PERCH Case Control

Nine sites in seven countries participated in PERCH, five in Africa and four in Asia.

This is a merged dataset containing epidemiological data from all study sites, and has been made available for online exploratory data analysis of aggregate data.

Data from this merged dataset cannot be downloaded. Instead, data from each country is available as a separate dataset; data from each country needs to be requested and downloaded separately:

PERCH Bangladesh Case Control
PERCH Kenya Case Control
PERCH Mali Case Control
PERCH South Africa Case Control
PERCH Thailand Case Control
PERCH The Gambia Case Control
PERCH Zambia Case Control

Background: Existing knowledge on the etiologies of childhood pneumonia is largely based on studies from the 1980s and 1990s. With expanded use of pneumococcal conjugate vaccine and Haemophilus influenzae type b conjugate vaccine and changes in host and environmental factors, a new evidence base to guide treatment and prevention strategies is needed.

Objectives:

Determine the etiology of pneumonia by integrating results from multiple specimens
Estimate the fraction of pneumonia attributable to pathogens for which vaccines are available or under development (e.g., pneumoniae, respiratory syncytial virus, parainfluenza virus, influenza and S. aureus) as well as other known, but poorly quantified causes of pneumonia in children
Assess risk factors for pneumonia

Methodology:

Geographic Location/Study Sites:

Dhaka, Bangladesh
Matlab, Bangladesh
Basse, The Gambia
Kilifi, Kenya
Bamako, Mali
Soweto, South Africa
Nakhon Phanom, Thailand
Sa Kaeo, Thailand
Lusaka, Zambia

Dates of Data Collection: August 2011-February 2014

Study Design: Case-control study

Eligibility Criteria: Cases were defined as children aged 1-59 months hospitalized with pre-2013 WHO-defined severe or very severe pneumonia. Controls were randomly selected from the community and age-frequency matched to cases.

Data Collection: Multiple specimens were collected from each case and control and tested for over 30 pathogens. Data were collected on clinical signs and symptoms, medical history, chest x-ray results, lab results, treatment, outcomes, and demographic, household, and environmental risk factors.

Study Documentation: The PERCH Background and Methods page provides links to download the study protocol, CRFs, and standard operating procedures (SOPs) for clinical, laboratory, data, and regulatory procedures.

Study protocol:

PERCH Study Protocol

Data Collection forms: (used to collect raw data)

PERCH Case Report Forms

Data Dictionaries:

PERCH Forms and Codes
PERCH Analytic Data Dictionary
PERCH EDC Data Dictionary

Standard Operating Procedures:

PERCH Clinical SOPs
PERCH Laboratory SOPs
PERCH Data Management SOPs
PERCH Regulatory SOPs

Additional Resources: PERCH has published two supplements in Clinical Infectious Diseases:

Rationale and Study Design for PERCH - contains 15 papers
Basis for the Primary Etiology Results - contains 23 papers
Requests to access the lung sound files and/or the chest X-rays should be directed to Christine Prosperi: [e-mail]

ClinEpiDB Data Integration: Epidemiological data files were provided to ClinEpiDB as SAS files. These datasets were converted to txt files and merged by unique ID and redundant or administrative columns were dropped from presentation on ClinEpiDB.org. All dates were obfuscated per participant through the application of a random number algorithm that shifted dates no more than seven days to comply with the ethical conduct of human subjects research.

Acknowledgements: Thank you to the families and children who participated in PERCH.

Financial Support: PERCH was supported by a grant from The Bill & Melinda Gates Foundation.

Ethics Statement: The study protocol was approved by the Institutional Review Boards or Ethical Review Committees for each of the seven institutions and at The Johns Hopkins School of Public Health.

Last Updated: July 26, 2024The Pneumonia Etiology Research for Child Health (PERCH) study was a prospective, multi-center, case-control study of respiratory disease in children 1-59 months of age. Each site recruited cases with severe or very severe pneumonia from hospitals, randomly selected controls from the community, and age-frequency matched cases and controls. Participant observations were assessed at enrollment, and cases were followed up at 24 hours, 48 hours, and 30 days. This is a merged dataset from all 7 study sites and had been made available for online exploratory data analysis of aggregate data.