Both the article and study suggest what I believe is a major failing in modern medicine: We should be measuring risk-adjusted outcomes for hospital treatments, then publishing a rating for each hospital on a bell curve. With such an approach, patients will naturally gravitate towards the better hospitals and the poorer hospitals will have an incentive to improve their procedures.
Here's something to consider: According to the article, "In 1964... the median estimated age at death for patients in Matthews’s center was twenty-one years, seven times the age of patients treated elsewhere...After Warwick’s report came out, Matthews’s treatment quickly became the standard in this country." If Matthews's treatment had been less spectacular - say only twice the life expectancy of patients treated elsewhere - how much longer would it have taken for his treatment to become the standard? Perhaps never?
[1] https://www.nytimes.com/2016/12/14/business/hospitals-death-...