Intervention Study of Face Mask and Hand Sanitizer to Reduce Influenza Transmission

2014-08-27 03:37:52 | BioPortfolio


Pharmaceutical interventions, particularly vaccination, have been the principal public health method of preventing and controlling seasonal influenza. Other less invasive non-pharmaceutical interventions include the use of face masks and hand hygiene. In Asia, masks are used extensively, especially in Japan, but with only circumstantial evidence of efficacy. Although there are numerous studies that support hand hygiene as an effective measure for decreasing transmission of enteric infections, there are fewer data assessing the effect of hand hygiene on respiratory illnesses. Therefore it is proposed that a study be conducted on the use of face masks with and without the use of hand hygiene in comparison to no intervention in residence housing units of the University of Michigan. The interventions will be carried out during a period of influenza transmission that is defined by surveillance conducted at the University Health Service. Data will be collected on use of the interventions and the occurrence of influenza and other illnesses. When illness occurs meeting a case definition, specimens for viral isolation will be collected in all intervention settings. Face masks could theoretically lead to a large reduction in transmission frequency of influenza, especially if most transmission occurs by the airborne (large or small droplet) route. Even a modest reduction could have a major effect during an influenza pandemic.



The overall objective of this study is to investigate the feasibility and effectiveness of various non-pharmaceutical interventions on the occurrence of influenza in residence units of the University of Michigan.

Specific Aims:

The three specific aims of this study are to: (1) determine the feasibility of using face masks and hand hygiene education among students living in residence hall units at the University of Michigan, (2) estimate the reduction of influenza infection and illness during laboratory documented influenza outbreaks in individuals and groups using one or more interventions compared to those not using such interventions, and (3) track occurrence of influenza cases in the participating residence units of the University.


Pharmaceutical interventions, particularly vaccination, have been the principal public health method of preventing and controlling seasonal influenza due to the availability, sufficient quantity and appropriate time before and during most influenza seasons. All this has changed with the threat of an influenza pandemic, potentially one caused by the (H5N1) avian influenza virus because a vaccine that can convincingly produce protection has not yet been possible to produce(4). This leaves Non-Pharmaceutical Interventions (NPI's) as the most attractive way to mitigate the impact of a disastrous pandemic. But making policy decisions will be a challenge, given the lack of substantive scientific information on the effectiveness of NPI's(8).While school closings have been used in some countries as a means of separation, the evidence is less than desirable, often due to the late implementation after absenteeism of teachers or students, rather than to prevent transmission.

General improvements in hand hygiene practices are well established in reducing certain infections(13) and serve as an effective measure for decreasing transmission of enteric infections. But little data exists that assesses respiratory infection outcomes. We conducted a review of the literature that examined the effect of hand hygiene interventions in relation to respiratory infection outcomes among individuals residing within the community setting (i.e. households, day-cares, and schools). Four studies reported a significant reduction attributable to hand hygiene education alone while four others reported no significant reduction. Another study reported a 51% reduction in respiratory symptoms attributed to an intervention that provided soap in addition to hand hygiene education compared with a control group that received nothing.(24) Three studies examining the use of waterless hand sanitizers (one study using alcohol-based hand sanitizers and two other studies using benzalkonium chloride hand sanitizers) reported a significant reduction in upper respiratory symptoms compared with control groups that did not use a waterless hand-sanitizer. Importantly, a randomized study conducted among students living in residence halls at a large university in Colorado reported a 30% reduction in upper respiratory-illness symptoms with the use of hygiene education and alcohol-based hand sanitizers compared to control dorms given nothing.(27)

Our review of the impact of hand hygiene interventions on respiratory symptoms suggests that: (i) the results of community based studies have been inconsistent but there are several studies that support the use of hand hygiene education, hand-washing, and waterless hand-sanitizer use as a relatively effective means for reducing respiratory symptoms; (ii) there is a lack of studies in which reports of respiratory illnesses are laboratory confirmed as to the etiology of the symptoms; (iii) droplet transmission may partly explain the reported lower effect estimates and inconsistent findings across hand hygiene intervention studies as compared to reported reductions in gastrointestinal infections.

