Open Access Research An evaluation of the 2012 measles mass vaccination campaign in Guinea Jean Gerard Tatou Doumtsop2,3,&, Emmanuel Roland Malano1, Ibrahima Telly Diallo1, Camara Sirimah1 1Ministry of Health and Public Hygiene, Guinea, 2WHO country office, Guinea, 3Ministry of Public health Cameroon &Corresponding author: Jean Gerard Tatou Doumtsop, WHO country office, Guinea, Ministry of Public health Cameroon Key words: Measles, immunization, evaluation, vaccination, coverage Received: 18/02/2013 - Accepted: 02/09/2013 - Published: 08/01/2014 Abstract Introduction: To estimate the post-campaign level of measles vaccination coverage in Guinea. Methods: Interview of parents and observation of measles vaccination cards of children aged 9 to 59 months during the mass measles campaign. A nationwide cluster randomized sample under health District stratification. Results: 64.2% (95%CI = 60.9% to 67.4%) of children were vaccinated and had their measles vaccination card. With respect to card and history 90.5% (95%CI = 88.3% to 92.3%) were vaccinated. The estimation was found to be between 72.7% and 81.9%. Coverage with card increased from 55.5% to 79.30% with the level of education of parents but that was not statistically significant, (X2(trend) =3.087 P= 0.07). However coverage with card significantly increased with profession from 55.1% for farmers followed by 59.2% for other manual workers to 73.8% for sellers, ending by 74.5% for settled technicians(X2 (trend) =12.16 P= 0.0005). For unvaccinated children, lack of information accounted for the main reason(37.03%) followed by parents' occupation(23.45%), parents' sickness (8.6%), children's sickness (4.9%) and others including vaccinators absent in the post or parents' belief that it was a door to door campaign. Conclusion: The mass measles vaccination campaign achieved an approximate coverage of 75%. Although not enough for effective control of measles, it has covered an important gap left over by the routine immunization coverage 42%. Appropriate measures are needed to improve coverage in routine immunization and specific actions should be taken to target farmers and other manual workers' families but also uneducated groups for both routine immunization and mass campaigns. Pan African Medical Journal. 2014; 17:4 doi:10.11604/pamj.2014.17.4.2475 This article is available online at: http://www.panafrican-med-journal.com/content/article/17/4/full © Jean Gerard Tatou Doumtsop et al. The Pan African Medical Journal - ISSN 1937-8688. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Pan African Medical Journal – ISSN: 1937- 8688 (www.panafrican-med-journal.com) Published in partnership with the African Field Epidemiology Network (AFENET). (www.afenet.net) Page number not for citation purposes 1 Introduction 840 children. In each household detailed explanation of the study objective was clarified to the family representative to have their informed consent. Data were recorded and analyzed using Epi-info Measles still kills more than half of children who die annually from software. Med-calc software was used to obtain the chi-square for vaccine preventable-diseases. Among children surviving from trend and Health map software was used for mapping of coverage measles, up to 10% will suffer of disabilities such as blindness, distribution per region. deafness and irreversible brain damage [1]. All six WHO regions have committed to measles elimination and five regions have set target dates. The America WHO Region achieved the goal in 2002, Results the western pacific Region aims at eliminating measles by the end of 2012 and the European and Eastern Mediterranean regions are accelerating their measles control activities in order to eliminate 866 households were investigated among which 853 (98.5%) with measles by 2015. In 2011 countries in the African region took on at least one target were selected for analysis. 549children were the goal to eliminate measles by 2020, and in 2010 the South-East vaccinated and had their vaccination cards. The corresponding Region adopted a resolution urging countries to mobilize resources vaccination coverage was 64.2% (95%CI = 60.9% to 67.4%). to support the elimination of measles, the target date for which was These Children were aged 11 to 61 months and 50.4% were male. under discussion [2]. In 2008 the world health assembly endorsed a Mean age was 32.46 months with standard deviation of target of 90% reduction in measles mortality by 2010 compared 14.03months. 