If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
In December 2020, over 500 residents of Eluru City were hospitalised with seizures and sudden loss of consciousness (LOC) resembling the neurotoxic effects of organochlorine poisoning after a flooding event during the last week of November 2020. We described the epidemiological investigation of outbreak and identified risk factors.
We performed descriptive analysis followed by 1:1 unmatched case-control study. Cases were identified through house-to-house search and review of medical records at district hospital. A case defined as sudden onset LOC or new-onset seizures in an Eluru resident aged ≥1 year, December 1–15, 2020 and a control as absence of neurological symptoms in a person aged ≥1 year selected randomly from same administrative division of the case. We compared cases and controls for possible risk factors and calculated adjusted odds ratio (aOR) with 95% confidence interval (CI). Biological and environmental samples were tested for contaminants.
We identified 545 cases (56% males), including one death. Seizures were reported in 491 (90%) cases. Median age was 27 years (interquartile range: 17–37 years) and 480 (88%) cases resided in urban area. Cases were clustered in administrative divisions supplied by municipal water reservoirs. Cases were more likely than controls to use municipal water as primary source of drinking water (aOR = 4.6, 95% CI = 1.6–13.0). High levels (average: 14.6 mg/l) of organochlorine compounds were detected in all municipal water samples (acceptable limit: <0.001 mg/l).
This investigation highlights water ingestion as an exposure pathway for environmental contaminants (organochlorines) in the community after largescale flooding. We recommended strengthening safe water surveillance in natural disaster response contingency plans in Eluru.
In December 2020, an outbreak of acute onset neurological illness was reported in the Eluru city of Andhra Pradesh, India. Over 500 residents were hospitalised at district hospital (DH), Eluru, with seizures and sudden loss of consciousness (LOC) in the first two weeks of December 2020.
Pesticides are broad range of chemicals used to control insects, weeds, fungi and various plant diseases. Neurotoxic effects of acute high-level exposure to pesticides is well-documented with most studies focussed on organophosphate (OP) insecticides; however some studies have also found the neurotoxic effects of other pesticides like organochlorines, carbamates etc.
OP insecticides have high acute toxicity leading to cholinergic syndrome initially which includes sweating, salivation, profound bronchial secretions, miosis, diarrhoea, tremors etc. and convulsions are uncommon; while organochlorines are moderately toxic leading to dizziness, tremors, vomiting and convulsions occur in severe poisoning.
Andhra Pradesh is situated on the southeast coast of India. Due to its long coastline, the state is highly vulnerable to natural disasters like floods and cyclones. Natural disasters such as floods and cyclones may precipitate pesticide entry into drinking water supply lines because of over loading of the water filtration systems in areas with extensive agricultural or industrial use.
Eluru, in the West Godavari district of Andhra Pradesh, is located in the delta areas of Godavari and Krishna rivers and experiences recurrent floods and drainage problems due to its geographical location.
An early warning signal was generated by the state surveillance unit, Integrated Disease Surveillance Programme (IDSP) on December 5, 2020, regarding clustering of cases with acute neurological illness in Eluru.
In response, a multidisciplinary team from the Ministry of Health and the National Centre for Disease Control (NCDC), including Epidemic Intelligence Service Officers (EISOs) was deployed on December 7, 2020. The objective of the investigation was to describe the epidemiology of the outbreak, identify the aetiology and associated risk factors, and provide evidence-based recommendations to prevent future outbreaks.
2.1 Confirmation of outbreak
We reviewed the registers of the emergency department of the district hospital (DH), Eluru and analysed the weekly trend of seizure cases from September 1–December 12, 2020.
2.2 Case definition and case finding
We defined a case as “sudden onset of LOC or episode of seizure/s in a person residing in West Godavari district, Andhra Pradesh from December 1–15, 2020". For case finding, a house-to-house survey of Eluru, district headquarters of West Godavari district, was conducted. Medical records at DH, Eluru were reviewed to identify all the cases reported from West Godavri district. Cases were interviewed for socio-demographic characteristics, clinical symptoms, details of hospitalisation, and exposure-related information.
2.3 Descriptive study
We described the cases in time, place, and person. For time analysis, we prepared the epidemic curve by date of onset of symptoms. For person analysis, we described the cases by socio-demographic (age, sex, residential area) and clinical profiles (type of symptoms, time of onset of symptoms, hospitalisations, outcome). We interviewed the cases using semi-structured questionnaire to assess non-point and point sources of exposure such as contaminated water, food, air, chemical, and pharmaceutical ingestion. For place analysis, administrative division-wise population attack rates and household attack rates were calculated. To identify epidemiologic linkage, spot maps were prepared along with elevated water reservoirs using ArcGIS online software.
From the descriptive study, we hypothesised an association between the illness outbreak and consuming contaminated municipal drinking water supply in the Eluru City of Andhra Pradesh.
