3

Suppose we say that the prevalence rate of some disease is $1$ in $50$. Does this make sense? Or should be convert this into a $ \%$? That is, should we say that the prevalence rate is $2 \%$?

gung - Reinstate Monica
  • 145,122
prevques
  • 31
  • Other responses have answered your question, but I just want to remind you that prevalence is about the phenomenon at a time or during a time period, so if you asked for a "suitable way to present" it, I'd suggest don't forget attaching the time reference. – Penguin_Knight Jun 05 '13 at 17:41

3 Answers3

11

Two thoughts in addition to the other answers:

  • prevalence is a fraction (ratio), not a rate.
  • a rate is a fraction where the units in enumerator and denominator differ. The difference is usually a time (duration) in the denominator. Examples: incidence rate, growth rate, decay rate.
    e.g. incidence rate: number of newly diagnosed disease X cases per (population size and time) - unit: per person years, that is 1 divided by a time-unit
  • A ratio always has the same unit in denominator and enumerator, and is therefore unitless. This is the case with your prevalence. Such numbers can be expressed as percent, permil, etc.

Even though you can express the prevalence as just a percentage (they are mathematically equivalent), there are recommendations against this because there are practical differences.

  • It is easy to forget explaining what exactly was in the denominator - and this can cause considerable chaos (see the paper below for some examples). Particularly for prevalences you need to be very explicit about the population in the denominator as prevalences can vary orders of magnitude between different patient populations.
    If you report prevalence as percentage, you really need to give the denominator: prevalence of disease X is 2 % of the blabla patient population
  • Gigerenzer goes one step further and recommends to report frequencies with whole numbers instead as this form is easier to grasp in an intuitive way (and does not even need more typing):
    1 out of 50 blabla patients or 2 out of 100 blabla patients
    IMHO, the first has the advantage of giving information about the precision of the estimate (20 out of 1000 is more precise than 1 out of 50), and in some cases may allow a better mental picture of the magnitude (1 out of 3 or 2 out of 7 patients vs. 30 out of 100). Of course, decimal fractions or ratios with powers of 10 in the denominator are easier to compare.

Literature e.g. Gigerenzer et al.: Helping doctors and patients to make sense of health statistics, Psychological Science in the Public Interest, 8 (2007), 53–96. (particularly sections "Transparent Numbers" and "Use frequency statements, not single-event probabilities")

cbeleites unhappy with SX
  • 38,684
5

In my mind there is no difference between the two. To say that the prevalence rate of a disease being 1 in 50 is saying that $\frac{1}{50}$ or 2% of all people suffer from the disease. The odds of having the disease are 49:1 against it.

Eric Peterson
  • 2,401
  • 16
  • 22
5

Because prevalence is the proportion of individuals afflicted with a certain condition in a cross sectional sample, it makes sense to report this value as a percent. Rates, unlike proportions, which need explicit unit-times in the denominator and have the hairy possibility of multiple events per individual in a unit person-time frame. This makes proportions more suitable for reporting in a percentage form.

AdamO
  • 62,637
  • I find your 2nd sentence very confusing. Do you mean: "Rates need explicit unit-times in the denominator and have the hairy possibility of multiple events per individual in a unit person-time frame. Unlike rates, proportions are more suitable for reporting in a ratio form" (trying to sort out to what the "which"s and "them" refer)? – cbeleites unhappy with SX Jun 05 '13 at 10:33