Toxicovigilance in the UK during the COVID-19 pandemic

Abstract

This study assesses the impact of the COVID-19 pandemic on enquiry numbers to the United Kingdom (UK) National Poisons Information Service (NPIS) to inform public health interventions. Phone enquiries were extracted and analysed from the UK Poisons Information Database (UKPID) in calendar years 2018 to 2020. Data collected included information on call, patient, and exposure characteristics for all enquiries and for specific enquiries involving selected agents of interest. During the first national lockdown there was a significant decrease in the proportion of enquiries about intentional exposures (20.6%) compared to pre lockdown (26%) and control periods (22.6%) (P < 0.001), potentially due to reluctance to attend healthcare facilities. Exposures regarding dental analgesics also increased significantly in lockdown (5.15%) compared to pre-lockdown (2.6%) and the control period (3.4%) (P < 0.001) likely due to limited access to dental services. Enquiries about hand sanitisers exposures showed a significant increase during lockdown (92) compared to pre-lockdown (32) and control (45) (P < 0.001) likely due to increased availability. No significant increases in exposures were identified for selected pharmaceuticals of interest and only a small number of enquiries relating to exposures with essential oils, methanol, Chinese herbal medicines antihistamines and deliberate bleach ingestion which offered reassurance from a public health perspective. This data led to public health interventions such as tweets, updates to guidance and advice to healthcare professionals and the public. Toxicovigilance is a key public health tool and poison centre enquiry data is helpful in identifying toxic exposures for any future pandemics or large-scale chemical incident.

Introduction

Toxicovigilance is the active process of identifying and assessing the threat or potential toxicity if an individual or population are exposed to consumer products, pesticides, pharmaceuticals, industrial chemicals, controlled substances, and natural toxins. Toxicovigilance involves monitoring data to identify trends in poisoning exposures and the emergence of new risks associated with toxic substances, as well as assessing the effectiveness of preventive measures.¹ On 30th January 2020 the World Health Organization (WHO) declared COVID-19 a Public Health Emergency of International Concern (PHEIC), only downgrading this public health emergency status on 5th May 2023 following a steady decline in associated mortality. Over the course of the pandemic, several preventive novel treatments (including a range of chemical and pharmaceutical preparations) were proposed and discussed within and outside the medical community. Social distancing measures and associated changes in behaviour also had the potential to impact the risk of exposure to domestic chemicals, such as hand sanitisers or cleaning products. The UK Health Security Agency (UKHSA, previously Public Health England during the study period) collaborated with the UK National Poisons Information Service (NPIS) to analyse UK poisoning data and evaluate whether the COVID-19 pandemic may have contributed to any potential public health risks related to toxic substances.

The NPIS is a national service that provides expert toxicological advice across England and the devolved health administrations of Wales, Scotland, and Northern Ireland. Four NPIS units share the responsibility for providing a 24-h telephone enquiry service to support frontline healthcare professionals diagnose and treat poisoned patients. Real-time data is recorded for each call enquiry and stored on a central database, the UK Poisons Information Database (UKPID). Data collected on UKPID can be used to study trends in enquiry numbers for the UK and the characteristics of the cases reported such as demographics, clinical features, treatments given or recommended and where possible clinical outcome. The NPIS also manages the content present on TOXBASE, an online database that healthcare professionals can use to obtain further information on chemical hazards and clinical treatment guidance. The NPIS also collects data on TOXBASE accesses to the system for specific poisons, which can be used to identify potential public health signals.

Toxicovigilance activities over the course of the COVID-19 pandemic were informed by anecdotal reports of poisoning or exposures from NPIS staff, as well as through Enhanced Event-Based Surveillance (EEBS) of global poisoning trends. EEBS was achieved through screening of international media, government announcements, and press releases issued by public health bodies for one year from 1 April 2020. The results of this work is detailed more extensively in a separate publication,² but a short summary for context is provided here.

