Two Keys to Success in Selecting “Evidence-based” Legionella Control Measures

Updated November 22, 2024

One of the important criteria for a Legionella control measure is that it be evidence-based — meaning there is scientific evidence the measure will likely be effective in minimizing Legionella.

Two keys to success in selecting “evidence-based” Legionella control measures are to (a) determine the best source of data (“evidence”) for a particular measure and (b) correctly obtain and interpret the data.

Determining the best source of evidence

The best source of evidence for the principle on which a Legionella control measure is based is usually one or more peer-reviewed, published scientific studies, preferably studies performed in buildings rather than only in laboratories. The best source of evidence for the specific application of the principle is usually from the building where the measure is to be applied. Put more simply, scientific studies of several building water systems are often needed to establish principles, but for a given building, the best evidence for specific applications is from that building.

The distinction is important. Insisting on scientific studies to support every specific procedure could result in missed opportunities to prevent Legionnaires’ disease. It could also lead to the wrong answer because of differences between your building and the ones studied.

Consider flushing as an example. Let’s say one of the members of your water management team suggests ensuring all faucets are used for at least two minutes every week, which may require flushing faucets that are infrequently used. Other team members may agree that scientific studies show stagnation in domestic water (plumbing) systems promotes Legionella growth but refuse to implement flushing procedures until published studies show exactly how often and long to flush each faucet.

Studies have shown that minimizing stagnation is important for Legionella control. However, since every building is different and there are many variables among buildings, no study is likely reliable for determining the details – frequency, breadth (e.g., number of faucets), and duration of flushing. What a study may conclude for a building or group of buildings may not be effective in other buildings. Such studies could even do more harm than good if researchers attempt to draw broad conclusions from too little data – the team may end up flushing too much or too little, or not at all.

Thus there’s a point at which the best source of data (evidence) transfers from scientific studies of other buildings to your building. In the flushing example, the best data for determining how often and long flush, or whether to flush at all, will typically be disinfectant and temperature test results in your building, not other buildings. The design of the plumbing system, and Legionella test results, should also be considered. (If you have access to LAMPS training notes, see number 4.031 for details about establishing a flushing program.)

There’s nothing wrong with questioning a frequency, duration, chemical concentration, or other detail. If you decide to implement a control measure but change the recommended frequency from once a week to once a month or twice a week, that’s fine, particularly if you are testing (properly) for Legionella to validate your decisions. But refusing to implement the control measure altogether until a scientific study tells you exactly how often, how long, and how much could be a decision that’s difficult to defend if someone gets Legionnaires’ disease from your water system — especially if you are not testing for Legionella to validate your program. You wouldn’t recommend that a loved one continue smoking until scientific studies determine exactly how many cigarettes a day will result in cancer.

In general, then, look to studies for principles but to your building-specific data for the details.

Correctly obtaining and interpreting data

If you rely on a scientific study, you must correctly apply the data generated in that study to make good decisions. Don’t just read the conclusions drawn by the authors. Some researchers draw conclusions that are too broad based on the data. Read through the methods the researchers used for generating the data, look at the data carefully, and then draw your own conclusions before reading the authors.’ Incorrectly interpreting the evidence generated by a study is no better than having no study to cite.

If you rely on Legionella test results to validate your water management plan, then collect and apply the data properly (Click here for the training program, “Sampling Water Systems to Validate Legionella Control”). Sample often enough. Collect enough samples each round. Choose the right sample locations and sample types. Collect samples properly. Record data thoroughly. Have the samples tested by a highly qualified laboratory with acceptable reporting limits and speciation? And, interpret the results correctly.

A reasonable approach?

Consider the following approach to establishing and maintaining effective, “evidence-based” Legionella control measures:

1. Establish specific control measures based on principles supported by scientific studies.
2. Properly test for Legionella to validate your program.
3. Monitor key water quality parameters, preferably with sensors, to get frequent feedback on conditions conducive to Legionella growth.
4. Adjust your control measures based on test results, then go back to number 2.

Do you think this is a reasonable approach? If not, what criteria do you use for ensuring Legionella control measures are evidence-based? Please comment below.