- Research article
- Open Access
- Open Peer Review
Pertussis resurgence in Toronto, Canada: a population-based study including test-incidence feedback modeling
© Fisman et al; licensee BioMed Central Ltd. 2011
- Received: 19 March 2011
- Accepted: 7 September 2011
- Published: 7 September 2011
Pertussis continues to challenge medical professionals; recently described increases in incidence may be due to age-cohort effects, vaccine effectiveness, or changes in testing patterns. Toronto, Canada has recently experienced increases in pertussis incidence, and provides an ideal jurisdiction for evaluating pertussis epidemiology due to centralized testing. We evaluated pertussis trends in Toronto using all available specimen data, which allowed us to control for changing testing patterns and practices.
Data included all pertussis culture and PCR test records for Greater Toronto from 1993 to 2007. We estimated incidence trends using Poisson regression models; complex relationships between disease incidence and test submission were explored with vector autoregressive models.
From 1993 to 2007, 26988 specimens were submitted for testing; 2545 (9.4%) were positive. Pertussis incidence was 2 per 100,000 from 1993 to 2004 and increased to 10 per 100,000 from 2005-2007, with a concomitant 6-fold surge in test specimen submissions after the introduction of a new, more sensitive PCR assay. The relative change in incidence was less marked after adjustment for testing volumes. Bidirectional feedbacks between test positivity and test submissions were identified.
Toronto's recent surge in pertussis reflects a true increase in local disease activity; the apparent size of the outbreak has likely been magnified by increasing use of pertussis testing by clinicians, and by improved test sensitivity since 2005. These findings may be applicable to changes in pertussis epidemiology that have been noted elsewhere in North America.
- Polymerase Chain Reaction Assay
- Test Volume
- Great Toronto Area
- Vector Autoregressive Model
Pertussis is a highly contagious respiratory tract infection caused by the gram negative bacterium Bordetella pertussis or less commonly by B. parapertussis . The disease is classically characterized by three stages : (i) a nonspecific catarrhal stage; (ii) a subsequent spasmodic stage involves the characteristic paroxysmal cough with inspiratory whoop, and lasting two to eight weeks; and (iii) a convalescent phase . The pathogenesis of the disease is not fully understood but involves both direct toxic effects of bacterial endo- and exotoxins, and also indirect effects of toxins on the host immune response, including a reduction in lymphocyte circulation [1, 4, 5]. While children and adults of any age may develop pertussis, severe sequelae (including encephalopathy and pneumonia) are most common in infants aged < 6 months [6–8]. The incidence of pertussis has decreased dramatically in wealthy countries since the implementation of widespread vaccination programs (e.g., from ~170/100 000 in the 1930s to fewer than 20/100 000 in the 1970s in Canada) pertussis is still an important source of mortality worldwide. The disease remains one of the leading causes of infant mortality,[6–8] and causes approximately 300 000 deaths in 50 million cases per year [7, 10].
In high income countries, elimination of pertussis has not occurred despite immunization, and apparent disease incidence has increased in recent decades [3, 6, 11]. Recent outbreaks, some with infant deaths, in Nottingham and Derby (England) , California , Ireland  and New South Wales, Australia  have been a source of considerable concern. It has been suggested, however, that increasingly sensitive PCR methods for the diagnosis of pertussis may have contributed to apparent, rather than real, increases in disease incidence [7, 9, 16–18]. Other factors invoked to explain pertussis increases in countries with high rates of immunization include genetic changes in the pathogen, [19, 20] waning immunity following both vaccination and infection,[1, 3, 7] reduced potency of vaccines following the switch to an acellular vaccine,[21, 22] more widespread laboratory testing due to increased awareness of the disease in adults, and lack of immunity or waning immunity in specific age cohorts . The role played by young adolescents and adults in disease spread, and the recognition of a gradual loss in immunity after natural infection or vaccination, have led several countries to advocate booster dosing of pertussis vaccine for young teens and adults [16, 25–27].
Surveillance of pertussis is complicated by differing manifestations of the disease in infants, adolescents and adults , with the disease presenting as apnea or poor feeding in infants [2, 12], whereas adults and adolescents often present only with prolonged cough , or minimal symptomatology . Nasopharyngeal swab culture is the historical gold-standard diagnostic modality for pertussis, but is sensitive only early in the infection, with a sensitivity < 50% after three weeks of illness or cough . Serology is even less sensitive but can be used to diagnose pertussis late in the course of infection . In recent years, the sensitivity and rapidity of polymerase chain reaction (PCR) have brought this testing modality into widespread use, even if testing is not well-standardized between laboratories [8, 16–18]. The World Health Organization's (WHO) most recent guidelines define a pertussis case as clinically confirmed (without laboratory confirmation) or laboratory confirmed (and meeting the clinical case definition). Prior to 2008, Canadian case definitions permitted the classification of a case of pertussis as "confirmed" based on a positive laboratory test (including PCR) alone [29, 30].
