A comparison of the regulatory frameworks governing microbial testing of drinking water in three Canadian provinces

Abstract

In Canada, the provinces have primary responsibility for water management, including drinking water management; the federal government has comparatively more limited water-related responsibilities. Currently, Canada does not have a harmonized federal drinking water strategy or law (although it does have non-binding guidelines); thus provinces have developed different approaches to drinking water management. In this study, key features of the current regulatory frameworks for microbial testing in drinking water quality management are examined in Canada’s three most populous provinces – British Columbia, Ontario and Quebec. Salient regulatory features including the types of drinking water samples, water quality standards, testing frequency and system size are described and compared. Distinct differences were found in provincial approaches to drinking water quality assessment, and microbial water quality management is variable not only among, but also within, provinces. This finding of inter- and intra-province variability in drinking water quality management shows that these three provinces are approaching similar challenges in different ways. In turn, these different approaches demonstrate that regulatory frameworks can be adapted in response to drinking water management challenges. Regulatory frameworks should be flexible and adaptable to new knowledge and scientific developments, such as molecular testing methods, in order to facilitate their translation into water management tools.

Au Canada, les provinces sont les principales responsables de la gestion des ressources en eau; comparativement, le gouvernement fédéral a une responsabilité plus limitée dans ce domaine. En l’absence d’une stratégie fédérale ou de législation nationale sur l’eau potable (malgré l’existence de ligne directrices non-contraignantes), les provinces ont élaboré des approches différentes pour la gestion de l’eau potable. Dans cette étude, la variabilité des caractéristiques clés des cadres réglementaires de gestion de la qualité de l’eau potable par l’analyse microbiologique a été examinée dans les trois provinces canadiennes les plus peuplées – soit en Colombie-Britannique, en Ontario et au Québec. Les éléments régulateurs saillants, y compris les types d’échantillons d’eau potable, les normes de la qualité de l’eau, la fréquence des analyses microbiologiques et la taille des systémes de distribution, sont décrits et comparés. Des différences prononcées ont été identifiées entre les approches sur la gestion et l’évaluation microbiologique de la qualité de l’eau potable, qui sont non seulement variable entre, mais aussi au sein, des provinces. Cette constatation de la variabilité inter- et intra- provinciale en matiére de gestion de la qualité de l’eau potable dénote que ces trois provinces ont choisi d’approcher des défis semblables de manières différentes. Ces différences démontrent aussi que les cadres règlementaires peuvent aussi être adaptés pour répondre aux difficultés locales que peut présenter la gestion de l’eau potable. Ainsi les cadres règlementaires devraient être flexibles et adaptables aux développements scientifiques, tels que les méthodes d’analyse moléculaires, afin de faciliter leur translation en outils de gestion de l’eau.

Introduction

Water resources across Canada’s provinces and territories are under ever-increasing pressure. Population growth, climate change, increasing land use and aging infrastructure all contribute to growing challenges surrounding water quality and quantity, Canada-wide. Traditional resource management approaches are unlikely to be successful in addressing multifactorial impacts on water resources; alternative and novel approaches will be necessary. Monitoring of microbial water quality, from source to tap, is an internationally accepted cornerstone of drinking water quality assessment; to fully implement this approach across Canada, novel strategies are needed. Alternative microbial water quality monitoring approaches, including quantitative polymerase chain reaction (qPCR) and microbial source tracking (MST), have been developed in response to limitations in current approaches (chiefly, culturing of indicator bacteria), which are slow and non-specific (Hrudey et al. 2006; Signor and Ashbolt 2006). However, to date, there has been limited uptake of emerging approaches for routine monitoring of all types of water; Canadian jurisdictions do not regulate the use of molecular methods for microbial testing of water, for indicator organisms or pathogens. The lack of uptake of molecular methods extends beyond Canadian jurisdictions; molecular methods are absent from the most recent edition of Standard methods for the examination of water and wastewater (Rice et al. 2012) and are mentioned only in reference to viruses (Adenovirus, Astroviruses and Caliviruses) in the World Health Organization (WHO) Guidelines for drinking water quality (WHO and OECD 2008). The United States Environmental Protection Agency (USEPA) has demonstrated the utility and appropriateness of qPCR as an alternative to culture-based methods for assessment of beach monitoring (USEPA Office of Water 2010) and has moved towards regulation of qPCR for enterococci for beach monitoring (US Clean Water Act, §304(a); USEPA Office of Water 2012). Molecular methods for water quality assessment are extensive in the peer-reviewed literature, published by university and government scientists in Canada, the US, the European Union and beyond, but there is limited uptake at a regulatory level (Brettar and Höfle 2008; Dunn et al. submitted). To consider how to best integrate emerging water quality assessment tools into provincial legislative frameworks in Canada requires a review of existing water quality regulatory frameworks.

