Response to Letter to the editor: "A quantitative bias analysis to assess the impact of unmeasured confounding on associations between diabetes and periodontitis".