Scholarly article on topic 'Evaluation of response strategies against epidemics due to Neisseria meningitidis
 C in Niger'

Evaluation of response strategies against epidemics due to Neisseria meningitidis C in Niger Academic research paper on "Health sciences"

0
0
Share paper
Academic journal
Trop Med Int Health
OECD Field of science
Keywords
{""}

Academic research paper on topic "Evaluation of response strategies against epidemics due to Neisseria meningitidis C in Niger"

Article Type: Original Article

Evaluation of response strategies against epidemics due to Neisseria meningitidis C in Niger

Halima Boubacar Maïnassara1,2,3, Garba Idé Oumarou1, Bassira Issaka1, Ali Sidiki1, Issa Idi1, Jean-Paul Moulia Pelat1, Arnaud Fontanet2,4 and Judith E. Mueller2,5

1Centre de Recherche Médicale et Sanitaire, Niamey, Niger

2 Institut Pasteur, Emerging Diseases Epidemiology Unit, Paris, France

3Université Pierre et Marie Curie, Cellule Pasteur UPMC, Paris, France

4Conservatoire National des Arts et Métiers, Chaire Santé et Développement, Paris, France

5EHESP French School of Public Health, Sorbonne Paris Cité, Rennes, France

Abstract

Objective: To inform public health recommendations, we evaluated the effectiveness and efficiency of current and hypothetical surveillance and vaccine response strategies against Neisseria meningitidis C meningitis epidemics in 2015 in Niger.

Methods: We analyzed reports of suspected and confirmed cases of meningitis from the region of Dosso during 2014 and 2015. Based on a definition of epidemic signals, the effectiveness and efficiency of surveillance and vaccine response strategies were evaluated by calculating the number of potentially vaccine-preventable cases and number of vaccine doses needed per epidemic signal. Results: 4,763 weekly health area reports, collected in 90 health areas with 1,282 suspected meningitis cases were included. At a threshold of 10 per 100,000, the total number of estimated vaccine-preventable cases was 29 with district-level surveillance and vaccine response, 141 with health area-level surveillance and vaccination, and 339 with health area-level surveillance and district-level vaccination. While being most effective, the latter strategy required the largest number of vaccine doses (1.8 million), similarly to the strategy of surveillance and vaccination at district level (1.3 million); whereas the strategy of surveillance and vaccination at health area level would have required only 0.8 million doses. Thus efficiency was lowest for district-level surveillance and highest for health area-level surveillance with district-level vaccination.

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/tmi.12815

This article is protected by copyright. All rights reserved.

Conclusion: In this analysis, we found that effectiveness and efficiency were higher at health area-level surveillance and district-level vaccination than for other strategies. Use of Neisseria meningitidis C vaccines in a preventive strategy thus should be considered, in particular as most reactive vaccine response strategies in our analysis had little impact on disease burden.

Keywords: Response strategies, Meningitis, Epidemics, Niger, Surveillance

INTRODUCTION

Since the introduction of a serogroup A conjugate meningococcal vaccine (PsA-TT, MenAfriVac®) in the African meningitis belt in 2010, the frequency and intensity of bacterial meningitis epidemics have decreased and reached an exceptionally low level in 2013-2014 [1]. The remaining few epidemics have been due to meningococcal serogroups W and X, which are not targeted by preventive vaccination and had caused several sporadic epidemic events since 2000 [2; 3; 4; 5]. Neisseria meningitidis (Nm) C has rarely been detected in carriage or meningitis in the African meningitis belt for about 30 years, with the last NmC outbreak in the region dating back to 1979 [6]. However, NmC re-emerged in an epidemic in Northern Nigeria during 2013-2014 [7], with the largest NmC epidemic ever reported during 2015, causing 8,502 cases in Niger and 2,845 suspected cases in Nigeria [8]. This epidemic emergence of NmC meningitis emphasizes the need for continued serogroup-specific surveillance for rapid detection of epidemics. For over one decade, the control of meningitis epidemics in Niger has been based on a reactive vaccination strategy using serogroup AC or ACW polysaccharide vaccines combined with adapted treatment protocols [9]. Field experience showed that reactive vaccination often is implemented too late to influence the epidemic curve [10].

