OBJECTIVE. We sought to describe the rates of pertussis hospitalization among infants by using databases that do not rely on passive reporting and compare with results obtained from the passive national surveillance system.
METHODS. The incidence of infant pertussis hospitalization in 1993 to 2004 was determined by using 2 national hospitalization discharge databases (Nationwide Inpatient Sample and Kids’ Inpatient Database) and the National Notifiable Disease Surveillance System/Supplemental Pertussis Surveillance System. Rates were determined for separate age groups among infants <1 year of age. Pertussis complications and procedures were examined by using the Kids’ Inpatient Database.
RESULTS. In 1993 to 2004, the pertussis hospitalization rates for infants ≤2 months of age were generally stable, by the discharge databases. The incidence of infant pertussis hospitalization obtained from the Nationwide Inpatient Sample and Kids’ Inpatient Database was ∼2 times greater than that obtained from the passive reporting system. Infants 1 to 2 months of age had the highest incidence (239 hospitalizations per 100000 live births in the 2003 Kids’ Inpatient Database). An annual average of 2678 hospitalizations occurred in 2000 and 2003; 86% occurred in infants ≤3 months of age. Among those with ages provided, 95% of infants who required mechanical ventilation and all of those who died were ≤3 months of age.
CONCLUSIONS. Pertussis hospitalization incidence rates among the youngest infants were generally stable in 1993 to 2004 and were highest for infants 1 to 2 months of age. The impact of the new adolescent and adult tetanus-diphtheria-acellular pertussis vaccines on infant pertussis should be monitored through such discharge databases. Additional vaccination strategies should be evaluated to protect infants as early in life as possible.
In the prevaccine era, pertussis was one of the major childhood illnesses, from which almost no child escaped, and was a leading cause of child death. With the institution of pertussis vaccination in the United States in the 1940s, pertussis morbidity and mortality rates decreased dramatically.1,2 Over time, the childhood vaccination program reached increasingly more children; from 1994 onward, ≥90% coverage for ≥3 doses of pertussis-containing vaccine was achieved among children 19 to 35 months of age.3 Despite full implementation of the vaccination program, however, an increase in pertussis cases among infants ≤4 months of age reported to the national passive surveillance system was observed between the 1980s and the 1990s.2 This increase in reported cases was accompanied by an increase in reported infant pertussis deaths, which suggested that the overall burden of pertussis among young infants had increased.4
These patterns of reported pertussis were based on the national passive surveillance system, which consists of the National Notifiable Disease Surveillance System and the Supplemental Pertussis Surveillance System (SPSS).2 State health departments voluntarily report pertussis cases to the national system. Local health departments, laboratories, clinicians, and other health care personnel report cases to the state health departments, following state laws and regulations.5 Because such passive systems rely on reporting and reporting practices can change over time, trends observed with passive surveillance data are difficult to interpret. Therefore, we used nationally representative hospitalization discharge databases to describe the incidence of pertussis hospitalization among specific age groups of infants 0 to 11 months of age in 1993 to 2004, and we compared the results with the passive data. We also used the discharge databases to examine complications and procedures for infants hospitalized with pertussis.
Data on live births were obtained from the National Center for Health Statistics.6,7 To estimate the number of live births in each 1-month age cohort, the number of live births reported for each calendar year was divided by 12. Rates of pertussis hospitalizations per 100000 live births were estimated for infants <1 year of age and for various age groups. Single-month ages were combined on the basis of similarity of rates and the US schedule of infant pertussis immunization (doses at 2, 4, and 6 months of age).8 Gender-specific live births6,9 were used as denominators for gender-specific pertussis rates. Annual incidence rates of pertussis hospitalizations among infants were derived from 3 databases, that is, the national SPSS and 2 nationally representative hospital discharge databases, the Nationwide Inpatient Sample (NIS) and the Kids’ Inpatient Database (KID).
The SPSS is a supplement to the National Notifiable Disease Surveillance System, through which cases of notifiable diseases are reported to the Centers for Disease Control and Prevention. Additional detailed information on reported pertussis cases (eg, age at cough onset [in months] and hospitalization status) is also collected. Before 1996, SPSS data were submitted on surveillance worksheets from state health departments to the Centers for Disease Control and Prevention; beginning in 1996, SPSS information was able to be transmitted via the National Electronic Telecommunication Surveillance System.2 Pertussis cases reported by the 50 states and the District of Columbia as confirmed or probable10 were used in this analysis.
