Preplanned National Measles Vaccination Campaign at the Beginning of a Measles Outbreak Sierra Leone,

Transcription

1 SUPPLEMENT ARTICLE Preplanned National Measles Vaccination Campaign at the Beginning of a Measles Outbreak Sierra Leone, David E. Sugerman, 1 Amadou Fall, 2 Marie-Thérèse Guigui, 3 Michael N'dolie, 4 Terry Balogun, 4 Alie Wurie, 5 and James L. Goodson 1 1 Centers for Disease Control and Prevention, Global Immunization Division, Atlanta, Georgia; 2 World Health Organization, Inter-Country Support Team for West Africa, Ouagadougou, Burkina Faso; 3 UNICEF, Health Specialist for Measles, Yellow Fever, and Health Emergencies, Regional Office for West and Central Africa, Dakar, Senegal; 4 Expanded Programme on Immunization, World Health Organization, and 5 Ministry of Health and Sanitation, Expanded Programme on Immunization, Freetown, Sierra Leone Background. Large-scale measles outbreaks occurred throughout Africa from In Sierra Leone, in November 2009, preceding a measles supplemental immunization activity (SIA), the largest measles outbreak in a decade started. Methods. We analyzed data from the national measles case-based surveillance system, developed a susceptibility profile of the population, and calculated vaccine effectiveness (VE) among children months of age. Results. From November 1, 2009 to July 13, 2010, 1,094 confirmed cases, including 9 deaths, were reported; 716 (66%) were,5 years of age. B3 genotype was identified. Measles attack rates per 100,000 population were highest among infants aged 6 8 months (56.4) and in Bo district (49.4). Districts with higher estimated SIA coverage tended to have lower attack rates (Spearman Correlation Coefficient ), p Among 473 cases with information on vaccination status, 222 (47%) were unvaccinated; estimated VE was 74%. The 2009 measles SIA led to 165,000 fewer estimated susceptible individuals. Conclusions. The 2009 measles SIA reduced the overall magnitude of the outbreak, though routine and SIA coverage was insufficient to prevent it entirely. Maintaining high coverage through routine services and SIAs in all districts and conducting follow-up SIAs prior to the end of the low transmission season may prevent future outbreaks. Global goals for measles control were first set by the World Health Assembly in 1989 [1] and by the World Health Summit for Children in 1990 [2]. In sub-saharan Africa during the 1970s, the Expanded Programme of Immunizations (EPI) was established and included a single dose of measles vaccine at 9 months of age for measles control. In 2000, the countries of the World Health Organization (WHO) African Region (AFR) Potential conflicts of interest: none reported. Supplement sponsorship: This article is part of a supplement entitled ''Global Progress TowardMeasles Eradication and Prevention of Rubella and Congenital Rubella Syndrome,'' which was sponsored by the Centers for Disease Control and Prevention. Reprints or correspondence: David E. Sugerman, MD, MPH, Global Immunization Division, Centers for Disease Control and Prevention, 1600 Clifton Rd, NE; MS-E05, Atlanta, GA (ggi4@cdc.gov). The Journal of Infectious Diseases 2011;204:S260 S269 Published by Oxford University Press on behalf of the Infectious Diseases Society of America (print)/ (online)/2011/204s1-0033$14.00 DOI: /infdis/jir110 became part of a global initiative with a goal to reduce the number of measles deaths by 50% by 2005, compared with the number of such deaths in 1999 [3]. To achieve this goal, the WHO and the United Nations Children s Fund (UNICEF) recommended the measles mortality reduction strategy in Africa that included improving case management, establishing case-based surveillance with laboratory confirmation of suspected measles cases, providing the first dose of measlescontaining vaccine (MCV1) for all children, and providing a second opportunity for measles vaccination through supplemental immunization activities (SIAs), regardless of previous vaccination status or history of measles disease [4]. The SIA component of the strategy included an initial wide-age-range catch-up SIA to target children from 9 months through 14 years of age to rapidly reduce the number of individuals who are susceptible to measles and included periodic follow-up SIAs every 3 5 years S260 d JID 2011:204 (Suppl 1) d Sugerman et al

