Surveillance for laboratory confirmed influenza
» Background
Influenza remains the most common infectious disease causing
death and hospitalization in the developed world. Because diagnostic
tests have been both unavailable and underutilized, estimates
of deaths due to influenza rely on projections from population
statistics and known disease activity. The absence of diagnostic
tests, and the fact that many hospitalizations and deaths are
due to the complications of influenza rather than influenza itself,
means that the impact of influenza is often underestimated by
clinicians. Previous studies have identified that less than 50%
of influenza related hospitalizations are captured by the diagnostic
categories including influenza and pneumonia, and <10% by the
diagnostic category influenza.
The advent of routine rapid testing for influenza, the availability
of new drugs for its treatment, and the attention focused on viral
respiratory illnesses in the wake of SARS all mean that there
has been a substantial increase in diagnostic testing for respiratory
viruses, and a new appreciation for the impact of viral respiratory
illnesses.
We hypothesize that the changes noted above will mean that there
will be a dramatic increase in influenza testing and diagnosis
in the next 1-3 years, and that this will be associated with a
new appreciation of the impact of influenza. Our goal is to provide
clinicians in south-central Ontario with a description of the
clinical features and outcomes of serious influenza-related illness
in their population, and to determine whether there are laboratory
or hospital-specific differences in the use of influenza diagnostic
tests.
» Methods
From July 1, 2004 to June 30, 2007, hospitals in Toronto, Peel
Region, Durham, Halton, Hamilton, Simcoe and York will perform
surveillance for hospital admission associated with the isolation
of influenza, or a rapid test positive for influenza from any
diagnostic specimen. The microbiology laboratories serving these
and the three largest private laboratories serving nursing homes
and physician's offices telephone the central study office whenever
an isolate of influenza is identified from an in-patient unit
or the emergency department. The patient will be screened to identify
whether or not they were admitted to hospital. Annual audits of
cultures and rapid influenza testing from all laboratories will
be performed to evaluate reporting accuracy. In addition, reported
cases will be compared to culture results from the Ontario public
health laboratory, which performs viral cultures for most hospital
laboratories.
Initial demographic data (date of birth and postal code of residence)
are available from the reporting laboratory. When a microbiology
laboratory telephones the central study office and the case is
identified as eligible, a "tracking record" is initiated
on which is recorded the date, reporting lab, patient initials,
study code number, hospital, physician name and telephone number.
The office notifies the study nurse responsible for the hospital
involved, and the patient is approached for consent to participate.
If the patient refuses, their name is removed from the tracking
record. If the patient agrees to the study, data will be collected
by interview and chart review. The data collection and consent
forms are attached. For subjects incompetent to consent, consent
will be obtained from the substitute decision maker. The consent
may be verbal, if the patient has been discharged (or was not
admitted), or if the substitute decision maker is not at the hospital.
If verbal consent is obtained, an information letter will be provided
for the patient/substitute decision maker (either mailed, or left
with nursing staff to be picked up), to explain the study, and
provide contact information. For nursing home residents, the facility
administration will also be contacted to inquire as to whether
vaccine may have been administered as part of an institutional
program. Where no substitute decision maker is available, or in
cases for which the patient or family cannot be contacted, clinical
data will be collected by chart review only.
» Analysis
This is primarily a descriptive study. Data will be provided
annually to hospitals on the number of admissions diagnosed with
influenza, and a summary of the clinical features of these illnesses.
For those patients resident in metropolitan Toronto and Peel region,
influenza vaccine efficacy will be assessed by comparing rates
of vaccination in patients with population rates of vaccination
in age group and illness matched cohorts being obtained from another
study. For comparisons of utilization of tests, laboratories will
be asked annually to provide the number of influenza tests ordered
(from in-patient and emergency departments), and hospitals will
be asked, if possible, to provide the total number of patients
admitted with a diagnosis of influenza or pneumonia (ICD10 codes
480-487). The rate of hospitalization due to influenza detected
in this surveillance will also be compared to the rates determined
by traditional models of hospitalization rates and disease activity
(project currently being undertaken by collaborators) in the population
area.