To inform public health recommendations, we evaluated the effectiveness and efficiency of current and hypothetical surveillance and vaccine response strategies against NmC meningitis epidemics during 2015 in Niger, similar to a recently published analysis suggesting a higher efficiency of health-center level surveillance for serogroup W and epidemics after elimination of meningococcal serogroup A[11]. Our approach builds on evidence for the localized character of meningitis epidemics [12;13], and thus is based on surveillance data at the health-center level.

METHODS

Databases

First, to provide a description of NmC epidemiology in Niger during 2015, we used data from laboratory-confirmed meningococcal meningitis from routine countrywide surveillance in Niger, spanning from 1st July 2014 to 30 June 2015. Laboratory confirmation was based on polymerase chain reaction

(PCR) [14], which was performed on 65% (841/1282) of suspected cases reported for this period. For the present analysis, weekly case counts were analyzed at the health center level.

Second, to evaluate surveillance and response strategies based on suspected case reporting, we compiled data on weekly suspected bacterial meningitis case counts reported by health centers from 1st July 2014 to 30 June 2015. In the context of routine countrywide epidemiologic surveillance, suspected case reports are available only aggregated at the district level. Due to limited resources for this compilation of original data at the health center level, we compiled these data only for Dosso region. This region was selected because it had shown the highest NmC meningitis incidence during 2015, along with the capital region Niamey, where, however, the patient referral structure is less appropriate for a health center level analysis.

Dosso region is located in the Southern part of Niger, bordered by the region of Tahoua in the East, Tillabery in the North and the West, Benin and Nigeria in the South. It had 2,372,411 inhabitants in 2015 (14% of the Niger population) and was organized in 5 health districts and 94 health areas (HA).

Statistical analysis

We used standard epidemiological methods to describe the NmC epidemic. We calculated weekly incidence rates (WIR) as a weekly number of suspected cases per 100,000 inhabitants and cumulative annual incidence (AI) as number of suspected or confirmed cases per 100,000 inhabitants during an epidemiological year. The Institut National de la Statistique provided the number of inhabitants per village according to the 2001 national census. We applied a mean annual growth rate of 3.3% [15]. We identified epidemic thresholds (ET) as WIR of suspected meningitis in Dosso region, based on receiver-operator curves for the detection of high AI [11; 12]. First, we chose as the primary reference standard an AI above the 95th percentile of AI in all HA in the database (172 per 100,000) and as the secondary reference an AI above the 97.5th percentile (263 per 100,000). The optimal epidemic threshold was chosen with regard to their sensitivity and specificity in detecting HA according to the reference standards. An epidemic was defined as WIR in a HA that exceeded the corresponding threshold for at least 1 week. We tested 22 thresholds of WIR at the HA level for their performance in detecting high AI, ranging from 1 to 200 cases per 100,000 during one week, named ET1, ET2 etc. Based on this definition of epidemic signals, we evaluated the effectiveness and efficiency of surveillance (ie, data analysis) and vaccine response strategies by calculating the number of potentially vaccine-preventable cases and number of vaccine doses needed per epidemic signal.

Three strategies were evaluated as previously described [11]. Briefly, we approximated the number of potentially vaccine-preventable cases (Nvp) as:

Nvp = N3w * PNm * VC * VE; where N3w is the number of suspected cases observed in the surveyed HA or district from three weeks after exceeding the threshold; PNm is the percentage of suspected cases confirmed as Nm, estimated as 50% in both epidemic and endemic periods; VC is the expected vaccine coverage during a mass campaign in response to an outbreak, estimated at 80%; and VE is the expected vaccine effectiveness, estimated at 80%. We calculated the number of vaccine doses needed per epidemic event as the target age group 1-29 years of the total population, which was 74% for the study region; and the total number of preventable cases in the target population for 100,000 vaccine doses. To evaluate the sensitivity of our estimates to a longer delay from signal detection to effective vaccine protection, we varied this delay from three to six weeks.

All analyses were performed in R Studio software version 2.15.3 [16]. Maps were created with QGIS software version 1.5.0 [17].

Ethical aspects

All data were collected via the national surveillance system after authorization by the Ministry of Public Health. Therefore, National Ethics Committee agreement was not required. Patients were, however informed of the reason why their CSF sample was being taken and they received free care during the meningococcal season. Each health structure received the individual results of submitted samples. The retrospective disclosure of information to other partners occurred without any mention of the patient's identity.

