The health risk represents the probability of damage, illness or human death as a result of the environmental risk factor effect [1
]. Health risk assessment from a contaminated natural environment is therefore an important tool from the point of view of protecting the health of human populations, and in a broader context, also from the point of view of maintaining sustainable development. The health risk assessment methodology was formulated in the 1980s by the US Environmental Protection Agency (US EPA)—so called the method of Human Health Risk Assessment [1
]. Its main principles apply with some modifications to date [2
] and have become the basis for the legislative elaboration of health risk assessment procedures within the European Union [7
] and also in individual countries, e.g., in the Slovak and Czech Republic [9
The formulation of basic concepts and procedures for health risk assessment and their unification at the legislative level currently allows, in addition to a qualitative approach to the assessment of adverse effects of elements/substances on humans, the quantitative determination of risk levels in relation to these effects.
Quantitative determination of health risk is expressed by the hazard quotient (HQ) for non-carcinogenic risk (threshold effect) or as individual lifetime cancer risk (ILCR), and optionally, annual population cancer risk (APCR), expressed in terms of the probability, i.e., number of cases of cancer per capita per year (non-threshold effect).
The current methodological procedures for the health risk calculation only deal with increased contents of harmful substances/elements. They assess possible adverse effects of various substances/elements that occur at contents above the limit, or reference dose. Reference doses of different harmful elements are set for different environmental compartments, and thus, e.g., human health risk from increased As content in soils, air, drinking/groundwater or in various foods (e.g., vegetables) is estimated. However, these procedures do not assess the health risk due to the deficient content of various—especially biogenic—essential elements necessary for healthy human development. A classic example of such elements is deficient content of iodine, fluorine, iron and several trace elements (Se, Zn and Cu) or essential macro-elements e.g., Ca and Mg [11
]. Deficient elements are currently added to food (e.g., iodine in table salt, fluorine in toothpaste) or are available in the form of various nutritional supplements (Zn, Se, Fe or others). Important essential elements include Ca and Mg, whose low contents in drinking water are often associated with an increased incidence/mortality from several serious chronic diseases, especially cardiovascular diseases [15
]. There are far fewer studies on cancer, respiratory or gastrointestinal diseases and most of them are only ecological types of studies, so the strength of the evidence is still low. However, there is the growing evidence on the important role of sufficient intake of Ca and Mg in the immune system [18
], which may provide explanation on the missing pathological mechanism involved. These relationships are so far known exclusively as the threshold type; therefore, only the concept of non-carcinogenic risk and the calculation of HQ will be worked on.
The aim of the present article is to propose a new methodology for the health risk calculation from the deficit content of the essential elements and to test it on the example of the content of Ca and Mg in drinking water of the Slovak Republic. Previous works using artificial neural network calculations confirmed that Ca and Mg were the elements most affecting the health of the population in the Slovak Republic [22
]. Their effect is up to two orders of magnitude higher than the effect of classical contaminants in drinking water, such as potentially toxic elements or nitrates. Under their deficit contents, the relative mortality for the main causes of death in the Slovak Republic (i.e., cardiovascular and oncological diseases, diseases of the digestive and respiratory system) increases significantly by 55%–120%. The average life expectancy decreases by up to five years with a deficient content of Ca and Mg. The basis for calculating this type of health risk is to determine the average daily required dose (ADRD), the dose of the element necessary for the healthy development of a person. The ADRD can be derived in two ways. The first way is very simple, using various recommendations or limit values, if any. The second way is much more complicated; it requires a comparison of the health status of populations with various contents of elements in the environment (e.g., drinking water and soil), thus including the use of epidemiological studies. From ADRD, ADMD can be calculated.
The reference dose and similar concepts (tolerable daily intake and acceptable daily intake) work with the total exposure of elements or substances; thus, these are the combined exposure of ingestion, inhalation and dermal exposure, as far as they are relevant for the substance. In our model, we differ because we only consider the intake of elements by drinking water—there is a certain parallel in the assessment of health risks of airborne contaminants, where HQ is calculated as the ratio of the concentration of a substance in the air and the reference concentration for this substance (both quantities in units mg·m−3
). This important premise, that the effect of insufficient intake of essential elements by drinking water is to some extent independent of food intake of the given element, is derived mainly from animal experiments. They showed that even though animals receive a complete diet, which ensures their need for essential elements for one hundred percent, the low content of these elements in served drinking water will also have a negative effect on their health [24
]. However, the premise is also indirectly confirmed by epidemiological studies carried out in developed countries. Although there is no problem with malnutrition of the population in developed countries, there are still significant differences in the mortality of the population among localities with different Ca and Mg contents in drinking water.
