Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
Bordetella pertussis continues to circulate and cause disease even in populations with high vaccination coverage of infants and children. Waning of vaccine-induced immunity1–4 is cited as an important factor contributing to this persistent problem.5–8 Despite widespread vaccination coverage (ie, 95% for the primary series9,10), the reported incidence of infant pertussis in the United States tripled in the past 2 decades from 34.2 cases/100,000 in the 1980s to 103.5 cases/100,000 infant population in 2003.11,12 Infants too young to have completed their primary vaccine series account for the majority of pertussis related complications, hospitalizations and deaths.12,13
One strategy for protecting infants is to increase herd immunity by vaccinating close contacts. France was the first country to introduce a booster dose for adolescents in 1998,14 followed by Germany, other European countries, Canada, Australia and Japan.15 France and Germany have also recommended a booster for parents and health care workers in contact with young children.14,16 The United States Advisory Committee on Immunization Practices (ACIP) recently recommended that all adolescents and adults receive a single booster dose of Tdap (tetanus toxoid, reduced diphtheria toxoid and acellular pertussis vaccine) in place of Td (tetanus and diphtheria toxoids vaccine).17,18 Modeling has shown that the effect on infant pertussis of routinely vaccinating all adolescents and adults critically depends on the proportion of infants infected by family members.19
Several studies have investigated the transmission of pertussis within households using a variety of study designs: case control studies in outbreak settings,20–22 investigations using general and hospital-based surveillance data,5,23–25 and secondary analyses using vaccine efficacy trial data.26 These studies reported that parents (20%–48%)5,23–25,27 and siblings (19%–53%)5,23–25 were common sources of infection for infants for whom a source was identified. However, interpretation of these findings is subject to important limitations. In these studies, the primary source of infection could not be identified in 47%–60% of infant index cases. Further, except in one small study,27 the diagnosis of the index case was not always laboratory confirmed, and identification of source cases relied predominantly on a clinical diagnosis of pertussis. It remains therefore unclear who transmits pertussis to young infants.
This study was designed to determine more precisely who transmits B. pertussis to young infants. We collected comprehensive clinical and biologic diagnostic data on close contacts to infant cases with laboratory confirmed pertussis.
Study Setting and Population.
A hospital based prospective study was conducted in France (4 hospitals), Germany (4 hospitals), the United States (3 hospitals) and Canada (1 hospital) from February 2003 to October 2004 (see list of participating hospitals of the Infant Pertussis Study Group). Data were analyzed at the University of North Carolina at Chapel Hill. Institutional Review Boards at each hospital and UNC approved the study.
Three categories of participants were included in the study: infant index cases, household contacts and close nonhousehold contacts. Index cases were eligible if the infants were aged ≤6 months, unvaccinated or partially vaccinated (<3 doses) and diagnosed with laboratory confirmed pertussis (confirmed by culture or polymerase chain reaction [PCR]). Household contacts of the index cases were defined as persons living in the same residence as the infant index case during the month preceding pertussis diagnosis in the infant. Nonhousehold contacts were eligible if, during the month before symptom onset in the index case, they were either a full-time caretaker (>30 hours/week), or a person with an acute cough illness lasting ≥7 days and had at least 10 hours of contact per week. Potential participants were excluded if their residence was too distant for the family to attend for evaluation.
Upon enrollment of the infant index case, data on age, gender, pertussis vaccination status, type and duration of pertussis symptoms, laboratory test results, family composition, and type of child care were collected. Enrollment of the index case and contacts was required to occur within one week after the diagnosis of pertussis.
All household members of the index case were recruited upon informed consent and interviewed face-to-face using a standard questionnaire to obtain demographic and clinical data to identify potential nonhousehold sources of pertussis. A respiratory sample (nasopharyngeal aspirate or swab) for culture and PCR detection of B. pertussis and a blood sample for IgG antipertussis toxin (anti-PT) antibody detection were collected. One month later, follow-up data on symptoms and a convalescent blood sample were collected. Potential nonhousehold sources of infection were contacted by telephone, recruited, and followed the same study protocol as household contacts.
Diagnostic Laboratory Procedures and Quality Control.
PCR and serum samples were sent frozen to the Reference Laboratory for Whooping Cough and other Bordetelloses, Institut Pasteur in Paris. PCR samples were analyzed by real-time PCR using the IS481 target according to recommendations of the PCR pertussis consensus group.28 Serum samples were analyzed using a standardized enzyme-linked immunosorbent assay (ELISA) to quantify IgG anti-PT, as previously described,29 and using purified PT from sanofi pasteur. Assay cut-off was set at ≥20 EL U/mL.
