Brugueras et al. BMC Public Health (2019) 19:144 https://doi.org/10.1186/s12889-019-6448-3 RESEARCH ARTICLE Open Access Does the economic recession influence the incidence of pertussis in a cosmopolitan European city? Sílvia Brugueras1,2,3* , Cristina Rius1,2,3, Joan-Pau Millet1,2,4, Martí Casals5, Joan A. Caylà1,4 and Barcelona Pertussis Working Group Abstract Background: In the last few years, pertussis has re-emerged worldwide. The aim of this article is to study how the incidence of the disease has evolved in Barcelona city over a 16-year period, and determine which factors are associated with the evolution of the disease. We discuss the causes of the observed changes considering different possibilities such as vaccination coverage, vaccine effectiveness, increased surveillance or the effect of the current economic recession. Methods: We performed a cross-sectional, observational, population-based descriptive study using data for the 2000–2015 period from the notifiable diseases register maintained by Barcelona Public Health Agency. We used Poisson regression to compute adjusted odds ratios (aOR) and their corresponding 95% confidence intervals (CI). Results: A total of 1791 cases were registered. The incidence of the disease increased throughout the city from 2011 onwards. While children under 1 year of age had the highest-incidence and were the most at risk (aOR = 27. 18, CI:23.51–31.44), we found that the age of affected children was higher in the last years. Incidence proportion (PRR) was lower among foreign-born children than native children (PRR = 0.43 CI:0.32–0.58). In the whole-cell vaccine period (2000–2004), the percentage of cases under 1 year of age who received the vaccine was lower than in 2005–2015 when the acellular vaccine was used (p = 0.01), suggesting a lower efficacy of the acellular vaccine. However, vaccination coverage in children under 6 years remained high (~ 90%), and there were no significant year-to-year variations (p = 0.757). Moreover, there did not appear to be any significant restrictions in medical care. According to the index of disposable household income (DHI), pertussis incidence increased from 2011 onwards in all neighbourhoods and remained higher in those with lower DHI. Conclusions: The noteworthy increase in pertussis incidence does not seem to be due to the economic recession, but to other factors here described. Keywords: Bordetella pertussis, Cities, Economic recession, Epidemiology, Pertussis vaccine, Whooping cough Background complication in children, it may occur in all age groups Pertussis is an infectious respiratory disease caused by the and there has been an increase in the age of detected cases Gram-negative Bordetella pertussis bacterium. Clinical of pertussis [1]. Adults suffer from less severe symptoms symptoms include paroxysmal cough and inspiratory stri- that are more difficult to diagnose, which results in in- dor, which may be followed by periods of apnea [1]. Al- creased risk of transmission within the family [2]. though it is more common and is associated with more Pertussis is a Notifiable Disease (ND) that continues to be a public health issue worldwide despite broad vac- * Correspondence: silviabruguerastorrella@gmail.com 1 cination coverage [3–5]. Globally, data from the Euro- Servei d’epidemiologia, Agència de Salut Pública de Barcelona, Barcelona, Spain pean Centre for Disease Prevention and Control (ECDC) 2 Centro de Investigación Biomédica en Red de Epidemiología y Salud show an increase in the incidence of pertussis in the Pública (CIBERESP), Barcelona, Spain European Union (EU) from 2011 onwards. Several Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Brugueras et al. BMC Public Health (2019) 19:144 Page 2 of 9 countries, including Spain, The Netherlands, Denmark, contacts. The study of contacts consists in reviewing the Belgium, Ireland and the United Kingdom (UK) wit- vaccination status of the contacts and adapting it to the nessed an increase in incidence from 2011 onwards, immunisation schedule and administering prophylaxis to while that in Italy and Greece has remain generally all contacts, following the pertussis guidelines of the stable [1, 6, 7]. Non-European countries with broad vac- from the health department of Catalonia. In order to cination coverage, such as the United States (US), know the vaccination status of the whole population Australia and Israel have also witnessed a re-emergence there is a record in the Shared Medical History of Cata- of pertussis. However, several reports indicate that this is lonia (HCCC) of the total population that receives not a global re-emergence of the disease [3, 7–9]. healthcare in Catalonia. The immunization coverage in In several countries, such as the US, Mexico, Croatia and Barcelona has remained high. During the 2010–2015 Thailand, economic recessions have been shown to have an period, primary vaccination among 1 year old children impact on communicable diseases because of the subse- was between 90 and 95% and the complete vaccine quent increase in inequalities between different population coverage in young people aged 17 years was 68–77% groups. Disease transmission may increase because of in- (source: Sistema d’Informació dels Serveis d’Atenció Pri- creased contact, poorer living conditions, poorer access to mària SISAP, Institut Català de la Salut). The immunisa- treatment, and lower treatment adherence. In addition, tion program has remained the same during the study public funding allocated to preventing such diseases may period except for the replacement of the