by Karla Gale

Last Updated: 2008-01-29 17:19:19 -0400 (Reuters Health)

The Centers for Medicaid and Medicare Services (CMS) often make national coverage determinations based on clinical trials among subjects whose demographics differ substantially from those of Medicare beneficiaries, according to researchers at the University of California at San Francisco School of Medicine.

As noted in the January 28 Archives of Internal Medicine, CMS policy is guided by a Medicare Evidence Development and Coverage Advisory Committee (MedCAC) that reviews technology assessments prepared for new medical products or services.

"I was appointed to the MedCAC in 2002, and my first meeting was called because Guidant had requested expansion of defibrillator coverage in the Medicare population," co-author Dr. Rita F. Redberg told Reuters Health. "In reviewing the technology assessment, it was obvious that the data were based on a population that was different than the population we were being asked if defibrillators would provide benefit for.

"It was difficult to make that extrapolation, 'how would the Medicare population do based on a study in middle-aged healthy men?'" she added. "It made me think that this was not an isolated incident."

To examine the relevance of data typically reviewed for the CMS, Dr. Redberg and Sanket S. Dhruva performed a meta-analysis of trials included in cardiovascular technology assessments considered by the MedCAC between 1998 and 2006.

Included were 141 peer-reviewed publications with data on 40,000 individuals. Topics investigated were noninvasive diagnostic imaging, behavioral interventions, percutaneous and transmyocardial revascularization, ventricular assist devices, cardioverter defibrillators, and ambulatory blood pressure monitoring.

Compared with Medicare beneficiaries, clinical trial cohorts tended to be younger (average age 60.1 versus 70.8 years) and mostly male (75.4% versus 43.7%), the authors report. Fewer than half of the trials were located in the US, and only seven reported race.

Many trials specifically excluded older patients or women, and most excluded patients with comorbidities such as kidney disease, low ejection fraction, and diabetes.

"The clinical trials primarily relied on to inform national coverage decisions simply do not reflect the Medicare patient population," the investigators write. To rectify the under-representation of Medicare beneficiaries, clinical trials must enroll more women and elderly individuals, they say.

Noting that these changes will "take a while," Dr. Redberg said, "what could be done immediately is publishing subgroup analysis by age and sex, either in journals or online. Then one could do a meta-analysis and see, are the results the same for women and men? Are they the same for people over 65 and under 65? Making that data available even without doing anything else would be a step forward.

"To be good fiscal stewards of our healthcare dollars, particularly for public programs like Medicare, we have to have good data about what we're covering," she asserted.

Arch Intern Med 2008;168:136-140.

Copyright Reuters 2008. Click for Restrictions