The Greater Toronto Area (GTA) is Canada's largest metropolitan area. All pertussis testing in the GTA is performed in only two laboratories (the Public Health Laboratory--Toronto (PHLT) and the laboratory of the Hospital for Sick Children (HSC), which share strong historical linkages, making it possible to evaluate not only case counts (according to pre-2008 Canadian case definitions) via laboratory data, but also to evaluate the impact that testing volume and testing practice may have had on measured disease epidemiology. The region experienced a documented outbreak of pertussis in 2005-2006 [7, 9, 16], and experienced a prolonged increase in pertussis incidence in association with the introduction of novel testing methodologies. Our objectives were to evaluate changes in pertussis epidemiology in this large urban setting and to define the relative contributions to disease incidence that may be attributable to laboratory testing submission volumes, changing testing technologies and underlying disease epidemiology.
Demographics and Laboratory Characteristics
The Greater Toronto Area (GTA) is the most populous metropolitan area in Ontario, with a population of 5,113,149 in 2006 over 7,125 km2. The GTA is located in the southern part of Ontario and is comprised of five administrative regions (Toronto, York, Halton, Peel and Durham), all of which have separate public health units charged with disease prevention and control activities. Pertussis is a notifiable infectious disease in Ontario and must be reported to regional public health agencies if compatible with the national case definition: during the period reported here, a confirmed case was one with either isolation of B. pertussis or a positive PCR assay for B. pertussis, regardless of symptomatology . Classical symptoms of pertussis (paroxysmal cough, post-tussive vomiting or apnea, and inspiratory whoop) are considered confirmatory in the absence of laboratory testing of individuals with known epidemiological links to another confirmed case . Typical case management strategies employed by public health authorities include evaluation of vaccine status, with provision of catch-up or booster vaccination as appropriate, and provision of post-exposure antimicrobial therapy to close contacts of cases .
GTA pertussis testing is conducted by two laboratories: the Public Health Laboratory--Toronto (PHLT) and the Hospital for Sick Children (HSC). The PHLT provides primary diagnostic services for hospitals and healthcare practices in the GTA, as well as confirmatory testing and reference laboratory services for the entire province. HSC's laboratory performs testing principally on children from the hospital's inpatient and outpatient areas, and the hospital's emergency room. Data included here cover all pertussis culture and PCR tests conducted at these two sites from 1993 to 2007. Serological data are not available. Prior to 1999, only culture isolation of B. pertussis was available.
Evolution of Testing Procedures
Data on respiratory specimen testing at the PHLT were available from January 1993 to January 2007. Since January 1999, the PHLT has performed parallel culture and PCR on all respiratory specimens submitted for pertussis testing, with culture and identification of B. pertussis performed as described elsewhere . The PCR assay introduced at the PHLT in 1999 was a qualitative assay for the presence of the IS481 gene, an insertion sequence that appears in multiple copies in the pertussis genome, and which is not present in several other Bordetella species of public health importance (e.g., B. parapertussis, B. bronchoseptica). The introduction of PCR in January 1999 was associated with a change in specimen collection procedures for pertussis testing, with the collection media changed from nutrient-rich, antibiotic impregnated Regan-Lowe medium, to phosphate-buffered-saline (PBS) medium. In May 2005, the PHLT introduced a real-time, quantitative PCR assay incorporating a fluorescent probe for enhanced assay sensitivity. Data on respiratory specimens submitted to the HSC laboratory were available from October 1999 to January 2007; a qualitative IS481-directed PCR was used on all specimens submitted at HSC during this time period, and culture was not performed.
Age- and sex-specific population estimates for the Toronto Census Metropolitan Area were obtained from the 1996, 2001, and 2006 Canadian census ; data for inter-census years were estimated through linear interpolation and extrapolation. We estimated crude and adjusted incidence of pertussis testing and disease occurrence in the Greater Toronto Area using specimens linked by postal code forward sortation area (FSA, the first three alphanumeric characters of the six character Canadian postal code) to the GTA, or which were identified as having been obtained in one of the five jurisdictions comprising the GTA (when FSA was absent). A small number of test specimens processed by the HSC laboratory were obtained from individuals with home FSA outside the GTA. HSC is a major pediatric tertiary/quaternary care center in Ontario, and consequently provides care to many patients who reside outside the GTA. When non-GTA specimens were obtained from individuals identified as hospital inpatients, these test specimens were included in counts for the Toronto health region. When non-GTA specimens processed by HSC were identified as being from outpatients, and when specimens were identified as having been provided by outside laboratories for reference testing, specimens were excluded from further analysis.