whole-cell vac- be affected by austerity policies, resulting in decreased care cine for the acellular vaccine. From 2004 to 2005 the for individuals in greatest need [10, 11]. The current global acellular vaccine was administered at all doses. economic recession started in 2008, and as yet there are no data on trends in the pertussis incidence in the context of Case definition this economic recession at local level. Cases with a clinical suspicion of pertussis were those in In the current study we: (i) describe the clinical and which the typical clinical coughing symptoms were epidemiological features of pertussis in Barcelona be- present for two weeks or longer, followed by one or sev- tween 2000 and 2015; (ii) analyse how the incidence of eral of the following symptoms: paroxysmal coughing, the disease evolved; (iii) evaluate the effect of the type of inspiratory stridor, vomit or apnea after coughing. No la- vaccine administered among children under 1 year old boratory confirmation was available in such cases. A per- (whole-cell vs acellular vaccine); (iiii) determine which tussis case was considered confirmed in patients with factors are associated with the increasing incidence in confirmed infection by B. pertussis using laboratory diag- the city between 2000 and 2015 considering variables nostic tests, based on culture or Polymerase Chain Reac- such as the gender, the age group, the socioeconomic tion (PCR). Confirmed cases also included patients who status and the period. Through a description of the dis- showed typical clinical symptoms and were epidemiolog- ease over the years, we discuss the potential etiological ically related to any other case that had already been factors of this increase and analyse whether the eco- confirmed with laboratory tests. Both suspected cases nomic recession had an impact in pertussis incidence. and confirmed cases were considered pertussis cases for the purposes of the current study [13]. Methods Design Variables Cross-sectional, observational, population-based descrip- Socio-demographic variables included gender, age (divided tive study. in 4 groups according to the most common distribution of pertussis: < 1, 1–14, 15–29 and > 30 years old), country of Population origin (native, foreign-born) and socio-economic status Pertussis cases of Barcelona notified by clinicians, micro- (divided in 5 groups, see below). Epidemiological variables biologists and public health nurses to the register of no- were the following: conclusion (confirmed case, suspected tifiable diseases (ND register) maintained by the case), association with other cases (sporadic, associated) Epidemiology Service of the Barcelona Public Health and contact tracing. Clinical variables included symptoms Agency (BPHA), Spain, between January 1st, 2000 and (cough, paroxysmal coughing, inspiratory stridor, vomiting December 31st, 2015. The population of Barcelona dur- after coughing, apnoea, fever, pneumonia and convul- ing the period was between 1.503.884 inhabitants in sions), evolution (curation, death), vaccination, proper 2000 and 1.609.550 in 2015 [12]. vaccination, laboratory diagnosis (culture, PCR) and hos- pital admission. Pertussis-related activities Vaccination variables included the number vaccinated Pertussis cases are reported to the BPHA, which carries cases (1 or more doses), non-vaccinated cases and cases out the follow-up of the cases and conducts the study of of unknown vaccination status. Brugueras et al. BMC Public Health (2019) 19:144 Page 3 of 9 Proper vaccination status was defined as follows: city’s local census to obtain data on the total population people older than 2 years who had been given at least 4 in the city each year, distributed according to gender, age vaccination doses; toddlers between 1 and 2 years of age group, and neighbourhood. The DHI was constructed by who had been given at least 3 doses (proper primary Barcelona City Council Statistics Department [15]. vaccination), babies between 6 months and 1 year who had received at least 2 doses; babies between 3 and 6 Statistical methods months who had received at least 1 vaccination doses; We conducted a descriptive study of pertussis cases in all babies less than 3 months old [14]. The age groups Barcelona between 2000 and 2015, and calculated the used to analyse the vaccination variables were under 1 annual incidence proportion of pertussis in the city over- year old and under 6 years old. all, and according to sex, age group, DHI, and country of To determine the socio-economic status of people suf- origin (number of cases per 100,000 residents). Inci- fering from pertussis, we classified neighbourhoods of dence proportions were obtained by dividing the number residence -assigned according to the mailing address of notified pertussis cases by the total population in each provided in the survey- according to the index of dispos- group. We also computed the incidence trend line that able household income (DHI). The DHI had the follow- best fit the observed data. Incidence proportion rate ra- ing categories: very high (> 159), high (126–159), average tio (PRR) in foreign-born under 15 years old versus na- (100–125), average-low (between 79 and 99), and low tives was calculated. (below 79). The DHI was constructed by analysing the As subanalysis, we compared vaccination coverage be- following 5 indicators [15]: fore and after the economic recession. We also deter- mined the proportion of cases among children under 1 1. Percentage of people with a degree: number of year old who had received either whole-cell or acellular