2 to target children born since the previous SIA [5]. Following implementation of the measles mortality reduction strategy starting in 2001, the 50% measles mortality reduction goal was achieved [3], and a new goal was established to achieve 90% reduction by 2010, compared with measles mortality in 2000 [6]. Although substantial progress toward this goal was made from 2000 through 2008, several large measles outbreaks occurred throughout the region during the period [7], including an outbreak in Sierra Leone. Sierra Leone is located in West Africa, with an estimated population of 5.8 million in 2010 [8], and is administratively divided into 4 provinces (Northern, Eastern, Southern, and Western Area) and 14 districts. Emerging from a crippling, decade-long civil war ( ) that led to.50,000 deaths and a million internally displaced persons [9], Sierra Leone remains one of the least developed nations in the world [10] and has the world s highest mortality rate among those,5 years of age (290 deaths per 1000 live births) [11]. In addition, after the civil war, donor funds dropped, health personnel dispersed, and the already limited health infrastructure was destroyed [12]. In Sierra Leone, EPI was established in 1974 [12]. Measles vaccination coverage remained low (,50%) until the push toward universal child immunization (UCI) occurred in the 1980s [13]. During the period , prior to the civil war, routine MCV1 coverage increased from 36% to 75%, and the annual number of reported measles cases decreased from 3625 to 830 [14]. During the civil war, MCV1 coverage estimates varied considerably each year (28% 79%), leading to a resurgence of measles with 3575 reported cases in 2000 (Figure 1). As part of the WHO AFR measles mortality reduction strategy adopted by member states in 2001, SIAs started in Sierra Leone in November 2003 with a nationwide catch-up SIA targeting children 9 months 14 years of age with measles vaccination and had reported administrative coverage of 93% (administrative coverage is calculated by dividing the total number of doses administered to children in the target age group by the number of children in that age group) (Table 1). Three years later, in November 2006, a nationwide follow-up SIA was conducted that targeted children 9 months 4 years of age; 100% administrative coverage with measles vaccine was reported [13]. In 2009, a second nationwide follow-up SIA targeted children 9 59 months of age with measles vaccination and those 9 months of age and older with yellow fever vaccination in 6 select districts. Following this SIA, administrative coverage with measles vaccination was 101%; however, a post-sia cluster survey estimated measles vaccination coverage was 84% by verbal history and 66% by vaccination card [15]. In Sierra Leone, following Figure 1. Number of measles cases reported and estimated percentage of children who received the first dose of measles-containing vaccine through routine services, Sierra Leone, *Confirmed cases of measles reported to the World Health Organization (WHO) and the United Nations Children's Fund (UNICEF) through the Joint Reporting Form Regional Office for the Africa Region ( ). Vaccination coverage data from Sierra Leone Expanded Programme of Immunizations ( ) and WHO and UNICEF estimates ( ) [14]. à Supplemental immunization activity (SIA). Sierra Leone Measles Outbreak, d JID 2011:204 (Suppl 1) d S261

3 S262 d JID 2011:204 (Suppl 1) d Sugerman et al Table 1. History of Measles Supplemental Immunization Activities in Sierra Leone, Year Dates Target age 2003 b 1 6 Nov 9 Months to 14 years Target population Age in November 2009 Integrated interventions 2,585,894 6 Years and 9 months to 21 years 2006 b Nov 9 59 Months 721,063 3 Years and 9 months to 7 years and 11 months 2009 d Nov 9 59 Months 824,366 9 Months to 4 years and 11 months Admin a Measles vaccine coverage (%) Cluster survey None 93. Vitamin A, mebendazole, and LLIN c (,5 years of age) in all 14 districts Yellow fever vaccine (R9 months of age) in 6 districts plus oral polio vaccine, albendazole, and vitamin A (,5 years of age) in all districts (95% confidence interval, 52 97) by card; 84 (95% confidence interval, 81 87) by verbal history NOTE. SIA, supplemental immunization activity. a Administrative coverage, calculated by dividing the total number of doses administered to children in the target age group by the census-estimated number of children in that age group. Coverage.100% is due to an underestimation of the target population or vaccination of children from other areas. b 2006 National Measles-Malaria Campaign Report, 2006, performed by the Sierra Leone Ministry of Health and Sanitation [13]. c Long-lasting insecticidal nets. d Yellow Fever and Measles Post-Campaign Immunization Coverage Survey in Sierra Leone, 29 December January 2010 performed by the World Health Organization, Statistics Sierra Leone, and Expanded Programme of Immunizations, Sierra Leone [15].

