Saturday, October 5, 2013

Don't BUGG me!

JAMA has just published the BUGG (Benefits of Universal Glove and Gown) study online (free full text here). This important, well-designed study was led by Anthony Harris (nice video of Anthony discussing the study here). It's a 9-month, multicenter, cluster randomized study in 20 medical and surgical ICUs that compares universal contact precautions (i.e., gowns and gloves for all patient care) to "standard" contact precautions (i.e., gowns and gloves for the care of patients with epidemiologically important organisms). The primary outcome evaluated was acquisition of MRSA or VRE. Patients were cultured for both organisms on admission and discharge from the ICU.

In a nutshell, the findings were as follows:
  • There was no significant difference in the rate of acquisition of MRSA and VRE combined.
  • When MRSA and VRE were evaluated separately, there was no difference in the acquisition of VRE, but there was a significant reduction in MRSA acquisition in the universal contact precautions group with an incremental benefit of 3 fewer MRSA acquisitions per 1,000 patient days.
  • There was no difference in device-related infections (CLABSI, CAUTI, or VAP) between the two groups and no difference in mortality.
  • There was no difference in adverse events between the two groups when evaluated by the IHI Global trigger tool (for what that's worth...).
  • Hand hygiene rates were higher in the universal gowns/gloves study arm.
  • As might be expected, there were fewer patient visits by healthcare workers in the universal gown/glove study arm.

So, how do we put this study into perspective? Should the study entice hospitals to begin universal gloving and gowning in the ICU setting? 

Let's assume you have a 15-bed ICU that admits 1,250 patients yearly with an average length of stay of 4 days (i.e., 5,000 patient-days annually). Assuming 10.5% of patients require contact precautions (this proportion comes from the control arm in the BUGG study), 131 patients would require isolation. Alternatively, under universal contact precautions, all 1,250 patients would be isolated. So by isolating an additional 1,119 patients we would prevent an additional 15 patients from acquiring MRSA (i.e., 3 per 1,000 patient days). Assuming 20% of the colonized patients go on to develop infection, 3 additional MRSA infections would be prevented with universal contact precautions. Bottom line: to prevent 3 additional infections we needed to isolate an additional 1,100 patients. Given that I'm a utilitarian and that I believe that the burden of contact precautions on patients is high, my assessment is that the benefit of universal gloves and gowns is outweighed by the overall burden on patients. Now it's true that MDR-GNRs and C. difficile weren't evaluated in the study so we may not be evaluating the full benefit of the intervention. But for now, don't BUGG me--I'm still pushing universal chlorhexidine bathing, high rates of hand hygiene compliance, and no isolation of patients with MRSA or VRE.

Addendum (10/6/13):  More on this study in Time.

1 comment:

  1. What I wonder is how the layout of the units (open pods versus single rooms) affect the association of universal gowns and gloves and acquisition of MRSA. I wonder this because with open pods, objects tend to be touched by providers taking care of neighboring patients, becoming a "bridge" for bacteria. Would these surfaces be more contaminated with providers wearing universal gowns and gloves than otherwise? My observation is that open layout units in my hospital tend to have higher problems with horizontal transmission of organisms than elsewhere; although this observation might be influenced by other colinear covariates.