Berkson’s paradox

Berkson’s paradox happens when given two independent events, if you only consider outcomes where at least one occurs, then they become negatively dependent.  More technically, this paradox occurs when there is ascertainment bias in a study design. Let me provide an example. Consider the case where patients can have diabetes or HIV.  Assume that patients have a positive probability of…

The problem with p-values

Interesting article in Aeon on why p-values may not be the best way to determine the probability we are observing a real effect in a study. Tests of statistical significance proceed by calculating the probability of making our observations (or the more extreme ones) if there were no real effect. This isn’t an assertion that…

AA and selection bias

This video that discusses whether alcoholics anonymous actually improves the outcomes of alcoholics who attend the meeting.  More broadly, the video the AA treatment effect discussion serves as an example for expounding on some fundamental statistical issues such as selection bias, randomization, intention to treat, marginal effect, instrumental variables, and others.

US Healthcare Spending Projections

In 2016 we will hit a milestone: national health spending per capita is projected to exceed $10,000 for the first time.  This estimate is from an article by Keehan et al. (2016).  In this paper, CMS’ Office of the Actuary (OACT) estimates costs not only this year but over the coming 10 years.  According to their projects,…

The Intuition behind Bayes Theorem

Bayes Theorem is well-known in law of probability. Mathematically, you could write it as: P(A|B)=P(A and B)/Pr(B) = P(B|A)*P(A)/P(B). An interesting interview in Scientific American with Decision theorist Eliezer Yudkowsky explains Bayes Theorem more intuitively. I might answer that Bayes’s Theorem is a kind of Second Law of Thermodynamics for cognition. If you obtain a…

Average prevalence of “sickness”

Despite the large number of illnesses defined by the International Statistical Classification of Diseases and Related Health Problems (ICD) disease coding system, health systems need to know how many encounters they are likely to experience each month. One gauge for this is the prevalence of sickness in the population. A paper by White et al.…

What are cure fraction models?

Many people are familiar with survival models. Survival models measure the probability of survival to a given time period. The “problem” addressed by these models is that some people are “censored”, in other words, the do not die in the sample time period. Although longer survival is good in practice, for statisticians it is problematic…

How much should you bet?

This is an interesting question to ask.  If you are going to the casino, in most cases, the answer is $0.  The odds are stacked against you.  But what if the odds are in your favor, or you believe that your own predicted probability of winning differs from that of the bet? The easy answer…