Skip to main content

Table 7 Description of studies included in a review of effectiveness of workplace social distancing to reduce influenza transmission, 2000–2017

From: Effectiveness of workplace social distancing measures in reducing influenza transmission: a systematic review

First author, year published

Study design

Influenza strain and transmissibility (R0)

Population, setting, and number of people (n)

Intervention(s) and comparatora

Relevant outcomes

Epidemiological studies

 Rousculp, 2010 [27]

Cohort (participants surveyed at baseline and monthly from October 2007 to April 2008)

Seasonal influenza A(H3N2), 2007–2008

Employees of three large US companies (national retail chain, transportation, and manufacturing) (n = 793 employees with ILI)

Single: Can work from home

Comparator: Cannot work from home

Attended work for ≥1 day while ILI symptoms were most severe (which would result in transmission to co-workers)

 Kumar, 2012 [28]

Cross-sectional (survey completed from January 22 to February 1, 2010)

2009 influenza A(H1N1) pandemic

Random sample of US adults from the Knowledge Networks online research panel (n = 2079)

Single: Can work from home

Comparator: Cannot work from home

ILI during April 2009 to date survey completed

 Lee, 2010 [29]

Cohort (June 22 to October 9, 2009)

2009 influenza A(H1N1) pandemic

Singapore military personnel (n = 907)

Multiple: Standard pandemic plan (provided general health education on respiratory and hand hygiene and advised to seek medical care if ill) + segregation of units into subgroups as small as 20 individuals (including having different activity and meal times, and times of entry and exit from camp) + daily temperature and symptom monitoring with provision of home medical leave

Comparator: Standard pandemic plan

Seroconversion to 2009 influenza A(H1N1); time to peak (based on onset of symptoms among those who seroconverted)

Modeling studies

 Timpka, 2016 [30]

Model based on an ontology system. Mean duration of outbreak in the reference model = 92 days

Future pandemic strain

General population in Linkoping municipality, Sweden (n = 136,000)

Single: Social distancing that decreases workplace influenza virus transmission probability by 50%

Comparator: No intervention

Cumulative influenza attack rate

 Zhang, 2012 [31]

Agent-based model. Simulated for 200 days. Intervention is triggered at a threshold of 0.02%

R0 = 1.9

General population in Singapore (n = 480,000, representing a 10% sample)

Single: Team-based rotational workforce shift for 6 weeks (i.e., each company or institution splits its employees into two work teams and minimizes contacts between the teams through 7-day rotations of staying at home or attending work)

Comparator: No intervention

Cumulative and peak influenza attack rates; peak attack day

 Mao, 2011 [32]

Agent-based model. Simulated for 200 days. Intervention is triggered at a threshold of 0.1%

Seasonal scenario (R0 = 1.4) and a pandemic scenario (R0 = 2.0)

General population in Buffalo, New York, USA (n = 985,001)

Single: Weekend extension by 3 days (Sat, Sun, Mon, Tues, Wed).

Comparator: No weekend extension (Sat, Sun)

Cumulative influenza attack rate; daily new cases

 Xia, 2013 [33]

Compartmental model. Simulated for 200 days

2009 influenza A(H1N1) pandemic. R0 = 1.5

General population in Hong Kong, China (n = 7 million)

Single: Workplace contact reduction

Single + Vaccination: Workplace contact reduction + vaccination with a coverage of 2.5%

Comparator: No intervention

Peak infectious population; days to peak

 Milne, 2008 [34, 40]

Agent-based model. Intervention is triggered before the introduction of the first infected case

Pandemic strain.

R0 = 1.5, 2.0, 2.5

General population in Albany, Australia (n = 30,000)

Single: Workplace nonattendance (each person attending a workplace has a 50% chance each day of staying home instead of attending the workplace)

Multiple: Workplace nonattendance + school closure + case isolation + community contact reduction

Comparator: No intervention

Cumulative and peak influenza attack rates

 Milne, 2013 [35]

Agent-based model. Intervention is triggered 2 weeks after the first case

Pandemic strain. R0 = 1.88

General population in Madang, Papua New Guinea (n = 35,000)

