OBJECTIVES: Poor compliance and failure to persist with drug therapy are of potential economic significance. The objective of this study is to assess the impact of medication non-compliance and non-persistence on economic evaluations in osteoporosis. METHODS: A Markov microsimulation model with a lifetime horizon and a societal perspective was used to analyse the impact of non-adherence to biphosphonate therapy on costs (drug, disease and total), on outcomes (Quality-Adjusted Life Years and number of fractures saved) and on the cost-utility of biphosphonate therapy versus no treatment. Analyses were performed for caucasian women aged 70 years with a diagnosis of osteoporosis (t-score ≤ -2.5). The relationship between compliance and fracture efficacy were taken from published sources and drug cost was proportionate to compliance level. To model persistence to therapy, we assumed that women can stop therapy after 3 months, 6 months, 1 year, 2 years or 3 years. RESULTS: Full adherence to therapy resulted in a QALY gain of 0.0397, a 7.6% reduction in the number of fractures and a higher cost of €383 compared to no treatment. Lower compliance was associated with a decrease in QALY gain, a reduction in the number of fractures saved and a higher disease cost compensated by a lower drug cost. The cost per QALY gained for biphosphonate therapy versus no treatment increased progressively with decreasing compliance and was €9,653 €29,570 and €46,389 at 100%, 50% and 20% of compliance respectively. Realistic persistence assumption (with full compliance) resulted in a lower QALY gain (only 0.0165) and a higher cost per QALY gained (€12,479), CONCLUSIONS: This study indicated that non-compliance and non-persistence to osteoporotic therapy result not only in worsening health outcomes, but also in a significant change in cost-effectiveness. Therefore, the effects of non-compliance and non-persistence should be an integral part of economic evaluations in osteoporosis.