4 implementation of the measles mortality reduction strategy, estimated MCV1 coverage increased from 37% in 2000 to 66% in 2008, and reported cases decreased from 3575 measles cases in 2000 to a total of 1916 cases during (ranging from 0 cases in 2007 to 649 cases in 2001) (Figure 1). The average estimated annual measles incidence decreased by 95%, from 10.0 cases per 100,000 population during to 0.5 cases per 100,000 population during [16]. However, on 1 November 2009 and immediately prior to the implementation of a nationwide follow-up SIA on 24 November, a laboratory-confirmed measles outbreak began, with initial cases reported from Bombali, Bo (the country s second largest city), Koinadugu, and surrounding districts (Port Loko, Kenema, and Tonkolili). In response to the measles outbreak, an outbreak investigation was conducted jointly by the Sierra Leone Ministry of Health and Sanitation, WHO, UNICEF, and the Centers for Disease Control and Prevention (CDC). To identify the likely cause of the outbreak and to describe measles epidemiology, we reviewed measles vaccination coverage data through routine service delivery and SIAs and analyzed measles case-based surveillance data. We report the findings of the outbreak investigation along with recommendations for preventing potential measles outbreaks in the future. METHODS Description of the Outbreak A suspected measles case was defined as a generalized maculopapular rash and fever and at least one of the following: cough, coryza (runny nose), or conjunctivitis in any person from 1 November 2009 through 13 July 2010 [17]. Surveillance officers used individual case investigation forms to collect data on suspected cases (age, sex, address, number of measles vaccine doses received, and data of last measles vaccination) and entered these into a case-based database. An attempt was made to collect serum samples within 30 days of rash onset for laboratory testing; confirmation was made by detection of measles immunoglobulin (Ig) M antibody at the WHO regional reference laboratory at the Pasteur Institute in Abidjan, Ivory Coast, using a standard commercial indirect enzyme-linked immunosorbent assay (ELISA) (Enzygnost for IgM; Siemens) [18]. Following the WHO AFR measles surveillance guidelines [17], laboratory confirmation of cases was discontinued after the outbreak was confirmed as measles, and additional cases were line-listed by district health officers. Line-listed cases met the suspected measles case definition and were confirmed by an epidemiological link established for cases that did not have a specimen collected for laboratory testing (ie, contact with a laboratoryconfirmed case who had rash onset within the preceding 30 days or lived in the same or adjacent district with a laboratoryconfirmed case) [17]. Epidemiologic and laboratory data were entered in the national measles case-based surveillance database. An outbreak of measles was defined as R3 laboratory-confirmed measles cases in a health facility or district in 1 month [17]. Vaccination coverage through routine services or SIAs was calculated using the administrative method by dividing the total number of doses administered to children in the target age group by the number of children in the target age group according to the 2004 census. An outbreak-related case was defined as a confirmed measles case with date of rash onset during the period 1 November July Agespecific and district-specific attack rates were calculated by dividing the number of confirmed measles cases in each age group or district by the total population in that age group or district according to the 2004 census. Oral fluid specimens were collected from a convenience sample of 14 individuals with laboratory-confirmed measles cases who were identified during site visits to health centers in Western Area, Bonthe, and Port Loko districts to determine the measles virus genotype. The specimens were shipped to the WHO regional measles reference laboratory at the National Institute for Communicable Diseases (NICD) in Johannesburg, South Africa. Standard real-time reverse-transcription polymerase chain reaction (RT-PCR) was used to amplify a region of the measles genome for sequence analysis and genotyping. Specific complimentary DNA of measles virus nucleoprotein was synthesized by RT-PCR and aligned, and phylogenetic trees were constructed using standardized methods [19]. All new sequences were submitted to GenBank (National Institutes of Health). Vaccine Effectiveness Vaccine effectiveness (VE) was calculated using the screening method with the formula VE51-[(PCV/(1-PCV))*((1-PPV)/ PPV)], where PCV refers to the proportion of cases vaccinated and PPV refers to the proportion of the population vaccinated [20, 21]. The analysis was restricted to cases in individuals