In Canada, water management is shared by the federal and provincial governments. The federal government’s water-related authorities include navigable waters, fisheries and transboundary waters. Provinces are primarily responsible for water management and there is considerable diversity in provincial legislation and policies regarding water management (Hill et al. 2007, 2008). A federal agency, Health Canada, coordinates the Federal-Provincial-Territorial Committee on Drinking Water (FPTCDW) to develop national health-based drinking water guidelines, the Canadian drinking water quality guidelines (including the Guideline technical documents, collectively the Guidelines) (FPTCDW 2006). In collaboration with the FPTCDW, the Canadian Council of Ministers of the Environment (CCME) has developed a pan-Canadian water quality monitoring framework and policy (CCME 1999, 2008, 2011; CCME Water Quality Task Group, Sub-Group on Water Quality Monitoring 2006) and has promoted the use of the Guidelines to protect community water supplies. The Guidelines and other pan-Canadian drinking water-related policies are voluntary and non-enforceable and each province (and territory) determines whether, and to what extent, to adopt the Guidelines. Since provinces are not obligated to adopt the Guidelines, drinking-water quality management is not standardized across the country. Canada remains among the few member countries of the Organization for Economic Cooperation and Development (OECD) that do not have enforceable national drinking water quality standards (Bakker and Cook 2011).

Using document analysis, specific components of the regulatory frameworks for testing microbial contamination of drinking water were compared in Canada’s three most populous provinces – British Columbia, Ontario and Quebec. Despite the importance of monitoring at all stages of water flow from its source to a consumer’s tap, current regulations are almost entirely concerned with drinking water consumed “at tap”; therefore, this review also has this focus. The review found significant differences in key features of provincial regulatory frameworks for microbial testing in drinking water quality management. These key differences will impact the integration of novel testing methodologies in Canada.

Methods

The legislation and regulations pertaining to microbial water quality testing in British Columbia, Ontario and Quebec were reviewed. These three provinces were selected because they are the most populous provinces, the largest geographically, and each has large urban areas as well as remote rural communities. Additionally, the review included policy documents (of relevant ministries) and grey literature relating to microbial water quality management in Canada. Comparison was made of four key features of provincial regulatory frameworks specific to microbial testing in drinking water quality management. The focus was on current provincial legislation (including regulations and policies) related to drinking water quality, and pan-Canadian policy documents made over the last decade related to drinking water quality.

Results

Table 1 presents the key legislation and regulations pertaining to microbial drinking water quality management in each of the three provinces, and indicates the ministries responsible. This study focuses only on drinking water quality legislation and regulations; a broad survey of all environmental water quality legislation and regulations is therefore beyond the scope of this study.

Table 1. Provincial legislation, regulations, and ministries pertaining to microbial water quality management.

	British Columbia	Ontario	Quebec
Water quality standards	No total coliforms	No total coliforms	No total coliforms
	No Escherichia coli or fecal coliforms	No E. coli	No E. coli or fecal coliforms
	Note that in practice most labs test for E. coli and not fecal coliforms “the reference to fecal coliforms in section 9 of the regulation and in Schedule A may be redundant” (BCOO 2008).	(heterotrophic plate counts)	Starting 8 March 2013, test results for E. coli and fecal coliform will be required
Testing frequency and regularity	Population-based; no regularity prescribed; sample type not specified	Population-based; regularity prescribed; sample type specified	Population-based; regularity prescribed; sample type specified
Types of water samples	Raw, treated, distribution	Raw, treated, distribution	Raw, treated, distribution
Small systems specifications	All systems with two or more connections are subject to regulation, but systems serving 500 or fewer individuals may be exempt from certain regulatory requirements	Separate regulation for small systems that may reduces the testing frequency and sample types required	Currently all systems serving 20 or fewer individuals, or one or more enterprises (or both of those) are exempt

Four key features of microbial water quality testing were selected for comparison – water quality standards, types of drinking water samples tested, frequency and regularity of testing, and system size specifications. Table 2 presents these results and the following discussion details the findings. As noted, this review of the regulatory frameworks focuses only on municipal water supplies (at tap); it does not include individual water supplier permits which may demand water suppliers satisfy requirements in addition to the regulations, nor does it include watershed (at-source) microbial monitoring.

