Giraffe Effect

Chapter summary

  • When reading articles, look at the whole picture and the real issues. The real issue is not obesity alone. Consider metabolic syndrome.
  • Do not believe everything you read. Research, investigate and experiment.
  • The Giraffe Effect: science and politics focus on the big, visible issue, while the real consequences may lie elsewhere.

What is the giraffe effect?

  • The term comes from data analyst Pini Yakuel. Giraffes are portions of data that dominate the rest and hide important insights.
  • Imagine watching animals on a nature reserve. A giraffe stands out, and the smaller animals all look small.
  • If you look away from the giraffe, you see the differences in size among the others are much larger than you thought.
  • When data contains very large members, important differences among the rest disappear from view. Sometimes this leads to wrong conclusions.

Obesity as the giraffe

  • Everyone looks at obesity. The popular press quotes the 65% of people who are overweight or obese.
  • The less visible animals:
    • 38% of people have metabolic syndrome.
    • About 15% have chronic kidney disease (CKD). In Australia 13% of people are understood to have CKD, against 4% with diabetes, and one in seven hospital admissions involves CKD.
    • 4% have life-limiting diabetes.
    • 20% of normal-weight people have metabolic syndrome and are on the path to chronic disease, out of the limelight.
  • These lower-profile groups are the ones using up much of the health system.

Different views

  • Obesity as the target: Most health policy, media and research funding focus on weight and BMI.
  • The author’s view: Weight is only a marker. The target should be metabolic health, including normal-weight people who are “thin on the outside, fat on the inside”.
  • A caution: Obesity and metabolic syndrome overlap heavily, so the author is not saying obesity does not matter.

Take-away points

  • Ask what the headline number hides.
  • Measure waist, blood sugar and blood pressure, not only weight.
  • Good health is possible at a range of weights, and poor health at a normal weight.

References

  • Yakuel, P. (2013) Beware of the giraffes in your data. Optimove / GigaOM.
Scroll to Top