Risk of pertussis by age, gender, season, and time period were evaluated using census-derived person-time denominators, with differences in risk evaluated through calculation of risk differences and 95% confidence limits. We compared age distributions for cases by time period (prior to 1999, from 1999 to May 2005, and after May 2005) using the Wilcoxon rank-sum test. Data were missing for subject gender (5.9%) and age (3.0%). In our primary analysis, individuals missing values for age were replaced using random age values drawn from a log-normal distribution with a mean and standard deviation identical to that of known ages. Missing gender was replaced randomly in proportion to known gender distributions. Sensitivity analyses were also performed, with records with missing data excluded from analyses.
Specimen factors associated with test positivity, and with culture or PCR positivity, using categorical methods, as well as univariable and multivariable logistic regression models that included age and gender of the specimen source person, laboratory performing the tests, and year and season of submission (modeled using "fast Fourier transforms" (FFT)  that provided an underlying sinusoidal seasonal baseline).
Temporal trends in pertussis occurrence for the population as a whole, and for individual demographic groups were explored using multivariable Poisson regression models that included both seasonal smoothers (FFT as described above) and also oscillatory FFT with 6-year periodicity, to account for the previously described multi-year periodicity of pertussis . The assumption of Poisson-distribution of errors was evaluated with the deviance statistic  and found to be appropriate (P = 1.00). The impact of changing testing procedures on test positivity was evaluated by introducing step functions at time points where HSC data were added to the time series, and where test procedures were modified (i.e., with the introduction of PCR by the PHLT and HSC laboratories, and with the modification of PCR testing procedures by the PHLT). Typical time series models of public health surveillance data are not able to account for the impact of the volume of test submission on apparent disease incidence (i.e., lack so-called "testing denominators"); test submission volumes in multivariable models were included, to evaluate the significance of their contribution to model fit using the log-likelihood test . We performed exploratory analyses in which we evaluated the differential effects of test volume on apparent pertussis incidence both by incorporating multiplicative interaction terms into models, and by performing restriction analyses in which testing effects were evaluated in models restricted to a single testing period (i.e., prior to 1999, from January 1999 to April 2005, and from May 2005). We evaluated differences in testing-adjusted and unadjusted covariates using the meta-analytic Q-statistic , and evaluated heterogeneity in testing effects by period using the Wald chi-squared test . Model choice was guided by balancing model fit against parsimony through minimization of Akaike's information criterion .
We hypothesized that the relationship between test submission and test positivity would be complex and bidirectional, and assessed the relationship between submission and positivity through construction of vector autoregressive models, with evidence of causality sought through the application of Granger's test . Briefly, this statistical test (widely applied in econometrics) seeks to identify cause and effect relationships by comparing model prediction when dependent variables and independent variables (at variable lags) are reversed . Evidence of stationarity of time series using the Dickey-Fuller unit root test  was assessed.
All data analysis was performed with Stata version 11 (StataCorp, College Station, TX). The study was approved by the Research Ethics Board of the University of Toronto.
Epidemiological Profile of Pertussis in the Greater Toronto Area
Crude and Stratum-Specific Rates of Laboratory-Confirmed Pertussis in Greater Toronto Area, 1993-2007
Rate per 100,000 Person-Years (95% Confidence Interval)
Prior to Introduction of PCR (January 1993-December 1998)
Initial Introduction of PCR Testing (January 1999-April 2005)
Introduction of Highly Sensitive PCR (May 2005-)
Poisson Regression on Demographic Risks, Temporal Trends, and Testing Effects Associated with Reported Pertussis Risk in the Greater Toronto Area, Canada
Exclude Testing Volume
Include Testing Volume
IRR (95% C.I.)
IRR (95% C.I.)
