Page 10: Canadian Guidelines for Domestic Reclaimed Water for Use in Toilet and Urinal Flushing

Appendix B: Additional risk assessment information and calculations

Health-based targets

Health-based targets are the "goal-posts" or "benchmarks" that have to be met to ensure the safe use of recycled water. In Canada, common forms of health-based targets are numerical guideline values and/or performance targets for chemical and microbiological hazards. In relation to chemicals, a guideline value is generally the concentration or measure of a water quality characteristic that, based on present knowledge, does not pose any significant risk to the health of the consumer over a lifetime of consumption. Guideline values for microbiological hazards focus on reducing acute risks and generally rely on monitoring for indicator organisms. Performance targets describe the reduction in risk to be provided by measures such as treatment processes (aimed at reducing hazards) and on-site controls (aimed at reducing both hazards and exposure). The wide array of microbiological pathogens makes it impractical to measure for all of the potential hazards; thus, performance targets are generally framed in terms of categories of organisms (e.g., bacteria, viruses and protozoa) rather than individual pathogens.

Disability-adjusted life years (DALYs)

The most recent edition of the World Health Organization's (WHO) Guidelines for Drinking-water Quality (WHO, 2004) adopts 10-6 disability-adjusted life year (DALY) as a reference level of risk. In Canada, the Federal-Provincial-Territorial Committee on Drinking Water has also chosen to use this target as an acceptable level of risk from microbiological contaminants in drinking water. The Australian Guidelines for Water Recycling (NRMMCEPHC, 2006) also cites this level of risk. Havelaar and Melse (2003) note that the concept of the DALY has been introduced as a common unit of risk to compare different health effects that vary in severity-for example, from mild diarrhoea to the most severe outcome, death. The basic principle of the DALY is to weigh each health effect for its severity, using standardized severity weights provided within the Global Burden of Disease project (Murray and Lopez, 1996). This weight is multiplied by the duration of the health effect and the number of people affected by the particular outcome. When all of the health outcomes caused by a particular agent are summed, the result is an estimate of the burden of disease attributable to this agent. The key advantage of the DALY as a measure of public health is cited as its aggregate nature, combining years of life lost (quantity) with years lived with disability (quality).

Other authorities use measures such as risk of infection or risk of illness. The U.S. Environmental Protection Agency target is a risk of infection of 10-4 from pathogens in drinking water (one additional infection per 10 000 people) (U.S. EPA, 2004). The reference level of 10-6 DALY is approximately equivalent to a lifetime additional risk of cancer of 10-5 (i.e., 1 case per 100 000 people) or, for a diarrhoea-causing pathogen with a low fatality rate, an annual risk of illness of 10-3 for an individual. To place this level of risk in a Canadian context, there are approximately 1.3 cases of enteric disease annually per person in this country. The reported rate of diarrhoeal illness for specific pathogens (from all routes of exposure) in Canada (for the year 2004, rate per 100 000 population) is shown in Table B1.

Table B1: Rates of selected notifiable diseases in Canada, 2004 Table B1 Footnote a
Notifiable disease Rate per 100 000 population
Age group:
all ages
Age group:
1-4 years

Table B1 Footnotes

Table B1 Footnote 1

From PHAC (2005).

Return to Table B1 footnote a referrer

Campylobacteriosis 30.22 60.90
Cryptosporidiosis 1.85 11.56
Giardiasis 13.08 47.29
Shigellosis 2.35 5.55
Verotoxigenic E. coli (O157:H7) 3.36 13.15

Dose-response models

Risk assessments are commonly based on data and dose-response models developed from human feeding studies. Log-normal, beta-Poisson and exponential distributions (Table B2) can be used to determine probabilities of infection following exposure to different doses of the pathogen (Haas et al., 1999). The dose from reclaimed water used for flushing toilets is expected to be low, as the water is not intended for ingestion. The dose is derived from the potential for accidental ingestion and exposure, as described in Section 4.3.

Table B2: Dose-response relationships for reference organisms
Organism Distribution Model ParametersTable B2 Footnote a

Table B2 Footnotes

Table B2 Footnote 1

a and r are parameters describing probability of infection; d = dose; N50 = median infective dose; P = probability of infection. Model parameters are as described in Haas et al. (1999), except for C. parvum, where the value calculated in from Messner et al. (2001) is used.

Return to Table B2 footnote a referrer

Enteric virus (rotavirus) Beta-Poisson P = 1 - [1 + d/N50(21/a - 1)]-a a = 0.27
N50 = 5.60
Bacterium (E. coli O157:H7) Beta-Poisson P = 1 - [1 + d/N50(21/a - 1)]-a a = 0.2099
N50 = 1120
Protozoan (Cryptosporidium parvum) Exponential P = 1 - exp(-rd) r = 0.018

Risk characterization

Using a burden of disease approach, the risk characterization in these guidelines uses the information from the hazard identification, dose-response and exposure assessments to estimate the magnitude of risk. A deterministic approach is used here to calculate a health-based target for the reference pathogens in the reclaimed water. This approach uses single estimates for exposure volumes and number of exposure events (e.g., point estimates), which has the disadvantage of neglecting to address variability and uncertainty and also tends to rely on conservative and even worst-case values. A stochastic analysis would help address these disadvantages, but would require more information than is currently available. A sample risk characterization is shown in Table B3. Single estimates are used for exposure volumes and number of exposure events. The estimates used are believed to be conservative. Formulae used in the calculations are shown in Box B1.

