PrAna: an R package to calculate and visualize England NHS primary care prescribing data

Assessment of restriction measures implemented during COVID pandemics on community lifestyle choices via wastewater-based epidemiology

SARS-CoV-2 pandemic affected lifestyle habits, and the mental health and wellbeing of people around the world. In this manuscript, two towns (Paulton and Radstock) and two cities (Bath and Bristol) in Southwest England (> 1 million people) were monitored for two years using Wastewater-Based Epidemiology to assess impacts of COVID (including management measures such as lockdowns and movement restrictions) on community lifestyle choices: illicit drugs, legal stimulants, abused pharmaceuticals and pain pharma usage. Results were triangulated with key dates captured during the pandemic (national lockdowns, restrictions and social distancing measures, etc.). This highlighted a reduction in cocaine intake (as benzoylecgonine) (community average: -36 %) during the first lockdown and an increase in illicit drugs usage after the 3rd national lockdown (community averages for amphetamine: +8 %, cocaine/benzoylecgonine: +39 %, ketamine:+70 %) when restrictions were removed, and social interaction and recreational activities increased. There was a reduction in the intake of caffeine (as 1,7-dimethylxanthine) (community average:-39 %) after 3rd national lockdown coinciding with pubs reopening while nicotine intake (via cotinine) remained stable indicating lack of impact of COVID on smoking habits. Pain pharma often used in pain management resulting from injuries linked with sport activities (naproxen and diclofenac) showed decrease in usage due to lockdown restrictions in physical exercise and access to gyms/sport facilities.

Understanding treatment of pain during SARS-CoV-2 pandemic in a two-year intercity longitudinal study using wastewater-based epidemiology

SARS-CoV-2 pandemic had a significant impact on the society, economy, and health of people around the world with consequences that need to be better understood for future pandemic preparedness. This manuscript provides insights into the usage of pharmaceuticals for pain treatment management throughout SARS-CoV-2 pandemic. Four towns and cities with a total population of > 1 million people covering an area of 2000 km2 in South West England were monitored for twenty-four months. Results showed different patterns in pain pharma usage, with small towns having higher population normalised daily loads (PNDLs) than big cities for majority of pain killers studied. This is likely due to demographics of these cities with smaller cities having older population. Per capita consumption of non-steroidal anti-inflammatory drugs (NSAIDs) increased compared to pre-pandemic usage in line with SARS-CoV-2 infections (ibuprofen and acetaminophen), while body pain drugs (diclofenac and naproxen) decreased in line with restrictions and closure of sports facilities. Changes in population normalised daily intake (PNDI) of pain killers were particularly apparent during the 1st and 3rd national lockdown. Comparison of PNDIs with prescriptions highlighted differences related to medication availability (OTC drugs) and patients' nonadherence (prescribed drugs). In addition, several instances of direct disposal events across the catchments were observed which raises an issue of lack of pharma compliance and general understanding of potential environmental impacts from pharma usage.

A new Wastewater-Based Epidemiology workflow to estimate community wide non-communicable disease prevalence using pharmaceutical proxy data

This manuscript introduces a new wastewater-based epidemiology workflow for estimation of non-communicable diseases (NCDs) prevalence by using wastewater-based epidemiology (WBE) and pharmaceuticals/their metabolites as proxies for NCDs prevalence. 83 targets were selected (54 parent pharmaceuticals and 29 metabolites). Three critical aspects were tested: (i) Solid-Phase Extraction - Ultra-Performance Liquid Chromatography and Tandem Mass Spectrometry (SPE-UHPLC-MS/MS) method performance, (ii) biomarker stability under variable storage conditions (during sampling and long-term storage) and (iii) accounting for human metabolism in WBE back-calculations. High stability of most analytes was observed under tested storage conditions. A few exceptions include diazepam, dihydroketoprofen and 5-hydroxy-lansoprazole. Analyte recoveries varied between 75% and 125% for most analytes. MDLs ranged from 0.2 ng L-1 to 5.6 ng L-1, while MQLs from 0.2 ng L-1 to 16.8 ng L-1. The overall average method accuracy and precision were: 99.5% and 4.0% respectively. A fully validated method was tested using community wastewater in the Southwest of England to estimate pharmaceutical usage, test metabolism correction factors established and compare results with prescription data. The new WBE method for NCD approximation allowed for the estimation of the daily usage/intake of 69 NCD targets with a standardized approach and a consistent reporting format.

