No evidence that patient choice in the NHS saves lives.

The Health and Social Care Bill 2011 has been framed to abolish direct parliamentary control and public accountability for the National Health Service (NHS) in England. In the face of enormous public opposition to the Bill, the UK Government stood down the legislative process between April and June, 2011. Prime Minister David Cameron used the temporary pause to advance the case for the Bill and argued “Put simply: competition is one way we can make things work better for patients. This isn’t ideological theory. A study published by the London School of Economics found hospitals in areas with more choice had lower death rates.” The study to which Cameron referred was a working paper by Zack Cooper and colleagues. However, contrary to Cooper and colleagues’ claims, their study did not show a causal inverse relation between patient choice and death rates. A statistical association is not the same as causation. As set out by Bradford Hill in his seminal paper, certain factors must be considered when determining whether a statistical association is likely to be causal: ”experiment” or study design, plausibility of intervention and outcomes, strength, consistency, specifi city, coherence, temporality, and quality of data. Cooper and colleagues’ study does not meet scientifi c standards. In the absence of evidence proving that competition improves health, Cooper and colleagues’ work should not be cited as scientifi c evidence in support of choice, competition, or the current market-oriented Health and Social Care Bill 2011. A revised version of the study, published in The Economic Journal, clarifi ed points of detail, but Cooper large comparative studies, one reporting data from two academic institutions and one from a multicentre community-based cohort, both noted—after many adjustments for case-mix and disease risk—substantially improved outcomes after surgery compared with radiation. The community-based analysis also recorded, as did Warde and colleagues, better out comes after either surgery or radiation than after androgen deprivation monotherapy. In both studies, diff erences between treatments were small for men with low-risk disease, and increased progressively as risk rose. Warde and colleagues have provided the strongest evidence to date that androgen deprivation therapy alone for men with high-risk prostate cancer is not adequate. These patients require an aggressive, multimodal approach incorporating prostate-directed local therapy. However, the crucial question—whether the optimum initial strategy should include radiation combined with androgen deprivation therapy, or surgery followed by selective radiation on the basis of pathological fi ndings and early biochemical outcomes— is still open. The defi nitive answer will only come through trials of men with high-risk disease randomly assigned to receive surgery or radiation as an initial treatment.