Oscillation, 6-year Period
1.15 (1.06 to 1.24)
1.14 (1.05 to 1.23)
0 to 4 years
57.76 (50.91 to 65.53)
38.59 (33.84 to 44.00)
5 to 9 years
17.16 (14.78 to 19.93)
17.34 (14.94 to 20.14)
10 to 14 years
13.14 (11.24 to 15.36)
14.81 (12.66 to 17.32)
15 to 19 years
3.57 (2.82 to 4.52)
4.30 (3.40 to 5.44)
20 and over
Introduction of Novel, Highly Sensitive PCR (May 2005)
3.75 (3.44 to 4.09)
2.31 (2.09 to 2.55)
Test Submission Volume (per 10 additional test specimens)
1.57 (1.53 to 1.61)
Akaike's Information Criterion
The impact of test submission volume on apparent pertussis incidence was diminished after May 2005 as opposed to before May 2005 (IRR per 10 specimens submitted 1.30, 95% CI 1.26 to 1.35 vs. IRR 7.93, 95% CI 6.80 to 9.26, P for heterogeneity < 0.001). When this time period was excluded there was no heterogeneity in the effect of test submissions before as opposed to after the introduction of PCR in January 1999 (IRR per 10 specimens submitted 9.30, 95% CI 7.74 to 11.18 vs. IRR 4.59, 95% CI 3.35 to 6.27, P = 0.14).
Patterns in Laboratory Testing in the Greater Toronto Area
Seasonal oscillation in pertussis incidence using fast Fourier transforms were incorporated into Poisson models for the time period prior to May 2005. Pertussis incidence, incidence of culture-confirmed pertussis, incidence of pertussis without culture confirmation, and pertussis test submissions all displayed statistically significant annual seasonality (P for oscillation < 0.001 for all evaluations). However, while test positivity for both culture- and PCR-positive pertussis displayed autumn seasonality (peak occurrence weeks 36 to 38), pertussis testing peaked in winter (week 49) (Figure 1 ). Heterogeneity in phase terms was assessed using the meta-analytic Q-statistic and found significant heterogeneity in waves (Q-statistic 778.2 on 3 d.f., P < 0.001); however, the wave forms for seasonality of test-positivity were homogeneous (Q-statistic 2.4 on 3 d.f., P < 0.31).
Clinical, Temporal and Testing Characteristics Associated with PCR-Positive, Culture-Negative Specimens (N = 1858), 1999-2005
Age > 20 (referent)
Introduction of Novel PCR (May 2005)
Vector Autoregressive Model Evaluating Prediction of Pertussis Test Positivity by Lagged Pertussis Positivity and Test Submissions
Independent Variable and Lag
Entire Study Period
Prior to Novel PCR
(1993 to April 2005)
(May 2005 to December 2007)
Coefficient (95% CI)
Coefficient (95% CI)
Coefficient (95% CI)
1 week lag
0.49 (0.41 to 0.58)
0.20 (0.12 to 0.28)
0.56 (0.34 to 0.79)
2 week lag
0.42 (0.33 to 0.51)
0.22 (0.14 to 0.30)
0.55 (0.31 to 0.79)
3 week lag
0.16 (0.07 to 0.25)
0.18 (0.10 to 0.27)
0.10 (-0.14 to 0.34)
4 week lag
-0.22 (-0.31 to -0.14)
0.07 (-0.01 to 0.16)
-0.42 (-0.64 to -0.20)
1 week lag
0.02 (0 to 0.04)
0.01 (-0.01 TO 0.02)
0.02 (-0.03 to 0.07)
2 week lag
-0.06 (-0.08 to -0.04)
-0.02 (-0.04 TO 0.01)
-0.09 (-0.15 to -0.03)
3 week lag
0.03 (0.01 to 0.06)
0.01 (-0.01 TO 0.03)
0.07 (0.01 to 0.13)
4 week lag
0.01 (0 to 0.03)
0.00 (-0.01 TO 0.02)
0.02 (-0.02 to 0.07)
1 week lag
1.41 (1.03 to 1.79)
0.44 (0.08 TO 0.80)
1.81 (0.78 to 2.83)
2 week lag
0.61 (0.21 to 1.01)
0.57 (0.21 TO 0.93)
0.94 (-0.15 to 2.02)
3 week lag
-0.05 _0.45 to 0.36)
0.17 (-0.20 TO 0.53)
-0.14 (-1.24 to 0.96)
4 week lag
-1.04 (-1.42 to -0.67)
-0.20 (-0.56 to 0.16)
-1.30 (-2.28 to -0.32)
1 week lag
0.66 (0.58 to 0.74)
0.49 (0.41 to 0.57)
0.67 (0.45 to 0.90)
2 week lag
0.01 (-0.08 to 0.11)
0.22 (0.13 to 0.31)
-0.15 (-0.41 to 0.12)
3 week lag
0.25 (0.15 to 0.35)
0.12 (0.04 to 0.21)
0.39 (0.12 to 0.66)
4 week lag
-0.08 (-0.16 to 0)
0.05 (-0.03 to 0.12)
-0.11 (-0.30 to 0.09)
Despite a greater than 10-fold reduction in the burden of pertussis-related mortality and morbidity in high- and middle-income countries with the introduction of immunization, this disease continues to challenge clinicians and public health policymakers [11, 42]. These challenges are rooted in fundamental questions regarding the epidemiology and natural history of pertussis, and the difficulty encountered by mathematical modelers in building well calibrated simulations for this disease imply that substantial gaps in knowledge still exist . We evaluated the epidemiology of pertussis over a 14-year period in a large urban-suburban region of Ontario, Canada. In the Greater Toronto Area, pertussis incidence was approximately stable in the early 1990s, increased somewhat with the introduction of polymerase chain reaction (PCR) in 1999, and then increased markedly with the introduction of a new, highly sensitive PCR assay. Because this was a laboratory-derived, population-based analysis (due to the fact that all pertussis testing in the region is performed by the two laboratories participating in this study), we had the opportunity to evaluate both changes in disease epidemiology and changes in testing volumes and test technology, and to examine the interplay between these two distinct processes.
