The commonest source of such infections are unwashed hands of doctors and medical attendants. In fact, a recent study of several intensive care units in the US showed that hands were washed on only one quarter of the necessary occasions. The report claims that "hand hygiene (HH) is the single most important factor in the prevention of health care-acquired infections".
Though extremely trivial, hand washing by doctors and nurses before attending to each patient is one of the most vexatious of problems. Atul Gawande has documented the challenge posted by it and chronicled how some hospitals have tried to create a culture of hand-washing by using checklists and by redesigning hand hygiene systems to make hand-washing easier and automatic. However, the amount of efforts required makes such approaches difficult to replicate on scale. In this context, NYT Fixes points to a technological solution being piloted in a few hospitals in the US,
"Every health care worker wears an electronic badge. When she washes her hands or uses alcohol rub, a sensor at the sink or dispenser or her own badge smells the alcohol and registers that she has washed her hands. Another sensor near the patient detects when her badge enters a room or the perimeter around a patient that the hospital sets. If that badge shows that her hands were recently washed, it displays a green light or something else the patient can see. If she hasn’t washed, her hands, the badge says so and emits a signal to remind her to do so. The sensor also sends this information to a central data base. Information about the hand-washing practices of a particular unit, shift or individual is instantly available."
In other words, the sensors nudge hospital staff to wash their hands before they attend to their patients. They bridge the last mile gap by bringing in a point-of-care compliance monitoring system that nudges health care workers to follow hand hygiene standards.
Though these systems are largely in pilot phase, initial results are encouraging. Its prohibitive costs means that it will be some time before they become affordable for even the richer hospitals. However, the simplicity of the solution, RFID tags attached to sensors means that once its utility is established they could be quickly commercialized at affordable rates.
See Proventix's nGage, HyGreen's Hand Hygiene Recording and Reminding System, BioVigil's Health Care Badge and Alcohol Sensor, and Patient Care Technology System’s Amelior 360.
Update 1 (30/4/2011)
The second part of Fixes article by Tina Rosenberg is available here. Peter Pronovost's landmark paper in the NEJM (made famous by this New Yorker article of Atul Gawande) which chronicled the experiment of 103 ICUs in hospitals across Michigan that used a five-point checklist to prevent infections in central line catheters is available here.
The five-point checklist - wash hands; cover the patient with sterile drapes; clean the skin with chlorhexidine antiseptic; do not insert catheters into the groin area; remove catheters as soon as they are no longer needed - reduced the the median rate of infection to zero within 3 months and was sustained for the remaining 15 months of the follow-up.
Update 2 (2/9/2011)
A study of handwashing in the Journal of Psychological Science by David Hofmann and Adam Grant found that changing the messages posted in the hospital environment from "Wash Your Hands to Protect Yourself" to "Wash Your Hands to Protect Your Patients" could motivate some doctors and nurses to wash their hands more frequently.
They measured the change in soap use when they put up different signs by the dispensers. One sign read "Hand Hygiene Prevents You from Catching Diseases". Another read "Hand Hygiene Prevents Patients from Catching Diseases". And a third sign, which served as a control, had a generic message: "Gel In, Wash Out". The patient-focused sign produced a 33 percent increase in the amount of soap and disinfectant used per dispenser over a two-week period, compared with the other signs.
In a second phase of the study, trained observers recorded how often doctors and nurses physically washed or disinfected their hands. The sign urging doctors to think about patients produced a roughly 10 percent spike in hand washing compliance, a jump that was small but statistically significant.