aged months at the time of rash onset; the estimated proportion of the population vaccinated during the period was based on WHO and UNICEF MCV1 coverage estimates. Susceptibility Profile A spreadsheet-based formula was used to calculate the estimated number of susceptible individuals for each birth cohort in Sierra Leone during the 30 years prior to the start of the outbreak [22]. The surviving birth cohort for each year was calculated by applying an annual population growth rate (1.8%) estimated from the 2008 Demographic and Health Survey [23] to the 2004 census data. The estimated number of individuals susceptible to measles by age in November 2009 was calculated based on Sierra Leone EPI MCV1 administrative coverage estimates during and the WHO and UNICEF coverage estimates for each year Sierra Leone Measles Outbreak, d JID 2011:204 (Suppl 1) d S263

5 Figure 2. Number of confirmed cases of measles by week of rash onset (n 5 970), Sierra Leone, 1 October July *Cases reported to the Sierra Leone Ministry of Health and Sanitation (MOHS) measles case-based surveillance system and confirmed by the MOHS as either immunoglobulin (Ig) M antibody positive or epidemiologically linked to an IgM antibody positive case. Supplemental immunization activity (SIA). during [14], measles SIA administrative coverage for 2003 and 2006, and an estimate from a cluster survey for the 2009 measles SIA vaccination coverage by verbal history. For the susceptibility profile calculation, the following assumptions were used: VE was 85% through routine services delivery and 90% through SIAs, infants,6 months of age were protected by maternal antibodies, unvaccinated children 6 months 14 years of age were considered to be susceptible, half of unimmunized individuals years of age were protected by natural immunity, and adults R20 years of age were no longer susceptible [22]. We also assumed that the likelihood of vaccination at each opportunity was independent (ie, those covered by routine services had equal chances to be vaccinated by SIAs) and, therefore, multiplied the remaining susceptible individuals after routine immunization by the SIA coverage in that year. To control for the likely inflation of SIA administrative coverage, we subtracted the 17% difference between the reported coverage by verbal history in the 2009 SIA cluster survey (84%) and reported administrative coverage (101%) after the 2009 SIA from the administrative coverage reported in the 2006 and 2003 SIAs. Data Analysis Data analysis was performed using Excel (Microsoft Corporation); Epi Info for Windows, version (CDC); and SAS, version 9.2 (SAS Institute). Correlations of 2 continuous variables were quantified using Spearman s rank order correlation coefficient. RESULTS Description of Outbreak and Measles Supplemental Immunization Activities From 1 January through 31 October 2009, 5 laboratoryconfirmed measles cases were detected (during epidemiologic weeks 6 13). From 1 20 November 2009 (epidemiologic weeks 44 46), a cluster of laboratory-confirmed measles cases was detected in Bombali, Bo, and Koinadugu (Figure 2). The planned nationwide measles SIA targeting children 9 months to 4 years of age was implemented November. National administrative coverage was reported as 101% (ranging from 99% in Tonkolili to 108% in Western Area Urban Districts), whereas the estimated national coverage by cluster survey was 84% (ranging from 71% in Tonkolili to 100% in Kailahun) (Table 2). Bo, with the highest attack rate (49 cases per 100,000 population) was not selected for the cluster survey, although it had reported low routine coverage from 2002 through 2009 (69%), whereas Bombali (34 cases per 100,000 population) had 83% SIA coverage by cluster survey. The 3 districts with the highest SIA coverage by cluster survey, Kambia (92%), Kailahun (100%), and Western Area Rural (91%), had the fewest reported measles cases (range, 0 4 cases) during the outbreak. Although not statistically significant, districts with higher estimated campaign coverage tended to have lower measles attack rates (Spearman correlation coefficient ). S264 d JID 2011:204 (Suppl 1) d Sugerman et al

