post by Bill Gardner
Many Americans are, rightly, dissatisfied with the quality of their care. Kelvin said that you can't change what you can't measure, so it is essential to develop standards for good care. This isn't always a straightforward task.
I was at a meeting yesterday that discussed candidate quality of care measures for adolescent health. We discussed whether screening for human papillomavirus (HPV, see the transmission electron micrograph to the right) in adolescent girls should be a quality of care standard. HPV infection, as you probably know, can cause cervical cancer. Making this a standard would mean that primary care doctors would be evaluated on the proportion of adolescent girls in their practice who have been screened for HPV. Everyone on the panel agreed that screening adolescent girls for HPV should be a quality of care standard.
The interesting question was whether boys should be screened as well. It seemed unjust to have girls only be screened, given that boys are also likely to carry the virus. However, HPV is asymptomatic for most men. Although men can get cancer of the penis and anus from HPV, these are rare, so for the most part men are not harmed by HPV. So a doctor asking a boy or his parent for consent to carry out an HPV screen is mostly doing this on behalf of his future female sexual partners.
So, has the doctor failed to give her patient good care if the boy refuses a procedure that is unlikely to benefit him? Quality measures are often used to incentivize doctors in pay-for-performance schemes. Do you want to incentivize the doctor to get her male patients to act on behalf of others? You could make the quality parameter be offering the screen, not completing it. This is slightly odd, in that no one benefits from the offer per se, and two doctors with identical offer rates but disparate completion rates would receive the same incentives. Does it make sense to have doctors incentivized to get girls to complete the screen, but incentivized to just offer it to boys? Measurement itself has a cost, to the doctor and to the health plan. Is just offering an HPV test sufficiently important to measure?
In the end, we decided to just leave screening for HPV as a quality standard for treatment of female patients only. No one felt fully comfortable with the decision. Interestingly, I didn't note any difference of opinion between the men and the women at the table.
Progress in measuring the quality of health care is slower than many would like. I hope this anecdote helps explain why this is so.