While face masks are used extensively in Asia, there is only circumstantial evidence of efficacy. We were unable to identify any data in English directly assessing the efficacy of face masks in the community setting as a measure for preventing respiratory infections. Indeed, as indicated by the recently published Institute of Medicine report, there are few data available on the differential value of different face masks, other than the N95 masks, which are not feasible for large scale use in terms of both practicality and supply. Face masks could theoretically have a major effect of lowering transmission frequency, especially if most of it occurs by the airborne (large or small droplet) route.

We were unable to identify any studies that utilized combined intervention efforts that targeted hand-to-hand transmission and prevention of respiratory droplet transmission through the use of face masks. Many of these studies measured both respiratory and gastrointestinal illness outcomes and the percent reduction in respiratory symptoms was of lower magnitude as compared to gastrointestinal symptoms for all hand hygiene interventions (i.e. education, soap, or hand-sanitizer). This suggests that other modes of transmission, such as droplet spread in addition to hand-to-hand interventions, should be assessed as a means for further reducing respiratory illness transmission in the community setting.

Further, community based intervention studies that target prevention of droplet transmission, via face masks in addition to hand-to-hand transmission along with alcohol-based antiseptic use, are needed along with culture confirmed respiratory illness data.


The proposed study is a randomized intervention designed to examine the effectiveness of face masks alone and layered face masks and hand hygiene interventions for reducing transmission of influenza among students residing in University of Michigan residence halls during laboratory defined influenza seasons. Randomization will be by geographically separate housing unit. In the first year, they will be assigned to each of the interventions and to the observed comparison to bring the total number of participants to 750 in each of the interventions and the comparison (total number: 2250). Students living in each of the residence halls will be asked if they wish to participate in the study. Those giving informed consent will be asked to complete a survey each week to ascertain illness occurrences. When influenza illness is documented to be occurring in the University setting, the interventions will be implemented. During that period, the survey will be expanded to include questions on intervention use, as appropriate for the particular housing unit. Also during this period, those with illnesses meeting a case definition will be asked to provide specimens for virus identification. This collection will not be limited to those agreeing to the use of the interventions, since informed consent will be obtained for all specimen collections.

Plans for the second year will be modified based on the results of the first. It is felt that the study should be large enough in the first year so that useful results can be obtained. Policy decisions may need to be made sooner rather than later. However, since it is difficult to estimate what proportion of residents will elect to participate and how well they will use the interventions the study may need to be expanded in the second year to additional housing units. Indeed, compliance data will be an important outcome of the study. Also, it may be possible to use targeted interventions with antivirals among direct contacts of cases (in conjunction with treating physicians in the University Health Service) and to get more information on viral shedding from the cases themselves.

The number of students at the University of Michigan who live in residence halls is 9,200. In order to obtain a representative sampling of residence halls across the University of Michigan campus, we have consulted with Carole S. Henry who is the director of student housing. She and her staff have been active participants in the overall design of the study and its feasibility. They have recommended targeting six residence halls and one residence complex that are representative of the overall student population and are also the best candidates for participation in the first year of the study. The six separate residence halls on the central campus area are of moderate size ranging from 137 to 500 residents. Since the target halls are located throughout the campus, they provide representative housing units for students that attend undergraduate classroom locations and utilize shared facilities, such as athletic centers, cafeterias, computing centers, and meeting places. Three of the units are all female and the others are co-ed residence halls. The units represent a distribution of the overall student populations as some are more specialized dormitories for students interested in specific areas of study and others are more general living quarters for a wide variety of student populations. There are additional halls that are potential candidates for participation in the second year; most are larger and will require more time for preparation in terms of discussing the interventions, instruction in the their implementation and other required procedures than will be available in the first year.

Intervention Methods:

Randomization will be clustered by residence hall since it is likely that incidence of influenza infection will be correlated among subjects within residence halls. This methodology has statistical implications, but also benefits, since the proportion of individuals participating will likely vary by unit. The intervention residence halls will receive one of the following two interventions: (i) masks alone or (ii) masks plus alcohol-based hand sanitizers (62% ethyl alcohol) and hygiene education. The remaining halls will serve as comparison units. The control dorms will be matched to the intervention dorms on basic demographic distributions, such as age range and geographical proximity. Residence halls that are randomized to the control arm will not receive any intervention products (i.e. masks, hygiene education materials, or alcohol-based hand sanitizer). For geographically matched residence hall, the specific randomization assignment will be carried out centrally (with student representation) by drawing random numbers from a hat, so that selection will be open, public and transparent.