In the SPSS, age is defined as the age at cough onset and no information on age at hospital admission is available. Hospitalization status was missing/unknown for some infants in the SPSS (Table 1). The proportion of infant cases with missing hospitalization status increased from 2% in 1993 to 27% in 2004. For 2002 to 2004, hospitalization status was missing for all cases in 5 to 7 states, and those states contributed 27% to 58% of all cases with missing status. A proportion of all cases with missing status were assumed to be hospitalized, and this proportion was determined through an imputation procedure. Hospitalization status (yes, no, or unknown) was stratified according to year and age group. Within each stratum, the proportion of infants who were hospitalized among infants with known status was then applied to infants with missing status, to estimate the number who were hospitalized but not reported as hospitalized.
The NIS and the KID are both part of the Healthcare Cost and Utilization Project, established by the Agency for Healthcare Research and Quality.11 National estimates derived from these sources were weighted to represent the 50 states, the District of Columbia, Puerto Rico, and the US territories combined and were calculated by following Healthcare Cost and Utilization Project recommended weighting procedures, using the SAS Surveymeans procedure.12
The NIS is a nationally representative database of hospital inpatient stays that is published annually and contains information on ∼5 million to ∼8 million inpatient stays (discharges) for patients of all ages.13,14 The NIS is a stratified probability sample of US hospitals. The number of states contributing to the NIS has increased over time (eg, 8–11 states before 1993, 17 states in 1993, and 37 states in 2004); because of concerns about coverage, the Agency for Healthcare Research and Quality recommends that trend analyses begin with the year 1993.15 Therefore, we began our study period with 1993. Up to 15 diagnoses and 15 procedures can be included per hospitalization. For the primary analysis, infant pertussis hospitalizations were selected by searching all International Classification of Diseases, Ninth Revision, Clinical Modification discharge diagnosis codes among infants <1 year of age for 2 codes in category 033 (whooping cough [includes pertussis]), namely, 033.0 (Bordetella pertussis) and 033.9 (whooping cough, unspecified organism). The age in months at hospital admission was calculated by using the age in days at admission divided by 30.4375. For cases from states where the age in days was not provided but the age in months was provided, the age in months was used as provided. A few states (eg, 5 states in 2003 and 2004) reported age at admission only in years and neither age in days nor age in months was available. To include all infant pertussis hospitalizations in the age group analyses, cases with age reported only as 0 years were weighted by using the standard NIS weighting procedure to give the national estimate of the number of such cases for that calendar year. We then distributed the cases in single-month age categories, following the weighted distribution of pertussis cases for which the age in months was available. The median proportion of weighted pertussis cases for which the age in months was imputed was 13.5% (range: 0%–29%) for 1993 to 2004. Confidence intervals (CIs) (95%) were calculated for the NIS rates on the basis of only nonimputed data, by using recommended weighting procedures.
The KID is a nationally representative database of pediatric inpatient stays that contains information on 2 million to 3 million inpatient discharges among children from ∼3000 hospitals in states participating in the Healthcare Cost and Utilization Project.16,17 KID data were available for 1997 (22 participating states), 2000 (27 states), and 2003 (36 states). The primary analysis was performed as described for the NIS. Five states in 2000 and 4 states in 2003 reported age at admission in years only. The proportions of weighted pertussis cases for which we imputed the age in months were 12% in 1997, 22% in 2000, and 23% in 2003. National estimates of the numbers (and 95% CIs) of pertussis hospitalizations were obtained by following recommended weighting procedures.
For the 2 most-recent KID sampling years (2000 and 2003), all additional diagnosis codes for infants with pertussis were reviewed for known pertussis complications (eg, pneumonia) and possible underlying health conditions (eg, congenital heart disease); procedure codes were also reviewed for expected procedures (eg, mechanical ventilation and extracorporeal membrane oxygenation) (Appendix⇓). Proportions with complications/procedures were estimated on the basis of weighted numbers.