RESULTS

In the eight regions of Niger, a total 1,144 NmC, 206 NmW, 1 NmX, 127 Streptococcus pneumonia (Sp), 8 Heamophilusinfluenzae (Hi) cases were confirmed among 4,318 Cerebro-Spinal Fluids (CSF) samples sent to CERMES from July 2014 to June 2015. Five regions and 25 districts reported at least one NmC case, with median WIR at district level of 0.62 per 100,000 (ranging from0.09 to 8.84 per 100,000) and median AI at 3 per 100,000 (ranging from 0.1 to 43.4 per 100,000). During the same period, 93 HA reported at least one NmC case with median WIR at 4.31 (ranging from 0.09 to 47.2) and median AI rate at 10.8 ranging from 0.09 to 135 per 100,000 (Figure 1). The two regions with highest annual NmC incidence were Dosso with 15.7 (372 cases/2 372 411 inhabitants) and Niamey with 43.4 (496 cases/1 142 248 inhabitants) per 100,000.

In the five districts of Dosso region, we included 4,763 weekly HA reports, collected in 90 HA with 1,282 suspected meningitis cases. Among the 841 meningitis suspected cases sent from the region of Dosso and analyzed by PCR at CERMES, 372 were due to NmC, 72 to NmW, 25 to Sp, 1 to Hi,

6 indetermined Nm, 364 negative and 1 not available result. At least one case of NmC was found in 33.0% (N=31) of HA of Dosso region (Figure 2): 39 NmC cases found in 2 HA of Dosso district; 202 NmC cases found in 13 HA of Doutchi district; 12 NmC cases found in 4 HA of Boboye district; 78 NmC cases found in 8 HA of Gaya district and 6 NmC cases found in 3 HA of Loga district. 32% (N=115) and 30% (N=109) of confirmed cases, respectively, occurred among 5- to 9-year-old and 10-to 14-year-old children (Figure 3). The highest annual incidence rate was observed among 10- to 14-year-old and 15- to 19-year-old persons with 76 and 73 cases per 100,000 inhabitants, respectively.

The median AI rate in the districts of Dosso region was 43 suspected cases per 100,000 (range 11-86 cases per 100,000), with the highest rates found in the districts of Gaya, Doutchi and Dosso with 86, 73and 43 cases, respectively, per 100,000 inhabitants. The median AI at the HA level was 8 cases per 100,000 inhabitants, with substantial heterogeneity between HA of the same district (Figure 4). The highest AI rates were 624 cases per 100,000 inhabitants in the HA of Karakara in the district of Gaya and 475 cases per 100,000 inhabitants in the HA of Tombokoirey in the district of Dosso.

The median WIR at the HA level during the meningitis season from January through May was 0 (range 1-220)cases per 100,000 inhabitants, with the highest WIR being 220 cases per 100,000 in the HA of Tombokoirey (district of Dosso) during calendar week 19 (Figure 5).

The thresholds that had the best performance to detect AI beyond the 95th and the 97th percentiles were at HA level WIR of 20 and 30 cases per 100,000 respectively and WIR of 9 and 10 respectively, at the district level.

Based on the current strategy of district level surveillance, the total number of epidemic events in Dosso region from July 2014 to June 2015 varied from 2 to5, depending on the chosen incidence threshold (ranging from 15 to 2 per 100,000). There would have been 22 to 45 events based on HA level surveillance (threshold ranging from 7 to 30), leading to 4 or 5 vaccine campaigns at district level. At a threshold of 10 per 100,000, the total number of estimated vaccine-preventable cases in Dosso region was 29 with district-level surveillance and vaccine response, 141 with HA-level surveillance and vaccination, and 339 with HA-level surveillance and district-level vaccination (Table 1). While being most effective, the latter strategy required the largest number of vaccine doses (1.8 million), similarly to the strategy of surveillance and vaccination at the district level (1.3 million),while the strategy of surveillance and vaccination at HA level would have required only 0.8 million doses. Thus efficiency was lowest for district-level surveillance (2.29 vaccine-preventable cases per 100,000 doses) and highest for HA-level surveillance with district-level vaccination (19.28 cases per 100,000 doses). The gain in effectiveness from moving to a finer geographical level of intervention was greater than for any change of threshold. When the assumed delay between epidemic signal and ef-

fective protection from vaccine antibody increased from 3 to 6 weeks, efficiency fell in all strategies, in particular for the HA level surveillance and vaccination strategy, which became clearly less efficient than the strategy based on HA level surveillance and district level vaccination, but remained more than twice as efficient than the districtlevel surveillance and vaccination strategy. District level surveillance and vaccination at a threshold of >10 cases per 100.000 did not prevent any cases with a delay of 5 weeks or longer (Figure 6).