Throughout the pandemic, there were international reports regarding increased exposure to certain agents including household products, hand sanitisers and methanol. There were also several reports of people self-medicating to prevent or treat COVID-19 infections with substances including but not limited to conventional and veterinary pharmaceuticals, supplements, and herbal remedies. Potentially hazardous exposures were also influenced by COVID-19 related policies and behaviours, for example, there were international reports that local authorities were unable to respond to environmental issues including algal blooms and proliferation of toxic plants due to restrictions from lockdown measures. Similarly, border closures appeared to result in the proliferation of highly toxic illicit drugs in certain countries (e.g. Canada) which resulted in significant increases in the numbers of serious poisonings. Misinformation spread about the use of other chemicals (e.g. bleach) and alcohol to treat COVID-19 also contributed to inappropriate use and poisoning. These reported trends were assessed by Public Health England (PHE) and the NPIS for their risk and potential to emerge as a public health issue in the UK and those that did were incorporated into the UK toxicovigilance work.

This process was undertaken throughout the COVID-19 pandemic and findings were shared across stakeholders across government responding to the incident to support public health messaging and interventions which are described later in this paper.

Methods

For this study, UKPID was retrospectively interrogated over the period of 1 January 2018 to 28 February 2021. Each year, week 1 commenced on the first Monday to allow for annual comparison (2018, 1 January; 2019, 7 January; 2020, 6 January; 2021, 4 January). Enquiries involving hand sanitisers in children under 6 yr, and those related to dental analgesic therapeutic errors were selected for more detailed review. Specific product categories were considered based on their potential to be influenced by the COVID-19 pandemic and were selected for further analysis based on Enhanced event-based surveillance and anecdotal reports received from specialists in poisons information. These included reported ingestion of hand sanitiser, cleaning products, methanol, essential oils, mushroom/fungi, certain pharmaceuticals, as well as inhalation of cleaning products and intentional bleach ingestion / injection.

Figures 1, 3, 5, 7, 9 illustrate the enquiry numbers (or % of total enquiries) over a 52-week period to allow comparison with key UK government public health interventions to illustrate potential impact on enquiry numbers. The event markers include the release of the COVID-19 Action Plan on the 3rd of March 2020, and stricter “first national lockdown” measures on the 24th of March 2020 (see further below). An event marker has also been added on the 11th May 2020, which coincides with the launch of the recovery guidance,³ and initiated the easing of lockdown measures in England and the change in public health messaging from “Stay at Home” to “Stay Alert.” Additional event markers are included when the tier alerting system was introduced (medium, high and very high alerts) on 14^th October 2020,⁴ the 2^nd national lockdown on 5^th November 2020⁵ and when the tier system was re-introduced on the 2^nd December 2020 as part of the COVID-19 Winter Plan.⁶ For enquiries related to dental analgesics, an event marker is also included for when the non-emergency dentists began to re-open on the 8th June 2020.⁷

Fig. 1.

Trend of phone enquiries (all agents) reported to the National Poisons Information Service over time during 2020 (COVID-19) in comparison to 2018 and 2019.

Fig. 3.

Trend of phone enquiries (all agents) in children under 6 reported to the NPIS over time during 2020 (COVID-19) in comparison to 2018 and 2019.

Fig. 5.

Trend of phone enquiries (all agents) with intentional (self harm) circumstances reported to the NPIS over time during 2020 (COVID-19) in comparison to 2018 and 2019.

Fig. 7.

Trend of phone enquiries regarding dental analgesic preparations reported to the National Poisons Information Service over time during 2020 (COVID-19) in comparison to 2018 and 2019.

Fig. 9.

Trend of phone enquiries regarding hand sanitisers in children under 6 reported to the National Poisons Information Service over time during 2020 (COVID-19) in comparison to 2018 and 2019.

In addition, to statistically evaluate the immediate impact of COVID-19 lockdown on enquiries to the NPIS weekly numbers of enquiries received from 23^rd March 2020 to 8^th June 2020 (Lockdown) were compared to weekly numbers received in the twelve weeks prior to lockdown (Pre-Lockdown: 30^th December 2019 to 16^th March 2020) and a comparable control period in 2019 (25^th March 2019 to 10^th June 2019). Due to sample size and expected non-normality statistical testing was performed using the non-parametric Kruskal Wallis test to test differences in the weekly enquiry rate. It was not possible to undertake these tests for cleaning products as the data was not collected for weeks 1 to 8 before the first lockdown due to low enquiry numbers. For the categorical variables relating to circumstance and location the Chi square test of independence was performed. Post-hoc analyses were made with adjustment for multiple testing in each case. For each of the selected categories time trends, location, circumstances and IPCS/EC/ EAPCCT Poisoning Severity Score (PSS)⁸ were analysed. This is a validated four-point grading system for poisoning, ranging from 0 (no signs or symptoms related to toxicity) to 3 (severe or life-threatening toxicity). Fisher’s exact test was used to analyse PSS data. Statistical analysis was performed using R (version X). These tests are illustrated in Figs 2, 4, 6 and 8.

