Unbiased Analysis of Today's Healthcare Issues

Addressing Type S vs. Type M errors with Bayesian Hierarchical Modeling

In statistics, most statistical tests aim to trade off type I and type II errors.  Type I error is the incorrect rejection of a true null hypothesis, in other words a false positive.  Type II errors are the incorrect retaining a false null hypothesis; in other words, a false negative.  Oftentimes, the null hypothesis is posed […]

Read the rest of this entry »

Bayes vs. Fisher

An interesting People of Science video compares the approaches of two titans of statistics, Ronald A. Fisher and Thomas Bayes.

Read the rest of this entry »

Cancer survival around the world

An interesting study measuring trends in cancer survival between 2000 and 2014 found, unsurprisingly, that patients in more developed countries had better survival. For women diagnosed with breast cancer between 2010 and 2014, 5-year survival rates reached 89.5% in Australia and 90.2% in the United States, but generally varied worldwide and remained low in some […]

Read the rest of this entry »

Which cancer treatment is best?

This seems like a straightforward question, but clearly depends on what you mean by “best”.  Some drugs will be more efficacious and have more adverse events; other drugs may be less efficacious but have fewer adverse events.  What if a one drug shows an 80% improvement in progression free survival (PFS), but a 50% improvement in overall […]

Read the rest of this entry »

Cancer deaths are rising…is that a good thing?

A JAMA Oncology paper estimating the global burden of cancer is getting a lot of attention in the press.  The study’s key findings are: In 2015, there were 17.5 million cancer cases worldwide and 8.7 million deaths. Between 2005 and 2015, cancer cases increased by 33% A 33% increase in cancer cases!!! There must be an […]

Read the rest of this entry »

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 […]

Read the rest of this entry »

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 […]

Read the rest of this entry »

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.

Read the rest of this entry »

Trends in Life Expectancy among Older Americans, by Race

It appear that most of the gains in longevity and reductions in disability among elderly Americans accrued to Caucasians. Using data from the 1982 and 2004 National Long Term Care Surveys and the 2011 National Health and Aging Trends Study, Freedman and Spillman (2016) find the following: We examine changes in active life expectancy in […]

Read the rest of this entry »

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, […]

Read the rest of this entry »