Department^Codes + Standards^Testing the waters
Facts to consider when deciding whether to sample for Legionella

By Matthew R. Freije

The Joint Commission on Accreditation of Healthcare Organizations (JCAHO), under EC 1.7, indicates proactive efforts to minimize Legionnaires' disease by requiring hospitals to "reduce the potential for organizational-acquired illness ... managing pathogenic biological agents in cooling towers, domestic hot water and other aerosolizing water systems."

While the standard does not mandate water testing for Legionella bacteria, many institutions sample anyway, periodically collecting water samples from cooling towers and various points of the domestic water system for Legionella culture by a qualified laboratory.

For years, experts have disagreed about routine water sampling for Legionella--some are in favor of it and some aren't. The disagreement among experts has left many health facility managers and infection control coordinators--the people who are truly on the front lines of the battle against the disease--confused about what to do. The tragedy is that the real losers of the debate are the patients who contract a preventable disease.

Facilities managers shouldn't wait for environmental sampling to be mandated by a code or standard. They should review the facts now so they can make informed decisions for their institutions. The basis for the decision is almost as important as the decision itself. If managers choose to sample, they should know why. And if they choose not to sample, they should also know why.

Review the facts

Reviewing the following facts will help you make an informed decision and explain the reasons for it with confidence:

  • The debate is over routine sampling. Virtually everyone agrees that environmental sampling should be conducted after a case of Legionnaires' disease is confirmed in an effort to pinpoint the source of contamination. The disagreement is about whether to sample routinely, in the absence of disease, to ascertain risk.
  • Government agencies differ in recommendations. The Centers for Disease Control and Prevention (CDC) generally advocates environmental sampling only after a case of Legionnaires' disease has been identified, although CDC officials have conceded that routine sampling may be appropriate in hospitals that treat high-risk patients.

    The U.S. Occupational Safety and Health Administration (OSHA) does not require sampling, but advises that "analysis of water samples from a source suspected of being contaminated with Legionella pneumophila is a valuable means of identifying potential sources of the disease" in Section II, Chapter 7, of its 1996 technical manual.

    The Maryland Department of Health and Mental Hygiene, in a report available at www.dhmh.state.md.us/html/legionella.htm, says that "Plumbing systems within acute care hospitals should be routinely sampled for Legionellae."

    The Allegheny County (Pa.) Health Department was the first local agency in the United States to issue a guideline that included water testing for hospitals. The document is available at www.legionella.org.

  • Some of the disagreement is due to perspectives. Perhaps one reason there has not been consensus is that the debaters have different perspectives. Government officials must think in terms of what practices should be mandated for all buildings in an entire country or state. Variables from one building to another make it nearly impossible to develop Legionella sampling guidelines that will fit every situation all the time. In contrast, a facility manager or consultant views the issue in terms of what is responsible for a given facility. A CDC official who contends that sampling should not be mandated for all hospitals may very well agree with a facility manager's decision to conduct routine sampling in a particular facility.
  • Sampling can be a logical part of a proactive program but should not be the focus. If an institution aims to design, operate and maintain domestic water systems and cooling towers to minimize Legionella bacteria, periodic water tests can provide feedback as to whether the preventive measures are working and may indicate specific maintenance. Sampling can therefore lead to better-targeted preventive maintenance.

    Sampling should not be the heart of the prevention program. Some health facilities conduct sampling but ignore flaws in the design or maintenance of mechanical systems.

  • Legionella is not everywhere. Legionella is not necessarily ubiquitous, and many hospitals find zero positive samples. Others find only one or two positives out of 20 or so samples per screening. In the table accompanying this article, six surveys involving a total of 264 hospitals in the United States, United Kingdom and Canada showed that nearly half of the facilities found no Legionella in the domestic water system. All samples collected from the domestic water systems were negative in 121 (46 percent) of the hospitals.

    A higher percentage of the hospitals may have found Legionella if cooling towers had been included in the survey, but low counts (less than 10 ppm) of Legionella bacteria in cooling towers are not likely to present a high risk to patients if the towers are equipped with efficient drift eliminators and are located at least 100 feet from outdoor air intakes and operable windows.

  • Studies have shown that the presence of Legionella in water samples can indicate risk of disease. It's possible that no cases of Legionnaires' disease will occur over a period of time in a building in which Legionella proliferate in water systems. But, just as failing to wash hands increases the risk of infection, allowing the bacteria to proliferate increases the risk of disease.
  • Test results usually give clues, but not conclusions. Some ask, "Why should we test if we don't know what to do with the results?" Good question. It's important to understand how to interpret Legionella test results.

    If you insist on a yes/no answer--our institution is okay or it isn't--you will probably be frustrated. Although OSHA and others have provided tables that list appropriate responses to various Legionella counts in buildings occupied by individuals in generally good health, there is no specific action level above which there is risk and below which there isn't.

    The Allegheny County guideline suggests that hospitals take action based on the percentage (30 percent) of samples positive in a given screening rather than the Legionella level per sample. The 30-percent rule is useful, but there are exceptions. For example, if only 15 percent of the samples test positive, but the positive samples were collected from faucets in bone marrow transplant patient rooms, corrective action is needed because these patients are at significant risk.

