Introduction
India has committed to measles elimination by 2023 [1]. Efforts are being made towards the elimination goal by achieving and sustaining vaccination coverage of 95% with two doses of a measles-containing vaccine (MCV) at the national and sub-national levels. As per the immunization schedule of India, the first dose of MCV is administered between 9 and 12 months of age and the second between 16 and 24 months [2]. Supplementary immunization with MCV irrespective of vaccination status was also conducted in 2018–2019. Case-based surveillance and an established network of 27 laboratories across the country aid in the country’s efforts towards elimination. India achieved the global standard non-measles non-rubella discard rate of 2 per 100,000 population during the last six months of 2021 [3].
In 2022, India had recorded 12,773 measles cases till November [Reference Vaidyanathan4]. India topped the Centers for Disease Control and Prevention (CDC) global list of countries with measles outbreaks from November 2022 to April 2023 [5]. Measles cases were reported from the states of Maharashtra [6, 7], Gujarat [8], Kerala [9], Uttar Pradesh [10], and Madhya Pradesh [11] from the latter part of 2022 with Maharashtra being the worst affected [8].
Between 17 and 23 September 2022, six suspected measles cases were reported from a slum area of Eastern Mumbai, India, predominantly belonging to a religious minority. Measles incidence in Eastern Mumbai in 2022 showed an upward trend with 0.83/million population compared to 0 in the previous three years. An outbreak was declared on 20 September 2022 after the detection of measles IgM antibodies by enzyme-linked immunosorbent assay (ELISA) in five of the six reported case-patients. We investigated the outbreak to describe its epidemiology, ascertain measles vaccination status, and provide recommendations for its control.
Methods
Case definitions
We defined a suspect case [12] as the occurrence of any maculopapular rash illness with fever in a person of any age in the slum area of Eastern Mumbai between 17 June 2022 and 20 February 2023 for a house-to-house case search. We further classified these cases as probable and laboratory-confirmed (immunoglobulin M (IgM) ELISA-positive) cases. We defined a measles death as death within 30 days of onset of rash without any obvious causes like trauma, accident, or snake bite in a person of any age within the last 90 days [12].
Descriptive epidemiology
Mumbai City Municipal Corporation initiated an active house-to-house search for cases immediately after the outbreak confirmation in the Eastern Mumbai slum, which continued till February 2023 when the outbreak was declared closed. During the active case search, data were collected on the demography, symptoms, vaccination status for MCV, laboratory testing, and results for suspect cases on a paper-based form and compiled in Google Sheets. We constructed an epidemic curve by plotting the distribution of cases by the date of onset of fever. We drew a spot map of the cases among the households in the slum area. We obtained the age and sex distribution of the under-5 child population and the 5–15-year population from the municipal corporation for the calculation of attack rates and case fatalities by age and sex.
Analytical epidemiology
We hypothesized that failure to vaccinate or vaccination failure of the MCV could be associated with the outbreak. To test these hypotheses, we conducted an unmatched case–control study, separately for the under-5 and 5–15-year age groups among the households in the outbreak area. We defined a case as an individual having maculopapular rash illness with fever in the specified age groups in the last 90 days. We defined a control as an individual in the specified age group who did not satisfy the criteria for a case. Cases and controls were enrolled from different households.
Using the OpenEpi software, we calculated a sample size of 133 cases and 133 controls each in the under-5 and 5–15-year age groups assuming an odds ratio (OR) of 0.5 associated with vaccination, 57% of cases vaccinated with one dose of MCV (unpublished data from Mumbai Municipal Corporation), a 95% confidence interval (CI), and a 1:1 case–control ratio.
We collected information from suspected measles cases and controls about the receipt of routine and supplemental measles vaccination along with reasons for not vaccinating, if unvaccinated, and duration of residence in Mumbai using a structured form by interviewing the caregiver of the children through face-to-face interviews and reviewing the vaccination cards. Wherever a vaccination card was not available, a recall of vaccination administration was asked from the caregiver.
We calculated the OR for one and two doses of MCV and vaccine effectiveness (VE) as VE = 1 – OR separately for the under-5 and 5–15-year age groups. We also performed a stratified analysis by sex, availability of vaccination cards, and migrant status (defined as less than ten years of residence in Mumbai). We used Epi Info version 7.2 for the statistical analysis.