Within each of the randomized residence halls assigned to receive an intervention, our goal is to recruit as many students as possible. We are going to provide incentives for participation, but recruitment to use of face mask may be difficult. Therefore, the largest incentive will be provided to participants who are randomly assigned to the intervention arms. Participants will be asked to complete weekly surveys for intervention compliance and reporting of symptoms as described in the following sections. This will begin as soon after funding as possible and will be handled through the University's e-mail system and, which will also be a means of obtaining informed consent to the intervention. The school year starts at the beginning of September and all students, including freshman, will have e-mails addresses assigned and available to us.

Starting early to mid-November, specimens for virus identification will be collected from individuals meeting a case definition presenting for care at the University Health Service. This has been done in the past from a small number of patients, but will be expanded for the purpose of timing the onset of influenza transmission on the campus. The case definition will be the presence of cough and at least one constitutional symptom (fever, chills or body aches). This definition is derived from one used in the trials of oseltamivir, modified based on recent analysis of predictors of influenza virus positivity of cases of influenza-like illness. The case definition will be used throughout the study and will be employed to make decisions, during the course of the outbreak, on collecting specimens from residents of residence halls. Its specificity and sensitivity will be evaluated and revised as necessary.

The use of face masks and alcohol in the intervention will be put into effect only when two or more virus positive illnesses are detected in a single week. Rapid tests will be available, if necessary, to help in identifying cases if laboratory confirmation may be delayed. The use of face masks will continue until the outbreak is recognized by virology to be coming to a close or after the passage of 6 weeks, whichever occurs first. It is felt that there needs to be a guaranteed duration in order to assure compliance with face mask use.

Face Masks Residence halls receiving interventions (i) and (ii) will receive face masks and will be asked to comply with their use at least while in the residence hall when it is known that influenza virus is transmitting on campus. A week's supply of face masks and plastic bags for disposal will be placed in a plastic bag along with instructions for use. The packets of face masks will then be placed in a unmarked envelope with a label of the participants name and room number. The unmarked face mask packets will then be placed in each participant's residence hall mailbox each week, during the intervention. The residence halls have shared eating facilities and other common areas that may provide areas of mixing with residents from other halls and potential influenza transmission; students can and do eat in any residence hall and spend additional time in them as well. They will be requested to wear the masks when out of the halls, but this may be more difficult to implement. Health education approaches will be used to get as much use as possible. As indicated below, we will ask what proportion of time the masks are worn. The choice of the specific face masks will be carried out in conjunction with the CDC project officers. We propose using standard surgical masks; these have been stockpiled by the University Health Service as part of pandemic preparedness. It is likely that N-95 masks would be impractical, both for experimental use in a study such as this, but, more importantly, not a realistic intervention for public health purposes in the target population. Thus, the surgical mask appears to be a good choice for evaluation as it is relatively inexpensive and easy to don. The face mask intervention would require a once a day change during use to ensure proper protection and a hygienic barrier. It is clear from the recent Institute of Medicine report that little is known about long term use or reuse of face masks, so variations in the number of times in which a face mask is changed is a variable that could be assessed in this study.

The use of face masks will continue for 6 weeks. While normal surgical masks are relatively simple to use, there will be instructions given as to the correct and incorrect way to utilize them. As part of the evaluation of level of use, staff observers in each facility will estimate what proportion of individuals actually use them properly. This will be done on a weekly basis throughout the intervention period and observations will take place in common rooms, hallways, and residence hall dining facilities. This will supplement the self reports that will be collected by e-mail from all participants.

Hand Hygiene Education and Antiseptic Product Use All students in residence halls receiving intervention (ii) will be provided with a comprehensive hand hygiene education component in addition to mask use. The educational component will include posters and pamphlets that provide background information on transmission of influenza, an explanation of the effectiveness of hand hygiene for reducing transmission, and guidelines for proper hand hygiene, including the need for hand hygiene at critical points (i.e. after sneezing or other contact with respiratory secretions). The hand washing campaign will consist of posters on bulletin boards and weekly email hand hygiene reminder messages to participants in the intervention residence halls. The messages will encourage regular hand-washing and hand-sanitizer use to prevent transmission of influenza within living quarters. The educational component of the study will include a video for both the control arm and the intervention arms. The control video will address information on influenza and an overview of the study. The intervention video will address both information on influenza and an overview of the study, in addition to hand hygiene education and the use of hand sanitizer and the use of face masks. Both videos are available on the website ( and on Survey Monkey linked to the consent form.