Multiple: Workplace nonattendance (each person attending a workplace has a 50% chance each day of staying home instead of attending the workplace) + community contact reduction (50%) + school closure

Comparator: No intervention

Cumulative influenza attack rate; daily incident cases

 Miller, 2008 [36]

Discrete event model. Intervention is triggered after the first infection in city

Influenza A(H3N2) in population with no prior immunity

General population in San Antonio, Texas, USA (n = 1.4 million)

Multiple: Reducing the number of community contacts to 50% by day 21 after first infection (representing closing schools and churches, banning public gatherings, and encouraging people to work from home)

Comparator: No intervention

Cumulative influenza attack rate

 Andradottir, 2011 [37]

Individual-based compartmental model. Simulated for 180 days. Intervention is triggered at a threshold of 0.01%

2009 influenza A(H1N1) pandemic.

R0 = 1.4

General population in Hamilton, Canada (n = 649,565)

Multiple: Social distancing (20% reduction in workplace and general community contacts) + school closure

Multiple + Vaccination: Social distancing + school closure + vaccination of up to 35% of the population with a vaccine with 40% vaccine efficacy (60-day delay in vaccination after trigger threshold is reached)

Multiple + Antiviral: Social distancing + school closure + antiviral treatment and prophylaxis of up to 10% of the population

Comparator: No intervention

Cumulative influenza attack rate

 Perlroth, 2010 [38]

Agent-based model. Intervention is triggered at a threshold of 0.1%

R0 = 1.6, 2.1

General population in a small US town (n = 10,000)

Multiple: Adult social distancing (work contacts reduced by 50%) + child social distancing + school closure

Multiple + Antiviral: Adult social distancing + child social distancing + school closure + antiviral treatment and prophylaxis

Comparator: No intervention

Cumulative influenza attack rate

 Halloran (Imperial-Pitt model), 2008 [39]

Agent-based model. Results at day 180 of the epidemic are reported. Intervention is triggered at a threshold of 1%

Future pandemic strain. R0 = 1.9, 2.4, 3.0

General population in Chicago, Illinois, USA (n = 8.6 million)

Multiple + Antiviral: Workplace social distancing (workplace contacts reduced by 50%) + community social distancing + school closure + interventions within the households of ascertained influenza cases (antiviral treatment and prophylaxis, home isolation of cases, and quarantine of household contacts)

Multiple: Above, excluding antiviral treatment and prophylaxis

Comparator: No intervention

Cumulative influenza attack rate

 Halloran (UW/LANL model), 2008 [39]

Agent-based model. Results at day 180 of the epidemic are reported. Intervention is triggered at a threshold of 1%

Future pandemic strain. R0 = 2.1, 2.4, 3.0

General population in Chicago, USA (n = 8.6 million)

Multiple + Antiviral: Workplace social distancing (workplace contacts reduced by 50%) + community social distancing + school closure + interventions within the households of ascertained influenza cases (antiviral treatment and prophylaxis, home isolation of cases, and quarantine of household contacts)

Multiple: Above, excluding antiviral treatment and prophylaxis

Comparator: No intervention

Cumulative influenza attack rate

 Halloran (VBI model), 2008 [39]

Agent-based model. Results at day 180 of the epidemic are reported. Intervention is triggered at a threshold of 1%

Future pandemic strain. R0 = 2.1, 2.4, 3.0

General population in Chicago, USA (n = 8.6 million)

Multiple + Antiviral: Workplace social distancing (workplace contacts reduced by 50%) + community social distancing + school closure + interventions within the households of ascertained influenza cases (antiviral treatment and prophylaxis, home isolation of cases, and quarantine of household contacts)

Multiple: Above, excluding antiviral treatment and prophylaxis

Comparator: No intervention

Cumulative influenza attack rate

  1. Abbreviations: ILI influenza-like illness, R 0 basic reproduction number, Imperial/Pitt Imperial College and the University of Pittsburgh, UW/LANL University of Washington and Fred Hutchinson Cancer Research Center in Seattle and the Los Alamos National Laboratories, VBI Virginia Bioinformatics Institute of the Virginia Polytechnical Institute and State University
  2. aUnless otherwise stated, the modeling studies assumed that the duration of social distancing in workplaces was for the entire influenza outbreak period