6 Table 2. Measles Cumulative Attack Rates and Measles Vaccine Coverage by District of Residence, Sierra Leone, November 1, 2009 July 13, SIA coverage District Population Routine coverage a,% Cluster survey b, % Admin, % Cases c Attack rate, cases per 100,000 population Bo 769, Bombali 689, Koinadugu 456, Kenema 844, Tonkolili 593, Port Loko 770, Western area urban 764, Bonthe 231, Kono 579, Pujehun 387, Moyamba 426, Kambia 461, Kailahun 616, Western area rural 169, Total 8,474, NOTE. SIA, supplemental immunization activity. a Mean measles-containing vaccine coverage reported by Expanded Programme of Immunizations, Sierra Leone, from b Yellow Fever and Measles Post-Campaign Immunization Coverage Survey in Sierra Leone, 12/29/2009-1/30/2010 performed by the World Health Organization, Statistics Sierra Leone, and EPI Sierra Leone [15]. c Confirmed measles cases reported to the Sierra Leone Ministry of Health and Sanitation (MOHS) measles case-based surveillance system and confirmed by the MOHS as either immunoglobulin (Ig) M antibody positive or epidemiologically linked to an IgM antibody positive case. During weeks of 2009, 31 additional cases were reported (Figure 2); of these, 75% were in individuals aged 9 months to 4 years of age who were therefore eligible for vaccination during the SIA. Starting in week 1 of 2010, weekly case counts sharply increased, peaked at 92 cases during week 7 in February 2010, and gradually decreased to 26 cases in week 28 during July From 1 November 2009 to 13 July 2010, a total of 1094 confirmed measles cases were reported, and 12 of the 14 districts had R3 laboratory-confirmed measles cases. Of the 1094 outbreak cases, 970 (89%) included day of rash onset, and 1083 (99%) had information on age. Among these cases, 292 (27%) were in individuals,1 year of age and 424 (39 %) were in individuals 1 4 years of age; age-specific attack rates were highest (56.4 cases per 100,000 population) among infants 6 8 months of age and those 9 11 months of age (40.6 cases per 100,000 population) (Table 3). Of the 1094 cases (100%) with information on district, 380 (35%) were from Bo, 233 (21%) were from Bombali, and 126 (12%) were from Koinadugu; these 3 districts had the highest attack rates in the country. Of the 1094 outbreak cases, 473 (43%) had information on vaccination status and 462 (42%) had information on age and vaccination status. Among those individuals with information on vaccination Table 3. Measles Attack Rate by Age Category, Sierra Leone, 1 November July 2010 Age group No. of cases a Percentage of cases Attack rate, cases per 100,000 population Percentage vaccinated b 0 5 Months Months Months Years Years Years R15 Years Total 1, NOTE. SIA, supplemental immunization activity. a Confirmed measles cases reported to the Sierra Leone Ministry of Health and Sanitation (MOHS) measles case-based surveillance system and confirmed by the MOHS as either immunoglobulin (Ig) M antibody positive or epidemiologically linked to an IgM antibody positive case (10 cases were missing age data). b The number of cases reporting prior vaccination through either routine services or SIA divided by all cases. Sierra Leone Measles Outbreak, d JID 2011:204 (Suppl 1) d S265

7 Figure 3. Confirmed measles cases (n ) and estimated susceptible individuals (n 5 268,900) by age, Sierra Leone, November 2009 July *Supplementation immunization activity (SIA); the bracket indicates the age group targeted by the SIA. status, 222 (47%) were unvaccinated. A majority of cases in all age groups were either unvaccinated or had unknown or undocumented vaccination status (Table 3). Among the outbreak cases, 9 deaths were reported, including 6 from the Médecins Sans Frontières Gondama Hospital in Bo district. Among the 9 reported measles deaths, the mean age of case-patients was 11 months (range, 5 months to 3 years), and 4 (44%) of the deaths occurred in individuals with reported vaccination. Virus sequence information was obtained from 11 (73%) of the oral fluid samples. All 11 sequences represented by GenBank accession numbers HQ HQ were genotype B3 and were closely related to sequences obtained from measles cases detected in West Africa during Measles genotype B3 is the most frequently detected measles genotype in sub-saharan Africa [24]. fluctuation during the civil war (Figure 1). All birth cohorts since 1990 were targeted by at least 1 measles SIA, and children 3 4 years and 6 7 years of age were targeted by 2 SIAs. At the start of the outbreak, in November 2009, children who were 3 years of age were targeted by both the 2006 and 2009 SIAs, and children aged 6 years were targeted by the 2003 and 2006 SIAs (Figure 3). Infants 6 8 months of age were too young to be eligible for the 2009 SIA and were considered to be susceptible to measles infection due to waning maternal antibodies. From the catch-up SIA in November 2003 until just before the second follow-up SIA in November 2009, the estimated cumulative number of susceptible individuals was times the surviving annual birth cohort, with,1000 total reported measles cases during this timeframe. The 2009 SIA reduced the estimated number of susceptible individuals by 1 annual birth cohort (Table 4). Vaccine Effectiveness Among the 424 case-patients who were months of age, 156 (36%) were vaccinated. VE among children months of age was estimated to be 74%, assuming the PPV to be 68% and assuming that all case-patients with missing or unknown vaccination status were unvaccinated. Susceptibility Profile The WHO and UNICEF estimates of MCV1 coverage were %75% before 1995 and ranged from 28% to 79% during , with DISCUSSION The measles outbreak was the largest in Sierra Leone in the past decade and was likely caused by an accumulation of individuals who were susceptible to measles, primarily due to nonvaccination. This outbreak was much smaller than other outbreaks that have occurred throughout sub-saharan Africa during [7]; the November 2009 SIA was estimated to have reduced the total number of susceptible individuals by 1 S266 d JID 2011:204 (Suppl 1) d Sugerman et al