Hand sanitizers were made available in some areas of University housing since a norovirus outbreak occurred in one of the residence halls. However, it is not clear whether students have been using these products regularly or if they own a pocket-sized hand sanitizer for daily use. Research suggests that providing a portable hand-hygiene product is an important motivating factor for compliance. Therefore, the intervention dorms will be given a portable alcohol-based hand sanitizer and a large stationary pump alcohol-based hand sanitizer to use throughout the study period in addition to basic hand hygiene education materials and messages. In order to quantify the effect of the alcohol-based hand sanitizer use on influenza virus transmission, regular self reports of hand-sanitizer use will be collected and the products provided will be monitored for use (by weighing) and re-supplied as needed on a weekly basis during the influenza season. It is likely that some subjects in the other arms may also be utilizing alcohol-based hand sanitizer and it would be unethical to ask these students to reduce the use of these products during influenza season. Therefore, control participants will also be required to provide information on type and name of product used and daily frequency of use, should they be using any.

Control Residence Halls Students residing in the control housing units will be asked to participate in the disease surveillance aspect of the study. They will be contacted by e-mail, be asked to consent to participate in a wellness study in which they will report on respiratory and enteric illness symptoms on a weekly basis. The students will be compensated for this participation but not at the level of those who are in the intervention arms.

Hygiene Face Mask and Illness Surveillance Participants in both the intervention and control housing units will be asked to report information on a weekly basis concerning use of the interventions (where appropriate) and the occurrence of respiratory and enteric illness symptoms. Questions on the health attitudes and knowledge will be collected at the start and end of the study season. Dr. Aiello has extensive experience monitoring, surveying, collecting data, and analyzing hand hygiene practices among clinicians as well as among individuals living in households in the community setting. This survey will be modified for residence hall participants so that frequency of use of face masks can be similarly determined, including attitudes about wearing them. It will be provided electronically through the use of a professional survey service (

Both intervention compliance and outcome measures will be collected via electronic survey. This will include mask use, alcohol-based hand sanitizer use, and details regarding respiratory symptoms and enteric symptoms. The symptomatic questions will be based on those used in the Tecumseh Community Health Study. During the first year, the survey instruments will be assessed for reliability for use among our target student populations in residence halls and may be modified in the second year. As indicated above, in order to encourage those who agree to complete the reports through the collection period, honoraria will be paid. They will be calculated based on the weeks the report is completed, and paid at the end of the season.

Collection of Specimens for Virus Identification All residents in the intervention and comparison residences will receive small magnetic cards, modified after those currently used in the vaccine study, describing the case definition. Whenever an illness occurs which meets the case definition (presence of cough and at least one constitutional symptom (fever, chills or body aches)), an individual will be asked to have a specimen collected for virus identification. Information on symptoms, duration of illnesses and other relevant characteristics will be obtained. In all cases, the course of the illness will be monitored, with follow-up collection of severity and duration information everyday for at least seven days after the reported onset of symptoms.

Particular attention will be placed on tracking cases of laboratory confirmed influenza in the housing units. This will allow determination of whether contacts who are using the interventions are actually being infected. During the second year, it may be possible to use targeted prophylaxis in among identified contacts. We will also use the documented presence of influenza on the campus and in the housing units as a way of improving compliance.

Laboratory Methods for Virus Identification:

All specimens for virus identification will be collected by throat swab and placed in veal infusion broth transport medium. This transport medium has been found to protect viruses for days during shipment from remote sites and to be suitable for PCR tests as well.

Specimens will be inoculated into two tubes of cell culture. Typically purchased rhesus monkey kidney is used, although MDCK cells with trypsin are also available. At 3-4 days and at one week, the cultures are hemadsorbed using guinea pig red cells. Positives are passed to additional tubes of cell culture. Initial typing is carried out by fluorescent antibody techniques. Subtyping of type A viruses are carried out by hemagglutination-inhibition. For specimens collected in the influenza vaccine efficacy study, specimens are also processed by real time PCR. This has been found to double the identification frequency for type A viruses. However, this procedure can not be carried out concurrently on additional specimens and is not included in the current budget. The technique could be carried out at a later time from frozen specimens.

