Context: Measles imparts high morbidity and mortality in low-income countries with weak health infrastructure. The coverage of measles vaccination in Kerala which is best performing state of India in indicators of health has also not reached the elimination level and outbreaks of measles continue to occur. Aim: The aim of the paper was to study the profile of measles outbreaks in Kerala during the years 2007-2008. Settings and Designs: The study design was cross-sectional and data of the measles surveillance project of Kerala were analyzed . Results: The total number of clinically suspected measles outbreaks in Kerala during the 2-year period of 2007 and 2008 was 29. But only 15(53.6%) were found to be ‘measles only’ outbreak. The total number of epidemiologically linked measles cases was 718 (314 in 2007 and 404 in 2008). The cases that were immunized but developed the disease were 187 (28.6%), the number of cases that were not immunized was 355 (54.3%) and those whose immunization status was unknown were 112 (17.1%). The mean age of occurrence of disease was higher in the vaccinated group as compared to not vaccinated group. Two deaths were recorded in the study. Conclusions: Significant morbidity and mortality due to measles do occur in the most developed state of India. The epidemics were clustered in some districts. The study indicated an age shift in occurrence of measles cases among who received the vaccination.
Keywords: Measles epidemic, measles in India, measles in Kerala, measles outbreak, measles surveillance
Measles is a highly infectious systemic childhood disease , affecting predominantly the respiratory system. It is endemic in almost all parts of the world  but is associated with high mortality and morbidity in developing countries.  Measles killed 7-8 million children a year and caused an estimated 135 million cases per year worldwide before the introduction of vaccination in the 1960s.  But even in 2007 there were 197 000 measles deaths globally which account for nearly 540 deaths everyday and 22 deaths every hour.  The striking fact is that more than 95% of measles deaths occur in low-income countries with weak health infrastructure. 
In India measles contribute significantly to the childhood morbidity and mortality. Measles vaccine was introduced in to the universal immunization program (UIP) of the country in 1985. Number of cases has come down from 252,000 cases in 1987 to 36900 cases in 2007.  Kerala, the southern state of India, which is well known for its better health indicators is also endemic for measles. The total number of measles cases decreased from 23156 in 1988-89  to 4937 in 2003.  Despite the fact that the number of cases and deaths due to measles are declining, it continues to occur both as sporadic as well as outbreaks even though scientific studies reveal that measles eradication is technically feasible with available vaccines.  But the vaccination coverage against measles in Kerala, the best performing state of the nation during the year 2005-06, was only 82.1%.  It has not reached the elimination level and outbreaks of measles continue to occur. The aim of the paper is to study the profile of measles outbreaks in Kerala during the years 2007-2008 there by contribute in identifying the gaps of measles elimination.
The study design was cross-sectional and data of the measles surveillance project of Kerala was used for analysis . The study was conducted from February to April 2009. A case of Measles was defined by the project as an epidemiologically confirmed case of measles that was investigated as a part of measles only outbreak. Epidemiologically confirmed case is a subject who meets the clinical case definition and is linked epidemiologically to a laboratory confirmed case. The study protocol was approved by the human ethical committee of Government Medical College, Thiruvananthapuram.
The setting was Kerala state, India. As of 2001 census the total population of the state is around 31.80 million and a population density of 819 per sq km. The state is divided in to 14 revenue districts . The measles surveillance project was initiated in the state from March 2007. Under the measles surveillance project an investigation is triggered when,
A preliminary search is initiated by the District Immunization Officer (DIO) and the Epidemic Response Team (ERT) is sent to conduct a comprehensive measles outbreak investigation. The health workers of each area conducts a house to house search to find clinical measles cases fitting the case definition that has occurred in the last 3 months and lists them down. The ERT identifies at least five cases of measles and deputes a medical officer to take the blood specimen from cases with onset of rash 4-28 days prior to visit. Samples are tested for antibodies against measles and rubella in a WHO accredited laboratory. The results of which are used to classify the outbreak as measles only outbreak (at least one sample positive for IgM measles and none for rubella), rubella only outbreak (one or more samples positive for IgM rubella and none for measles) or mixed outbreak (some samples positive for IgM measles and some for IgM rubella).
