BMJ  2003;326:1397-1398 (21 June), doi:10.1136/bmj.326.7403.1397-b

Letter

What hospital mortality league tables tell you

EDITOR—Jacobson et al take issue with Dr Foster's publication of mortality league tables and pose four main questions.¹

Firstly, they ask what the data mean, citing changes in the type of care provided in hospitals. Hospital standardised mortality ratios were designed to be a robust measure of in-hospital mortality, taking into account differences in patient mix with explicit adjustment for 80 different diagnoses, admission method, age, sex, and length of stay. We found that after adjustment, death rates show no bias against hospitals with more patients staying more than 28 days, or against hospitals with more geriatric beds.² We agree that hospital mortality is only one outcome measure and is included as such in the "wealth of information" in the hospital guide.

Secondly, they ask whether the results are a valid measure of what they purport to be, acknowledging Dr Foster's ongoing enhancements to the quality of the analysis since publishing hospital standardised mortality ratios for the first hospital guide. They acknowledge that rankings published last year cannot be directly compared to this year's but not that comparable trends in hospital standardised mortality ratios over the past three years were included in the 2003 Good Hospital Guide (perhaps a consequence of submitting their critique before the guide was published).

Thirdly, they question primary diagnosis, and we emphasise that the hospital standardised mortality ratios is a summary measure incorporating 80% of in-hospital deaths. Misclassification between different diagnoses will not greatly affect the overall figure. Hospital standardised mortality ratios based on primary diagnosis on discharge or primary diagnosis on admission are highly correlated (r=0.96).

Finally, they ask what value the data would add to hospital performance were they accurate. We acknowledge that many factors contribute to hospital mortality and we adjust for several of them. We found in our original analysis that other factors including socioeconomic deprivation did not account for the variability in hospital mortality.³ To help trusts explain their hospital standardised mortality ratios, Dr Foster offers a more detailed analysis, allowing trusts to drill down by diagnosis, admission method, age, sex, year, and length of stay.⁴

We strongly agree that there must be openness about clinical performance.⁵ Patients should be able to gain access to information about the relative performance of a hospital or a particular service or consultant unit.⁵ We believe that we are taking major steps towards fulfilling this aim.

Dr Foster Unit at Imperial p.aylin{at}imperial.ac.uk Department of Epidemiology and Public Health, Imperial College London, St Mary's Campus, London W2 1PG

Dr Foster Unit at Imperial b.jarman{at}imperial.ac.uk Department of Epidemiology and Public Health, Imperial College London, St Mary's Campus, London W2 1PG

Dr Foster Limited, Sir John Lyon House, London EC4V 3NX

---

Competing interests: The Dr Foster Unit at Imperial is funded by Dr Foster Limited.

References

1. Jacobson B, Mindell J, McKee M. Hospital mortality league tables. BMJ 2003;326: 777-8. (12 April.)[Free Full Text]
2. Jarman B, Bottle A, Aylin P, Taylor R. Mortality rates. Dead confusing. Health Service J 2002;112: 26-7.
3. Jarman B, Gault S, Alves B, Hider A, Dolan S, Cook A, et al. Explaining differences in English hospital death rates using routinely collected data. BMJ 1999;318: 1515–20.[Abstract/Free Full Text]
4. Jarman B, Cook A, Taylor R. Performance league tables [electronic response to Howell et al. Performance league tables]. bmj.com 2002. bmj.com/cgi/eletters/324/7336/542#20565 (accessed 3 June 2003).
5. Bristol Royal Infirmary Inquiry. The inquiry into the management of care of children receiving complex heart surgery at the Bristol Royal Infirmary. www.bristolinquiry.org.uk (accessed 18 Apr 2003).

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

Relevant Articles

Hospital mortality league tables

Bobbie Jacobson, Jenny Mindell, and Martin McKee
BMJ 2003 326: 777-778. [Extract] [Full Text] [PDF]

Explaining differences in English hospital death rates using routinely collected data

Brian Jarman, Simon Gault, Bernadette Alves, Amy Hider, Susan Dolan, Adrian Cook, Brian Hurwitz, and Lisa I Iezzoni
BMJ 1999 318: 1515-1520. [Abstract] [Full Text] [PDF]