Drinking water quality indices: a systematic review

Programa de Pós-graduação em Tecnologia Ambiental. Universidade de Santa Cruz do Sul (UNISC), Avenida Independência, n° 2293, CEP: 96816-501, Santa Cruz do Sul, RS, Brazil. E-mail: rodrigoklamt@hotmail.com Departamento de Ciências da Vida. Universidade de Santa Cruz do Sul (UNISC), Avenida Independência, n° 2293, CEP: 96816-501, Santa Cruz do Sul, RS, Brazil. E-mail: adilson@unisc.br Departamento de Ciências da Saúde. Universidade de Santa Cruz do Sul (UNISC), Avenida Independência, n° 2293, CEP: 96816-501, Santa Cruz do Sul, RS, Brazil. E-mail: marig@unisc.br Corresponding author. E-mail: lobo@unisc.br

(FUNASA), water for human consumption must not contain pathogenic microorganisms and be free of bacteria from coliform group, which are indicators of fecal contamination, being represented mainly by Escherichia coli (FUNASA, 2009).
Currently, there is a lack of WQI's for drinking purposes, especially those focused on simpler parameters, at low cost, generating qualitative information accessible to the population. In this context, this study uses the systematic review methodology developed by Petersen et al. (2008), aimed at identifying, analyzing and interpreting data on the use of WQI's for human consumption, trying to find the most relevant publications, the possible research gaps, and their challenges, but mainly to identify the main parameters used and the composition of WQI's.

METHODOLOGY
This study is characterized as a systematic review, defined by the use of data from the literature, seeking to integrate information from a set of studies carried out separately on a given topic, which may present conflicting or coincident results (Sampaio, 2007). A scientific article that applies the systematic review as a methodology aims to identify, analyze and interpret all available evidence related to a specific research question (Kitchenham and Charters, 2007), in this case addressed to the use of WQI's for human consumption. This type of methodology discusses not only the conclusion, but also all activities related to the main theme. Thus, a systematic study collects data on the places where the activity occurs and the media in which it was published. The methodology consists of elaborating the research questions, designing the research process and defining the criteria for filtering the results (Cooper, 2016).

Research questions
The review covers articles published from January 2000 until March 2020, over a period of 20 years. The first general question (GQ1) is to identify which WQI's were applied to classify water quality (Table 1). From this, a specific question (FQ1) and a statistical question (SQ1) were defined.

Reference Questions
General question (GQ1) Which indices were applied to classify the quality of drinking water?
Specific question (FQ1) Which variables were found in the research?
Statistic question (SQ1) Where were the surveys published?

Research process
According to Petersen et al. (2008), three stages are defined during the research process: (1) Specify the research keywords, (2) Choose the databases to apply the research keywords, and (3) Get the results. The first step begins by identifying the main terms and their most relevant synonyms. Thus, scientific articles were searched using the keywords "drinking water" and "water quality index", in the PubMed, Scielo, ScienceDirect and Web of Science databases, with a custom interval between 2000 and 2020.

Applying filters
In order to filter the most relevant scientific information, the following criteria (C) were included: C1. The study must be published in a scientific magazine or newspaper.
C2. The study must be directly related to WQI's for consumption.
C3. The study must be a complete work.
C4. Keywords must be included in the title.
The selected articles were stored in EndNote X7 © software (Thomson Reuters), where the texts were organized in specific folders for each research database. The next filter was based on the three-step approach introduced by Srinivasan Keshav (2007). The first step was a quick scan and consists of reading the title, summary and introduction. In the second step, only the section and subsection headings are read. Finally, in the third step, the mathematical content (if any) is examined to determine the underlying theoretical foundations. Table 2 shows the filtering process, with criteria C1 to C4 applied in each stage. Based on the results obtained, the articles were organized by title, year of publication, authors, databases, technologies and work objectives.

GQ1 -Which indices were applied to classify water quality?
A WQI is a tool to present a numerical expression derived cumulatively to define a certain level of water quality. In other words, a WQI summarizes large amounts of water quality data in simple terms (for example, excellent, good and bad) to report to management authorities and non-technical people consistently. The WQI concept is based on the comparison of water quality parameters with their respective regulatory standards, and gives a unique value to water quality, which translates the list of constituents and their concentrations present in a sample. The formulation and use of indices have been strongly advocated by agencies responsible for water supply and water pollution control. Table 3 shows the WQIs for consumption that have been used.
In some studies, more than one WQI was used, being modified for consumption purposes, and several citations followed the standards of the World Health Organization, which deals with drinking water guidelines. The Indian Standard Bureau (ISB) is an institution that regulates and provides details on the permitted and desirable standard specifications and limits of various parameters in drinking water. The CCME Water Quality Index, based on a formula developed by the British Columbia Ministry of Environment, is a quality index that summarizes complex data on water quality to facilitate its communication to the public. The NSF index was developed to provide a standardized method for comparing the water quality of various bodies.  Table 3. Water quality indices used and respective references.

