What are my chances, doc?

Abstract

I have only been successfully sued once in my journalistic career—over a story about hospital death rates. I had obtained details of death rates after general surgery across the country that showed a six-fold variation and I produced a league table that appeared in the newspaper for which I then worked. Unwisely, as it turned out, I made critical remarks about the areas with the worst rates, on the basis of a report suggesting that extra deaths could indicate incompetent surgeons or poor nursing care. The named hospitals took umbrage and, on the advice of the paper's lawyer, the dispute was settled quickly. That was over 15 years ago. 5 years later, in 1999, the UK Government released the first hospital death rates for England. Last year, a decade later, a new boundary was crossed when the Government announced publication of death rates for every hospital on the National Health Service (NHS) Choices website. 60 years after it was founded, NHS patients could at last find out what their US cousins had known for years: how likely they were to leave hospital alive. Death rates are used in many countries to measure quality of care; they are also used globally to compare the performance of countries. But some argue that they paint too crude a picture and should be replaced with other outcome measures, such as morbidity and disability. How times change. In the early 1990s, anyone who had the temerity to question variations in outcomes risked attracting the attention of m'learned friends. Today, we seek more sophisticated and accurate measures, the better to identify those variations. The publication of death rates changed the terms of the debate. But as quality measures they never really cut the mustard. Roger Boyle, the UK Government's National Director for Coronary Heart Disease, explained why years ago. He described his experience with his father who was a high-risk heart patient: “I knew the outcomes for all the heart surgeons in the area but I did not choose the one with the lowest mortality. I went to the surgeon who had most experience in dealing with my father's particular problem.” This is the challenge—how to put the average patient (and their family) in the position of the expert—and raise standards at the same time. We need measures that have credibility with practitioners and intelligibility for the rest of us. Death rates may have served their purpose, but we should be cautious before jettisoning simple measures.