Two cases of human cutaneous anthrax in Bosnia and Herzegovina , September 2014

J Arapović (jarap@sve-mo.ba)1,2, S Skočibušić1,2, B Jelavić2,3, H B Ivanković1,2, M Jurić2,3, D Mamić4, S Grgić1,2, J Lesko2,3, M Leventić2,3, I Soldo2,5, J Ravlija1,6, J Nikolić1,2 1. Department of Infectious Diseases, University Hospital Mostar, Mostar, Bosnia and Herzegovina 2. School of Medicine, University of Mostar, Mostar, Bosnia and Herzegovina 3. Department of Otorhinolaryngology and Maxillofacial Surgery, University Hospital Mostar, Mostar, Bosnia and Herzegovina 4. Department of Epidemiology and Public Health, Public Health Institute of Livno, Livno, Bosnia and Herzegovina 5. Department of Radiology, University Hospital Mostar, Mostar, Bosnia and Herzegovina 6. Department of Epidemiology, Public Health Institute of Federation of Bosnia and Herzegovina, Sarajevo, Bosnia and Herzegovina

intravenously daily for the next 10 days.The patient also received local ophthalmic antimicrobial therapy (tobramycin solution).Clinical recovery was followed by regression of local swelling and peeling of the dark crust and normalisation of CRP levels.

Background
B. anthracis is a sporulating Gram-positive bacterium.The main routes of human exposure to spores are by inhalation, ingestion, contact with skin and injection of contaminated drugs [1,2].In most patients (95%), the disease manifests as cutaneous anthrax, whereas the remaining 5% of cases present with inhalational or gastrointestinal syndromes [3].Of these two forms, the most severe is inhalational anthrax, which is usually fatal if left untreated.In such cases, often acts of bioterrorism are suspected [1][2][3][4].Cutaneous anthrax on the other hand generally develops after direct contact with infected animals or animal products.Symptoms typically appear up to 17 days after contact [5], but in an experimental model, the incubation time can be up to 58 days [6].Thus, cases of cutaneous anthrax are reported as an occupational disease, which mostly occurs in farmers, butchers, dealers of hides and animal hair, wool sorters and veterinarians [4].Characteristic of cutaneous anthrax is the appearance of a malignant pustule surrounded by oedema at the infection site [7].Due to the use of antibiotics, the mortality rate is typically under 1%, but if left untreated the fatality rate can reach 20% [2,7].
Anthrax has been used in bioterrorism.The only documented terrorist attack with anthrax spores was in 1993 by the Aum Shinrikyo cult in Japan; there were no cases [8].During deliberate release of weaponised anthrax spores in 2001 in the United States, 28 people tested positive for B. anthracis in nasal swabs, 22 became ill and five people died [1,6].Sporadic anthrax in livestock was last reported in Bosnia and Herzegovina in 2010 [9][10][11], but no confirmed cases of cutaneous anthrax in humans had been reported in this country in the last two decades.In 2011, sporadic human cutaneous anthrax cases due to contact with infected livestock were reported in Serbia [12], which borders with Bosnia and Herzegovina, and in Romania [13].Sporadic livestock anthrax cases have also been reported in the past 25 years in areas of Croatia bordering Bosnia and Herzegovina [9,14,15].
The two cases reported here should be considered as an alert to the medical community in Bosnia and Herzegovina to the possible appearance of anthrax in humans arising from exposure to infected animals.

Discussion
The last reported case of anthrax in humans in Bosnia and Herzegovina was in 1992, a woman from the rural south-western part of the country, who was hospitalised and treated for cutaneous anthrax in Split, Croatia [16].A possible human anthrax case in Bosnia and Herzegovina -a person from neighbouring Serbia -was reported in 2009 in Bosnia and Herzegovina's epidemiological bulletin [17], but the infection was not confirmed by laboratory analysis or epidemiological investigation (unpublished data).
Floods in Bosnia and Herzegovina and neighbouring countries in May 2014 could wash and mobilise B. anthracis spores from deeper layers in the soil.In addition to classical infection by contact with bacteria or spores, B. anthracis might also be transmitted by insects [16,18], which are especially active in warmer (summer) days, although the epidemiological impact of biting and/or non-biting insects in anthrax infection is not yet fully understood.
The structure of the epidemiological and veterinary services in Bosnia and Herzegovina should also be borne in mind regarding anthrax control.One of the two political entities in the country is the Federation of Bosnia and Herzegovina, which is divided into 10 independently structured counties.Every county has its own department of public health and veterinary department, leading to challenging control, coordination and prevention of infectious diseases through appropriate vaccination of livestock.
As described by Kracalik et al. for Georgia [19,20], vaccination of livestock and education of cattle workers could be the basis of an effective strategy for the prevention of anthrax also in Bosnia and Herzegovina and thus for the avoidance of anthrax spread.Thus, a longterm vaccination programme in livestock coordinated by the government would be helpful as an important measure to prevent the spread of anthrax in Bosnia and Herzegovina.
The human cases reported here should serve as a reminder to healthcare personnel in Bosnia and Herzegovina to be vigilant.

Surveillance and outbreak reports
The impact of a national routine immunisation programme initiated in 1999 on Hepatitis A incidence in Israel, 1993 to 2012 H Levine (hlevine@hadassah.org.il) 1 , E Kopel 2 , E Anis 1,2 , N Givon-Lavi 3 , R Dagan 3  1.Braun School of Public Health and Community Medicine, Hebrew University-Hadassah, Jerusalem, Israel 2. Division of Epidemiology, Ministry of Health, Jerusalem, Israel 3. Pediatric Infectious Disease Unit, Soroka University Medical Center and the Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel Data on long-term impact of universal national vaccination programmes against hepatitis A are lacking.We aimed at evaluating the impact on hepatitis A incidence of the Israeli toddlers-only universal routine two-dose vaccination programme against hepatitis A initiated in 1999.All hepatitis A episodes reported to the national surveillance system from 1993 to 2012 were analysed in relation to the vaccination programme and coverage.Mean vaccine coverage in Israel between 2003 and 2010 was 92% for the first dose, given at 18 months of age, and 88% for the second dose, given at 24 months.The annual hepatitis A incidence declined from a mean of 50.4 per 100,000 in the period between 1993 and 1998 to a mean of <1.0, during the period from 2008 to 2012, representing a reduction of >98%.The decline was evident in all ages and ethnicity groups, including unvaccinated populations.Of the 1,247 cases reported nationwide between 2002 and 2012, the vaccination status could be ascertained in 1,108 (89%).Among them, only 20 (2%) were reported be vaccinated with one dose and three (<1%)

Introduction
In many low-income countries, hepatitis A virus infects more than 80% of the population by late adolescence [1].The infection is also common in middle and highincome countries [1].
The disease caused by hepatitis A virus is usually more severe with increasing age.Hepatitis A virus is mostly transmitted from person to person by the faecal-oral route, however, common source outbreaks related to contaminated water or food occur.High risk groups include persons with chronic liver disease, men who have sex with men, people who inject drugs, travellers to countries where hepatitis A is endemic and children in communities with consistently elevated rates of hepatitis A infections [2].
The long-term impact after the introduction of new immunisation programmes is dependent not only on the vaccine itself, but on vaccination coverage and its sustainability as well.Hepatitis A vaccine is highly effective when given in two doses, six to 12 months apart [3].Furthermore, in Argentina, the recent introduction of a single dose of hepatitis A vaccine in the universal immunisation plan to 12 months-old children resulted in a profound impact on disease in all ages with a magnitude comparable to that of a two-dose schedule within a short time [4].No data on long-term population impact of hepatitis A national immunisation plan (NIP) exist, not enabling long-term definitive predictions [5].The 2012 World Health Organization (WHO) position paper on hepatitis A vaccines stated that following its introduction, the assessment of hepatitis A vaccine impact is important, using information on morbidity generated by surveillance and study data [6].
Until 1999, Israel was considered a country with intermediate hepatitis A endemicity [7].Between 1993 and 1998 incidence rates were 30 to 70 per 100,000 population, with higher rates in the non-Jewish population [8].The first hepatitis A vaccine licensed in Israel was in 1996 [8].Initially, vaccines were used sporadically, except for targeted vaccination in the Israeli military that started in 1997 [9].In July 1999 however, Israel was the first country to introduce hepatitis A vaccine to its NIP as a two-dose schedule, at ages 18 and 24 months, with no catch-up campaign.We previously reported the early impact of the programme, five and a half years after its introduction [8].In brief, the annual incidence declined by 95% or more when comparing the period from 2002 to 2004 to that between 1993 and 1998.
The decline was most prominent in the vaccinated age group (1-4 years), but was remarkable in all age groups, demonstrating herd protection.Other studies, based on records of a large health maintenance organisation in Israel, showed a reduction of 88% in hepatitis A incidence from 1998 to 2004 and of 95% from 1998 to 2007 [10,11].
We report here on the long-term (14 years) impact of the toddlers-only universal hepatitis A two-dose vaccination programme initiated in 1999 on hepatitis A incidence in all ages in Israel.