Results
Description of the outbreak
We identified 358 cases (including six laboratory-confirmed) and four deaths among the under-15 population of 3,156 residing in approximately 3,100 houses in the slum, with an overall attack rate of 11.3% and 1.1% case fatality. Attack rates were highest in the 0–24-month age group (38.8%) and boys (13%) (Table 1). The median age of case-patients was 30 months (interquartile range (IQR) 16–48 months). Cough, coryza, conjunctivitis, and pneumonia were seen in 94%, 94.3%, 51.6%, and 0.8% of case-patients, respectively, at first presentation.
Distribution of case-patients by age and sex, Eastern Mumbai slum, India, September 2022–February 2023
The index case had febrile rash illness in the third week of September 2022. Majority of the case-patients reported onset of fever in the second week of November 2022 (n = 132, 36.9%) and peaked by the end of the second week. Thereafter, the number of cases declined. All four measles deaths occurred before the first peak. Outbreak response immunization was initiated in the third week of November and coincided with the peak. A second smaller peak was seen in the first week of December 2022. The two peaks were one incubation period apart, suggesting propagated mode of spread through person-to-person transmission. The last case was reported during the second week of January 2023 (Figure 1). Cases were distributed across the slum area (Figure 2).
Distribution of cases by date of onset of fever in Eastern Mumbai slum, India, September 2022–February 2023.
Distribution of cases by place in Eastern Mumbai slum, September 2022–February 2023.
Among controls, 117 (71%) and 82 (63%) among the under-5 age group and 85 (88%) and 64 (82%) among the above-5 age group had received one and two doses of MCV, respectively. Overall, 46 (16.3%), 149 (51.9%), and 91 (31.7%) case-patients reported receiving zero, one, and two doses of MCV, respectively. Despite supplementary immunization activities in 2018–2019, none of the case-patients reported receiving three doses of MCV. Only about 50% of case-patients in the 9–12-month age group had received the first dose of MCV. 42.5% of male case-patients and 19% of female case-patients in the 13–24-month age group had received two doses of MCV. The percentage of case-patients vaccinated with either one or two doses of MCV was similar across both sexes except in the 13–24-month and above 120-month age groups where vaccination among female was lesser than among male case-patients (Table 2).
MCV status of case-patients eligible for one and two doses by age and sex, Eastern Mumbai slum, India, September 2022–February 2023
a Case-patients eligible for one dose of MCV only.
b Case-patients aged 16 months and above eligible for two doses of MCV.
The reasons reported by the caregivers for incomplete vaccination of children (n = 70) were the child being sick (59%), family-related problems (14%), lack of knowledge about immunization sessions (10%), adverse events following immunization (10%), lack of faith in vaccination (4%), and non-availability of staff for vaccination (3%).
Vaccine effectiveness
VE was 40% (95% CI: 5–62%, p-value 0.43) and 64% (95% CI: 23–73%, p-value <0.01) for one and two doses of MCV, respectively, in the under-5 age group. The VE in the 5–15-year age group was 64% (95% CI: 17–84%, p-value <0.01) for one dose and 70% (95% CI: 28–88%, p-value 0.04) for two doses of MCV. There was no difference in VE by sex, availability of vaccination cards, and migrant status in both age groups (Table 3).
Effectiveness of MCV in the under-5 and 5–15-year age groups by dose, stratified by sex, availability of vaccination card, and place of permanent residence, Eastern Mumbai slum, India
a Vaccination card was unavailable for the above-5 age group; hence, the stratum was excluded from the analysis.
Discussion
We report an outbreak of measles in a slum area of Eastern Mumbai where a high proportion of eligible children missed either one or two doses of MCV, resulting in a widespread outbreak with a propagated mode of spread. The VE was low, albeit higher for two doses among all age groups, with the child’s sickness being the main reason for not receiving vaccination.
Globally, the progress towards measles elimination substantially declined in 2020 due to suboptimal measles surveillance and missed vaccination [Reference Dixon, Ferrari and Antoni13]. According to the World Health Organization (WHO), nearly 40 million children missed MCV dose(s) in 2021 leaving them highly vulnerable to infection, leading to an estimated nine million cases and 128,000 deaths globally.14 India was next only to Nigeria among nations that contributed to the highest number of infants who missed their first dose of MCV in 2020 [14].