residents aged ≥25 years with a diploma, degree, or vaccination, and used univariate analysis to evaluate po- Ph.D. as a percentage of the total number of tential significant differences. residents aged ≥25 years To study the possible risk factors for the incidence of 2. Unemployment rate: number of unemployed people notified pertussis cases at the multivariate level, a gener- as a proportion of the total workforce alized linear model (GLM) was used assuming the fre- 3. Private cars/1000 residents: number of cars owned quency of the cases followed a Poisson distribution. We by neighbourhood residents*1000/total residents performed this analysis using aggregated data (with inci- 4. New high-power cars/total of new cars: number of dence as the dependent variable) to determine the effect cars less than two years old with more than 16 fis- of the following variables on the incidence trend: gender, cal horsepower/the total number of cars under two age group, DHI, and period. The model was adjusted for years old (owned by residents; fiscal horsepower re- the variables gender, age group, DHI and period. As the fers to a tax on high-powered vehicles in Spain) offset of this model the logarithm of the number of 5. Price of second-hand housing: Unit price of second- population exposures in each group was used. We ex- hand homes in euros. cluded 6 cases because they lacked data for the neigh- bourhood variable. Using the year variable, we created a new variable, All analyses were performed using SPPS v18.0 and the period, to divide the years of the study into three pe- statistical package R (The R Foundation for Statistical riods: 2000–2004, 2005–2010 (after vaccines were chan- Computing, Vienna, Austria), version 3.1.1. Statistical ged from whole-cell vaccines to acellular vaccines) and significance was set at p < 0.05. 2011–2015 (years of the epidemic). We also created a new variable to identify the pre-recession (2000–2008) Results and recession periods (2009–2015). A total of 1791 cases of pertussis were included, of which To evaluate the effect of the type of vaccine adminis- 6.3% were diagnosed between 2000 and 2004, 16.0% be- tered, we created a vaccination period variable and classi- tween 2005 and 2010, and 77.8% between 2011 and 2015. fied children under 1 year old in either the whole-cell There was 87.7% of cases classified as confirmed cases. vaccine period (2000–2004) or the acellular vaccine period Globally, 41% of cases were confirmed by diagnostic tests, (2005–2015), depending on the type of vaccine received. 13.9% by epidemiological contact, and 45.2% using both We also considered incidence differences between methods. The remaining 12.3% of cases were classified as children under 15 years old foreign-born and natives. suspected cases. 55.9% of the cases were female, 84.9% were natives, and we observed an increase in the average Information sources age of cases during the study period. Most cases were Data were obtained from epidemiological surveys carried from low-average and low DHI neighbourhoods. We also out by the Barcelona ND register. We used Barcelona observed an increase in the detection of associated cases Brugueras et al. BMC Public Health (2019) 19:144 Page 4 of 9 in the 2011–2015 period, as well as in the number of con- neighbourhoods. From 2011 onwards, incidence in- tact studies performed (Table 1). creased in all neighbourhoods, and remained higher in The following clinical symptoms were observed in those with lower DHI (Fig. 1c). cases: coughing that lasted longer than two weeks In terms of country of origin, we found that the inci- (77.8%), paroxysmal coughing (61.3%), inspiratory stri- dence in children under 15 years old was higher among dor (27.2%), post-coughing vomiting (29.5%), apnoea natives than among foreign-born (PRR = 0.43 CI: 0.32– (9.5%), fever (5.9%), pneumonia (0.9%), and convulsions 0.58) (Fig. 1d). (0.1%). There were three deaths during the study period, Multivariate analysis showed that the increased inci- one in 2008 and two in 2012. dence observed between 2000 and 2015 was associated In most cases, hospital admission involved children with being female (aOR = 1.12, CI:1.02–1.23) and being under 1 year (252/274, 92.0%), and there was a general a case during the 2011–2015 period (aOR = 1.75, decrease over time in the percentage of cases under 1 CI:1.54–2.00). We also observed trend related to year of age who were admitted to hospital (Table 1). age-group, with the strongest association among chil- Vaccination coverage remained high, although we dren less than 1 year old (aOR = 27.18, CI:23.51–31.44). found that the proportion of cases with unknown vac- In terms of socio-economic status, we did not observe cination status increased during the 2011–2015 period, a marked differences between neighbourhoods, although period in which cases were also found to be generally there was a general tendency (Table 3). older. More than 90% of children under 6 years (N = 791) were vaccinated, 88.9% were properly vaccinated Discussion according to their age, and 5.7% were non-vaccinated. In this study we have described evolution of pertussis in Note that parents who are against vaccines only Barcelona city between 2000 and 2015. We observed a accounted for 3.9% of all cases in children under 6 years, steady increase in incidence since 2010, which is in line and 68.9% of cases involving non-vaccinated children. with the trends observed elsewhere in Spain and in other Vaccination coverage before (91.3%) and during (90.6%) European countries [1, 7]. The trend in pertussis inci- the