Influence of Adverse Ecological Factors on the Incidence of Malignant Neoplasms

. An analysis of the leading sanitary-ecological factors of the environment affecting the incidence rates among the population of the Republic of Karakalpakstan in 2009-2018 found that between the indicators of the primary incidence of malignant neoplasms and water samples of open reservoirs that are not standard in chemical composition, strong direct correlations in the Northern (rxy = 0.89) zone, average direct correlations in the Western (rxy = 0.67) and Central (rxy = 0.57) zones were established. Similarly, strong connection was found in the Chimbay (rxy = 0.73) district, moderate correlation in Nukus (rxy = 0.44) and Ellikkala (rxy = 0.66) districts. The moderate correlation of the dynamics of malignant neoplasms with indicators of chemical pollution of tap water in the Chimbay (rxy = 0.33) district, well water in Kanlykul (rxy = 0.32) and atmospheric air in the city of Nukus (rxy = 0.41) was revealed and in the Republic of Karakalpakstan (rxy = 0.39). The solution to the problem lies in providing for the population of the selected districts with centralized drinking water that meets the state standard.


Introduction
At Karakalpak State University, together with the Medical Institute of Karakalpakstan and the Republican Center for Sanitary and Epidemiological Welfare (abbreviated RC SEW; former abbreviation RC SSES) a research is being carried out to study and assess the most important environmental pollution in the context of cities, regions and conditionally allocated zones, to determine their impact on human health, as well as identify the most contaminated areas, and to identify risks which may lead to diseases.
At the end of the 20th century and at the beginning of the 21st century, as a result of a sharp reduction in the water flow from the Amudarya and Syrdarya rivers into the Aral Sea, the actual disappearance of the Aral Sea, the ecological balance was disturbed in the Aral Sea region, the external environment was degraded, an extreme condition for people's habitation arose [1,2,3,4].
Polluted atmospheric air, soil and water serve as a source of pollution of plant and livestock products, later used by the population as products of food [2,3,4,5]. One of the diseases that needs to be studied in connection with harmful ecological factors in the Republic of Karakalpakstan (hereinafter: RK) is malignant neoplasms (hereinafter: MN).
The purpose of this study is to identify and evaluate the leading sanitary-ecological factors of the environment that affects human health and appears on incidence level of malignant neoplasms among the population of cities, districts and conditionally allocated zones of the RK.

Data
The data for the study was f-12-health reports from the Health Institute of the Ministry of Health ( The dynamics of the incidence of MN with a diagnosis for the first time (incidence per 100 000) in the RUz and the RK, tends to decrease slowly (Table 1). According to the statistical reports of oncological dispensaries, in 2010 in the RUz, intensive incidence rates for malignant neoplasms amounted to 67.7 (per 100 000 population; from now on the unit of incidence is not indicated but considered to be "per 100 000 population"), among rural residents 79.0 cases, among women 73.4 cases. In 2018, these values were 71.0, 82.6 and 82.9, respectively. Breast cancer was the No. 1 form of MN in RUz in 2010 (11.9 cases), followed by stomach cancer (9.5) and circulatory and lymphatic tissue (8.6). In the RK, forms of MN (in descending order) are esophageal cancer (17.6), stomach cancer (10.4) and cervical cancer (10.0) [6].
In 2018, the form of oncological incidence in the whole RUz most often was breast cancer (10.9), followed by stomach cancer (5.7), and cervical cancer (5.0). In the RK in 2018, (in decreasing order) esophageal cancer has accounted for 12.0, stomach cancer 10.4, breast cancer 9.8 and cervical cancer 8.7 cases.
Compared with 2009, in 2018 in the RUz there was an increase in the incidence of breast cancer by 45.3%, in the RK by 22,5%.
According to the MH of the RUz [9], the incidence rates of MN in 2012 amounted to 156.0 in Kazakhstan, 114.0 in Kyrgyzstan, 72.8 in Tajikistan, and 64.5 in Uzbekistan.
In 2010, cancer mortality in Uzbekistan amounted to 34.5, while in Karakalpakstan 43.0, therefore the number of cases in the RK 24.6% above than the RUz average. And in 2018, cancer mortality in Uzbekistan amounted to 41.1, in Karakalpakstan 51.4, indicating again relevantly (25%) more cases in the RK. The above regional differences in the primary oncological incidence and the excess of the incidence rate of Karakalpakstan over Uzbekistan prompted us to disclose their causes.
Given the uneven distribution of incidence by territory and by time, the territory of the RK is conditionally divided into 4 zones: the Western zone (Muynak, Kungrad, Kanlykul and Shumanay districts), the Northern zone (Takhtakupyr, Karauzyak, Chimbay, Kegeyli districts), the Central zone (Nukus city, Khodjeyli, Takhiatash and Nukus districts), and also the Southern zone (Amudarya, Beruni, Ellikkala and Turtkul districts).
In order to increase the visibility of the tendencies, the studied 10-year period is divided into 2 five-year periods: 2009-2013 and 2014-2018.
When studying the dynamics of the incidence of MN with a first established diagnosis, in the second five-year period compared with the first, an increase in the incidence rate was found in the Western (by 1,5%), Northern (by 1,8%), Central (by 7,6%) and Southern (5,1%) zone ( Table 2). In the next five-year period, the regional (RK) average was 68.9; relatively high primary incidence rates of MN were recorded in Chimbay (82.