A large proportion of children missing their MCV dose(s) indicate the problem of failure to vaccinate, which could have resulted in the accumulation of susceptible children in the slum area. This is corroborated by higher vaccination rates among the controls. The occurrence of 38% of cases among children aged > = 3 years also suggests suboptimal coverage of routine immunization in the area. The restrictions imposed as a result of the COVID-19 pandemic may have led to an accumulation of susceptible population in the community at large. The restrictions ranged from social distancing norms, which deterred parents from bringing children for immunization to programme-related factors including suspension of mass vaccination campaigns and disruption of the supply chain of vaccines, in addition to the diversion of human resources towards pandemic control leading to delayed or missed vaccination, thereby creating a suitable milieu for the outbreak [Reference Ackerson, Sy and Glenn15, Reference Ho, Gurung and Mirza16]. Since overcrowding facilitates transmission, the population living in the slums was particularly vulnerable. Vaccine hesitancy, influenced by religious, social, cultural, and personal factors, may have also played a role [Reference Erchick, Gupta and Blunt17, Reference Krishnamoorthy, Kannusamy, Sarveswaran, Majella, Sarkar and Narayanan18]. A recent study by Tamysetty et al. in Eastern Mumbai’s slums highlighted concerns about vaccine safety, dependence on family members for vaccination decisions, fear, and lack of faith in immunization in the context of COVID-19 vaccination [Reference Tamysetty, Babu and Sahu19], some of which may be applicable to hesitancy to measles vaccination as well. Childhood sickness as the most commonly reported reason for missed vaccinations could be a socially desirable response implicitly suggesting vaccine hesitancy in this population. In addition to the above-mentioned factors, a common social scenario comprising large family size, lower family income, and children of working mothers increases the chances of missed vaccination among children residing in slums [Reference Hu, Li, Luo, Lou, Qi and Xie20–Reference Singh, Sahu, Agrawal and Vashi22].
Poor VE can result from failure to vaccinate [23]. A seropositivity of 80–85% is reported in children who receive two doses of the MCV long after the vaccination [Reference Hansashree, Verma, Rawat, Sankhyan and Bharti24], and the interval between the doses does not seem to affect the sero-protection [Reference De Serres, Sciberras, Naus, Boulianne, Duval and Rochette25]. The VE estimated in our study is much lower than the expected based on global estimates of MCV1 VE of 77% for the Southeast Asian region based on a meta-analysis [Reference Uzicanin and Zimmerman26]. This could be due to lower age at administration of measles vaccine [27] and programme-related factors, including suboptimal vaccine storage, handling, and administration [Reference Cutts, Smit and Colomno28]. Potentially inaccurately documented dose due to recall bias could have led to low estimates of VE in our study. Further implementation research is needed to understand the determinants of low VE factors and potential strategies to improve them in the Indian context.
Vaccine failure can be primary or secondary depending on whether or not the individual ever developed an immune response to the vaccine and can be host-related (immunodeficiency, age-related, malnutrition) [Reference Rytter, Kolte, Briend, Friis and Christensen29–Reference Wu, Xue and Zhang31] or vaccine-related (incomplete coverage of strains, intramuscular route of administration rather than subcutaneous, manufacturing differences) [Reference Pollard and Bijker32]. Children with secondary vaccine failure might be those who initially received and responded to the MCV, and over time, their protective immunity diminished, rendering them susceptible to the disease once again [Reference De Serres, Sciberras, Naus, Boulianne, Duval and Rochette25]. However, this needs to be further investigated by humoral and cell-mediated immune responses to identify whether vaccine failure was primary or secondary [Reference Hickman, Hyde and Sowers33, Reference Ward, Boulianne, Ratnam, Guiot, Couillard and De Serres34].
