Guidelines for Canadian Drinking Water Quality: Guideline Technical Document - Haloacetic Acids

12.0 Rationale / * Errata

Because haloacetic acids are formed in drinking water primarily as a result of chlorination of organic matter present in raw water supplies, it is important to recognize the substantial benefits to health associated with disinfection by chlorination. The use of chlorine has virtually eliminated waterborne microbial diseases, because of its ability to kill or inactivate essentially all enteric pathogenic microorganisms, including viruses and bacteria from the human intestinal tract. Chlorine is the most convenient and easily controlled disinfectant; it is a strong oxidant for which a residual can be maintained in the distribution system to prevent bacterial regrowth. Although the use of chlorine can lead to the formation of DBPs such as HAAs, efforts to manage HAA levels in drinking water must not compromise the effectiveness of disinfection.

HAAs and THMs are the two major groups of DBPs found in drinking water and generally at the highest levels. The concentrations of these contaminants can be used as indicators of the total loading of all DBPs that may be found in drinking water supplies. In the absence of information on other DBPs, control and management of HAAs and THMs should reduce exposure to and risk from other by-products. When appropriate drinking water treatment strategies are implemented to reduce HAAs and THMs, the levels of other halogenated DBPs may also be reduced in the process.

There are sufficient scientific data available to derive health-based targets for four HAAs: MCA, DCA, TCA and DBA. MCA is classified as Group IV (unlikely to be carcinogenic to humans). A health-based target concentration of 0.1 mg/L can be calculated for MCA in drinking water, based on changes in body, liver, kidney and testes weights observed in rats. DCA is classified in Group II (probably carcinogenic to humans), based on sufficient evidence in animals and inadequate evidence in humans. A health-based target concentration of 0.01 mg/L can be calculated for DCA in drinking water, based on liver tumours observed in both mice and rats. TCA is classified in Group III (possibly carcinogenic to humans), based on limited evidence of carcinogenicity in experimental animals and inadequate evidence in humans. A health-based target concentration of 0.3 mg/L can be calculated for TCA in drinking water. Although animal studies have shown a link between exposure to TCA and liver tumours in mice only, it is still uncertain whether the mechanism causing these tumours is relevant to humans. MBA is classified in Group VI (unclassifiable with respect to carcinogenicity in humans), based on inadequate data from animal studies. No health-based target concentration can be established for MBA at this time. DBA is classified in Group II (probably carcinogenic to humans), based on sufficient evidence in animals and inadequate evidence in humans. A health-based target concentration of 0.002 mg/L can be calculated for DBA in drinking water, based on tumours in several organs observed in both mice and rats.

Recent Canadian exposure data for surface water sources show that the HAAs consistently found at the highest concentrations in Canadian distribution systems were DCA and TCA.

Although the proportion of each HAA will vary according to conditions, DCA can commonly represent 40-60% of the total HAA concentration. DBA makes up only a small fraction (generally less than 6%)* of the total HAA concentration.

The removal of HAAs after their formation in drinking water supplies is not considered to be the best approach to reduce exposure to HAAs. The most efficient and practical way to reduce HAA concentrations in finished waters is to prevent their formation, primarily through the removal of organic precursors. Although pH adjustments may help reduce HAA formation, they may cause a corresponding increase in the formation of other DBPs, including THMs.

Although health-based targets can be established for four of the five HAAs, and considering the technological limitations associated with reducing individual HAA levels in drinking water while maintaining effective disinfection, the Federal-Provincial-Territorial Committee on Drinking Water is establishing a MAC of 0.08 mg/L (80 µg/L) for Total HAA5 in drinking water based on a running annual average rather than individual guidelines. This is consistent with the approach taken by the U.S. EPA, which established a maximum contaminant level based on best available technology for these same HAAs.

---

Errata

Section 12, page 61

Although health-based targets can be established for four of the five HAAs, and considering the technological limitations associated with reducing individual HAA levels in drinking water while maintaining effective disinfection, the Federal-Provincial-Territorial Committee on Drinking Water is establishing a MAC of 0.08 mg/L (80 µg/L) for Total HAA5 in drinking water based on a running annual average rather than individual guidelines. This is consistent with the approach taken by the U.S. EPA, which established a maximum contaminant level based on best available technology for these same HAAs.

---

The MAC of 0.08 mg/L for total HAAs in drinking water is established, using a risk management approach based on the following considerations:

• 1a. Although the health-based target for DBA is the lowest value calculated, it is also found as a very small proportion of the HAA mixture and is not considered to be the appropriate driver for a health-based guideline. Mean levels of DBA observed are well below the health-based target of 0.002 mg/L for DBA.
• 1b. The health-based target for DCA is the lowest value that should be considered. The concentration of DCA in drinking water representing an "essentially negligible" risk is 0.01 mg/L, but this level cannot be achieved in distribution systems without compromising the effectiveness of disinfection.
• 2. The PQLs, based on the capability of laboratories to routinely measure HAAs within reasonable limits of precision and accuracy, vary according to the individual HAAs and the method used. However, all the PQL values are well below the MAC.
• 3. The MAC must be achievable at reasonable cost. Based on the limited available Canadian data, it is estimated that 88% of existing treatment plants serving populations over 5000 and 56% of existing treatment plants serving populations less than 5000 can be expected to achieve HAA concentrations of 0.08 mg/L. By optimization of treatment processes and the use of advance treatment technologies to remove organic compounds prior to disinfection, it is possible to reduce HAA5 below 0.08 mg/L.

The estimated lifetime cancer risk associated with the ingestion of drinking water containing HAAs at 0.08 mg/L is greater than the range that is considered generally to be "essentially negligible" (i.e., between 10^-5 and 10^-6). Based on the incidence of liver cancer in animal studies for DCA, the estimated lifetime risk associated with ingestion of water containing HAAs at 0.08 mg/L is 3.2 x 10^-5 to 4.8 x 10^-5 for DCA proportions of 40 % to 60%, respectively, of total HAAs. Although exposure to HAAs at the guideline level may carry a lifetime risk from DCA that is higher than would normally be considered negligible, this risk is calculated using a very conservative approach that typically overestimates potential risk.

It is recommended that utilities strive to maintain HAA levels as low as reasonably achievable without compromising the effectiveness of disinfection. As part of its ongoing guideline review process, Health Canada will continue to monitor new research in this area and recommend any change(s) to the guideline that it deems necessary.