DISCUSSION

In this analysis, we found that effectiveness and efficiency were higher for HA level surveillance and district level vaccination compared to other strategies. Although the efficiency decreased in all strategies with the variation of the delay between epidemic signal and the effective protection from vaccine antibody, the strategy of HA level surveillance and district level vaccination remained more efficient than the other strategies. The gain in effectiveness from moving to a finer geographical level of intervention was greater than for any change of threshold. Serogroup C of Neisseria meningitis has been the most common serogroup in 2015 epidemic in Dosso region and affected more children who were fortunately part of the vaccination campaign target. However, a global shortage of meningococcal vaccines limited this campaign. Considering that vaccine shortage could lead to a large spread of the outbreak, the international health community helped Niger government to combat this outbreak by providing vaccine against serogroups A, C, Y and W [18]. In this situation of vaccine shortage, the definition of the more efficient and effective strategy for vaccine response is essential. Giving the magnitude of the 2015 outbreak, and time to acquire vaccines, the strategy of surveillance at the HA level and vaccination of the whole district seemed to be the most appropriate strategy for NmC outbreaks. This strategy can allow rapid detection of cases and more effective vaccine coverage maintained at a good level over time. It had also a good level of efficiency maintained over time, although requiring more quantity of vaccine doses. Nevertheless, for our simulated situation of NmA elimination, effectiveness was highest for HA level surveillance and district vaccination [11], whereas efficiency was optimized for HA vaccination. So, the district level response should be maintained for NmC, but the surveillance for outbreak detection should be brought to the HA level.

We also found that only half of HA in which NmC circulated experienced an epidemic, and there was great heterogeneity between NmC incidences of HA in the same district. At least one NmC case was found in 33% of the 90 HA of Dosso region, most of them without epidemic situation, suggesting that the presence of NmC does not necessarily lead to an epidemic, but that co-incidence with some epidemiogenic co-factors is required. These epidemiogenic factors are likely more present at the border of Nigeria and along the Niger river which Nigerian (Niger) part is used to join Nigeria and

Mali. As previously hypothesized by Mueller & Gessner [19], such a factor could be responsible for the surge in carriage prevalence observed during localized epidemics [20], and micro-epidemics of viral respiratory infections could be involved. Epidemics were more frequent on the border with Nigeria and the region of Dosso was the first affected by the outbreak in Niger. This could be the result of the presence of the NmC as localized epidemics in the North of Nigeria in 2013 [7] associated to a low immunity of population to NmC. Indeed, vaccinations were less frequent in the population since 2010 with the introduction of serogroup A meningococcal meningitis vaccine (PsA-TT). Behavioral and cultural factors, such as high travel activity at the Niger-Nigeria border may have interacted with the precedent factors and favorited the 2015 burden of NmC in Niger. For a better understanding of NmC emergence and expansion, carriage studies should be performed regularly in the meningitis belt particularly for the re-emerging serogroup C of Nm.

The high heterogeneity of incidences even within epidemic or non-epidemic districts concords with the results of NmA epidemics in Burkina Faso [12] and Niger for the period of 2002-2012 [21]. Although the fact that the epidemic largely affected Dosso region, this heterogeneity was visible, suggesting the necessity of surveillance at a finer spatial level than the district for a rapid detection of localized epidemics.

Children aged 5-14 years were most affected and those aged 10-19 years most at risk for NmC in our study. The most affected group of age found in this study for NmC was similar to that of NmX and NmA in Burkina and Togo [5].

Other authors observed that the recent NmC outbreak in Nigeria was not unusual in regard to its seasonal variation, with maximum number of cases being reported in the dry season [7]. We also found that the period of its occurrence, the peak of the season, and the magnitudes of its incidence were not unusual. The peak incidence rates of NmC were comparable to reported NmA rates [22].