As can be seen from Table 6
, the HQd
values for Ca, Mg and (Ca + Mg) are in a narrow range of 2.94–2.95, which corresponds to the medium risk of chronic disease development in the sense of the US EPA. In contrast, meta-analyses of more valid epidemiological studies [15
] result in odds ratio or relative risk of death from cardiovascular disease for water with a higher magnesium content at the level of about 0.75–0.80, which is still a statistically significant risk, but relatively low compared to other known risk factors (for cardiovascular diseases), as it means about a 20%–25% decrease in mortality. However, in the case of high number of cardiovascular diseases in the population, the impact of such relatively low risk factor on public health can be significant. The relatively high level of risk of chronic diseases (HQd
e.g., for Mg = 2.94) fully corresponds to the health status of people supplied with soft drinking water. Mortality from the most common causes of death in Slovakia (cardiovascular and oncological diseases) in the “soft” water group is more than 50% higher compared to the population supplied with hard water. Mortality from the digestive and respiratory systems is two times higher compared to the population supplied with hard water.
The narrow range of three average HQd (2.94–2.95) is quite surprising. Not for hardness because it is only a function of the sum of Ca + Mg, but surprising is the similarity of HQd for calcium and magnesium, although both have different functions and effects in the body. The only explanation of this finding is that the effect of magnesium is key for the four diagnoses monitored and its different content in drinking water in the monitored municipalities contributes significantly to the identified health status. However, because the presence of calcium in water is usually associated with magnesium (i.e., soft waters have low contents of both elements, while harder waters are rich in Ca and Mg), the correlation between health status and Ca content in water is similar to that for magnesium. However, any mutual interaction of the two elements cannot be ruled out.
Of all 2037 individual diagnoses listed in the International Classification of Diseases (ICD), 10th revision [33
], there was no better HI value in the “soft” water group compared to the “hard” water group. When focusing on individual causes of death, the highest difference in the causes of death was observed for diagnosis I21
, acute myocardial infarction, 3.23 times higher in the “soft” water group, diagnosis I25
, chronic ischemic heart disease, 2.17 times higher in the “soft” water group and for diagnosis C34
, bronchial and lung malignancy, 2.17 times higher in the “soft” water group than in the “hard” water group. As the highest and the most noticeable difference is the number of deaths from the diagnosis of G80
-infantile cerebral palsy. Only two cases were recorded over the 15 years reviewed in the “hard” water group and up to 34 cases in the “soft” water group. Due to the fact that the cases did not form local or time clusters, it can be assumed that the higher incidence was not caused by any infectious epidemic but another factor or factors, which we hypothesise to be the different hardness of drinking water (see [28
]). Clearly, the health status of the population supplied with soft drinking water is much worse than the health status of the population supplied with hard drinking water.
Calcium and magnesium are among the oldest recognized essential elements needed for human health [12
]. Adequate amounts of Ca and Mg are required for the proper functioning of human body and maintenance of homeostasis. The recommended daily doses of Ca and Mg for adults are between 700–1200 mg·day−1
and 320–420 mg·day−1
], respectively. Although dairy products are the most prominent source of Ca in the diet, drinking water is another important source of Ca. Water is a major component of the human body and is involved in many body functions, including the transport of nutrients and the removal of waste and toxins. In order to maintain good hydration and body balance, daily intake should reach 1.2 to 2.5 L of water, although needs may vary depending on the age, physical activity or climatic conditions [43
]. Since the 1990s, studies have been conducted to assess the bioavailability of Ca contained in calcium-rich mineral water compared to the bioavailability of Ca consumed in dairy products [44
]. These studies show that the bioavailability of Ca from Ca-rich drinking water is equal to or even higher than that from milk and dairy products.
Several epidemiological studies in North America and Europe confirm that children and adults who consume Western type diet have low intake of Mg (30%–50% of the recommended nutritional dose) [50
]. For example, 45% of US citizens are at risk of magnesium deficiency and 60% of adults are under ADI (Average Dietary Intake) [51
]. Drinking water is an important source of Mg when it contains Mg at concentrations of around 30 mg·L−1
Calcium and magnesium are found in drinking water of natural origin almost exclusively in the form of real ions and are thus fully accessible to the human body. In various foods, Mg is often bound in the form of complex organic compounds, and therefore its bioavailability to the human body is lower [55
]. Under the insufficient intake of these elements by food and their borderline deficit, minor intake from drinking water might play a decisive health role. People who regularly consume drinking water with low Mg content, for example 5 mg·L−1
, compared to people consuming water with elevated Mg content (e.g., 30 mg·L−1
) have a daily Mg intake by 50 mg lower, assuming 2 L of water daily. This difference represents approximately 15% of the recommended daily dose. The same is true for Ca.