Fig. 2.

Comparison of total phone enquiries to the XXX comparing for the 12 week period following 1^st national lockdown to the 12 weeks preceding lockdown and a control period 2019.

Fig. 4.

Comparison of total phone enquiries in children under 6 yr to the NPIS for the 12 week period following 1^st national lockdown to 12 weeks preceding lockdown and a control period (2019).

Fig. 6.

Comparison of total number of phone enquiries where circumstances were intentional (self harm) to the National Poisons Information Service for the 12 week period following 1^st national lockdown to the 12 weeks preceding lockdown and a control period (2019).

Fig. 8.

Comparison of total phone enquiries for dental analgesia preparations made to the National Poisons Information Service for the 12 week period following 1^st national lockdown to 12 weeks preceding lockdown and a control period (2019).

Results

General trends

The NPIS recorded a total of 38,805 exposure related enquiries in 2020–2021, 38,691 in 2019–2020 and 39,641 in 2018–2019 as shown in Table 1, indicating no change in annual rate of enquiries during the COVID-19 pandemic.

Table 1.

Total enquiry numbers and percentage of enquires by location of poisoning and circumstances.

	Total	Weekly mean (range)	Location (% of total)				Circumstances (% of total)
			Home	Work	School	Other	Intentional	Accidental	Other
2020–2021 2/3/20–28/2/21	38,805	746.3 (473–882)	34,372 (88.6)	671 (1.7)	198 (0.5)	3,564 (9.2)	8,483 (21.9)	19,570 (50.4)	10,752 (27.7)
2019–20 4/3/19–1/3/20	38,691	746.2 (555–866)	33,370 (86.0)	1,084 (2.8)	411 (1.1)	3,939 (10.2)	8,870 (22.9)	19,183 (49.4)	10,751 (27.7)
2018–2019 5/3/18–3/3/19	39,641	762.3 (564–861)	34,111 (86.0)	1,116 (2.8)	408 (1.0)	4,006 (10.1)	8,521 (21.5)	20,076 (50.6)	11,044 (27.9)

There was a reduction over the 6-week period starting w/c 9^th March 2020 (week 9) corresponding with the initial implementation of the COVID-19 action plan. Figure 1 indicates that the most significant reduction occurred in w/c 23^rd March 2020 (week 12), the week of the first UK lockdown before returning to within normal range by w/c 20 April. Other lockdown events did not result in the same reduction effect in 2020.

The median weekly enquiry total for all agents increased during the lockdown period (782 per week compared to 743 in the pre-lockdown period). This increase was not statistically significant (P = 0.741) potentially due to greater weekly variability in weekly enquiry numbers during lockdown (Fig. 2).

Home or domestic is reported to be the most common location for exposures in all years, with other locations consistently contributing to <12% of total enquiries. For the 12-week lockdown period 90.5% of exposures were at home compared to 87.15% in the pre-lockdown period and 85.9% in the control period (P < 0.001). Annual exposures at work and school both showed a potential minor reduction in 2020–2021. Decreases in the proportion of enquiries relating to exposure at work during lockdown compared to pre-lockdown and control periods (Χ2 (2 df) =75.15, P < 0.001) and decreases in enquiries relating to exposures at school in the same reference periods (Χ2 (2 df) =87.72, P < 0.001) were both statistically significant.

Enquiries relating to children <6 yr as a percentage of total enquiries shows an increase in 2020–2021 when compared with 2019–2020 between the first lockdown (week 12) and week 31 after which they returned to expected rates compared with previous years (see Fig. 3). During lockdown 29.9% of enquiries (see Fig. 3) concerned exposures in children <6 yr of age, an increase from 26% in the pre-lockdown period (P = 0.004). This increase is also significant when compared to the control period where 25.8% of enquiries concerned exposures in children <6 yr of age (P = 0.001).