This surveillance study has a number of limitations. Our analysis is based on routine surveillance data with little patient information, such as individual or village level vaccination status, or clinical presentation of cases. We chose only one region for our analysis of surveillance strategies, and giving the fact that this region was one of the most affected by the outbreak, we could have overestimated the incidence of NmC and the number of epidemic events. The demographic data used for incidences calculation were those of 2001 to which we applied a constant grow rate, which may not take into account the exact spatio-temporal, cultural and ethnical variations of the population.

Data linkage of suspected and confirmed meningitis cases highlighted opportunities and improvements in meningococcal surveillance in Niger. Matching of the two datasets has been possible due to enhanced surveillance activities conducted since 2002. The availability of PCR since 2002 [23] may have also improved our ability to microbiologically confirm meningococcal disease diagnosis

during 2015 outbreak. This may also explain the larger proportion matched and better serogroup information for 2015 meningitis epidemic in Niger.

The changing epidemiology of meningococcal meningitis in the meningitis belt with emergence of non-A serogroup was considered a potential risk following the introduction of a monovalent serogroup A vaccine [24], while the epidemic emergence of NmC was rather unexpected. Given the quick geographic expansion of the serogroup from 2013, it is possible that NmC outbreaks will occur regularly in the region. Use of NmC vaccines in a preventive strategy thus should be considered, in particular as most reactive vaccine response strategies in our analysis showed limited impact on disease burden.

ACKNOWLEDGMENTS

We acknowledge the contribution of all the staff of the "Centre de RechercheMédicale et Sanitaire" and health centers of Niger who participated to the meningitis surveillance, the "Direction de la Surveillance et de la Riposte aux Epidémies", the "Direction des Statistiques", the "Institut National de la Statistique", the World Health Organization, the CDC/MenAfriNet and the United Nations International Children's Emergency Fund. This work was supported by the CDC/MenAfriNet and the Office of International Cooperation of the Principality of Monaco. Surveillance activities were funded by the Ministère Français des Affaires Etrangères (Fonds de Solidarité Prioritaire), Sanofi Pasteur and the« Ministère de la Santé Publique du Niger/Fonds commun». The "Cooperation Technique Belge" funded the first health area map's support.

REFERENCES

1. World Health Organization. Meningococcal disease control in countries of the African meningitis belt, 2014.WklyEpidemiol Rec 2015March27;90(13):121-132.

2. MacNeil JR, Medah I, Koussoubé D, et al. Neisseria meningitidis serogroup W, Burkina Faso, 2012. Emerging Infect Dis2014; 20(3):394-399.

3. Boisier P, Nicolas P, Djibo S et al. Meningococcal meningitis: unprecedented incidence of serogroup X- related cases in 2006 in Niger. CID 2007; 44:657-63.

4. Gagneux SP, Hodgson A, Smith TA et al. Prospective study of a serogroup X Neisseria meningitidis outbreak in northern Ghana. J. Infect. Dis. 2002; 185:618-626.

5. Delrieu I, Yaro S, Tamekloé TAS et al. Emergence of epidemic Neisseria meningitidis serogroup X meningitis in Togo and Burkina Faso. PLoS ONE 2011; 6(5)e19513.

6. Broome CV, Rugh MA, Yada AA, et al. Epidemic group C meningococcal meningitis in Upper Vol-ta, 1979. Bull World Health Organ. 1983;61(2):325-30.

7. Funk A, Uadiale K, Kamau C, et al. Sequential outbreaks due to a new strain of Neisseria meningitidis serogroup C in northern Nigeria, 2013-14. PLoS currents Outbreaks 2014; doi:10.1371/currents.outbreaks.b50c2aaf1032b3ccade0fca0b63ee518.

8. World Health Organization. Preparedness for outbreaks of meningococcal meningitis due to Neisseria meningitidis serogroup C in Africa: recommendations from a WHO expert consultation. WklyEpidemiol Rec 2015 Nov20;90(47):633-644.

9. WHO. Control of epidemic meningococcal disease. WHO practical guidelines. Lyon, France: Fondation Marcel Mérieux; 1995.