Incidence Rates of Pertussis Hospitalization
In the SPSS, the proportion of infants with pertussis who were hospitalized was highest among infants 0 months of age (84%) and decreased with increasing age (Table 1). Infants 1 to 2 months of age had the highest incidence of pertussis hospitalization (Fig 1), followed by infants 0 months of age. Imputation of hospitalization status for infants without reported status (Table 1) modestly increased rates of pertussis hospitalizations for each year (Fig 1) and suggested an increase in incidence over the study period among infants ≤2 months of age (Fig 1).
NIS and KID
Pertussis hospitalization rates derived from the NIS were generally stable from 1993 through 2004 for infants ≤2 months of age (Fig 2). Rates for infants 3 to 11 months of age seemed to decrease. The findings were similar across the 3 years for which data were available from the KID. The rank order of rates according to age group was the same as in the SPSS, with infants 1 to 2 months of age having the highest rates. Including infants with imputed values for age in months, peak rates in the NIS for infants 1 to 2 months of age occurred in 1993 (305 hospitalizations per 100000 live births), 1999 (259 hospitalizations per 100000 live births), and 2004 (292 hospitalizations per 100000 live births). In each age group, the rates derived from the KID were approximately the same or slightly higher than those derived from the NIS (median KID/NIS ratio: 1997: 1.1:1; 2000: 1.2:1; 2003: 1.1:1) (Table 2 and Fig 2). Incidence rates for female infants were slightly higher than those for male infants (female/male ratio for <1 year of age in the KID: 2000: 1.13:1; 2003: 1.06:1).
When imputed data were included in the rate estimates, the pertussis hospitalization rates in the KID and SPSS were similar for infants 0 months of age but KID rates were ∼2 to 2.5 times higher than SPSS rates for other age groups (median KID/SPSS ratio: 1997: 2.6:1; 2000: 2.3:1; 2003: 2.0:1) (Table 2). For infants 1 to 2 months of age, 2003 hospitalization rates were 239 hospitalizations per 100000 live births (KID), 209 hospitalizations per 100000 live births (NIS), and 114 hospitalizations per 100000 live births (SPSS).
In the single-month age groups, 1-month-old infants had the highest incidence of pertussis hospitalization in the KID (Fig 3). In the KID for 2000 and 2003, 61% of pertussis hospitalizations involved infants 1 to 2 months of age and 86% of hospitalizations involved infants 0 to 3 months of age.
Complications in KID in 2000 and 2003
Pertussis was listed as the primary diagnosis in 80% and 78% of total cases in 2000 and 2003, respectively. For infants with a non–pertussis code listed first, the top codes listed first were those for acute bronchiolitis due to another organism (21% in 2000 and 28% in 2003) and respiratory syncytial virus (RSV) infection/pneumonia/bronchiolitis (20% and 28%, respectively).
Of the complications examined, apnea/respiratory distress was the most frequently diagnosed complication in both years (Table 3), coded on average for 19% of infants. Pneumonia (all types, and “pneumonia, in whooping cough”), the second most common complication, was most frequently coded among infants 0 months of age (14%–18%). Gastroesophageal reflux was coded for 12% of infants and RSV infection for 5%. In 2000, the infants with encephalitis for whom age in months was provided were 3 months of age; in 2003, the infants were 0 to 2 months of age.
Mechanical ventilation was coded for 11% and 15% of infants 0 months of age in 2000 and 2003, respectively, and the proportion decreased with increasing age. Of the infants who underwent mechanical ventilation for whom age in months was provided, 95% were ≤3 months of age; the oldest in 2000 was 4 months of age, and the oldest in 2003 was 5 months of age. Extracorporeal membrane oxygenation was coded for 14 infants in 2000 (oldest: 3 months of age) and 7 infants in 2003 (oldest: 2 months). A fatal outcome was recorded for 18 infants in 2000 and 14 infants in 2003, all ≤3 months of age.
A small proportion of infants <1 year of age had diagnosis codes that indicated possible significant underlying cardiac or pulmonary disease (≤4% per category) (Table 3). The lengths of hospitalization stay were similar in the 2 years, and the youngest infants had the longest stays (Table 3).
