Surgical morbidity is a significant public health issue worldwide. It is estimated that >230 million surgical procedures are performed each year, with an estimated mortality of at least 0.4% and morbidity of between 3% and 17%. Furthermore, there are potentially far-reaching consequences of a complicated perioperative course, because perioperative morbidity is associated with reduced long-term survival. In this review, we examine the factors that are associated with surgical outcomes. Issues related to the delivery of health care, such as structure, process, and resource utilization, have been shown to vary within and between institutions, leading to differences in both morbidity and mortality after surgery. Patient-related factors, in particular comorbid illness, functional capacity, and cardiovascular health, are also related to perioperative risk, and may be assessed using risk stratification models, exercise testing, and biomarker assays. The strengths and weaknesses of each of these techniques are discussed. We also review the strengths and limitations of the measures used to assess outcome after surgery, including patient-centered variables such as mortality and morbidity scores, and patient-related outcome measures. Finally, we suggest the direction of future work, which should be aimed at improving the precision of tools for describing perioperative risk, and of the measures used to assess the outcomes and quality of surgical health care. These tools are the building blocks of high-quality clinical trials, epidemiological studies, and quality improvement programs.