2.4 Case-control study
A 1:1 case-control study was conducted in the district headquarters Eluru of West Godavari district, where most of the cases were reported, to test the hypothesis generated from the descriptive study. A sample size of 103 case-participants and 103 control-participants was calculated using Epi Info version 126.96.36.199 with a 95% confidence level, 80% power, exposure of 65% among controls,
Based on the descriptive study, a different case definition was formulated for the case-control study. A case-participant was defined as a person with sudden onset LOC or new-onset seizures from December 1–15, 2020, residing in Eluru City aged ≥1 year. A control-participant was defined as the absence of LOC or new-onset seizures from December 1–15, 2020, in a resident of Eluru City aged ≥1 year. Case-participants and control-participants with a past history of seizures were excluded. Cases were selected randomly from the case line list before visiting the case household. For each case household, a control household was selected randomly from the household enumeration list from the same administrative division. From the selected household, one control was selected, using the nearest birth date of any year to the date of visit, and interviewed. In an administrative division, households of cases and controls were approached and interviewed using semi-structured questionnaire at the same time of the day to avoid any selection bias related to the field team's time of visit or the availability of working persons at home. The semi-structured questionnaire enquired about their sociodemographic profile, clinical history (past and present), history of substance abuse, household water supply, its storage, treatment and recent change in the quality of drinking water.
2.5 Data analysis
For data collection, we used Epicollect5 software. Data were compiled and analysed using Microsoft Excel and Epi Info version 188.8.131.52. We calculated proportions and attack rates. Continuous variables were dichotomised at the median. Chi-square test and Fisher's exact test was used as a test of significance, and the odds ratio (OR) was calculated with a 95% confidence interval (CI). We conducted multivariate logistic regression analysis to calculate adjusted odds ratio (aOR) adjusted for risk factors that were significant at p<0.05 in bivariate analysis. Age was included in the model as a potential confounder for face validity. All independent variables were assessed for multicollinearity using a correlation coefficient matrix before inclusion in the model.
2.6 Laboratory and environmental investigation
As part of a broader investigation by a multi-disciplinary team of experts that was constituted by the federal government; clinical specimens (serum, urine, and stool) from the hospitalised patients in the postictal state and water samples from the water reservoirs were tested for biological and chemical contamination at the Indian Council of Medical Research (ICMR) National Institute of Virology (NIV), Pune and Anoor Envirochem and Food Analytical Laboratory (AEFAL), Vijayawada respectively. Four administrative divisions of Eluru (two each from Town 1 and Town 2) reporting majority of the cases were selected for water testing for chemical contamination.
The municipal corporation water treatment plant at Eluru was inspected, and key informant interviews were conducted with public health officials from water and sanitation departments of the municipal corporation to understand the details of the water supply system, quality monitoring, sanitation and use of insecticides in vector control activity.
2.7 Ethical considerations
This investigation was undertaken as part of an emergency public health response to identify the cause of an outbreak of neurological illness for early intervention. All statutory permissions were obtained from NCDC and Integrated Disease Surveillance Programme. The investigation was conducted consistent with applicable state and central government law (Epidemic Diseases Act no.3, 1897). Strict data protection protocols reviewed by NCDC were followed while collecting information from cases and controls.
3.1 Descriptive epidemiology
We observed an increase in the reported seizure cases in the emergency department of DH, Eluru in the first week of December 2020, compared to the previous weekly threshold of 1–2 cases per week from 1 September−November 22, 2020. During December 1–15, 2020, a total of 545 cases (56% males) with neurological illness were reported, with one death (case fatality rate: 0.2%). The maximum cases (182 [33%]) were observed on December 5, 2020, and the last case was reported on December 12, 2020 (Fig. 1).
The deceased person was a 45-year-old male with no past medical history and no history of any substance abuse. By occupation, he was an electrician and driver. He had multiple episodes of generalised tonic-clonic seizures on December 6, 2020, and was given anti-convulsant treatment at DH, Eluru. The patient suffered status epilepticus leading to cardiopulmonary arrest and despite resuscitation efforts died that same evening.
The median age of cases was 27 years (interquartile range: 17–37 years) with 26.5% (143) of cases in the age group 11–20 years and 23% (122) in the age group 31–40 years. Seizures were present in 90% (491) of cases, while 10% (545) of cases had LOC. Other symptoms reported include loss of bladder/bowel control 3% (14), frothing 2% (9), vomiting 1.5% (8) and fever 0.5% (3) (Table 1).
Table 1Demographic, symptomatic, and epidemiological description of cases during an acute neurological illness outbreak, West Godavari District, Andhra Pradesh, India, December 1–15, 2020 (n = 545).