Methods
Acute infectious hepatitis has been notifiable by law since 1950 in Israel and hepatitis A cases have been reported separately from other infectious hepatitis cases since 1993 [8].The Division of Epidemiology at the Ministry of Health collects and reviews reports from all districts and collates data weekly and annually.While there is no official criteria for the diagnosis of hepatitis A disease, reports of cases will usually be discarded unless there is a positive laboratory test result for anti-hepatitis A virus IgM antibodies or epidemiologic linkage with a previous serologically-confirmed case [8].from Jewish (ca 80%) or non-Jewish populations [8].The non-Jewish population includes mostly the Moslem Arab population.Overall and age/ethnicity specific annual incidence rates per 100,000 population were calculated.
Data on vaccination coverage of first and second doses by year, district and ethnicity were calculated based on reports provided to the Division of Epidemiology from all 15 public health districts in Israel, as previously reported [8].In Israel, ca 95% of all routine immunisations are given free of charge in public sector motherchild health centres.Immunisation coverage rates are based on doses of specific vaccines given in these centres per number of newborns residing in each of the 15 public health districts.In eight districts with a large number of annual births, a systematic 16.7% sample of newborns (i.e.those born every sixth calendar day) is selected for calculation of coverage.The vaccination coverage for the first hepatitis A vaccine dose is calculated at the age of two years and for the second dose at the age of three years.

Vaccine uptake
Among the Jewish population, the mean annual vaccine coverage between 2003 and 2010 for the first hepatitis A vaccine dose was 89% (range: 82-92%) and for the second dose, 84% (range: 78-90%) (Figure 1).The respective figures among the non-Jewish population were 97% (range: 94-98%) and 94% (range: 92-98%).The coverage of both first and second hepatitis A vaccine doses between 2003 and 2010 was comparable to that in the period from 2001 to 2002 for the Jewish population and higher than that of 2001 to 2002, for the non-Jewish population.The overall mean annual vaccine coverage for 2003 to 2010 was 92% for the first dose and 88% for the second dose, compared with 90% and 85% in the period from 2001 to 2002.Vaccination coverage differed by districts and years, with pockets of low coverage, most notably in the Tel Aviv district.In this district, inhabited mainly by a Jewish population, vaccination coverage for the second dose ranged between 56% and 79% during 2003 to 2010.

Hepatitis A incidence
The overall pre-vaccination hepatitis A incidence in the years from 1993 to 1998 was 50.4 per 100,000.Shortly after the initiation of the programme in July 1999, a sharp decline in incidence occurred within the following three years reaching 2.3 or less per 100,000 after 2002 [8] (Figure 2).From 2008 to 2012, the rates remained stable with a mean of < 1.0 per 100,000.This represents a persistent reduction of > 98% compared with the pre-vaccine period (p < 0.001; chi-squared test).The incidence decline in hepatitis A morbidity was evident in all age groups and in both Jewish and non-Jewish ethnic groups (Figure 3).Furthermore, during the period from 2003 to 2011, a reduction in the occurence of large outbreaks occurred.No outbreaks comprising five cases or more were reported during this entire period (data not shown).Between March 2012 and February 2013, one large outbreak was reported in the Tel Aviv district, with 80 cases in total, mainly young male adults.An urban cluster of people who inject drugs and homeless men,

Figure 2
Annual incidence rates of hepatitis A per 100,000 population, Israel, 1993Israel, -2012 The start of the routine vaccination programme is marked by an arrow.Incidence/100,000 inhabitants which comprised ca 15% of the cases including the index case, was identified (data not shown).
Of the 1,247 cases reported nationwide between 2002 and 2012, vaccination status could be ascertained in 1,108 (88.9%).Of these, 1,085 (98%) were not vaccinated, and 20 (2%) had received one dose.Only three cases (<1%) were reported to be vaccinated with two doses before onset of illness.The latter three cases were adults, with questionable verification of vaccination status in two cases, and immunosuppression in the third one [12].

Discussion
We show sustained success of the Israeli NIP in almost complete elimination of hepatitis A morbidity and transmission.Our findings are of universal importance since Israel was the first country to include hepatitis A vaccine into the NIP in 1999.The differences in hepatitis A incidence by ethnicity prior to the NIP were eliminated.The sustained reduction was evident in all ethnicity groups, thus showing the potential of a vaccination programme to reduce health disparities, as was shown in Arizona, United States [13].
Sporadic cases occured, mostly in high-risk individuals and most commonly among travellers to endemic areas outside Israel [14].
Further support for this reported reduction stems from the elimination of hepatitis A outbreaks in school children during the study period, from a range of eight to 48 outbreaks in the southern district between 1993 and 1998 to zero outbreaks in the period from 2001 to 2005 [15].Only one outbreak with more than five cases (n = 80) occurred during the entire period, which occurred mainly among young adults (in the county with the lowest vaccination rate).
The vast majority of cases in our study had not received hepatitis A vaccinations in the past, while hepatitis A

Figure 3
Annual incidence rates of hepatitis A per 100,000 population, by specific age and ethnic groups, Israel, 1993-2012 In each panel, the start of the routine vaccination programme is marked by an arrow.cases among those vaccinated with one dose were rare, pointing to the high protection given by the vaccine on the individual level.Such cases highlight the need to ensure a full vaccination schedule among individuals susceptible to both hepatitis A exposure and vaccination failure [12].Recent reports suggest memory and persistence of immunity even after one hepatitis A vaccine dose in adults [16][17][18] and high short-term effectiveness after one dose only when given to children at one year of age, as part of the NIP [4].

Jewish population
Additional support to the decline observed in this study, is a similar decline in hepatitis A incidence in Israel Defence Forces soldiers (individuals > 18 years of age), based on the military surveillance system, as well as a decline in the proportion of hepatitis A IgG seropositive recruits with time [9].Furthermore, the virtual elimination of hepatitis A IgG seropositivity rates in 18 month-old toddlers (pre-vaccination age) living in a previously hyperendemic area after introduction of hepatitis A NIP (from 16.2 to 19.6% in 1991 through 2000 to 0% in 2003 through 2007) suggests that the virus circulation in the community is close to being eliminated [19].
The rapid decline and sustained very low incidence following the introduction of toddlers-only hepatitis A NIP in Israel, strongly supports the claim that the decline is due to the vaccination programme.Our findings are in line with similar observations in other countries with diverse and heterogeneous epidemiology, including Argentina, Belarus, China, Italy and Spain, where the implementation of routine vaccination of children in one or more age cohorts in all or part of the country was followed by immediate and extensive overall declines in hepatitis A incidence [1].
Other factors, such as improved hygiene or sociodemographic changes, cyclic trends, and vaccination beyond the NIP, should be considered as possible additional explanations to the declining incidence.However, the immediate and rapid effect post hepatitis A vaccine introduction speaks for only a small, if at all, role of these factors in the events.Cyclic pattern of disease incidence with peaks every five to 10 years has been noted in some low-income countries with temperate climates [1].In our case, the long-term 14 years follow-up, makes cyclic trend a very unlikely explanation for this decline.Some individuals were immunised beyond the NIP due to occupational, travel, medical or other reasons.However, the limited scope of this immunisation points against significant contribution to the decline.
The overall benefit to society of the near complete elimination following introduction and maintenance of NIP is extensive.Previous cost-benefit model predicting hepatitis A NIP in Israel during the 1997 to 2014 period showed a societal benefit:cost ratio of 2.54:1 [20].However, the real benefit was higher than predicted, due to herd protection effect in unvaccinated individuals.
Our study was limited by the passive surveillance system.However, as our surveillance system, reporting and diagnosis methods were mainly unchanged between 1993 and 2012, our data provide strong evidence for the continuous success of the vaccination programme.This was validated in the past with active surveillance [8].Our vaccination coverage estimation is limited and might be somewhat biased.As the denominator for vaccination coverage estimation is based on births, a number of children residing in Israel and not reported to the Ministry of Health (such as children of immigrants), may have lower vaccination coverage, leading to biased over estimation of vaccination coverage.However, these represent a small minority.Furthermore, since medical service is universal and provided free of charge, we expect similar vaccination coverage in these populations.
A major strength of the current study lies in it being a long-term (20 years) prospective surveillance of hepatitis A incidence, including the six years before initiation of the NIP.An additional strength of the study is its nationwide population-based nature, covering the entire Israeli population, and including age and ethnicity sub-group analyses.
In conclusion, the results of this long-term impact study document that the toddlers-only universal routine two-dose vaccination programme is highly effective, and resulted in the sustained near elimination of hepatitis A in Israel.
Spatial distribution and cluster analysis of a leishmaniasis outbreak in the south-western Madrid region, Spain, September 2009 to April 2013 Since July 2009, there has been a community outbreak of leishmaniasis in south-west Madrid, Spain.The present study used the spatial distribution of cases to investigate the connection between the outbreak and a recently built peri-urban park.We included 157 cases of cutaneous (CL) and 90 cases of visceral (VL) leishmaniasis diagnosed at Fuenlabrada University Hospital between July 2009 and April 2013.CL and VL cases were geo-referenced and incidence rates by census tract were calculated.To identify high-risk areas, the spatial autocorrelation between individual cases was estimated.In a next step, areas where risk of disease was significantly increased were identified by cluster analysis.Higher incidence rates and the areas with highest intensity of CL and VL were located in the north-western part of the municipality.The most likely cluster of CL comprised three census tracks with relative risk (RR) = 11.5 (95% confidence interval (CI): 9.2-13.6).Two additional significant VL clusters were detected, the most likely one with RR = 9.2 (95% CI: 7.3-11.1).In addition, we found one significant VL cluster in the immigrant population (RR = 12.8; 95% CI: 9.3-16.1).The spatial pattern of leishmaniasis transmission revealed a relation between the outbreak and the suspected risk area.