It was observed during the outbreak that measles cases occurred among vaccinated children, which suggests the possibility of vaccine failure in addition to failure to vaccinate as a possible cause of this outbreak. The VE was particularly low among the under-5 population who had received only one dose of MCV, suggesting possible primary vaccine failure among those who had not received the second dose of MCV. Though we do not have primary data from the outbreak area, undernutrition as high as 25–30% is known to exist among the urban slum population in Mumbai [Reference Huey, Finkelstein and Venkatramanan35], which might be one of the factors responsible for poor VE, resulting from frequent infections and consequent missed vaccination as well as vaccine failure. The prevalence of known risk factors for secondary vaccine failure, such as malnutrition, and the presence of other comorbidities need to be assessed in this area. These factors for secondary vaccine failure are relevant for children who had received both doses of MCV.
Our study has certain limitations. Firstly, we could not achieve the desired sample size in the 5–15-year age group, leading to a lower power in estimating the VE in this age group. Secondly, the vaccination status, wherever vaccination cards were not available, was based on recall by nearly half of the study participants in the under-5 and all of the 5–15-year age groups, and more so among the cases than the controls. This could have led to differential misclassification in vaccination status, resulting in the VE estimate biased in either direction. Thirdly, a measles infection prior to 90 days and conferred immunity thereof could have influenced the VE estimates. Finally, the study was conducted in a slum area in Mumbai; hence, the VE estimates cannot be generalized to the entire city.
In order to achieve elimination status, measles vaccine coverage of 95% with two doses is recommended by the WHO [36]. To regain momentum towards the 2023 elimination goal, catch-up vaccination efforts through government strategies are necessary to improve vaccine coverage wherever deficient. In addition to improving coverage, it may also be prudent to explore if a policy decision by the Indian government for additional vaccination at a later age will aid in improving the overall immune status of the community.
In order to prevent future outbreaks, improving vaccination coverage through targeted community awareness and outreach programmes and focusing on tracing and vaccinating children who missed doses due to illness need to be strengthened. Increased community engagement, involving religious leaders and influencers, can help build trust. In-depth research for understanding reasons for non-vaccination, a multi-stakeholder analysis of gaps in the implementation of the immunization programme, and the evaluation of the immunization programme in Eastern Mumbai, particularly in slum areas, can help inform future strategies to reduce missed opportunities for vaccination.
Outbreak response immunization and additional routine vaccination camps to complete the immunization of missed and dropout children were carried out by Mumbai civic authorities. It is worth noting that there was a delay in initiating the outbreak response immunization in the area, which could have possibly prevented such a large outbreak. All practitioners in the area, private hospitals, and nursing homes were informed to notify early of all cases of fever with rash irrespective of age. They were also instructed to isolate and adequately treat the cases and administer vitamin A to all cases of fever with rash to prevent complications. All school headmasters were instructed to report any case of suspected measles and to restrain such students from attending school to limit the spread of measles. Meetings with religious leaders were conducted to improve vaccine hesitancy resulting from religious grounds. The last case of measles was reported on 17 January 2023.
Data availability statement
Data supporting this study can be obtained upon reasonable request from the author.
Acknowledgements
The authors acknowledge the partial funding support from the ICMR-National Institute of Epidemiology, India. The authors acknowledge the administrative support from the EHO office, the Public Health Department, and the Municipal Corporation of Greater Mumbai, India. The authors acknowledge Ms. Aishwarya Nighojkar and Dr. Dhruv Kalawadia for helping in data collection for analytical epidemiology.
Author contribution
R.M.Y., T.B., and M.M. conceptualized the study. R.M.Y., U.B., V.K.K., J.W.V.T., S.B., T.B., and M.M. planned the study methods and data collection formats. M.G., A.G., and U.W. supervised the descriptive data collection, facilitated the study, and managed the outbreak. R.M.Y. and U.B. collected data for analytical epidemiology. R.M.Y. accessed and verified the descriptive data, analysed the data, wrote the first draft, and edited the manuscript. M.D.V., S.B., T.B., and M.M. reviewed and edited the manuscript. T.B. supervised the study and acquired funds for the study. All authors reviewed and finalized the final draft of the manuscript.
Funding statement
This study was partially supported by funding from the ICMR-National Institute of Epidemiology, India. The funding source had a role in study design; ethical approval; collection, analysis, and interpretation of data; writing and editing of the report; and the decision to submit the paper for publication.
Competing interest
The authors declare none.
Ethical standard
The study was approved by the Institutional Human Ethics Committee of the Indian Council of Medical Research-National Institute of Epidemiology, Chennai, India.
Open access
