Journal of Infection and Public Health 17 (2024) 102568

Contents lists available at ScienceDirect

Journal of Infection and Public Health

journal homepage: www.elsevier.com/locate/jiph

Original Article

Norovirus outbreaks due to contaminated drinking water and probable

person-to-person transmission, Kerala, India, 2021 ]]
]]]]]]
]]

a,1 b,1 c d
Amjith Rajeevan , Manikandanesan Sakthivel , Nikhilesh Menon , Sachin KC ,
Harisree Sudersanan e, Ramya Nagarajan f, Mohankumar Raju b, Sharan Murali f,
Chethrapilly Purushothaman Girish Kumar g, Anukumar Balakrishnan h, Renuka Raveendran i,
Dineesh Perumbil j, Devaki Antherjanam k, Sherin Joseph Xavier Kallupurackal l,
⁎
Bipin Balakrishnan m, Nandu Krishna n, Sibin Samuel o, Prabhdeep Kaur g, ,
Manoj Vasant Murehkar p
a
District Nodal officer Ardram mission, Pathanamthitta, Health services, Kerala, India
b
Consultant, ICMR-National Institute of Epidemiology, Chennai, Tamil Nadu, India
c
State Nodal Officer Lab Network, Health services, Kerala, India
d
District Nodal officer Ardram mission, Kannur, Health services, Kerala, India
e
Assistant Professor, Microbiology, Govt TD Medical college, Alappuzha, Kerala, India
f
Scientist-B, ICMR-National Institute of Epidemiology, Chennai, Tamil Nadu, India
g
Scientist-F, ICMR-National Institute of Epidemiology, Chennai, Tamil Nadu, India
h
Scientist -E, ICMR-National Institute of Virology, Kerala, India
i
District Medical officer, Wayanad, Health services, Kerala, India
j
District surveillance officer, Wayanad, Health services, Kerala, India
k
Senior consultant, NHSRC, Delhi, India
l
Superintendent, Taluk hospital, Vythiri, Wayand, Kerala, India
m
Epidemiologist Wayanad, National Health Mission, Wayanad, Kerala, India
n
Medical officer, NHM, Wayanad, Kerala, India
o
Epidemiologist, NHM, Wayanad, Kerala, India
p
Scientist-G & Director, ICMR-National Institute of Epidemiology, Chennai, Tamil Nadu, India

a r t i cl e i nfo a bstr ac t

Article history: Background: In July 2021, the Alappuzha district in Kerala, India, reported an unexpected number of acute gastro­
Received 3 June 2024 enteritis (772) cases (Outbreak A). On October 10, 2021, a university in Wayanad, Kerala, reported 25 acute gastro­
Received in revised form 8 October 2024 enteritis cases (Outbreak B). We described both the outbreaks and determined the agent, source and risk factors.
Accepted 15 October 2024
Methods: We defined a suspected case as the occurrence of vomiting or at least three episodes of loose
stools within 24 h and a confirmed case as those with stool samples/rectal swabs positive for norovirus. We
Keywords:
did a matched case-control study in Outbreak A and a retrospective cohort study in Outbreak B. We cal­
Disease Outbreaks
Disease Transmission culated the adjusted odds ratio (aOR) in outbreak A, relative risk (aRR) in outbreak B and population at­
Groundwater tributable fraction (PAF). We tested stool and water samples for bacteria and viruses.
India Results: We identified Group II norovirus in stool samples in both outbreaks and 4/5 water samples in
Norovirus Outbreak A. Suspected norovirus infection was associated with drinking inadequately boiled water from the
municipal water supply in outbreak A [aOR: 4.5; 95 % C.I: 1.2–15.8; PAF: 0.23] and well water in hostels in
outbreak B [aRR: 2.2; 95 % C.I: 1.2–3.9; PAF: 0.15]. In Outbreak A, groundwater from tube wells was mixed in
the municipal water supply overhead tanks without chlorination.
Conclusion: The gastroenteritis outbreaks were caused by Group II norovirus due to the consumption of
inadequately boiled contaminated groundwater (outbreak A) and well water (outbreak B). We re­
commended superchlorination of overhead tanks and wells and boiled water for drinking.
© 2024 The Authors. Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health
Sciences. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/li­
censes/by-nc-nd/4.0/).