Of 545 cases, data were available for 529 cases. Of 529 cases, 96% (506) had sought treatment, and 90% (310/345) of cases had recovered at the time of interview. Among the 79% (428) hospitalised cases, 33% (142) cases were discharged on the same day and 39% (165) cases on the next day of admission (Table 1).
About 88% (480) of cases were residents of urban Eluru, and 12% (65) resided in surrounding suburban areas. Of the 62 administrative divisions of Eluru City, 95% (59) reported cases. The overall household attack rate was 0.9% (464/53,839), with 1.3% (241/18,062) in town 1 and 0.6% (223/35,777) in town 2. The highest household attack rate was 6.4% (21/330) in a residential community A in town 2. The overall population attack rate was 0.3% (478/1,56,714) and the highest population attack rate was 1.8% (20/1117) in the residential community B in town 2.
Municipal water was the source of drinking water in 97% (526/543) of case households, and 44% (231/520) of cases were drinking water without any treatment. Ten days prior to illness, 28% (149/523) of cases reported greenish-yellow discolouration and foul-smelling municipal water supply in their household. In addition to reports about poor water quality, 7% (34/493) reported taking allopathic or ayurvedic medications for other pre-existing illnesses, 3% (14/516) reported a travel history to another town/village, 1% (7/506) had attended a social gathering like a family function or school, and 1% (6/523) complained of foul-smelling air/fumes.
3.2 Case-control study
As part of the 1:1 case-control study, 103 case-participants and 103 control-participants were enrolled. Both groups were comparable by sex, family size, monthly household income, and employment status. However, controls were slightly older than the cases (p < 0.05) (Annexure 1).
Using a multiple logistic regression model, consumption of only municipal water in the household (aOR = 4.6, 95% CI = 1.6–13.0) was associated with acute onset neurological illness after adjusting for potential confounders and risk factors (Table 2). In a subset analysis among 93 case-participants and 68 control-participants who were drinking municipal water only, boiling water before drinking (OR = 0.3, 95% CI = 0.2–0.7) or use of RO (reverse osmosis) water at household level (OR = 0.1, 95% CI = 0.01–0.5) was associated with lower odds of illness.
Table 2Environmental factors associated with acute onset of neurological illness, case-control study, Eluru, Andhra Pradesh, December 2020.
Among the clinical samples sent to NIV, Pune and AEFAL, 13% (4/30) of serum samples showed an intermediate level of Japanese Encephalitis (JE) IgM antibodies, and 17% (5/30) of samples were positive for both dengue IgM and chikungunya IgM. Of the 30 nasal swabs, one was positive for Influenza type A. All cerebrospinal fluid samples (5/5), urine samples (11/11) and stool samples (15/15) were negative for any viral infection.
3.4 Environmental investigation
The main source of drinking water is the Godavari River. The water from this river is stored in a summer water storage tank which supply raw water to the Eluru municipal corporation water treatment plant. After treatment of raw water at the Eluru municipal corporation water treatment plant, the treated water is transported to 26 elevated water reservoirs through 9 pumping stations. From these water reservoirs, all the administrative divisions of the city receive drinking water between 0630–0730 h and 1700–1830 h, except administrative division number 34 (supplied by borehole well water). The time of onset of symptoms peaked around morning (0700–0900 h) and night-time (1900–2100 h), coinciding with the timings for municipal water supply. All cases were reported in the administrative division that received water from the municipal water reservoirs; administrative division number 34 which was supplied by borehole well water reported no cases (Fig. 2).
No records of routine drinking water quality tests were available from the water treatment plant; the only records available were of the electric voltage of pumps and level of residual chlorine in drinking water before distribution.
Water samples collected and tested by AEFAL showed the presence of a high level of organochlorine pesticides (dichlorodiphenyltrichloroethane, dichlorodiphenyldichloroethylene, and dichlorodiphenyldichloroethane) in all (4/4) water samples collected from the water reservoirs in Eluru (Fig. 2). The average level of organochlorine pesticides detected was 14.6 mg/L (acceptable limit <0.001 mg/L). Water samples from the two locations of Town 2 also showed the presence of the herbicide Alachlor, with an average level of 10.89 mg/L (acceptable limit: <0.001 mg/L) (Fig. 3).
We describe a common source, short exposure outbreak of self-limiting acute neurological illness associated with drinking municipal water, potentially contaminated with pesticides after the Nivar cyclone. Illness was generally self-limiting as three-fourth of the cases were discharged within two days, and the majority completely recovered without any residual symptoms.
Most cases were concentrated in the urban area; however, some rural areas nearby were also affected. The outbreak affected all the administrative divisions of Eluru City supplied by municipal water except three, one of which was supplied by underground bore water. Time distribution of cases suggested a short exposure and a common source origin of the agent.
The West Godavari district of Andhra Pradesh is an agrarian district where agriculture contribute 40% of gross district domestic product. The major products cultivated in the district include paddy, sugarcane, banana and coconut.