8 Table 4. Leone Susceptibility Profile Showing Estimated Number of Individuals Susceptible to Measles by Age in November 2009, Sierra Age in Nov 2009 Birth year Routine coverage in birth year a,% SIA coverage, % b annual birth cohort, likely blunting the magnitude of the outbreak. The wide age distribution of cases (patient age, 0 44 years) indicates that the accumulation of susceptible individuals occurred over several years; however, the large proportion of case-patients,5 years of age (66%) suggests recent gaps in the quality of SIA implementation and routine vaccination services. Routine measles vaccination coverage remains suboptimal, and national estimates have never reached.80%. Despite reported administrative coverage.100% for the 2009 measles SIA, survey coverage by verbal history was only 84%, and 48% of outbreak cases occurred among children in the age group that was targeted by the SIA. These findings highlight the risk of an outbreak due to the buildup of susceptible individuals among the,5-year-old population from vaccine ineligibility (age of 6 8 months), low vaccination coverage through routine services (60% 71%), and vaccination coverage below 95% during the 2006 and 2009 SIAs. With 2009 SIA No. of susceptible individuals (31000) c Cumulative no. of susceptible individuals (31000) No. of susceptible individuals (31000) d Without 2009 SIA Cumulative no. of susceptible individuals (31000) 0 5 Months Months Months Year Years Years Years Years Years Years Years Years Years Years Years Years Years Years Years Years Years Years NOTE. SIA, supplemental immunization activity. a World Health Organization and United Nations Children s Fund estimates of coverage with the first dose of measles-containing vaccine. For 4 years with missing information ( , 1989), the estimate from the previous year with an estimate was used. b For the 2009 SIAs, the coverage estimate by cluster survey was used [15]. For the 2003 and 2006 SIAs, administrative coverage was calculated by dividingthe total number of doses administered to children in the target age group by the census-estimated number of children in that age group and subtracting the difference (17%) between the estimated coverage by verbal history in the 2009 SIA cluster survey and the 2009 reported administrative coverage. c Children,6 months of age or R20 years of age were considered to be 100% immune. In 2009, the WHO released revised measles outbreak response guidelines that include recommendations for conducting selective and nonselective outbreak response immunization (ORI) [25]. Selective ORI includes providing measles vaccination through routine service sites for all unvaccinated children 6 59 months of age or for a specific age group based on the measles epidemiology and is recommended when a measles outbreak is suspected. Nonselective ORI refers to a mass vaccination campaign targeting all children in a specific age group and geographic area and is recommended when a measles outbreak is confirmed and adequate resources are available. The measles outbreak started 3 weeks before implementation of the planned nationwide measles SIA, at the end of the low season for measles virus transmission. The SIA may have served as a nonselective ORI and potentially prevented additional measles cases; however, cases continued to be reported after the SIA, with high attack rates among children,5 years of Sierra Leone Measles Outbreak, d JID 2011:204 (Suppl 1) d S267