We examined pertussis hospitalizations among infants during the period 1993 to 2004 by using 2 national inpatient discharge databases and the national surveillance system. The rates derived from the NIS suggest that the incidence of pertussis hospitalizations in infants ≤2 months of age was generally stable during this period. In contrast, rates from the SPSS were approximately one half the rates from the NIS or KID, and the imputed SPSS data suggested an increase in incidence for infants ≤2 months of age. The increase seems to rely heavily on the results from 2004, the year with the greatest number of pertussis cases reported (25827 in all age groups combined) since 1959.18 Although an infant must be diagnosed as having pertussis to be counted in any of the systems, cases in the SPSS must also be reported by a health care professional to the health department. Improved reporting over the study period likely accounted for the increase in hospitalization rates for the youngest infants observed in the SPSS. However, the proportion of infants for whom hospitalization status was missing in the SPSS increased over time, which indicated that reporting did not improve in all aspects.
With the exception of infants 0 months of age, incidence rates of pertussis hospitalizations from the NIS and KID were ∼2 times greater than those from the SPSS. Sutter and Cochi19 performed a capture-recapture analysis for the period from 1985 to 1988 and found that only 34% of hospitalizations for infants diagnosed as having pertussis had been reported to the SPSS. Because it was not possible to link discharge records in the NIS and KID databases with individual records from another source, we could not determine whether NIS or KID rates still underestimated the true rates of hospitalization for diagnosed pertussis. Underestimation also could occur if pertussis was never diagnosed because of limitations in pertussis laboratory diagnostic testing20,21 or atypical clinical presentation.
The highest incidence of pertussis hospitalization occurred among infants 1 month of age, younger than the age at which the first dose of diphtheria/tetanus/acellular pertussis (DTaP) vaccine is generally recommended (2 months of age). The incidence for infants 0 months of age was lower than for infants 1 month of age. One explanation for this lower rate is that, because the incubation period is usually 7 to 10 days (range: 5–21 days), infants who are hospitalized at 0 months of age would have to be exposed during the first 2 weeks of life and develop significant symptoms quickly after a short incubation period. Those with a longer incubation period and/or more-gradual onset of symptoms would be 1 month of age at hospitalization. Less exposure to pertussis among infants 0 months of age may also explain lower rates in infants 0 months of age. Finally, although contemporary US data demonstrate low levels of maternal pertussis antibodies in newborns, it is theoretically possible that there is some degree of protection from transplacentally transferred antibody that diminishes by 1 month of age.22–25
From the peak incidence rate in the KID among infants 1 month of age, the rate decreased precipitously to 3 months of age and continued to decrease to 6 months of age. This pattern is most likely explained by the corresponding increase in pertussis vaccine coverage, as assessed by the National Immunization Survey (Fig 4). 3,26,27 The 60% lower incidence rate among infants 3 months of age, compared with infants 1 month of age, suggests that even 1 dose of DTaP vaccine provides protection against hospitalization with pertussis. Among infants 2 to 3 months of age with pertussis, Tanaka et al2 found lower odds ratio of hospitalization (0.69; 95% CI: 0.61–0.79) for infants who had received 1 dose of DTaP vaccine, compared with unimmunized infants.
The lower hospitalization rate among infants 3 months of age, compared with infants 1 month of age, might also result from factors other than immunization. For example, physicians might be more likely to hospitalize a 1-month-old infant than a 3-month-old infant with suspected pertussis, regardless of immunization status. Even if the younger group could be immunized, there might be physiologic (eg, cardiorespiratory) differences between these ages that would still necessitate more pertussis hospitalizations among the youngest infants.28 However, infants ≤2 months of age have the highest incidence of pertussis hospitalization, and routinely giving the first dose of DTaP vaccine as soon as infants reach 2 months of age (or perhaps even 6 weeks of age, the lowest end of the eligibility range) is warranted.
The youngest infants hospitalized with pertussis had the greatest risk of requiring mechanical ventilation or extracorporeal membrane oxygenation, having encephalopathy, or dying. Our proportions of hospitalized infants <1 year of age who had apnea/respiratory distress, pneumonia, seizures, or encephalopathy or who died were similar to those reported by O'Brien and Caro,29 who used discharge databases from 4 states covering 1996 to 1999. In their study, 14% of infants received care in a NICU or PICU, a variable not available in our databases. Data from national inpatient active surveillance networks in Canada30 and Australia31 reported rates of pneumonia, seizures, encephalopathy, and death among infants similar to those we obtained from the KID.