The effect of Mg deficiency on the increased incidence of cardiovascular diseases was confirmed by several independent meta-analyses [15
]. The meta-analysis of Gianfredi et al. [41
] additionally found a statistically significant protective effect of water Ca on cardiovascular diseases. In addition, decreased vascular flexibility and higher arterial age of people drinking soft water were confirmed by Rapant et al. [56
]. It was also shown that people drinking Ca and Mg deficient water had a shorter lifespan [16
The effect of low Mg content and partly also Ca in drinking water on increased mortality from oncological diseases was proved by several studies [55
]. The data in our work show that the deficient Ca and Mg contents in drinking water have a significant effect on increased mortality and diseases of the digestive and respiratory system. Although no advanced epidemiological study has yet been performed to directly confirm the relationship to drinking water, the negative effect of insufficient Ca and Mg intake on the state of the immune system, which plays a key role in the etiology of respiratory and digestive diseases, is known. Our ADMD calculation is based on the MRC value derived specifically for the Slovak Republic according to the national epidemiological and health data. This is essentially the preferred approach for health risk assessment by the US EPA method—giving preference to data on the dose-effect relationship and exposure from the same population as the one being assessed. However, this is rather an exception when routinely using the health risk assessment method because such locally produced data are quite rare and mostly dose-response data generated by other studies in other parts of the world have to be taken over. A typical example is the dose-response of arsenic where its limit content in drinking water recommended by the US EPA or WHO was, until recently, based on the half-century-old results of an epidemiological study from Taiwan [60
]. Researchers evaluating the Ca and Mg deficiency in water for a population in another part of Europe or the world can therefore use our MRC values but they should be aware of possible differences in drinking water quality, eating habits and health status of the Slovak population.
The Uncertainty Factor value used in this study is also questionable. The value of UF in methodologies for calculating health risks introduced by the US EPA is mostly in the range of 1–10. This factor takes into account all possible uncertainties as well as the possible impact of other health determinants. In our case, UF values of 2 and 3 were taken into account. Based on the results from the application of artificial neural networks [23
], UF value of 2 was finally selected. In this way, MRC of Ca and Mg could be determined, at which mortality from the four main causes of death in the Slovak Republic reaches an average. If the UF value of 3 and the subsequently derived MRC were used, the health status would already be better than the average value.
As our assessment revealed a medium, thus relatively serious level of risk, and we are talking about the need for corrective action, the question naturally arises as to what measures are relevant and feasible in this case. The simplest measure seems to recommend to the local population that they should try to compensate the deficit of Ca and Mg from drinking water by changing or modifying the diet. Such a recommendation carries essentially no risk, but it is not possible to estimate how effective it will be. On the one hand, the willingness of the population to change their lifestyle, including nutrition, is not usually high, and on the other hand, we do not even know whether this would actually be reflected in the improvement of health. Our work is based on the assumption that the intake of Ca and Mg by drinking water and its effect are, to some extent, independent of food intake. We also do not know of any intervention study that would confirm the effectiveness of compensating water deficit with food. Therefore, measures should be aimed primarily to increase the content of these elements in water. It can be in the form of recommendation to the population to buy bottled water with higher content of these elements or, still better, to increase the hardness of the supplied drinking water. This can be achieved by selecting a more suitable water source, which may not be available at all in a certain geological area, or by treating the water to increase Ca and Mg contents. Although such treatment is carried out for many soft waters, it is usually done in the simplest and cheapest way, which consists in increasing Ca content in order to reduce the corrosive power of the water. A more suitable technology could also supply Mg, and in addition to the technological function, also fulfil a health function. This is the direction of our next work.
Calcium and magnesium are not regulated as obligatory indicators either under the WHO guidelines or under an EU directive or other international recommendation. In the currently completed revision of the EU Directive 98/83/EC, at least one qualitative requirement was added to replenish the content of minerals in water, the content of which was significantly reduced due to the treatment or conditioning: “This applies particularly to waters undergoing treatment (demineralization, softening, membrane treatment, reverse osmosis, etc.). Where water intended for human consumption is derived from treatment that significantly demineralizes or softens water, calcium and magnesium salts could be added to condition the water in order to reduce possible negative health impact, as well as corrosion or aggression of water and to improve taste. Minimum concentrations of calcium and magnesium or total dissolved solids in softened or demineralized water could be established taking into account the characteristics of water that enters these processes.” [61
Of the 28 EU countries, the minimum content of Ca, Mg or hardness in national standards as a minimum recommended value is regulated in only 11 countries, with varying degrees of binding force [62
]. In terms of the results of the present study, probably the most correct standard for drinking water is in the Czech Republic. The Czech standard recommends minimum Mg and Ca contents of 10 mg·L−1
and Ca 30 mg·L−1
, respectively. However, this applies only to water in which Mg and Ca contents are reduced by the treatment. In addition, the standard recommends ranges of the following elements as an optimum in terms of health impact: Mg 20 up to 30 mg·L−1
and Ca 40 up to 80 mg·L−1
]. At least, these Ca and Mg contents in drinking water can be recommended to keep the population at the lowest possible health risk.