An integrated One Health framework for holistic evaluation of risks from antifungal agents in a large-scale multi-city study

A new framework for retrospective mass spectral data mining for antifungal agents (AFs) and Wastewater-Based Epidemiology (WBE) was developed as part of One Health framework to tackle risks from AFs. A large scale, multi-city study was undertaken in South-West England. Key drivers of AFs in the catchment were identified with communal wastewater discharges being the main driver for human AFs (fluconazole, ketoconazole) and agricultural runoff being the main driver for pesticide AFs (prochloraz, prothioconazole and tebuconazole). Average WBE-estimated human used fluconazole and ketoconazole PNDIs (population normalised daily intake) exceeded 300 mg day-1 1000 inh-1 and 2000 mg day-1 1000 inh-1. This is much higher than PNDPs (population normalised daily prescriptions <40 mg day-1 1000 inh-1 and <80 mg day-1 1000 inh-1 for fluconazole and ketoconazole respectively). This was expected due to both prescription and over-the-counter usage, and both oral and topical applications. Pesticide AF, prothioconazole had PNDIs <40,000 mg day-1 1000 inh -1, which gave intake: 0.43, 0.26, 0.07 mg kg-1 in City A, B, and C, likely due to accounting for external/non-human sources. This is higher than the acceptable daily intake (ADI) of 0.01 mg kg-1bw day-1, which warrants further study. Intake per kg of body weight estimated using tebuconazole was 0.86, 1.39, 0.12, 0.13, and 2.7 mg kg-1 in City A-E respectively and is likely due to external/non-human sources. Intake calculated using its metabolite was 0.02 and 0.01 mg kg-1 in City B and C respectively, which aligned with ADI (0.03 mg kg-1bw day-1). The environmental risk assessment of AFs indicated low/medium risk from fluconazole, prochloraz, and tebuconazole, medium risk from epoxiconazole, prothioconazole's metabolite, and tebuconazole, and high risk for prothioconazole in river water. High risk was estimated from fluconazole, epoxiconazole, prothioconazole and its metabolite, tebuconazole, ketoconazole in wastewater samples, which is important during raw sewage discharge events via sewer overflows.

Wastewater-based epidemiology for comprehensive community health diagnostics in a national surveillance study: Mining biochemical markers in wastewater

This manuscript showcases results from a large scale and comprehensive wastewater-based epidemiology (WBE) study focussed on multi-biomarker suite analysis of both chemical and biological determinants in 10 cities and towns across England equating to a population of ∼7 million people. Multi-biomarker suite analysis, describing city metabolism, can provide a holistic understanding to encompass all of human, and human-derived, activities of a city in a single model: from lifestyle choices (e.g. caffeine intake, nicotine) through to health status (e.g. prevalence of pathogenic organisms, usage of pharmaceuticals as proxy for non-communicable disease, NCD, conditions or infectious disease status), and exposure to harmful chemicals due to environmental and industrial sources (e.g. pesticide intake via contaminated food and industrial exposure). Population normalised daily loads (PNDLs) of many chemical markers were found, to a large extent, driven by the size of population contributing to wastewater (especially NCDs). However, there are several exceptions providing insights into chemical intake that can inform either disease status in various communities or unintentional exposure to hazardous chemicals: e.g. very high PNDLs of ibuprofen in Hull resulting from its direct disposal (confirmed by ibuprofen/2-hydroxyibuprofen ratios) and bisphenol A (BPA) in Hull, Lancaster and Portsmouth likely related to industrial discharge. An importance for tracking endogenous health markers such as 4-hydroxy-2-nonenal-mercapturic acid (HNE-MA, an oxidative stress marker) as a generic marker of health status in communities was observed due to increased levels of HNE-MA seen at Barnoldswick wastewater treatment plant that coincided with higher-than-average paracetamol usage and SARS-CoV-2 prevalence in this community. PNDLs of virus markers were found to be highly variable. Being very prevalent in communities nationwide during sampling, SARS-CoV-2 presence in wastewater was to a large extent community driven. The same applies to the fecal marker virus, crAssphage, which is very prevalent in urban communities. In contrast, norovirus and enterovirus showed much higher variability in prevalence across all sites investigated, with clear cases of localized outbreaks in some cities while maintaining low prevalence in other locations. In conclusion, this study clearly demonstrates the potential for WBE to provide an integrated assessment of community health which can help target and validate policy interventions aimed at improving public health and wellbeing.

