Assessing trends in mortality in 121 U.S. cities, 1970-79, from all causes and from pneumonia and influenza.

The Centers for Disease Control receives weekly reports of mortality due to all causes and to pneumonia and influenza from 121 cities and counties in the United States. To assess the epidemiologic applicability of these data, the trends of death rates based on data compiled by the Centers for Disease Control's mortality reporting system (CDC-MRS) from 1970 through 1979 were compared with trends derived from national mortality statistics compiled by the National Center for Health Statistics (NCHS). In general, CDC-MRS trends in death rates from all causes and from pneumonia and influenza followed patterns similar to those shown by mortality statistics for the entire nation. CDC-MRS data were particularly sensitive to annual fluctuations in the nationwide rate of death from pneumonia and influenza among the elderly population. However, because of higher death rates among residents of the CDC-MRS reporting areas, in addition to other ascertainment biases, CDC-MRS death rates--from all causes and from pneumonia and influenza--consistently exceeded NCHS rates for the nation. Moreover, for each age group, trends based on CDC-MRS reflected an underestimate of the rate of decline in mortality observed over time according to NCHS data. It is concluded that despite its limitations, the CDC-MRS provides mortality data that are both timely and useful for epidemiologic purposes.

The low risk of hepatitis B in rural hospitals. Results of a seroepidemiologic survey.

To determine the risk of hepatitis B virus infection for rural hospital employees, we obtained serum and a completed questionnaire from each of 2,064 employees of 11 rural hospitals. Only 96 (4.7%) employees had hepatitis B virus markers. Increased marker prevalence was significantly associated with prior residence in a city with a population of greater than 100,000 (odds ratio, 2.9; 95% confidence interval, 1.6 to 5.2) and increasing blood contact (odds ratio, 1.6; confidence interval, 1.2 to 2.3); however, the association with blood contact was not significant when we limited analysis to the 836 employees who had never lived in a city. We conclude that the risk of hepatitis B virus infection for these rural hospital employees is low, probably because the incidence of hepatitis B in rural areas is low.