Tests and results
We presume that the dynamics of the level of oncological incidence for regions under similar living conditions, lifestyle and nutritional features should be similar; significant influences are exerted by harmful external environmental factors resulting from the Aral ecological disaster. Therefore, we attempt to connect the dynamics of the primary incidence of MN with the main harmful sanitary-ecological factors that formed in the South Aral Sea region with variations of chemical pollution of drinking water and atmospheric air ( Fig. 2).
At the end of 2018, 62% of the population of the Republic was provided with tap water, the rest uses well water (mainly water from tubular wells) and water from open reservoirs.
First of all, we compared the levels of primary oncological incidence ( Table 2) with indicators of non-normative water samples of open reservoirs by chemical composition (by hardness and mineralization) ( Table 3).
The strength of the relationship between the primary incidence rates of oncological diseases (Table 2) and the indicators of water samples from open reservoirs, exceeding the maximum permissible concentration (MPC) of chemicals and indicators (Table 3) was analized for each zones and districts. As an example, the time series are plotted on Fig 1. for the Western zone of the RK.    The strength of the relationship between the curves on Fig 1. (and in other cases), correlation coefficient according to Pearson [10] has been determined. The calculation was carried out on a computer program Formulas. Though the time series is "short", i.e. including 10 samples only (k=10), to decide on the dependence of oncological cases and water quality measures, a hypothesis test should be conducted. The hypothesis-test for correlation coefficient can be performed by a Student's t-test with degrees of freedom k−2 [11], where the null-hypothesis is that the two time series are uncorrelated, i.e.
For 'large' number of samples, it is known that follows a Student's t-distribution with degrees of freedom k−2. If then the null-hypothesis should be accepted, thus the two time series are independent at a confidence level p. The values of t-distribution at k-2=8 freedom for probability of 0.8, 0.9 and 0.95 are 1.397, 1.806 and 2.306, respectively. Accordingly, the dependence can reliably by assumed, when the null-hypothesis is rejected, i.e.
Note, however, that the results may be distorted due to the small number of samples. According to [12], in case of small samples, the distribution may differ.  Table 4. In the Table those cells has been emphasized by green color, where correlation has been accepted: light green is used for cases accepted on 0.8 probability level (the value of exceeds 1.397), darker green shows the cases, which are accepted on 0.9 probability level (the value of exceeds 1.806), while the darkest green shows those cases, which could have been accepted on 0.95 probability level (the value of exceeds 2.306).
The results in Table 4 suggest no systematic correlation by time or by location. The strong correlation in Chimbay is mainly a result of the strong correlation during the 2009-2014 period, which cannot be observed anymore. A notable (a quite unfavorable) feature is that in Muinak, Kalykul and in Amudarya there was no correlation observed in the 2009-2013 period, but it became relevant in the 2014.2018 period.
Then we compared the levels (incidence) of primary oncological diseases (Table 2) with the indicators of tap water samples, the percentage of which exceeded the MPC of chemical substances and indicators ( Table 5).
Rationing of drinking tap water is based on the application of the State Standard of Uzbekistan (UzSS) 950-2011 "Drinking water. Hygienic Requirements and Quality Control" (SS-state standard) updated every 10 years. The district laboratories of the sanitary-epidemiological service measured the hardness, mineralization (dry residue) in the water sample. This dry residue in our regions mainly consisted of nitrates, sulfates and chlorides. No pesticide contamination of water and food has been detected over the past 10 years. They have ceased to be used in agriculture for 25 years.
The results of the comparison are shown on Table 6. In the Table light green is used for cases accepted on 0.8 probability level (the value of exceeds 1.397), darker green shows the cases, which are accepted on 0.9 probability level (the value of exceeds 1.806), while the darkest green shows those cases, which could have been accepted on 0.95 probability level (the value of exceeds 2.306).
The levels of primary oncological diseases depended on chemical contamination ( Table  2 and Table 5) at several locations in the Western, Northern and Southern regions during the 2014-2018 period. It is quite unfavourable situation as at most locations no strong correlation was observed before. As for the Central region, no relevant correlation was observed for the 5 year periods (apart from Khojeli during the 2009-2013 period), but the correlation seems to be convincing for the 10 year period (Fig.3).    Table 6. Correlation between the primary incidence rates of MN in the RK per 100 000 population (Table 2) and the percentage of tap water samples exceeding the MPC of chemical substances and indicators ( Comparison of the levels (incidence) of primary oncological diseases (Table 2) with the indicators of well water samples exceeding the MPC in terms of chemical indicators (Table  7) is shown on Table 8. In the Table light green is used for cases accepted on 0.8 probability level (the value of exceeds 1.397), darker green shows the cases, which are accepted on 0.9 probability level (the value of exceeds 1.806), while the darkest green shows those cases, which could have been accepted on 0.95 probability level (the value of exceeds 2.306).
According to the results, the correlation seems to be stronger by time in every district of the Western region, in contrary, when the average values for the Western region are analysed, the tendency is opposite: the correlation is stronger in the 2009-2013 period.
In the Northern region correlation at 0.8 probability level can be observed at several districts during the 2009-2013 period, which has been absent for the 2014-2018 period, apart from Karauzyak, where the correlation became notably stronger.
In the Central region there is an obvious correlation (at least at 0.8 probability level) for both 5 year periods, but it cannot be detected for the whole 10 year long time series.
Not any dependence in the Southern region of the RK has been observed.