Contacts were classified into one of 4 pertussis outcome categories: (1) laboratory confirmed symptomatic case, (2) epidemiologically linked symptomatic case, (3) laboratory confirmed asymptomatic infection, and (4) not infected. Outcomes were assigned using a case definition algorithm (Fig. 1 available online only). Laboratory confirmed symptomatic cases reported symptoms of rhinorrhea or cough and met at least one of the following criteria: positive culture, positive PCR,28 a ≥4-fold change30 in anti-PT IgG levels in paired serum samples, or a single anti-PT IgG antibody titer ≥125 EU/mL.31–33 For immunized children aged 3 months through 2 years, or 4 years through 7 years, single anti-PT titer results were not used for laboratory confirmation because recent vaccination may have influenced the IgG titers.8
Epidemiologically linked cases were persons in contact with the index case in the month preceding symptom onset in the index case and who had an acute cough illness lasting ≥2 weeks, and had no laboratory confirmed pertussis.34
Laboratory confirmed asymptomatic cases met the same criteria as laboratory confirmed cases, but did not report any cough or cold symptoms at either visit. Those classified as uninfected failed to meet any of the above criteria.
A primary case was defined as a laboratory confirmed or epidemiologically-linked case with symptom onset between 7 and 30 days before symptom onset in the index case.25,35–39
The primary analysis consisted of a complete case analysis in which only contacts with complete critical diagnostic information (ie, PCR and serology) were included. Multiple source cases were allowed for an index case when more than one laboratory confirmed index case reported symptom onset on the same day.
Sensitivity Analyses Investigating the Effect of Missing Data.
To determine the impact of missing data on the study results 2 sensitivity analyses were conducted. In a highly conservative analysis, the strict complete enrollment analysis, the entire family was included only if every household member was enrolled and critical diagnostic results were available for every contact. In a less restrictive analysis, the all participant analysis, data on all participants were analyzed. When PCR and/or serology results were missing, pertussis outcome was assigned using the case definition algorithm (Fig. 1, available online only).
Sensitivity Analyses Investigating Infants for Whom a Primary Case was not Identified.
Additional sensitivity analyses investigated why the primary analysis failed to identify a source case for some infants despite complete enrollment and diagnostic investigation. In the expanded primary case definition analysis, the timing of symptom onset to meet the primary case definition was expanded to 2–48 days before symptom onset in the index case based on evidence from early transmission studies of pertussis.40,41 In the asymptomatic case analysis, contacts with asymptomatic laboratory confirmed infection were included as potential source cases42 in families where no symptomatic primary case (using the expanded primary case definition) was identified.
Data management and statistical analyses using Mantel-Haenzel χ2 tests were conducted in SAS 9.1.3 (Cary, NC). Ninety-five percent confidence intervals were calculated using Stata 8 (College Station, TX).
Ninety-five infant index cases (France = 15, Germany = 16, United States = 62, Canada = 2) and 404 contacts (France = 60, Germany = 67, United States = 269, Canada = 8,) were enrolled. Of the eligible index cases and household contacts, 84.1% and 86.0%, respectively, participated. A mean of 4.2 (range = 1–13) contacts were enrolled per index case, and 29 (30.5%) of the index cases had at least one nonhousehold contact enrolled. There were 206 adult household members (aged ≥18 years, except in the case of emancipated minors), 127 child household members (aged <18 years) and 71 nonhousehold contacts enrolled.
The mean age of the infant index cases was 2.9 months (range = 0.53–6.9 months); 70.5% were vaccinated at the appropriate age (ie, had received the correct dose no later than 30 days past the country specific scheduled date). Thirty-four (35.8%) infants had received one dose of pertussis vaccine and 8 (8.4%) had received 2 doses. Associations between age and number of received vaccine doses with being hospitalized are presented in Table 1.
The distributions of collected diagnostic specimens for PCR and serology among close contacts to the infant index cases, contact level with the index case, and vaccination history stratified by type of contact are presented in Table 2.
