To characterize associations between exposures within and outside the medical workplace with healthcare personnel (HCP) SARS-CoV-2 infection, including the effect of various forms of respiratory protection.

Case–control study.

We collected data from international participants via an online survey.

In total, 1,130 HCP (244 cases with laboratory-confirmed COVID-19, and 886 controls healthy throughout the pandemic) from 67 countries not meeting prespecified exclusion (ie, healthy but not working, missing workplace exposure data, COVID symptoms without lab confirmation) were included in this study.

Respondents were queried regarding workplace exposures, respiratory protection, and extra-occupational activities. Odds ratios for HCP infection were calculated using multivariable logistic regression and sensitivity analyses controlling for confounders and known biases.

HCP infection was associated with non–aerosol-generating contact with COVID-19 patients (adjusted OR, 1.4; 95% CI, 1.04–1.9; P = .03) and extra-occupational exposures including gatherings of ≥10 people, patronizing restaurants or bars, and public transportation (adjusted OR range, 3.1–16.2). Respirator use during aerosol-generating procedures (AGPs) was associated with lower odds of HCP infection (adjusted OR, 0.4; 95% CI, 0.2–0.8, P = .005), as was exposure to intensive care and dedicated COVID units, negative pressure rooms, and personal protective equipment (PPE) observers (adjusted OR range, 0.4–0.7).

COVID-19 transmission to HCP was associated with medical exposures currently considered lower-risk and multiple extra-occupational exposures, and exposures associated with proper use of appropriate PPE were protective. Closer scrutiny of infection control measures surrounding healthcare activities and medical settings considered lower risk, and continued awareness of the risks of public congregation, may reduce the incidence of HCP infection.

To identify risk factors of patients placed in airborne infection isolation (AII) for possible pulmonary tuberculosis (TB) to better predict TB diagnosis and allow more judicious use of AII.

Case-control, retrospective study at a single tertiary-care academic medical center. The study included all adult patients admitted from October 1, 2014, through October 31, 2017, who were placed in AII for possible pulmonary TB. Cases were defined as those ultimately diagnosed with pulmonary TB. Controls were defined as those not diagnosed with pulmonary TB. Those with TB diagnosed prior to admission were excluded. In total, 662 admissions (558 patients) were included.

Overall, 15 cases of pulmonary TB were identified (2.7%); of these, 2 were people living with human immunodeficiency virus (HIV; PLWH). Statistical analysis was limited by low case number. Those diagnosed with pulmonary TB were more likely to have been born outside the United States (53% vs 13%; P < .001) and to have had prior positive TB testing, regardless of prior treatment (50% vs 19%; P = .015). A multivariate analysis using non–US birth and prior positive TB testing predicted an 18.2% probability of pulmonary TB diagnosis when present, compared with 1.0% if both factors were not present.

The low number of pulmonary TB cases indicated AII overuse, especially in PLWH, and more judicious use of AII is warranted. High-risk groups, including those born outside the United States and those with prior positive TB testing, should be considered for AII in the appropriate clinical setting.

Ventilator bundles encompass practices that reduce the risk of ventilator complications, including ventilator-associated pneumonia. The impact of ventilator bundles on the risk of developing ventilator-associated events (VAEs) is unknown. We sought to determine whether decreased compliance to the ventilator bundle increases the risk for VAE development.

Nested case-control study.

This study was conducted at 6 adult intensive care units at an academic tertiary-care center in Tennessee.

In total, 273 patients with VAEs were randomly matched in a 1:4 ratio to controls by mechanical ventilation duration and ICU type.

Controls were selected from the primary study population at risk for a VAE after being mechanically ventilated for the same number of days as a specified case. Using conditional logistic regression analysis, overall cumulative compliance, and compliance with individual components of the bundle in the 3 and 7 days prior to VAE development (or the control match day) were examined.

Overall bundle compliance at 3 days (odds ratio [OR], 1.15; P=.34) and 7 days prior to VAE diagnosis (OR, 0.96; P=.83) were not associated with VAE development. This finding did not change when limiting the outcome to infection-related ventilator-associated complications (IVACs) and after adjusting for age and gender. In the examination of compliance with specific bundle components increased compliance with chlorhexidine oral care was associated with increased risk of VAE development in all analyses.

Ventilator bundle compliance was not associated with a reduced risk for VAEs. Higher compliance with chlorhexidine oral care was associated with a greater risk for VAE development.

Infect Control Hosp Epidemiol 2018;39:637–643