|
|
EA Sidman, NL Smith, LA Jackson, University of Washington,
GL. Burke, Wake Forest University, PHM Chaves, Johns Hopkins University,
TG Evans, University of California, Davis, AB Newman, University
of Pittsburgh, DS Siscovick, University of Washington
Excess mortality occurs during influenza outbreaks, and cardiovascular
diseases are a major contributor to the excess mortality not directly
attributed to influenza or pneumonia. Inflammatory responses to
infections may trigger cardiovascular events or promote atherosclerosis.
We examined whether influenza vaccination late in life was associated
with reductions in all-cause, coronary heart disease (CHD), and
non-CHD mortality in the Cardiovascular Health Study, a longitudinal,
population-based, study of 5,888 adults aged =65 years. Participants
reported their influenza vaccination status at baseline and annually
during follow-up. Data from 1993-1997 were used. Fatal events occurring
throughout follow-up and during periods in which influenza was and
was not expected to be circulating were assessed. Cox regressions
were carried out with influenza vaccination status modeled first
as a time-dependent variable, and then as a non-time-dependent variable
based on participants' joint 1992/1993 vaccination status. Using
the time-dependent method, vaccine recipients were approximately
15% less likely to die than were non-recipients after controlling
for demographic and health status differences (all-cause: HR = 0.84,
95% CI = 0.73 - 0.95; CHD: 0.81 [0.63 - 1.05]; non-CHD: 0.85 [0.73
- 0.99]). No associations were observed between time-dependent vaccination
status and all-cause or non-CHD mortality during influenza seasons.
In non-influenza periods, vaccine recipients were at lower risk
of all-cause and non-CHD mortality relative to non-recipients (all-cause:
0.72 [0.61 - 0.86]; non-CHD: 0.68 [0.55 - 0.84]). Lower risks of
CHD death were observed during influenza and non-influenza periods
but were non-significant. Non-time-dependent analyses comparing
participants vaccinated in both 1992 and 1993 with those unvaccinated
in both years were generally similar, except that stronger inverse
associations were observed between vaccination status and CHD mortality
(influenza seasons: 0.52 [0.31 - 0.86]; non-influenza periods: 0.67
[0.43 - 1.05]; combined: 0.60 [0.43 - 0.84]). Lower risks of mortality
associated with influenza vaccination late in life may reflect an
effect of vaccination on acute or chronic disease risk, or selection
bias if vaccination is a marker for other healthy behaviors.
View
Slides
|
