Systematic and Day-to-Day Effects of Chemical-Derived Population Estimates on Wastewater-Based Drug Epidemiology

Abstract

Population size is crucial when estimating population-normalized drug consumption (PNDC) from wastewater-based drug epidemiology (WBDE). Three conceptually different population estimates can be used: de jure (common census, residence), de facto (all persons within a sewer catchment), and chemical loads (contributors to the sampled wastewater). De facto and chemical loads will be the same where all households contribute to a central sewer system without wastewater loss. This study explored the feasibility of determining a de facto population and its effect on estimating PNDC in an urban community over an extended period. Drugs and other chemicals were analyzed in 311 daily composite wastewater samples. The daily estimated de facto population (using chemical loads) was on average 32% higher than the de jure population. Consequently, using the latter would systemically overestimate PNDC by 22%. However, the relative day-to-day pattern of drug consumption was similar regardless of the type of normalization as daily illicit drug loads appeared to vary substantially more than the population. Using chemical loads population, we objectively quantified the total methodological uncertainty of PNDC and reduced it by a factor of 2. Our study illustrated the potential benefits of using chemical loads population for obtaining more robust PNDC data in WBDE.

Critical review on the stability of illicit drugs in sewers and wastewater samples

“…Numerous efforts have focused on improving the accuracy (trueness and 130 precision) of different analytical methods, and inter-laboratory studies were conducted in order to evaluate and harmonize the different analytical procedures being used (Castiglioni et al, 2013;Thomas et al, 2012). Uncertainties associated with the population estimates have recently been conceptually reduced (Lai et al, 2015;O'Brien et al, 2014), as well as the uncertainty related to wastewater sampling (Ort et al, 2010). Furthermore, sample collection, storage and preparation 135 methods have been evaluated, deducing critical, substance-specific parameters: i) solvent and temperature used during the evaporation of solid-phase extraction (SPE) extracts, and ii) biomarker stability in the wastewater matrix and silanisation of glassware (Baker and Kasprzyk-Hordern 2011a).…”

Wastewater-Based Epidemiology of Cocaine in the Brazilian Federal District: Spatial Distribution, Weekly Variation and Sample Preservation Strategies

“…Also for this reason, analytical data concerning benzoylecgonine in wastewater samples has been used as a near real-time tool to estimate cocaine consumption for a given population. 8,9 Several works have explored different aspects of the WBE approach including other cocaine metabolites, 10 estimation of other consumed drugs, 10-12 influence of recreational events, [13][14][15][16] consumption dynamics in restricted facilities, [17][18][19] temporal and spatial variability, [20][21][22] stability of target chemicals, 23,24 refinement on estimates calculation, [25][26][27] innovations in sample preparation and analysis, [28][29][30] among others. In Brazil, the first work to put the WBE approach into practice was designed to assess cocaine consumption on regions served by six wastewater treatment plants (WWTP) installed in the Brazilian Federal District (FD).…”

Using wastewater-based epidemiology to estimate drug consumption—Statistical analyses and data presentation

“…WWTP coverage areas could be approximately matched with census blocks or tracts to create better estimates with better measurement error approximations, although daytime population change estimates may be available only for larger towns and cities. Work is ongoing to try to better estimate the size of the underlying population using a range of other measureable analytes (see, e.g., Brewer et al, 2012;Castiglioni et al, 2013;Lai et al, 2015Lai et al, , 2011Ort et al, 2014a). Finally, researchers could obtain data from mobile phone companies during the average day and average night to understand daily migration patterns.…”

“…Substituting a larger or smaller error component estimate has relatively small effects on the resulting confidence intervals in comparison to inter-municipality variation (sensitivity analyses using sampling error of 5% and 20% for MDMA appear in Supplement section S 9.6; others available upon request). Furthermore, more advanced methods of estimating error via Monte Carlo simulation are available, but the linear error propagation used here (equation 3) is generally thought to be an acceptable approximation especially if individual error components are not too large (Lai et al, 2015;Ort et al, 2014b). Finally, the results presented here are index loads of excreted substances, not estimates of the number of users or the average ingestion of a given parent substance.…”

“…Uncertainty associated with a composite sample within a day representing the whole day is subsumed in estimates of within-year sampling error. We thus have four sources of uncertainty, associated with sampling, chemical analysis, flow measurement, and population estimation, which are commonly assumed to be independent (e.g., Lai et al, 2015;Ort et al, 2014bOrt et al, , 2010a as no physical or conceptual dependence has been reported.…”