The burden of city's pain treatment - A longitudinal one year study of two cities via wastewater-based epidemiology

This paper explores Wastewater-Based Epidemiology (WBE) as a tool enabling understanding of city's pain treatment in an intercity longitudinal study. An intensive 13-month monitoring programme was undertaken in two adjacent urban areas in South-West England: a small commuter town Keynsham and the city of Bath (>180 samples collected). The study has shown a great potential of using triangulated WBE and National health Service (NHS) prescription data in understanding pain treatment in two contrasting communities with strong apparent seasonal patterns of short pain medications vs chronic pain treatment as well as the type of treatment used (e.g. oral vs topical). Community-wide usage of Non-Steroidal Anti-inflammatory Drugs (NSAIDs) and paracetamol in the intercity study is population size and season driven with the highest usage recorded in winter months. This contrasts with other pain pharmaceuticals, especially those used for chronic pain, where no/limited seasonal usage was recorded. Unmetabolized NSAIDs are, to a large extent, directly disposed of into the sewerage system bypassing metabolism due to their topical application. This is particularly apparent in winter months with naproxen showing the highest seasonal variability. Pharma/met (ratio of pharmaceutical and its metabolite concentration) analysis allows for tracking topical (non-metabolic) application/down-the-drain disposal of pharmaceuticals with frequent instances of direct disposal of NSAIDs into the sewerage system observed. Normalisation of pharma markers to population size shows comparable estimates of pharma usage in the two cities confirming population as the main driver of pharma loads in wastewater. Variable application patterns of pain pharmaceuticals make back-calculation of intake more convoluted. Intake calculated using percentage excretion of parent NSAIDs will likely lead to overestimation, as it is assumed that NSAIDs are subject to extensive metabolism (this is not the case for topical applications). Intake calculated using percentage excretion of metabolites (or parent compound) as consumption markers leads to underestimation of NSAIDs usage due to contributions from topical application not being accounted for. Prescription data indicates cumulative internal and topical usage, but the data ignores large proportion of over-the-counter usage. Therefore, we have proposed a combined approach allowing for estimation of total usage including, and differentiating between, topical application and oral administration.

Human population as a key driver of biochemical burden in an inter-city system: Implications for One Health concept

This paper tests the hypothesis that human population and city function are key drivers of biochemical burden in an inter-city system, which can be used to inform One Health actions as it enables a holistic understanding of city's metabolism encompassing all of the activities of a city in a single model: from lifestyle choices, through to health status and exposure to harmful chemicals as well as effectiveness of implemented management strategies. Chemical mining of wastewater for biophysico-chemical indicators (BCIs) was undertaken to understand speciation of BCIs in the context of geographical as well as community-wide socioeconomic factors. Spatiotemporal variabilities in chemical and biological target groups in the studied inter-city system were observed. A linear relationship (R² > 0.99) and a strong positive correlation between most BCIs and population size (r > 0.998, p < 0.001) were observed which provides a strong evidence for the population size as a driver of BCI burden. BCI groups that are strongly correlated with population size and are intrinsic to humans' function include mostly high usage pharmaceuticals that are linked with long term non-communicable conditions (NSAIDs, analgesics, cardiovascular, mental health and antiepileptics) and lifestyle chemicals. These BCIs can be used as population size markers. BCIs groups that are produced as a result of a specific city's function (e.g. industry presence and occupational exposure or agriculture) and as such are not correlated with population size include: pesticides, PCPs and industrial chemicals. These BCIs can be used to assess city's function, such as occupational exposure, environmental or food exposure, and as a proxy of community-wide health. This study confirmed a strong positive correlation between antibiotics (ABs), population size and antibiotic resistance genes (ARGs). This confirms the population size and AB usage as the main driver of AB and ARG levels and provides an opportunity for interventions aimed at the reduction of AB usage to reduce AMR. Holistic evaluation of biophysicochemical fingerprints (BCI burden) of the environment and data triangulation with socioeconomic fingerprints (indices) of tested communities are required to fully embrace One Health concept.