⁎
Correspondence to: R127, second main road, Tamil Nadu Housing Board, Ayapakkam, Chennai, Tamil Nadu 600077, India.
E-mail address: [email protected] (P. Kaur).
1
Contributed equally to this work and shared first authorship

https://doi.org/10.1016/j.jiph.2024.102568
1876-0341/© 2024 The Authors. Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

Background between October 1 and November 15, 2021 (outbreak B). We defined
a confirmed case as a suspected case who tested positive for nor­
Norovirus causes 685 million cases yearly worldwide and is ovirus in a stool sample or rectal swab using reverse transcription
considered the most common cause of acute gastroenteritis (AGE) polymerase chain reaction (RT-PCR).
[1]. Norovirus infection remains a problem in high-middle- and low-
income settings [2]. However, deaths due to norovirus happen pri­
Case search
marily in low-and-middle-income countries. The burden of nor­
ovirus infection in India remains unclear due to a lack of
Outbreak A: We stimulated the ongoing passive surveillance at
surveillance. A community-based cohort study among children
the public health facilities in the district by requesting them to re­
below three years in Southern India estimated that norovirus ac­
port all case patients meeting the definition of a suspected case.
counted for 40 % of diarrhoeal illnesses with an incidence of 14 per
Outbreak B: We actively searched for suspected case-patients
100 child years. [3]. In the hospital-based studies from India, nor­
among the students and staff of the university using a self-ad­
ovirus detection rate among under-five children with AGE ranged
ministered online questionnaire. We also circulated the phone
from 6 % to 25 % [4–7]. Norovirus was the second most common
number of DSU and motivated the students to report if they devel­
pathogen identified, next to rotavirus [3–5]. A majority (70–85 %) of
oped symptoms.
norovirus-associated diarrhoeal episodes in children are caused by
group II (GII) norovirus in India [3,4,8].
Norovirus is highly transmissible, making it the leading cause of Descriptive epidemiology
AGE outbreaks worldwide [1,2]. Humans are the only reservoir. The
transmission can be foodborne, waterborne, direct person-to- We analysed the distribution of suspected case patients in both
person, or through fomites [2]. Most Norovirus outbreaks from high- outbreaks by time, place and person.
income countries are attributed to person-to-person transmission Outbreak A: We drew an epidemic curve with the reporting date.
followed by food contamination [2,9]. However, the risk of norovirus We calculated the median [interquartile range (IQR)] age in years
outbreaks may differ in low-and-middle-income countries with poor and the attack rate by the local self-government areas (village pan­
sanitation and hygiene conditions. In India, one norovirus outbreak chayat/municipality and wards within the municipality) and by
has been documented in the literature to date, which was due to gender.
food contamination [10]. Outbreak B: We drew an epidemic curve with the date of
We report the investigation of two acute gastroenteritis out­ symptom onset. We calculated the median (IQR) age in years and the
breaks due to norovirus in southern India conducted by the officers attack rate by hostels and gender. We described the proportions for
of India-Epidemic Intelligence Service (EIS), a fellowship program in various symptoms reported by probable case patients. We also cal­
field epidemiology. culated the secondary attack rate among the household members of
The first outbreak (Outbreak A) was in Alappuzha, a coastal the students with suspected norovirus infection when they went
district in Kerala, a southern state in India, in June-July 2021. The home during a common holiday. We collected information on the
district surveillance unit (DSU) received reports of increased AGE number of total and symptomatic household members through the
among children from public health facilities. The second outbreak online questionnaire.
(Outbreak B) was in Wayanad, a hilly district in Kerala, in October In both outbreaks, we conducted key informant interviews to
2021. The DSU of Wayanad district received reports of a sudden in­ understand the sequence of events, the water distribution system,
crease in AGE within a university campus (university X) from the and potential risk factors that could aid hypothesis generation.
local primary health centre.
Analytical epidemiology
Methods
We tested the hypothesis generated through the descriptive
epidemiology using analytical studies during both outbreaks. In both
Outbreak settings
outbreaks, we defined inadequately boiled water as unboiled or
mixed with unboiled water after boiling.
Alappuzha municipality, the site of outbreak A, is the major
urban area within the Alappuzha district, with small peri-urban
towns surrounding it. The municipality area spans about 65 km2 Outbreak A
with a population of about 250,000. Despite being a coastal area, it is We conducted a 1:1 age and gender-matched case-control study
generally prone to water scarcity. in two wards with the highest attack rate in Alappuzha municipality.
Outbreak B occurred in a university with 500 students and 100 We defined a case as a suspected norovirus case patient from one of
staff in the Wayanad district. Most students stayed in hostels. There the two wards and control as a resident of either of the wards
were two hostels within the campus (one each for boys and girls) without vomiting and diarrhoea. We selected controls from the
and two hostels (one each for boys and girls) outside the campus. nearest house of the case patients after matching for gender and age
Except for the girl’s hostel outside the campus, the other three within a five-year interval. We calculated the sample size for the
hostels had individual canteens and kitchens. The students from this matched case-control study using the “epiR” package in R software
girl’s hostel had their food from the canteen of the girl’s hostel lo­ [11]. With an expected proportion of controls drinking inadequately
cated within the campus. There were separate residential quarters purified water as 40 % and an odds ratio of 3.0, we estimated the
for the staff within the campus. The residents of the hostels usually sample size as 120 (60 cases and 60 controls) with a 95 % confidence
seek care at the local primary health centre for any minor illness. level and 80 % power.
We interviewed the cases and controls in their houses using a
Case definition semi-structured questionnaire. We conducted univariate conditional
logistic regression and calculated the matched odds ratio (MOR). We
We defined a suspected case as the occurrence of at least one selected the risk factors with a p-value < 0.2 in the univariate ana­
episode of vomiting or at least three episodes of loose stools within lysis for the multivariable conditional logistic regression and calcu­
24 h in a resident of Alappuzha between June 23 and July 25, 2021 lated the adjusted matched odds ratio (aOR) with a 95 % confidence
(outbreak A) or among students or staff of University X, Wayanad, interval.