Background
Leishmaniasis is a parasitic disease caused by more than 20 protozoan species of the genus Leishmania.It is transmitted through the bite of female sandflies of the genera Phlebotomus (Old World) and Lutzomyia (New World).An ecological system in which a Leishmania species is maintained indefinitely is usually formed by one principal reservoir host, e.g.dogs for Leishmania infantum in both the Old and New Worlds [1].Recent reports indicate that in Spain, other animals such as wild carnivores, rabbits and hares may play a role in the maintenance of the system, occasionally bringing the parasite from its enzootic focus into contact with humans [2,3].
In Spain, leishmaniasis is endemic and both the visceral (VL) and the cutaneous (CL) forms are caused by L. infantum.Phlebotomine sandflies belonging to the subgenus Larroussius serve as vectors, while the dog is the main reservoir [4].The real prevalence of leishmaniasis in Spain is unknown.Notification of leishmaniasis is not mandatory at national level (mandatory only in 12 of 17 regions) and there is no national leishmaniasis control programme [5].Moreover, Leishmanias sp.cases are underascertained and underreported in Spain [6].
In the autonomous region of Madrid, notification of leishmaniasis has been mandatory since 1997 and is the most common zoonotic disease affecting infants and in patients with human immunodeficiency virus (HIV) infection [7,8].Between 2000 and 2009, the annual incidence rate was around 0.5 per 100,000 population (between 12 and 25 leishmaniasis cases per year) [9].During the last quarter of 2010, the case number increased fivefold.Subsequent research confirmed that an outbreak of leishmaniasis had been ongoing since July 2009 in the south-western area of the region of Madrid, with most cases occurring in the city of Fuenlabrada [5,10].
Most of the L. infantum isolates obtained from patients in the outbreak area presented an uncommon strain: the combined genotype L-920 [2].The initial hypothesis to explain this outbreak postulated risk areas for bites by Phlebotomus sandflies infected with this new strain, creating a temporospatial clustering of infected individuals.Because leishmaniasis was found in only 7% of the dogs examined in the outbreak area, other reservoirs were suspected and investigated.Around 30% of the hares studied were positive for Leishmania parasites [3,5], suggestive of a sylvatic transmission cycle possibly linked to a recently built park called Bosquesur, adjacent to the urban area of Fuenlabrada [11].
Spatial pattern analysis has been found to be useful to better understand disease transmission of parasitic diseases when there is a strong correlation between the spatial distribution of the disease and its hosts [12,13].We aimed to assess the spatial distribution of CL and VL cases and the cluster occurrence within the city of Fuenlabrada (Madrid) as well as the distribution of the cases in relation with the park Bosquesur.

Study area
Fuenlabrada is a city and municipality located in the

Population
The population data was obtained from the 2011 annual statistic published by the City of Fuenlabrada, stratified by sex and age, at census track level [15].Fuenlabrada had an area of 39.1 km 2 and an estimated population of 204,100 in 2011, of whom 75.3% were in the age group of 16-64 years and 15.4% were immigrants.

Spatial analysis
Geo-referencing of all VL and CL cases was carried out using Google Earth to obtain the geographical

Bosquesur park
Continuous urban fabric Industrial or commercial units Road and rail networks Construction sites Sport and leisure facilities Arable land coordinates of patients' residences.In order to study the features of the territory, we obtained the map of Fuenlabrada by census track from the National Institute of Statistics (2011) [16] and downloaded images from the National Plan for Aerial Orthography (PNOA) project [17].

Spatial distribution
We calculated the incidence rates for CL and VL by census track adjusted by four age groups (0-14, 15-44, 45-64, ≥ 65 years) and by sex.We plotted the map with rates to understand the distribution of the disease at census track level.We calculated the local Moran's index in order to study the local indicator of spatial autocorrelation (LISA) between the rates.This index assesses local associations by comparing local averages to global average [18].Its significance is estimated by generating a reference distribution using 999 random permutations.The LISA significance map includes the following categories: 'high-high' indicates clustering of high value rates (positive spatial autocorrelation), 'low-high' indicates that low value rates are adjacent to high value rates (negative spatial autocorrelation), 'low-low' indicates clustering of low value rates (positive spatial autocorrelation), 'high-low' indicates that high values are adjacent to low value rates (negative spatial autocorrelation), and 'not significant indicates that there is no spatial autocorrelation.
In order to understand the risk distribution, we estimated the spatial smoothing distribution of CL and VL by means of the kernel density function.This function is based on the quadratic kernel function described by Silverman [19].This method is an interpolation and smoothing tool used to generalise the position of a point to an area.Kernel density estimation fits a curved surface over each case such that the surface is highest above the case and zero at a specified distance (the bandwidth) from the case.
To estimate the distance to the park Bosquesur, which was the main suspected risk area according to previous investigations [3,11,20], we drew a polygon around this park.Distances to other landscape elements were also assessed.We created buffers with different distances: 500, 1,000 and 2,000 m around the park Bosquesur to measure the distances of the cases´ addresses.Then, we calculated the average nearest neighbour index (ANNI) to calculate the distance between the location of each case and their nearest neighbour.This method was used as first approach to test whether or not the cases were clustering.If the average distance was below the average for a hypothetical random distribution, the distribution of the analysed characteristics was considered to be clustered.The index is expressed as the ratio of the observed distance divided by the expected distance.Thus, if the index is < 1, the pattern exhibits clustering, while index of > 1 indicates a trend towards dispersion.A significance level of 99% was chosen in the analysis.

Cluster analysis
Spatial clusters were analysed using the SaTScan spatial statistic estimator developed by Kulldorff [21].To assess CL and VL spatial clusters in the entire population, and separately in the immigrant population, we Maps were drafted using the free-licence source: Spanish National Institute of Geography.National Plan for Aerial Orthography (PNOA project).Available from: http://www.ign.es/ign/layoutIn/actividadesFotoTelePNOA.do[15].

A.Cutaneous leishmaniasis B. Visceral leishmaniasis
Bosquesur park LISA Not Significant High-High High-Low Low-High Low-Low used the scan statistic estimator to perform a purely spatial analysis, based on the assumption of a Poisson distribution.This method consists of creating a circular window which scans the entire study area.In our study we restricted the spatial window to a maximum radius of the average distance between cases (250 m).
The radius of the centroid varies continuously in size from zero to the specified upper limit, in our case 250 m.The circle with maximum likelihood and containing more cases than expected is denominated the most likely cluster.
An increase in observed cases above the number expected was assessed using Monte Carlo test simulations (999 replications) with a 95% confidence interval.
For the spatial analysis, we used Arcgis version 10.0, free software GeoDa and SaTScan.Data analysis was performed using SPSS version 18.0.

Spatial distribution
The incidence rates of CL and LV in the different census tracks varied between zero and 1,003 and between zero and 613 cases per 100,000 population, respectively.Figure 3A shows the distribution of CL rates.Higher rates of CL (between 500 and 1,100 cases/100,000 population) were observed in three census track in the north of the municipality, while the highest rate for VL was found in a different northern census track (Figure 3B).
The LISA for CL and VL rates are shown in Figure 4. Significant clusters of high values for CL and VL were detected: a CL hotspot (p < 0.005) in a census track located in the north-west of the municipality, and two significant VL hotspots (p < 0.005) in two census tracks with high values adjacent to census tracks with low value rates.
The average distance among the cases (bandwidth) was 250 m.Following spatial smoothing, areas with high intensity for CL were identified in the north of the city, close to the park Bosquesur (Figure 5A).The distribution of the VL cases is shown in Figure 5B; high intensity areas were located in the northern part of the municipality.
Table 1 summarises the cases distribution and distance to the park.We observed a decreasing trend in CL and VL occurrence in relation to the distance to the park.Up to 75% of CL cases and 70% of VL cases lived in places less than 1 km from this suspected risk area.
The ANNI was 0.505 (p < 0.001) for CL and 0.582 (p < 0.001) for VL throughout the study area.Both for CL and VL, the pattern exhibited clustering.The observed average distances were 98 and 270 m and the expected average distances were 193 and 463 m for CL and VL, respectively.