We found that the surge in pertussis incidence in the GTA from 2005 to 2007 likely reflected a combination of factors, with a true, underlying increase in disease risk coinciding with the introduction of a highly sensitive assay that probably identified cases that would have been missed with earlier testing modalities. Surges in apparent disease incidence with the introduction of highly sensitive assays for pertussis have been described in other settings , but the process we identify here is not as simple as increasing case counts as a result of increased test sensitivity. Instead, through the use of time series analysis, we were able to identify positive feedback loops, whereby increasing test positivity (due to either increasing incidence or increased test sensitivity) led to increased test submissions by clinicians over subsequent weeks. Such positive feedbacks were previously noted by us in the context of legionellosis  but to our knowledge this is the first time such feedbacks have been described for pertussis.
The ability to control for both test volumes and testing technology provides a more nuanced view of the state of pertussis epidemiology in this jurisdiction than would be possible using usual surveillance data sources, which focus on "positive" cases alone, and do not consider frequency of testing. In this instance, adjustment for test frequency reduced the relative risk of pertussis in the youngest children, and also diminished the apparent impact of the introduction of a highly sensitive PCR assay on pertussis incidence. Unfortunately, few jurisdictions have the ability to adjust for testing practices and volumes when analyzing disease trends, but our findings show that these factors may be important, particularly when real or artifactual surges in disease create public concern, leading to further increases in testing. This is a concept which has application beyond pertussis; for example, Vickers and Osgood  evaluated an increase in Chlamydia incidence in the Canadian province of Saskatchewan which has centralized testing such that test volumes are available; in a mathematical model, they found that changes in test volume alone, rather than changes in testing characteristics or true underlying Chlamydia risk, were likely responsible for the surge in observed Chlamydia incidence in that province. These authors noted "diminishing returns" from increasing test submissions, as the marginal identification of cases declines with expansion of testing. Statistical evidence for such a phenomenon is seen in our analysis too, in the diminished (though still statistically significant) association between test submission volumes and estimated relative risk from May 2005 onwards.
We identified other important changes in the apparent epidemiology that appear to be driven by testing practices rather than disease activity per se. For example, using spectral decomposition, we found that whereas pertussis in the GTA displays (expected and typical) autumn seasonality, but pertussis testing displays wintertime seasonality (perhaps due to the use of pertussis testing in individuals with undifferentiated cough illness due to wintertime respiratory viruses). This might be expected to influence observed seasonal patterns of disease occurrence, and again confirms the degree to which a nuanced understanding of disease epidemiology requires some knowledge of how information on disease trends is obtained at the level of the microbiology laboratory.
Like any epidemiological analysis, our analysis is subject to limitations. Key among these relates to measurement. Although an attempt was made to describe trends in pertussis in a population with centralized testing, it is unlikely that all individuals with pertussis underwent testing, and inasmuch as both likelihood of symptoms and likelihood of testing depend on age, some of the effects reported here may relate to differential testing of different groups. This is a limitation of any population-based study of pertussis epidemiology. In addition, we had no access to information on individuals with clinical pertussis without laboratory testing, though given universal healthcare coverage in this region and the ready availability of laboratory tests, it is unlikely to have had a major influence on key findings.