2
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

Outbreak B Table 1
We conducted a retrospective cohort study among all the stu­ Attack rate of norovirus infection by gender and the age distribution during norovirus
outbreaks, Kerala, India, 2021.
dents in the university. We defined a case as a suspected norovirus
case-patient. We did not calculate sample size as we included all the Outbreak A (in Alappuzha municipality)
students for the study. We then collected the data with a self-ad­ Characteristic n N AR* per 1000
ministered online questionnaire using Google Forms, circulated Age in years, median (IQR) 7 (3 −22)
through the university administration and student leaders. Gender
Female 324 103574 3
We did univariate binary logistic regression to calculate the un­
Male 289 112205 3
adjusted risk ratio (RR). We selected the risk factors with a p-value Total 613 215780 3
< 0.2 in the univariate analysis for the multivariable analysis to Outbreak B (in University X)
calculate the adjusted risk ratio (aRR) with a 95 % confidence in­ Characteristic n N AR* per 1000
terval. Age in years, median (IQR) 21 (18 −23)
Gender
We used STATA software version 17 for the analysis [12]. In both Female 35 217 161
outbreaks, we calculated the population-attributable fraction (PAF) Male 4 130 31
for the significant risk factors (p-value < 0.05) in the multivariable Total 39 347 112
analysis. *AR - Attack rate
IQR – Interquartile range

Environmental investigation
superchlorination was done in all the public overhead water tanks in
We inspected the water supply system and measured the re­ Alappuzha municipality. There were no reports of suspected case-
sidual chlorine level using the ortho toluidine test and chloroscope patients after July 21, 2021.
both at the supply and the user end during both outbreaks. The attack rate of the suspected cases was highest in the
Alappuzha municipality (4 per 1000 population), followed by the
Outbreak A surrounding peri-urban areas (Fig. 2). Within Alappuzha munici­
We sent drinking water to look for faecal contamination and to pality, there were eight public overhead water tanks (Fig. 2). The
test for norovirus. attack rate was highest in the areas supplied by overhead tank B (766
per 100 population) and tank E (308 per 1000).
Outbreak B Key informant interviews revealed that the primary water
We collected water samples from boys’ and girls’ hostels and sources for domestic use are the public water supply system, private
tested them for bacterial contamination. However, we could not test reverse osmosis (RO) plants and the wells within individual houses.
the water samples for norovirus due to the lack of availability of a The public water supply system of Alappuzha municipality and the
sample transportation facility to an advanced viral laboratory. surrounding eight peri-urban areas received water from a water
treatment plant (WTP). The WTP was located outside the munici­
Laboratory investigation pality and received water from a nearby river. The WTP treated the
water and pumped it into a common canal (public water supply
We sent stool samples (12 in Outbreak A and six in Outbreak B) to system), which ran across six peri-urban areas before reaching the
the nearest public health laboratories for bacteriological analysis. municipality area. Most households used the public water supply for
We also sent stool samples and rectal swabs (nine in outbreak A and drinking water. The engineers in the municipality indicated that the
eight in outbreak B) to a research institute to test for viral pathogens septic tanks in most households were not lined by bricks when
using real-time reverse transcription polymerase chain reaction (RT- constructed, making them unprotected from seepage of water into
PCR). Samples were tested for Shigella, Salmonella and Vibrio the ground. Also, in June 2021, there was a food festival wherein a
Cholera under bacteriological analysis and for rotavirus, norovirus, locally acclaimed food item named chicken biriyani, prepared using
astrovirus and sapovirus under virological analysis. rice and chicken, was sold at the doorstep to raise funds for the
RNA was extracted from 20 % faecal suspension in Minimal disaster relief fund. Hence, water contamination and food (chicken
Essential Medium (Sigma) using QIAamp Viral RNA mini kit biryani) were the possible sources of infection.
(Qiagen). Water samples were concentrated as per the protocol of
Qiu Y et al., with slight modifications [13] and RNA extraction was
done using the Trizol method [14]. Screening for Norovirus was Outbreak B
performed using genotype-specific real-time reverse transcription We identified 39 suspected case-patients among 347 students
PCR (RT-PCR) [15]. who responded to the online questionnaire, with an attack rate of
112 per 1000 students. The median (IQR) age of the suspected case-
Results patients was 21 (18−23) years (Table 2). The attack rate was higher
among girls than boys (161 vs 31 per 1000). The most common
Descriptive epidemiology symptoms were vomiting (67 %), followed by loose stools (64 %),
malaise (23 %) and abdominal pain (20 %). Of the 156 household
Outbreak A contacts of students, 91 (58 %) developed symptoms.
In total, 772 suspected case-patients of norovirus were reported, The index case was reported on October 4, 2021. Cases continued
with an attack rate of 3 per 1000 population. Most (n = 613, 79 %) of to occur intermittently till October 23. The university was closed on
these cases were from Alappuzha municipality. The median (IQR) October 25 due to the outbreak. The suspected case-patients were
age of the suspected case-patients was 7 (3−22) years (Table 2). The reported again after the university reopened on November 1 and
attack rate was similar among males and females (2 per 1000). peaked with five cases on November 4. The university administra­
The number of suspected case-patients started increasing on tion super-chlorinated the wells on November 5. No cases were re­
June 26, 2021, following an interruption in the treated water supply ported after November 11, 2021.
after a rain spell (Fig. 1). The number reached its maximum peak on The attack rate was highest in the two girls’ hostels [within
June 28 with 83 case-patients. Cases continued with multiple campus – 16/124 (13 %), outside campus – 18/78 (23 %)], followed by
smaller peaks on July 7, July 15 and July 20. On July 7, the boys’ hostel within the campus [4/90 (4 %)]. No case-patients

