Medicare Article Analysis

Solution Preview

1. Analyze data analysis to determine whether the appropriate statistical procedures were used to analyze data.

This review article did not use any statistical procedures, they only cited the results of others. This article was basically a summary of an article in the same issue of JAMA, by McWilliams et al.

I looked at that article, and found that they looked at "Household survey data from the nationally representative Health and Retirement Study were used to analyze differences in receipt of basic clinical services by adults in 1996 and 2000, before and after becoming eligible for Medicare at age 65 years." This means this was a longitudinal study.

The sample size was 2203. Before they became eligible for Medicare, they fell into 3 groups: continuously uninsured (n = 167), intermittently uninsured (n = 216), or continuously insured (n = 1820).

The variables used were "reports of receiving cholesterol testing, mammography (in women), prostate examination (in men), and treatment of arthritis and hypertension in the prior 2 years" in both 1996 and 2000.

In the statistical analysis, the authors looked at the difference in whether an individual received treatment in 1996 and 2000 (before and after Medicare). They did this for each of the variables (the different kinds of testing and treatment mentioned above). Their description is:

Differences between insurance status groups before and after Medicare eligibility and the differences between these differences were estimated with 95% confidence intervals (CIs). For example, to calculate the difference between differences in cholesterol testing for continuously insured and uninsured adults, variable uk was defined for each respondent, k, according to the following formula: uk = ...