Among the 95 families enrolled, 3 had no contacts for whom complete data were available; one family had twin infant index cases with identical contact patterns and was analyzed as a single case. Of the remaining 91 (96.8%) unique families, 91 index cases and 347 (85.9%) contacts had complete data. The prevalence of pertussis was 65.1% among all close contacts included in this analysis. The distribution of case-type is shown in Table 2. A source of infection was identified for 44 (48.4%) index cases. The majority (79.5%) of primary cases were laboratory confirmed. For 4 index cases, multiple primary cases were assigned as more than one laboratory confirmed primary case reported symptom onset on the same day.
Among the 44 infants for whom a primary case was identified, household members were more often identified as primary cases than nonhousehold contacts (75.5% versus 24.5%). Parents accounted for 55.1% of the primary cases identified (18 mothers and 9 fathers), followed by siblings (16.3%), aunts/uncles (10.2%), friends/cousins (10.2%), grandparents (6.1%) and part-time caretakers (2.0%) (Table 3a). Among the 8 siblings identified as primary cases, 4 (two 3 year olds, a 9-year-old and an 11-year-old) had a documented history of age appropriate pertussis immunization, one 15-year-old reported no previous vaccination and 3 siblings (aged 4 years, 17 years and 23 years) reported unknown vaccination status. Five of these cases were from the United States, 2 from Germany and one from France.
When calculating the proportion of source cases among all index cases, parents were identified as the source case for 30% of infants, siblings 9%, aunts/uncles 5%, friends/cousins 5%, grandparents 3% and part-time caretakers 1%.
Among the 47 (51.6%) index cases for whom no source case could be identified, complete enrollment and complete laboratory diagnostic tests for all household and eligible nonhousehold contacts were available for 21 (44.7%), indicating potential transmission from a casual contact. For the remaining 26 families, there was at least one identifiable close contact who did not participate in the study, limiting our ability to draw conclusions regarding the source of infection for the infant.
Sensitivity Analyses Investigating the Effect of Missing Data.
In the conservative strict complete enrollment analysis 49 (52.1%) families were excluded due to missing data for one or more contacts. Among the 45 infants and 193 contacts included in the analysis, a source case was identified for 62.2% of index cases. In the all participant analysis, all data for the 94 index cases and 404 contacts were included and a source case was identified for 53.2% of index cases. The differences in the distribution of relationships of the source cases in these sensitivity analyses and the primary analysis were small (0.2%–5.1%), with overlapping 95% confidence intervals. [Table 3b, available online only.]
Sensitivity Analyses Investigating Infants for Whom a Primary Case was not Identified.
In the expanded primary case definition analysis, a primary case was identified for an additional 12 (13.2%) index cases. Seven primary cases had symptoms 31–48 days before onset in the index case, and 5 primary cases had symptoms 2–6 days before onset in the index case. A primary case was thus identified for 61.5% of all infants, when compared with 48.4% in the primary analysis. The distribution of the relationships of the primary case to the index case remained similar. [Table 3b, available online only.]
When transmission from asymptomatically infected close contacts was assumed possible, a source case was identified for an additional 15 (16.5%) index cases beyond those identified when using the expanded case definition, 6 of which were confirmed by serology only, 5 by PCR only, and 4 by both PCR and serology. In this sensitivity analysis, a primary case was identified for 71 (78.0%) index cases. Again, the distribution of relationships to the infant index cases remained similar to that in the primary analysis. [Table 3b, available online only.]
Stratified Analysis by Continent.
Data from the European sites were compared with the North American sites. No differences were observed in the demographic characteristics at α = 0.05. The mean age of European parents was 32.6 years and the mean age of North American parents was 30.0 years (P = 0.07). The distributions of primary cases by relationship to the infant and by age were not statistically significantly different between the 2 continents (data not shown).
This study shows that parents (48%–55%), siblings (16%–21%) and nonhousehold close contacts (18%–29%) are important sources of pertussis transmission to young infants. Although these findings corroborate those from prior studies5,23–25,27 (Table 4), they were robust to sensitivity analyses as evidenced by the relatively small range (<8%) in point estimates among each stratum of source cases. Additionally it had fewer limitations than previous studies. The prospective study design limited recall error in reporting the timing of symptom onset in contacts. The collection of biologic specimens (PCR and serology) on all household members, irrespective of symptoms, allowed laboratory confirmation of symptomatic cases and identification of asymptomatic infection. Inclusion of close nonhousehold contacts allowed investigation of their role in pertussis transmission to young infants. Analysis of PCR and anti-PT ELISA at a reference laboratory avoided misclassification because of potential interlaboratory variation in test performance. Further, the relatively large sample size allowed us to draw robust conclusions about who transmitted infection to young infants.