9 age. Children,9 months of age were too young to be eligible for routine measles vaccination or for the 2009 SIA, and maternal antibodies may have waned, placing them at risk in the face of an outbreak. Some of the 132 cases and 3 deaths that occurred among infants 6 8 months of age might have been prevented if WHO-recommended ORI strategies were implemented, including providing measles vaccination through routine service sites starting at 6 months of age (selective ORI) or through targeted campaigns in affected districts (nonselective ORI) if resources were available. However, measles vaccine efficacy is greatly reduced when given to infants,9 months of age [26, 27]; therefore, it is recommended by the WHO that infants who receive measles vaccination before the age of 9 months must be revaccinated after the age of 9 months with at least a 1-month interval between doses [25]. If measles vaccination coverage through routine services and SIAs was higher prior to the outbreak, then infants,9 months of age may have been otherwise protected by herd immunity. The outbreak in Sierra Leone began at the tail end of the rainy season, the low-transmission period for measles virus (August December) and peaked during the high-transmission that occurs during the dry season (January July); a similar pattern has been observed in previous large measles outbreaks in West Africa [28 30]. The WHO AFR SIA field guide suggests that SIA implementation should be during the low-transmission season for measles virus [17], although it does not specify early, middle, or late. Considering that large outbreaks may start during the end of the low transmission season for measles virus leading to established circulation in the community, the optimal timing for SIAs in West Africa may be during August September and prior to the end of the low-transmission season. It is postulated that countries are at risk of a measles outbreak when the number of susceptible children of preschool age exceeds the size of an annual birth cohort [31]. In this outbreak, which occurred 3 weeks before the 2009 SIA, the estimated number of susceptible children % 5 years of age was 1.6 times greater than the annual birth cohort, although, following the 2009 SIA, the estimated number of susceptible individuals decreased to nearly half the annual birth cohort, likely reducing the overall number of cases. Forty-four cases (6%) were in patients who were.21 years of age and who were not reached by the 2006 follow-up SIA or 2003 catch-up SIA. A reduction in cases was noted from children targeted by one SIA (children 2 and 5 years of age) to those reached by overlapping SIAs (children 3 4 years and 6 7 years of age). These findings suggest inadequate coverage during each SIA and demonstrate the benefits of expanding the age range for follow-up SIAs. Finally, based on the susceptibility profile, Sierra Leone was at risk for an outbreak from 2003 until the start of the 2009 SIA. To prevent measles outbreaks, nationwide SIAs should be implemented in countries before a build-up of a high number of susceptible individuals occurs, especially when measles virus continues to circulate in neighboring countries or when large measles outbreaks are occurring throughout the region. The findings of this investigation should be considered in light of several limitations. The extent of community exposure to measles virus and underreporting of measles cases from health facilities, including those cases in individuals who never sought treatment, was unknown; our findings are only representative of cases reported to the surveillance system. The majority of outbreak-related cases were in individuals with missing vaccination status, which prevented a more accurate VE screening value. A higher estimate of PPV would result in higher VE and would support our conclusion that vaccine failure did not play a substantial role in the outbreak. The 2003 and 2006 SIA coverage estimates were based on administratively collected data and were not validated by a population-based probability survey; if either the numerator or denominator figures were inaccurate, then coverage may have been incorrect, as evidenced by administratively reported SIA coverage.100%, thereby artificially lowering the estimated number of susceptible individuals. Despite these limitations, the Sierra Leone measles outbreak investigation illustrated the benefit of measles SIAs, while highlighting the challenges that low vaccination coverage presents to measles control. Ongoing suboptimal routine measles vaccination coverage in Sierra Leone will lead to periodic measles outbreaks in the future. To ensure that all children receive the recommended 2 doses of measles vaccine, improved measles vaccination coverage through both routine services and SIAs is needed. Follow-up SIAs must attain high (.95%) vaccination coverage in all districts and should be completed prior to the end of the low-transmission season. Funding World Health Organization (to A. F., M. N., and T. B.), United Nations Children s Fund (to M. -T. G.), the Ministry of Health and Sanitation of Sierra Leone (to A. W.), and the Centers for Disease Control and Prevention (to D. E. S. and J. L. G.). Acknowledgments We thank the Expanded Programme of Immunization staff in Sierra Leone; Chantal Akoua-Koffi, Hervé Kadjo, Marius Adagba N Takpé, Jeannie Gbahouo, Bertin Kouakou, and Aboulaye Ouattara from the Pasteur Institute in Abidjan, Ivory Coast; Sheilagh Smit, from National Institute for Communicable Diseases in Johannesburg, South Africa; Likang Xu (Centers for Disease Control and Prevention), for database management and statistical assistance; and Balcha Masresha (World Health Organization African Regional Office), for his guidance and support during this investigation. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the MOHS of Sierra Leone. References 1. World Health Organization. WHA In: Handbook of resolutions and decisions of the World Health Assembly and the Executive Board, 2nd ed. Vol III.( ). Geneva, Switzerland: WHO, S268 d JID 2011:204 (Suppl 1) d Sugerman et al