economic recession remained very similar (p = dence in recent decades follows a U-shaped curve, both 0.757). Analysing vaccination percentage according to in Spain and in other countries, such as the US and the DHI, we found that richer neighbourhoods (very high UK. This is a similar pattern to that described for other DHI) included a higher percentage of non-vaccinated communicable diseases, and requires greater focus on cases, as well as a higher percentage of non-vaccinated this disease and its control [6, 7, 9, 16]. Despite the peri- children under 6 years and under 1 year (p < 0.05). Simi- odicity of the disease we can see in other studies that larly, among vaccinated cases, the highest percentage of the last epidemic wave of pertussis shows an increasing incorrect vaccination was observed in very high DHI trend of the disease [6, 14, 17]. neighbourhoods in children under 6 years, and in high This increase does not seem to be related either to DHI neighbourhoods in children under 1 year (Table 2). limitations in healthcare assistance (healthcare for com- In the whole-cell vaccine period (2000–2004), 88.0% of municable diseases was maintained in Barcelona, despite cases under 1 year were vaccinated versus the 95.5% of the economic recession and Royal Decree 16/12 which cases under 1 year vaccinated in the acellular vaccine hinders access to the healthcare card among immigrants period (2005–2015) (p = 0.01). 82.2% of these were prop- [18]) or to vaccination issues (vaccination coverage in erly vaccinated according to their age between 2000 and Barcelona remained high over the years). The global rise 2004, and 94.6% in the 2005–2015 period (p < 0.001). in cases with unknown vaccination status was due to an We observed a sustained increase (R2 = 0.748 see expo- increase in the age of cases; such a rise is not observed nential curve, Fig. 1a) in the incidence of pertussis in Bar- in the under 6 years and under 1 year age groups. Al- celona 2011 onwards, from 3.1 cases/100,000 residents in though the action of the anti-vaccination movement 2000 to 16.2/100,000 in 2011, and, following a brief de- have been associated with a higher disease risk, this re- cline, a rise to 37.0 cases/100,000 in 2015. This trend was mains a minority group in Barcelona [19]. Several Span- similar in both males and females (Fig. 1a). ish studies have shown the need for new vaccination In terms of age, the highest incidence observed during strategies and more effective vaccines, given that a high the study period was among children under 1 year old. The percentage of cases is properly vaccinated [6, 14, 20]. incidence remained very low in the rest of the population Globally, several authors have suggested that vaccination until 2011, after which we observed an increase among 1- strategies should be reformulated, as a result of their to 14-year-olds that persisted until 2015 (Fig. 1b). limited efficacy [5, 7, 21]. Grouping cases according to DHI, we did not observe The percentage of cases under 1 year of age who re- a stable distribution during years of low incidence, al- ceived the vaccine was lower in 2000–2004 (when the though the incidence was generally higher in poorer whole-cell vaccine was used) than in 2005–2015 (in Brugueras et al. BMC Public Health (2019) 19:144 Page 5 of 9 Table 1 Distribution of pertussis cases according to epidemiological variables 2000–2004 N (%) 2005–2010 N (%) 2011–2015 N (%) Total N (%) All 112 (6.3) 286 (16.0) 1393 (77.8) 1791 (100.0) Confirmed 97 (86.6) 220 (76.9) 1253 (89.9) 1570 (87.7) Suspected 15 (13.4) 66 (23.1) 140 (10.1) 221 (12.3) Gender Female 69 (61.6) 160 (55.9) 772 (55.4) 1001 (55.9) Male 43 (38.4) 124 (43.4) 621 (44.6) 788 (44.0) Age group < 1 year 83 (74.1) 128 (45.1) 221 (15.9) 432 (24.1) 1–14 years 20 (17.9) 89 (31.3) 782 (56.1) 891 (49.7) 15–29 years 6 (5.4) 22 (7.7) 63 (4.5) 91 (5.1) > 30 years 3 (2.7) 45 (15.8) 327 (23.5) 375 (20.9) Country of origin Native 83 (74.1) 220 (76.9) 1217 (87.4) 1520 (84.9) Foreign-born 7 (6.3) 26 (9.1) 122 (8.8) 155 (8.7) Unknown 22 (19.6) 40 (14.0) 54 (3.9) 116 (6.5) Index of Disposable Household Income (DHI) Very high 4 (3.6) 15 (5.2) 77 (5.5) 96 (5.4) High 4 (3.6) 18 (6.3) 77 (5.5) 99 (5.5) Average 19 (17.0) 58 (20.3) 325 (23.3) 402 (22.4) Low-average 41 (36.6) 90 (31.5) 498 (35.8) 629 (35.1) Low 41 (36.6) 105 (36.7) 413 (29.6) 559 (31.2) Unknown 3 (2.7) 0 (0.0) 3 (0.2) 6 (0.3) Case type Associated 18 (16.1) 127 (44.4) 824 (59.2) 969 (54.1) Sporadic 94 (83.9) 159 (55.6) 569 (40.8) 822 (45.9) Contact tracing Yes 36 (32.1) 216 (75.5) 1130 (81.1) 1382 (77.2) No 76 (67.9) 70 (24.5) 263 (18.9) 409 (22.8) Vaccination Yes 92 (82.1) 236 (82.5) 1005 (72.1) 1333 (74.4) No 8 (7.1) 25 (8.7) 97 (7.0) 130 (7.3) Unknown 12 (10.7) 25 (8.7) 291 (20.9) 328 (18.3) Hospital admission in < 1 year (N = 432) Yes 64 (77.1) 83 (64.8) 105 (47.5) 252 (58.3) No 19 (22.9) 45 (35.2) 116 (52.5) 180 (41.7) Barcelona, 2000–2015 which the acellular vaccine was used). Moreover, the lat- profile of currently vaccines [24]. The increased incidence ter period was characterised by a higher proportion of could also be due to faster, more sensitive diagnostic tests, properly vaccinated cases. These results agree with the improved epidemiological surveillance, better registry sys- hypothesis of lower efficacy of the acellular vaccine than tems, and greater awareness by healthcare professionals the whole-cell vaccine [7, 22, 23]. [3, 7]. However, US- and UK-based study suggests that Apart from the lower efficacy of the acellular vaccine, asymptomatic transmission could be key for