This very interesting paper by John Wennberg, Doug Staiger et al develops a new approach for calibrating risk adjustment models so as to not over adjust for the higher intensity of coding that results when there are more visits. In short, in markets where doctors do more visits, they will also tend to code more diagnoses. Their approach takes into account both visits and diagnoses to improve model fit. There are many further questions one could ask, but this is destined to be an influential paper.

• Wennberg JE, Staiger DO, Sharp SM, Gottlieb DJ, Bevan G, McPherson K, Welch HG. Observational intensity bias associated with illness adjustment: cross sectional analysis of insurance claims. BMJ 2013;346:f549.
  [Supplemental table]

http://www.bmj.com/highwire/filestream/632298/field_highwire_article_pdf/0/bmj.f549