Non-pharmaceutical interventions may quell influenza pandemic spread

Issue: September 2007

During the 1918-1919 United States influenza pandemic, nonpharmaceutical interventions reduced the spread of the influenza pandemic, and health officials today may want to use this historical lesson as a teacher in preparing for the next pandemic, according to CDC researchers.

“In planning for future influenza pandemics, nonpharmaceutical interventions should be considered for inclusion as companion measures to developing effective vaccines and medications for prophylaxis and treatment,” researchers wrote in a recent study published in The Journal of the American Medical Association.

“Non-pharmaceutical interventions appear to have had a significant effect in mitigating the 1918 pandemic in the United States, however for these interventions to work, one must act early before the epidemic gets well established; applying a layered approach is better than putting all your interventions in one basket; and the measures must be sustained through the peak of transmission or until a vaccine can definitively rescue a population,” Martin Cetron, MD, of the division of global migration and quarantine of the CDC, told Infectious Diseases in Children.

For a “Category 5 Severe Pandemic,” the CDC’s most recent interim community mitigation guidance recommends activation of nonpharmaceutical interventions when a pandemic virus outbreak is first confirmed in a state or metropolitan region.

1918-1919 influenza data

The researchers studied recorded data of the 1918-1919 influenza pandemic in the United States and assessed the nonpharmaceutical interventions implemented in 43 of the largest cities during this time.

Researchers examined mortality data from the U.S. Census Bureau’s Weekly Health Index, which contains the most complete weekly pneumonia and influenza mortality data in United States urban areas during the 1918-1919 influenza pandemic.

Either the day of the first reported influenza case in a particular city or the calendar day of the first recorded influenza death minus 10 days were estimated as the report of the first case of pandemic influenza, while excess weekly P&I death rates above 2 times the baseline was defined as the acceleration phase in pandemic related mortality.

Nonpharmaceutical interventions were noted by a review of at least two daily high-circulation newspapers for each city at the time of the pandemic.

These interventions were then grouped into three major categories:

School closure.
Public gathering bans.
Isolation and quarantine.

The researchers also considered a category of subsidiary nonpharmaceutical interventions such as altered work schedules, closure or regulation of businesses, face mask ordinances and transportation restrictions, though these ancillary measures were not included into the analysis.

“Based on an estimated 10-day time between disease onset and death, we estimated that the association of nonpharmaceutical interventions with reductions in estimated death rates occurred 10 days after their actual date of implementation,” the researchers wrote in the study.

An analysis of variance model was constructed to test the association of the timing of nonpharmaceutical interventions with mortality.

The dependent variables were the weekly estimated death rates; and independent variables were the epidemiological week, city and status of every combination of nonpharmaceutical interventions.

Nonpharmaceutical intervention

Researchers noted 115,340 influenza deaths among the 43 cities over a 24-week period.

At least one of the three interventions was adopted by every city and 15 cities cohesively applied all three categories together.

At some point during the study period, 15 cities (35%) implemented isolation or quarantine interventions alone; 22 (51%) closed schools alone; and six cities (14%) banned public gathering alone.

The combined school closures and public gathering bans represented the most common combination among 34 cities (79%) with an average duration of four weeks; and 40 cities (93%) implemented school closings in combination with other nonpharmaceutical intervention categories.

Twenty-five schools reported closing their schools once; 14 were closed twice; and one school in Kansas City, Mo., closed three times. On average, schools were closed for six weeks.

New York City, New Haven, Conn. and Chicago never officially closed their schools, but later reported a student absenteeism rate of 45% or more at the peak of the epidemic, according to the researchers.

Those cities that implemented interventions earlier had greater delays in reaching peak mortality, lower peak mortality rates and lower total mortality than other cities.

“Physicians should be prepared to educate their patients about the importance of staying home when they are sick. They should prepare families for home care of those loved ones with pandemic flu who are not critically ill. This home care should be delivered in a safe way that prevents household spread,” Cetron said.

This study was a joint collaboration of the CDC with the University of Michigan. – by Jennifer Southall

For more information:

Markel H, Lipman HB, Navarro JA, Sloan A, Michalsen JR, Stern AM, Cetron MS. Nonpharmaceutical interventions Implemented by U.S. cities during the 1918-1919 influenza pandemic. JAMA. 2007;298:644-654.