226 children not having their vaccination card were with 2000. Estimated global measles mortality decreased 74% from reported as having a history of vaccination during the campaign 535300 deaths in 2000 to 139300 in 2010[3]. Accordingly, the from their parents' declaration. Then a total of 775 children were World Health Assembly committed again in May 2010 to endorse a vaccinated on the basis of vaccination cards and vaccination history series of interim measles control targets for 2015 which include during the campaign and the corresponding vaccination coverage exceeding 90% coverage with first dose measles containing vaccine was 90.5% (95%CI = 88.3% to 92.3%) Parents were aged 14 to 82 (MCV1) nationally, exceeding 80% vaccination coverage in every years and 71.7% were children's mothers. Single parent district, reducing annual measles incidence to < 5cases per million, represented 12.8%. Mean age of parent was 31.53years with maintaining that level and reducing measles mortality by more than standard deviation of 11.35 Years. Parents were mostly uneducated, 95% compare with 2000 estimates [4]. This commitment was set in 55.2% and housewives, 51.8% (Table 1). 81 children, thus 9.5% line to achieve the Millennium development goals4(MDG4) which (95%CI= 7.7% 11.7%) had no vaccination card and no history of aim to reduce the overall number of death among children by two- vaccination during the campaign. Their mean was 27.48months with thirds between 1990 and 2015[5].The proportion of children standard deviation of 12.13months. 48.1% were men. For reasons vaccinated against measles was adopted as an indicator to measure of non vaccination, lack of information accounted for 37.3%, progress towards this MDG4 and the 2020 measles elimination parents' occupation for 23.45%, parents' sickness for 8.6%, objective in Africa region. The highly infectious nature of measles children's sickness for 4.9%, vaccinators absent in the post during virus requires maintenance of very high levels of population the visit for 3.7%, believes that the campaign was a door to door immunity. Supplementary Immunization Activities(SIA) conducted campaign for 3.7% and vaccination post far to reach for 2.46% every two, three or four years depending on the quality of routine (Table 2). Although there was an increase trend of coverage with immunization currently play an important role in protecting children vaccination card and the level of education, our data did not provide in countries unable to achieve and maintain high and homogenous a strong relationship as the Chi square for trend was not significant vaccination coverage through routine immunization systems[6]. In (X2 trend = 3.08 P= 0.07'>X2 trend = 3.08 P= 0.07) (Figure 1) Guinea, the objective for the 2012 measles SIA was set at 95% However there was a significant increase trend of coverage and vaccination coverage and the activities was implemented from June parents' profession (X2(trend) =12.16 P= 0.0005'>X2(trend) =12.16 29th to July 5th targeting estimated 2209623 children aged 9 to 59 P= 0.0005) (Figure 2). Vaccination Coverage varied from one months. Assessment was carried out 2 months later. region to another and we noted a constant overestimation of administrative coverage in all except Boke region (Figure 3). Methods Discussion A nationwide cross-sectional survey conducted 2months after measles national immunization days. Population was divided into Let us admit that the "estimated vaccination coverage" (EVC) is the homogeneous clusters. The list of clusters was provided by the cut point to be used when talking about measles vaccination national institute of demography. For each Health District one coverage in Guinea in 2012. The EVC lay between the proven cluster was selected using a computer random process but two vaccination coverage with card and the unsure vaccination coverage clusters for the two biggest Health Districts. Then a total of 40 which encompassed children with vaccination cards and those with clusters from the 38 Districts of the Country were randomly history of vaccination. Most often evaluation of measles campaigns selected. 