3
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

Table 2
Factors associated with probable norovirus infection during norovirus outbreaks, Kerala, India, 2021.

Outbreak A, Community - Age and gender-matched case-control study

Risk factor Cases (%) Controls (%) MOR* 95 % C.I p value Adjusted OR 95 % C.I PAF# 95 % C.I
(N = 60), n (N = 60), n

Drinking inadequately boiled water 18 (30) 7 (12) 4.6 1.3 −16.2 0.016 4.5 1.2 −15.8 0.23 0.14 −0.31
Had restaurant food within 4 days prior 13 (22) 6 (10) 3.3 0.9 −12.1 0.070 3.1 0.8 −11.9
to symptom onset
Wash hands without soap before eating 56 (93) 57 (95) 1.5 0.3 −8.9 0.657
Participated in biriyani festival 2 (3) 3 (5) 0.7 0.1 −3.9 0.657
#
*MOR – Matched Odds Ratio; PAF – Population Attributable Fraction
Outbreak B, University - Retrospective cohort study
Risk factor N n % RR† 95 % C.I P value Adjusted RR 95 % CI PAF# 95 % CI
Roommate with symptoms Yes 84 26 31 6.3 3.4 −11.6 < 0.001 3.4 1.6 −6.9 0.47 0.17 −0.66
No 263 13 5
Drinking inadequately boiled water Yes 41 10 24 2.6 1.4 −4.9 0.004 2.2 1.2 −3.9 0.15 0.02 −0.26
No 306 29 9
Washing hands before eating Without soap 201 31 15 2.8 1.3 −6.3 0.010 1.6 0.7 −3.8
With soap 129 7 5
Staying in either of the girls’ hostel Yes 317 38 12 3.6 0.5 −25.2 0.198 1.2 0.1 −11.2
No 30 1 3
Eating food from hostel mess Yes 261 29 11 0.9 0.5 −1.8 0.841
No 86 10 12

†RR - Risk Ratio; #PAF – Population Attributable Fraction Variables included - age, gender, drinking inadequately boiled water, roommate with symptoms, washing hands before
eating, staying in either of the ladies hostel

