Quantitative Assessment Of Uncertainty And Variability Of Sewage Epidemiology Based On Statistical Method

Sewage epidemiology is a new method of investigating drug abuse and consumption in recent years.Although it has been applied in many fields,there are still some problems,among which the key scientific problem is uncertainty.Probability distribution is one of the ways to solve uncertainty.Probability based Monte Carlo analysis is an important method of uncertainty analysis.Point estimation and interval estimation of input parameters and determination of distribution type are prerequisites for Monte Carlo analysis.Taking the methamphetamine abuse and prevalence in Dalian as an example,this paper makes point estimation,interval estimation and distribution type determination of the related input parameters.Firstly,the influent samples from 11 sewage plants were collected in 10 months in 2017,and the concentrations of methamphetamine were analyzed.The real-time population based on the concentration of ammonia nitrogen was estimated.The abuse and prevalence of methamphetamine in Dalian were predicted.The results showed that the amount of methamphetamine abuse per capita in Dalian changed significantly in different months.The average abuse of methamphetamine in Dalian in spring,summer,autumn,and winter was 188 ± 68 ?g/d,377 ± 259 ?g/d,18.4± 12.5 ?g/d and 40.9 ± 24.2 ?g/d,respectively.The abuse of methamphetamine per capita in east Dalian is higher than that in other areas.This area is the main scenic tourism and resort area in Dalian and the abuse of methamphetamine among tourists may cause high abuse in this area.According to the predicted per capita abuse,frequency and dose,and the proportion of the population,the prevalence of methamphetamine among adults(15-64 years of age)fluctuated between 0.05%and 1.4%.In order to verify the reliability of the calculation results,the uncertainty and variability of methamphetamine abuse per capita and prevalence in Dalian were analyzed based on probability analysis of Monte Carlo simulation.Anderson-Darling test was used to determine the logarithmic normal distribution of methamphetamine in sewage and the normal distribution of ammonia nitrogen concentration in sewage,flow rate and population data based on ammonia nitrogen in sewage.The bootstrap method based on resampling technique was used to determine the distribution characteristics of the characteristic values of small sample data such as abuse frequency,abuse dose and correction factor.Based on the above distribution and characteristics,the uncertainty and variability of the methamphetamine abuse and prevalence in Dalian were analyzed.The results showed that the methamphetamine abuse and prevalence per capita in Dalian were 204 ?g/d(95%Cl:31-550 ?g/d)and 0.42%(95%CI:0.02-1.94%),respectively.In this paper,the distribution types of large and small sample data in sewage epidemiology are determined to improve the accuracy of the investigation results of sewage epidemiology,and provide a scientific basis for drug control and drug abuse prevention.