80 investigators were trained including role play to carry are done immediately after to improve on recall bias. 2 months out data collection with a structured questionnaire. For each cluster, length time can be enough to consider the effect of a recall bias, the investigators started in the middle and allocated numbers on especially when children have gotten additional routine vaccination folded papers to each of the four directions. Papers were tossed and during this length time confusion is likely to be possible. In fact the one selected gave the direction to follow. The first household on children of unsure parents became randomly classified easily as the direction was the starting one. In the household, when there vaccinated or not vaccinated depending on the subjectivity of the were one or more target children, only one was selected using the investigator of the household. Vaccination Coverage with history same random process as for the selection of the direction to follow. may overestimate if investigators are too sensitive or underestimate The Child was observed and his immunization card was requested. A if they are too specific but these are unknown indicators. In structured interview was then conducted to the parent in charge of assessing the validity of interview information in estimating him. At the end of the process investigators fully thanked the family community immunization levels in USA, Comstock observe among and moved to the nearest household until 18 to 22 children were 494 people of all age that 99.2% were serologically immune selected per cluster in order to have at least the expected sample of whereas 83.4% were immune according to interview information. Page number not for citation purposes 2 He further compared the validity of history per place of residence coverage was still low and a population-based cluster survey (rural or urban), annual income of head of households and age reported 86% coverage [15]. Lack of information has also been group and in all instances the proportion of immune persons was reported to be the main cause of non vaccination during mass somewhat understated by interview results varying from 81.7% to immunization campaign in many other countries like Burkina Faso, 87.6% as compare to the 99.2%. However, although comparison Congo and South Africa. For the perspective of measles elimination for measles was inconclusive because nearly every participant was in 2020, an important gap is still to be met for measles SIA serologically immune, interview information was poorly correlated campaign in Guinea in term of correlation between administrative with the serological findings for mumps and poliomyelitis and coverage, 103% and the coverage of the survey post campaign, understated for rubella and tetanus [7]. A study in hillsborough 75%. Experience of East European countries currently experiencing county,USA assessing correlation between history of either measles measles elimination proves that one good indicator can the good or vaccination and serologic immunity shown that positive history match between administrative coverage and that of the LQA and positive antibody were concordant on 317 over 374 ( sensitivity coverage post campaign [16]. (Armenia, 96.8% &95.8%, Tajikistan of 84.8%) and negative history and negative antibody were 97.8% &96.6%, Turkmenistan 96% &97.6%) comparatively, an concordant on 14 over 30(specificity of 47%) [8]. Applying these alternative for better handling of measles control and moving indicators to our unsure 226 children vaccinated based only on towards elimination would be: 1) Implementation of follow-up history, we can estimate that 226*0.85*0.47= 90 are effectively campaign given the very low capacity to provide second opportunity vaccinated and 136 are not effectively vaccinated, thus an through routine immunization, 2) Introduction of LQA coverage "estimated coverage of (549+90)/853 = 75%". A similar analysis surveys post-campaign. 3) Immunization efforts specifically was carried on in a comparable country in terms of anthropology, targeting underserved groups (farmers and other manual workers, culture and socioeconomic status, Soudan. Variation rate between non classic educated and other non-educated). vaccination coverage with vaccination card and vaccination coverage with vaccination card plus history of vaccination was 23%. Our data showed a variation rate of 26%. The study further Conclusion concluded that illiterate mothers had remarkably good recollections of their children measles vaccination status and therefore accurate The mass measles vaccination campaign achieved an approximate estimation could rely only on mothers' reports irrespective of the coverage of 75%. Although not enough for effective control of assumptions made about mothers who were unsure about children's measles, it has covered an important gap left over by the routine vaccinations status. For example, if children of mothers who were immunization coverage 42%. Appropriate measures are needed to unsure are assumed not to have received measles vaccine, the improve coverage in routine immunization and specific actions sensitivity of mother's reports would be 87% and the specificity should be taken to target farmers and other manual workers' 79% conversely, if such women are assumed to have