Harder than you can imagine.

Decline in autopsy rates

The highest autopsy rates have been reported from a small proportion of hospitals (mainly academic hospitals); in general, it is likely that hospitals in which autopsies are not performed outnumber those in which autopsies are performed. It has been estimated that in the United States before 1970, autopsies were performed in 40% to 60% of all cases involving hospital deaths; in recent years, that number has decreased to approximately 5%.[1] Data from the United States National Center for Health Statistics (NCHS) have shown that the autopsy rate has decreased from 19.1% of all deaths in 1972 to 8.3% in 2003.[2] More recent data from an August 2011 NCHS Data Brief have shown a slight increase in the autopsy rate from 8.3% to 8.5% for 2007, with approximately half of the autopsies performed for deaths due to disease and half due to external causes.[3]

Autopsy rates in countries other than the United States have also decreased substantially. In Australia, the rate decreased approximately 50% between 1992 and 2003.[4] Data from the Danish National Institute of Health showed that in Denmark, the autopsy rate decreased from 45% in 1970 to 16% in 1990.[5] Data from Halifax, Nova Scotia, showed that the rate of clinical (ie, hospital) autopsies decreased from 30% in 1987 to 20% in 1999.[6]

Autopsies supply significant information

The present data suggest that, among the approximately 850 000 individuals dying in US hospitals each year,59- 60 a major diagnosis remains clinically undetected in at least 8.4% of cases (71 400 deaths). The data also suggest that approximately 34 850 of these patients might have survived to discharge had misdiagnosis not occurred, but this estimate depends on the accuracy of the designator of class I error. Although, this second number is more speculative, given the dependence of class I error estimates on methodological features of the primary studies, it can be considered in the context of the Institute of Medicine's estimates of 44 000 to 98 000 preventable deaths per year due to medical error.61 These latter estimates have been debated,62- 64 but the studies from which they were derived may not have detected many of the errors reported in our analysis.