10. WHO. Recommended standards for surveillance of selected vaccine-preventable diseases. WHO/V&B/03.01; 2003.

11. Maïnassara HB, Paireau J, Idi I, et al. Response strategies against meningitis epidemics after elimination of serogroup A meningococci, Niger. Emerg Infect Dis. 21(8)Aug. doi:http://dx.doi.org/10.3201/eid2108.141361:1322-1329.

12. Tall H, Hugonnet S, Donnen P, et al. Definition and characterization of localised meningitis epidemics in Burkina Faso: a longitudinal retrospective study. BMC Infectious Diseases 2012; 12:2, http:/www.biomedcentral.com/1471-2334/12/2.

13. Paireau J, Girond F, Collard JM et al. Analysing spatio-temporal clustering of meningococcal meningitis outbreaks in Niger reveals opportunities for improved disease control. PLoSNegl Trop Dis 2012; 6(3):e1577.

14. Chanteau S, Sidikou F, Djibo S et al. Scaling up of PCR-based surveillance of bacterial meningitis in the African meningitis belt: indisputable benefits of multiplex PCR assay in Niger. Trans R Soc Trop Med Hyg2006; 100(7):677-80.

15. Institut National de la Statistique - Niger. Annuaire statistique « Séries longues 1990-2006 ». Edition 2007

16. R Core Team, 2012. R A language and environment for statistical computing. R foundation for statistical computing, Vienna Austria. http://www.R-project.org.

17. QGIS Development Team, 2012. QGIS Geographic information system. Open source geospatial foundation project. http://qgis.osgeo.org.

18. Ministère de la Santé Publique - Sécretariat Général - Direction de la Surveillance et la Riposte aux Epidémies. Rapport de la gestion de l'épidémie de méningite de 2015 au Niger. Juillet 2015. 61 pages.

19. Mueller JE, Yaro S, Madec Y et al. Association of respiratory tract infection symptoms and air humidity with meningococcal carriage in Burkina Faso. Trop Med IntHealth 2008; 13: 1543-52. doi: 10.1111/j.1365-3156.2008.02165.

20. Koutangni T, Boubacar Mainassara H and Mueller JE. Incidence, Carriage and Case-Carrier Ratios for Meningococcal Meningitis in the African Meningitis Belt : A Systematic Review and Meta-Analysis. PLoS ONE 2015; 10(2); doi:10.1371/journal.pone.0116725.

21. Mainassara HB, Paireau J, Idi I et al. Serogroup-specific characteristics of localized meningitis epidemics in Niger 2002-2012 and 2015: a serogroup-specific analysis of surveillance data. PLoS One. 2016 Sep 22;11(9):e0163110. doi: 10.1371/journal.pone.0163110.

22. Mainassara HB, Sidikou F, Djibo S, et al. Epidemiological patterns of bacterial meningitis in Niger from 2002 to 2010. Science Journal of Public Health 2014; 2(2):58-63.doi:10.11648/j.sjph.20140202.13.

23. SidikouF, Djibo S, Taha MK et al. Polymerase chain reaction assay and bacterial meningitis surveillance in remote areas, Niger. Emerging Infectious Diseases 2003;9(11):1486-1488.

24. Dakar Discussion Group on Priorities for Research on Epidemic Meningococcal Disease in Africa, Altmann D, Aseffa A, et al. Priorities for research on meningococcal disease and the impact of serogroup A vaccination in the African meningitis belt. Vaccine 2013;31:1453-7.

Corresponding author: Halima Boubacar Maïnassara, Centre de Recherche Médicale et Sanitaire, 634 Boulevard de la Nation YN034, PO box 10887, Niamey, Niger. Phone +227 20 75 20 40 / 45, Fax +227 20 75 31 80, Email :halima@cermes.org /amhalima@gmail.com

Table 1. Comparison of different strategies of surveillance and meningococcal vaccine response in terms of effectiveness and efficiency for preventing meningitis casesin Dosso region, Niger, 2015. This region had 2.37 million inhabitants.