Cluster analysis
Four significant CL clusters were identified through spatial cluster analysis (p < 0.001, Figure 6A).The most likely cluster comprised three census tracks in which 24 cases were diagnosed during the study period, while the number of expected cases was 2.46 (relative risk (RR) = 11.50 and p < 0.005).Other secondary clusters were located in the north of the municipality.
The VL spatial cluster analysis detected two significant clusters (p < 0.001, Figure 6B; Table 2).The most likely cluster comprised one census track with eight cases and an expected number of cases of 0.95.The RR was 9.15 (p < 0.005).The other secondary cluster was located in the north-eastern part of the municipality.
When we analysed the data according to the migrant status, we found one significant cluster for VL in the immigrant population, with four observed cases, 0.35 expected cases and a RR of 12.75 (p < 0.005) (Figure 6C).This cluster included one census track located in the north of the municipality, different from the one detected for VL in overall population.No significant cluster of CL was identified through spatial cluster analysis in the immigrant population.

Discussion
Our study revealed the spatial characteristics of human CL and VL during an outbreak in Fuenlabrada (Madrid) using geographic information system (GIS) tools and spatial statistical analysis.Similar approaches have been used in Spain to investigate the spatial distribution of the sandfly vector and canine leishmaniasis [4,22].However, to our knowledge this is the first attempt to implement spatial techniques to assess the distribution and cluster occurrence of human CL and VL cases in an outbreak in Europe.
We have described the evolution of the VL and CL cases in Fuenlabrada separately, as analysing them together may have resulted in misinterpretation.VL was more common in men than in women, while no sex differences were found for CL.This seems to be concordant with findings from previous epidemiological studies indicating that VL occurs more frequently among adult men [20,23].Although it has been hypothesised that sex hormones play a role in the modulation of immunity against leishmaniasis [24], the explanation for this trend still remains uncertain.
In our research, the census tracks with the highest incidence of CL were different from the ones with highest incidence of VL.It is difficult to determine whether the different distribution of CL and VL cases determined by disease onset, vector distribution, disease transmission pattern and/or presence of the host.
Leishmaniasis is known to be a diverse and complex disease [1], therefore further studies may be needed to investigate this particular finding.
Although the spatial distribution of CL and VL leishmaniasis did not overlap perfectly in space, both were spatially clustered along the border of a park.Risk associated with other landscape elements within the municipality were analysed without significant results.Leishmaniasis is one of the main parasitic diseases of the world for which the transmission profile includes landscape elements and environment [25].In our research, all spatial methods deployed showed that the northern peripheral census tracks were the most heavily affected.Moreover, CL and VL incidence rates and the distance to the park Bosquesur were spatially correlated.Finally, the cluster analysis also showed that the most likely CL and VL clusters were close to this park.
Our results were in accordance with previous research that emphasises that this disease can be associated with urbanisation close to vegetation areas [23,26,27].Worldwide, leishmaniasis outbreaks have been related to human activities close to or within forested areas: road construction, building of dams, irrigation schemes and horticulture development, and establishment of new residential colonies lead to intrusion into the sylvatic cycle of the disease [27,28].The park Bosquesur is a man-made natural area, located between the towns of Alcorcón, Leganés, Fuenlabrada, Getafe and Pinto.During the outbreak period, leishmaniasis cases were reported from these five towns [10].Adjacent to this newly planted area, public construction work was carried out on one of the main roads (M-407) to Fuenlabrada in 2009-10.All these recent changes at peri-urban level may have altered the ecology of this area and the transmission dynamics of leishmaniasis.
The sum of observed cases in the four significant clusters of CL was 61 (39% of the total 157 CL cases included in the analysis), while the percentage of VL cases included in the significant clusters was even lower, 14% (n=13) of the 90 VL cases.This difference in case clustering can also be observed in the epidemiological curve (Figure 2), although we cannot draw conclusions in this regard as we did not carry out a temporospatial analysis.Other exposure pathways are also plausible.As shown by de Almeida et al., the pattern of leishmaniasis is not static and the disease may spread from one area of a municipality to another [29].
On the other hand, the use of a patient's home residence as a marker of the place of infection may not be accurate because contact between host and vector can have occurred outside the home.
The park has several recreational areas with footpaths and bicycle trails [30].Social networks, recreational activities and other interactions between human settlements and the peri-urban natural environment may play a role in the distribution of the disease in this context [26,28].Nevertheless, these establishments are attended mostly during daytime.First appearance of active P. perniciosus females occurs at sunset and the density peak is usually reached around 23:00-24:00 [5,11].Therefore, the highest probability of transmission may be not associated with the presence of individuals inside the park, but rather in homes or outdoor places close to, but outside the park.
According to Arce et al., patients were asked during the outbreak investigation about recreational habits, but neither particular areas nor classic environmental risk factors were identified [10].It should be noticed that only cases were assessed during the official investigation.We believe that the outbreak investigation would have benefited from a control group in order to evaluate exposures and risk factors.
A high percentage of VL cases (39%) were immigrants and we identified a cluster of VL in this particular group.This cluster did not overlap with the VL cluster detected in overall population.We also carried out the analysis in the non-immigrant population only (data not shown) and did not find any relevant difference compared to the analysis on entire population.According to official data from the local authorities, 15.4% of the Fuenlabrada population in 2011 were immigrants, of which 14.9% are from Sub-Saharan Africa [15].The percentage of immigrants in the study area was similar to that of other urban areas in Spain [31].Several factors could have been responsible for the high percentage of VL in immigrants: migration of non-immune people from areas where L. infantum is not endemic [1,32], differences in host immune responses [33], and differences in living habits and/or health-seeking behaviour.These factors should be explored in further investigations.Our study has several limitations.One relates to the long incubation period of the disease, rendering determining and interpreting the information related to transmission site difficult.Control measures may have affected the spatial distribution of cases.We are aware of disinfection activities undertaken since 2011; although no information on the targeted sites is publically available, we cannot discard their impact on the case distribution.
We chose Kulldorff's method for the spacial cluster analysis because it has several advantages [34]: it adjusts for population density and confounding variables (e.g.age and sex); there is no pre-selection bias since the clusters are selected without prior hypothesis on their location, size or time period; the statistical test takes into account multiple testing and delivers a single p value; if a cluster is detected, its location is specified.
In a molecular typing study, Chicharro et al. found that the outbreak was not caused by a single parasite strain, as four combined genotypes were found.At Fuenlabrada Hospital, the ITS-LOMBARDI type was isolated from all serotyped cases.This L. infantum ITS type has been present in this region since at least 1992 [2].A high density of P. perniciosus has also been observed in the park [10].In previous research carried out in central Spain, a correlation was found between vector density and cases living in areas between villages or at the limits of a village [22].Future entomological and molecular typing studies could benefit from our results, as a combined methodology could allow more precise conclusions regarding the transmission patterns.

Conclusion
Although our study design did not allow establishing causal associations, the methodology can be considered useful in generating hypotheses during an outbreak investigation.Future work should examine the role of vector density, seroprevalence of Leishmania in canine and other possible reservoirs, climate variability, socio-economic conditions, land use and changes made by humans to the habitat over a longer time span in the study area.This will allow building accurate risk maps and targeting prevention and treatment interventions in these high-risk areas in a timely manner.
Since 2005, pneumococcal vaccination has been recommended in the Italian national immunisation programme (NIP), but only for risk groups (e.g.asplenia, immunocompromised subjects, patients with chronic diseases) [2,3].However, it was in the following years increasingly offered also to other target groups, dependent on the vaccination policies in the individual regions [2].Since 2008, pneumococcal vaccination has been included in the NIP free of charge for all newborns; it was administered as a three-dose schedule (during the first year of life) concomitantly with the hexavalent vaccine against diphtheria (D), tetanus (T), acellular pertussis (aP), Haemophilus influenzae type b (Hib), hepatitis B virus (HBV) and inactivated poliovirus (IPV) [4,5].Thus, in routine practice, PCV and the hexavalent vaccine were usually administered concomitantly to children in a single vaccination session.
Clinical trials evaluating the effect of co-administration of PCV13 with the hexavalent vaccine showed a comparable safety and immunogenicity profile as those evaluating co-administration of PCV7 and the hexavalent vaccines when given in routine practice; among systemic adverse events, fever was more common in subjects vaccinated with PCV13 compared with PCV7 [6,7].Like other vaccines, PCV can provoke fever, which could trigger a febrile seizure [7,9].However, those clinical trials were not sized to detect less common or specific adverse events following immunisation (AEFI), and only involved a selected paediatric population (children with risk conditions were excluded).Therefore, post-marketing surveillance remains essential.The routine monitoring of spontaneous reports collected by the Italian Pharmacovigilance Network (IPN) in the first year after introduction of PCV13 in Italy showed a slightly higher frequency of serious AEFI with PCV13 than with PCV7; this trend was more evident for neurological events and when PCV13 was co-administered with the hexavalent vaccine [10].This finding led us to further investigate the safety of PCV13 in combination with the hexavalent vaccine.
The first objective of this study was to estimate the risk of all types of AEFI, neurological events, and convulsions following the co-administration of PCV13 with the hexavalent vaccine, using the co-administration of PCV7 with the hexavalent vaccine as a reference.The second objective was to compare the risk of all types of AEFI, neurological events, and convulsions following the co-administration of PCV13 with the hexavalent vaccine vs single administration of PCV13 and the hexavalent vaccine in different vaccine sessions.