In summary, the epidemiology of pertussis in a large urban-suburban region of Ontario, Canada with centralized pertussis testing over a 15-year period identified surges in pertussis incidence in this jurisdiction from 2005 onwards. The ability to control for laboratory test technology and test volume allowed identification of numerous components to this rise in incidence, including improved test sensitivity and a previously undescribed (for pertussis) "positive feedback loop" whereby increasing reported disease activity appeared to cause subsequent surges in laboratory test submissions by clinicians. We were also able to detect distinctive seasonality of pertussis identified by PCR testing without culture confirmation. In conclusion, the microbiology laboratory is a key component of smart, integrated disease surveillance infrastructure; where possible, trend estimates for pertussis and other communicable diseases should incorporate data on laboratory test volumes and technologies, in order to inform smarter and more efficient public health policy.
This work was supported by funding from GlaxoSmithKline (Dr. Fisman), and by a University of Toronto Faculty of Medicine Clinical Research Experience--Medical Student Award (to Dr. Hauck). Funders played no role in acquisition of data, design of the study, analysis of data, or drafting the manuscript but were given an opportunity to comment on an advanced draft of the manuscript prior to journal submission.
- Halperin SA: Pertussis and Other Bordetella Infections. Harrison's Principles of Internal Medicine. Edited by: al Fe. 2008, New York: McGraw Hill Medical, 933-937. 17Google Scholar
- Galiza EP, Heath PT: Pertussis. Medicine. 2009, 37 (12): 635-637. 10.1016/j.mpmed.2009.09.007.View ArticleGoogle Scholar
- Raguckas SE, VandenBussche HL, Jacobs C, Klepser ME: Pertussis resurgence: Diagnosis, treatment, prevention, and beyond. Pharmacotherapy. 2007, 27 (1): 41-52. 10.1592/phco.27.1.41.View ArticlePubMedGoogle Scholar
- Carbonetti NH: Immunomodulation in the pathogenesis of Bordetella pertussis infection and disease. Current Opinion in Pharmacology. 2007, 7 (3): 272-278. 10.1016/j.coph.2006.12.004.View ArticlePubMedGoogle Scholar
- Smith AM, Guzman CA, Walker MJ: The virulence factors of Bordetella pertussis: a matter of control. FEMS Microbiol Rev. 2001, 25 (3): 309-333. 10.1111/j.1574-6976.2001.tb00580.x.View ArticlePubMedGoogle Scholar
- Van Boven M, Ferguson NM, Van Rie A: Unveiling the burden of pertussis. Trends in Microbiology. 2004, 12 (3): 116-119. 10.1016/j.tim.2004.01.002.View ArticlePubMedGoogle Scholar
- Wood N, McIntyre P: Pertussis: review of epidemiology, diagnosis, management and prevention. Paediatric Respiratory Reviews. 2008, 9 (3): 201-212. 10.1016/j.prrv.2008.05.010.View ArticlePubMedGoogle Scholar
- Crowcroft NS, Pebody RG: Recent developments in pertussis. Lancet. 2006, 367 (9526): 1926-1936. 10.1016/S0140-6736(06)68848-X.View ArticlePubMedGoogle Scholar
- Greenberg DP, Doemland M, Bettinger JA, Scheifele DW, Halperin SA, Waters V, Kandola K: Epidemiology of pertussis and haemophilus influenzae type b disease in Canada with exclusive use of a diphtheria-tetanus-acellular pertussis- inactivated poliovirus-haemophilus influenzae type b pediatric combination vaccine and an adolescent-adult tetanus-diphtheria-acellular pertussis vaccine: Implications for disease prevention in the United States. Pediatric Infectious Disease Journal. 2009, 28 (6): 521-528. 10.1097/INF.0b013e318199d2fc.View ArticlePubMedGoogle Scholar
- Yeh SH: Pertussis: persistent pathogen, imperfect vaccines. Expert Rev Vaccines. 2003, 2 (1): 113-127. 10.1586/14760522.214.171.124.View ArticlePubMedGoogle Scholar
- Bamberger ES, Srugo I: What is new in pertussis?. European Journal of Pediatrics. 2008, 167 (2): 133-139. 10.1007/s00431-007-0548-2.View ArticlePubMedGoogle Scholar
