OBJECTIVES: Health Technology Assessment (HTA) is the basis for drug reimbursement in many countries, however, it is not always clear how different factors are incorporated and weighted by HTA bodies. This research aims to develop a model that identifies the drivers of successful HTA submission in five European HTA agencies. METHODS: Single drug assessments conducted between January 2011 and April 2015 were extracted from a database of HTA evaluations. The HTA outcome (recommendation for National Institute for Health and Care Excellence (NICE), Scottish Medicines Consortium (SMC) and Zorginstituut Nederland (ZIN); ‘Amélioration du service médical rendu’ (ASMR) for Haute Autorité de Santé (HAS) and benefit rating for Gemeinsamer Bundesausschuss (G-BA)) and the clinical, economic and societal data underlying the assessment were collected. Factors influencing the outcome were determined by univariate and multivariate analyses. RESULTS: In total, 1019 HTAs were identified: 108 for G-BA, 464 for HAS, 68 for NICE, 218 for SMC and 69 for ZIN. Each model started with 63 explanatory variables, which were then refined by omitting variables that showed many missing values. A backward regression model was used to develop the models with the best fit. Models were chosen following several tests of goodness of fit: Akaike information criterion (AIC), likelihood-ratio, Wald-test and residual chi-square, resulting in country-specific models that predict factors with the best fit. CONCLUSIONS: Univariate and multivariate backward regressions were used to develop models for understanding the key drivers of HTA recommendations. In most countries, clinical factors appeared to be predictive factors for success. In countries performing economic evaluations, mainly economic factors appeared to be predictive factors. Societal factors were identified less frequently as predictive factors for recommendation. The difference in observations per country was a limitation for this research. Further research could expand the time horizon in order to increase the observations.