Topic 11B: Hierarchies of Evidence

1 Introduction to hierarchies of evidence

In this computer lab, we will discuss Hierarchies of Evidence (see, e.g., Greenhalgh (1997)) and evaluate several studies under this framework.

Consider the following figure:

As we have already seen throughout this subject, not all research is considered equal. The above diagram provides a ranking of different types of evidence, in order from strongest (1. Systematic reviews and meta analyses) to weakest (7. Expert opinion).

We have discussed some of the above types of evidence in Topics 3B and 4B. However, the types of evidence we have discussed usually relate to the results of only one research paper. For even stronger evidence, a systematic review may be undertaken, whereby a collection of studies on the same or a related topic are gathered, and their results are reviewed and analysed (see, e.g., Ahn and Kang (2018)). As part of this systematic review, a meta analysis may be carried out. This involves combining the statistical results of the individual studies to estimate an overall effect. Because systematic reviews and meta analyses combine the results of multiple research papers, they are considered the strongest type of evidence.

1.1

In your own words, write a sentence explaining each of the seven types of evidence displayed in the above diagram. For some types, the following will be helpful to refer to:

2 Evaluating studies

2.1

Recall the article (Coren 2016) we considered in Computer Lab 1B which discussed how dogs feel about hugs. Of the seven types of evidence, what type of evidence do you think is provided by the article?

2.2

Recall the paper called, Soft Drink Consumption: Do We Know What We Drink and Its Implication on Health? (Martín et al. 2018), which we considered in the Topic 10B Computer Lab. Identify which type of evidence is provided in the article.

2.3

Consider this paper (Dai et al. 2019) which looks at preventing diabetes incidence in people with prediabetes. Identify which type of evidence is provided by the article.

2.4

Consider this paper (Unit 2005) which looks at the efficacy and safety of cholesterol-lowering treatment. Identify which type of evidence is provided by the article.

References

Ahn, EunJin, and Hyun Kang. 2018. “Introduction to Systematic Review and Meta-Analysis.” Korean Journal of Anesthesiology 71 (2): 103.
Bowden, Jack, and Chris Jackson. 2016. “Weighing Evidence ‘Steampunk’ Style via the Meta-Analyser.” The American Statistician 70 (4): 385–94. https://doi.org/10.1080/00031305.2016.1165735.
Bristol, TARG. 2017. “Data Visualisation and Manipulation.” Youtube. 2017. https://www.youtube.com/watch?v=i675gZNe3MY&t=94s.
Coren, Stanley. 2016. “The Data Says "Don’t Hug the Dog!".” Psychology Today. 2016. https://www.psychologytoday.com/au/blog/canine-corner/201604/the-data-says-dont-hug-the-dog.
Dai, Xia, Lu Zhai, Qingyun Chen, Joshua D Miller, Lirong Lu, Cunyi Hsue, Lin Liu, et al. 2019. “Two-Year-Supervised Resistance Training Prevented Diabetes Incidence in People with Prediabetes: A Randomised Control Trial.” Diabetes/Metabolism Research and Reviews 35 (5): e3143.
Greenhalgh, Trisha. 1997. “Getting Your Bearings (Deciding What the Paper Is About).” BMJ 315 (7102): 243–47.
Martín, Ismael San Mauro, Elena Garicano Vilar, Mario Rincón Barrado, and Víctor Paredes Barato. 2018. “Soft Drink Consumption: Do We Know What We Drink and Its Implication on Health?” Mediterranean Journal of Nutrition and Metabolism 11 (1): 1–10.
Unit, Epidemiological Studies. 2005. “Efficacy and Safety of Cholesterol-Lowering Treatment: Prospective Meta-Analysis of Data from 90 056 Participants in 14 Randomised Trials of Statins.” Lancet 366 (9493): 1267–78.


These notes have been prepared by Amanda Shaker. The copyright for the material in these notes resides with the authors named above, with the Department of Mathematics and Statistics and with La Trobe University. Copyright in this work is vested in La Trobe University including all La Trobe University branding and naming. Unless otherwise stated, material within this work is licensed under a Creative Commons Attribution-Non Commercial-Non Derivatives License BY-NC-ND.