The epidemiological factors of the outbreaks and the attributes of the epidemiologically linked cases were analyzed separately. One category of variables were related to the outbreaks like the year and location of outbreaks, number of cases, duration, death and the number of samples positive for measles. The other category of variables were related to the epidemiologically linked cases like there socio-demographic characters, clinical features of the exanthematous fever and mortality.
Univariate analysis was done by estimating the means (Standard Error) of quantitative variables (eg., age) and proportions in case of qualitative variables (eg., sex, place, vaccination status). Bivariate analysis such as unpaired t test, ANOVA (to estimate the statistical significance where the dependent variable is quantitative) was also used.
The total number of clinically suspected measles outbreaks in Kerala, that were investigated during the 2-year period of 2007 and 2008 was 29. But only 15 (53.6%) were found to be ‘measles only’ outbreak. The rest of the outbreaks (14) were either ‘rubella only’ outbreaks or ‘mixed outbreaks’ of both measles and rubella. The total number of epidemiologically linked measles cases during the years 2007-2008 were 718 (314 in 2007 and 404 in 2008). But only 654 (91%) cases were available and used for analysis.
District wise distribution of outbreaks and cases
Among the 14 districts only six districts had confirmed measles outbreak in the years 2007-2008. The locations of the outbreaks are illustrated in [Figure 1]. The districts that were consistently and severely affected were Malappuram district in the north of Kerala and Thiruvananthapuram district in the southern most part of Kerala [Table 1]. Rest of the 8 districts did not have any confirmed measles outbreaks [Table 1].
Seasonal distribution of cases
The months which had the most number of cases were April, June and July in 2007, while in 2008 it was August, September and October. The [Figure 2] shows a shift of the peak of epidemic toward the later months in 2008 as compared to that of 2007.
Duration of epidemic
The maximum and minimum duration of the measles outbreak in 2007 were 92 days from Malppuram district and 28 days from Thiruvananthapuram district. The mean (Standard error) duration of an epidemic in 2007 was 64.6 (2.34) days. The same for the year 2008 were 120 days from Malappuram district and 10 days from Kozhikode district. The mean (SE) duration of an epidemic in 2008 was 67.3 (3.66) days. In both the years the maximum duration of epidemics was noted in Malappuram district. The mean duration epidemics in Malappuram district was 85 days while that for the rest of the districts was only 59 days. But this difference is not statistically significant (t= 1.434, df=12, P= 0.177).
The mean (SE) age of the cases in the year 2007 was 72.3 (4.1) months and in 2008 was 78.5 (3.3) months. An age shift is noted between these 2 years, but this difference is not statistically significant. However, to confirm this shift the data of a few more years may be needed.
In the year 2007 males contributed to 57.5% of the cases while in 2008 it was 49.8%, overall for both the years males contributed to 52.8% of cases.
The cases that were immunized but developed the disease were 187 (28.6%), the number of cases that were not immunized was 355(54.3%) and those whose immunization status was unknown were 112 (17.1%).
The mean (SE) age of those who were vaccinated were 88.74 (4.06) months, while for those unvaccinated were 57.21 (2.72) months and for those whose status was unknown was 115.02 (8.92) months.
There was a significant association between the time (age) of development of disease and vaccination status. The mean age of occurrence of disease is higher in the vaccinated group as compared to not vaccinated group [Table 2].
Two deaths were recorded in the study. They were an 11-month-old female child from Malappuram district and a 192-day-old child from Thiruvananthapuram district during the year 2007. Both were female unvaccinated children. No deaths were reported among the 404 cases in 2008. Overall two deaths from 714 cases were recorded giving a case fatality rate of 2.8 per 1000 cases.
The less number of outbreaks in 2007 could have been due to poor reporting because measles surveillance was introduced only in March 2007. The system may have taken some time in 2007 to be fully implemented. The data in the coming years may shed light to the true epidemiological distribution of cases.
Among the 14 districts in Kerala state, only six districts ever had a confirmed measles only outbreak since the implementation of measles surveillance program. The districts of Malappuram and Thiruvananthapuram showed consistent and severe out breaks in both the years. Malappuram district is located toward the north of Kerala and consistently reported low vaccination coverage.  The poor vaccination coverage may be an important reason for the outbreaks. The coverage in this district is less than 50% for 3 UIP vaccines for the age group 12-23 months (DPT 3, OPV 3 and Measles).  The coverage for measles vaccination was only 42.9% .  The vaccination coverage against measles in Thiruvananthapuram district was 850 % in 2008-09.  Thiruvananthapuram district though better in its coverage also had the maximum number of outbreaks. This may be due to pockets of areas in the district where the immunization coverage is low.