Finally, in contrast to prior studies, we explored alternative hypotheses for index cases for whom no source could be identified. We evaluated the effect of missing data resulting from nonparticipation by household members, potential transmission from casual contact in the community, the effect of increasing the sensitivity of the definition for primary cases, and the potential role of individuals with asymptomatic laboratory confirmed infection in the transmission of pertussis to infants. We also expanded the period of infectiousness and reduced the incubation period based on data from Stocks,40 who used cough plates to determine the start and end of the infectious period among sequential cases within a household. In our study, we identified a substantial number of source cases with symptom onset 2–6 days and 31–48 days before onset in the index case. The finding that individuals with asymptomatic laboratory confirmed infection were the only pertussis cases among close contacts for 16.5% of index cases raises questions about the dogma that asymptomatic “carriers” do not transmit pertussis.43,44
These results are important for evaluating the impact of current vaccination strategies and predicting the effect of new vaccination strategies on infant pertussis. The study corroborates the findings from Olin et al45 that infant vaccination protects against hospitalization. Although it is vital to maintain high coverage rates and timely vaccination of infants and children against pertussis, we did not find that nonvaccination of siblings was a driving force for infant pertussis in this population. Although adolescents (aged 13–18) accounted for 16%–21% of the primary cases identified (10% were nonhousehold contacts and 6% were siblings in the primary analysis), their relative role in transmission may be greater, because 66% of the infants did not have an adolescent contact in this study. However, the results of this study suggest that vaccination of parents, especially mothers, is of primary importance for the control of infant pertussis. The effectiveness of vaccinating adolescents will depend on the proportion of families in which adolescents are in close contact with infants, and the vaccine coverage achieved.
In this population of young children, the role of nonhousehold caretakers as sources of pertussis was small as only one (2.0%) part-time caretaker and no full-time caretakers were identified as a source case. It is possible that individuals who served a role as informal caretakers identified themselves as relatives (aunts, grandparents). In addition, because only 8 (8.5%) infants enrolled had a caretaker, the relative role of caretakers as a source of pertussis for infants who do have a caretaker other than their parents may be greater than the absolute 2% identified in the study.
The generalization of these findings to other settings may be limited as the socioeconomic characteristics, epidemiologic patterns of pertussis, social contact patterns and immunization strategies may differ between countries. However, the finding that the results were not significantly different between the North American and European continents likely indicates the external validity of the study. Another limitation is the inability to determine the temporality of infection between the index case and those asymptomatically infected. It is also possible some asymptomatically infected individuals failed to recall and report mild symptoms.
We could not identify a source case in 22.0%–51.6% of infants. While this may be due to missing data and/or variability of residual immunity among contacts, it may also suggest that transmission from casual contact may be an important source of transmission and warrants further investigation. If casual contacts play an important role, it is likely that optimal prevention of pertussis among young infants will not be achieved until universal vaccination of adolescents and adults is highly successful.
In conclusion, this study provides robust evidence that household members are responsible for 73%–82% of B. pertussis transmission to infants for whom a source case could be identified, and for 39% of transmission to all infants with pertussis in this study. Parents accounted for half of all B. pertussis transmissions to infants. It is therefore likely that the implementation of the recent ACIP recommendation for adult and adolescent vaccination could substantially reduce the burden of infant pertussis, if high coverage rates among those in contact with young infants can be achieved.
The authors thank Benoit Soubeyrand and Luc Hessel for help in selection of sites in Europe; Marie Maitre, Martha Doemland, Vivian Jusot, and Zahia Latti for coordinating the shipment of specimens and case report forms, and facilitating communication between centers. The authors also appreciate the statistical help from Charles Poole and Michael Hudgens, UNC School of Public Health; and Philippe André, Pia MacDonald for editorial comments. The authors thank Marie-José Quentin-Millet and Sanofi Pasteur for the provision of purified pertussis toxin.
Funding sources: We gratefully acknowledge the Institut Pasteur Foundation, Sanofi Pasteur and Sanofi Pasteur-MSD for the unrestricted grant that funded this study. We acknowledge that Remy Teyssou and David Greenberg are employees of Sanofi Pasteur and contributed to the design of the study and review of the manuscript. Aaron Wendelboe and Annelies Van Rie had full access to all data, and take responsibility for the integrity of the data and the accuracy of the data analysis.