Intentional poisoning

Fig. 5 illustrates the proportion of enquiries about intentional exposures (i.e. self-harm) received by the NPIS throughout the pandemic from Week 1 to week 4 (January 2020) to week March 2021 per 4-week period. In the 4-week period (weeks 8 to 12) just prior to the first lockdown there was a potential spike observed in the proportion of intentional enquiries (27.9%) in 2020. However, during the lockdown period 20.6% of enquiries were intentional exposures, a significant reduction (P < 0.001) from the pre-Lockdown (25.9%) and control (22.6%) periods (see Fig. 6).

Dental analgesics

An analysis of NPIS enquiries involving exposures to paracetamol (and compound preparations), aspirin, other non-steroidal anti-inflammatory drugs (NSAIDS), and codeine-containing preparations were undertaken. These were further narrowed by enquiry records that mentioned preparations that were taken to treat dental pain (UKPID retrospectively interrogated for the terms “dental,” “tooth,” “teeth”).

The median number of enquiries concerning dental analgesia was 41 per week (IQR = 13.5) during lockdown compared to 18 per week in pre-lockdown (IQR = 10.5) and 26.5 per week (IQR = 8) during the Control period. As a percentage of total calls this equates to 5.15% in lockdown, 2.59% in pre-lockdown and 3.42% in the control period (P < 0.001) (Fig. 7). This change appears to remain for the study period following the opening of non-emergency dental practices on the 8^th June 2020 (week 23) to 28^th February 2021 (week 60) as the number of dental analgesic enquiries remain high at (38.2 average per week) compared to 2019 (23.8 per week) and 2018 (20.1 per week).

The majority of enquiries involving dental analgesics resulted in either none or minor symptoms at the time of enquiry (Poisoning Severity Score 0 or 1).⁸ Fisher’s exact test showed no evidence for an association between study period (Lockdown/Pre-Lockdown/Control) and PSS (P = 0.074). The data does not suggest a change in severity of poisoning for cases of dental analgesia in the immediate aftermath of lockdown.

Hand sanitisers

During lockdown a median of 10 enquiries per week (IQR = 5) were received which concerned exposure to hand sanitisers. This compares to a median of 5.5 per week (IQR = 3) during pre-lockdown and a median of 4.5 per week (IQR = 2) during the control period. As a percentage of total enquiries (Fig. 10) there was an increase from 0.56% during control and 0.73% during pre-lockdown to 1.44% during lockdown (P < 0.01) for the comparison between pre-lockdown and lockdown).

Fig. 10.

Comparison of total phone enquiries for hand sanitisers in children under 6 yr made to the National Poisons Information Service for the 12 week period following 1^st national lockdown to 12 weeks preceding lockdown and a control period (2019).

Hand sanitiser enquiries concerning children aged less than 6 yr increased from 32 exposures (median 2 per week) in the control period to 45 (median 4 per week) during pre-lockdown and 92 (median 7 per week) during lockdown (P < 0.001). The initial increase in hand sanitiser enquiries was observed at around the time of launch of the COVID-19 action plan at the beginning of March 2020 (week 9) and remained high throughout the study period (Fig. 9).

The majority of enquiries involving hand sanitisers resulted in exposures with either no or minor symptoms (Poisoning Severity Score 0 or 1) appeared similar when comparing the two review periods (99% vs 97%). The number of patients developing either moderate or severe features of ethanol intoxication from hand sanitiser ingestion increased from 1 to 7 cases (0.4% vs 1%). Two adult males developed severe features of ethanol toxicity following intentional ingestion of large amounts of >70% concentrated ethanol-based hand sanitiser. For enquiries concerning hand sanitisers there was no evidence of association between study period (Lockdown/Pre-Lockdown/Control) and PSS (P = 0.219, Fisher’s exact test.