AEFI reports and administered doses
We retrieved paediatric spontaneous reports to IPN after vaccination with PCV (7-and 13-valent) or the hexavalent vaccine occurring from 1 January 2009 and 31 December 2011.Data retrieval from IPN took place in mid-2012.No exclusion criteria for cases of AEFI were adopted.Details on the vaccinations received during the three-year period, such as the number of doses of PCV7, PCV13 and the hexavalent vaccine administered to the paediatric population (0-2 yearsold) both as single or concomitant vaccination, were available from four Italian regions, Emilia-Romagna, Lombardy, Tuscany and Veneto, and pooled at regional level.These regions are situated in the north and in the centre of Italy.All our analyses were limited to these regions, covering 22.6% of the resident Italian population and representing 77.8% of the spontaneous reports collected in the IPN during 2011 [11].The corresponding paediatric population consisted of more than 217,000 children per year [12].The birth cohort of children in 2012 had the following distribution by region: 41,397 in Emilia-Romagna, 96,602 in Lombardy, 32,473 in Tuscany, 46,588 in Veneto.The vaccine coverage (as completed vaccine course) during 2011 for DTaP, Hib, HBV, and IPV (included in the hexavalent vaccine) was at least 95% in all regions involved [13].
Each report was identified through a unique anonymised code and it was not possible to directly identify the person.We used information on the vaccinee (age, sex), the event(s) (type, date of onset, seriousness and outcome), vaccine(s) administered (type, trade name and date of administration), region and local healthcare facility indicated on the reporting form for the adverse drug reaction (ADR).
We considered two 18-month time periods to estimate the AEFI incidence rates (IR) of PCV7 and PCV13 (alone or co-administered with the hexavalent vaccine), from 1 January 2009 to 30 June 2010 for PCV7 and from 1 July 2010 to 31 December 2011 for PCV13.All participating regional pharmacovigilance centres were asked to reconfirm the information reported for each case to exclude duplicates, increase data completeness and identify possible misclassification between PCV7 and PCV13.In Italy, during the period of interest, a single hexavalent vaccine was available.
Spontaneous ADR reports were grouped in three different categories, i.e. all the AEFI reported, neurological events, and convulsions.Neurological events were identified through the analysis of the reported AEFI and the related preferred terms coded according to the standardised medical terminology developed by the International Conference of Harmonization (MedDRA) [14].Only cases with at least one preferred term leading to the primary MedDRA system organ classification (SOC) 'nervous system disorders' were considered as neurological events [15].In addition, all serious cases were further evaluated (on the basis of the preferred terms, verbatim and other information included in the ADR reporting form) to confirm their inclusion/exclusion from the neurological events analysis.
All neurological reports were then evaluated to identify cases of convulsions (both febrile and afebrile) according to a pre-defined case definition.Cases reporting terms such as 'seizure' or 'convulsion' or 'convulsion and fever' were classified as convulsion/febrile convulsion by default.Cases with less specific terms (e.g.'tonic-clonic movements', 'hypertonia', 'oculogyric crisis') were classified as convulsion only when two terms indicating loss of consciousness and generalised motor manifestation appeared together in the same report; this is in accordance with current case definition guidelines [16].

Statistical analysis
IR of AEFI were calculated by dividing the number of the AEFI reports by the number of administered vaccine doses (expressed per 100,000 doses).The administered doses in the group receiving single administrations of PCV13 and the hexavalent vaccine at different times were calculated as the sum of the administered doses of PCV13 alone and the hexavalent vaccine alone.Confidence intervals (95% CI) were calculated using the Poisson distribution.Incidence rate ratios (IRR) were estimated using univariate Poisson regression.IR and IRR were estimated for three AEFI groups: (i) all types of AEFI, (ii) neurological events, (iii) convulsions (febrile and afebrile).An attempt to stratify AEFI groups by seriousness of event was made.A standard pharmacovigilance definition for seriousness of cases was used: "An adverse reaction corresponds to any untoward medical occurrence that at any dose results in death, is life-threatening, requires inpatient hospitalisation or prolongation of existing hospitalisation, results in persistent or significant disability or incapacity, is a congenital anomaly/birth defect [17]" All events which did not meet the criteria for seriousness were thus considered as non-serious.Because seriousness was analysed as reported and was not modified during the analysis, we need to consider that misclassification may have occurred.This study was not intended to investigate risks by seriousness of events, and the stratified analyses should be viewed as a hypothesis generator only.Statistical analyses were carried out using STATA software version 11.2 (Stata Corporation, College Station, United States).

ADR reports and administered doses
According to the NIP, administration of PCV (7-or 13-valent) together with the hexavalent vaccine was at least four times more frequent than single administration of PCV (7-or 13-valent) or the hexavalent vaccine (Table 1); the number of co-administered doses of PCV (7-or 13-valent) with the hexavalent vaccine was comparable.
Overall, 883 spontaneous reports of AEFI with PCV7, PCV13 or the hexavalent vaccine either as single or concomitant administration were retrieved in the IPN during the period 2009 to 2011 (Table 1); 107 reports were serious (12.1%), including two deaths of which one occurred two days after vaccination in a child suffering from perinatal hypoxic ischaemic encephalopathy and the other was sudden infant death syndrome.At least one neurological event was reported in 15.4% of the reports (136 of 883); 68 (50.0%) neurological events were serious.Of the 136 neurological events, 41 (30.1%) were cases of convulsions and 32 of the 41 were reported as serious.In particular, we found that of 41 cases of convulsion, 26 were reported as febrile convulsion, while the remaining 15 cases were afebrile convulsions; given the small sample we considered convulsion (both febrile and afebrile) as a single category in the analyses.
The majority of reports (n = 537), 15.4% of which were serious (n = 83), occurred after co-administration of PCV7 or PCV13 with the hexavalent vaccine (n = 232 and n = 305, respectively).The hexavalent vaccine alone was found to be administered in 256 AEFI reports, 5.5% of them serious (n = 14), while we found only 90 reports following a single administration of PCV7 or PCV13, 10 of which were serious (four with PCV7 and six with PCV13).
Reports of any AEFI (serious and not serious) were equally frequent for any of the three vaccines, whether administered alone or concomitantly.In contrast, serious AEFI, neurological events and convulsions were observed more frequently when PCV13 and the hexavalent vaccine were administered together.
Overall, the demographic characteristics of children experiencing an AEFI were comparable across vaccine groups (Table 2).However, according to the vaccine strategy adopted in Italy, children up to five completed months of age accounted for ca 70% of reports following co-administration of PCV (7-or 13-valent) with the hexavalent vaccine, while the majority of reports following single administration of PCV7 (67%) or PCV13 (80%) occurred in children that were at least 12 months-old.
On average, onset of the AEFI was on the same day as vaccine administration in 75.4% of the reports, ranging from a minimum of 55.8% for single PCV7 to a maximum of 82.2% for the co-administration of PCV7 with the hexavalent vaccine.The majority of the AEFI reported (86.6%) resolved or improved.No difference in event onset or ADR outcome was found between the two types of co-administrations across different AEFI groups (Table 2).

PCV7 or PCV13 co-administered with the hexavalent vaccine: AEFI incidence rates and incidence rate ratios
Overall, no differences in IR (for all AEFI, neurological events or convulsions) were observed following PCV7 co-administration or PCV13 co-administration (Table 3).
The IRR for all AEFI following the co-administration of PCV13 with the hexavalent vaccine compared with the co-administration of PCV7 with the hexavalent vaccine was 1.08 (95% CI: 0.91-1.28).Similarly, the IRR for neurological events or convulsion following co-administration of PCV13 with the hexavalent vaccine were 1.27 (95% CI: 0.85-1.89)and 1.43 (95% CI: 0.70-2.91),respectively.None of the three comparisons were statistically significant.When taking into account the seriousness of reactions, the IRR for all types of AEFI and for neurological events indicated an increased risk reaching statistical significance, and a similar risk trend was observed for serious convulsions (Table 3).

Co-administration of PCV13 with the hexavalent vaccine vs single administration of PCV13 and the hexavalent vaccine: AEFI incidence rates and incidence rate ratios
The AEFI IR for the co-administration of PCV13 with the hexavalent vaccine were compared with those observed when PCV 13 and the hexavalent vaccine were administered in different vaccination sessions.Overall, a lower significant IR for all AEFI was observed following the co-administration of PCV13 and the hexavalent vaccine (Table 4).In case of neurological events and convulsions the IR did not differ between the two groups.When considering the IRR, co-administration of PCV13 with the hexavalent vaccine was also associated with a smaller number of all types of AEFI compared with the single administration of PCV13 and the hexavalent vaccine at different times (IRR: 0.59, 95% CI: 0.49-0.72)(Table 4).This protective effect was not observed for neurological events or convulsions (IRR = 1.44; 95% CI: 0.77-2.67 and IRR = 1.46; 95% CI: 0.50-4.25,respectively).Taking into account the seriousness of events (serious AEFI overall, serious neurological events and serious convulsions), the IRR showed a trend of increased risk associated with coadministration, although statistical significance was not reached and CI were wide.

