BMJ  2005;330:324 (12 February), doi:10.1136/bmj.330.7487.324-b

News

Cardiac mortality in children in Oxford hospital is not excessive

London

Fears that mortality in children who needed heart surgery in Oxford was excessively high—suggesting a possible repeat of the incidents at the Bristol Royal Infirmary—are unfounded, an investigation has shown.

Two UK studies on survival rates after heart surgery across the United Kingdom in children less than 1 year old, which were published in the BMJ last year, prompted the Department of Health to request an investigation by Thames Valley Strategic Health Authority.

The first study was an analysis of survival rates for 2000-1, which relied on returns to the central cardiac audit database ( BMJ 2004;328: 611-20[Abstract/Free Full Text]). It found no differences among the 13 UK centres.

A later study, however, which used hospital episode statistics for 1991-2002, found a death rate of 11% at Oxford Radcliffe Hospitals NHS Trust. This exceeded the 4% national average for the period ( BMJ 2004;329: 825-7[Abstract/Free Full Text]). Case mix could not account for the discrepancy, the authors said.

The latest investigation involved an in-depth comparison of the two sets of data and included the most up to date local and national outcomes figures for this group of patients.

Last week's report concluded, "Oxford does not have a mortality rate which is statistically significantly different from the national average for hospitals conducting open cardiac surgical procedures on children aged under 1 year."

On the basis of the central cardiac audit database returns, an 8% death rate was a more accurate reflection of national figures than the 4% rate in the hospital episode statistics, the report found. And it recommended that the database should be used for all future comparisons of outcomes in children's heart surgery in the NHS. The Department of Health now plans to develop the cardiac database further, said a spokesperson.

Both sets of data had their place, said the report, but it urged that every effort be made to ensure the compatibility of the different coding systems used to collect the figures.

Although satisfied that the investigation had dispelled fears about excessive death rates, Robert Sherriff, director of public health and health policy at Thames Valley Strategic Health Authority and chairman of the investigative steering group, told the BMJ that trying to compare the two sets of data had been a major challenge. "It would be a good thing if different research groups could work together. They are in a better position to compare and contrast their data than a strategic health authority," he said. (See p 319.)

---

Report of the Paediatric Cardiac Surgery Steering Group is available at www.tvsha.nhs.uk

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Relevant Article

Surgery for congenital heart conditions in Oxford

Bruce Keogh
BMJ 2005 330: 319-320. [Extract] [Full Text] [PDF]