    Test results usually indicate specific changes to make in the operation or maintenance of mechanical systems, rather than confirming a building as safe or unsafe. If a high percentage of samples test positive, it's clear that corrective action is needed. What's more often the case, though, is that one or two samples out of 15 to 20 test positive.

    In such instances, one must consider what makes certain parts of the domestic water system conducive to Legionella growth. In the author's experience, the answer is often obvious, and appropriate risk reduction measures can be implemented.

  • Sampling cannot replace preventive maintenance or patient testing. You must have pure motives. Don't collect samples with the notion that a clean test report will justify inadequate environmental preventive measures or patient testing. You still need to maintain water heaters, cooling towers and piping to minimize Legionella growth conditions, and test patients for Legionella per CDC recommendations.

    Moreover, it is presumptuous to assume that a building is free of Legionella bacteria even when all samples test negative because Legionella can be in sites not sampled, and certain conditions can cause Legionella counts to soar unexpectedly. Perhaps the safest way to view results is to listen only to bad news: Consider corrective measures if samples test positive, but continue preventive measures if the results are negative. Don't let a clean test report give you a false sense of security.

    Sampling may be smart risk management if you implement preventive and corrective measures, but not if you don't. Legionnaires'-related lawsuits are not uncommon; several are in process as of this writing. Some attorneys encourage building owners to be proactive in implementing Legionella-preventive measures in water systems, but others suggest a head-in-the-sand approach, contending the best way to avoid claims is to know nothing and do nothing.

    Based on experience as an expert witness, two observations are offered: (1) The only way to avoid Legionella-related claims is to ensure that no one in your building contracts the disease, and that is best accomplished by maintaining water systems to control Legionella; and (2) If you find Legionella in environmental samples, you should take reasonable corrective action. If you do, the sampling program will likely help you defend a lawsuit if a case of Legionnaires' occurs despite your preventive efforts. If you don't take corrective action, the sampling could hurt your defense.

  • Legionella problems can be mitigated at a reasonable cost. As mentioned above, finding positive samples does not necessarily mean that a disinfection system is required. In some cases, minor adjustments are enough--changing pump cycling, cleaning hot water tanks periodically, removing dead piping, increasing hot water temperatures and replacing equipment that would have been replaced soon anyway. A worst case scenario: If a domestic water disinfection system is required, a 200- to 500-bed hospital with two hot-water loops will spend about $50,000 to $75,000, including installation.
  • Sampling can be an impetus for risk reduction. Routine environmental screenings make health care professionals more cognizant of Legionella. The engineering department will be more diligent in Legionella-related preventive maintenance, just as a daily weigh-in will make a boxer more cognizant of his eating habits. And clinicians will be more likely to suspect the disease, and to order diagnostic laboratory tests for patients, especially if Legionella is found in the water.

    Sampling should also increase communication between facility management, infection control and the medical staff. Ideally, the facility manager will be alerted whenever legionellosis is detected in patients so that he or she will know to investigate mechanical systems, and medical staff members will be alerted whenever water samples are positive so that they can be especially watchful for cases of legionellosis.

  • Cost is always a factor. A hypothetical 300-bed hospital with two cooling towers, two hot-water tanks and one decorative fountain will spend $1,440 to $2,340 per screening assuming it's collecting 18 samples and paying a commercial laboratory $80 to $130 per culture.

    You should figure four screenings for the first year, then two to four in subsequent years if previous test results are favorable. There's no need to pay a consultant to collect samples provided your employees receive training or thorough written instructions. The laboratory fees therefore represent the total cost, other than incidentals such as sampling bottles, swabs and overnight shipping to the laboratory.

    Careful attention required

    A reactive approach to Legionnaires' disease--ignoring water systems until a case of disease is identified--is no longer an option for JCAHO-accredited institutions. Whether or not you make environmental sampling a part of your proactive program is a decision that deserves careful attention. Some day you may have to explain your decision to a distraught patient or surviving family member, a nosy reporter or a ruthless attorney.

    Matthew R. Freije, president of HC Information Resources, Fallbrook, Calif., is the author of the book "Legionellae Control in Health Care Facilities: A Guide for Minimizing Risk" and the document "Management Plan for Legionella and Other Waterborne Pathogens." He also teaches a Legionella prevention training course. His firm is not affiliated with a lab and does not perform,
    nor profit from, laboratory culture services. He can be reached at mf@hcinfo.com.

    Hospital surveys for Legionella contamination of water distribution systems

    Reference Location Hospitals Percent with Legionella Isolate
    HMSO United Kingdom 40 70% Legionella pneumophila, Serogroup 1
    Alary Quebec 84 68 L pneumophila, Serogroups 1-8
    Vickers Western Pennsylvania 15 60 L pneumophila, Serogroups 1-6
    Patterson United Kingdom 69 55 L pneumophila Legionella species
    Marrie Nova Scotia 39 23 L pneumophila Legionella longbeachae
    Liu United Kingdom 17 12 L pneumophila, Serogroups 1,4,6
    Note: This table was excerpted and adapted from "Resolving the Controversy on Environmental Cultures for Legionella: A Modest Proposal" by Victor L. Yu. Infect Control Hosp
    Epidemiol 1998;19:893-897. The data do not include samples taken at cooling towers.

    To respond to this article, please click here.

    Click here for a FREE subscription to Health Facilities Management.