Cleaning products

Data regarding inhalation exposure to fumes from household bleaches, disinfectants and cleaning agents were analysed from weeks 9 to 33 (2^nd March to 23^rd August 2020) and compared with the corresponding timeframe in 2019 (4^th March to 25^th August 2019). Data demonstrate greater number in 2020 (194; 1.05% of total) compared with the same period in 2019 (164; 0.87% of total) but note that these differences were not statistically significant. Enquiries relating to inhalational exposure to fumes from the processes of mixing two cleaning agents together were also extracted for analysis (Fig. 11). Enquiries relating to exposure to fumes from mixing two cleaning agents increased to 107 (0.58% of total enquiries) in 2020 compared to 91 (0.48% of total enquiries) in 2019 but similarly this increase was not statistically significant.

Fig. 11.

Trend of phone enquiries regarding inhalation of cleaning products reported to the National Poisons Information Service over time during 2020 (COVID-19) in comparison to 2018 and 2019.

Reported poisoning severity score comparing 2020 with 2019 for the same timeframe also demonstrated no clear difference (2020/2019; none 22%/21%; minor 72%/71%, moderate 4%/6%, severe 1%/1% and not applicable or unknown 2%/2%).

Pharmaceuticals

Due to anecdotal UK or international reports enquiry data was retrieved on several pharmaceuticals including hydroxychloroquine, dexamethasone, azithromycin, lopinavir and nebulised interferon from January 2021 to August 2021 and compared to the equivalent period in 2019. There were no potential signals identified for any of these preparations following an analysis of phone enquiry (UKPID) data. In addition, given its toxicity and media reports of widespread international use, a TOXBASE alert was setup whenever a user accessed the drug toxicity information for hydroxychloroquine. Users were asked (via a pop-up box) “If patient related, was your patient exposed to hydroxychloroquine for the prophylaxis/treatment of COVID-19 symptoms?”. Between the 6th July 2020 and the 28^th February 2021 there were 223 TOXBASE enquiries in relation to patients. From these enquiries 7 patients were identified as meeting the criteria listed above.

Miscellaneous agents

We analysed all NPIS cases about methanol reported to NPIS during the 12-month period prior to and after March 2020 shown but saw no statistically significant increase in the number of methanol cases or cases from March 2020 compared to the same 12-mo period from March 2019. There was no notable increase in cases involving illicit alcohol after March 2020. There was a small increase in cases involving the use of a branded disinfectant containing methanol at low concentrations. NPIS enquiries were also monitored to identify cases where a product was used inappropriately to protect against or to treat infection with COVID-19. A small number of cases reported inappropriate use of a range of agents including essential oils, hydrogen peroxide, Chinese herbal medicines, antihistamines, and domestic cleaning agents. However, no significant trends were identified.

Discussion

Introduction

It was reassuring from a public health perspective that, based on enquiries received by the National Poisons Information Service, there was no evidence to indicate that potentially hazardous practices of deliberately ingesting bleach, essential oils or methanol for the treatment or prophylaxis of COVID-19 symptoms was occurring frequently in the UK, in contrast with global reports. For instance, surveillance of international media reports noted a rise in the number of cases of methanol poisonings during the early stages of the COVID-19 pandemic.² These case reports included exposure to counterfeit hand sanitisers and home-made cleaning products and consumption of illicit alcohol that contained high concentrations of methanol. There was some limited evidence based on TOXBASE® database access data that a small number of individuals had taken hydroxychloroquine after the MHRA had advised against its use in the treatment of COVID-19 on the 16th June 2020.⁹ However, there were no evidence of any moderate or severe poisoning from subsequent phone enquiries to the NPIS.

General trends

The NPIS enquiry number trends during the COVID-19 pandemic demonstrate an initial reduction in total enquiry numbers followed by return to normal trends within a few weeks. This was in contrast to poisons control centre data from other countries, for instance Canada that reported a 35% increase in enquiries in March 2020¹⁰ or Jordan (91% increase)¹¹ or the average of 21 European countries (4.5%).¹² This is likely explained by the UK NPIS not being a public access service, so the reduction during the initial implementation of the COVID-19 action plan in the UK resulted in a decrease in enquiries received from NHS 111 and NHS 24 who are the initial point of contact for members of the public. However, they had to adapt their systems to cope with the overwhelming number of enquiries received when they became the main point of public contact for COVID-19 related issues. Similar trends were not observed with later lockdowns presumably because NHS 111 and NHS 24 had increased their capacity by the later phases of the pandemic. Of note, a poisons unit in Egypt reported a total decrease in enquiries of 26% but this included patient attendance at a toxicology clinical treatment unit.

