Evaluate risk before choosing respirators

Hospital chooses PAPRs to raise protection

Choosing a respirator is partly a matter of numbers. Each one has a rating for its filtration and seal, which reveals how much of the contaminated ambient air would still reach the lungs.

But assessing the risk in health care is more complicated than a single calculation — and so is selecting the proper respirator.

"Each individual institution must make a decision about risk: What risk is acceptable for your institution?" says John Schaefer, CIH, HEM, CPEA, associate director of health safety and environment at Johns Hopkins University and Medical Institutions in Baltimore. Schaefer presented a model for respiratory selection at the recent annual conference of the American Industrial Hygiene Association.

Risk is clearly related to the type of infectious disease your health care workers will encounter. XDR-TB, or extensively drug-resistant tuberculosis, would be of even greater concern than drug-susceptible TB. Pandemic influenza will, at least initially, call for higher levels of precautions than seasonal influenza.

The respiratory risk of infection for the health care worker is influenced by the time exposed and the ventilation of the room. For example, 12 air exchanges per hour will create a lower-risk environment than six air exchanges per hour, the minimum required by the Centers for Disease Control and Prevention for airborne isolation rooms.

Chemical hazards have a designated exposure limit, but infectious organisms are not that clear-cut. Infectivity of organisms differs — and may vary even among patients. Some patients may release a high concentration of organisms in coughing or even breathing. During SARS, for example, some patients were dubbed "superspreaders" because of their heightened infectivity. Aerosol-generating procedures, such as bronchoscopy or intubation, clearly also increase risk.

For many diseases, infectivity isn't well understood. It is theoretically possible for someone to become infected from a single infectious particle. But hospitals can use data from measles and tuberculosis — two airborne diseases — to gauge the potential effectiveness of respirator types for viruses and bacteria, says Schaefer.

Risk data on measles, TB

One study of airborne spread of measles in an elementary school produced some quantifiable data on infectivity. The index case produced 93 quanta of airborne infection per minute. The probability of infection from an hour of exposure without respiratory protection was 29%, or one in 3.5.1

Wearing a fit-tested N95 would have reduced that probability to 7%, or one in 24, calculated Schaefer. By using a powered air-purifying respirator (PAPR), the probability would have been 1% or less, depending on the fit factor of the model.

A 1982 study of nosocomial tuberculosis found that a patient in an isolation room with six air exchanges per hour released 60 quanta of infectious particles.2 That amounted to a 19% probability of infection after an hour of exposure without a respirator, or a risk of one in five. With an N95, the risk would drop to 5%, or one in 20, and would be less than 1% with a PAPR, Schaefer says.

"From an industry hygienist point of view, for someone to become ill or die due to workplace exposure is unthinkable," says Schaefer. Infection control, however, seeks to minimize transmission but accepts some level of risk.

Ultimately, hospital administrators must consider factors that include cost, employee comfort and preference, and regulatory issues, he notes. "This is just another way of adding a factor for risk of infection," he says.

Selecting reusable respirators

Johns Hopkins has chosen to maximize protection — but also to benefit from respirators that can be reused. For example, the hospital's emergency department recently decided to stock elastomeric half-face respirators for pandemic influenza preparedness.

If a pandemic lasted a month, N95s would cost about $40 a person, assuming that health care workers used two masks a day for 20 days, says Schaefer. An elastomeric respirator with a cartridge that can be cleaned with bleach or alcohol costs $20, he says. The hospital also would not need to worry about lack of supplies or unavailability of the fit-tested models.

Meanwhile, the hospital purchased about 200 PAPRs, which are available on every floor and don't need to be fit-tested.

Yet some health care workers prefer the N95, and Johns Hopkins still provides those. The hospital convened a committee that included frontline health care workers and selected models based on comfort and ease of use. Employees who are assigned N95s receive annual fit-tests.

Johns Hopkins is now evaluating its respiratory protection for pandemic influenza preparedness. Schaefer and his colleagues are reviewing departments and job descriptions to determine which employees would be at the greatest risk during a pandemic. The respiratory protection will differ based on that risk assessment, he says.

"We're going to inform people of the potential risk they have and what type of device they will be using to protect themselves," he says.

Ultimately, Schaefer hopes that employees will feel confident in the protection the hospital is providing — and will be more likely to report for work in the case of a pandemic.


1. Riley EC, Murphy G, and Riley RL. Airborne spread of measles in a suburban elementary school. Am J Epidemiol 1978; 107: 421-432.

2. Cantanzaro A. Nosocomial tuberculosis. Am Rev Respir Dis 1982; 125:559-562.