Systematic review of empiric studies on lockdowns, workplace closures, and other non- pharmaceutical interventions in non-healthcare workplaces during the initial year of the COVID-19 Pandemic: Benefits and selected unintended consequences

Background: We conducted a systematic review aimed to evaluate the effects of nonpharmaceutical interventions within non-healthcare workplaces and community-level workplace closures and lockdowns on COVID-19 outcomes in workers or the general community and on selected mental health and labor market outcomes. Methods The inclusion criteria included systematic reviews, randomized controlled trials, and non-randomized studies. The exclusion criteria included qualitative, and modeling studies. Electronic searches were conducted using MEDLINE, Embase, and other databases from January 1, 2020, through May 11, 2021. Risk of bias was assessed using the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) tool. A qualitative synthesis was performed. Results A total of 60 studies met the inclusion criteria. There were 40 studies on COVID-19 outcomes, 15 on anxiety and depression symptoms, and five on unemployment and labor force participation. There was a paucity of studies on physical distancing, physical barriers, and symptom and temperature screening within workplaces. Workplace closures significantly reduced COVID-19 incidence or growth rate (five of seven studies) and reproduction number (three of four studies) in the general community. Lockdown significantly reduced COVID-19 incidence and case growth rate (22 of 23 studies), reproduction number (10 of 11 studies), and COVID-19 mortality and death growth rate (seven of seven studies) in the general community. Lockdown significantly increased depression symptoms (10 of 15 studies), but the effect on anxiety symptoms was inconsistent. Lockdown increased unemployment (five studies) and decreased labor force participation (three studies). The risk of bias for most of the studies on COVID-19 or labor market outcomes was moderate or serious. The risk of bias for the studies on anxiety or depression symptoms was serious or critical. Conclusions Empiric studies that assessed the effect of workplace closures and lockdowns suggest that these measures helped reduce the impact of COVID-19, albeit with notable secondary (unwanted) effects. There is a pronounced paucity of studies on the effect of interventions, including symptom and temperature screening as well as less disruptive nonpharmaceutical interventions such as physical distancing measures within still-open workplaces. Addressing the gaps in the evidence base would be important for informing future pandemic preparedness. Systematic review registration number: PROSPERO registration # CRD42020182660.