Under the banner ''Statistics for the Rest of Us,'' the Journal in 2007 published a critique of case series reports to determine whether surgical resection is effective in improving survival for patients with malignant pleural mesothelioma. 1 The article concluded,''We owe it to today's patients with mesothelioma to get it right. The epidemic is still rising in Europe and there will be many patients in the next 10 to 20 years. We owe it to the rest of the world where asbestos was less well controlled or remains uncontrolled. Equally important is this: we owe it to our own scientific integrity.''There have been some randomized controlled trials, 2-4 but it looks like business as usual-more uncontrolled case series with their inherent limitations.The criteria for observational data to be accepted as evidence have been well described: when an intervention visibly, quickly, and reproducibly alters the course of a disease, cause and effect may be accepted as evident. 5,6 Tension pneumothorax, empyema, cataract, hip fracture, leaking aneurysm, and aortic stenosis are familiar examples. Surgery for mesothelioma meets none of the criteria for acceptance on experience alone. Surgery is performed neither to avert imminent death nor for immediate benefit, nor is cure evident. 7 There are too many biologic and therapeutic variables to determine whether in truth death is delayed; it may even be hastened. Heterogeneity in the progression of the disease provides some patients who naturally survive long without operation, so in the absence of fair comparators it is not possible to confidently attribute the life of any individual patient with mesothelioma to an operation done a year or two previously.At the very least, we should have better estimates of what would be the expected survival of patients, comparable with those operated on, if cared for without surgery. Reports of surgical series have been presaged with gloomy statements that survival is likely to be as short as ''4 to 12 months,'' but From the