In the continued vein of how greater availability of bigger data is shaping our daily lives, I was struck by the news of the release of data showing hospital billing rates to Medicare. While most health policy researchers had a sense that disparities in billing amounts for the same services exist, quantifying these stark differences sharpens our understanding of billing practices. Some hospitals are billing up to 4 times the national average for same services. You don’t have to be a data scientist to know that 4 times higher than the national average is a notable disparity.
In other news, the IRS 501(c)(4) application scandal can’t escape a bit of data-driven commentary. Nate Silver responds to part 2 of the scandal, that conservatives were singled out for individual and business audits, with evidence of how many Obama and Romney voters were likely audited last year by the IRS (480,000 and 380,000, respectively). His overall point is worth repeating – that a few anecdotal stories do not lead to reliable conclusions from big datasets. In this case, we can’t make conclusions on a few stories from a “dataset” of 1.5 million audits a year. As he states, “a handful of anecdotal data points are not worth very much in a country of more than 300 million people.”