the the increased incidence could be due to a genetic change re-emergence of pertussis [25]. Moreover, pertussis often in the bacteria. Bacterial antigens are known to change goes unrecognized among patients. The lack of character- over time, and they no longer coincide with the antigenic istic symptoms is more common in adults, as opposed to Brugueras et al. BMC Public Health (2019) 19:144 Page 6 of 9 Table 2 Vaccination status in pertussis cases according to age group and disposable household income. Barcelona, 2000–2015 Age group Vaccinated N (%) Non-vaccinated N (%) Unknown vaccination Proper vaccination in status N (%) vaccinated cases N (%) All categories (N = 1791) Overall 1333 (74.4) 130 (7.3) 328 (18.3) 1146 (86.2) < 6 years 718 (90.8) 45 (5.7) 28 (3.5) 635 (88.9) < 1 year 406 (94.0) 15 (3.5) 11 (2.5) 371 (92.3) Very high DHI (N = 96) Overall 65 (67.7) 12 (12.5) 19 (19.8) 52 (80.0) < 6 years 37 (78.7) 8 (17.0) 2 (4.3) 31 (83.8) < 1 year 16 (76.2) 4 (19.1) 1 (4.8) 15 (93.8) High DHI (N = 99) Overall 73 (73.7) 11 (11.1) 15 (15.2) 60 (82.2) < 6 years 36 (85.7) 4 (9.5) 2 (4.8) 32 (88.9) < 1 year 18 (94.7) 0 (0.0) 1 (5.3) 15 (83.3) Average DHI (N = 402) Overall 280 (69.7) 29 (7.2) 93 (23.1) 239 (85.4) < 6 years 148 (88.6) 9 (5.4) 10 (6.0) 127 (85.8) < 1 year 95 (94.1) 3 (3.0) 3 (3.0) 87 (91.6) Average-low DHI (N = 629) Overall 464 (73.8) 42 (6.7) 123 (19.6) 401 (86.4) < 6 years 236 (91.1) 15 (5.8) 8 (3.1) 212 (89.8) < 1 year 120 (94.5) 5 (3.9) 2 (1.6) 109 (90.8) Low DHI (N = 559) Overall 447 (80.0) 36 (6.4) 76 (13.6) 394 (88.1) < 6 years 257 (95.2) 9 (3.3) 4 (1.5) 233 (90.7) < 1 year 153 (96.8) 3 (1.9) 2 (1.3) 145 (94.8) DHI disposable household income children. In this sense, a review of studies from several despite correct vaccination coverage [4, 9, 31–33]. This countries showed the high seroprevalence of B. pertussis change highlights a need for a booster dose in vaccin- compared to the number of diseases that were notified ation schedules during teenage years. [26]. This findings emphasise the need for better and The decline in the number of cases under 1 year of age maintained epidemiological surveillance in the entire who were admitted to hospital could be because population. formerly only the most severe cases were notified, be- While children under 1 year are the most vulnerable cause of improved hospital admission criteria, or be- age group, and the group with the highest disease inci- cause the disease appears in less severe forms that do dence, we observed an increase among children older not require hospitalization. This could also be related to than 1 year and among teenagers. This increase in the earlier more precise diagnoses, and the use of faster, age of cases may be due to greater awareness among more sensitive diagnostic techniques. healthcare professionals, leading to increased diagnosis Moreover, native children aged 0–14 years show higher among teenagers and adults (symptoms are less severe disease incidence than among foreign-born children, in in these groups and thus harder to identify). The ab- spite of the economic recession. This could be because sence of characteristic symptoms results in greater diag- low-income countries still use the whole-cell vaccine, nosis delay, such that other individuals are exposed to which provides a longer-lasting immunity. In addition, infected individuals for longer. This increase in age immigrant children from countries with a high incidence could also be due to reduced acquired immunity over of pertussis would have higher natural immunity. time, which is attributed to the change in vaccine from Examining neighbourhoods according to DHI, we found whole-cell to an acellular vaccine and to reduced natural that those with very high DHI had a lower incidence of immunity after several years of low disease incidence, pertussis. However, although these neighbourhoods have which does not allow immunity to be strengthened fewer cases, they show a higher percentage of through naturally occurring disease [3, 6, 22, 23, 27, 28]. non-vaccinated and not correctly vaccinated cases. Thus, This decrease in immunity may have caused increased some cases of infection may have occurred because of im- transmission between older children, teenagers and proper vaccination. As for all communicable disease, most adults, who would in turn transmit the disease to youn- cases of pertussis cluster in disadvantaged neighbour- ger children [2, 29, 30]. Other studies have also de- hoods in large cities, where poorer living conditions such scribed increased reporting of cases at older ages, as poor hygiene or overcrowding in small apartments Brugueras et al. BMC Public Health (2019) 19:144 Page 7 of 9 A C B D Fig. 1 Incidence of pertussis in Barcelona, 2000–2015. a Incidence of pertussis in Barcelona according to sex, 2000–2015. b Incidence of pertussis in Barcelona according to age group, 2000–2015. c Incidence of pertussis in Barcelona according to index of Disposable Household Income, 2000–2015. d Incidence of pertussis among under 15-year-olds according to country of origin, Barcelona 2011–2015 leads to increased disease transmission [34–36]. However, note that health-related actions should not be limited to these neighbourhoods, as multivariate analysis shows that Table 3 Analysis of factors associated with pertussis incidence all neighbourhoods in the city are at risk of a similar in- in Barcelona, 2000–2015 crease in incidence. This shows that, despite the economic Variables Categories cOR 95%CI aOR 95%CI recession, neighbourhoods with low DHI have not been