FQ1 -Which variables were found in the research?
The researchers explored each index as a performance measure that aggregates information, reflecting the influence of physical, chemical and biological variables of water quality conditions. These are based on quality parameters with their respective regulatory standards, presenting a single value for water quality. Table 4 shows the parameters used and their respective references. Table 4. Parameters used to assess the quality of drinking water and respective references. Total hardness (9) Batabyal and Chakraborty (2015) Iron (9) Batabyal and Chakraborty (2015) Calcium (9) Batabyal and Chakraborty (2015) Magnesium (8) Batabyal and Chakraborty (2015) Total coliforms (7) Avvannavar and Shrihari (2008) Barium (1) Gharibi et al. (2012).
The parameters most used to assess water quality were pH, nitrate, turbidity, chloride and sulfate, cited in 10 of the 16 articles selected in this study (62,5% of the total). The inorganic contaminant of greatest concern in groundwater is the nitrate ion, NO3 − , which normally occurs in aquifers in rural and suburban areas. Nitrate in groundwater comes mainly from four sources: application of nitrogen fertilizers, animal manure in plantations; soil cultivation; human sewage deposited in septic systems; and atmospheric deposition (Baird and Cann, 2011).
A recent concern deals with the increase in nitrate ion levels in drinking water, particularly in well water in rural localities, the main source of this nitrate being leaching from cultivated land into rivers and water flows. Excess nitrate ion in drinking water is worrying because it causes blue baby syndrome in newborns; in adults, it may be responsible for causing stomach cancer and increase the likelihood of breast cancer in women (Baird and Cann, 2011).
Chloride in drinking water comes from natural sources, sewers, industries, effluents, urban runoff containing defrost salt, and saline intrusion. High concentrations of chloride increase the metal corrosion rates in the distribution system, depending on the alkalinity of the water, and give a salty taste to water. The taste limits for the chloride anion depend on the associated cation and are in the range of 200-300 mg L -1 for sodium, potassium and calcium chloride. Excess concentrations of 250 mg L -1 are more likely to be detected by taste, but some consumers may get used to low levels of chloride-induced flavor. No health-based guideline proposes a concentration value for chloride in drinking water (WHO, 2017).
The presence of sulfate in drinking water can cause a noticeable taste, and very high levels can cause a laxative effect in unaccustomed consumers. Sulfates occur naturally in various minerals and are used commercially, mainly in the chemical industry. They are discharged into water in industrial waste and by atmospheric deposition; however, the highest levels generally occur in groundwater and are from natural sources. In general, the average daily intake of drinking water, air and food sulfate is approximately 500 mg, with food being the main source. However, in areas with drinking water supplies containing high levels of sulfate, drinking water can be the main intake source. No health-based guidelines are proposed for sulfate. However, due to gastrointestinal effects resulting from drinking water containing a high sulfate content, it is recommended that health authorities are notified of drinking water sources containing sulfate concentrations above 500 mg L -1 (WHO, 2017).
pH is a measure of the concentration of H + ions in water. The balance of hydrogen and hydroxide ions (OH − ) determines how acidic or basic water is. In chemically pure water, the H + ions are in equilibrium with the OH − ions and their pH is neutral, that is, equal to 7. The main factors that determine the pH of water are dissolved carbon dioxide (CO2) and alkalinity. pH has a direct relationship with the quality of water for human consumption. Water with a high carbon dioxide content, reduced total alkalinity and low pH is considered aggressive. Aggressive waters are those that tend to dissolve calcium carbonate. Water with a low pH promotes corrosion of metal pipes and fittings, namely copper, lead and zinc, which can cause problems such as a metallic or sour taste (ERSAR, 2008;WHO, 2011).
A pH greater than 8.5 may indicate that the water is hard. It does not pose a health risk, but it can cause esthetic problems, such as incrustations in pipes and equipment, causing a decrease in the inside diameter; alkaline flavor; difficulty in obtaining foam; and formation of insoluble precipitates on clothes. Turbidity has a potential health effect: it is often associated with the presence of microorganisms originating in surface runoff from soil (WHO, 2017).
The presence of microbiological agents (Escherichia coli, coliforms), can generate acute and generalized effects on the health of consumers without changing the flavor, odor or color of water. The greatest public health risk of microbes in water is associated with drinking water that is contaminated with human and animal excrement, although other sources and routes of exposure may also be significant. Outbreaks transmitted by water have been associated with inadequate water treatment and unsatisfactory management of drinking water distribution.
There are few chemical substances in the water causing health problems, except for massive water contamination, such as in environmental accidents, and it is commonly impossible to consume the water due to its taste, odor or general appearance. Therefore, variations in the concentration of chemical elements in the water, even if for short periods they present values above those allowed does not necessarily mean that it is not fit for consumption. The proportion by which the permitted values can be exceeded and the period for which this situation can be prolonged without prejudice to health will depend on each substance or element considered.
The problems related to the elements or chemical substances present in water for human consumption are mainly due to the negative effects to health after prolonged periods of exposure to organic contaminants and toxic heavy metals. The radioactive aspects depend on naturally occurring radioactive isotopes that start from the uranium decay series, such as lead 210 mg L -1 and uranium 238 mg L -1 and 234 mg L -1 . The presence of radionuclides, even in normal circumstances, must be considered.
As for the acceptability aspects of drinking water, the population bases them on properties such as taste, color, odor, appearance, turbidity and other parameters perceptible to its taste. Esthetically unacceptable water will decrease consumer confidence, leading to complaints and, more importantly, to the use of water from less secure sources. Consumers have no means of judging the quality of their drinking water, but their attitude towards drinking water supplies and suppliers will be largely affected by the aspects of quality that they are able to perceive with their own senses. Consumers consider unacceptable water that is not colorless or which has an unpleasant taste or smell, even though these characteristics may not have direct health impacts.
It is important to consider whether existing water treatment and distribution practices can affect the acceptability of drinking water, as well as managing changes and operations to minimize the risk of acceptability problems. For example, chloramination that is not properly managed can lead to the formation of trichloramines, which can cause unacceptable taste and odor. Other problems can be indirect, such as internal deposits in pipes when the flow in distribution systems is disturbed or altered.
The water quality parameters used to assess the quality of drinking water were classified into microbiological, chemical, radioactive and acceptance aspects, as described in the Guidelines for Drinking-water Quality (WHO, 2017). The parameters not included in these categories were classified in "others" (Table 5).
In Brazil, according to the Consolidation Ordinance nº 5/2017 of the Ministry of Health (Brasil, 2017), some physical, chemical and bacteriological parameters must be followed to ensure their potability, considering their respective regulatory standards listed in Annex XX (Table 6), which deals with the control and surveillance of water quality for human consumption. All water intended for human consumption from an individual alternative water supply solution, regardless of how the population accesses it, is subject to water quality monitoring.