Legislated drinking water quality standards

In all of the provinces studied, microbiological contaminant testing of water is restricted to bacterial indicators: Escherichia coli, fecal coliforms and total coliforms. Treatment requirements are largely driven by the need to reduce the presence of other microorganisms (i.e. parasites and viruses). In some instances, utilities voluntarily test for additional microorganisms such as Giardia, Cryptosporidium, and algae. Utilities may be required, in their operating permit, to test for additional microbial parameters that are not required in provincial legislation. In Quebec, systems supplied in whole, or in part, from at-risk surface water or groundwater must ensure the mandatory elimination of at least 99.99% of Giardia, Cryptosporidium and viruses; however, direct pathogen testing is not required (Regulations to the Environmental Quality Act, Reg. c. Q-2, r. 40, art. 5).

Table 2. Comparison of four components of provincial microbial water quality testing regulations.

	British Columbia	Ontario	Quebec
Legislation	Drinking Water Protection Act, 2001	Safe Drinking Water Act, 2002	Environment Quality Act
Regulations	BC Reg. 200/2003	O. Reg. 170/03; O. Reg 169/03; O. Reg. 318/08; 319/08	c. Q-2, r. 40
Ministry responsible	Ministry of Health	Ministry of Environment (municipalities); Ministry of Health and Long-Term Care (small water systems)	Ministry of Sustainable Development, Environment and Parks

Legislation in all three provinces prescribes microbiological water quality standards for total coliforms and E. coli: there should be no detectable presence of either in a 100-mL water sample; this is an element of the Guidelines that the three provinces are each in accordance with. In Ontario, under Regulation 170 of the Clean Water Act, at least 25% of distribution samples and weekly treated samples must also test for heterotrophic plate count (HPC), a bacterial indicator test (used largely for water treatment process control) that is not required by legislation in either British Columbia or Quebec (see O. Reg. 170/03, Schedules 10 and 11). In the future, Quebec may test for more microbial contaminants as provincial reports on water quality testing have recommended that regulations require testing for a wider range of pathogens (Québec, Ministère du Développement durable, de l’Environnement et des Parcs 2006; Québec, Direction des politiques de l’eau 2012b).

Types of water samples tested

Water samples are typically collected from three source types: raw, treated and distribution system water (Table 3). Ontario legislation and regulations define the sample types. The Ontario definitions are not directly applicable in other provinces, but they are used here to describe the different natures of the sample types. A distribution sample is

a water sample that is taken, in the drinking water system’s distribution system or in plumbing that is connected to the drinking water system, from a point significantly beyond the point at which drinking water enters the distribution system or plumbing. [O. Reg. 170/03, s.1(1)]

Table 3. Frequency and types of water sampled by province.

Province	Population	Sampling Frequency	Type of Water Sample	Regularity of Sampling
BC	< 5000	4 samples per month	Not defined	Not defined
	5000–90,000	1 sample per 1000 people
	> 90,000	90 samples
		plus 1 sample per 10,000 people in excess of 90,000
ON	< 100,000	8 “distribution” samples per month	Distribution, treated and raw	At least one sample per week (distribution must be treated more frequently than raw samples)
		plus 1 sample per 1000 people served (per month)
	> 100,000	100 ‘distribution’ samples per month		[at least] three samples per week
		plus 1 sample per 10,000 people served in excess of 100,000		(distribution must be treated more frequently than raw samples)
QC	< 21 individuals / enterprises	Drinking water regulations do not apply
	21–1000	Min. 2 samples per month	Distribution and raw	If # samples < 4 then samples should be collected at 7-day intervals (at least)
	1001–100,000	1 sample per 1000 people		Evenly distributed throughout sampling period if possible equal number of sample per week. 1 sample a week for raw water samples over 120 days.
	> 100,001	100 samples per month plus 1 additional sample per 10,000 people in excess of 100,000

A raw water sample is taken from “outside a drinking water system that is a source of water for the system,” before the water has been treated [Safe Drinking Water Act 2002, s.2(1)]. A treated sample is not defined in the legislation but in practice; treated samples are collected at the end of the treatment process, before the water enters the distribution system.