disproportionately affected. Gender Male 1 1 Our multivariate analysis showed a remarkable associ- Female 0.87 0.80–0.96 1.12 1.02–1.23 ation between the pertussis and children under 1 year of Age group < 1 year 23.13 20.13–26.57 27.18 23.51–31.44 age. Thus, this age group should be the focus of resources 1–14 years 6.64 5.89–7.50 6.80 6.03–7.69 and future health-related actions, and it would be also be 15–29 years 1.70 1.34–2.12 1.76 1.39–2.20 necessary to promote vaccination in pregnant women. > 30 years 1 1 Although economic recessions have previously been linked to negative impacts on child health [37], the DHI V.high-high 1 1 current recession does not seem to have affected access Average 1.21 1.02–1.44 1.31 1.10–1.55 to the healthcare system or vaccination coverage, and Average-low 1.63 1.39–1.91 1.64 1.40–1.94 foreign-born children were not more seriously afflicted Low 1.71 1.46–2.02 1.62 1.37–1.91 than native children. Thus, the increase detected is due Period 2000–2004 2.23 1.78–2.77 1.11 0.88–1.38 to other factors mentioned above, such as vaccine effect- 2005–2010 1 1 iveness, reduced immunity, gene changes in B. pertussis, and the availability of new diagnostic tests. 2011–2015 1.24 1.09–1.41 1.75 1.54–2.00 A possible limitation of the study is under-diagnosis and Adjusted for the variables gender, age group, DHI and period DHI disposable household income, cOR crude odds ratio, aOR adjusted odds under-notification of the disease, mostly in adult popula- ratio, CI confidence interval tion, as well as the absence of data on their vaccination Brugueras et al. BMC Public Health (2019) 19:144 Page 8 of 9 status. Nonetheless, this is a population-based study of Guasch, Maria Rosa Pascual (Centre d’Atenció Primària Via Barcino, Barcelona, Barcelona city that includes 16 years of data from the Spain), Carme Palasi (Centre d’Atenció Primària Sardenya, Barcelona, Spain), Anna Luz Moro (Centre d’Atenció Primària Larrard, Barcelona, Spain), Manel BPHA registry. In addition, socio-economic level was de- Enrubia (Centre d’Atenció Pediàtrica Integral Barcelona Esquerra, Barcelona, termined using a DHI indicator created by Barcelona City Spain), Gemma Monsó (Centre d’Atenció Primària Rio de Janeiro, Barcelona, Council. Given the high number of cases and their Spain), Joan Martí Solé (Centre d’Atenció Primària Sant Andreu, Barcelona, Spain), Paloma Rodríguez, Helena Pallarès, Antoni Salvà (Consorci d’Atenció follow-up, we consider this to be a valid study of consider- Primària de Salut Barcelona Esquerra, Barcelona, Spain), Ángeles Ferrández able scientific interest. (Centre d’Atenció Primària Ciutat Meridiana, Barcelona, Spain), Ana María Hostalot (Centre d’Atenció Primària Pare Claret, Barcelona, Spain), María Carmen Muñoz (Hospital Sant Joan de Déu, Barcelona, Spain). Corresponding Conclusions autor: Anna de Andrés; aandres@aspb.cat The noteworthy increase in pertussis incidence does not seem to be due to the economic recession, but to other Ethical approval and consent to participate All data were treated and analysed according to the ethical guidelines for factors such as vaccine effectivity or genetic change in human experimentation defined in the Declaration of Helsinki (1964) and the bacteria. The increased incidence could also be due reviewed by the World Health Organization in Fortaleza (Brazil) in 2013 [38]. to faster, more sensitive diagnostic tests, improved epi- Data and registers were made anonymous to guarantee their confidentiality. The study complies with Spanish Organic Law on Personal Data Protection demiological surveillance, better registry systems and (15/1999) [39]. greater awareness by healthcare professionals. Another This study was evaluated and approved by the Clinical Research Ethics key factor could be asymptomatic transmission and the Committee of Parc de Salut Mar (CEIC – Parc Salut Mar, project 2013/5246/I). Consent to participate was not applicable for this study. absence of characteristic symptoms among infected people. Although we observed an increase in the age of Funding cases, children under 1 year continue to be the most vul- No additional funding was required for the preparation of this article. nerable age group, and the group with the highest dis- ease incidence. According to the index of DHI, we Availability of data and materials The datasets generated and/or analysed during the current study are not observed that neighbourhoods with low DHI have not publicly available due to the data belong to the Barcelona Public Health been disproportionately affected. However, pertussis in- Agency that is a public institution where data of residents of Barcelona are cidence increased from 2011 onwards in all neighbour- collected and kept. According to the data protection law and the regulations of the entity, data can not be disseminated or shared but are available from hoods and remained higher in those with lower DHI. the corresponding author on reasonable request. In conclusion, we consider that the following measures should be undertaken to improve the epidemiology of the Authors’ contributions disease: (i) prioritize preventive measures to prevent trans- SB participated in the design and development of the study, carried out the data analysis and interpretation, and drafted the manuscript. JPM formulated mission to most vulnerable groups including children the idea, participated in the design of the study and supervised the analysis under 1 year such as vaccinate pregnant women; (ii) inten- and interpretation of the results, and the drafting of the manuscript. CR sify epidemiological