BOSTON, Mass. (January 8, 2008)Many recent studies have found that Medicare spending across the country varies greatly. But despite these spending differences, aggregate health outcomes tend to be the same no matter which region a person lives in. Because of this, some policy makers have determined that there is no value to the excess costs in high-spending areas.
A new study that focuses on colorectal cancer as a model suggests this is not correct. While it finds that patients in low spending areas ultimately fare just as well those in high spending areas, the authors find that all care is not alike.
In certain cases the increased spending *is* beneficial, says Harvard Medical School professor of health care policy Mary Beth Landrum, lead author on the study that will be published in the January/February issue of Health Affairs. The focus should not simply be on cost containment, but rather on targeting care to the patients who we know will benefit.
For this study, Landrum and colleagues Nancy Keating and Ellen Meara, also Harvard Medical School faculty, looked at a cohort of 55,549 patients, who were all diagnosed with colorectal cancer between 1992 and 1996, and who were all over age sixty-five and enrolled in Medicare. These patients lived in various locations throughout the US, in high-spending Medicare areas like Los Angeles and Detroit, low-spending areas such as Iowa, Seattle, and Utah, and more moderate spending areas like San Francisco and Connecticut. The authors analyzed various aspects of their cancer care, including mortality rates three years post diagnosis.
Although increased spending did not yield improved patient outcomes overall, the authors found that not all increased spending was necessarily wasteful.
For example, chemotherapy for patients in stage 3 colorectal cancer is very helpful, and people in these high-spending areas receive it and greatly benefit from it, says Landrum. But
'/>"/>
| Contact: David Cameron david_cameron@hms.harvard.edu 617-432-0441 Harvard Medical School Source:Eurekalert |