The majority of poisons related exposures are suspected to occur in the home, and this remained the case during the COVID-19 pandemic. There was a notable increase in exposures at home during lockdown periods which coincided with reductions in poisonings occurring in schools and workplaces. The increased percentage of total enquiries identified in children under 6 yr mirrors that found in poisons centres globally.^10,11,13,14 It is thought to be a result of the combination of children being at greater risk outside a structured environment, and due to increased availability of specific products during COVID-19, discussed in further detail below in the hand sanitisers section. An example of increased exposures in the home is that of poisons centre in Egypt that reported a 12.4% increase (as a proportion of total enquiries) in enquiries relating to corrosives and household products.

Intentional poisonings

The initial decrease in proportion (and total numbers) of intentional enquiries during the first lockdown may reflect that the public were reluctant to attend hospitals during the first lockdown period which may have resulted in less self-harm occurring or if self-harm was occurring it was less likely to be reported to a healthcare facility and hence result in an enquiry to the NPIS. A survey of 21 poisons centres (including the UK) across Europe noted a decrease from 20% to 17% in intentional exposures (.¹² A UK study¹⁵ looking at presentations to hospitals of patients who have self-harmed found that during the 12 weeks following lockdown there was a large reduction in the number of self-harm presentations to hospitals by adults compared to the preceding year and was more pronounced in women than men.¹⁵ International data from the COVID-19 Suicide Prevention Research Collaboration (ICSPRC) established during the pandemic showed no increase in risk of suicide during the early months of the pandemic across 21 countries and several countries (mostly medium- to high- income areas) appear to have seen fewer suicides than expected. Potential explanations for this included a reduction in stresses and increased social integration during lockdowns. This is also supported by data during the first and second world wars showing a decrease in suicide rates.¹⁶ Although our data and some of the literature have identified a decrease in self- harm or suicide during the first lockdown it is possible that impact on mental health leading to self-harm or suicide might be delayed. For instance, another UK study in Manchester found that reductions in primary care recorded self-harm persisted to May 2021, though they were less marked than in April 2020 during the first national lockdown. They concluded that the observed reductions could represent longer term reluctance to seek help from health services¹⁷ and could reflect that individuals continue to self-harm at similar rates but do not present to health services. However, our study suggests that poisons centre data on intentional poisonings could potentially be used as a proxy for an increase or decrease in self-harm behaviour in any subsequent significant large scale public health events.

Analgesics for dental pain

Specialists in Poisons Information taking enquiries for the UK National Poisons Information Service noticed a substantial increase in the number of dental enquiries reported throughout the first few weeks of the COVID-19 pandemic. Within weeks of the first lockdown on the the number of dental enquiries had significantly increased which triggered further investigation into the dental enquiry data. Our data show strong evidence for an increase during the first lockdown and enquiry numbers remained higher than previous years data throughout the study period.

In general, dentists were forced to close to routine appointments and reserved for only the most severely affected dental patients. The nature of COVID-19 meant that healthcare professionals performing aerosolised treatments were at particular risk and as the pandemic developed, access to face masks that provided higher protection (e.g. FFP3 and above) became increasingly difficult. In addition to this, with the public fear of catching the virus many people refrained from attending healthcare facilities including dentists, preferring to self-medicate at home. In addition, since the first lockdown in March 2020, until restrictions were lifted slightly during June 2020 the way in which dental surgeries were operating had changed. This included measures of just having one dentist in a room at a time, sterilising rooms and changing PPE between patients and closing waiting rooms. This meant waiting lists for treatments increased in length as dental surgery capacity was reduced.¹⁸ Data for Autumn 2020 is harder to analyse as the devolved administrations introduced lockdowns for different durations, at different times but it is still very clear to see that dental enquiries to the UK NPIS were much higher than previously noted for that time of year. When the UK went into a third full lockdown immediately after Christmas 2020 the trend remained the same with elevated dental enquires being received for excessive therapeutic doses being consumed at home. It should be noted that dental analgesic enquiry numbers were relatively low and only one enquiry in 2020 reported severe symptoms, two reported moderate symptoms, and all others were mild or asymptomatic at the time of enquiry.¹⁹ However, delayed symptoms following analgesic overdose (e.g. paracetamol) are possible and may not necessarily have been captured in these figures due to lack of follow-up data. Following liaison with PHE’s Dental Public Health team and the Chief Dental Officer, PHE disseminated warnings around the use of dental analgesics through avenues such as the NHS England Primary Care Bulletin²⁰ (Fig. 12) and provided links to relevant guidance on pain relief for dental pain.²¹ At that point in the pandemic the increase in dental analgesic enquiries was not a trend that has been reported by other countries and led to PHE informing other poisons centres via the World Health Organization.

