Study outlines deficiencies in American health care

Technology seen as a key ingredient in changing system-based shortcomings.

By Andis Robeznieks, AMNews staff. July 7, 2003.

Suspicions that the use of recommended health care procedures is far from universal were confirmed by a new study. This has led to calls for systematic changes in health care delivery.

American adults, on average, receive only a little more than half the measures recommended for their conditions, said the study published in the June 26 New England Journal of Medicine. The study concluded there are serious threats to the well-being of the American public because of "the gap between what we know works and what is actually done."

With this article

See related content

"My greatest hope is that we can stop debating whether we have a problem and start working on solutions," said the study's lead author, Elizabeth A. McGlynn, PhD, associate director of Santa Monica, Calif.-based Rand Health, the nation's largest independent health policy research organization.

Dr. McGlynn said the study was the largest of its kind ever conducted and the most comprehensive in scale. It included participants from across the nation with a wide range of both health conditions and insurance coverage.

Rand researchers interviewed and reviewed the health care records of nearly 7,000 adults in 12 metropolitan areas and measured 439 indicators for 30 acute and chronic conditions. Quality indicators were chosen by Rand physicians after a review of established and proposed national quality guidelines. These indicators were then approved by four nine-person panels whose members were nominated by appropriate specialty societies.

Those surveyed were found to have received only 54.9% of recommended health care measures. This included: 54.9% of preventive care measures, 53.5% of acute care measures and 56.1% of the care recommended for chronic conditions.

Only 55% of those surveyed received recommended health care measures.

Shortcomings that were highlighted included people with hypertension getting 64.7% of recommended care and 24% of diabetic participants getting the recommended three or more glycosylated hemoglobin tests over two years to assess treatment and identify early complications.

Dr. McGlynn speculated that the reason for the shortcoming is at least partly that medicine is still basically practiced the same way it was 100 years ago.

She added that the model includes a doctor with handwritten notes on a paper chart (which may not have information from other doctors) trying to figure out from memory what is needed to treat and evaluate a patient -- during a 17-minute visit.

"The nature of that intervention hasn't changed a lot, but the nature of what patients need has," Dr. McGlynn said. "We're asking doctors to behave like supercomputers, asking them to act quickly without all the needed information in front of them."

In some ways, she added, the study results are a "remarkable tribute" to the dedication of physicians who must pull together disparate elements of a fragmented health care system to treat their patients.

Charles M. Kilo, MD, president of the Portland, Ore.-based GreenField Health System and a fellow at the Institute for Healthcare Improvement, agreed that the study results had confirmed observers' concerns but said that there were also many reasons to be positive.

People with hypertension get only 65% of recommended care.

"Despite 10 years or more of focus on quality improvement, there's little documented information that we've improved so far," Dr. Kilo said. "But we now have both the content knowledge of what we need to do and the technology to make it possible."

He explained that efforts to improve quality are hampered by barriers of leadership, culture, structure and financing, but said all of these could be overcome.

"I'm less pessimistic about these barriers than I am optimistic about where we're going," Dr. Kilo said. "But I'm also realistic and not surprised by what the data show."

Primary care deserves attention

James Mold, MD, a director for the Oklahoma City-based Oklahoma Center for Family Medicine Research, said that although he has problems with the study's assumption that all people of a certain age or condition require the same interventions, the results are what he would have predicted and are worthy of further study.

"Primary care practices currently do not have the systems in place to make sure that all the effective treatment options are at least considered for every potentially eligible patient," Dr. Mold said. "However, until the health care system moves from a disease-oriented model to a person-centered, goal-directed one, it will be impossible to do any better."

Dr. Kilo called for better use of information technology and a different way of financing health care so that primary care gets the attention it deserves. "The primary care we have today is not the primary care we need," he said, adding that it is impractical to think that primary care physicians can properly address the health needs of the 2,000 to 3,000 people they see each year in short office visits.

"Primary care is continually devalued," he said. "As long as that continues to be the case, we will not solve either the cost or the quality problems."

Dr. McGlynn suggests that some type of public subsidy might be needed to get doctors plugged into the electronic information systems that can help them do a better job.

"We definitely have to do something different to get different results. Something has to change, and I think the starting point is information systems. That's not the silver bullet, but it's something that's required."

PROFESSIONAL ISSUES