Vaccine Effectiveness Case-Control Studies: Out of Control?

• Activity Number: 411
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Elizabeth Zell*+ and Jennifer Verani and Michelle Groome and David Moore and Cynthia Whitney and Shabir A Madhi
• Companies: CDC and CDC and Respiratory and Meningeal Pathogens Research Unit and Respiratory and Meningeal Pathogens Research Unit and CDC and National Institute for Communicable Diseases
• Keywords: Vaccine Effectiveness ; Case-control study ; pneumonia ; Hospital Controls ; Community Controls

Abstract

Case-control studies are commonly used to evaluate vaccine effectiveness (VE) following licensure. As many developing countries introduce newer, more costly vaccines, evaluation of the impact of these vaccines is important to justify their continued use. The use of hospital controls (HC) has become a common approach for these evaluations; rather than the more difficult community controls (CC). In an evaluation of pneumococcal conjugate VE against presumed bacterial pneumonia, we compared and contrasted VE estimates generated using HC and CC among HIV-uninfected children. Cases and HC were enrolled at Chris Hani-Baragwanath Hospital, Soweto, South Africa. CC were identified through birth registers at the same hospital. All controls were matched on case date of birth. Vaccinations received =14 days before case onset dates were considered valid doses. Dose- and age-based crude VE estimates for hospital controls ranged from 3-50% and for community controls from 23-65%. Although not statistically different, point estimates were higher using CC. Here we compare and contrast findings from these two groups to make recommendations for future VE studies using the case-control methodology.