Fig. 12.

Examples of public health messages in relation to toxicovigilance.

Hand sanitisers

The use of hand sanitising products containing often large amounts of ethanol by volume (60–95%) is recommended where washing with soap and water is not possible.²² Hand sanitiser bottles are typically small enough to carry in a pocket or handbag, and this can lead to adults assuming these products as being innocuous. However, their often colourful, pleasantly fragranced preparation and small size can capture the attention of children. However, their high ethanol concentration presents a very real toxicological risk in cases of accidental ingestion by children and less commonly rare intentional ingestions in adults. The hazards of hand sanitiser exposures to children have been reported in a study of 647 paediatric exposures over a 7-year period. The median age in this cohort was 2-yr and although only a small number of patients developed symptoms (4%), the potential for ethanol poisoning was recognised.²³

Sales of hand sanitiser products in the United States have been increasing substantially since 2003 with over $70 million worth of hand sanitiser sales in 2006 in the US alone.²³ During the early phase of the COVID-19 pandemic, hand sanitiser sales in the UK increased by 255%.^24,25 Despite the increase in their sales, this is not reflected in an equally sharp increase in the number of enquiries to the NPIS. One reason for this is that the majority of hand sanitiser exposures are likely to be managed by healthcare professionals without necessarily contacting the NPIS by phone, instead consulting TOXBASE. Hand sanitiser enquiries started to increase around the time a public information campaign advising the public to wash hands frequently began in the UK²⁶ and supported by the launch of the COVID-19 action plan.²⁷ Our data also show that hand sanitiser enquiries remained high throughout the pandemic despite easing and stricter lockdown measures.

During the COVID-19 pandemic there has been a substantial increase in hand sanitiser exposures reported to Poison Centres globally,^14,28,29 for example 6% of hand sanitiser exposures reported to the Croatian Poison Control Centre between January 1^st 2020 and the 30^th June 2020 involved children (<18 yr), 91% of whom were less than 6 yr of age.²⁹ As a result of the toxicovigilance data collected by NPIS, several public health actions were instituted. PHE updated handwashing and surface cleaning advice to include advice on the safe storage of hand sanitisers.²⁶ Public health messaging encouraged parents to store hand sanitisers safely and inform them of the hazards associated with their exposure, particularly to young children (Fig. 12).

In addition, as part of the toxicovigilance workstream PHE liaised with the Office for Product Safety Standards (OPSS) to create medical alerts when two hand sanitiser products were found to be contaminated with hazardous concentrations of methanol. Two case reports from Australia illustrate the necessity to disseminate public health advice about the dangers of alcohol-based hand sanitisers to children. Ingestion of 50 mL of a 62% hand sanitiser resulted in a 6-year-old girl being hospitalised and intubated to protect their airway following persistent vomiting, and a 5-year-old girl underwent urgent magnetic resonance imaging and a lumbar puncture under general anaesthesia as part of a differential diagnosis that was eventually confirmed as ethanol toxicity.³⁰ Ingestion of approximately 55 mL of a 70% alcohol-based hand sanitiser in a 3-year-old girl resulted in a blood ethanol concentration of 260 mg/dL. The child developed hypotension and was admitted to paediatric intensive care for ethanol intoxication.²⁸

