OBJECTIVES: Although self-administration “pen” devices offer certain advantages, their higher cost has prevented its wider dissemination in patients with diabetes. The aim of this study was to determine their cost-effectiveness, compared to conventional vial and syringe, in adult patients in Colombia. METHODS: We designed a cost-effectiveness Markov model, with annual cycles, five-year time horizon (with analysis at 3 and 10 years), 5% discount rate, and third party payer perspective (Colombian health system). Transition probabilities and utilities in quality adjusted life years (QALY) were obtained from a systematic review of the literature. Costs, in 2014 Colombian pesos (1 euro = 2,660 COP), were obtained locally from different sources (following the methodology proposed by IETS, the Colombian HTA agency). Cost-effectiveness threshold was three times per capita GDP (equivalent to € 17,547 per QALY). RESULTS: The total cost of treating a patient with a syringe and vial over five years would be € 1,002, while the use of pen device would represent € 1,226. In the same period, the patient would gain 3.1709 and 3.1849 QALY, respectively, representing an incremental cost effectiveness ratio (ICER) of € 15,302 per additional QALY gained. By extending the time horizon, the ICER lowers (€ 10,424 in 10 years). Different discount rates, from 0 to 12%, do not substantially alter the results. With a price reduction of 31% the pen device reaches the threshold of 1 per capita GDP. In the sensitivity analysis, the variables related to hypoglycemia (frequency and costs of treatment) are the ones that most modify the results. Cost-effectiveness, however, is maintained under different scenarios. CONCLUSIONS: Given the assumptions and limitations of this model, the pen-like devices for self-administration of insulin are cost-effective, compared with administration by vial and disposable syringes, in adult patients with diabetes in the Colombian context.