Other diagnoses were also made in infants with pertussis. RSV coinfection (5% in our study) was not unexpected, given the high incidence of RSV in infants. Crowcroft et al32 found RSV coinfection in 9 (36%) of 25 infants with pertussis who required ICU care in the United Kingdom during the non-RSV season. In our study, approximately 10% of infants were coded as having gastroesophageal reflux. It is not known whether this was considered a separate condition or a diagnosis describing the feeding problems (eg, posttussive vomiting) caused by pertussis. Few infants had additional codes for underlying cardiac or respiratory conditions. Information on gestational age was not available in the KID.
Estimates of complications from the KID based on small numbers should be interpreted with caution because they are unreliable. In addition, deaths that occurred at home or in the emergency department would not be captured in the KID. The numbers of infant pertussis deaths estimated from the KID (18 deaths in 2000 and 14 deaths in 2003), however, are remarkably close to the numbers of infant pertussis deaths reported to the National Notifiable Disease Surveillance System/SPSS (17 deaths in 2000 and 15 deaths in 2003) (Centers for Disease Control and Prevention, unpublished data, 2006).
Our analyses had other limitations. Age was calculated differently in the SPSS (age at cough onset) and the databases (age at hospital admission), which might have affected comparisons. However, because infants with pertussis who are hospitalized are most likely admitted within several days after illness onset, the difference in age definition was not a likely explanation for the higher rates in the NIS or KID.
An important limitation of the rate estimates concerns missing data for hospitalization status in the SPSS and for age in months in the NIS and KID. The proportion of cases with missing hospitalization status in the SPSS increased over time; for several states in recent years, none of the reported pertussis cases had information on hospitalization status. Although we think that a certain proportion of these infants were truly hospitalized, the accuracy of our imputation procedure is not known. We imputed data on age in months for infants whose ages were coded only as 0 years in the NIS and KID. However, we have no reason to think that the distribution of age in months at hospital admission among the 4 or 5 states with age coded as 0 years was different from that for the states with available data on age in months. Finally, the accuracy of International Classification of Diseases, Ninth Revision, coding for pertussis in the NIS or KID is not known.
With a goal of reducing infant exposure to B pertussis, the Advisory Committee on Immunization Practices recently recommended that, as a priority group, persons in close contact with infants should be immunized with tetanus/diphtheria/acellular pertussis vaccine (TdaP).20,33 This vaccine is also recommended for all adolescents and adults ≤65 years of age. Implementing these recommendations may lead to decreased circulation of B pertussis in the general population. In addition to strategies to reduce infant exposure, other strategies that should be fully evaluated are passively protecting infants by vaccinating pregnant women and actively immunizing infants early (even before 2 months of age).34
For any illness, changes can occur over time in hospitalization practices, testing practices, and methods used to make the diagnosis. The degree to which such changes affect trends observed in hospitalization rates from any data source can be difficult to quantify. Surveillance systems, like the SPSS, that rely on reported data are also subject to changes in reporting, another factor that is difficult to quantify and to control for in analyses. Assuming that the higher rates in the NIS and KID better reflect the true incidence of pertussis hospitalizations, our analysis suggests that, at least beginning with 1993, a large margin exists in which improved reporting to the SPSS could account for any observed increase in infant pertussis hospitalizations.
Nationally representative discharge databases are useful for monitoring infant pertussis hospitalizations. The impact on infant pertussis of the use of the new TdaP vaccines for adolescents and adults should be monitored through such discharge databases.
We are grateful to Qian Li (National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention) for providing coverage data from the National Immunization Survey and to Lynne McIntyre (National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention) for her valuable editorial assistance.
- Accepted August 15, 2007.
- Address correspondence to Margaret M. Cortese, MD, Centers for Disease Control and Prevention, National Immunization Program, 1600 Clifton Rd, MS E61, Atlanta, GA 30333. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
The findings and conclusions of this article are those of the authors and do not necessarily represent the views of the funding agency.
What's Known on this Subject
Incidence rates of reported infant pertussis among infants ≤4 months of age and reported infant pertussis deaths increased from the 1980s to the 1990s, according to national passive surveillance data, despite a fully implemented vaccination program.
What This Study Adds
Recent data on the incidence of pertussis hospitalization among infants in the United States from national discharge databases are presented; rates were generally stable among the youngest infants and were 2 times higher than those estimated from the passive system. Infants 1 to 2 months of age had the highest rates.
