Guidelines recommend restrictive red blood cell transfusion strategies. We conducted an observational study to examine whether the rate of peri-operative red blood cell transfusion in the USA had declined during the period from 01 January 2011 to 31 December 2016. We included 4,273,168 patients from all surgical subspecialties. We examined parallel trends in rates of the following: pre-operative transfusion; prevalence of bleeding disorders and coagulopathy; and minimally invasive procedures. To account for changes in population and procedure characteristics, we performed multivariable logistic regression to assess whether the risk of receiving a transfusion had declined over the study period. Clinical outcomes included peri-operative myocardial infarction, stroke and all-cause mortality at 30 days. Peri-operative red blood cell transfusion rates declined from 37,040/441,255 (8.4%) in 2011 to 46,845/1,000,195 (4.6%) in 2016 (p < 0.001) across all subspecialties. Compared with 2011, the corresponding adjusted OR (95%CI) for red blood cell transfusion decreased gradually from 0.88 (0.86–0.90) in 2012 to 0.51 (0.50–0.51) in 2016 (p < 0.001). Pre-operative red blood cell transfusion rates and the prevalence of bleeding disorders decreased, whereas haematocrit levels and the proportion of minimally invasive procedures increased. Compared with 2011, the adjusted hazard ratios (95%CI) in 2012 and 2016 were 0.96 (0.90–1.02) and 1.05 (0.99–1.11) for myocardial infarction, 0.91 (0.83–0.99) and 0.99 (0.92–1.07) for stroke and 0.98 (0.94–1.02) and 0.99 (0.96–1.03) for all-cause mortality. Use of peri-operative red blood cell transfusion declined from 2011 to 2016. This was not associated with an increase in adverse clinical outcomes.