OBJECTIVES: To develop and validate predictive models that identify members with higher risk of medication non-adherence and increased total healthcare cost over a 12-month period in a managed care setting. METHODS: The study included members insured under a commercial healthcare plan who filled ≥ 1 prescription for any of seven targeted medication classes for common chronic diseases between October 2010 and May 2014. Pharmacy and medical claims during the four months before and six months after the member's first prescription for a targeted medication (index date) were used to generate 85 baseline member variables. These variables were tested for potential model inclusion to separately predict medication non-adherence (proportion of days covered <80%) and total healthcare costs during the 12-month follow-up period. Total costs included pharmacy and medical costs from outpatient, emergency room, and inpatient visits. Members were randomized 3:1 to the development or validation samples. The development sample was used to estimate and refine model parameters. The validation sample was used to evaluate the final model's performance based on c-statistic and R-squared values. Medication non-adherence was predicted using a logistic model. Total healthcare cost was predicted via a generalized linear model with a log link function and gamma distribution. RESULTS: Among the 70,502 and 23,505 members included in the development and validation samples, respectively, baseline prevalence of medication non-adherence ranged from 37% to 73%, depending on the medication class. Baseline adherence and cost were the most important predictors. Predictive performance improved when other variables, such as member demographics and comorbidities, were added to the baseline adherence only model (c-statistic increased from 0.81 to 0.89; p<0.0001). The cost model's R-squared value was 0.43. CONCLUSIONS: The models demonstrated good predictive performance and could be used together to identify members with potential non-adherence to medications and greater healthcare costs for intensive clinical interventions.