in care. Objective: The purpose of this study is to examine the relationship between attendance in counseling and retention in medication-assisted treatment (active prescription for buprenorphine) at three or more months for individuals experiencing homelessness. Methods: This retrospective cohort analysis examines the relationship between attendance in counseling and retention in medication-assisted treatment for three or more months for individuals experiencing homelessness being treated at a Federally Qualified Health Center (FQHC) and Health Care for the Homeless (330h) Program grantee in San Diego County, California. The Behavioral Model for Vulnerable Populations is the framework used for this study. The cohort included 306 adults experiencing homelessness who had one prescription for buprenorphine and participated in a medication-assisted treatment program between 2017 and 2019. Descriptive univariate, bivariate, and multivariable modeling was conducted in SPSS. A binomial logistic regression for the association between participation in therapy and retention at three months was used to test the hypothesis that counseling appointments are positively associated with higher retention rates for individuals on buprenorphine for opioid use disorder. Covariates included sociodemographics, polysubstance use, comorbid diagnosis categories and healthcare utilization. Results: The sample of 306 patients had a diagnosis of opioid use disorder and a prescription for buprenorphine. The sample was 64.4% men, almost exclusively white, and 35% unsheltered. Of the sample, 97 patients were retained at 3 months and 209 were not. Results from the logistic regression model showed that counseling appointments were positively associated with retention at three months (OR=1.59, p<0.001). Conclusion: Findings from this study inform future MAT program retention and advance knowledge in the field.