Can giving patients information such as individual hospital and surgeon death rates help them make better decisions about where to go for treatment? John Appleby reveals the problems with a US example

Published:25/04/2002, Volume II2, No. 5802, Page 29

It may seem axiomatic that patients - and society in general - would benefit from more and better information on the performance of health services.

And in particular, information which helps patients to choose the consultant/hospital likely to maximise a desired outcome (such as survival) or a bundle of 'outcomes' (which would allow patients to trade off aspects of their treatment and its clinical outcomes to suit their own needs or wants).

The latest coronary artery bypass graft (CABG) surgery report card publication from the New York State Department of Health 1, for example, argues that the process of reporting risk-adjusted mortality data (the number of deaths) by hospital and surgeon has been instrumental in achieving 'the excellent outcomes that are evidenced in this report...' But a recent US study that looked at the impact of the New York and Pennsylvania states' CABG surgery performance report cards suggests patient information may not be as beneficial as expected.

2The authors point out that public disclosure of patient outcomes may address important information asymmetries in markets for healthcare which, importantly, can lead to competitive pressure on poor-performing providers to improve.

But such pressure may lead to changes in provider behaviour which would be undesirable from the points of view of patients and society.

Though the research found that report cards led to better 'matching' of patients to providers (that is, the 'best' hospitals treated greater proportions of sicker patients than before), it also found the introduction of report cards led to providers tending to select healthier patients relative to trends in other states.

Apart from this adverse selection, publishing performance data also appeared to increase the volume of CABG surgery - but only for healthier patients.

Further, after report cards were introduced, substitute revascularisation procedures such as angioplasty reduced in numbers for both sick and healthy patients. Overall, the research suggests that report cards led to higher healthcare spending on healthy and sick patients - but with no discernible health benefits to healthy patients and 'adverse' consequences for sicker patients.

The researchers say that the findings are based on only a few years' data and that in the longer run reporting on performance may have positive benefits.

Moreover, with more sophisticated systems to minimise gaming by providers, report cards could be made more beneficial.

How this applies to the NHS as it starts to pursue opportunities for patient choice and respond to demands for more information for patients is hard to say. Will clinicians improve their performance through adverse selection? This may at first sight seem difficult within the NHS, but choosing where to work on the basis of a local population's general health state is not unheard of in the UK.

In any case, with decisions on many patients' treatments as much an art as a science, clinical thresholds (to admit, to operate, type of procedure etc) could change in ways which improve (statistical) performance but leave patients little better off. l REFERENCES 1CABG mortality rates in New York State:

www. health. state. ny. us/nysdoh/consumer/heart/homehear. htm 2Dranove D, Kessler D, McClellan M, Satterthwaite M. Is More Information Better? The effects of 'report cards' on health care providers. NBER Working Paper No. w8697. National Bureau of Economic Research, Cambridge, Massachusetts, 2002.