British Columbia legislation requires water suppliers to monitor all three types of water samples at the frequency established by the regulations and by the supplier’s operating permit. A water supplier must “monitor its drinking water source [raw water], the water in its system [distribution water] and the water it provides [treated water]” [Drinking Water Protection Act 2001, s.11(1)(a)]. However, the regulation in British Columbia does not detail the frequency, regularity or testing of different types of samples. Rather, the testing requirements for raw and distribution water are detailed in water supplier operating permits [Drinking Water Protection Act 2001, s. 8(3)]. Since operating permits were not available for review, the researchers could not determine further details of sample testing practices in British Columbia. However, in general practice, most British Columbia water suppliers only regularly collect and test samples of treated water.

For each type of system and sample, Ontario has particular requirements for the number of samples, the regularity, and the type of microbial tests performed (see O. Reg. 170/03). Distribution system samples are tested most frequently, followed by raw water samples and then treated water samples, as will be discussed further below in the section on testing frequencies. Only large municipal residential systems in Ontario are required to ensure that treated water samples are collected and tested for microbiological contamination.

As in British Columbia and Ontario, Quebec regulations require the sampling of all three types of water. Distribution samples are to be taken at the tap, where the water is put at the disposal of users, of which at least 50% of samples must be taken at the outermost limits of the distribution system (c. Q-2, r. 40, art. 11 and 12). In Quebec, raw water is defined as “water that has not yet been treated to become potable”; it is sampled and tested prior to treatment (Québec, Direction des politiques de l’eau 2012a). Raw water samples will be tested according to the category of the source. There are three categories: systems supplied by surface water or groundwater that is at risk for microbial alteration by surface water (c. Q-2, r. 40, art. 5); systems supplied by non-disinfected groundwater considered to be vulnerable (c. Q-2, r. 40, art. 13), and systems where the previous microbial test results failed to comply with bacterial parameters (c. Q-2, r. 40, art. 39). Treated water samples must be collected at the outlet of the facility or reservoir before the water enters the distribution system. As of March 2013, Quebec requires that systems serving more than 20,000 persons install software capable of providing a continual calculation of virus and protozoan (Giardia and Cryptosporidium) removal efficiencies during treatment disinfection (c. Q-2, r. 40, art. 22). Microbiological samples are not collected to calculate these removal efficiencies; rather, they are based on models and non-microbiological parameters measured through continuous monitoring.

Frequency and regularity of testing for microbial markers

In all of the provinces studied, frequency of testing is based on the size of the population served by the water system (Table 3). In British Columbia, testing frequencies have been adopted, without alteration, from those recommended in the Guidelines. The testing frequency schedule is as follows: water supply systems serving populations of up to 5000 are sampled four times per month; systems serving populations of 5000 to 90,000 must conduct one sample per month for every 1000 members of the population; and systems serving populations of more than 90,000 must take 90 samples, plus one sample for every 10,000 persons in excess of 90,000, per month (BC Reg. 200/2003, Schedule B). While British Columbia adopts the sampling frequencies recommended in the Guidelines, it has not incorporated the suggestion that samples be taken at regular intervals throughout the month into its regulations. Thus, water systems in British Columbia may collect samples of treated water at irregular intervals (i.e. sampling four times on a single day may comply with the testing frequency schedule).

Ontario Regulation 170 of the Clean Water Act prescribes different sampling frequencies depending on the type of water sample. A large municipal residential system is defined as a municipal drinking-water system that serves a major residential development and serves more than 100 private residences; it must test distribution, treated and raw samples. Large residential systems must test treated and raw water samples weekly for microbiological contamination. For all systems, distribution samples must be tested more frequently than raw water samples; only large residential systems must test treated water samples (see O. Reg. 170/03, Schedules 10, 11 and 12). Systems serving populations of 100,000 people or fewer must take eight distribution samples per month, plus one additional sample for every 1000 people served, per month. Systems serving populations greater than 100,000 must take 100 distribution samples per month plus one additional sample for every 10,000 people served every month. Ontario prescribes regularity of sampling collection, requiring at least one sample per week for systems serving fewer than 100,000 people; systems serving populations greater than 100,000 must collect three samples weekly (O. Reg. 170/03, Schedule 10).

