Cross Connection and Backflow Prevention – Underutilized Protection for Potable Water

April 17, 2018 | Blog

Written By: Jean Holloway

Cross connections and accompanying backflow are a common occurrence in plumbing and piping systems of every kind, and one of the most frequently overlooked hazards to potable water, and therefore, to public health. Backflow is a common occurrence because there are so many everyday opportunities to cross-connect potable and non-potable water supplies. Cross connections are so frequently overlooked because most people outside the water or plumbing industry don’t realize the hazards that the backflow they cause can pose to drinking water and therefore to public health and safety.

A cross connection is any direct, unprotected contact of a potable water supply with a non-potable or potentially contaminated liquid, solid or gas supply. Defined in that fashion, it would seem a simple matter to prevent such cross connection and thus prevent contamination and resulting illness – just don’t connect the pipes. The reason it’s not that simple is that cross connections and backflow events can occur in everyday situations that are as common as garden hoses, sink sprayers and other mundane items that don’t appear to present any hazard at all. They also don’t have to be the result of one line being directly connected to the wrong supply or outlet, as the name “cross connection” may imply.

Anytime a potable water line or fixture is in close enough proximity to a non-potable material source or container could be an opportunity for cross connection. If a garden hose (the most common source of all backflow) is left on and dropped into a bucket, chemical sprayer, child’s swimming pool or any other vessel containing non-drinkable material, there is the potential for backflow into the potable water supply. The key is a difference in pressure between the two connections, flows or vessels.

There are two types of backflow – backpressure and backsiphonage. Backsiphonage occurs when the supply side pressure is less than atmospheric pressure, and non-potable materials are sucked into the potable water system in much the same way that a beverage is sucked through a drinking straw. Backpressure occurs when pressure on the discharge side of a connection is greater than the supply side, and non-potable substances are pushed back into the potable water system, in the same way, blowing back through that same drinking straw pushes liquid back into the glass.

Events like a water line break that cause a sudden drop in supply pressure can cause back siphonage. The sudden drop in water supply pressure sucks whatever is available into the line and back into the water system. If whatever is available happens to be toxic or disease-causing, the potential exists to contaminate the potable water system and to make a lot of people sick.

Elevated storage tanks or high rise buildings with inadequate backflow protection can cause backpressure. The greater pressure caused by a taller “column of water” pushes back on the supply line and causes potential contamination to flow back into the potable water system. Firefighting water demand or the use of booster pumps can also cause backflow, along with a host of other everyday opportunities to draw or force non-potable material into a potable water supply. This potential is just as true for a single house using an on-site well as it is for a public community water supply such as those owned by municipalities. “Backflow situations have undoubtedly affected health since the origin of plumbing.” From 1981 to 1998 the Centers for Disease Control documented 57 waterborne disease outbreaks directed related to cross connections, outbreaks which resulted in 9,734 illnesses. EPA’s issue paper on the topic of cross connection and backflow further estimates that outbreaks related to cross connection are under-reported, in large part because the resulting illnesses are not recognized as backflow-related.

So, given the potential for poisoning an entire population or contaminating an entire community’s water supply indefinitely, why isn’t the value of cross connection and backflow prevention programs more recognized and why aren’t such programs more prevalent? One reason may be that the topic is not a “hot button” health issue like AIDS or an illness that conjures up images of horrible deaths like Cholera. Another reason is that the usual waterborne illness presents as gastrointestinal upset that can be mistaken for everything from food poisoning to the seasonal flu It often takes multiple cases presented in a short time span and a good deal of investigation before an outbreak of illness is even traced back to a drinking water cross connection point of origin.

A third reason is that “…the value of cross-connection control may not be readily apparent” when the magnitude of potential illness is not recognized or the drinking water as a source of the illness is not even considered. When public officials consider the cost-benefit value of establishing a cross connection control and backflow prevention program, they are hard-pressed to spend public funds on something that doesn’t have readily apparent “bang for the buck” in benefits to public health and welfare. This reluctance is even more likely if it is hard to demonstrate a direct connection between contamination, illness outbreak and cross connection.

Cross connections are not only hard to prove as a source of illness, they are often un-reported or at least under-reported. One state official estimated that about 1,200 backflow incidents occur in his state per year, yet his state only reported 15 such incidents from 1970 to 2002.3 Another official estimated that the incidence of actual backflow events may be greater than reported by a factor of 10.4 When the University of Southern California Foundation for Cross Connection Control and Hydraulic Research (USC FCCCHR) prepared a “Summary of Case Histories” in 1993, they covered 397 incidents over a period of 90 years from 1903 to 1993. The Chief Engineer of the foundation estimated that 90% of incidents were not documented sufficiently to their standards to even be included in the study. This lack of sufficient documentation usually results from the fact that many illnesses are not directly attributable to cross connection and resulting backflow, and are simply chalked up to “something you ate,” a “stomach bug,” or some other nebulous source.

Besides detrimental health effects, cross connections can cause operational problems and additional costs that negatively affect all water system customers and drinking water consumers. A cross connection can allow corrosive materials like acids or carbon dioxide to enter a drinking water system. Even if not present in sufficient levels to cause illness, corrosive substances in the distribution system can cause many problems and associated costs by hastening the deterioration of pipes and connections, leaching toxic metals like lead and copper from pipes, causing taste, odor and color problems, increasing scaling and precipitate in pipes and bringing on any number of public relations problems associated with all those things.

Backflow from a cross connection, regardless of source, can introduce microbes into the water distribution system, where they can attach to pipe walls and multiply into a biofilm. These biofilms are often impervious to the usual disinfectants. Even if the microbes are not disease-causing in and of themselves, the biofilms they form can trap and concentrate nutrients, which promote the growth of pathogenic organisms. Backflow can also introduce such nutrients that promote the proliferation of existing biofilms.

Once the dangers of cross connection and backflow are recognized, there are some signs to look for in identifying a backflow incident before it causes widespread illness or becomes a wider operational problem. Some possible signs include an increase in customer complaints, drops in system operating pressure, drops in the disinfectant residual, water meters running backward and the number of coliform detects. An increase in customer complaints of taste, odor and/or color can be a primary indicator of a possible backflow incident, but like a drop in operating pressure, it can mean that the event has already occurred rather than something that can aid in preventing an incident before it happens.

Entry into the system from a backflow incident can also cause an atypical drop in disinfection residual by introducing organic carbons or other “reducing” agents that use up available disinfectant. A sudden increase in total coliforms can be an indication that such contaminants have entered, possibly through a backflow event. Likewise, when check valves or similar apparatus fail, water meters can run in reverse, indicating a change in the direction of water flow (backflow). All these indicators are signs that an incident has already occurred and none will help prevent those incidents unless the signs are used to detect and remedy the source of the backflow or cross connection before future incidents can occur.

The best protection from the cross connection is a systematic program of identification, prevention and testing to ensure that events are stopped BEFORE they happen rather than after the problems occur. This program can incorporate plumbing standards and codes, local laws and ordinances, construction inspection, backflow device and assembly testing, training and, most of all, diligence and alertness on the part of system personnel for any indications of potential or actual backflow incidents. Documentation can help to ensure proper reporting, but more importantly, proper recognition and prevention of this very common and potentially dangerous threat to public health and safety. Local officials as a body may not recognize the potential for a public health crisis presented by backflow, but any individual who becomes ill, even temporarily after a backflow incident can attest to the value of preventive measures.


This artile was produced by the Southeastern RCAP (SERCAP). Learn more about SERCAP at