Dartmouth Researcher Challenges AMA Data

In the wake of healthcare reform, data is becoming more important than ever in issues such as the need for primary care physicians.

The findings of a recent Dartmouth Institute for Health Policy and Clinical Practice report say mortality and hospitalizations are significantly lower in areas where there are more primary care doctors who work fulltime and are taking care of patients in ambulatory care and office settings.

Looking past those general findings, however, reveals other issues. The report also takes aim at the American Medical Association's database of physicians and questions the accuracy of its findings, and its breakdown of primary care physicians in certain areas, the co-author told me. Ironically, the Dartmouth study is published in the Journal of the American Medical Association.

"The (data) that the AMA is using to estimate the number of primary care doctors is not accurate," say Chiang-Hua Chang, MS, PhD, an instructor in Dartmouth Institute for Health Policy and Clinical Practice. "What the AMA is using does not accurately reflect what the doctors are doing." What the listed primary care doctors are actually doing, says Chang, is "going into subspecialties."

Essentially, "primary care doctors listed on AMA are not all doing primary care, many of them are actually providing specialty care," she says.

Recognizing the difference is not only important to improve primary care clinician measurement, but also reflects the drift of physicians into nonprimary care careers. The manner in which primary care physicians are practicing is important in the healthcare reform era, especially with the established great need for primary care physicians. As Chiang-Hua Chang noted, the training capacity of family medicine and internal medicine may have "disappointing patient benefits if the resulting physicians are primary care in name only."

Currently many clinicians licensed and counted as primary care doctors actually work in hospitals, emergency departments, in research, or in public health, or may not take care of Medicare beneficiaries at all, Chang says. In some cases, they have gone on to specialize in other fields such as cardiology.

"In healthcare, we are trying to figure out how many primary care doctors are really in the U.S.," Chang explained. "We don't have a good number; based on estimates that are inaccurate. It may be more certain in some areas, than in others."

The number of U.S. medical school students going into primary care has dropped 51.8% since 1997, according to the American Academy of Family Physicians, which predicts a shortage of 40,000 family physicians in 2020.

Chang explains a key element in evaluating the number of physicians in which there is a projected "large shortage of general internists and family physician to care for a growing number of elderly patients."

The Dartmouth study offers a "cautionary note" that having more physicians trained in primary care in an area, by itself "does not ensure substantially lower mortality, fewer hospitalizations or lower costs."

That other cautionary tale in Chang's report involves the AMA data itself. As Chang and co-authors note in the report, "Despite a widespread interest in increasing the numbers of primary care physicians to improve care and moderate costs, the relationship of the primary care physician workforce to patient level outcomes remains poorly understood."

At issue is the AMA's data Masterfile Dataset of doctors who list themselves as primary care providers. The AMA data is the "most commonly-used national measure of primary care physician workforce," as Chang notes.

"The AMA Masterfile misclassifies some specialist physicians, such as hospitalists and emergency department physicians, as general internists and family physicians," according to the Dartmouth report.

Sources for the AMA dispute some of Chang's findings, saying that the association relies on many databases for information. However, the AMA did not immediately respond for comment on the Dartmouth report.

As far as Chang is concerned, research and healthcare planning that relies on the AMA Masterfile "will not accurately measure the "primary care workforce."

 "Many of the physicians who would be classified as providing primary care by the AMA Masterfile were either not providing care to fee for service Medicare beneficiaries or were providing nonambulatory or specialty care."

 "Similarly," it adds, "a significant proportion of the physicians delivering primary care services as per Medicare were not classified as office-based primary care physicians in the AMA Masterfile."

In driving home its point in the study, the researchers decided then not to rely solely on the AMA Masterfile because of questions of its accuracy, Chang says. As part of their review, they also used clinical primary care FTEs derived from Medicare claims for beneficiaries as a "secondary measure of primary care physician workforce."

The researchers based their findings on a sample of more than 5 million Medicare beneficiaries in in 2007 in 6,542 primary care service areas and their hospitalization claims for 12 conditions that can be avoided when good care is provided within a clinic or office visit.

Chang says she believed it was important for her study to verify exactly where funds were allocated related to primary care services, particularly in outpatient settings.

"If they allocate and put the resources only to train primary care that may not be much help," she said.