Quebec also has a frequency schedule based on the size of the population being served, as well as the type of water, but this approach differs from that of the Guidelines. Currently, drinking water quality regulation in Quebec does not apply to systems exclusively serving fewer than 21 persons, one or more enterprises (e.g. a commercial or agricultural establishment, excluding hospitals, schools and jails), or both (c. Q-2, r. 40, art. 4, 10, 10.1). As of March 2012, all enterprises installing new drinking water treatment equipment or modifying (maintenance or replacements with same equipment are not included) such equipment will be subject to the requirements for microbial monitoring of filtered or disinfected water (c. Q-2, r. 40, art. 4; Québec, Direction des politiques de l’eau 2012a, 23). All other systems must test distribution and raw water according to the appropriate schedules. Distribution water samples are tested at least twice per month for water systems serving from 21 to 1000 persons (c. Q-2, r. 40, art. 11). Systems serving between 1001 and 8000 persons are required to take at least eight samples per month, while systems serving between 8001 and 100,000 persons must take one sample per 1000 persons. Systems serving populations in excess of 100,001 persons must take 100 samples per month and one additional sample for every 10,000 persons (over the first 100,000) served monthly (c. Q-2, r. 40, art. 11). Quebec specifies that collection of samples should be evenly distributed through the month. In Quebec, raw water testing depends on the source of the water. In systems supplied in whole or in part by surface water or groundwater at risk for microbial alteration by surface water, sampling must be done weekly for 120 days (c. Q-2, r. 40, s. 5; Québec, Direction des politiques de l’eau 2012a, 23–26). For systems supplied in whole or in part by non-disinfected groundwater that is considered to be vulnerable, samples must be collected once a month from captured or stored raw water supplies to monitor E. coli and Enterococci (if vulnerability is bacterial) or F-specific coliphage viruses (if vulnerability is virological) (c. Q-2, r. 40, art. 13). In systems for which previous microbial test results failed to comply with one of the bacterial parameters, samples must be taken at the tap based on population size as prescribed in the legislation (c. Q-2, r. 40, art. 39). Quebec has been steadily increasing its sampling and frequency requirements for microbial monitoring, which currently amount to about 80% of water quality assessment data with a compliance rate with microbial standards of 99.7% during the years 2005 to 2009 (Québec, Direction des politiques de l’eau 2012b). The recent amendments to the Regulation respecting the quality of drinking water and changes implemented in March 2013 are meant to be more inclusive of technical improvements (e.g. new filtration and disinfection requirements for new treatment equipments) by integrating new developments from the lab (e.g. monitoring of E. coli can no longer be substituted for fecal coliforms and elimination rates are now required for two protozoa); scientific data regarding other parameters are being assessed for possible future inclusion (c. Q-2, r. 40; Québec, Direction des politiques de l’eau 2012a).

Small system specifications

In British Columbia, only small systems serving up to 500 individuals during any 24-hour period may be exempted from the regulations. An exemption may be granted if the system does not provide water for human consumption or food preparation purposes, and is not connected to a water supply system that provides water for human consumption and food preparation purposes (i.e. hospitals and schools); or if each recipient of water from the system has a point of entry or point of use treatment then the supplying system will not be subject to the regulation (BC Reg. 200/2003, s. 3.1).” Even if small systems are not exempted from the regulation, they are granted relief from some regulatory requirements. For example, operators of small systems will not need to meet the same certification standards as other water supply systems [BC Reg. 200/2003, s. 4(2)].

In the Ontario regulatory framework important features of water systems include not just the size of the population the system serves, but also whether the system is municipal or not, residential or not, and year-round or seasonal. Small and large municipal residential drinking water systems are subject to the same regulations managed by the Ministry of Environment, but to different microbial contaminant testing protocols (see O. Reg. 170/03, Schedules 10 and 11). Large municipal residential systems must test raw, treated and distribution samples; small municipal residential systems are not required to test treated water samples, and requirements for other types of samples are less frequent and depend on the treatment protocol in place, which is usually linked to the type of drinking water source.

In 2008, Ontario created a separate regulatory framework for small drinking water systems, of which there are five categories:

1. Large municipal, non-residential drinking water systems that serve such facilities as municipally-owned airports and industrial parks, and large sports and recreation facilities.

2. Small municipal, non-residential drinking water systems that serve such facilities as small community centres, libraries, and sports and recreation facilities.

3. Non-municipal, seasonal residential drinking water systems that serve such facilities as private cottages on communal drinking water systems.

4. Large non-municipal, non-residential drinking water systems that serve such facilities as large motels and resorts.

5. Small non-municipal, non-residential drinking water systems that serve such facilities as motels, restaurants, gas stations, churches, and bed and breakfasts [O. Reg. 318/08, s. 1(1)].

