Summary Background Administration of vaccines by needle-free technology such as jet injection might offer an alternative to needles and syringes that avoids the issue of needle phobia and the risk of needle-stick injury. We aimed to assess the immunogenicity and safety of trivalent influenza vaccine given by needle-free jet injector compared with needle and syringe. Methods For this randomised, comparator-controlled trial, we randomly assigned (1:1) healthy adults (aged 18–64 years) who attended one of four employee health clinics in the University of Colorado health system, with stratification by site, to receive one dose of the trivalent inactivated influenza vaccine Afluria given either intramuscularly with a needle-free jet injector (Stratis; PharmaJet, Golden, CO, USA) or with needle and syringe. Randomisation was done with a computer-generated randomisation schedule with a block size of 100. Because of the nature of the study, masking of participants was not possible. Immunogenicity was assessed by measurement of the hemagglutination inhibition antibody titres in serum for the three viral strains included in the vaccine. We included six coprimary endpoints: three strain-specific geometric mean titre ratios and the absolute differences in three strain-specific seroconversion rates. The immune response of the jet injector group was regarded as non-inferior to that of the needle and syringe group if both the upper bound of each of the three 95% CIs for the strain-specific geometric mean titre ratios was 1.5 or less, and the upper bound of the three 95% CIs for the strain-specific seroconversion rate differences was less than 10 percentage points. We used t test for group comparison. This study is registered with ClinicalTrials.gov, number NCT01688921. Findings During the 2012–13 influenza season of the northern hemisphere, we allocated 1250 participants to receive vaccination by needle-free jet injector (n=627) or needle and syringe (n=623). In the intention-to-treat immunogenicity population, all participants with two serum samples were included (575 in the jet injector group and 574 in the needle and syringe group). The immune response to Afluria when given by needle-free jet injector met the criteria for non-inferiority for all six coprimary endpoints. The jet injector group met the geometric mean titre criterion for non-inferiority for the A/H1N1, A/H3N2, and B strains (upper bound of the 95% CI for the geometric mean titre ratios were 1·10 for A/H1N1, 1·17 for A/H3N2, and 1·04 for B strains). The jet injector group met the seroconversion rate criterion for non-inferiority for the A/H1N1, A/H3N2, and B strains (upper bound of the 95% CI of the seroconversion rate differences were 6·0% for A/H1N1, 7·0% for A/H3N2, and 5·7% for B strains). We recorded serious adverse events in three participants, none of which were study related. Interpretation The immune response to influenza vaccine given with the jet injector device was non-inferior to the immune response to influenza vaccine given with needle and syringe. The device had a clinically acceptable safety profile, but was associated with a higher frequency of local injection site reactions than was the use of needle and syringe. The Stratis needle-free jet injector device could be used as an alternative method of administration of Afluria trivalent influenza vaccine. Funding Biomedical Advanced Research and Development Authority (BARDA), PATH, bioCSL, and PharmaJet.