Background: The influence of misconceptions and related socio-demographics on metformin use could hamper adherence to medications. This study aimed to assess the rates and causes of metformin nonadherence and to investigate knowledge, attitudes, and practices (KAP) on misconceptions of metformin use including the association with socio-demographic variables.Methods: An observational analytical cross-sectional study was conducted at the diabetes clinic of Karapitiya Teaching Hospital in Galle, Sri Lanka. Causes of metformin non-adherence, associations with sociodemographics, and KAP on misconceptions on metformin use were assessed using the chi-squared test, ttests, and ANOVA using IBM SPSS Statistics for Windows, Version 26.0 (Released 2019; IBM Corp., Armonk, New York, United States) (p<0.05).Results: Metformin non-adherence was reported as 55%. Use of complementary and alternative therapies was 14.7%. Fear of major organ failure was the commonest (20.5%) reason quoted within the non-adherence group (N=223). Socio-demographic factors like ethnicity, lower education, unemployment, use of complementary and alternative therapies, and obtaining medications for other diabetes-related diseases significantly influenced adherence to the metformin-prescribed doses (p<0.05). Among all participants (N=400), the most common misconception was that long-term use of metformin caused organ damage (kidney 72.5%, liver 64.3%, and heart 34.8%), while 44% believed higher doses (two tablets or more for a day) caused organ damage. The KAP scores were reported as 24.5% with low, 52.7% moderate, and 22.7% satisfactory levels. Significantly lower KAP scores were associated with lower education levels and patients obtaining complementary and alternative therapies (p<0.05). Conclusion: Misconceptions are not merely kept in mind but lead to non-adherence with metformin doses prescribed and warrant evidence-based educational interventions with the high-risk groups.