OBJECTIVES: To apply value of information analysis (VoI) to a cost-effectiveness model of a Levonorgestrel intrauterine system LNG-IUS 13.5mg (total content) in order to identify parameters worth investigating to reduce remaining uncertainty around the reimbursement decision.  LNG-IUS 13.5mg is a novel low-dose hormonal intrauterine contraceptive system for up to 3 years use and was shown to be cost-effective over this horizon vs. commonly used short-acting contraceptive methods in women aged 20-29. METHODS: VoI analyses were performed using an existing US cost-effectiveness model with LNG-IUS 13.5mg. Expected Value of Perfect Information (EVPI) and the parameter EVPIs (EVPPI) for the following groups were estimated: probability of failure (unintended pregnancy (UP)); probability of discontinuation during year 1 and subsequent years; and follow-up consultation costs. Individual EVPPIs were calculated for probability of failure of injectable contraceptives and LNG-IUS 13.5mg and first year discontinuation probability of LNG-IUS 13.5mg. EVP(P)I were calculated using different willingness to pay (WTP) values for an UP and extrapolated to population values. RESULTS: The estimated EVPI was $69, $62 and $40 at WTP of $300, $400 and $700 per UP, respectively. EVPPI analysis suggested that probabilities of failure and first year discontinuation were the most influential, accounting for the majority of uncertainty of the reimbursement decision.  Population EVPI estimates suggested that further research in this area was potentially worthwhile. CONCLUSIONS: Comparative effectiveness research focusing on the failure rate of injectable contraceptives and the first year discontinuation rate of LNG-IUS 13.5mg is potentially worthwhile. Expected value of sample information will have to be compared to the actual costs of such research.