The "recovery rate" for cancer is another fraught area. As more and more screening is being conducted, they are picking up cancers that are smaller and much less rapidly growing. Basically, there are three categories of cancer: turbo-cancer that grows so fast that by the time you find it you're basically going to die in a year or so, treatment does little to nothing for this, normal speed cancer, which grows at a steady pace and is responsive to treatment, and slow cancer, which grows so slowly that you're likely to die of old age before it ever becomes a problem. Treatment for this type of cancer is pointless and the side effects of the treatment are often much worse than the cancer itself. Screening increases the detection of the last two categories. But, the mortality rate from the last category is very low. The issue is that there is no differentiation of these categories of cancer in the mortality figures, so the very slow growing cancers artificially increase the 5 year survival statistics. There's an interesting book on this, called Overdiagnosed, which goes into how the over use of screening tests is leading to unnecessary treatment for diseases, including cancer, that is resulting in an overall increase in all cause mortality. Effectively, they are treating people with drugs that might be appropriate if the condition they had was life threatening, but they really don't have a life threatening condition. So the iatrogenic harms are felt, but there is no benefit.