In Ireland, the National Centre for Pharmacoeconomics (NCPE) appraises the cost effectiveness of technologies in response to requests from the Health Service Executive (HSE).  A large number of reimbursement decisions are based upon the appraisal of company Health Technology Assessments (HTAs).  HTAs are conducted in accordance with existing agreed Irish HTA Guidelines.  These guidelines do not specify the requirement of Expected Value of Perfect Information (EVPI) analysis. OBJECTIVES: To examine the application of EVPI analysis as part of the formal HTA process. METHODS:  There is no set cost-effectiveness threshold in Ireland; however, technologies with ICERs > €20,000/QALY are less likely to be reimbursed.  This threshold was considered here. EVPI estimates (at €20,000/QALY) were determined directly from the PSA results of company economic models.  Estimates were scaled up to 10 year population EVPI (PEVPI) levels.   NCPE recommendations on reimbursement were recorded. RESULTS: The NCPE have estimated PEVPI values on nine company economic evaluations to date.  All evaluations were for newly licensed technologies; eight were pharmaceuticals and one was a diagnostic.  Two technologies had ICERs > €120,000 with PEVPI estimates > €20 million; reimbursement was not recommended.  Two technologies dominated the relevant comparators and one had an ICER < €10,000/QALY.  All PEVPI values were below €1 million.  Reimbursement of all three technologies was recommended.  The four remaining technologies had ICERs in the range of €21,000/QALY -€30,000/QALY; their PEVPI values ranged from about €1.5 million - €35 million.  Reimbursement was not recommended.  In two cases (original PEVPI values of €2.4 million and €35 million respectively) the manufacturer subsequently revised the price.  Reimbursement was then recommended.  There was no formal reanalysis of PEVPI.  CONCLUSIONS: To date, the formal analysis of PEVPI has not affected the decision to accept or reject technologies with ICERs lower and higher than €20,000/QALY respectively.