- ↵US Public Health Service. Reported Incidence of Selected Notifiable Diseases: United States, Each Division and State, 1920–50. Washington, DC: US Public Health Service; 1953:240–242
- ↵Centers for Disease Control and Prevention. National Notifiable Diseases Surveillance System. Available at: www.cdc.gov/epo/dphsi/nndsshis.htm. Accessed August 2, 2007
- ↵Martin JA, Hamilton BE, Menacker F, Sutton PD, Mathews TJ. Preliminary births for 2004: infant and maternal health. In: Health E-stats. Hyattsville, MD: National Center for Health Statistics; 2005. Available at: www.cdc.gov/nchs/products/pubs/pubd/hestats/prelimbirths04/prelimbirths04health.htm. Accessed January 24, 2007
- ↵Matthew TJ, Hamilton BE. Trend analysis of the sex ratio at birth in the United States. National vital statistics reports; vol 53 no 20. Hyattsville, Maryland: National Center for Health Statistics. 2005. Available at: www.cdc.gov/nchs/data/nvsr/nvsr53/nvsr53_20.pdf. Accessed Jan 24, 2007
- ↵SAS Institute. SAS/STAT User's Guide, Version 8. Cary, NC: SAS Institute; 1999
- ↵Agency for Healthcare Research and Quality. Overview of the Nationwide Inpatient Sample. Available at: www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed December 7, 2006
- ↵Agency for Healthcare Research and Quality. Introduction to the HCUP Nationwide Inpatient Sample. Available at: www.hcup-us.ahrq.gov/db/nation/nis/NIS_Introduction_2004.pdf. Accessed December 7, 2006
- ↵Houchens RL, Alixhauser A. Using the HCUP Nationwide Inpatient Sample to estimate trends (updated for 1988–2004). HCUP Methods Series Report #2006-05 Online. August 16, 2006. US Agency for Healthcare Research and Quality. Available at: www.hcup-us.ahrq.gov/reports/2006_05_NISTrendsReport_1988-2004.pdf. Accessed Dec 6, 2006
- ↵Agency for Healthcare Research and Quality. Overview of the Kids’ Inpatient Database (KID). Available at: www.hcup-us.ahrq.gov/kidoverview.jsp. Accessed December 7, 2006
- ↵Agency for Healthcare Research and Quality. Introduction to the HCUP Kids’ Inpatient Database (KID), 2003. Available at: www.hcup-us.ahrq.gov/db/nation/kid/Introduction_to_KID_2003.pdf. Accessed December 7, 2006
- ↵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.2006;55 (RR-3):1– 34
- Healy CM, Munoz FM, Rench MA, Halasa NB, Edwards KM, Baker CJ. Prevalence of pertussis antibodies in maternal delivery, cord, and infant serum. J Infect Dis.2004;190 (2):335– 340
- Healy CM, Rench MA, Edwards KM, Baker CJ. Pertussis serostatus among neonates born to Hispanic women. Clin Infect Dis.2006;42 (10):1439– 1442
- ↵Smith PJ, Hoaglin DC, Battaglia MP, Khare M, Barker LE. Statistical methodology of the National Immunization Survey, 1994–2002. Vital Health Stat 2.2005; (138):1– 55
- ↵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.1999;28 (6):1238– 1243
- ↵Crowcroft NS, Booy R, Harrison T, et al. Severe and unrecognised: pertussis in UK infants. Arch Dis Child.2003;88 (9):802– 806
- ↵Kretsinger K, Broder KR, Cortese MM, et al. Preventing tetanus, diphtheria, and pertussis among adults: use of tetanus toxoid, reduced diphtheria toxoid and acellular pertussis vaccine: recommendations of the Advisory Committee on Immunization Practices (ACIP) and recommendation of ACIP, supported by the Healthcare Infection Control Practices Advisory Committee (HICPAC), for use of Tdap among health-care personnel. MMWR Recomm Rep.2006;55 (RR-17):1– 33
- ↵Belloni C, De SA, Tinelli C, et al. Immunogenicity of a three-component acellular pertussis vaccine administered at birth. Pediatrics.2003;111 (5):1042– 1045
- Copyright © 2008 by the American Academy of Pediatrics