Sensitivity Analysis for Observational Studies with Recurrent Events

Area of Science:

• Epidemiology
• Genetics
• Infectious Diseases

Background:

• Sickle cell trait (Haemoglobin AS) is a common genetic adaptation in malaria-endemic regions.
• The protective effect of Haemoglobin AS against severe malaria is well-established, but its impact on malaria fevers requires further investigation.
• Observational studies are crucial for understanding disease patterns but are susceptible to confounding.

Purpose of the Study:

• To investigate the effect of Haemoglobin AS on the incidence of malaria fevers in children.
• To assess the robustness of this association against potential unmeasured confounding.
• To introduce and apply a novel sensitivity analysis method for recurrent event data.

Main Methods:

• Observational study design analyzing malaria fever incidence.
• Application of a new sensitivity analysis technique for recurrent event data.
• Evaluation of the influence of potential unmeasured confounders on the observed association.

Main Results:

• Strong evidence suggests Haemoglobin AS significantly reduces the hazard rate of malaria fevers in children.
• The protective association remains robust even when considering potential unmeasured confounding.
• A hypothetical unmeasured confounder would need substantial influence to negate the observed protective effect.

Conclusions:

• Haemoglobin AS confers significant protection against malaria fevers in children.
• The findings are robust to unmeasured confounding, strengthening the causal inference.
• The developed sensitivity analysis method is valuable for future observational studies on recurrent events.