vaccinated families but also uneducated groups for both routine immunization their children the sensitivity of mother's would then be 95% and the and mass campaigns. specificity would drop to 70% [9]. Applying these indicators to our unsure parents will provide the validity of the interview (Table 3). The "estimated vaccination coverage" based on the validity of the Competing Interests interview was between 72.7% and 81.94%, 75% is included in the interval and so can be a good approximation of the cut point for measles vaccination coverage in Guinea in 2012. Based on this cut The authors declare no competing interests. point which takes into account the validity of the interview The distribution of coverage per region was estimated (Table 4) and a mapping generated (Figure 4). Authors’ Contributions A catch up measles vaccination campaign is an opportunity to deliver booster and to catch up on children hard to reach due to Jean Gerard Tatou Doumtsop: study design, data collection and poor socioeconomic conditions or geographical reasons in measles supervision, data recording and analysis, edition of the final report, control country like Guinea. But when the trend of vaccination edition of this manuscript. Emmanuel Roland MALANO: Study coverage increases with the socioeconomic level (profession and design, data collection and supervision, data recording and analysis, education) during mass vaccination as it is usually the case during edition of the final report Ibrahima Telly Diallo: Study design, data routine immunization it suggests that efforts are still needed to collection and supervision, data recording and analysis, edition of improve on equity and the same who escaped routine immunization the final report. Sirimah Camara: Study design, data collection and are probably left over. Missing to cover the gap due to supervision, data recording and analysis, edition of the final report. socioeconomic discrepancy in the access to immunization is often All the authors have read and approved the final version of the observed during mass immunization [10] suggesting that specific manuscript. actions are still needed to make mass immunization an equal opportunity for every child. However a coverage of 75% can be satisfactory if we consider measles routine coverage that is 42% Tables and figures [11] (Figure 3, unpublished data, Ministry of health, February 2012) and the experience of other countries in the past (9-75% in Burkina Faso, 10-95% in Nigeria, 40-75% in Columbia) [12] though Table 1: Characteristics of parents not enough for effective control of measles. For example, the same Table 2: Reasons for non vaccinations experience was gotten in South Africa where a cluster survey Table 3: Validity of the interview results conducted immediately before and two months after the mass Table 4: distribution of coverage per region based on the validity of campaign shown an increase from 55% for routine coverage to 72% interview and the considered cut point for mass campaign coverage [12]. Measles campaigns have not Figure 1: Vaccination coverage with card according to parents’ always been satisfactory in low income countries and often haven't level of education reached the estimated target [13, 14]. Despite extending the Figure 2: Vaccination coverage with card according to parents’ duration of the campaign in Kabul by 7 days and sending external profession monitors to search door-to-door for missed children, reported Page number not for citation purposes 3 Figure 3: Comparison of routine and campaign measles vaccination coverage 10. Zuber Patrick LF et al. Vaccination de masse contre la rougeole Figure 4: Map of estimated vaccination Coverage per region of dans des zones urbaines du Burkina Faso, 1998. Bulletin de Guinea l'Organisation mondiale de la Sante. 2001 ; 79(4) : 296- 300. PubMed | Google Scholar References 11. Revue externe du programme élargi de vaccination de la Guinée. Février 2012. Guinée.WHO. PubMed | Google Scholar 1. Simons E, Ferrari M, Fricks J, Wannemuehler K, et al. Assessment of the 2010 global measles mortality reduction 12. Kessler S, Melendez D. speeding up child immunisation. World goal: results from a model of surveillance data. Lancet. 2012 Health Forum. 1987; 8(2): 216-220. PubMed | Google Jun 9;379(9832):2173-8. PubMed | Google Scholar Scholar 2. World Health Organization. Global measles and rubella 13. Maureen k et al. Evaluation of mass measles immunisation strategic plan 2012-2020. 201 campaign in a rapidly growing peri-urban area. South African http://www.who.int/immunization/newsroom/Measles_Rubella Medical Journal. 