Discussion
In this study, we found a trend of a slightly increased risk for neurological events or convulsion after coadministration of PCV13 with the hexavalent vaccine when compared with the co-administration of PCV7 with the hexavalent vaccine.No increased risk emerged when the comparison concerns all types of AEFI.Two factors could be responsible for this observation.The first is the Weber effect, i.e. the increased attention paid by the healthcare professionals (HCPs) and the public to the launch of the new product (PCV13) in mid-2010 replacing the old product (PCV7) which may have led to an increased incidence of reported cases.The second factor is an overall increased vaccine reporting trend over the years in Italy, fuelled by the launch of several active pharmacovigilance projects [18].
The second relevant finding of this study regards the comparison between different vaccination strategies with PCV13 and the hexavalent vaccine (i.e.whether co-administered or not).Co-administration of these two vaccines showed a slightly increased risk trend for neurological events or convulsion.However, when considering all AEFI, co-administration had a protective effect compared with single administration.Thus, the comparison between the two vaccination strategies seemed to favour co-administration, which is the current immunisation practice in Italy), over single administration.It should be pointed out that the protective effect of co-administration on all AEFI may have been influenced by the fact that the probability for an AEFI to be counted twice is higher when PCV13 and the hexavalent vaccine are given on different days: one AEFI may occur with PCV13 on one day and another AEFI may occur with the hexavalent vaccine on another day.
Only two studies have been published on the postmarketing surveillance of PCV, both based on data from the United States (US) and none aimed at evaluating PCV given in co-administration [19,20].The first was a descriptive study by Wise et al., presenting ADR reports following PCV7 [19]; the second article by Tseng et al. was based on active surveillance data and investigated the association between pre-specified events (including febrile seizure) with PCV13 or PCV7; it did not find any signal of an increased risk for febrile convulsion [20].However, this study did not report the CI for risk estimates of febrile seizure and did not consider the effect of co-administration of PCV with other vaccines or other vaccination strategies.Of note, Tseng et al. discussed an ancillary analysis in children receiving PCV13 and influenza vaccine concomitantly; they found a risk trend for febrile seizure 1.35 times higher (95% CI: 0.93-2.00)than the risk after receiving PCV7 and influenza vaccine concomitantly [20].
A study based on active surveillance data was conducted in the US to investigate a signal of febrile seizures in young children after receipt of trivalent inactivated influenza vaccine (TIV) during the 2010/11 season.This study showed a risk of febrile seizure when TIV was co-administered with PCV13 that was higher than after single administration of TIV or PCV13 [21].These findings were validated by an observational study conducted in the US which used prospective data collection and showed a higher risk of fever when TIV and PCV13 were co-administered than when either of these vaccines was administered without the other [22].Increasing evidence suggests that co-administration and type of vaccine co-administered could play a role in the occurrence of an event such as convulsions following PCV.Overall, the results of our study are coherent with available literature data [19][20][21][22].
A strength of our study is that it used spontaneous reports from the general population exposed to vaccines without pre-defined exclusion/inclusion criteria.Moreover, our setting involved the whole paediatric population undergoing routine immunisation practice in the four regions included in our analysis.As these regions represent about a quarter of the country's total population, we consider our results to be representative for Italy as a whole.
The quality of spontaneous reporting data in Italy is high since an evaluation of each single case and validation of each report by a trained HCP is required [23].Furthermore, all cases reporting neurological events were reviewed, and a predefined criterion to identify convulsions was applied.
The main limitation of this study resides in the nature of spontaneous reporting data.Although they allow the detection of less common events such as convulsions, they may contain partial information reported in a narrative way and collected heterogeneously.Moreover, the percentage of under-reporting could be significant, leading to a systematic under-estimation of cases; this would lead to decreased power of the study in detecting differences in the risks of AEFI, ultimately causing a lack of statistical significance and a wider CI.The higher risk estimates for serious cases can be expected to be affected by this methodological limitation.Seriousness has been analysed as indicated on the reports, not modified during the analysis, and could have introduced misclassification; the analyses by seriousness should thus be taken only as hypothesis-generating.Indeed, for a better interpretation of the study findings, it is advisable to refer to the overall risks within each ADR category.No risk adjustments for factors representing potential confounders (underlying diseases, age, sex) was feasible since this information was not available at individual level for the subjects undergoing vaccination (only administered doses pooled at regional level were retrieved); however, age, sex, time of onset and ADR outcome can be considered to be balanced among the spontaneous reports in each vaccine group, and the influence of these covariates on the risk estimates could be residual.The lack of anamnesis for neurological conditions both in the AEFI reports and in each vaccinee's record is important missing information.Finally, we were not able to carry out any analyses by received dose since this information is not systematically included on the spontaneous report forms.
Even though safety surveillance based on spontaneously reported data is not intended to provide evidence on causality, it is a useful method to rapidly quantify associations between vaccines and less common adverse events.Moreover, preliminary findings based on surveillance data could help in designing further investigations to deliver more robust evidence.

Conclusion
Our analysis showed a trend of a slightly increased risk of neurological events or convulsions following vaccination with PCV13 compared with PCV7 when both were used in routine vaccination practice with the hexavalent vaccine.Similarly, we found an increased risk of neurological events or convulsions (although not reaching statistical significance) when PCV13 was co-administered with the hexavalent vaccine compared with single administration of both vaccines at different times.Given the limitations highlighted, our findings cannot be considered conclusive.Moreover, it should be underlined that such risks should be viewed in the context of the overall benefit of both vaccines.
While we continue monitoring reports of less common and potentially serious AEFI, further research should be conducted using different data sources that also account for dosing schedule, subjects' characteristics and co-morbidities.This study indicates that the evaluation of co-administration of PCV with other vaccines during a single session is a relevant issue for public health research.Our findings may also contribute to pooled estimates together with those of similar investigations.

Assessment of human influenza pandemic scenarios in Europe
C Napoli 1 , M Fabiani 1 , C Rizzo (caterina.rizzo@iss.it) 1 , M Barral 2 , J Oxford 3 , J M Cohen 4 , L Niddam 5 , P Goryński The response to the emergence of the 2009 influenza A(H1N1) pandemic was the result of a decade of pandemic planning, largely centred on the threat of an avian influenza A(H5N1) pandemic.Based on a literature review, this study aims to define a set of new pandemic scenarios that could be used in case of a future influenza pandemic.A total of 338 documents were identified using a searching strategy based on seven combinations of keywords.Eighty-three of these documents provided useful information on the 13 virusrelated and health-system-related parameters initially considered for describing scenarios.Among these, four parameters were finally selected (clinical attack rate, case fatality rate, hospital admission rate, and intensive care admission rate) and four different levels of severity for each of them were set.The definition of six most likely scenarios results from the combination of four different levels of severity of the four final parameters (256 possible scenarios).Although it has some limitations, this approach allows for more flexible scenarios and hence it is far from the classic scenarios structure used for pandemic plans until 2009.

Introduction
Before the 2009 influenza A(H1N1) pandemic, most European Union (EU) Member States had developed preparedness plans in order to timely respond to an eventual pandemic.Many of these plans involve explicit or implicit planning assumptions on what can be expected during a pandemic and on how a pandemic virus might behave [1].
The response to the emergence of the 2009 influenza A(H1N1) pandemic was the result of a decade of pandemic planning, largely centred on the threat of an avian influenza A(H5N1) pandemic.However, the influenza A(H5N1) and the 2009 pandemic influenza A(H1N1) viruses have markedly different characteristics in terms of mortality among confirmed cases and human-to-human transmission [2,3].Moreover, the 2009 pandemic influenza A(H1N1) virus caused illness that did not require hospitalisation in the vast majority of cases, and was a highly transmissible virus among humans spreading to several countries within days [3,4].
In this situation, the severity assessment applied during the 2009 influenza A(H1N1) pandemic using a variety of indicators leading to a qualitative assessment in three levels (i.e.mild, moderate and severe) was not specific enough to guide interventions [5,6].
After the pandemic, the International Health Regulations (IHR) Pandemic Review Committee encouraged the World Health Organization (WHO) to develop and utilise measures to assess the severity of every influenza epidemic by applying, evaluating and refining tools to measure severity every year [7].WHO has recently developed a new document for Pandemic Influenza Risk Management [8].
The 2009 influenza pandemic highlighted the importance of quantitatively defining different scenarios; severity should be assessed as early as possible during a pandemic and continually re-assessed as the pandemic evolves and new information becomes available.
This work has been conducted in the frame of the European Commission project FLURESP (Costeffectiveness assessment of European influenza human pandemic alert and response strategies) with the aim to define a set of scenarios to be used for a future pandemic planning (www.fluresp.eu).

