What is Mahalanobis distance? Most people know what Euclidean distance is…it is the shortest distance between any two points. In other words, its what we typically think of when we think of distance – the distance we would measure with a ruler, and the one given by the Pythagorean formula. Unlike Euclidean distance, Mahalanobis distance [...]
Read the rest of this entry »How Missing Data affects Physicians’ P4P Bonuses
Pay-for-performance programs often offer bonuses (or penalties) for physicians, hospitals and other providers based on the quality of care patients receive. Measuring quality of care, however, is often difficult. For chronic conditions, for instance, many patients eligible for outcome measures may be lost to follow-up. This issue can potentially affect provider evaluations and bonus payments. [...]
Read the rest of this entry »CMS Chronic Conditions Dashboard
CMS has just released a new interactive tool that allows users to examine chronic conditions among Medicare beneficiaries. The CMS Chronic Conditions Dashboard presents statistical views of information on the prevalence, utilization and Medicare spending for Medicare beneficiaries with chronic conditions. The Dashboard displays information on a set of predefined chronic conditions available in the [...]
Read the rest of this entry »2011 National Health Expenditures
According to research from the CMS Office of the Actuary, healthcare spending growth is decelerating. As published in Health Affairs: In 2011 US health care spending grew 3.9 percent to reach $2.7 trillion, marking the third consecutive year of relatively slow growth. Growth in national health spending closely tracked growth in nominal gross domestic product (GDP) in 2010 [...]
Read the rest of this entry »Add to Your Skills Toolkit: Bootstrapping Confidence Intervals
In previous posts, I have explained how to create bootstrap estimates for a variety of statistics. Doing so is fairly simple and involves a 3 step procedure: Step 1: Using the observe data, create m boostrap data sets by using random resampling with replacement. Step 2: Calculate the statistic of interest for each bootstrap data [...]
Read the rest of this entry »Statistics from the Medicare Trustees Report
Basic Statistics In 2010, 47.5 million people were covered by Medicare: 39.6 million aged 65 and older, and 7.9 million disabled. About 25 percent of beneficiaries have chosen to enroll in Part C private health plans that contract with Medicare to provide Part A and Part B health services. Total benefits paid in 2010 were [...]
Read the rest of this entry »Biases
All economists are familiar with the problem of selection bias. In non-randomized samples, patients may choose to be in either the treatment or control group based on factors which are also related to the outcome of interest. Even if researchers can design a study that fully controls for selection bias, robust studies must also account [...]
Read the rest of this entry »Medicare Dashboards
For which drugs does Medicare spend the most money? For which inpatient hospital treatments does Medicare have the highest expenses. CMS’s new Dashboards provide an easy to use source to access these high level summary statistics. You can find this information here: Medicare Inpatient Hospital Dashboard (website, description) Medicare Prescription Drug Benefit Dashboard (website, description) [...]
Read the rest of this entry »Share of Federal Budget Spent on Health Care Jumped in 2009
In 2008, 38 percent of the federal government’s revenue was spent on health care. In 2009, however, this figure jumped to 54 percent of total revenues. Although federal health spending only increased by 17.9%, a decline in revenues of a similar magnitude caused this large change. Surprisingly, state and local spending on healthcare barely budged. In [...]
Read the rest of this entry »Econometric Methods for Fractional Response Variables
Oftentimes, people use the following rule of thumb: if the dependent variable is continuous, use OLS; if binary use a logit or probit. But what should you do if your dependent variable is fraction between 0 and 1. To use a logit or probit one would have to unnecessarily transform the dependent variable into binary [...]
Read the rest of this entry »