Study site and population

This study employed specific criteria [21] for site selection, and finally selected two sites in China: Yiwu, Zhejiang province and Yongcheng, Henan province. Yiwu, located in the central-western part of Zhejiang province, has a large number of temporary residents. Yongcheng, located in the central part of Henan province, with a lower population density.

In our study, the population involved in calculating pertussis hospitalization was the long-term residents in Yiwu and Yongcheng, which were defined as the residents who have lived in these cities for more than six months. Population data were obtained from Health Statistics and the Immunization Program Management of the Centers for Disease Control and Prevention of the study sites. In 2021, Yiwu had a resident population of 1,859,390, served by 24 medical institutions. Meanwhile, Yongcheng with a resident population of 1,250,409, served by 41 medical institutions. Together, a total of 3.11 million residents were covered by the two sites.

Healthcare utilization and attitudes surveys (HUAS)

A healthcare utilization survey (HUAS) was conducted in Yiwu and Yongcheng from 2019 to 2020 for the selection of surveillance hospitals (SHs) selection. The sampling flow of HUAS is shown in Supplementary Fig. 1, and the detailed description of methodology regarding HUAS has been published in Yu, J. et al. [21].

The hospitalization coverage rate was calculated by dividing the number of people who expressed a preference to be hospitalized in specific hospitals by the number of HUAS respondents, as shown in formula [1]. Hospitals with a collective hospitalization coverage rate close to 80% were selected as SHs. In Yiwu, Yiwu Center Hospital, the Fourth Affiliated Hospital of Zhejiang University School of Medicine, and Yiwu Maternal and Child Healthcare Hospital were selected as SHs. In Yongcheng, Yongcheng City People’s Hospital, Yongcheng City Central Hospital, and Yongcheng Maternal and Child Healthcare Hospital were selected as SHs.

$$\:Hc\:=\frac&No.\:of\:people\:who\:expressed\:a\:preference\cr&\quad\:to\:be\:hospitalized\:in\:SHs\:at\:age\:group\:i\end}$$

(1)

Hc, hospitalization coverage rate of SHs at age group i; SHs, surveillance hospitals; i, each age group.

Patient enrollment

All hospitalized patients admitted to SHs any day of the week, except for national holidays, were screened as the following three steps to identify suspected pertussis cases: In the first step, patients diagnosed with the International Classification of Diseases 10th Revision (ICD-10) codes listed in Supplementary Table 1 were included. These codes include A37 (Pertussis), J00-J22 (Acute respiratory infections, influenza and pneumonia), J40-J47 (Chronic lower respiratory diseases), R05 (Cough), R09.2 (Respiratory arrest), P22 (Respiratory distress of newborn), P28.2 (Cyanotic attacks of newborn), P28.3 (Primary sleep apnea of newborn), P28.4 (Other apnea of newborn), and P28.5 (Respiratory failure of newborn). In the subsequent step, patients who met ≥ 1 of the criteria of suspected pertussis case were selected (Fig. 1). In the final step, patients who met ≥ 1 of the exclusion criteria were excluded (Fig. 1).

Fig. 1

Enrollment and testing procedure of 366 suspected pertussis cases (from June 1, 2021 to May 31, 2022 in Yiwu, Zhejiang province, and Yongcheng, Henan province, China). HUAS, healthcare utilization and attitudes surveys; SHs, surveillance hospitals; EMRs, electronic medical records. * For individuals aged 1 year old and older, pertussis-like symptoms were defined as paroxysmal cough, inspiratory whoop, post-tussive vomiting and cough increasing at night; for infants aged less than 1 year old, pertussis-like symptoms were defined as apnea, paroxysmal cough, inspiratory whoop, and post-tussive vomiting

Patient data collection

During enrollment, face-to-face interviews were conducted to collect sociodemographic, clinical, and epidemiological data of the enrolled cases. Clinical signs and symptoms, including paroxysmal cough, post-tussive emesis, inspiratory whooping, increasing cough at night, apnea, expectoration, polypnea, cyanosis, fever (self-reported), runny nose, lacrimation, sore throat, hoarseness, chest pain, weakness, night sweat, headache, and myalgia, were documented by clinicians.

Specimen collection and laboratory testing

Nasopharyngeal swabs of the enrolled cases were collected during the study period. Yongcheng CDC performed culture (Charcoal Agar, Thermo Fisher Scientific) and PCR (Fluorescent PCR, Pertussis bacterium nucleic acid test kit, bioPerfectus technologies) with specimens collected in Yongcheng SHs. Yiwu CDC only performed culture (Charcoal Agar, Thermo Fisher Scientific) with swab specimens collected in Yiwu SHs, and samples collected for PCR analyses were transferred to the central laboratory of China CDC for Bordetella pertussis infection analysis (Triple-Channel Real-Time PCR Method, B. pertussis, B. parapertussis and B. holmesii Detection Kit, MABSKY).

Data management and analysis

Data collected in the study were centrally managed by China CDC through an internal online platform (http://10.249.35.12/task/15). Local staff at the two sites entered completed forms into the system, and data was securely uploaded via a virtual private network. Data analysis was conducted using R software (version 3.1.0).

Pertussis hospitalization rates were calculated by dividing the number of lab-confirmed pertussis cases by the catchment population, calculated as the product of hospitalization coverage rate of SHs and the resident population. The 95% credible intervals (95%CrI) of hospitalization coverage rate of SHs were calculated using the bootstrap method with 1,000 replications. Pertussis hospitalization rate by age group was calculated using relevant parameters derived from each age group, as shown in the formula [2]. The ages were defined by the hospitalization date.

$$\:Hp\:=\frac\:&\mathbf\mathbf.\:\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf\mathbf\cr&\quad\:\mathbf\mathbf\:\mathbf\mathbf\mathbf\:\mathbf\mathbf\:\mathbf\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf\mathbf\:\mathbf\end}\mathbf\times\:\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\mathbf\:\mathbf\mathbf\:\mathbf\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf\mathbf\:\mathbf\:\mathbf\mathbf\:\mathbf\mathbf\mathbf\:\mathbf\mathbf\mathbf\mathbf}$$

(2)

Hp, pertussis hospitalization rate; Hc, hospitalization coverage rate of SHs at age group i; SHs, surveillance hospitals; i, each age group.

Individual case review on immunization history

The vaccination information of hospitalized pertussis cases aged ≤ 5 years was reviewed. A dose of vaccination was considered to be receive if illness onset was ≥ 14 days after vaccination [22], and the month groups of children aged < 1 year were classified by date of illness onset to better understand whether infants were protected by vaccines in a timely manner.