DENVER – Children exposed to secondhand tobacco smoke who are admitted to the hospital for influenza are more likely to require admission to the intensive care unit and have a longer hospital stay than their peers who are not exposed to secondhand smoke.
These effects are even greater for children with chronic illnesses who are exposed to secondhand smoke, Dr. Karen M. Wilson reported at the annual meeting of the Pediatric Academic Societies.
Dr. Karen M. Wilson
An estimated 18% of children aged 3-11 years are regularly exposed to secondhand tobacco smoke inside the home, said Dr. Wilson, assistant professor of pediatrics at the University of Rochester (N.Y.).
Although secondhand smoke exposure is associated with worse outcomes for children’s illnesses, including respiratory syncytial virus and asthma, "the effect of secondhand smoke exposure on influenza severity in children is unclear," she noted. "More than 40% of preschool children experience influenza at some point. In adults, tobacco smoke increases the risk of influenza infection and the risk of complications."
To determine if children hospitalized with influenza who are exposed to secondhand smoke have more severe illness, Dr. Wilson and her associates conducted a review of 169 medical charts at Golisano Children’s Hospital in Rochester. They generated a list of patients aged 0-15 years with a discharge diagnosis of influenza between 2002 and 2009. The influenza diagnosis was verified by laboratory review.
Measures of severity included intensive care unit admission, defined as admission or transfer to the ICU at any time during the stay; need for mechanical ventilation, defined as any documentation of endotracheal intubation during the stay; and length of stay.
Exposure to secondhand smoke was assessed by any documentation of presence or absence of secondhand smoke exposure by any provider. "Any documentation of exposure was considered exposed; documentation of no exposure was considered not exposed," Dr. Wilson said.
She reported findings from 113 children who were included in the final analysis. Of these, 46 (41%) were exposed to secondhand smoke and 67 (59%) were not. The average age of the 113 children was 4 years, and 50% were male. Of the 113 children, 58% were white, 22% were black, 8% were Hispanic, and 3.5% were Asian; race/ethnicity was unknown in the remaining 8.5%. Fewer than half of the children (44%) had public health insurance.
More than three-quarters of the children (78%) had influenza A. In addition, 25% had asthma, 25% had an underlying chronic condition, 14% had documentation of prematurity, 19% required ICU care, and 6% required mechanical ventilation.
None of the potential covariates – including asthma, prematurity, and chronic conditions – were significantly associated with secondhand smoke exposure. However, children exposed to secondhand smoke were significantly more likely to require ICU admission (31% vs. 10% for children with no exposure) and mechanical ventilation (13% vs. 2%, respectively).
The mean length of stay was 2.1 days for children who had no chronic condition or exposure to secondhand smoke, 2.5 days for children who had no chronic condition but had exposure to secondhand smoke, 3.5 days for children who had a chronic condition but no exposure to secondhand smoke, and 11 days for children who had a chronic condition and were exposed to secondhand smoke.
In a logistic regression model controlling for age, gender, race, and type of insurance, exposure to secondhand smoke was significantly associated with ICU admission but chronic conditions were not.
In a logistic regression model limited to exposure to secondhand smoke and chronic conditions, chronic conditions were associated with the need for mechanical ventilation but exposure to secondhand smoke was not.
In a negative binomial regression model using the log-transformed length of stay, controlling for age, gender, race, insurance type, and chronic conditions, exposure to secondhand smoke was associated with an increased length of hospital stay, with an incident rate ratio of 1.9.
Dr. Wilson acknowledged certain limitations of the study, including its single-center design "and the potential for errors in documentation and abstraction," she said. "The exposure measure was reliant on provider documentation ... but provider documentation is more likely to underestimate secondhand smoke exposure in children, so we probably misclassified some children as being non–smoke exposed."
In addition, "there may be other covariates that we were not able to measure because we don’t have documentation in the chart," she said.
Despite such limitations, Dr. Wilson said that the findings support the notion of considering secondhand smoke exposure in risk stratification for children admitted with influenza. "Greater efforts are needed to help parents eliminate their children’s exposure to secondhand smoke," she said. "Parents of children with chronic illness should be aware of the risk of secondhand smoke exposure, and children exposed to secondhand smoke should be a priority group for influenza immunization."