The Ontario Ministry of Health and Long-Term Care (OMHLTC) manages the separate regulatory framework for small systems pursuant to the Health Protection and Promotion Act (OMHLTC 2010; Health Protection and Promotion Act 1990). More particularly, small systems are “administered by local boards of health. Public health inspectors will conduct a site-specific risk assessment for every small drinking water system in the province” (OMHLTC 2010). These risk assessments will be used to prepare a “directive” for the system that will specify “requirements for water testing, treatment options or training” that when implemented by owners and operators will “keep their drinking water safe” (OMHLTC 2010). Until the risk assessment is completed for a system, the system is subject to the requirements of the transition regulation which includes two different water sample microbial contaminant testing protocols for the five types of small water systems (O. Reg. 318/08).

Currently, Quebec regulations do not distinguish systems only on their size; systems serving exclusively enterprises (and those serving a combination of one or more enterprises and less than 21 persons) are exempt from requirements set under the Regulation Respecting the Quality of Drinking Water. Nonetheless, besides this exemption, the regulatory framework for microbial monitoring is applicable to all systems, with the exception that systems serving fewer than 21 persons are exempt. Additionally, while not exclusive to microbial contaminant testing, other requirements under the Quebec regulation differ based on system size; for example, systems serving less than 501 persons are not required to do a continuous calculation of removal efficiencies for microbial parameters in treated water, and a similar exemption is also provided to non-municipal systems supplied by a municipality that is subject to the regulations (c. Q-2, r. 40, art. 22 and 21.1). As of March 2013, Quebec regulations have changed E. coli testing frequency requirements: systems supplied in whole or in part with raw water from surface water or groundwater susceptible to microbiological alteration by surface water and that serve more than 1000 persons will be required to monitor E. coli concentrations of their raw water. Systems that rely on the same source (or type of source) but serve fewer than 1001 persons are not required to monitor E. coli in raw water (but remain required to achieve prescribed elimination efficiencies for viruses, Giardia, and Crytosporidium) (c. Q-2, r. 40, art. 5.1, Québec, Direction des politiques de l’eau 2012a, 27). Systems serving between 1001 and 5000 persons require testing at least once a month, and systems serving more than 5000 persons require testing at least once a week.

Discussion

This study illustrates some key differences among three selected Canadian provinces in microbial contaminant testing of drinking water. All three provinces have the same bacterial water quality standards, tolerating no detectable presence of total coliforms, fecal coliforms or E. coli in a treated water sample, consistent with international standards and guidelines (Council of the European Union 1998; WHO 2003; USEPA 2009; ISO 2012). Additional indicators (enterococci and coliphage) are mentioned in Quebec legislation, but not in Ontario or British Columbia legislation. Pathogen-specific testing (i.e. Cryptosporidium or adenovirus) is not described in the legislation, nor are alternative indicators. In addition to E. coli standards, the US also has legally enforceable standards for Cryptosporidium, Giardia and enteric viruses (zero milligrams per litre) that apply to public water systems; however, tests for analytes other than E. coli are required only if water systems are determined to be at risk after an initial assessment phase (USEPA 2009). In the European Union member states also have a legally enforceable standard for Enterococci (zero in 100 mL at the consumer’s tap) (Council of the European Union 1998). In addition to the EU Directive requirement, England and Wales also has legally enforceable standards for E. coli, Enterococci, coliforms (zero in 100 mL both at the consumer’s tap and in the service reservoirs and water treatment plants), as well as Clostridium perfringens (zero in 100 mL at the supply point) (Water, England and Wales 2000).

Ontario prescribes in detail the types of samples that must be tested for microbial indicators as well as the frequency and regularity with which samples need to be collected. Quebec generally requires treated, distribution and raw water samples to be tested at particular frequencies on a regular basis. British Columbia requires that water suppliers monitor the three types of water samples, but the regulations do not prescribe frequencies and regularity of sampling. Individual water supplier operating permits may detail individual sampling procedures. In practice, treated water samples are the most frequently tested. It is unclear why British Columbia has been less prescriptive in sampling requirements than either Quebec or Ontario, but this may be attributed to the oversight role the legislative framework grants to the regional drinking water officer in the Health Authority; this structure is likely intended to address the considerable geographic variance in provision of potable drinking water in British Columbia.