Other cleaning products

Enhanced event-based surveillance and reports in the literature are noted to have discussed increased reports exposure to cleaning products potentially related to excessive cleaning in home, workplace, and industrial settings.³¹ The risk of household products including cleaning agents to children in the UK particularly during the pandemic has also recently been highlighted.³² The process of mixing two chemicals together is a well-recognised and toxicological issue as the resultant release of chlorine gas can quickly penetrate deep into the lungs causing features of respiratory irritation, including cough, dyspnoea, and in more severe cases leading to pulmonary oedema and acute respiratory distress syndrome (ARDS) especially when exposure is prolonged. Pulmonary injury may be delayed for up to 72 h to develop³³ so there is the potential for the link between cases of respiratory irritation relating to inhalation of chlorine fumes causality to be missed. Phone enquiry data for exposure to fumes from mixing of cleaning agents did not demonstrate the anticipated spike during first lockdown, or over the study period. However, the decrease in total enquiries during this period, and overall low numbers in which to identify patterns are a likely explanation for this. However given the international reports of poisoning, PHE highlighted the hazards of mixing of chemicals (Fig. 12) in the PHE Chief Nurse newsletter,³⁴ the Regional PHE Children and Young People’s newsletter³⁵ and the PHE guidance on cleaning of non-healthcare settings outside the home.³⁶ Numerous international poisons centres (e.g. USA and Italy) dealing with larger enquiry volumes reported significantly increased numbers of chlorine related exposure during the COVID-19 pandemic^37–39 and a European study noted an increase in disinfectants from 1.9 to 5.2% and cleaning products from 4.4 to 5.7%.¹² However, another poisons centre in Italy reported a similar issue to our phone enquiry data in that numbers were too low to report statistical significance.⁴⁰ Case reports were also identified that reported life-threatening toxicity from deliberate exposure to cleaning agents as a treatment for COVID-19,⁴¹ no similar cases were reported to the UK NPIS.

Conclusion

The COVID-19 pandemic influenced phone enquiries to the UK National Poisons Information Service especially in relation to increases in enquiries in children under 6 due to spending more time at home, dental analgesia lack of access to dental services and the increased availability of hand sanitisers during the first lockdown. This provided Public Health England with the evidence to be able to initiate public health actions in relation to tweets, updates to guidance and advice to healthcare professionals and the public. Toxicovigilance is a key public health tool for early identification of toxic exposures in any large-scale chemical incidents or future pandemics.

Limitations

NPIS data cannot reflect the true incidence of poisonings in the UK as advice is not sought on all cases of exposure, but the collated information does provide a good indication of overall trends to identify toxicovigilance “signals,” particularly when the data is compared to that of previous years.

Lockdown measures under the devolved governments of Wales, Scotland, and Northern Ireland deviated from patterns observed in England throughout 2021. Of the four nations, England has the largest population and therefore data from England is more prevalent in overall NPIS data. However, the authors acknowledge that variations in lockdown and reopen dates across the devolved administrations may have had confounding effects on overall NPIS data and therefore the results contained within this study.

There was limited capacity for following up on clinical outcomes in individual cases during the peak of the pandemic due to pressures on frontline health services. Accordingly, the Poisoning Severity Scores reported in this study refers to severity at time of enquiry, which may not reflect the worst severity of these cases.

Author contributions

Nicholas Brooke: Conceptualisation, Project Administration, Investigation, Methodology, Formal Analysis, Writing – Original draft preparation.

Gillian Carter: Investigation, Formal analysis, Software, Methodology, Data Curation, Writing – Review and Editing.

Pardeep Jagpal: Investigation, Formal analysis, Software, Methodology, Data Curation, Writing – Review and Editing.

Meghan Cook: Investigation, Resources, visualisation, Writing – Review and Editing.

Sian Harbon: Investigation, Formal analysis, Software, Methodology, Data Curation.

Steven Jones: Formal analysis, Software, methodology, data curation, validation visualisation.

David Cole: Formal analysis, Software, Data Curation, visualisation.

Sian Harbon: Investigation, Formal analysis, Software, Methodology, Data Curation.

Elaine Donohoe: Investigation, Formal analysis, Software, Methodology, Data Curation, Writing – Review and Editing.

Muhammad Elamin: Supervision, Writing- Review and editing.

Ruben Thanacoody: Supervision, Writing – Review and Editing, Validation.

Raquel Duarte-Davidson: Conceptualisation, Supervision, Writing – Review and Editing, Validation.

Funding

No external sources of funding were used for this study.

Conflicts of interest: There are no conflicts of interest to declare for any of the authors.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.Nicholas.brooke@ukhsa.gov.uk