Antineuraminidase serum antibodies in natural influenza A and immunization with influenza vaccines

Parallel HI and virus-elution-from-erythrocytes-inhibition (a simplified method for titration of neuraminidase antibody) tests were used for examinations of 1117 blood serum specimens from 440 adults and children under study, 5250 single serum specimens from healthy subjects from birth to 65 years of age, 38 paired serum specimens from children who experienced influenza A/Texas/1/77 disease in the epidemic of 1979-1980, and 590 paired serum specimens from subjects immunized with influenza vaccines. In 7%-23% of influenza patients and immunized subjects antibody rise was observed to only one of the influenza A virus surface antigens, hemagglutinin or neuraminidase. The protective activity of antibody to influenza A virus neuraminidase was as good as that of antihemagglutinins. Both kinds of antibody interacted in protection against the disease. Antineuraminidase antibody was found to affect the decrease in severity of the infectious process in natural infection with influenza A. The formation of immunological memory in the system of synthesis of antihemagglutinins and antineuraminidase antibodies was shown to have features in common. The pattern of heterologous immune responses in immunized subjects and patients with influenza showed all antigenic varieties of neuraminidase N2 as well as neuraminidases N1 and N2 to share common cross-reacting determinants.     

Generalized Confidence Sets for a Statistically Indeterminate Random Vector

A problem is considered for the construction of confidence sets for a random vector, the information on distribution parameters of which is incomplete. To obtain exact estimates and a detailed analysis of the problem, the notion is introduced of a generalized confidence set for a statistically indeterminate random vector. Properties of generalized confidence sets are studied. It is shown that the standard method of estimation, which relies on the unification of confidence sets, leads in many cases to wider confidence estimates. For a normally distributed random vector with an inaccurately known mean value, generalized confidence sets are built up and the dependence of sizes of a generalized confidence set on the forms and parameters of a set of possible mean values is examined.