Table B3: Potential disease burdens for aerosols from toilet flushing
  Cryptosporidium Rotavirus E. coli O157:H7

Table B3 Footnotes

Table B3 Footnote 1

Concentrations of Cryptosporidium and rotavirus in raw sewage are taken from NRMMC-EPHC (2006); numbers of adenovirus are used as an indication of rotaviruses because of the lack of enumeration methods for rotavirus.

Return to Table B3 footnote a referrer

Table B3 Footnote 2

Concentration of E. coli O157:H7 is calculated assuming that 2% of the maximum number of generic E. coli enumerated in raw wastewater samples from Canadian cities are pathogenic (6.2 × 106; Payment et al., 2001). More information is needed to refine this estimate.

Return to Table B3 footnote b referrer

Table B3 Footnote 3

Based on log reductions shown in Tables D1 and D2 (see Appendix D); hazard concentrations reduced by secondary treatment, coagulation, filtration and disinfection.

Return to Table B3 footnote c referrer

Table B3 Footnote 4

Constants and models used to calculate risk of infection are shown in Table B2.

Return to Table B3 footnote d referrer

Table B3 Footnote 5

Havelaar and Melse (2003).

Return to Table B3 footnote e referrer

Table B3 Footnote 6

DALY per case based on Havelaar and Melse (2003).

Return to Table B3 footnote f referrer

Table B3 Footnote 7

The proportion of the population susceptible to developing disease following infection. The figure of 6% for rotavirus is based on the fact that infection is common in very young children. The 6% equates to the percentage of population aged less than five years.

Return to Table B3 footnote g referrer

Organisms per litre in source waterTable B3 Footnote 1Table B3 Footnote 2 2000 8000 1.2 × 105
Log reduction provided by treatmentTable B3 Footnote c 5 6 6
Exposure per event (L) 1 × 10−5 1 × 10−5 1 × 10−5
Dose per event (organisms) 2 × 10−7 8.0 × 10−8 1.2 × 10−6
Number of events per year 1100 1100 1100
Dose-response constantsTable B3 Footnote d 1.8 × 10-2 a = 2.7 × 10-1 a = 2.1 × 10-1
N50 = 1120
Probability of infection per organism 1.8 × 10−2 2.7 × 10−1 4.8 × 10−3
Risk of infection (Pinf) (probability of infection per event) 3.6 × 10−9 4.6 × 10−8 6.0 × 10−9
Ratio of illness/infectionTable B3 Footnote e 0.70 0.88 0.53
Risk of illness (Pill) per event 2.5 × 10−9 4.0 × 10−8 3.2 × 10−9
Risk of illness (per year, i.e., 1100 events) 2.8 × 10−6 4.4 × 10−5 3.5 × 10−6
Disease burdenTable B3 Footnote f (DALY per case) 1.5 × 10−3 1.3 × 10−2 5.5 × 10−2
Susceptibility fraction (%)Table B3 Footnote g 100 6 100
DALY per year 4.2 × 10−9 3.5 × 10−8 1.7 × 10−8

Box B1: Formulae used in Table B3
Box B1: Formulae used in Table B3
1. Dose per event = Source water concentration × log reduction × exposure
2. Pinf = Dose-response models and parameters as shown in Table B2
3. Pinf per year = 1 − (1 − Pinf)N
where N = number of exposures per year
For lower levels of risk, this can be approximated to:
Pinf per year = Pinf × N
4. Pill per year = Pinf per year × ratio of illness to infection
5. DALY per year = Pill per year × DALY per case × susceptibility fraction

Another approach is to calculate treatment goals to achieve a health target of 10-6 DALYFootnote 3 for the specified uses of reclaimed water, based on the initial concentration of a reference pathogen in the untreated source water. The disease burden, in DALYs, is calculated from the estimated exposures to pathogens in the recycled water. Because the reductions depend on the initial concentrations and the associated exposure, uses involving higher exposures will require greater reductions of pathogens from treatment.

The log reductions required to reach a target of 10-6 DALY per year in treated reclaimed water can be calculated. Dose equivalents to 10-6 DALY (dalyd) can be determined using the formulae given below:

DALY per year = Pinf per year × N × ratio of illness to infection × DALY per case × susceptibility fraction

Since the target DALY per year is 10-6 in this example, this equation can be written to solve for the dose equivalent:

Dose Equivalent Equation

The equation used to calculate the pathogen dose that is equivalent to 10-6 Disability Adjusted Life Years per person per year. description follows
Text Equivalent
The dose equivalent is calculated by dividing the target Disability Adjusted Life Year (DALY) by the product of the following : the DALY weight per case, multiplied by the probability of infection by the organism, multiplied by the ratio of illness given infection, multiplied by the susceptibility fraction

Where concentrations of organisms in source water are known, required log reductions (Table B4) can be calculated with the following formula:

Log reduction = log (concentration in source water × exposure (L) × N ÷ dalyd)

Where: L = volume, in litres
N = number of times the exposure occurs in one year
dalyd = Doses equivalent to 10−6 DALY

Table B4: Required log reductions
Organism Dose
equivalentTable B4 Footnote a
Required log reductions
Based on aerosols from toilet flushing Based on cross-connectionTable B4 Footnote b

Table B4 Footnotes

Table B4 Footnote 1

Doses equivalent to 10-6 DALY.

Return to Table B4 footnote a referrer

Table B4 Footnote 2

Based on worst-case assumption of 1 person per 1000 (10-3) consuming 1 L/day for 365 days.

Return to Table B4 footnote b referrer

C. parvum 5.3 × 10−2 2.6 4.1
Rotavirus 5.5 × 10−3 4.2 5.7
E. coli O157:H7 7.1 × 10−3 5.3 6.8

Page details

Date modified: