OBJECTIVES: This study aims to assess the predictors of medication adherence in Type II Diabetes Mellitus (T2DM) patients of Quetta city, Pakistan. METHODS:  The study was designed as a questionnaire based, cross sectional analysis. Three hundred T2DM patients attending public and private hospitals were targeted for data collection. In addition to demographic and disease related information, Drug Attitude Inventory and Michigan Diabetes Knowledge Test were used to assess medication adherence and diabetes related knowledge respectively. Treatment satisfaction was assessed by patient’s experience towards health care professionals and available facilities. Descriptive statistics were used to elaborate patients’ demographic and disease related characteristics. Binary logistic regression was used to predict factors independently associated with medication adherence. SPSS v. 20 was used for data analysis and p< 0.05 was taken as significant. RESULTS: Patients in the current study reported moderate adherence with mean score of 4.94 ± 2.72. Age, gender, education, diabetes-related knowledge and treatment satisfaction were significantly associated (p<0.05) with medication adherence. The created model showed a significant goodness of fit as the Omnibus Test of Model Coefficient was highly significant (Chi square =11.342, p = 0.001, df = 4). Knowledge score had significant association (adjusted OR= 2.232, 95 % CI = 1.345 – 1.766, P < 0.001) with medication adherence. An increase in knowledge score of one point was associated with an increase in being good adherence to a factor of 2.232 provided controlling other confounding variables. CONCLUSIONS: The study presents a model that is associated with medication adherence of patient with T2DM, where disease-related knowledge shaped as a predictor. The healthcare practitioners and system should formalize and acknowledge patient education as key component of therapeutic plans.