## Dealing with time-censored cost data

We health economists deal with medical cost data all the time.  One challenge we all face is that the medical cost data is often censored.  The censoring may occur because the patient dies.  If you are using administrative health insurance claims data, censoring may occur because people switch their health plan and leave your sample.…

## 2SLS vs. 2SRI: Which to use with bivariate outcomes?

IV methods are usually implemented using a two‐stage approach where the first stage estimates an expectation of the endogenous variable conditional on measured confounders and one or more IV. The second stage model then predicts outcomes as a function of the estimated treatment values from the first stage, measured confounders, and potentially other control variables.…

## The problem with odds ratios

Many researchers use logit models to estimate the effect of specific variables on a binary (i.e., 0 or 1) outcome.  How are these models derived?  How are odds ratios calculated?  What are the problems with odds ratios?  I answer all these questions in this post, following a lovely summary by Norton and Dowd (2018). Deriving…

## Which inflation index should I use?

Many studies use data on health care costs from multiple time periods.  To make costs comparable over time, researchers often use an inflation index to translate previous years costs to current dollars.  The first question is, what inflation indices are available to make this adjustment.  A paper by Dunn et al. (2018) reviews the potential…

## 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…

## Bayes vs. Fisher

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

## Collecting data on health state utilities

If you had a given disease, how bad would it be?  Can we quantify people’s happiness based on different diseases?  One way to do this is to measure patient preferences using a concept known as health state utilities.  As stated in a recent ISPOR guidelines, health state utilities are “estimates of the preference for a given…

## Longitudinal Modelling of Healthcare Expenditures: Challenges and Solutions

Previous analyses–such as Basu and Manning 2009–have addressed the problem of mass of health care expenditures around \$0. In typical economic analyses, we assume that the dependent variable is normally distributed. In the case of health care expenditures, however, a large number of people have \$0 expenditures (i.e., healthy individuals). Further, among sick individuals that…

## The gold standard of scientific evidence

That is the title of my latest article in Pharmaceutical Market Europe. An excerpt is below. Randomised controlled trials (RCTs) are regarded as the gold standard of scientific evidence, and for good reason. By randomizsing a treatment across study arms, RCTs eliminate patient-treamtent selection bias, resulting in reliable causal inference. In contrast, in the real…

## Interesting paper measuring the option value

One key benefit new cancer treatments is not only that they improve survival, but also–in areas with a lot of treatment innovation–they that they allow some patients to live to receive the next treatment advance.  Although this concept may make sense intuitively, it is not clear how one could quantify this value. This is the…