Editor's Choice

Demonstrating accountability, fostering trust

BMJ 2004; 329 doi: https://doi.org/10.1136/bmj.329.7463.0-g (Published 19 August 2004) Cite this as: BMJ 2004;329:0-g
  1. Jane Smith, deputy editor (jsmith{at}bmj.com)

    “Plants don't flourish when we pull them up too often to check how their roots are growing.” Tom Treasure approvingly quotes this statement from Onora O'Neill's 2002 Reith lectures on trust and accountability (p 424). O'Neill's point is that “the culture of accountability that we are relentlessly building for ourselves [may actually damage] trust rather than supporting it.”

    Treasure is commenting on a study of the learning curve of newly independent surgeons doing coronary artery surgery (p 421). This study shows that mortality in patients operated on by surgeons in the first four years of independent practice is similar to that in patients operated on by more experienced surgeons but that mortality did decrease from 2.2% in the first year of independent practice to 1.2% in the fourth. Treasure's point is that coronary artery surgery is done in high volumes and outcomes are “extraordinarily good.” He seems to be suggesting that rather than continue to pore over the figures on outcomes of cardiothoracic surgeons, for ever diminishing returns, other surgery now merits closer scrutiny, such as some cancer surgery, where operations are counted in single figures per hospital and mortality is 10-20%.

    Yet other articles in this week's issue show that even for cardiothoracic surgeons there's still work to do in providing outcomes data—and the business is complicated. The inquiry into paediatric cardiac surgery deaths in Bristol in the 1990s recommended that patients must be able to obtain information on the relative performance of hospitals and surgeons, and the Americans have been publishing data on individual surgeons for a while. On p 450 Bruce Keogh and his colleagues from the Society of Cardiothoracic Surgeons of Great Britain review this experience. The big risk is that surgeons will avoid high risk cases—something that is not easy to determine; the other argument against individual-level data is that the performance of a surgeon cannot be separated from that of his or her hospital. The data from UK cardiothoracic surgeons that have been collected since 1998 partly confirms this: in cases where mortality fell above the threshold for review it has usually been process and organisation that's at fault rather than technical surgical ability. But ensuring data quality is hard, and the compliers of the data have to decide on their aim: to show safety or help patient choice. UK cardiothoracic surgeons opted to reassure patients about safety, but that leaves the question of how to help patients make choices. In his editorial (p 413) Vipin Zamvar suggests that the next steps need to include outcomes beyond mortality, such as morbidity and long term survival benefit.

    So, in terms of trust and accountability, the message from this week's BMJ is that cardiothoracic surgeons have been trying hard to do something that remains inherently difficult. Now others need to follow their path.


    View Abstract