Sir, your readers may not know that there is a significant correlation (99.26%) between the divorce rate in Maine, USA and the per capita consumption of margarine over 2000-2009.1 This may have little to do with dentistry, however, it raises an important counterpoint to an issue raised by my esteemed colleagues Dr de Gea Rico and Dr Williams in their recent letter to the BDJ.2
In the letter, they highlight the findings of Dayer et al.'s 13 year Lancet study into the changing incidence of infective endocarditis (IE) since the adaption of the NICE guidelines.3,4 (I would similarly urge my dental colleagues to remain vigilant for the sign and symptoms of IE especially in patients at high risk for the condition.)
The headline finding of the study was the significant increase of IE in England since the introduction of the 2008 NICE guidelines and abolition of antibiotics prophylaxis.
Importantly, the study highlighted that it could not prove a causal link between the cessation of antibiotics prophylaxis and an increase in IE incidence. There were no data on the causative organism which led to IE episodes. This is clearly outlined in the study and undermines any assumption that oral bacteria, such as Steptococcus viridans, may be causally linked to episodes of IE. In fact Dayer et al. speculated about other possible non-oral causes for the increase in IE incidence. Interestingly, the study also highlighted the incidence of IE increasing prior to the introduction of NICE guidelines, for unknown reasons, which may suggest a non-oral cause for the increase.
The authors of the study do point out that their analytical model was based upon several estimates which in itself could have led to potential errors. Readers should be wary of drawing causal conclusions from the data outlined in the IE study as much as the spurious suggestion that chemicals in margarine were responsible for matrimonial disharmony in the population of Maine in the previous decade.
The scope of Dayer et al.'s paper is commendable and its findings remain interesting. Though perhaps most importantly it keeps IE in the forethoughts of the dental profession and highlights the evolving nature of the research on which our clinical guidelines are built.
References
Spurious correlations. Available online at http://www.tylervigen.com/spurious-correlations (dccessed July 2015).
de Gea Rico A, Williams L R . The power of data. Br Dent J 2015; 218: 440.
Dayer M J, Jones S, Prendergast B, Baddour L M, Lockhart P B, Thornhill M H . Incidence of infective endocarditis in England, 2000-13: a secular trend, interrupted time-series analysis. Lancet 2015; 385: 1219–1228.
National Institute of Clinical Excellence (2008). Prophylaxis against infective endocarditis: Antimicrobial prophylaxis against infective endocarditis in adults and children undergoing interventional procedures. Available online at http://publications.nice.org.uk/prophylaxis-against-infective-endocarditis-cg64 (accessed July 2015).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Aslam, A. Research ideas: Correlation does not imply causation. Br Dent J 219, 49 (2015). https://doi.org/10.1038/sj.bdj.2015.585
Published:
Issue Date:
DOI: https://doi.org/10.1038/sj.bdj.2015.585
This article is cited by
-
The insight of why: Causal inference in Earth system science
Science China Earth Sciences (2023)