Infant Pertussis Study Group: Isabelle Poilane, MD (Hôpital Jean Verdier, Bondy, France), C. Berchiche, MD (Hôpital Edouard Herriot, Lyon, France), Albert Faye, MD, PhD, E. Bingen, PhD, and S. Bonacorsi, MD, PhD (Hôpital Robert Debré, Paris, France), Dominique Moissenet, MD (Hôpital Armand Trousseau, Paris, France), Bernard Belohradsky, MD and Michaela Hamm, MD, (von Haunersches Kinderspital, Ludwig-Maximilians-Universitaet, Munich, Germany), A. Hofmiller, MD (Kinderklinik der Technischen Universitaet, Munich, Germany), A. Koelzer (Kinderkrankenhaus 3. Orden, Munich, Germany), B. Kreuz, MD (Kinderklinik Harlaching, Munich, Germany), Klara Posfay-Barbe, MD and Marian G. Michaels, MD, MPH (Children's Hospital of Pittsburgh, Pittsburgh, PA).
1. Miller E, Gay NJ. Epidemiological determinants of pertussis. Dev Biol Stand
2. Tindberg Y, Blennow M, Granstrom M. A ten year follow-up after immunization with a two component acellular pertussis vaccine. Pediatr Infect Dis J
3. Wendelboe AM, Van Rie A, Salmaso S, Englund JA. Duration of immunity against pertussis after natural infection or vaccination. Pediatr Infect Dis J
. 2005;24(5 Suppl):S58–S61.
4. Wirsing von Konig CH, Halperin S, Riffelmann M, Guiso N. Pertussis of adults and infants. Lancet Infect Dis
5. Bisgard KM, Pascual FB, Ehresmann KR, et al. Infant pertussis: who was the source? Pediatr Infect Dis J
6. Greenberg DP. Pertussis in adolescents: increasing incidence brings attention to the need for booster immunization of adolescents. Pediatr Infect Dis J
7. Schellekens J, von Konig CH, Gardner P. Pertussis sources of infection and routes of transmission in the vaccination era. Pediatr Infect Dis J
. 2005;24(5 Suppl):S19–S24.
8. Pebody RG, Gay NJ, Giammanco A, et al. The seroepidemiology of Bordetella pertussis
infection in Western Europe. Epidemiol Infect
9. Centers for Disease Control and Prevention (CDC). National, state, and urban area vaccination levels among children aged 19–35 months—United States, 2002. MMWR Morb Mortal Wkly Rep.
10. Antona D, Bussiere E, Guignon N, Badeyan G, Levy-Bruhl D. Vaccine coverage of pre-school age children in France in 2000. Euro Surveill
11. Hopkins RS, Jajosky RA, Hall PA, et al. Summary of notifiable diseases—United States, 2003. MMWR Morb Mortal Wkly Rep
12. Tanaka M, Vitek CR, Pascual FB, et al. Trends in pertussis among infants in the United States, 1980–1999. JAMA
13. Floret D. [Pediatric deaths due to community-acquired bacterial infection. Survey of French pediatric intensive care units]. Arch Pediatr
. 2001;8(Suppl 4):705s–711s.
14. Bonmarin I, Laurent E, Guiso N, Njamkepo E. Renacoq: surveillance de la coqueluche à l'hôpital en 2002. BEH
]. Accessed March 18, 2006.
15. Forsyth K, Nagai M, Lepetic A, Trindade E. Pertussis immunization in the global pertussis initiative international region: recommended strategies and implementation considerations. Pediatr Infect Dis J
. 2005;24(5 Suppl):S93–S97.
16. Empfehlungen der Staendigen Impfkommision am Robert-Koch-Insitut. Epidemiol Bull.
17. Broder KR, Cortese MM, Iskander JK, et al. Preventing tetanus, diphtheria, and pertussis among adolescents: use of tetanus toxoid, reduced diphtheria toxoid and acellular pertussis vaccines recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep
18. Practices ACoI. ACIP Votes to Recommend Use of Combined Tetanus, Diphtheria and Pertussis (Tdap) Vaccine for Adults. Available at: http://www.cdc.gov/nip/vaccine/tdap/tdap_adult_recs.pdf
. Accessed April 5, 2006.
