Traditional value of medical innovation models estimate the beneﬁts of a new treatment to individuals who fall sick ex-post. However, economic theory suggests medical innovation has additional beneﬁts ex-ante - before susceptible individuals know whether or not they will fall sick. This paper presents a model for estimating the ex-ante value of new medical innovation, where risk-averse individuals who are susceptible to contracting a disease in the future derive value from the existence of a treatment, because risk-averse individuals prefer to insure against uncertainty in future survival outcomes. A healthy, risk-averse individual should value a new treatment for a disease and be willing to pay a small premium for it, even if he never uses it because of his future risk of the disease. I demonstrate that when individuals are risk-averse, the total ex-ante value is always greater than the total ex-post value. The diﬀerence between ex-ante and ex-post value is the insurance value of medical innovation due to risk aversion. I show that insurance value is highest for rare diseases, where aﬄicted patient populations are small, but vulnerable patient populations are large. For very rare diseases, up to 64 percent of the value of a cure is ex-ante insurance value. Using population-level estimates of disease risk for cancer, I ﬁnd that between 24 and 63 percent of the value of cancer-speciﬁc cures are comprised of ex-ante insurance value - estimates vary by the prevalence and mortality risk of each cancer type. These results suggest there will be under-provision of valuable innovation by the private biopharmaceutical market if only ex-post valuations are reimbursed by payers.