SQ1 -Where the surveys were published?
The articles were categorized in the four databases used in this systematic review. Table 7 shows the number of articles published by the database, highlighting ScienceDirect with 10 articles. No articles were published in the Scielo database. In a general way, little is found in the literature about comparative analysis related to the development, comparison and application of different WQI's; currently, there is great concern about controlling the quality of the water we consume. A series of indices are being developed and adapted for consumption.
The database that stood out most for the publication of studies on WQI within the selection criteria was ScienceDirect. It was observed that the impact of human activity was severe in most parameters. The maximum permitted values have exceeded the tolerable limits in some situations. Yet the main cause of deterioration in water quality is the lack of adequate sanitation, unprotected river areas and high anthropogenic activity. Water is essential to sustain life; there must be a satisfactory supply (adequate, safe and accessible) and it must be available to everyone. Improving access to contaminant-free drinking water can result in tangible health benefits. Every effort should be made to obtain drinking water that is as safe as possible.  Table 6. Water quality parameters in Brazil, following the Consolidation Ordinance nº 5/2017 of the Ministry of Health (Brasil, 2017).

Escherichia coli
Absence in 100 mL Heterotrophic Bacteria 500 UFC mL -1 Table 7. Classification of results.  Safe drinking water, as defined by authorities that determine its parameters, does not represent a significant health risk during the entire consumption life, including in those with sensitivities that can occur at different stages of life. Those most at risk for waterborne diseases are children and elderly people, especially when living in unsanitary conditions. Those who are generally at risk for waterborne illnesses may need to take additional measures to protect themselves from exposure to waterborne pathogens, such as boiling drinking water. Safe drinking water is required for all usual household purposes, including drinking, food preparation and personal hygiene. Therefore, the development of indices to assess the quality of drinking water is essential for each human being to have access to the quality of the water they are consuming.

CONCLUSION
This systematic review presented the search for WQI's for human consumption. In addition, it presented different models of variables associated with the quality of drinking water, in order to obtain a better result for water quality. However, none of the reviewed articles documented how a historical database could effectively improve this goal with the support of new technologies. The bibliographic references selected in this systematic review allow us to observe that, although the frequency of studies related to WQI's for drinking has increased in recent years, the number of studies is still insufficient to obtain effective scientific evidence. One of the gaps found during this research corresponds to evaluation of the parameters analyzed and their separation between microbiological, radioactive, chemical, acceptability and others, highlighting that the basic and essential requirements to guarantee the safety of drinking water are based on the health established by competent health system authorities and adequate surveillance systems.
We suggest that future research seek to propose regionalized WQI's for consumption, in order to consider local aspects in the evaluation process and to determine intervention priorities for health surveillance agencies. In addition, the development of computerized platforms for centralizing information on these WQI's, accessible to the public in real time, is essential for the effective implementation of programs for monitoring the quality of drinking water.