Masks plus hand hygiene reduced ILI in college dorm study

Jan 22, 2010 (CIDRAP News) – College students living in dorms reduced their risk of influenza-like illness (ILI) at the peak of the flu season by wearing surgical masks a few hours a day and practicing good hand hygiene, say researchers from the University of Michigan.

In the last 3 weeks of a 6-week intervention in 2007, students who wore masks and had hand hygiene training lowered their rate of ILI by a significant 35% to 51% compared with a control group, says the report, published online by the Journal of Infectious Diseases.

A second group that used face masks alone also had a lower rate of ILI than the controls in the last few weeks of the study, but the difference was not significant. And neither group had a significant reduction in the cumulative ILI rate for the entire 6 weeks of the intervention.

The study was authored by Allison E. Aiello of the University of Michigan along with colleagues from Michigan, the University of South Alabama, and the Centers for Disease Control and Prevention (CDC).

"Our results indicate that interventions to reduce the transmission of ILI during a winter season may have substantial effects among individuals who share crowded living conditions," their report says.

Surgical masks have been widely used for protection against flu and ILI, but evidence for their effectiveness in protecting wearers from infection (as opposed to reducing virus shedding by sick people) from flu has been sparse. The authors of the Michigan study say they know of no previous studies of the use of face masks in "open, noninstitutionalized populations" to protect healthy people from respiratory infections.

Seven dorms participated
The researchers used a cluster randomized design in which seven dorms were randomly assigned to one of three arms of the trial: face masks and hand hygiene (FMHH), face masks only, or control. The largest dorm became a mask and hand hygiene test site; four smaller dorms became mask-only sites, and the other two dorms served as the control group.

All the participants received basic online education on hand hygiene. The FMHH group also received written materials and a supply of alcohol-based hand sanitizer. The two mask-wearing groups received standard medical procedure masks and were asked to wear them as much as possible while in the dorm and encouraged to use them elsewhere as well. The intervention was started after the first flu case was identified on campus; it ran from Jan 22 to Mar 16, 2007, with 1 week excluded for spring break at the end of February.

Participants were asked to complete baseline and weekly online surveys about any respiratory illness symptoms and their use of interventions during the study. Study staff members were stationed in residence hall common areas to observe the students' compliance with the interventions.

The three groups totaled 1,297 participants, with 367 in the FMHH group, 378 in the mask-only group, and 552 controls. At baseline, the groups did not differ in variables such as stress level, smoking, alcohol use, exercise, and flu vaccination; but the control and mask-only groups reported better hand washing practices than the FMHH group.

During the intervention, 368 students had an ILI as defined either by self-report or a clinical report. The 94 students who were medically evaluated were tested for flu, and 10 tested positive, including 2 from the FMHH group, 5 from the mask-only group, and 3 from the controls.

Ten percent overall reduction in ILI
Without adjustments for other variables, the two intervention groups had about a 10% reduction in overall ILI incidence over the 6 weeks compared with the control group, but this was not statistically significant.

However, when the researchers analyzed the ILI rates for each week of the study and adjusted for other variables (age, sex, ethnicity, baseline hand washing practices, sleep quality, alcohol use, flu vaccination), they found that the FMHH group had significantly lower rates than the controls in weeks 4, 5, and 6 (35%, 44%, and 51%, respectively).

The adjusted rates for the mask-only group also were lower than those of the control group in weeks 4, 5, and 6, with reductions ranging from 28% to 42%, but this did not meet the authors' significance criterion of P<.025.

The authors say several factors may explain why a significant reduction in ILI was found only during the second half of the intervention period. One is that recruitment of students continued through the first 2 weeks of the intervention, increasing the sample size by 11%. Also, the proportion of FMHH participants who wore their masks more than the average 3.5 hours a day increased in the later weeks of the study, and confirmed flu cases on campus peaked in weeks 4 and 5.

Although only the FMHH group had a significant reduction in ILI in the later weeks of the study, the authors say the ILI rates didn't differ much between the two intervention groups. This suggests that the hand hygiene component didn't contribute appreciably to the reduction, they write. But they add that their study lacked the statistical power to detect small differences between the groups.

As for other limitations of the study, the researchers say most ILI cases probably were not flu, as 2006-07 was a mild flu season. Also, self-reporting might have introduced a bias, and direct observation of the participants' use of the interventions was limited.

Study wins praise, stirs debate
The study won praise from other experts for careful design and execution. And although it did not deal with respiratory protection for healthcare workers, some observers said the findings support the view that face masks offer sufficient protection for healthcare workers caring for H1N1 flu patients—contrary to the current CDC recommendation that such workers should wear N-95 respirators. But others said that conclusion goes too far.