Literature review and selection of parameters
The literature search was conducted by consulting Medline, restricting it to articles published until December 2011.Seven different sets of keywords were considered (Table 1).
A systematic selection procedure was conducted in two steps by two researchers independently.In the first step, the major topics of the articles were assessed by title and abstract.In this phase of the selection procedure, all articles reporting epidemiological data on influenza pandemics were included.In case of doubt on the article's relevant information, the article was included in the second selection step.
In the second step, the full text articles, previously selected , were assessed.Of each article included in the review, the following data were recorded: year of the study, year of pandemic referring to, country, and described parameters.
International technical reports were obtained by consulting the websites of the European Centre for Disease Prevention and Control (ECDC) and WHO.Influenza pandemic preparedness plans for the European Union/ European Economic Area (EU/EEA) countries were obtained from the ECDC website [9].We also considered relevant studies based on mathematical modelling published in the literature but not retrieved through the search strategy.
Parameters collected through the literature review were subsequently discussed within the FURESP Project by a panel of experts composed of collaborators from international (WHO and ECDC) and national public health organisations (from France, Italy, Spain and the United Kingdom) who selected the parameters to be used for defining scenarios.

Definition of severity profiles and scenarios
For each of the selected parameters, four severity profiles were defined.In order to set the profiles, ranges of variability for each of the parameters were categorised into a four-group scale, according to a quartile distribution.We then adjusted the ranges for each of the four groups, according to the suggestions made by the panel of experts.Based on the possible combination of the four severity profiles of each of the four parameters, a set of scenarios were defined.

Parameters selected
From the literature review we collected information on 13 parameters as potential candidates for defining the pandemic scenario.These parameters were divided into eight virus-related (R 0 ; CAR; age-specific CAR; CFR; communicability/generation interval; modes of transmission; incubation period; timing and duration of pandemic) and five health-system-related (CCR; HAR; ICAR; work absenteeism; BOR).
The panel of experts was of the opinion that some of the parameters collected through the literature review were more relevant for mathematical modelling than for public health purposes, and others were considered less relevant for defining scenarios; consequently, all these were excluded: R 0 ; age-specific CAR; communicability/generation interval; modes of transmission; incubation period; timing and duration of pandemic.For example, R 0 (the average number of secondary infections produced by a single infected individual while The search was restricted to articles published until December 2011. they are infectious, in an entirely susceptible population), incubation period, and the generation interval (defined as the mean duration between time of infection of a secondary infected individual and the time of infection of their primary infector), are measures of the degree of transmissibility of an infection and in combination might affect CAR.Age-specific CAR in most of the considered influenza pandemics were derived from studies conducted in small and selected communities not representative of the entire population, while the timing and duration of pandemic is expected to be from several weeks to a few months but will likely vary from country to country or within a single country.Therefore, these parameters were not considered in this study.Additionally, the contribution and clinical importance of potentially different modes of transmission of influenza are unknown and therefore were considered not relevant.
Thus, according to the opinion of the panel of experts, four parameters were selected to be used for defining scenarios for pandemic planning.The two virus-related parameters are listed below with their limitations: • CAR, the proportion of the population with clinical symptoms over a specified period of time.Some individuals may not develop symptoms severe enough to be readily identified as acute respiratory infection (ARI) or influenza-like-illness (ILI).
The measured CAR is thus not always the number of individuals who actually develop symptoms, and may also include the number of individuals seeking healthcare.• CFR, represented by the proportion of individuals who develop influenza symptoms, and die because of complications.The measured CFR could be affected by the laboratory confirmation that may be unavailable to validate the total number of cases.Moreover, the confirmation is likely biased to more severe cases.This results in an overestimation of the clinical severity of the disease, especially in case of people with underlying conditions that are at higher risk of death.
The health-system-related parameters deal with virulence (i.e. the ability of the virus to invade the tissues of the host and produce pathologic effects and complications) and impact (i.e. the effect on the healthcare sector) of the virus on the population.The most relevant health-system-resource utilisation parameters used to define pandemic scenarios are listed below with their limitations: • HAR, represented by the proportion of population hospitalised for confirmed influenza independently from the presence of complications.This measure is strongly affected by how the healthcare systems in different countries are structured.• ICAR, the proportion of hospitalisations for confirmed influenza that are treated in an intensive care unit (ICU) for influenza complications.
The ICAR could also be related to the level of virulence of the virus, since it is a proxy for the level of severity.

Literature review
A total of 338 documents (including technical reports and scientific articles and reviews) were identified using our search strategy with the seven sets of keywords.Of these, 17 were duplicated articles and 238 showed no relevant information on the selected parameters (Figure 1).In conclusion, 83 articles and documents reporting information on the parameters listed above were considered for this study.
The year of publication of these documents ranges from 2003 to 2011 with more than half of the documents published between 2009 and 2010.When evaluating the performance of keywords' combinations selected, 26% (83/321) of the detected documents provided useful information on the parameters for defining scenarios.The largest number of documents was detected using three sets of keywords (293/321, 91%) (Table 1).
The keywords combination 'human influenza pandemic description' provided the highest proportion of useful documents (10/28, 36%).
The range estimates for the parameters derived from the 83 selected documents and their specific references are listed in Table 2.

Figure 1
Flowchart outlining the selection of documents for the study on pandemic influenza scenarios in Europe The search was restricted to articles published until December 2011.Of the 83 relevant articles, 23 articles reported information on R 0 , with values ranging from 0.99 to 3.75.
Thirty-four reported data on CAR, whose values ranged widely between 0 and 50%, while eight documents provided some information on age-specific CAR.For the CFR, 30 articles showed a range between 0 and 25%.Only three articles dealt with the generation interval, whose range was 1.6-4.1 days.The duration of infectiousness, reported in five articles, ranged between one and 21 days.Only seven articles provided generic descriptions of possible modes of transmission: all of them reported the respiratory route by droplets of infected secretions and/or hand-face contact after touching a contaminated person or surface.The incubation period, described in 12 articles, ranged from 0.5 to seven days, while the pandemic duration varied from 0 to 180 days according to seven articles.Moving to health system resource utilisation parameters, CCR ranged from 14% to 73% (seven articles); HAR ranged from 0% to 27.5% (26 articles); ICAR ranged from 0% to 34% (13 articles); work absenteeism ranged from 0 to 40% (seven articles); and BOR was between 0% and 37% of total critical care bed capacity according to one article.
In some European countries (UK in particular) there were three waves associated with the 1918/19 pandemic [12].In the UK, the wave structure of this pandemic is not well understood; the final 1919 wave may have been a separate pandemic of a different virus to the 1918 waves.The smallest of the waves was in July-August 1918, the largest second wave was from October 1918 to January 1919, and the third wave was from February to April 1919 [13].Estimates of the national CAR vary in the UK, but suggest that nationally it was around 25% of the population (totalled over all waves).The highest CAR were observed in the young population.Estimates of the CFR are around 2%, relatively evenly spread across the population, though with an excess in young adults [12].The 1957/58 pandemic had one wave.Estimates of the CAR vary, but suggest that nationally it was around 30% of the population.

Parameters collected from the 2009 influenza pandemic
The recent influenza A(H1N1) pandemic in 2009/10 produced no major signal of excess deaths in the overall population [16], and most of the EU countries have reported data on those that have died from confirmed 2009 pandemic influenza A(H1N1) as a result of influenza A(H1N1) virus infection, but case ascertainment is unlikely to have been complete, and the true number is almost certainly higher [17][18][19].
Rates from the ILI surveillance systems across Europe showed that consultations were highest in the young.There was only one wave, except for the UK where two 'waves', one immediately following the other, were observed.There were high levels of background immunity among elderly.
In general, estimates of the CAR for the 2009 pandemic influenza A(H1N1) vary among countries: 0.01% in the central region of Portugal [20], 0.072% in Mexico [21], 18.3% in New Zealand [22], 30% in the Netherlands [23].Across Europe the estimated CAR was 30% [24].This variation reflects the different methods used to get the data: e.g.seroprevalence studies, epidemiological studies in different populations, mathematical models, etc.Other experiences in smaller groups of population provide additional results: 3.15% in a train in China [25], 4% in a primary school in China [26], 22% on a Peruvian Navy ship [27], 28.5% during an outbreak investigation in Nepal [28].
In the UK, figures used to track the epidemic suggest a CAR of 1-2% and modelling studies suggest that these estimates reflect only around 10% of those infected [29], which is consistent with the results of the serological analysis of the first wave [30].If only half of those infected were symptomatic, although possibly with very mild symptoms (as this is typical for influenza), the CAR would be around 5-10%.If so, estimates of the CFR are around 0.01% [31,32], but higher levels have been reported in the literature: up to 0.05% in the US [31], 0.1% in Spain [33], and 0.35% in Europe [24], 0.6% in Mexico [34].In terms of age groups, mortality was spread evenly across the age groups although most cases were reported in the younger age groups.
Definition of severity profiles and scenarios Table 4, shows the severity profile for each of the four selected parameters derived from the literature review and selected by the suggestions of the panel of experts.
With regard to CAR, the literature review reported data ranging from 0 to 50%.Nevertheless, since the maximum value of 50% refers to the extreme value reported during the 1889 'Russian' pandemic [13,[35][36][37][38][39], this value was excluded and, therefore, CAR maximum value was set at 35%.Also for CFR, from the literature review the observed values ranged from 0% to 25% (Table 2).However, some of the data collected from the literature review were estimates of CFR derived from different populations (often representing high-risk groups) and source of information (mortality associated with the 2009 pandemic influenza A(H1N1) was estimated 15 times higher than reported laboratory-confirmed deaths) [40].Moreover, when considering the influenza A(H5N1) avian influenza virus: CFR estimates reported by WHO for the ongoing outbreak is around 60% [41], even if, findings from a study based on surveillance and seroprevalence data published in 2008, reports estimates ranging from 14 to 33% [42].For this reason we set the maximum level of the CFR at 2.5%, as most of values from recent pandemics ranged from 0.01 to 2.5% [13,14,32] (Table 3).
HAR depends on the level of virulence of the pandemic virus.However, its estimation may be affected by access to healthcare, proportion of chronic medical conditions in the population, pregnancy, and the virus characteristics (e.g. the level of pre-existing immunity, and pathogenicity of the virus itself).In our literature review we obtained values ranging from 0 to 27.5%.As the extreme value refer to pandemic 1918/19 in the US Army Camps¸ according to the opinions of the panel of experts set the maximum value to 4% as this value was collected in the general population [5].
For ICAR, it ranged from 0 to 35% in the literature review, and we considered the maximum of 35% as 'reasonable worst case' given that it was derived from an accurate evaluation of laboratory-confirmed influenza hospitalised cases conducted in the US [43].
Based on the above described ranges of values, it is possible to define 256 different scenarios.Among these, we selected six reasonable scenarios, according to the opinion of the panel of experts (Figure 2).
In detail, scenario A represents a 'seasonal-like' influenza outbreak; scenario B describes a situation in which the virus is quite diffusive, with an important HAR and a low virulence and ICAR, similar to the 2009 pandemic.The high HAR in scenario B could also represent high-risk groups (e.g.elderly, individuals with underlying conditions).Scenario C and D represent a situation in which the CAR is high with a low and a high virulence, respectively.Moreover, since the scenarios described above do not take into account the age profile of the population and the proportion of individuals with chronic conditions, we consider a reasonable solution to retain scenario E and F that represent the worst case scenarios for at-risk groups.

