OBJECTIVES: : Meta-analysis (MA) of time to event survival data are most commonly performed using the individual summary statistic hazard ratio from each study, as an appropriate measure of effect. Currently there is no clear guidance regarding alternative novel methodologies of evidence synthesis using survival data which violates the proportional hazards (PH) assumption. The aim of this study was to assess: (i) the guidance from HTA bodies in relation to the MA of time to event survival data; (ii) technology assessments (TAs) submitted to NICE to determine the level of supporting information relating to the PH assumption accompanying MAs of time to event data in manufacturer submissions, and the response of reimbursement authorities. METHODS: HTA authorities guidelines (NICE, PBAC, IQWIG, CADTH, NCPE) were searched to identify information relating to the MA of time to event data. The NICE website was interrogated to identify TAs and the associated ERG/FAD reports in the oncology setting (published 2011–2014) reporting MAs of time to event data. RESULTS: Of the guidelines searched, the NICE, PBAQ and IQWIG guidelines for evidence synthesis refer to the consideration of the proportional hazards assumption when performing MA of time to event survival data. Of the most recent 60 NICE TAs, seven included the analysis of time to event data, however none commented upon the PH assumption. CONCLUSIONS: The impact of failing to consider the validity of the PH assumption for MA of time to event data in manufacturer submissions is unclear. The failure of trialists and statisticians to investigate the validity of the PH assumption for time to event data used in evidence synthesis may result in clinical decisions based on inappropriate methods.