In the accompanying JID editorial, Titus L. Daniels and Thomas R. Talbot of Vanderbilt University praised the study as "a well-designed cluster randomized study demonstrating that the use of a face mask combined with hand hygiene in a crowded community setting is helpful in preventing ILI."

Daniels and Talbot write further that the Michigan findings, combined with the difficulties associated with N-95 respirators (including fit-testing, cost, and the need to conserve them for use in other settings such as tuberculosis), "support a recommendation for face mask use, and not N95 respirators, to prevent transmission of influenza and ILI."

Neil Fishman, MD, president of the Society for Healthcare Epidemiology of America, expressed similar views. "This was a wonderful, excellently designed and executed investigation, and I think the results are very significant," he said, commenting that randomized cluster studies are very difficult to execute.

One limitation is that the study does not address the effect of hand hygiene alone in preventing flu transmission, an area where there is little evidence, said Fishman, who is director of the Department of Healthcare Epidemiology and Infection Control for the University of Pennsylvania Health System in Philadelphia. Hence, he said he is unsure whether he would recommend, as a matter of policy, using the combination of face masks and hand hygiene in group settings such as college dorms.

Also, the results shouldn't be generalized to the community at large, Fishman added. "I don't think people should be reading this study and deciding we should be walking around with face masks on during flu season."

But while the study didn't deal with N-95 respirators, he said, "I think it did demonstrate that face masks work. So it does provide adjunct data that N-95s are not necessary, at least for seasonal influenza."

Cautionary advice
Raymond Tellier, MD, MSc, a leading expert on flu transmission and a microbiologist with the Provincial Laboratory for Public Health in Calgary, Alta., also had praise for the study, but he cautioned against concluding too much from it.

"This work convincingly supports the contention that simultaneously practicing frequent hand hygiene and constantly wearing a surgical mask result in a significant reduction in the incidence of influenza-like illness," he commented by e-mail.

But because only 10 participants tested positive for flu, "it is not possible to conclude anything specific about influenza transmission as such," said Tellier, who is also an associate professor in the Department of Microbiology and Infectious Diseases at the University of Calgary.

He also took issue with the view that the study serves as evidence that surgical masks are as good as N-95 respirators for protecting healthcare workers from flu viruses.

He said participants in the study were supposed to wear the masks most of the time, a practice that helps protect other people from flu viruses spreading either by large droplets or aerosolized particles if the wearer is infected.

"In contrast, healthcare workers entering the room of a patient with influenza would encounter already aerosolized influenza virus, a very different scenario," he said. "To use the study of Aiello et al as an argument against the use of N-95 for influenza protection by healthcare workers, as done by Daniels and Talbot in their editorial comment, does not appear convincing."

Another infectious disease expert, Michael T. Osterholm, PhD, MPH, also cautioned against taking the findings as evidence that masks work just as well as N-95 respirators for protecting wearers from flu. He is director of the University of Minnesota Center for infectious Disease Research and Policy, publisher of CIDRAP News.

Noting that the intervention groups had no significant cumulative reduction in ILI for the overall study period, Osterholm said Daniels and Talbot's editorial comments go well beyond what the study actually showed.

He acknowledged that N-95 respirators, while they theoretically should work better than masks, pose practical problems. But he said the debate over masks versus respirators has generated strong feelings that lead to distorted views.

"The emotions have far outstripped the data. I've seen it in both studies and editorials, and I don't think this is any different," he said. "And to date there's limited and conflicting data that enlightens us on this issue. We just need better studies."

Editor's note: CIDRAP has received unrestricted funds from 3M Co. as part of its higher education giving. 3M manufactures both respirators and surgical masks.

Aiello AE, Murray GF, Perez V, et al. Mask use, hand hygiene, and seasonal influenza-like illness among young adults: a randomized intervention trial. J Infect Dis 2010 Feb 15 (early online publication) [Full text]

Daniels TL, Talbot TR. Unmasking the confusion of respiratory protecton to prevent influenza-like illness in crowded community settings (Editorial). J Infect Dis 2010 Feb 15 (early online publication) [Full text]

See also:

Nov 5, 2009, CIDRAP News story "Reanalysis changes findings in respiratory protection study"

Oct 2, 2009, CIDRAP News story "Study suggests masks rival respirators for flu protection"

Sep 3, 2009, CIDRAP News story "IOM affirms CDC guidance on N95 use in H1N1 setting"

Aug 12, 2009, CIDRAP News story "IOM hears diverse findings on PPE for flu"

Newsletter Sign-up

Get CIDRAP news and other free newsletters.

Sign up now»


Unrestricted financial support provided by

Bentson Foundation Unorthodox Philanthropy logo and text 'Leading Underwriter'3M logoGilead 
Grant support for ASP provided by


  Become an underwriter»