Discussion
In our study we defined a set of scenarios that may be useful for pandemic planning.We used the combination of four severity profiles of four epidemiological parameters to identify 256 possible scenarios that can be adapted over time and are far from the classic scenarios structure used for pandemic plans up to the 2009 influenza pandemic [1].Among the scenarios identified, on the basis of a literature review and of the opinion of the panel of experts, we selected the six most likely scenarios that synthesise the possible effect of an influenza outbreak with different characteristics (from a seasonal-like to a major event).
Historically, influenza pandemic planning has been based on an assessment of the 'reasonable worst case', derived from previous influenza seasons and pandemics in the 20th century, and thus has shown not to be appropriate during a moderate event, such as the 2009 pandemic [44].Other experiences reported a modelling approach using a combination of indicators leading to a qualitative assessment in three levels (i.e.mild, moderate and severe) [5]; that approach was considered not to be specific enough to guide interventions [6].Moreover, mathematical modelling based on preliminary epidemiological data is useful in defining the impact and the mitigation measures to be implemented during a pandemic.However, these models are strongly affected by the epidemiological parameters used and, even if they are able to explore a wide range of values, they need a specific set of scenarios to produce reliable results.For this reason, the use of scenarios in pandemic planning is crucial.
In our literature review, most of the selected articles and documents were observational studies, mathematical simulations, or reviews.Information referring to different world regions, different population subgroups and different influenza pandemics (mostly the 2009 influenza pandemic) over 100 years-period made comparison of results difficult.In fact, our results showed that most of the parameters values vary a lot between different countries and in different pandemics.For example, mortality rates often vary by age: agespecific mortality rates for 1957/58 and 1968/69 show a U-shaped pattern with a slightly increased CFR in the very young and an increasing one with older age [45].
On the other hand, during the 1918 pandemic, a higher mortality rate was observed in young adults followed by lower rates in other age groups [35].Moreover, variation in epidemiological parameters could also reflect differences in the surveillance systems (e.g. in case of different case definitions, time lag between influenza confirmation and death, etc.) and diagnostic methods.This heterogeneous information presented in the documents did not allow us to use parts of the data (e.g. the absolute number of deaths was neglected where the corresponding denominator to calculate CFR was missing).It should also be noted that we did not consider age groups and the proportion of people with other underlying conditions that are strictly related to the vulnerability of the population to a pandemic virus [6].Finally, estimates for the 2009/10 pandemic are likely to change as further data and studies become available after the literature review was conducted (December 2011).
The experience of the 2009 influenza pandemic showed that the EU countries had prepared for a pandemic of high severity but appeared unable to adapt their national and subnational responses adequately to a more moderate event.Knowledge of past pandemics is of substantial help when planning for a future one [46] and indeed the epidemiological aspects of the three 20th century influenza pandemics (1918/20, 1957/58, 1968/69) are of outstanding importance.However, modelling studies based on epidemiological parameters collected during the 20th century pandemics overestimated the impact of the 2009/10 pandemic [8].Furthermore, society has undergone major changes since 1918 (the scenario on which most pandemic plans and models before and during the 2009 pandemic have been based) and even since 1968, with an increased availability of ICUs and clinical countermeasures (such as vaccines, antivirals, etc.).
Thus, in June 2013, the WHO published the 'Pandemic Influenza Risk Management' [8].The approach taken in this document introduces a risk-based approach to pandemic influenza risk management and encourages countries to develop flexible plans, and to conduct risk assessments in order to prioritise the development of risk management programmes tailored to the hazards present.Our results are in line with the 'WHO Pandemic Influenza Risk Management' [8] and provide a description of possible scenarios of pandemic influenza considering key epidemiological parameters.The described scenarios allow severity assessments and provide the basis for developing flexible risk management plans over the course of a pandemic.
In the context of the FLURESP Project, the proposed scenarios have been used to select potential response strategies (clustered and ranked according to performance and efficiency using a multi-criteria analysis) in order to conduct cost-effectiveness evaluations to compare cost and performance of response strategies for each proposed scenario.
Our study, although not based on a standardised procedure, is supported by an extensive literature review and suggestions derived from a panel of experts.
In conclusion, our study provides an original template to categorise human influenza pandemic scenarios, useful for pandemic planning.Before using its outcomes, limitations should be taken into account by public health authorities dealing with pandemic planning.This study is the first step of the FLURESP project, whose objective is to define adequate public health responses and measures according to each scenario presented in this paper and to compare performance and cost-effectiveness of such measures.

Figure 2
Figure 2Six possible scenarios identified by a combination of the selected parameters, study on pandemic influenza scenarios in Europe

Table 1
Frequency of spontaneous reports of adverse events following vaccination, by seriousness and vaccine type administered alone or in co-administration, four regions in Italy, 2009 to 2011 (n = 883) AEFI: adverse events following immunisation; PCV: pneumococcal polysaccharide conjugate vaccine.

Table 2
Spontaneous reports of adverse events following vaccination, by sex, age group, and time to onset, four regions in Italy, 2009 to 2011 (n = 883)

Table 3
Comparison of incidence ratios between PCV7 vs PCV13 given in co-administration with the hexavalent vaccine, four regions in Italy, 2009 to 2011 (n = 537) AEFI: adverse events following immunisation; CI: confidence interval; IR: incidence rate; PCV: pneumococcalpolysaccharide conjugate vaccine.

Table 4
Comparison of incidence ratios of co-administration of PCV13 with the hexavalent vaccine vs separate administration in different vaccination sessions, four regions in Italy, 2009 to 2011 (n = 448) ADR: adverse drug reaction; AEFI: adverse events following immunisation; CI: confidence interval; IR: incidence rate; IRR: incidence rate ratio; PCV: pneumococcal polysaccharide conjugate vaccine.a The events following vaccine administration in different vaccination sessions were obtained as the sum of the number of ADR following single PCV13 (Q3-Q4 2010 and 2011) and the number of ADRs following a single hexavalent vaccine dose (Q3-Q4 2010 and 2011).

Table 1
Sets of keywords used in searching scientific articles and number of articles retrieved for the study on pandemic influenza scenarios in Europe

Table 2
Range of values and references for the main parameters selected for the study on pandemic influenza scenarios in Europe

Table 3
Relevant parameters collected in the four past influenza pandemics for the study on pandemic influenza scenarios in Europe Estimates of the CFR are around 0.1-0.2%.These average figures mask the considerable variation by age, most deaths being in the older adult population.However, the highest number of cases was registered in the young individuals[14].The 1968/69 pandemic came in two waves in Europe[15].Estimates of the national CAR vary, but based on comparisons with the epidemic in the United States (US), it may have been around 35% of the population[12].Estimates of the CFR are less than 0.2%.These average figures for mortality mask the considerable variation by age, again, with most deaths recorded in the older adult population.

Table 4
Severity profile for the selected parameters, study on pandemic influenza scenarios in Europe