Regulation of water systems based on size is the most striking difference in these three provincial approaches to microbial water quality testing. British Columbia provides some exemptions from the regulations for what it classifies as small systems – those serving fewer than 500 individuals. Quebec regulations have increased the frequency of testing requirements for raw water systems supplied in whole or in part by surface or groundwater considered vulnerable to microbial alteration from surface water and serving more than 1000 persons (c. Q-2, r. 40, art. 22.01). Ontario’s new small-systems approach also permits small systems to test for microbial water quality less frequently. On the assumption that approximately five individuals live in each private residence, British Columbia’s threshold of 500 individuals is approximately equal to Ontario’s threshold of fewer than 101 private residences. Quebec’s testing frequency schedule for smaller systems is not directly commensurate to British Columbia or Ontario. In Quebec’s current regulatory framework, systems serving fewer than 21 persons, or pertaining only to one or more enterprises, are exempt from the requirements of the regulations. In 2009, Quebec recognized system size as an impediment to compliance with the drinking water regulation. In a report on the implementation of drinking water regulations, the Ministry of Sustainable Development, Environment, and Parks found that size of the distribution facility was a key issue in the implementation of bacteriological monitoring parameters (Québec, Direction des politiques de l’eau 2012b). The report found that seasonal tourist facilities, as well as non-municipal residential distribution facilities, have faced more issues in complying with the prescribed regulations than have municipal residential distribution facilities. Specifically, the compliance issues were connected to the limited resources smaller systems had for testing and result reporting (Québec, Direction des politiques de l’eau 2012b). Indeed, in Quebec there are no regulations specific to systems serving less than 21 persons, while all distributors (serving 21 or more persons) must comply with general sampling requirements.

British Columbia also does not have a regulation specific to non-residential and/or seasonal water systems, but testing of these facility types is enforced at the health authority level. In British Columbia’s regulatory framework, all water supply systems are subject to the regulations if they have two or more connections (Drinking Water Protection Act 2001, s. 1 and 6). Small systems, serving fewer than 500 individuals, are exempted from certain regulatory requirements, but there is not a unique frequency schedule for testing for small systems [BC Reg. 200/2003, s.4(2)]. Frequency of sampling in British Columbia for systems serving a population of fewer than 5000 individuals is four tests per month of treated water. In Quebec, since 2012, microbial water testing of raw water from treatment systems supplied by surface water or groundwater vulnerable to surface water microbial alteration serving fewer than 1000 persons only needs to be tested periodically (which is not defined in the Quebec regulations), systems serving between 1001 and 5000 persons need to test monthly, and systems serving more than 5000 persons need to test weekly (c. Q-2, r. 40 s. 22.01). Quebec’s regulations prescribe different testing requirements depending on the raw water source; British Columbia regulations do not.

However, British Columbia provides other exemptions for small systems, including allowing point of service or point of entry treatment to negate the need for a water system to provide potable water. Also, in British Columbia, operators of small systems might not need to meet the qualifications standards of the regulations [Drinking Water Protection Act 2001, s. 9; BC Reg. 200/2003, s. 4; British Columbia Office of the Ombudsman (BCOO) 2008]. Thus far, these provisions have had limited success in addressing the challenges of microbial drinking water quality management in small systems in British Columbia: in 2008, the British Columbia Ombudsman reported that about 10% of British Columbians are served by small water systems, many of which are under boil water or water quality advisories (BCOO 2008). Part of the challenge in British Columbia is identifying small systems that might have been created before the Drinking Water Protection Act came into force, or are simply unknown to a drinking water officer (BCOO 2008). This latter point hints at the geography of British Columbia in which small systems may be isolated from major centres, possibly facilitating non-compliance, significantly reducing capacity and definitely challenging enforcement (BCOO 2008). British Columbia is currently developing a Small Water Systems Strategy, but health authorities report that they tend to focus their limited resources on water systems that serve larger populations (BCCO 2008).

Further, the particular geography of a community may contribute to variability in microbial water quality testing practices (Grover 2011; Grover et al. 2012). Isolated rural communities may be more challenging to access and therefore less likely to be inspected and to have samples collected for testing. Indeed, it is argued that “people living in rural settings with small supplies are far more likely to experience an outbreak of infectious disease than people receiving their water from a large public utility” (Schuster et al. 2005; Hunter et al. 2009, 3681; Wilson et al. 2009; Moffatt and Struck 2011). Certainly, in British Columbia, this may result in a permanent boil water advisory for a system, or a system that is excessively chlorinated when collecting samples is a challenge (Grover 2011; Grover et al. 2012).