Conclusions
The present investigation has made use of spatial and statistical analysis tools, so establishment of an analytical relation between the causes of high mortality from MN and the mentioned environmental factors would require further studies.
According to Tables 4, the strong relation between the incidence rates of MN in the RK per 100 000 population (Table 2)  The correlation between the primary incidence rates of MN and the percentage of tap water samples exceeding the MPC of chemical substances and indicators (see in Table 6  As it can be observed in Table 8, the correlation between the primary incidence rates of MN and the percentage of tap water samples exceeding the MPC of chemical substances and indicators shows strong correlation (at 0.95 confidence level) only occasionally, in Kungrad and Karauzyak. A very favorable feature is that there is no correlation at all in the Sourthern region of RK.
Since the small number of samples distorts the results, also lower confidence level should be considered. At 0.8 confidence level, the correlation between primary incidence rates of MN and non-normative water samples of open reservoirs by chemical parameters (Table 4) and also for the tap water samples (Table 6) was established at several districts and regions of the RK. Therefore, it is reasonable to search for a dependence analytically as well. The results for the well water samples ( Table 8) are not that convincing, but as open reservoirs may more properly reflect the actual water quality (as in open reservoirs the water is more homogeneous due to the diffusion of the chemical particles than in the water supply system), the source of the dependence is worth for an analytical investigation.
All in all, the results obtained allow us to conclude that the higher the proportion of non-normative water samples, i.e. chemical contamination of water in open reservoirs and non-normative samples of tap water and atmospheric air in specified districts, the higher primary malignant oncological incidence rates. The solution to the problem of the consumption of contaminated water by the population for drinking purposes is to maximize the coverage of the population of the allocated districts with centralized water supply and bring the quality indicators of the tap water supplied to the population in accordance with the state standard. Our results can contribute to planning the reduction of the pollution of the environment, providing a measure for a differential approach for specific territories of the RK.