To evaluate the effectiveness, as well as the cost-benefit of a hospitalwide physician educational programme in minimizing the spinal anesthesia (SA) failure rate. METHODS: A bupivacaine physician educational programme was initiated in June 2011 to minimize the SA failure rate in the Anesthesia Department of a tertiary-care hospital in Central Taiwan. We then used the patients recorded in the institution who underwent SA from January 2010 through December 2011, to evaluate the effectiveness as well as the cost-benefit of the programme, taking into consideration the cost of the training. To minimize potential impact of seasonal change, we compare two patient cohorts, the pre-training and the post-training cohorts, that were collected during July to December in 2010 and 2011, respectively. Logistic regression model was used to evaluate how well the application of the programme will predict SA failure rate, after controlling for potential impact of seasonal change. From the payer's perspective, the cost-benefit of the training was evaluated by calculating the cost needed under the SA failure rates before and after the training, based on the true number of patients undergoing SA after the training. The approved payments from the Bureau of National Health Insurance were used to do the calculation. RESULTS: There were 1841 and 2372 patients undergoing SA in the pre-training and post-training periods, respectively. The SA failure rate reduced from 2.66% to 1.39% in the two periods. The training is a predictor of lower SA failure rate (OR 0.57; 95% CI 0.41-0.79, pϭ0.001). A cost reduction of 3.13% was noted, indicating the cost-benefit of the training. CONCLUSIONS: The physician educational programme was associated with a significantly lower SA failure rate, with a RRR of 43%. Moreover, it had a significant cost-benefit profile from the payer's perspective.