Wednesday, November 08, 2006

A Lecture about Pay-for-Performance: Leaving the Reader Guessing About Conflicts of Interest

We have discussed before how the band-wagon for "pay-for-performance" has been rolling on inexorably, even though applying this seemingly logical mantra to physicians' practices is fraught with problems. (See our posts here and here.) Potential problems with proposed P4P programs include failure to adjust outcome measures for differences in patient characteristics, leading to perverse incentives; evaluating physicians on processes not under their direct control; using error-filled administrative data; emphasizing performance that saves money rather than improves patients' outcomes; and using measures that apply to only a fraction of medical problems and only certain specialties.

Nonetheless, practicing physicians continue to be subject to harangues that P4P is inevitable, so they might as well just submit to it. Two of my favorite blogs, DB's MedicalRants and Retired Doc's Thoughts, have recently been ranting (here and here) about the latest pitches for "pay-for-performance" (P4P) to appear in respected medical journals.

MedicalRants called these pitches "confusing." That may be an understatement.

Take in particular the article written by one John W Rowe MD, which was derived from the Harvey Kimball lecture Dr Rowe gave to the 2005 American Board of Internal Medicine Summer Conference. [Rowe JW. Pay-for-performance and accountability: related thems in improving health care. Ann Intern Med 2006; 145: 695-699.] Dr Rowe started by saying,

When I started my career in the late 1970s as an academic internist and geriatrician, I was skeptical of pay-for-performance, feeling that standards of care could not accurately assess the real benefits of my care of my frail elderly patients with multiple impairments. I still feel that way today because despite the rapid growth of the evidence base, we continue to measure relatively simple aspects of the process of care rather than measuring outcomes.

Dr Rowe later noted that current P4P metrics cannot evaluate management of chronic diseases or multiple comorbid conditions. To do so, he said,

We will need a much richer evidence base than is currently available.

Furthermore, he noted

Although most current standards are developed to evaluate primary care, specialty care accounts for a disproportionate share of health care costs.
So,

Standards must extend to specialty care and to complex patients with multisystem problems.

But despite acknowledging these problems, he still asserted,

I feel a sense of urgency.

So that P4P "now seems to be a useful strategy to pursue."

So who is Dr Rowe, and why should we heed his call for pay-for-performance, even though even he admits that current approaches have major deficiencies?

Trying to answer that question by simply reading the article turns out to illustrate how inadequate current policies on disclosure of conflict of interest are to cope with our current health care environment, dominated as it is by large health care organizations.

The Annals identifies Dr Rowe as being "from Columbia University, New York, New York." His address is listed as the Mailman School of Public Health there. Thus, his current position seems to be on the faculty of a well-known school of public health.

However, this appearance may be deceiving. See the addendum below for evidence that Dr Rowe is not on the faculty of the Mailman School, but rather arranged a donation to the School from his family foundation, and is currently on the School's Board of Overseers.

Dr Rowe did mention that he has "exprience as the leader of a large academic health science center and as chief executive officer of a major health insurer." The Annals lists his "potential financial conflicts of interest" as "employment: Aetna Inc." and "stock ownership or options (other than mutual funds): Aetna Inc." Thus, is he also a part-time employee of Aetna Inc., a large health insurer and commerical managed care organization, perhaps a part-time medical director, or analyst, who holds a few shares of stock in a retirement plan?

Worldly physicians, but not all Annals readers, may realize, instead, that Dr Rowe is not just a part-time mid-level employee, but is the former chief executive officer (CEO) of Aetna Inc. According to Aetna's most recent proxy statement, Dr Rowe retired as CEO recently, in Feb 14, 2006. However, the statement also reveals he still has a very important position with Aetna. He was has been Chairman of the Board since April 1, 2001, and still holds that position, although he plans to retire at the end of this year. In 2006, his salary was $1.1 million, his bonus was $2 million, his other annual compensation was slightly more than $202,000. In addition, in 2006, he received 911,904 stock options, and long-term compensation of over $4.5 million.

Furthermore, Dr Rowe owns far more than a few shares of Aetna stock. According to the proxy statement, he owns or controls the equivalent of 6,338,393 shares of Aetna stock, just under 1% of the total shares outstanding outstanding.

Thus, the magnitude and intensity of Dr Rowe's relationships with Aetna were hardly revealed by the few words of disclosure provided by the Annals of Internal Medicine, under the journal's current disclosure policy.

If the Annals, or any other medical journal for that matter, wants to let the former CEO, current Chairman, and major stock-holder of a large commerical managed care organization lecture us on pay-for-performance, a topic clearly related to the company's vested interests, that is the journal's right. However, the journal should clearly explain who the authors of its articles are.

Moreover, in my humble opinion, the journal should clearly explain the extent, nature, and magnitude of the financial interests of its authors that could relate to the topics they write about. Failure to distinguish between low-level employment and running the company, or between holding a few shares of stock in a retirement plan and owning 1% of a large company is misleading.

If disclosure is going to be an adequate solution to the conflict of interest issues now pervading health care, it must be complete enough so that readers can judge how much bias such conflicts may cause.

ADDENDUM (11/9/2006) When I wrote the post above, I assumed that Dr Rowe did have an academic appointment at Columbia, given the affiliation and address listed in the Annals of Internal Medicine article. However, as bast as I can tell, this is not so.
Searches of the Columbia University and the Mailman School of Public Health directories did not reveal a listing for him. The Mailman School's faculty list did not include his name. Google searching did show that Dr Rowe is on the Board of Overseers for the Mailman School. In addition, his family foundation has given $500,000 to the school to support the Center for History and Ethics in Public Health.
Thus, Dr Rowe appears not to have any current academic appointment at Columbia University or specifically at the Mailman School of Public Health.
The information given in the article about Dr Rowe's primary affiliation and address now also appears to be misleading.
In my humble opinion, medical and health care journals should list author affiliations and addresses that convey an accurate idea of authors' primary organizational affiliations.

1 comment:

Matthew Holt said...

Roy--that one is too easy for you! I think Annals the their ilk need to set up something more challenging next time!