1989; 76(4): 157-159.PubMed | Google _StrategicPlan_2012_2020.pdf. Accessed 24 April 2012. Scholar 3. World Health Organization. Measles vaccines, WHO Position 14. Berry Dj, Yach D, Hennink Mhj. An evaluation of the national Paper. Weekly Epidemiological Record. 2009; 84(35):349- measles vaccination campaign in the new shanty areas of 360. PubMed | Google Scholar Khayelitsha. South African Medical Journal. 1991; 79(8): 433- 436.. PubMed | Google Scholar 4. World Health Organization. Global eradication of measles, report by the Secretariat: Sixty-third world health assembly. 25 15. Dadgar N et al. Implementation of a Mass Measles Campaign March 2010. Geneva. World Health Organisation. Google in Central Afghanistan, December 2001 to May 2002. The Scholar Journal of Infectious Diseases. 2003; 187(Suppl 1): S186- 90. PubMed | Google Scholar 5. United Nations. The millennium development report 2009. 2009. New York. United Nations. PubMed | Google Scholar 16. Khetsuriani N et al. Supplementary Immunization Activities to Achieve Measles Elimination: Experience of the European 6. Strebel PM et al. A World without Measles. J Infect Dis. 2011; Region. The Journal of Infectious Diseases. 2011; 204 (suppl 204 (suppl 1): S1-3. PubMed | Google Scholar 1): S343-52. PubMed | Google Scholar 7. Comstock et al. Validity of interview information in estimating 17. Henderson RH & Sundarset T. Cluster sampling to assess community immunization levels. Health Services Reports.1973; immunization coverage: a review experience with a simplified 88(8): 750-757. PubMed | Google Scholar method. Bulletin de l' Organisation mondiale de la Santé. 1982; 60(2): 253-260. PubMed | Google Scholar 8. Levitt LP et al. Determination of Measles Immunity after a Mass Immunization Campaign. Public Health Reports. 1970; 85(3): 261-265. PubMed | Google Scholar 9. Gareaballah ET & Loevinsohn BP. The accuracy of mother's reports about their children's vaccination status. Bulletin of the World Health Organisation. 1989;67(6): 669- 674. PubMed | Google Scholar Page number not for citation purposes 4 Table 1: Characteristics of parents Characteristics of parents n % Single Parents 109 12.8 Marital status Married 743 87.2 Mothers 611 71.7 Fathers 126 14.8 Relation with child Grand parents 58 6.8 brothers 57 6.6 Others (Coranic etc...) 27 3.2 No education 469 55.2 Education Primary education 173 20.4 Secondary education 151 17.8 Tertiary education 29 3.4 farmers 107 12.7 Others manual workers 49 5.8 unemployed 46 5.5 Profession housewife 436 51.8 Salaried servants 26 3.1 sellers 122 14.5 Settled technicians 55 6.5 Table 2: Reasons for non vaccinations Reasons for non vaccination n % Lack of information 30 37.03 parents absents or busied 19 23.45 Parents’ sickness 7 8.64 Children’s sickness 4 4.93 Vaccinators absent in the post 3 3.70 Believes to a door to door campaign 3 3.70 Vaccination post far to reach 2 2.46 Ignorance of the importance 2 2.46 Age not out of range 1 1.23 Unknown 10 12.34 Page number not for citation purposes 5 Table 3: Validity of the interview results Children of unsure parents are assumed Children of unsure parents are assumed to to haven’t been vaccinated have been vaccinated (Sensitivity 87% and specificity 79%) (Sensitivity 95% and specificity 70%) status Effectively vaccinated Not effectively Effectively vaccinated Not effectively vaccinated vaccinated Vaccinated with history 226-155=71 226*0.87*0.79= 155 226*0.95*0.70=150 226-150=76 (n=226) Vaccinated with card 549 0 549 0 (n=549) Effectively vaccinated 71+549= 620 155 150+549 = 699 76 Estimated vaccination 620/853*100= 72.7% 155/853*100=18.2% 699/853*100 = 81.94% 76/853*100 = 8.9% coverage(EVC) Table 4: distribution of coverage per region based on the validity of interview and the cut point REGION Number of Number with Number with Effectively Not Estimated “Estimated children proven history of vaccinated effectively number of vaccination investigated vaccination vaccination on history vaccinated vaccinated Coverage” card on history (%) BOKE 109 97 12 5 7 102 93 CONAKRY 151 113 29 12 17 125 83 FARANAH 85 48 30 12 18 60 71 KANKAN 111 84 21 8 13 92 83 KINDIA 104 59 36 14 22 73 71 LABE 105 67 26 10 16 77 74 MAMOU 61 19 33 13 20 32 53 NZEREKORE 127 62 39 16 23 78 61 TOTAL 853 549 226 90 136 639 75 Page number not for citation purposes 6 Figure 1: Vaccination coverage with card according to parents’ level of education Figure 2: Vaccination coverage with card according to parents’ profession Figure 3: Comparison of routine and campaign measles vaccination coverage Page number not for citation purposes 7 Figure 4: Map of estimated vaccination Coverage per region of Guinea Page number not for citation purposes 8