Currently, each province approaches microbial water quality management in small systems differently. Ontario’s comprehensive scheme should attract the interest of other provinces. Despite the large expense in creating a separate regulatory scheme for specific types of small systems and individually assessing risk for those systems, such a customized approach is more likely to achieve the desired result of safer drinking water for the entire population. In drinking water management, the importance of compliance and enforcement cannot be underestimated. If requirements are judged by an owner or operator to be unreasonable against the risk profile of the system, then operators may choose not to comply, potentially threatening drinking water security for all users. Using enforcement to require unnecessary testing (and/or treatment) in water systems is an inefficient and expensive approach to a critical public health management issue. It has been well demonstrated that the more preventative steps that can be taken in microbial drinking water quality management, the better (Hrudey 2011).

This study demonstrates the disparity in microbial water testing regulations across the provinces, as well as the limits of existing legislation, including the limited uptake of the widely accepted framework for risk aversion approaches, the source-to-tap model. Looking forward, the results of this study suggest a few considerations regarding the introduction of alternative microbial testing approaches into existing legal frameworks. Novel tests would need to be introduced into each provincial framework for drinking water regulation. Integration of new tests would need to address current legislative features, including water type, sampling frequency and population size and type. Perhaps more importantly, any new test would need to provide equal, if not better, protection of public health through ongoing water quality monitoring.

Conclusions

Review and comparison of regulatory frameworks in three Canadian provinces for four features of microbial water quality testing – water quality standards, types of water samples tested, frequency and regularity of testing, and system size specifications – found that water quality standards in the three provinces are similar, but the other features are more variable. This study was designed to analyze the different approaches to microbial testing of drinking water across three provinces. The comparative effectiveness or impact of these different provincial approaches would be difficult to assess and has not been evaluated. However, the found differences are relevant when considering how new knowledge can be integrated into existing water management approaches. This includes considerations of how and when new methods for assessing microbial water quality testing might be integrated into existing regulatory frameworks, such as qPCR to replace culture-based detection for E. coli, or use of PCR for detection of pathogens of public health importance such as Cryptosporidium or adenoviruses that are rarely (or never) monitored under current practices.

Each province has responded differently to the challenge of managing microbial water quality in smaller systems: the creation of a separate regulatory framework for small drinking water systems (Ontario), exemptions for smaller systems (British Columbia), and systems serving less than 21 persons (Quebec). There is clearly some consensus among the provinces that unique regulatory approaches for systems of different sizes may be necessary to resolve the challenges small water systems have in meeting water quality requirements designed for larger systems. Variations in microbial water quality management within provinces may be as significant as variations among provinces in ensuring public health. That is, differences among provincial regulatory frameworks tell only part of the story when it comes to managing public health with respect to drinking water quality in Canada. Regulating drinking water systems based on size (and other factors in Ontario, including recreational use or year-round use and recommendations from the Walkerton Report (O’Connor 2002a, 2002b). is important for public health because small systems simply may not have the capacity to meet extant regulatory requirements (Kot et al. 2011). Adopting an approach that recognizes these capacity limits and manages the particular risks of small systems could serve to improve public health related to drinking water in small systems.

Two other critical issues of within province variation are how source water quality is managed in relation to drinking water quality and uneven compliance, and enforcement of regulatory frameworks across a province. These two issues, which are beyond the scope of this review, will contribute additional complexity to microbial water quality management among and within provinces.

That the provinces have different approaches to water quality assessment in the provision of safe drinking water, depending on the size of the system, shows that each province is approaching similar management challenges in different ways. Certainly, Ontario’s unique approach emerged from a unique context – shaped in part by lessons from the Walkerton tragedy and its associated political will for more efficiency, which ultimately made the province a leader in innovative drinking water management. Of course, the drinking water management context in each province is unique and, as such, each provincial regulatory environment would need to be tailored to the particular priorities that novel testing approaches can efficiently meet.

Acknowledgements

The authors would like to acknowledge the financial contribution of the Canadian Water Network, Genome Canada and Genome British Columbia. They would also like to thank Leila Harris (University of British Columbia) for her thoughtful review of an earlier version of the manuscript. Funding from the Alberman Fellowship at the Hebrew University of Jerusalem, Department of Geography is gratefully acknowledged.