10 pdf. Accessed 20 November United National Children s Fund. Plan of action for implementing the World declaration on the survival, protection, and development of children in the 1990s. New York, NY: UNICEF, unicef.org/wsc/plan.htm. Accessed 20 November Centers for Disease Control and Prevention. Progress in global measles control and mortality reduction, MMWR Morb Mortal Wkly Rep 2007; 56: World Health Organization (WHO) African Regional Office. Plan of action for measles mortality reduction in the African Region, Harare, Zimbabwe: WHO African Regional Office, Centers for Disease Control and Prevention. Measles mortality reduction-west Africa, MMWR Morb Mortal Wkly Rep 2008; 55: Bilous J, Eggers R, Gasse F, et al. A new global immunization vision and strategy. Lancet 2006; 367: Masresha B, Fall A, Eshetu M, et al. Measles Mortality Reduction and Pre-Elimination in the African Region, J Infect Dis doi: /infdis/JIR United Nations Population Division. World population prospects: the 2008 revision Accessed 18 March Human Rights Watch: Sierra Leone Getting away with murder, mutilation, rape. New testimony from Sierra Leone. 1999, 11(3A). Accessed 20 November United Nations Human Development Index. statistics/. Accessed 20 November United Nations Children s Fund State of the World s Children Maternal and newborn health. SOWC09-FullReport-EN.pdf. Accessed 20 November Senessie C, Gage GN, von Elm E. Delays in childhood immunization in a conflict area: a study from Sierra Leone during civil war. Confl Health 2007; 9: National Measles-Malaria Campaign Report, 2006: Sierra Leone Ministry of Health and Sanitation. FinalM-MCampaign.pdf. Accessed 20 November World Health Organization (WHO). WHO vaccine-preventable diseases monitoring system 2010 global summary. immunization_monitoring/en/globalsummary/countryprofileresult.cfm. Accessed 20 November Yellow fever and measles post-campaign immunization coverage survey in Sierra Leone, 12/29/2009 1/30/2010: World Health Organization, Statistics Sierra Leone, and EPI Sierra Leone. 16. Centers for Disease Control and Prevention. Progress toward measles control African Region, MMWR Morb Mortal Wkly Rep 2009; 58: World Health Organization African Regional Office. Measles surveillance guidelines, Jan php?option=com_content&view=article&id=2383:publicationsmeasles&catid=1974&itemid=2734. Accessed 16 March World Health Organization. Manual for the laboratory diagnosis of measles and rubella virus infection, 2nd ed. who.int/immunization_monitoring/labmanualfinal.pdf. Accessed 20 November Kremer JR, Nkwembe E, Bola Oyefolu AO, et al. Measles virus strain diversity, Nigeria and Democratic Republic of the Congo. Emerg Infect Dis. 2010; 16: McMorrow ML, Gebremedhin G, van den Heever J, et al. Measles outbreak in South Africa, S Afr Med J. 2009; 99: Farrington CP. Estimation of vaccine effectiveness using the screening method. Int J Epidemiol. 1993; 22: Goodson JL, Perry RT, Mach O, et al. Measles outbreak in Tanzania, Vaccine 2010; 28: Sierra Leone standard demographic and health survey &SrvyTp5ctry&cn5Sierra%20Leone. Accessed 20 November Rota PA, Featherstone DA, Bellini WJ. Molecular epidemiology of measles virus. Curr Top Micribiol Immunol 2009; 330: World Health Organization. Response to measles outbreaks in measles mortality reduction strategies; /WHO_IVB_09.03_eng.pdf. Accessed 20 November Kurubi J, Vince J, Ripa P, Tefuarani N, Riddell M, Duke T. Immune response to measles vaccine in 6 month old infants in Papua New Guinea. Trop Med Int Health 2009; 14: Caceres VM, Strebel PM, Sutter RW. Factors determining prevalence of maternal antibody to MV throughout infancy: a review. Clin Infect Dis 2000; 31: Guyer B, McBean AM. The epidemiology and control of measles in Yaoundé, Cameroun, Int J Epidemiol 1981; 10: Cummings DA, Moss WJ, Long K, et al. Improved measles surveillance in Cameroon reveals two major dynamic patterns of incidence. Int J Infect Dis. 2006; 10: Ferrari M, Grais R, Bharti N, et al. The dynamics of measles in sub- Saharan Africa. Nature 2008; 451: World Health Organization (WHO). Measles vaccines: WHO position paper. Wkly Epidemiol Rec 2009; 84: Sierra Leone Measles Outbreak, d JID 2011:204 (Suppl 1) d S269