- Smith C, Vyas H: Early infantile pertussis; increasingly prevalent and potentially fatal. European Journal of Pediatrics. 2000, 159 (12): 898-900. 10.1007/PL00008365.View ArticlePubMedGoogle Scholar
- Winter K, Harriman K, Schechter R, Yamada E, Talarico J, Chavez G: Pertussis -- California, January-June 2010. Morbidity and Mortality Weekly Report. 2010, 59 (26): 817-818.Google Scholar
- Barret AS, Ryan A, Breslin A, Cullen L, Murray A, Grogan J, Bourke S, Cotter S: Pertussis outbreak in northwest Ireland, January - June 2010. Euro Surveill. 2010, 15 (35):Google Scholar
- Communicable Diseases Report, NSW, November and December 2008. N S W Public Health Bull. 2009, 20 (1-2): 31-35.Google Scholar
- Waters V, Jamieson F, Richardson SE, Finkelstein M, Wormsbecker A, Halperin SA: Outbreak of atypical pertussis detected by polymerase chain reaction in immunized preschool-aged children. Pediatr Infect Dis J. 2009, 28 (7): 582-587. 10.1097/INF.0b013e318197fac1.View ArticlePubMedGoogle Scholar
- Taranger J, Trollfors B, Lind L, Zackrisson G, Beling-Holmquist K: Environmental contamination leading to false-positive polymerase chain reaction for pertussis . Pediatric Infectious Disease Journal. 1994, 13 (10): 936-937.View ArticlePubMedGoogle Scholar
- Lievano FA, Reynolds MA, Waring AL, Ackelsberg J, Bisgard KM, Sanden GN, Guris D, Golaz A, Bopp DJ, Limberger RJ, Smiths PF: Issues associated with and recommendations for using PCR to detect outbreaks of pertussis. Journal of Clinical Microbiology. 2002, 40 (8): 2801-2805. 10.1128/JCM.40.8.2801-2805.2002.View ArticlePubMedPubMed CentralGoogle Scholar
- Mooi FR: Bordetella pertussis and vaccination: The persistence of a genetically monomorphic pathogen. Infection, Genetics and Evolution. 2010, 10 (1): 36-49. 10.1016/j.meegid.2009.10.007.View ArticlePubMedGoogle Scholar
- Brinig MM, Cummings CA, Sanden GN, Stefanelli P, Lawrence A, Relman DA: Significant gene order and expression differences in Bordetella pertussis despite limited gene content variation. Journal of Bacteriology. 2006, 188 (7): 2375-2382. 10.1128/JB.188.7.2375-2382.2006.View ArticlePubMedPubMed CentralGoogle Scholar
- Weir E: Resurgence of Bordetella pertussis infection. Canadian Medical Association Journal. 2002, 167 (10): 1146-PubMedPubMed CentralGoogle Scholar
- Cherry JD: The science and fiction of the "resurgence" of pertussis. Pediatrics. 2003, 112 (2): 405-406. 10.1542/peds.112.2.405.View ArticlePubMedGoogle Scholar
- Ntezayabo B, De Serres G, Duval B: Pertussis resurgence in Canada largely caused by a cohort effect. Pediatric Infectious Disease Journal. 2003, 22 (1): 22-27. 10.1097/00006454-200301000-00009.View ArticlePubMedGoogle Scholar
- Wendelboe AM, Njamkepo E, Bourillon A, Floret DD, Gaudelus J, Gerber M, Grimprel E, Greenberg D, Halperin S, Liese J, MuÃ ± oz-Rivas F, Teyssou R, Guiso N, Van Rie A: Transmission of Bordetella pertussis to young infants. Pediatric Infectious Disease Journal. 2007, 26 (4): 293-299. 10.1097/01.inf.0000258699.64164.6d.View ArticlePubMedGoogle Scholar
- Wendelboe AM, Njamkepo E, Bourillon A, Floret DD, Gaudelus J, Gerber M, Grimprel E, Greenberg D, Halperin S, Liese J, Munoz-Rivas F, Teyssou R, Guiso N, Van Rie A: Transmission of Bordetella pertussis to young infants. Pediatr Infect Dis J. 2007, 26 (4): 293-299. 10.1097/01.inf.0000258699.64164.6d.View ArticlePubMedGoogle Scholar
- Broder KR, Cortese MM, Iskander JK, Kretsinger K, Slade BA, Brown KH, Mijalski CM, Tiwari T, Weston EJ, Cohn AC, Srivastava PU, Moran JS, Schwartz B, Murphy TV: 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. 2006, 55 (RR-3): 1-34.Google Scholar
- De Serres G: An Advisory Committee Statement (ASC). National Advisory Committee on Immunization (NACI). Statement on adult/adolescent formulation of combined acellular pertussis, tetanus, and diphtheria vaccine. Can Commun Dis Rep. 2000, 26: 1-8.PubMedGoogle Scholar