were from the boys’ hostel outside the campus and the staff quar­ Environmental and laboratory investigation
ters (Fig. 2).
Key informant interviews revealed that the university campus Outbreak A
had two wells, one common for boys’ and girls’ hostels and one for While none of the stool samples showed growth in bacter­
staff quarters. The hostels located outside the campus had individual iological analysis, six of the nine samples were positive for Group II
wells. The well water was the primary source of drinking water in all norovirus. The residual chlorine level was less than the re­
the hostels and quarters. The canteen in the boys’ hostel usually commended level of 0.2 PPM in all the public overhead tank stations
provided boiled water for drinking, unlike the two girls’ hostels. within the municipality. Two (tank B and tank E) out of 13 samples
Hence, we hypothesised that drinking inadequately boiled water from public overhead tanks in Alappuzha municipality showed a
was possibly associated with the suspected case-patients. high coliform count. Four (water from overhead tanks B and E, tube
well water at tank B station and household of one suspected case
patient) out of five water samples were positive for Group II nor­
Analytical epidemiology ovirus, establishing the groundwater contamination.
There was a heavy rain spell in Alappuzha between June 23–25.
Outbreak A On June 25, there was an interruption in the water supply from the
We interviewed 60 suspected cases and 60 age and gender- WTP due to a breakage in the pipe carrying water from the river to
matched matched controls. Drinking inadequately boiled water was the WTP. Hence, the pump operators within Alappuzha municipality
associated [MOR: 4.6, p < 0.05] with the suspected infection in used unchlorinated water from tube wells available at every over­
univariate analysis (Table 2). In multivariable analysis, drinking in­ head tank station to maintain a continuous water supply.
adequately boiled water was associated with the suspected infection Thus, heavy rain in Alappuzha just before the start of the out­
[aOR – 4.5; 95 % C.I: 1.2–15.8]. About 23 % [95 % C.I: 14 %−31 %] of the break would have increased the groundwater level, and the seepage
suspected case-patients were attributed to drinking inadequately of wastewater from the unbricked septic tanks could have con­
boiled water. taminated the groundwater. A breakage in the pipeline, which car­
ried water to the WTP, led to a drop in water supply from the WTP.
To maintain a continuous water supply, the pump operators pumped
Outbreak B the contaminated groundwater from the tube wells into the public
In the retrospective cohort study, 347/500 (69 %) students re­ (municipal) water supply system without chlorination, possibly
sponded to the online questionnaire. Those who drank inadequately leading to the outbreak.
boiled water had a 2.6 times higher risk of infection in the univariate
analysis [RR: 2.6, p < 0.05]. In addition, students who washed their Outbreak B
hands without soap [RR – 2.8, p-value: 0.010] and who had a Six of the eight stool samples tested positive for Group II nor­
roommate with symptoms [RR – 6.3, p-value < 0.001] had a higher ovirus, while none of them showed bacteriological growth. The re­
risk of developing infection (Table 2). In multivariate analysis, sidual chlorine level in the water samples from all the hostels was
drinking inadequately boiled water was associated with the sus­ less than 0.2 PPM. The water sample from the two girls’ hostel
pected infection [aRR – 2.2; 95 % C.I: 1.2–3.9]. In addition, having a showed the growth of faecal streptococci and had an unsatisfactory
roommate with symptoms [aRR – 3.4; 95 % C.I: 1.6–6.9] was asso­ presumptive coliform count.
ciated with the norovirus infection. While about 15 % (95 % C.I: There was a water stream running down the hill close to the two
3 %−27 %) of the suspected case-patients were attributed to drinking wells which supply water to the boys’ and girls’ hostels within the
inadequately boiled water, 47 % (95 % C.I: 15 %−64 %) were due to stay university and the girls’ hostel outside the campus. The stream
with a roommate with symptoms. passed through a village up the hill, and the residents used the

4
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

Fig. 1. Distribution of norovirus cases by date of reporting in two norovirus outbreaks, Kerala, India, 2021.

stream water for their domestic activities. Like outbreak A, there was In both outbreaks, the epidemic curve with multiple small peaks
heavy rain before outbreak B started. During the rain, the stream’s suggested continuous exposure to the source and probable person-
water level would have increased to spill outside the river and enter to-person transmission. The drop in daily case-patients in both
the well. Though the same well supplied water to the girls’ and boys’ outbreaks after the superchlorination of the drinking water source
hostel, a regular supply of boiled water in the boys’ hostel canteen favoured the claim of continuous exposure to the infection source.
could have prevented them from the infection. Our investigation findings in outbreak B were consistent with the
previous outbreaks with well-documented person-to-person trans­
Discussion mission [16–21]. The higher risk of infection among roommates and
symptoms among household members in outbreak B, when case-
We documented the investigation of two acute gastroenteritis patients visited home during holidays, indicated person-to-person
outbreaks caused by norovirus in Kerala, India. Despite the differ­ transmission. However, we could not construct a phylogenetic tree
ence in the outbreak settings, both outbreaks were caused by con­ to establish a person-to-person transmission as reported in one of
taminated drinking water and probably spread by person-to-person the norovirus outbreaks in a school in China [18].
transmission. The groundwater could have been contaminated by We identified the contaminated drinking water as the source of
the seepage of water from the unprotected septic tanks in outbreak infection in both outbreaks, groundwater in outbreak A and well
A. The well water in outbreak B could have been contaminated by water in outbreak B. Globally, there were several documented nor­
the seepage of water from a nearby stream, probably contaminated ovirus outbreaks due to groundwater, municipal water and well
due to domestic activities in a village up the hill. To our knowledge, water contamination similar to our study [19,22–35]. Identification
these were the first documented norovirus outbreaks due to con­ of norovirus in the water samples was not always possible in the
taminated water in India. outbreak setting, as expressed in several of these studies, as it

5
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

Fig. 2. Attack rate of norovirus cases by administrative divisions & water tanks (outbreak A) and by hostels (outbreak B), Kerala, India, 2021.

