TABLE III. Cases of specified notifiable diseases, United States, weeks ending
March 26, 1988 and March 28, 1987 (12th Week)

TABLE III. (Cont'd.) Cases of specified notifiable diseases, United States, weeks ending March 26, 1988 and March 28, 1987 (12th Week)

*For measles only, imported cases includes both out-of-state and international importations N: Not notifiable U: Unavailable *International Out-of-state

TABLE III. (Cont'd.) Cases of specified notifiable diseases, United States, weeks ending March 26, 1988 and March 28, 1987 (12th Week)

*Mortality data in this table are voluntarily reported from 121 cities in the United states, most of which have populations of 100,000 or more. A death is reported by the place of its occurrence and by the week that the death certificate was filed. Fetal deaths are not included.

**Pneumonia and influenza.

† Because of changes in reporting methods in these 3 Pennsylvania cities, these numbers are partial counts for the current week. Complete counts will be available in 4 to 6 weeks.

†Total includes unknown ages.

§ Data not available. Figures are estimates haced on average of past 4 weeks.

Current Trends

Autopsy Frequency

United States, 1980-1985

In approximately 14% of the 2,089,378 deaths reported in the United States in 1985, an autopsy was performed. Recent reports indicate that the frequency of autopsy has been declining and that the decline may have adversely affected the accuracy of determining the underlying cause of death (1). To assess the recent variation in autopsy frequency, mortality data collected by the National Center for Health Statistics, CDC, for the period 1980-1985 were analyzed. During that time, the proportion of deaths involving an autopsy gradually declined from 17% to 14%. Within each year, however, autopsy frequency varied substantially by cause of death. For this analysis, cause of death was grouped into six general categories: natural causes, unintentional injuries and poisonings,* suicide, homicide, external causes with undetermined intent, and unknown or unspecified causes. These groups correspond to the codes in the International Classification of Diseases, Ninth Revision (ICD-9) (Table 1).

The proportion of autopsies performed ranged from 12% among natural deaths to 97% among homicide deaths (Table 1). Deaths with unknown autopsy status were enumerated separately and excluded from the calculations. Although 12% of all records lacked autopsy data, the proportion of records without autopsy data varied from 2% among homicide deaths to 13% among natural deaths. In general, larger autopsy percentages are associated with smaller percentages of missing data.

Autopsies for natural deaths declined at least 0.5% every year during the period 1980-1985, from 13% in 1980 to 10% in 1985 (Figure 1). In contrast, the frequency of autopsies for deaths caused by unintentional injuries and poisoning increased from 46% to 51%, and the frequency among suicide deaths increased from 48% to 52%.

*This report uses the phrase "unintentional injury and poisoning" because there is no summary term in the ICD-9 manual to refer to codes E800-E949.

*Death certificate does not indicate whether an autopsy was performed.
*Deaths with unknown autopsy status not included in denominator.
$Refers to external causes.