How Rare is "Rare"?

What does it mean to say a cancer is rare? 

According to a recent article in Oncology Times, it depends on where you live. In the United States a type of cancer has to affect fewer than 200,000 people, while in Europe the figure is fewer than 5 out of 10,000.

[Plugging in the US Census Bureau figure of 310,838,574 for the U.S. population, a type of cancer is considered rare in the U.S. if it affects less than 6.4 out of 10,000.]

Where we get into trouble is when we define "rare" using incidence for one geographic location and prevalence in another, because only prevalence is affected by life expectancy.

• Incidence =   # patients who are newly diagnosed with same disease in a given year
• Prevalence = # patients who are living with same disease at a given moment in time

If life expectancy is long, the prevalence rises. For example, relatively few men are diagnosed with testicular cancer each year. But because the disease is frequently cured, the population of survivors of testicular cancer is sizeable, i.e., the prevalence is high.

If life expectancy is short, the prevalence is low. For instance, relatively many people are diagnosed with small cell lung cancer (SCLC). But because the disease is usually fatal fairly quickly, the population of SCLC survivors is small, i.e., the prevalance is low.

My next post will discuss why it matters.

 

Comments

[Debby]

Great topic. Looking forward to further discussion and analysis of figures.

[Felicity Lenes]

Great post! Similar to when you explored the issue of statistical vs. vernacular use of the word "significant." This is something many individuals may not know that can really help us understand our scientists and physicians!