Specifying a target trial protocol to estimate the effect of preconception treatment of periodontitis on time-to-pregnancy: A commentary and applied example.

BACKGROUND

The target trial framework was developed as a strategy to design and analyze observational epidemiologic studies with the aim of reducing bias due to analytic decisions. It involves designing a hypothetical randomized trial to answer a question of interest and systematically considering how to use observational data to emulate each trial component.

AIMS

The primary aim of this paper is to provide a detailed example of the application of the target trial framework to a research question in oral epidemiology.

MATERIALS AND METHODS

We describe the development of a hypothetical target trial and emulation protocol to evaluate the effect of preconception periodontitis treatment on time-to-pregnancy. We leverage data from Pregnancy Study Online (PRESTO), a preconception cohort, to ground our example in existing observational data. We discuss the decision-making process for each trial component, as well as limitations encountered.

RESULTS

Our target trial application revealed data limitations that precluded us from carrying out the proposed emulation. Implications for data quality are discussed and we provide recommendations for researchers interested in conducting trial emulations in the field of oral epidemiology.

DISCUSSION

The target trial framework has the potential to improve the validity of observational research in oral health, when properly applied.

CONCLUSION

We encourage the broad adoption of the target trial framework to the field of observational oral health research and demonstrate its value as a tool to identify directions for future research.