surveillance in the entire population. participated in the study design and supervised the development of the study, the analysis and the drafting of the manuscript. MC performed the statistical Abbreviations analyses and interpretation and commented on the manuscript. JAC aOR: adjusted odds ratio; BPHA: Barcelona Public Health Agency; participated in the idea and design, data analysis and interpretation, and CI: confidence intervals; cOR: crude odds ratio; DHI: disposable household critically revised the manuscript. The Barcelona Pertussis Working Group income; ECDC: European Centre for Disease Prevention and Control; participated in the acquisition of data through the diagnostic, treatment, follow EU: European Union; GLM: generalized linear model; ND: Notifiable Disease; up and notification of the pertussis cases, entered the data in the database, OR: odds ratio; PCR: Polymerase Chain Reaction; PRR: Incidence proportion revised the content of the manuscript critically and supervised and approved rate ratio; UK: United Kingdom; US: United States the draft of the manuscript. All authors read and approved the final manuscript. Acknowledgements Consent for publication The authors would like to thank all the public health nurses of BPHA for their Not applicable. support and contributions to this study. This study would not have been possible without the participation of the committed team from the BPHA. Competing interests We would also like to extend our thanks to the Barcelona Pertussis Working The authors declare that they have no competing interests. Group for his participation and assistance in this project. We are very grateful for the reporting of pertussis cases done in the years 2000 to 2015 by the health care centres and hospitals of Barcelona. Finally, we also would like to Publisher’s Note thank to Tomas Perez-Porcuna, Andrea Martin, Toni Soriano and Antoni Springer Nature remains neutral with regard to jurisdictional claims in Noguera for their contributions in the discussion of the results. published maps and institutional affiliations. Barcelona Pertussis Working Group: Pere Simón, Anna de Andrés, Roser Clos, Montse Ricart, Miriam Ros, Maria José Santomà, Pilar Gorrindo, Pilar Palau, Author details 1 Montse Cunillé, Eva Masdeu, Ingrid Avellanés (Agència de Salut Pública de Servei d’epidemiologia, Agència de Salut Pública de Barcelona, Barcelona, Barcelona, Barcelona, Spain), Jose Ángel Rodrigo, Gema Codina, Maria Teresa Spain. 2Centro de Investigación Biomédica en Red de Epidemiología y Salud Martin, Juan José González (Hospital Universitari Vall Hebrón, Barcelona, Pública (CIBERESP), Barcelona, Spain. 3Departamento de Pediatría, Obstetricia Spain), Sònia Brió (Hospital de la Santa Creu i Sant Pau, Barcelona, Spain), y Ginecología y Medicina Preventiva, Facultad de Medicina, Universidad Jessica Del Carpio, Andrea Otero (Hospital Universitari Dexeus, Barcelona, Autónoma de Barcelona, Barcelona, Spain. 4Tuberculosis Research Unit Spain), Álvaro Díaz (Hospital de Nens de Barcelona, Barcelona, Spain), Núria Foundation (fuiTB), Barcelona, Spain. 5Sport and Physical Activity Studies López (Hospital del Mar, Barcelona, Spain), Montse Vaqué, Pere Sala, Marta Centre (CEEAF), University of Vic-Central University of Catalonia (UVic-UCC), Calsina, Yolanda Meije (Hospital de Barcelona, Barcelona, Spain), Eulàlia Barcelona, Spain. Brugueras et al. BMC Public Health (2019) 19:144 Page 9 of 9 Received: 1 February 2018 Accepted: 17 January 2019 22. Sheridan SL, Ware RS, Grimwood K, Lambert SB. Reduced risk of pertussis in whole-cell compared to acellular vaccine recipients is not confounded by age or receipt of booster-doses. Vaccine. 2015;33(39):5027–30. 23. Fulton TR, Phadke VK, Orenstein WA, Hinman AR, Johnson WD, Omer SB. References Protective effect of contemporary pertussis vaccines: a systematic review 1. European Centre for Disease Prevention and Control. Annual and meta-analysis. Clin Infect Dis. 2016;62(9):1100–10. epidemiological report 2014 - vaccine-preventable diseases. Stockholm: 24. van Gent M, Heuvelman CJ, van der Heide HG, Hallander HO, Advani A, ECDC; 2014. Guiso N, et al. Analysis of Bordetella pertussis clinical isolates circulating in 2. Wendelboe AM, Njamkepo E, Bourillon A, Floret DD, Gaudelus J, Gerber M, European countries during the period 1998-2012. Eur J Clin Microbiol Infect et al. Transmission of Bordetella pertussis to young infants. Pediatr Infect Dis Dis. 2015;34(4):821–30. J. 2007;26(4):293–9. 25. Althouse BM, Scarpino SV. Asymptomatic transmission and the resurgence 3. Tan T, Dalby T, Forsyth K, Halperin SA, Heininger U, Hozbor D, et al. Pertussis of Bordetella pertussis. BMC Med. 2015;13:146. across the globe: recent epidemiologic trends from 2000 to 2013. Pediatr 26. Barkoff AM, Gröndahl-Yli-Hannuksela K, He Q. Seroprevalence studies of Infect Dis J. 2015;34(9):e222–32. pertussis: what have we learned from different immunized populations. 4. Heininger U, André P, Chlibek R, Kristufkova Z, Kutsar K, Mangarov A, et al. Pathog Dis. 2015;73(7). Comparative epidemiologic characteristics of pertussis in 10 central and 27. Schwartz KL, Kwong JC, Deeks SL, Campitelli MA, Jamieson FB, Marchand- eastern European countries, 2000-2013. PLoS One. 2016;11(6):e0155949. Austin A, et al. Effectiveness of pertussis vaccination and duration of 5. Sabbe M, Vandermeulen C. The resurgence of mumps and pertussis. Hum immunity. CMAJ. 2016;188(16):E399–406. Vaccin Immunother. 2016;12(4):955–9. 