As I was driving back from a very nice long week-end with my extended family (fireworks, festivals, baseball, wings, and art in a gritty urban setting) in metro Atlanta, I heard an NPR story on “the July effect.”  This effect is a suspicion (now with some evidence behind it) that health care in teaching hospitals is worse in July because of the inexperience of the new learners (or as I tell my residents “You are moving from a very experienced physicians at one level to a very inexperienced physician at the next level”). The report cites an increase in deaths in counties with teaching hospitals and  lack of a similar effect in counties with non-teaching hospitals in July as evidence for this effect. It only finds the effect with medical errors. I am grateful to Dr Carol Motley who traded calls with me so I could be with my family and she could work with the newly promoted physicians on this worst weekend of the worst month of the year.

From experience I will agree that the learning curve for a newly promoted physician is steep. I unfortunately know of no better method of training physicians. The author of the study in his interview cited a lack of surgical effect as evidence that surgical training is superior. I would argue that this suggests a certain randomness to his findings and we probably need to look more closely before discarding the entire training process. It does point out the need for close supervision of neophyte learners and the importance of good processes coupled with an assessment of outcomes to determine if the desired effect is being achieved.

It also identifies a need for continuation of an extensively supervised period of learning prior to neophyte physicians being transitioned into the “real world”. This process (known as residency training) is labor intensive if done correctly. It involves ongoing assessment of the learners progress towards achieving  six types of competence which the prototypical physician is expected to demonstrate in practice. We do this in part through close supervision of the learner in hopes of detecting potential errors before they are made. We also do this by collating thousands of individual observations on each of our learners and using them to assess progress towards achieving these competencies. This process is time consuming and expensive.

The current way of paying for this instruction is for the payor (usually Medicare)  to give money to the hospitals and hope that they pass this money on to those of us providing the instruction. The hospitals tend to see the cost of training residents as including a lot of costs not involving direct supervision and assessment residents by attending physicians (in part because we can bill for the service which covers a small part of the total cost). Consequently, we don’t see much of that money. Perhaps if more people are aware of the “July problem” the allocation of money to pay  for supervision of neophyte residents will be seen as important.