requires sophisticated water concentration methods and diagnostic the age and gender distribution of identified case-patients. The low
techniques, which are not widely available in low-and-middle-in­ median age in outbreak A could be because of better health-seeking
come countries such as India. However, we documented the pre­ behaviour for AGE among children compared to adults. Our third
sence of norovirus in the water samples in one of the outbreaks. limitation was that there could have been a recall bias while esti­
Notably, GII norovirus was identified as the causative agent ra­ mating the association between drinking inadequately boiled water
ther than Group I (GI), which is usually prevalent in waterborne and suspected norovirus infection in outbreak A, as the case-patients
norovirus outbreaks [34,36–38]. However, there had been instances might have remembered water boiling practices such as boiling time
when GII norovirus was implicated as the causative agent for wa­ better than the controls. However, the association between drinking
terborne outbreaks [26,35,39]. inadequately boiled water and norovirus infection was consistent in
Our investigations had several limitations. Firstly, we did not both outbreaks. We could also identify the norovirus in the water
investigate the contribution of person-to-person or fomite-borne samples in outbreak A, establishing the biological plausibility.
transmission in outbreak A (community) as we did in outbreak B Fourth, we could not confirm the presence of norovirus in water
(university hostels). The low population attributable fraction for the samples during outbreak B due to logistic reasons. However, we
inadequately boiled water in Outbreak A also suggested another confirmed the well water’s faecal contamination (E. coli). Two meta-
source or route of transmission. Multiple small peaks in the epi­ analyses of groundwater studies revealed that bacteria, especially
demic curve drawn based on the reporting date were consistent with E.coli, could indicate viral occurrence in the groundwater, though the
outbreak B and other similar outbreaks that reported person-to- correlation was weak [40,41].
person transmission. Second, we described the case-patients in In conclusion, the two outbreaks of acute gastroenteritis were
outbreak A using the data only from passive surveillance. We could caused by Group II norovirus due to the consumption of in­
not conduct an active case search due to an active COVID-19 wave. adequately boiled and inadequately chlorinated contaminated water.
Thus, the health-seeking behaviour of the people could have affected The investigations highlighted the importance of testing the stool