We are going to provide questions regarding influenza immunization in our study questionnaire for all participants. If the participant is interested in being vaccinated, we will refer them to health services for further information regarding obtaining vaccinations. By collecting data on vaccination use among participants, we will be able to control for vaccine use in our study analyses. Given the low rate of vaccination uptake among college age individuals, we do not expect vaccine use to be a significant confounder in our study design.

Statistical Design:

To estimate the reduction of influenza infection and illness during laboratory documented influenza outbreaks in individuals and groups using one or more interventions compared to those not using such interventions.

The primary purpose is to directly test the effectiveness of the interventions in year two. Prior to the outbreak, students randomized to intervention arm (ii) will already have been exposed to a hygiene education promotion program and will be using an alcohol-based hand sanitizer. During an increase in influenza cases in target residence halls, intervention arms (I & ii) will receive face masks. Cases will be identified and followed over time in each of the halls. Survey data and observations for intervention compliance will be collected. Using the data collected during year two, a series of analytic steps that accommodate the multivariate and clustered nature of the data will be conducted. Overall, the data analytic strategy involves the following:

1. We will construct a database using electronically collected survey and symptoms data. Data will be analyzed descriptively to determine normality, sufficient variability for analyses and to identify missing values, outliers, and illogical values. As needed, erroneous data will be corrected and transformations such as taking logarithms or categorization will be used for variables with high skewness or kurtosis.

2. Descriptive statistics such as frequency counts, percentages, means, medians, standard deviations, will be used to characterize each variable of interest. Point estimates and confidence intervals will be used to estimate population values on all variables. Descriptive statistics (point estimates and confidence intervals) for sub-samples of interest, such as age and gender, will also be presented. Contingency tables with appropriate statistical tests will be constructed to test for differences in key variables between intervention units and control units.

3. The effect of the intervention will be tested through the use of generalized estimating equations (GEE) as applied to statistical regression models for clustered data. The GEE methods are increasingly utilized in the epidemiological studies of clustered data and Dr. Aiello has successfully developed and applied these models for assessing the effect of household hygiene interventions on carriage of bacterial species and antibiotic resistance. A Poisson regression model with GEE will be used to examine the effect of each intervention on influenza case rates within the residence hall. Given the short time range of study follow-up we do not anticipate the need to impute missing data. However, should this be necessary, multiple imputation methods will be used to handle missing data. In all analyses, point estimates of the effects of interest will be presented together with standard errors, 95% confidence intervals, and p-values.

The main outcome for this aim will be the rate of influenza cases and rates of secondary influenza transmission comparing the intervention and control dormitories. Additional outcomes will include other non-influenza respiratory illnesses and enteric illness symptoms. Predictor variables will include the main effects of the intervention and control for potential confounders depending upon balance of the known confounders across intervention and control residence halls.

Study Design

Allocation: Randomized, Control: Uncontrolled, Endpoint Classification: Efficacy Study, Intervention Model: Factorial Assignment, Masking: Single Blind, Primary Purpose: Prevention




Ear Loop Procedure Mask, 62% Ethyl Alcohol Hand Sanitizer


University of Michigan
Ann Arbor
United States


Active, not recruiting


University of Michigan

Results (where available)

View Results


Published on BioPortfolio: 2014-08-27T03:37:52-0400

Clinical Trials [2321 Associated Clinical Trials listed on BioPortfolio]

Mouthwash With Chlorhexidine 0.12%/Ethyl Alcohol 7% Compared to Ethyl Alcohol 7%

Clinical triple-blind randomized controlled trial to assess the use of two different mouthwashes to reduce the oral colonization by gramnegative bacteria in patients with chemotherapy indu...

Alcohol Based Hand Sanitizer: Can Frequent Use Cause an Elevated Blood Alcohol Level?

Ethanol based hand sanitizers do not raise blood alcohol levels to a measurable amount. This study will determine to what extent, if any, heavy use of ABHS changes blood alcohol levels.

Effectiveness of a School-Based Hand Sanitizer Program on Reducing Asthma Exacerbations

Asthma is a common, serious illness among children in the United States. Improving hand cleanliness and hygiene may prevent the spread of viruses that can cause asthma exacerbations. This ...