- Ward JI, Cherry JD, Chang SJ, Partridge S, Lee H, Treanor J, Greenberg DP, Keitel W, Barenkamp S, Bernstein DI, Edelman R, Edwards K: Efficacy of an acellular pertussis vaccine among adolescents and adults. N Engl J Med. 2005, 353 (15): 1555-1563. 10.1056/NEJMoa050824.View ArticlePubMedGoogle Scholar
- Case definitions for diseases under national surveillance. Canada communicable disease report. 2000, 26s3: i-iv. 1-122Google Scholar
- Moghadas SM, Bowman CS, Rost G, Fisman DN, Wu J: Post-exposure prophylaxis during pandemic outbreaks. BMC Med. 2009, 7: 73-10.1186/1741-7015-7-73.View ArticlePubMedPubMed CentralGoogle Scholar
- 2006 Community Profiles. Last accessed September 27, 2010, [http://www12.statcan.ca/census-recensement/2006/dp-pd/prof/92-591/index.cfm?Lang=E]
- Committee on Infectious Diseases American Academy of Pediatrics: Pertussis. 1997 Red Book. Edited by: Peter GB, Hall CB, Halsey NA, Marcy SM, PIckering LK. 1997, Elk Grove, IL: American Academy of PediatricsGoogle Scholar
- Loeffelholz MJ, Sanden GN: Bordetella. Manual of Clinical Microbiology. Edited by: Murray PR, Baron EJ, Jorgenson JH, Landry ML, Pfaller MA. 2007, Washington, DC: ASM Press, 803-814. 9Google Scholar
- Scharlemann JP, Benz D, Hay SI, Purse BV, Tatem AJ, Wint GR, Rogers DJ: Global data for ecology and epidemiology: a novel algorithm for temporal Fourier processing MODIS data. PLoS One. 2008, 3 (1): e1408-10.1371/journal.pone.0001408.View ArticlePubMedPubMed CentralGoogle Scholar
- Gordon JE, Hood RI: Whooping cough and its epidemiological anomalies. Am J Med Sci. 1951, 222 (3): 333-361. 10.1097/00000441-195109000-00011.View ArticlePubMedGoogle Scholar
- Bithell JF: Geographical epidemiology. Handbook of Epidemiology. Edited by: Ahrens W, Pigeot I. 2007, Berlin: Springer, 861-888.Google Scholar
- Woolf B: The log likelihood ratio test (the G-test); methods and tables for tests of heterogeneity in contingency tables. Ann Hum Genet. 1957, 21 (4): 397-409. 10.1111/j.1469-1809.1972.tb00293.x.View ArticlePubMedGoogle Scholar
- Rothman K, Greenland S: Introduction to stratified analysis. Modern Epidemiology. Edited by: Rothman K, Greenland S. 1998, Philadelphia: Lippincott-Raven, 253-277. 2Google Scholar
- Harrell FE: Overview of maximum likelihood estimation. Regression Modeling Strategies. 2001, New York: Springer, 179-213.View ArticleGoogle Scholar
- Moineddin R, Nie JX, Domb G, Leong AM, Upshur RE: Seasonality of primary care utilization for respiratory diseases in Ontario: a time-series analysis. BMC Health Serv Res. 2008, 8: 160-10.1186/1472-6963-8-160.View ArticlePubMedPubMed CentralGoogle Scholar
- Bishai DM: Infant mortality time series are random walks with drift: are they cointegrated with socioeconomic variables?. Health Econ. 1995, 4 (3): 157-167. 10.1002/hec.4730040302.View ArticlePubMedGoogle Scholar
- Halperin SA: The control of pertussis--2007 and beyond. N Engl J Med. 2007, 356 (2): 110-113. 10.1056/NEJMp068288.View ArticlePubMedGoogle Scholar
- Nguyen HT, Rohani P: Noise, nonlinearity and seasonality: the epidemics of whooping cough revisited. J R Soc Interface. 2008, 5 (21): 403-413. 10.1098/rsif.2007.1168.View ArticlePubMedGoogle Scholar
- Outbreaks of respiratory illness mistakenly attributed to pertussis--New Hampshire, Massachusetts, and Tennessee, 2004-2006. MMWR Morb Mortal Wkly Rep. 2007, 56 (33): 837-842.Google Scholar
- Ng V, Tang P, Fisman DN: Our evolving understanding of legionellosis epidemiology: learning to count. Clin Infect Dis. 2008, 47 (5): 600-602. 10.1086/590558.View ArticlePubMedGoogle Scholar
- Vickers DM, Osgood ND: Current crisis or artifact of surveillance: insights into rebound chlamydia rates from dynamic modelling. BMC Infect Dis. 2010, 10: 70-10.1186/1471-2334-10-70.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://0-www.biomedcentral.com.brum.beds.ac.uk/1471-2458/11/694/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.