In terms of pesticide consumption, the two Indian neighbouring states of Andhra Pradesh and Telangana contribute to 24% of pesticide usage in the country with around 13,600 pesticide dealers in these two states. Pesticides are the substances used in various agronomic practices to control insects, pests, other plant diseases but their excessive and irrational use can lead to various health hazards like cancers, reproductive problems, nervous system disorders etc.
In India, insecticides are the most commonly used pesticides followed by herbicides and then fungicides and bactericides. Among insecticides, OPs are most commonly used followed by pyrethroids & neonicotinoid and then others like organochlorines, carbamates, etc.
The water samples collected during this investigation detected very high levels of organochlorine (OC) pesticides from water reservoirs. A separate toxicological investigation was conducted by India's National Institute of Nutrition (NIN) on biological specimens and water samples. They detected organophosphates (OP) in 92% (12/13) of water samples from the households of cases.
Another study conducted in non-outbreak setting in Hyderabad, India (state capital of Telangana), reported detection of OP pesticides in various water sources like drinking water from municipal corporation, underground water, river water etc.
The reason might be the contamination of rivers, lakes, water reservoirs and underground water due to the surface water runoff from agriculture fields and soil contaminated with pesticides.
The clinical profile of the cases correlates more with acute OC poisoning. Cases reported seizure as the predominant symptom, followed by rapid recovery. Acute OC pesticides most commonly stimulate the central nervous system, resulting in rapid self-limiting neurological symptoms (seizures, tremors, paraesthesia, etc.) within 1–6 h post-exposure and rapid recovery without any further evidence of toxicity. In comparison, acute OP poisoning is more fatal and it initially results in muscarinic symptoms (salivation, lacrimation, rhinorrhoea, sweating, and vomiting), followed by nicotinic symptoms (neuromuscular paralysis); central symptoms like seizures are uncommon.
The case-control study demonstrated that the odds of illness was nearly five times higher among those who were drinking water from municipal water supply as compared to those who were drinking bottled RO water. This finding strongly suggest that after Nivar cyclone, municipal water supply got contaminated with OC pesticides either at the source or in the supply chain as more than one-fourth of cases reported greenish-yellow discoloured and foul smelling municipal drinking water after heavy rains and flooding.
In our study, we could not ascertain the point of contamination which required a more speedy and comprehensive investigation with intersectoral coordination. In the immediate aftermath of the event, medical care of cases was prioritized along with public health actions like provision of alternate and safe drinking water. This highlights the critical need of health department readiness to deal with consequences of chemical releases following natural hazard events and disasters. Natural disasters such as heavy rainfall, flooding, cyclone, and other extreme events can lead to chemical releases in the environment by either directly damaging the infrastructure or through toxic runoff from inundated areas. These released chemicals can then mix with drinking water to cause adverse health impacts at the population level.
There was no significant finding on microbiological investigations in our study that could explain the clinical-epidemiological presentation. Environmental investigation showed that the water treatment plant lacked a dedicated water testing laboratory and had no available records on the use of insecticides.
This investigation highlights that in the aftermath of largescale flooding, water ingestion can serve as an acute exposure pathway for environmental pesticides in the community. This reinforces the need for strengthened safe water surveillance, provision of alternative water supply and community education on safe water storage in natural/man-made disaster response contingency plans when chemical and biological contamination is more likely to occur.
Source(s) of support
This public health activity was conducted by India Epidemic Intelligence Service (EIS) program of the National Centre for Disease Control (NCDC). The NCDC receives funding support for the India EIS Program through a cooperative agreement with the U.S. Centers for Disease Control and Prevention, Center for Global Health, Division of Global Health Protection. ICMR-NIV and other agencies involved in the investigation are public institutes supported by government funds. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This investigation did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Conflicting interest (if present, give more details)
The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the U.S. Centers for Disease Control and Prevention
We acknowledge Department of Health and Family Welfare, Andhra Pradesh for their cooperation and district health teams for their participation in the data collection process for this outbreak investigation.
Table 1Socio-demographic characteristics of cases and controls, case-control study, Eluru city, Andhra Pradesh, December 2020
No. of cases N = 103 (%)
No. of controls N = 103 (%)
Age, Median (range)
35 (1.5–66) years
37 (3–75) years
*Monthly household incom Median (range)
Rs. 10000 (1000–75,000)
Rs. 10,000 (1000–2,00,000)
#For occupation, denominator was employed persons i.e. 32 cases and 39 controls.
*For monthly household income, data is available for 96 cases and 100 controls.
Drinking Water, Sanitation, Hygiene and Housing Condition. NSS 76th Round. National Statistcal Office, Government of India,
July 2018-December 2018 (The Hindu Center. [cited 2021 Aug 26]. Available from:)