Total popula- Total preventable Total preventable

threshold N epidemic tion con- Total doses cases by vaccina- cases per 100,000

(WIR per signals/N in- cerned by the (1-29 yrs) tion doses

Strategy 100,000) terventions signal

Surveil-

lance vaccination

Health area Health area 7 45/45 1,257,206 930, 332 184 19.7

10 38/38 1,116,644 826,317 141 17.1

15 32/32 1,020,622 755,260 95 12.6

20 24/24 702,031 519,503 62 12.0

30 22/22 634,973 469,880 40 8.5

Health area District 7 45 /5 2,372,411 1,755,584 342 19.5

10 38/5 2,372,411 1,755,584 339 19.3

15 32/5 2,372,411 1,755,584 271 15.4

20 24/4 2,161,414 1,599,446 204 12.8

30 22/4 2,161,414 1,599,446 130 8.1

District District 2 5/5 2,372,411 1,755,584 286 16.3

4 4/4 2,161,414 1,599,446 173 10.8

7 4/4 2,161,414 1,599,446 89 5.6

10 3/3 1,735,748 1,284,454 29 2.3

15 2/2 956,919 708,120 17 2.3

Cases < 2 years of age were not excluded.

total: for entire study area Dosso region (population 2,372,411); WIR, weekly incidence rate per 100,000; HA, health area.

Cumulative incidences of NmC

□ O.l -4.9

□ 5 - 19.9

■ 20 - 49.9

■ 50 - 89.9

■ 90 - 135.9

Health areas without NmC cases

Dosso Region □

Health area

Figure 1: Cumulative annual incidences of confirmed Neisseria meningitidis (Nm) C per 100,000 in Niger during 2015 epidemiological year. Five of the eight regions presented at least one NmC case found in 90 health areas. Most of the health areas with high incidences were located in Dosso region.

Figure 2. Cumulative annual incidences of confirmed Neisseria meningitidis (Nm) C per 100,000 in health areas of Dosso region (Niger) during the 2015 epidemiological year.

The serogroups were identified by systematic polymerase chain reaction testing on cerebrospinal fluid. 31 health areas reported at least one confirmed NmC Case. White areas represent health areas without NmC cases. Grey areas represent health areas with at least one NmC case.

Figure 3: Distribution of PCR-confirmed Neisseria meningitidis C cases and annual incidence rate of NmC per 100,000 inhabitants according to the age group, Dosso region (Niger), 1st July 2014 to 30 June 2015.

Figure 4. Annual incidences of suspected meningitis in the 91 health areas and five districts of Dosso region (Niger), 2015. Annual incidences of health areas are presented by district. The health areas that are outliers are Birni urbain (NmW) and Kiota (NmC) in the district of Boboye; Tombokoirey (no sample received), Tessa_Dosso (NmC), Goroubankassan (no sample received), Barokoira (NmW) and Saboudey (no sample received) in the district of Dosso; Dankassari (NmC) in the district of Doutchi; Kakakara (NmC) in the district of Gaya.

CD S=L

CD 7=1

■—I

CD C-J

CD kiTi

■—I

CD LT]

Danks Dicundisu Karakara KDUtaumb-au TE3sa

TambDkDirey Birni urbain Ddlities^ Fadama

Gorou Eankaasan

Maikalgo

Z-abori

"1—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—r

1 3 5 7 5 11 13 15 17 19 21 23 25

Calendar week

Figure 5. Epidemic curves (suspected meningitis) of selected health areas of Dosso region, Niger, 2015. . Health areas with cumulative annual incidences higher than 100 cases per 100,000 inhabitants were included in this graph. Tombokoirey, Tessa Dosso and Gorou Bankassan belonged to Dosso district; Koutoumbou, Dioundiou, Karakara and Zabori belonged to Gaya district; Dankassari, Doumega, Fadama and Maikalgo belonged to Doutchi district, Birni urbain belonged to Boboye district.

3 4 5 6

Delay from epidemic detection to vaccine protection (Weeks)

Figure 6. Comparison of different strategies of surveillance and meningococcal vaccine response in terms of efficiency for preventing meningitis cases, in Dosso region (Niger), 2015. We considered three, four, five and six weeks delay between epidemic detection and effective vaccine protection. The strategies were surveillance and vaccination at health area (HA/HA) level, surveillance at health area level combined with vaccination of the district (HA/District) and surveillance and vaccination at district level (District/District). Incidence thresholds for epidemic response were 7, 10, and 15 weekly cases per 100,000 inhabitants for surveillance at the health area level, and 4, 7 and 10 for surveillance at the district level.