6
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

and water samples for norovirus during AGE outbreaks in India and Acknowledgement
norovirus surveillance of drinking water sources.
We took the following measures in both outbreaks: (1) aware­ We would like to thank Dr A P Sugunan, ICMR-National Institute
ness to boil the water before drinking and avoid mixing with un­ of Virology, Alappuzha, for his prompt support in testing the stool
boiled water, (2) superchlorination of overhead tanks and wells. (3) samples and for his valuable guidance in the epidemiological in­
Practice handwashing before eating and after using the toilet in vestigation. We also thank the Alappuzha municipality staff, who
outbreak B. Following the superchlorination in all overhead tanks gave us information about the geography and water supply system,
(outbreak A) and wells (outbreak B) and awareness of drinking only which guided our investigation. We extend our gratitude to the field
boiled water based on our recommendation, the number of case- staff who supported us in the active case search and analytical
patients reduced gradually. We further recommended the following studies.
long-term engineering measures: (1) promoting the building of
brick-lined septic tanks in outbreak A, (2) building taller protective Conflict of interest
walls around the wells above the level of the stream in outbreak B,
(3) encouraging the village residents up the hill to refrain from using None.
the stream for domestic activities either by supplying them the
stream water through a piped water supply or providing an alternate References
source of water for domestic use.
[1] Norovirus Worldwide | CDC [Internet]. 2022 [cited 2023 Jan 3]. Available from:
〈https://www.cdc.gov/norovirus/trends-outbreaks/worldwide.html〉.
Statements [2] Lopman B.A. Global Burden of Norovirus and Prospects for Vaccine Development
[Internet]. CDC Foundation; 2015 [cited 2023 Aug 23]. Available from: 〈https://
www.cdc.gov/norovirus/downloads/global-burden-report.pdf〉.
NA. [3] Menon VK, George S, Sarkar R, Giri S, Samuel P, Vivek R, et al. Norovirus
Gastroenteritis in a Birth Cohort in Southern India. PloS One
2016;11(6):e0157007.
Funding statement [4] Lo M, Mitra S, De P, Banerjee A, Deb AK, Miyoshi SI, et al. Genetic character­
ization and evolutionary analysis of norovirus genotypes circulating among
children in eastern India during 2018-2019. Arch Virol 2021
This research did not receive any specific grant from funding Nov;166(11):2989–98.
agencies in the public, commercial, or not-for-profit sectors. [5] Gupta S, Krishnan A, Sharma S, Kumar P, Aneja S, Ray P. Changing pattern of
prevalence, genetic diversity, and mixed infections of viruses associated with
acute gastroenteritis in pediatric patients in New Delhi, India. J Med Virol 2018
Mar;90(3):469–76.
Ethical statement [6] Gupta S, Singh KP, Jain A, Srivastava S, Kumar V, Singh M. Aetiology of childhood
viral gastroenteritis in Lucknow, north India. Indian J Med Res 2015
We conducted these investigations as part of the public health Apr;141(4):469–72.
[7] Chhabra P, Dhongade RK, Kalrao VR, Bavdekar AR, Chitambar SD.
response the outbreaks. However, we submitted the First
Epidemiological, clinical, and molecular features of norovirus infections in
Investigation Report of the outbreak to the Institute Human Ethics western India. J Med Virol 2009 May;81(5):922–32.
Committee (IHEC) of the Indian Council of Medical Research – [8] Nagamani K, Rani M, Reddy V, Rao P, Rajyalakshmi S, Pakalapaty S. Molecular
epidemiology of norovirus variants detected in children under five years of age
National Institute of Epidemiology (ICMR-NIE), Chennai, and got
in Hyderabad, India. Indian J Med Microbiol 2022;40(1):12–7.
approval for the investigation. [9] Wikswo ME, Roberts V, Marsh Z, Manikonda K, Gleason B, Kambhampati A, et al.
Enteric Illness Outbreaks Reported Through the National Outbreak Reporting
System—United States, 2009–2019. Clin Infect Dis Publ Infect Dis Soc Am 2022
Authors’ contributions Jun 10;74(11):1906–13.
[10] Girish R, Broor S, Dar L, Ghosh D. Foodborne outbreak caused by a Norwalk-like
virus in India. J Med Virol 2002;67(4):603–7.
AR and MS were involved in conceptualization, design, metho­ [11] Stevenson M., Sergeant E., Heuer C., Nunes T., Heuer C., Marshall J., et al. epiR:
dology, lead role in investigation, supervision, data analysis, pre­ Tools for the Analysis of Epidemiological Data [Internet]. 2024 [cited 2024 Oct 5].
paration of first draft and manuscript revision. NM and SKC were Available from: 〈https://cran.r-project.org/web/packages/epiR/index.html〉.
[12] StataCorp. 2021. Stata Statistical Software: Release 17. College Station, TX:
part of the design, investigation, supervision, data analysis and re­ StataCorp LLC;
vision of the manuscript. HS, CPGK and AB were involved in the [13] Qiu Y, Lee BE, Ruecker NJ, Neumann N, Ashbolt N, Pang X. A one-step centrifugal
methodology, investigation, data interpretation and manuscript re­ ultrafiltration method to concentrate enteric viruses from wastewater. J Virol
Methods 2016 Nov;237:150–3.
vision. RN, MR and SM contributed to methodology, supervision, [14] Rio DC, Ares M, Hannon GJ, Nilsen TW. Purification of RNA using TRIzol (TRI
data acquisition, and manuscript revision. RR, DP, DA, SJXK, BB, NK reagent). Cold Spring Harb Protoc 2010 Jun;2010(6):pdb.prot5439.
and SS were part of Investigation, supervision, data acquisition, data [15] Updated Norovirus Outbreak Management and Disease Prevention Guidelines
[Internet]. [cited 2023 Feb 14]. Available from: 〈https://www.cdc.gov/mmwr/
analysis and manuscript revision. PK and MVM were involved in preview/mmwrhtml/rr6003a1.htm〉.
conceptualization, design, methodology, data interpretation, critical [16] Xu Y, Zhu Y, Lei Z, Rui J, Zhao Z, Lin S, et al. Investigation and analysis on an
review and manuscript revision. AR and MS contributed equally to outbreak of norovirus infection in a health school in Guangdong Province, China.
Infect Genet Evol 2021 Dec 1;96:105135.
this paper and designated as co-first authors. All authors have read [17] Zhou H, Wang S, von Seidlein L, Wang X. The epidemiology of norovirus gas­
and agreed to the contents of the manuscript. troenteritis in China: disease burden and distribution of genotypes. Front Med
2020 Feb 1;14(1):1–7.
[18] Li J, Gao X, Ye YL, Wan T, Zang H, Mo PH, et al. An acute gastroenteritis outbreak
associated with person-to-person transmission in a primary school in Shanghai: first
Data Availability report of a GI.5 norovirus outbreak in China. BMC Infect Dis 2018 Jul 9;18(1):316.
[19] Giammanco GM, Bonura F, Urone N, Purpari G, Cuccia M, Pepe A, et al.
Data are available on reasonable request. Waterborne Norovirus outbreak at a seaside resort likely originating from mu­
nicipal water distribution system failure. Epidemiol Infect 2018
May;146(7):879–87.
[20] Solano R, Alseda M, Godoy P, Sanz M, Bartolomé R, Manzanares-Laya S, et al.
Declaration of Competing Interest Person-to-person transmission of norovirus resulting in an outbreak of acute
gastroenteritis at a summer camp. Eur J Gastroenterol Hepatol 2014
Oct;26(10):1160–6.
The authors declare that they have no known competing fi­
[21] Wikswo ME, Hall AJ. Outbreaks of Acute Gastroenteritis Transmitted by Person-
nancial interests or personal relationships that could have appeared to-Person Contact — United States, 2009–2010. Morb Mortal Wkly Rep Surveill
to influence the work reported in this paper. Summ 2012;61(9):1–12.

