Application of a self-controlled case series study to a database study in children

Area of Science:

• Pharmacoepidemiology
• Pediatric Drug Safety
• Health Informatics

Background:

• Post-marketing surveillance is vital for identifying drug safety issues in vulnerable pediatric populations often excluded from clinical trials.
• Electronic health databases offer a valuable resource for monitoring the safety of marketed pharmaceutical products.

Purpose of the Study:

• To evaluate the self-controlled case series (SCCS) method against cohort studies for pediatric pharmacoepidemiology using administrative data.
• To investigate the impact of defining eligible analysis periods and handling recurrent events on study outcomes.

Main Methods:

• Utilized an anonymized Japanese hospital administrative database (April 2003-August 2011).
• Calculated incidence rate ratios (IRRs) for diarrhea, bronchitis, and eczema associated with palivizumab in young children.
• Defined eligible periods as first-time inpatient stays (EPA) or continuous periods with short intervals between visits (EPB) for sensitivity analysis.

Main Results:

• Analyzed 70,771 patients, with 641 exposed to palivizumab.
• Reported age-adjusted IRRs for diarrhea, bronchitis, and eczema as 3.0, 10.3, and 16.9, respectively, using multiple events and EPB.
• Observed significant variations in IRRs between the two defined eligible periods, highlighting the method's sensitivity to period definition.

Conclusions:

• The self-controlled case series method shows promise for pediatric pharmacoepidemiological research.
• Sensitivity analyses are essential due to the within-individual comparison nature of the SCCS method.
• Careful consideration of inpatient and outpatient data handling is necessary for robust results.