FiteBac Hand Sanitizer in the Management of Hand Dermatitis in Adults

Hand dermatitis, (a rash, or eczema, on the hands) is a common skin condition that affects approximately 10% of the population. It can be caused by a wide variety of things, such as allerg...

Alcohol and Immune Response

This study investigates how alcohol affects the immune system and behavior in healthy adults. The study also will examine how an individual's typical drinking habits may affect the immune ...

PubMed Articles [9069 Associated PubMed Articles listed on BioPortfolio]

Occupational Exposure to Alcohol-Based Hand Sanitizers: The Diagnostic Role of Alcohol Biomarkers in Hair.

Ethyl glucuronide (EtG) and fatty acid ethyl esters (FAEEs) in hair are effective direct biomarkers of ethanol ingestion, whose analytical determination can be used to discriminate between chronic and...

Implementing a province-wide mandatory vaccinate-or-mask policy at healthcare facilities in British Columbia, Canada.

In 2012, British Columbia (BC) became the first Canadian province to implement an influenza prevention policy requiring healthcare workers (HCW) to either be vaccinated annually against influenza or w...

Evaluation of a novel method for the analysis of alcohol biomarkers: Ethyl glucuronide, ethyl sulfate and phosphatidylethanol.

Currently available markers and methods to evaluate alcohol consumption are indirect and suboptimal, or rely on self-report, which have inherent problems. Direct metabolites of alcohol, phosphatidylet...

Alcohols as Surface Disinfectants in Healthcare Settings.

Isopropyl alcohol and ethyl alcohol have been used as low-level disinfectants in healthcare settings for many years. Recent studies have found that ethyl alcohol inhibits protein synthesis in Escheric...

Hand-Sewn versus Stapled Closure of Loop Ileostomy: A Meta-Analysis.

Individual trials comparing hand-sewn with stapled closure of loop ileostomy show different outcomes due to lack of statistical power. A systematic review, with a pooled analysis of results, might pro...

Medical and Biotech [MESH] Definitions

Substances interfering with the metabolism of ethyl alcohol, causing unpleasant side effects thought to discourage the drinking of alcoholic beverages. Alcohol deterrents are used in the treatment of alcoholism.

Calcium-binding motifs composed of two helixes (E and F) joined by a loop. Calcium is bound by the loop region. These motifs are found in many proteins that are regulated by calcium.

A complication of gastrojejunostomy (BILLROTH II PROCEDURE), a reconstructive GASTROENTEROSTOMY. It is caused by acute (complete) or chronic (intermittent) obstruction of the afferent jejunal loop due to HERNIA, intussusception, kinking, VOLVULUS, etc. It is characterized by PAIN and VOMITING of BILE-stained fluid.

Species of the genus INFLUENZAVIRUS B that cause HUMAN INFLUENZA and other diseases primarily in humans. Antigenic variation is less extensive than in type A viruses (INFLUENZA A VIRUS) and consequently there is no basis for distinct subtypes or variants. Epidemics are less likely than with INFLUENZA A VIRUS and there have been no pandemics. Previously only found in humans, Influenza B virus has been isolated from seals which may constitute the animal reservoir from which humans are exposed.

Membrane glycoproteins from influenza viruses which are involved in hemagglutination, virus attachment, and envelope fusion. Fourteen distinct subtypes of HA glycoproteins and nine of NA glycoproteins have been identified from INFLUENZA A VIRUS; no subtypes have been identified for Influenza B or Influenza C viruses.

More From BioPortfolio on "Intervention Study of Face Mask and Hand Sanitizer to Reduce Influenza Transmission"

Quick Search


Relevant Topics

Public Health
Alternative Medicine Cleft Palate Complementary & Alternative Medicine Congenital Diseases Dentistry Ear Nose & Throat Food Safety Geriatrics Healthcare Hearing Medical Devices MRSA Muscular Dyst...

Within medicine, nutrition (the study of food and the effect of its components on the body) has many different roles. Appropriate nutrition can help prevent certain diseases, or treat others. In critically ill patients, artificial feeding by tubes need t...

Influenza or 'flu' is a respiratory illness associated with infection by influenza virus. Symptoms frequently include headache, fever, cough, sore throat, aching muscles and joints. There is a wide spectrum of severity of illness ranging from min...

Searches Linking to this Trial