Looking into such spot maps of other states and that of Kerala, bordering districts showed a number of outbreaks. There is a great need to be alert about the outbreaks occurring in neighboring states.
Even though measles is seen during winter and early spring (January-April), it was observed that the peak of out breaks happened in Kerala during April to June in 2007 and August to October in 2008. The period of epidemic in 2007 appeared in the middle of the dry season in Kerala and tapered toward beginning of 1 st monsoon season. The period of epidemic in 2008 appeared by the end of 1 st monsoon and tapered by the beginning of the 2nd monsoon season. Kerala does not have extremes of climates and do not experience severe heat or cold and a link of measles with such seasons is difficult to establish. This supports the fact that the virus can spread in any season . 
The duration of (mean) of epidemics in Malappuram district was 85 days while that for the rest of the districts was only 59 days. But this difference is not statistically significant, which may be because of few numbers. For Malappuram the number of waves is roughly 6 (85/14=6), after which the outbreak subsides down (the average incubation period till the onset of rash in measles is 14 days). The number of waves for the rest of the districts in Kerala is roughly 4 (59/14=4). The numbers of waves are likely to be depended upon the subset of susceptible population (unvaccinated) in the community coming in contact with the cases. If the vaccination coverage is high the number of waves (duration of epidemics) are likely to decrease.
The mean age of the cases in the year 2007 was 72.3 months (SE 4.1) and in 2008 was 78.5 months (SE 3.3). An age shift is noted between these 2 years which is similar to studies of other surveillance data, but this difference is not statistically significant. However, to confirm this shift the data of a few more years needs to collected and analyzed .
Among the cases of measles the numbers of unvaccinated persons were 355 (54.3%). This shows that the immunization coverage is low and substantiates the fact that failure to give the primary dose is the main cause of continuing high measles mortality and morbidity. To be able to give the primary dose (first opportunity) is the first and best option for intervention and prevention of disease occurrence and transmission in the community. It is alarming that the measles vaccination coverage in the state has decreased when comparing NFHS 2 (1998-99 )  and NFHS 3(2005-06)  data. This gives clue to impending outbreaks and epidemics in the state.
The mean age of attack of measles among those who were vaccinated was 88.74 months (SE 4.06), while for those unvaccinated were 57.21 months (SE 2.72). The higher age at incidence among the vaccinated group could be due to the waning immunity of the first dose or inadequate efficacy of the first dose. It has been shown that if a second opportunity is given it can slow the increase of susceptible children by vaccinating those who had missed the first dose and providing protection to those who did not develop an immune response after a single dose of measles vaccine.  The second dose comes at a later time in the life of the child and the seroconversion rate can be expected to be above 95% if given above 12 months of age.  This will help in better herd immunity hence drastically decrease the disease transmission in the community. It has also been shown that to interrupt transmission of measles in a population where it is endemic more than 95% of the population should be immune .  These findings emphasize the need to provide a second dose of measles vaccine.
A major limitation to the study is that even though the system is fully operational in the state, there are a lot of underreporting of measles cases. This may have underestimated the real problem of measles in the state.
Measles outbreaks continue to occur in the State with unequal geographical distribution. Malappuram district of northern Kerala where the vaccination coverage is low is a hot spot for Measles outbreaks. The determinants of higher number of outbreaks in Thiruvananthapuram district should be studied further. Measles do occur in vaccinated children as well but are older compared to unvaccinated children. The outbreaks can further be prevented by administering a second dose of measles vaccine to reduce the number of children who have missed their first dose or who have failed to respond with the first dose. This will drastically reduce the susceptible population. The state may face more and larger outbreaks of measles in the coming years if the current trends in vaccination coverage and policies continue. It is prudent to prevent these outbreaks by increasing the vaccination coverage and improving case management if the outbreaks occur.
Dr. Asha Raghavan, Surveillance Medical Officer for Polio and Measles, Kerala.
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2]
[Table 1], [Table 2]