New study questions the validity of publishing hospital mortality rates

A previous study of mortality rates for congenital heart surgery used routinely available hospital data that were misleading, according to a report published today on bmj.com which questions the validity of such data being made public.

Professor Westaby and colleagues found the system of information gathering used in the study had underestimated the number of infant deaths. In the previous BMJ study, published in 2004, Oxford had been singled out as having significantly higher mortality than the national average for open heart surgery on infants. Yet this new paper, using data from a different source - the Central Cardiac Audit Database - shows that the hospital’s mortality statistics were not actually different from the mean for all the centres (10 percent compared to 8 percent between 2000 and 2002).

The authors looked at a report from the ‘Dr Foster’ unit at Imperial College which was published in the wake of the inquiry into the Bristol congenital heart surgery deaths. That inquiry, which drew widespread publicity and had a profound effect on surgical practice in the UK, used Hospital Episode Statistics (HES) to compare mortality rates among cardiac surgical units across the country. The 2004 study by Dr Aylin described these mortality statistics.

The authors of the current study compared the mortality rates reported by the administrative HES database and an alternative system, the clinically based Central Cardiac Audit Database, for infants under 12 months undergoing cardiac operations. The statistics were gathered between 1st April 2000 and 31st March 2002.

They found HES did not provide reliable patient numbers or 30-day mortality data. On average HES recorded 20 percent fewer cases than CCAD and only captured between 27 percent and 78 percent of 30-day deaths, with a median shortfall of 40 percent.

In Centre A, with the largest number of operations, 38 percent of all patients were missed by HES and only 27 percent of the total deaths were recorded. Overall, mortality statistics were underestimated by 4 percent using HES data.

The authors say publication of inaccurate statistics detracts from public confidence and that: “If mortality statistics are to be released their quality must be beyond reproach.”

They acknowledge the media are keen to publish such statistics and pinpoint ‘Dr Foster’ who have pioneered this by providing newspapers with information on heart disease, for example, in return for a fee.

They conclude: “Given the problems with data quality, the imprecision of risk stratification models, and the confrontational agenda in the media, we question the value of placing mortality statistics in the public domain.”

Source: BMJ-British Medical Journal

Related stories:

Metabolic syndrome ups colorectal cancer risk
In a large U.S. population-based study presented at the 73rd Annual Scientific Meeting of the American College of Gastroenterology, metabolic syndrome patients had a 75 percent higher risk of colorectal cancer compared to those without metabolic syndrome.
Majority of osteoporosis patients not receiving calcium and vitamin D with treatment
New research published today at the annual meeting of the American Society for Bone and Mineral Research (ASBMR), Montréal, Canada, reveals that less than half (43%) of patients in Europe with osteoporosis are claiming to take both calcium and vitamin D supplementation with their osteoporosis treatment. Maximum benefit in managing osteoporosis can be achieved with combination therapy of an osteoporosis treatment (such as a bisphosphonate) with calcium and vitamin D supplementation, yet the majority of patients in this research claim they do not follow this approach.
Minorities less likely to know about breast cancer treatment options
Nearly half of women treated for breast cancer did not know that their odds of being alive after five years are roughly the same whether they undergo mastectomy or breast conserving surgery. Minority women were even less likely to be aware of this important factor of their treatment decision, according to a study from the University of Michigan Comprehensive Cancer Center.
Preterm birth contributes to growing number of infant deaths
Babies born too soon and too small accounted for a growing proportion of infant deaths, according to new statistics released today from the National Center for Health Statistics, (NCHS).
First worldwide analysis of cancer survival finds wide variation between countries
Cancer survival varies widely between countries according to a worldwide study published online today in Lancet Oncology.* More than 100 investigators contributed to the study.
Using biostatistics to detect disease outbreaks
An internationally recognised authority in biostatistics, Harvard University Professor Louise Ryan, is working with CSIRO researchers to help improve the way statistics are used to detect disease outbreaks in Australia.
Non-parental care of infants tied to unfavorable feeding practices
With more new mothers in the workplace than ever before, there has been a corresponding increase in the number of child-care facilities in the United States. At the same time, data from a variety of sources point to a growing prevalence of overweight infants and toddlers. Is there a connection?
Higher education associated with greater gains in mortality reduction from common cancers
Deaths due to the four most common cancers—lung, colorectal, prostate, and breast—have dropped substantially in the United States from 1993 to 2001 in working-aged individuals. However, not all Americans are equally likely to benefit from those gains. A study published in the July 8 online issue of the Journal of the National Cancer Institute shows that more highly educated individuals had mortality reductions in nearly all of these cancers, while less educated individuals had a mortality reduction in only one of the cancer types.

News discussion:

Medicine & Health news