OBJECTIVES: To map factors that influence HTA-agencies in their Health Technology Assessments (HTA) on type 2-diabetes agents in the UK (NICE), Scotland (SMC), Netherlands (CVZ) Germany (IQWIG) and Sweden (TLV). METHODS: To retrieve the HTA reports, a search was executed using the agencies websites with the following keywords: pioglitazone, rosiglitazone, sitagliptin, vildagliptin, exenatide, glargine, detemir, aspart, glulisine and lispro. If a report contained several drugs each drug was counted separately although a decision could involve a class of drugs. Decision parameters were clustered in three categories: efficacy, safety and health economics where each assessment could contain multiple parameters. Overall recommendation was classified in three categories: recommended restricted recommended and not recommended in relation to indication based on marketing authorisation. RESULTS: 35 reports were identified with 49 assessments. Twelve assessments lead to recommendation (24%), 23 to restricted recommendation (47%) and fourteen to no recommendation (29%). Reasons for recommending a treatment contained in 83% of cases one or more arguments related to efficacy, 33% to safety, and 66% to health economic aspects of drugs. Reasons for restricted recommendation were 70%, 39%, and 60%, and for not recommended were 100%, 57% and 21% respectively. Within each decision parameter the most common reason for restricting the market authorization indication was related to the drug not being cost-effective (57%). The most common reason for not recommending a drug was lack of long term data on efficacy (86%). CONCLUSIONS: Despite that large variations in results between agencies were observed, data demonstrating efficacy of the drug appeared to be the most important factor in getting a recommendation for type 2 diabetes treatment. A high incremental cost-effectiveness ratio was likely to lead to restrictions in indication (NICE, SMC and TLV) whereas lack of long term data could lead to the drug not being recommended (IQWIG and CVZ).