28. Klein NP, Bartlett J, Rowhani-Rahbar A, Fireman B, Baxter R. Waning protection after fifth dose of acellular pertussis vaccine in children. N Engl J 6. Sizaire V, Garrido-Estepa M, Masa-Calles J, Martinez de Aragon MV. Increase Med. 2012;367(11):1012–9. of pertussis incidence in 2010 to 2012 after 12 years of low circulation in 29. de Greeff SC, Mooi FR, Westerhof A, Verbakel JMM, Peeters MF, Heuvelman Spain. Euro Surveill. 2014;19(32). CJ, et al. Pertussis disease burden in the household: how to protect young 7. Sealey KL, Belcher T, Preston A. Bordetella pertussis epidemiology and evolution infants. Clin Infect Dis. 2010;50:1339–45. in the light of pertussis resurgence. Infect Genet Evol. 2016;40:136–43. 30. van der Maas NA, Mooi FR, de Greeff SC, Berbers GA, Spaendonck MA, de 8. Domenech de Cellès M, Magpantay FM, King AA, Rohani P. The pertussis Melker HE. Pertussis in the Netherlands, is the current vaccination strategy enigma: reconciling epidemiology, immunology and evolution. Proc Biol sufficient to reduce disease burden in young infants? Vaccine. 2013;31(41): Sci. 2016;283(1822). 4541–7. 9. Clark TA. Changing pertussis epidemiology: everything old is new again. J 31. Crespo I, Cardeñosa N, Godoy P, Carmona G, Sala MR, Barrabeig I, et al. Infect Dis. 2014;209:978–81. Epidemiology of pertussis in a country with high vaccination coverage. 10. Rechel B, Suhrcke M, Tsolova S, Suk JE, Desai M, McKee M, et al. Economic Vaccine. 2011;29:4244–8. crisis and communicable disease control in Europe: a scoping study among 32. Hellenbrand W, Beier D, Jensen E, Littmann M, Meyer C, Oppermann H, national experts. Health Policy [Internet]. Elsevier Ireland Ltd; 2011 Dec et al. The epidemiology of pertussis in Germany: past and present. BMC [cited 2017 January 23];103(2–3):168–75. Available from: http://www.ncbi. Infect Dis. 2009;9:22. nlm.nih.gov/pubmed/21820196. 33. de Greeff SC, de Melker HE, van Gageldonk PGM, Schellekens JFP, van der 11. Suhrcke M, Stuckler D, Suk JE, Desai M, Senek M, McKee M, et al. The impact Klis FRM, Mollema L, et al. Seroprevalence of pertussis in the Netherlands: of economic crises on communicable disease transmission and control: a evidence for increased circulation of Bordetella pertussis. PLoS One. 2010;5: systematic review of the evidence. PLoS One [Internet]. 2011 Jan [cited e14183. ];6(6):e20724. 34. Suhrcke M, Stuckler D, Suk JE, Desai M, Senek M, McKee M, et al. The impact 12. Ajuntament de Barcelona. Departament d'Estadística. Available at: http:// of economic crises on communicable disease transmission and control: a www.bcn.cat/estadistica/angles/index.htm systematic review of the evidence. PLoS One. 2011;6(6):e20724. 13. Generalitat de Catalunya. Departament de Salut. Definició de cas de les 35. Levri KM, Reynolds L, Liko J, Dott M, Robinson BF, Cieslak PR. Risk factors for malalties de declaració obligatòria. 11a ed. Barcelona: Generalitat de pertussis among Hispanic infants: metropolitan Portland, Oregon, 2010- Catalunya; 2006. Available at: https://scientiasalut.gencat.cat/bitstream/ 2012. Pediatr Infect Dis J. 2016;35(5):488–93. handle/11351/1089/definicio_malalties_declaracio_obligatoria_2010. 36. Zamir CS, Dahan DB, Shoob H. Pertussis in infants under one year old: risk pdf?sequence=1 markers and vaccination status--a case-control study. Vaccine. 2015;33(17): 14. Moraga-Llop FA, Campins-Martí M. Vacuna de la tos ferina. Reemergencia 2073–8. de la enfermedad y nuevas estrategias de vacunación. Enferm Infecc 37. Rajmil L, Fernandez de Sanmamed MJ, Choonara I, Faresjö T, Hjern A, Kozyrskyj Microbiol Clin. 2015;33(3):190–6. AL, et al. Impact of the 2008 economic and financial crisis on child health: a 15. Martí C, Casas E, Jorge Juan A, Martí J, Conde J, Figuerola G, et al. systematic review. Int J Environ Res Public Health. 2014;11(6):6528–46. Distribució territorial de la Renda Familiar per càpita a Barcelona. 38. Declaración de Helsinki de la AMM. Principios éticos para las investigaciones Ajuntament de Barcelona; 2007. Available at: http://www.bcn.cat/ médicas en seres humanos. [Internet]. Asociación Médica Mundial; 2013. publicacions/pdf/rfd.pdf (accessed 14 Oct 2016). Available at: http://www.isciii.es/ISCIII/es/contenidos/fd-investigacion/fd- 16. Reichman LB. The U-shaped curve of concern. Am Rev Respir Dis. 1991; evaluacion/fd-evaluacion-etica-investigacion/Declaracion-Helsinki-2013-Esp.pdf 144(4):741–2. 39. Ley Orgánica 15/1999, de 13 de diciembre, de Protección de Datos de 17. Broutin H, Guégan JF, Elguero E, Simondon F, Cazelles B. Large-scale Carácter Personal. [Internet]. Madrid: Noticias jurídicas; 2011. Available at: comparative analysis of pertussis population dynamics: periodicity, http://noticias.juridicas.com/base_datos/Admin/lo15-1999.html synchrony, and impact of vaccination. Am J Epidemiol. 2005 Jun 15;161(12): 1159–67. 18. Real Decreto-ley 16/2012, de 20 de abril, de medidas urgentes para garantizar la sostenibilidad del Sistema Nacional de Salud y mejorar la calidad y seguridad de sus prestaciones. (Boletín Oficial del Estado, número 98, de 24 de abril de 2012). 19. Salmon DA, Dudley MZ, Glanz JM, Omer SB. Vaccine Hesitancy. Causes, Consequences, and a Call to Action. Am J Prev Med. 2015;49(6S4):S391–8. 20. Moraga-Llop FA, Mendoza N, Muntaner A, Codina G, Fàbregas A, Campins M. Pertussis in fully vaccinated infants and children. Are new vaccination strategies required? Enferm Infecc Microbiol Clin. 2014;32(4):236–41. 21. Bolotin S, Harvill ET, Crowcroft NS. What to do about pertussis vaccines? Linking what we know about pertussis vaccine effectiveness, immunology and disease transmission to create a better vaccine. Pathog Dis. 2015;73(8).