19. Van Rie A, Hethcote HW. Adolescent and adult pertussis vaccination: computer simulations of five new strategies. Vaccine
20. Izurieta HS, Kenyon TA, Strebel PM, et al. Risk factors for pertussis in young infants during an outbreak in Chicago in 1993. Clin Infect Dis
21. Biellik RJ, Patriarca PA, Mullen JR, et al. Risk factors for community- and household-acquired pertussis during a large-scale outbreak in central Wisconsin. J Infect Dis
22. Lambert HJ. Epidemiology of a small pertussis outbreak in Kent County, Michigan. Public Health Rep.
23. Halperin SA, Wang EE, Law B, et al. Epidemiological features of pertussis in hospitalized patients in Canada, 1991–1997: report of the Immunization Monitoring Program—Active (IMPACT). Clin Infect Dis
24. Principales caracteristiques des cas de coqueluche identifies par le reseau Renacoq, 1996–2004. http://www.invs.sante.fr/surveillance/coqueluche/donnees_96_04.pdf
. Accessed March 9, 2006.
25. Baron S, Njamkepo E, Grimprel E, et al. Epidemiology of pertussis in French hospitals in 1993 and 1994: thirty years after a routine use of vaccination. Pediatr Infect Dis J
26. Schmitt-Grohe S, Cherry JD, Heininger U, et al. Pertussis in German adults. Clin Infect Dis
27. Crowcroft NS, Booy R, Harrison T, et al. Severe and unrecognized: pertussis in UK infants. Arch Dis Child
28. Riffelmann M, Wirsing von Konig CH, Caro V, Guiso N. Nucleic acid amplification tests for diagnosis of Bordetella infections. J Clin Microbiol
29. Simondon F, Iteman I, Preziosi MP, Yam A, Guiso N. Evaluation of an immunoglobulin G enzyme-linked immunosorbent assay for pertussis toxin and filamentous hemagglutinin in diagnosis of pertussis in Senegal. Clin Diagn Lab Immunol
30. van der Zee A, Agterberg C, Peeters M, Mooi F, Schellekens J. A clinical validation of Bordetella pertussis
and Bordetella parapertussis
polymerase chain reaction: comparison with culture and serology using samples from patients with suspected whooping cough from a highly immunized population. J Infect Dis
31. Giammanco A, Chiarini A, Maple PA, et al. European sero-epidemiology network: standardization of the assay results for pertussis. Vaccine
32. Versteegh FG, Weverling GJ, Peeters MF, et al. Community-acquired pathogens associated with prolonged coughing in children: a prospective cohort study. Clin Microbiol Infect
33. Baughman AL, Bisgard KM, Edwards KM, et al. Establishment of diagnostic cutoff points for levels of serum antibodies to pertussis toxin, filamentous hemagglutinin, and fimbriae in adolescents and adults in the United States. Clin Diagn Lab Immunol
34. Centers for Disease Control and Prevention (CDC). Pertussis—United States, 1997–2000. MMWR Morb Mortal Wkly Rep.
35. Deen JL, Mink CA, Cherry JD, et al. Household contact study of Bordetella pertussis
infections. Clin Infect Dis
36. Wirsing von Konig CH, Postels-Multani S, Bock HL, Schmitt HJ. Pertussis in adults: frequency of transmission after household exposure. Lancet
37. Heininger U, Cherry JD, Stehr K, et al. Comparative efficacy of the Lederle/Takeda acellular pertussis component DTP (DTaP) vaccine and Lederle whole-cell component DTP vaccine in German children after household exposure. Pertussis Vaccine Study Group. Pediatrics
. 1998;102(3 Pt 1):546–553.
38. Sprauer MA, Cochi SL, Zell ER, et al. Prevention of secondary transmission of pertussis in households with early use of erythromycin. Am J Dis Child
39. Trollfors B, Taranger J, Lagergard T, et al. Efficacy of a monocomponent pertussis toxoid vaccine after household exposure to pertussis. J Pediatr
40. Stocks P. Some epidemiological features of whooping-cough: a statistical investigation. Lancet
41. Kendrick P, Elderling G. A study in active immunization against pertussis. Am J Hyg
42. Broome CV, Preblud SR, Bruner B, et al. Epidemiology of pertussis, Atlanta, 1977. J Pediatr
43. Hewlett E. Bordetella species. In: Mandell GL, Bennett JE, Dolin R, eds. Principles and Practice of Infectious Diseases.
5th ed. Philadelphia: Churchill Livingstone; 2000. No. 1.
44. Mertsola J, Ruuskanen O, Eerola E, Viljanen MK. Intrafamilial spread of pertussis. J Pediatr
45. Olin P, Gustafsson L, Barreto L, et al. Declining pertussis incidence in Sweden following the introduction of acellular pertussis vaccine. Vaccine