7
A. Rajeevan, M. Sakthivel, N. Menon et al. Journal of Infection and Public Health 17 (2024) 102568

[22] Park JH, Jung S, Shin J, Lee JS, Joo IS, Lee DY. Three gastroenteritis outbreaks in with the Drinking Water Supply in a Rural Area in Galicia, Spain, 2021. Microbiol
South Korea caused by the consumption of kimchi tainted by norovirus GI.4. Spectr 2022 Jul 14;10(4):e01048-22.
Foodborne Pathog Dis 2015 Mar;12(3):221–7. [32] Werber D, Laušević D, Mugoša B, Vratnica Z, Ivanović-Nikolić L, Žižić L, et al. Massive
[23] Giammanco GM, Di Bartolo I, Purpari G, Costantino C, Rotolo V, Spoto V, et al. outbreak of viral gastroenteritis associated with consumption of municipal drinking
Investigation and control of a Norovirus outbreak of probable waterborne water in a European capital city. Epidemiol Infect 2009 Dec;137(12):1713–20.
transmission through a municipal groundwater system. J Water Health 2014 Sep [33] Scarcella C, Carasi S, Cadoria F, Macchi L, Pavan A, Salamana M, et al. An outbreak
1;12(3):452–64. of viral gastroenteritis linked to municipal water supply, Lombardy, Italy, June
[24] Cho HG, Lee SG, Kim WH, Lee JS, Park PH, Cheon DS, et al. Acute gastroenteritis 2009. Eurosurveillance 2009 Jul 23;14(29):19274.
outbreaks associated with ground-waterborne norovirus in South Korea during [34] Hewitt J, Bell D, Simmons GC, Rivera-Aban M, Wolf S, Greening GE.
2008–2012. Epidemiol Infect 2014 Dec;142(12):2604–9. Gastroenteritis Outbreak Caused by Waterborne Norovirus at a New Zealand Ski
[25] Koh SJ, Cho HG, Kim BH, Choi BY. An Outbreak of Gastroenteritis Caused by Resort. Appl Environ Microbiol 2007 Dec 15;73(24):7853–7.
Norovirus-Contaminated Groundwater at a Waterpark in Korea. J Korean Med [35] Maunula L, Miettinen IT, von Bonsdorff CH. Norovirus Outbreaks from Drinking
Sci 2011 Jan 1;26(1):28–32. Water. Emerg Infect Dis 2005 Nov;11(11):1716–21.
[26] Carol M, Guadalupe-Fernández V, Rius C, Soldevila N, Razquin E, Guix S, et al. A [36] Matthews JE, Dickey BW, Miller RD, Felzer JR, Dawson BP, Lee AS, et al. The
Waterborne Gastroenteritis Outbreak Caused by a GII Norovirus in a Holiday epidemiology of published norovirus outbreaks: a review of risk factors asso­
Camp in Catalonia (Spain), 2017. Viruses 2021 Sep;13(9):1792. ciated with attack rate and genogroup. Epidemiol Infect 2012
[27] Tryfinopoulou K, Kyritsi M, Mellou K, Kolokythopoulou F, Mouchtouri VA, Jul;140(7):1161–72.
Potamiti-Komi M, et al. Norovirus waterborne outbreak in Chalkidiki, Greece, [37] Maunula L, Von Bonsdorff CH. Norovirus genotypes causing gastroenteritis
2015: detection of GI.P2_GI.2 and GII.P16_GII.13 unusual strains. Epidemiol outbreaks in Finland 1998–2002. J Clin Virol 2005 Nov 1;34(3):186–94.
Infect 2019 Jan;147:e227. [38] Lysén M, Thorhagen M, Brytting M, Hjertqvist M, Andersson Y, Hedlund KO.
[28] Qin M, Dong XG, Jing YY, Wei XX, Wang ZE, Feng HR, et al. A Waterborne Genetic Diversity among Food-Borne and Waterborne Norovirus Strains Causing
Gastroenteritis Outbreak Caused by Norovirus GII.17 in a Hotel, Hebei, China, Outbreaks in Sweden. J Clin Microbiol 2009 Aug;47(8):2411–8.
December 2014. Food Environ Virol 2016 Sep;8(3):180–6. [39] Zhou X, Li H, Sun L, Mo Y, Chen S, Wu X, et al. Epidemiological and molecular
[29] Riera-Montes M, Sjölander KB, Allestam G, Hallin E, Hedlund KO, Löfdahl M. analysis of a waterborne outbreak of norovirus GII.4. Epidemiol Infect 2012
Waterborne norovirus outbreak in a municipal drinking-water supply in Dec;140(12):2282–9.
Sweden. Epidemiol Infect 2011 Dec;139(12):1928–35. [40] Fout GS, Borchardt MA, Kieke BA, Karim MR. Human virus and microbial in­
[30] ter Waarbeek HLG, Dukers-Muijrers NHTM, Vennema H, Hoebe CJPA. dicator occurrence in public-supply groundwater systems: meta-analysis of 12
Waterborne gastroenteritis outbreak at a scouting camp caused by two nor­ international studies. Hydrogeol J 2017;25(4):903–19.
ovirus genogroups: GI and GII. J Clin Virol 2010 Mar 1;47(3):268–72. [41] Hynds PD, Thomas MK, Pintar KDM. Contamination of Groundwater Systems in
[31] Jacqueline C, del Valle Arrojo M, Bellver Moreira P, Rodríguez Feijóo MA, the US and Canada by Enteric Pathogens, 1990–2013: A Review and Pooled-
Cabrerizo M, Fernandez-Garcia MD. Norovirus GII.3[P12] Outbreak Associated Analysis. PLoS ONE 2014 May 7;9(5):e93301.