Search :
Spring 2000

Research Magazine > ARCHIVE > Summer 99 > Article

Learning from Medicaid
by Jennifer T. Daly

For many, the thought of pharmaceutical research conjures up images of test tubes and microscopes. For Jeffrey Kotzan, it's a matter of millions of bits of computer data analyzed on a mainframe in the UGA College of Pharmacy.

Kotzan, a professor in UGA's pharmacy care administration program, studies statistics gleaned from anonymous claim records of Georgia's 1.24 million Medicaid recipients. His studies of pharmaceutical use in the Medicaid system provides the Department of Medical Assistance - the government agency that oversees Medicaid funding in Georgia - with information crucial to making informed policy decisions. The research also helps grant makers, most often pharmaceutical companies, better understand the effectiveness of certain drugs and drug policies in the Medicaid marketplace.

For example, Kotzan is studying the use of nonsterodial anti-inflammatory drugs (NSAIDS), which are prescribed regularly for patients with arthritis or other joint problems. In a small percentage of patients, NSAIDS have the unwelcome side effect of causing gastrointestinal bleeding. However, new drugs called "Cox II inhibitors" claim to carry a reduced risk of GI bleeding, but are more expensive.

The question from a Medicaid standpoint is: What is cheaper - paying for the new drugs or the side effects that are sometimes generated by the older medications? Pharmaceutical companies would like to know if these new drugs have a home in the Medicaid marketplace. From a patient standpoint, the question is simply: Which drug is most likely to be safe and effective?

To find the answers, Kotzan first searches UGA's Medicaid database for cases of gastrointestinal bleeding of any kind in the past three years. He then searches for NSAIDS prescriptions among the more than 30 million Medicaid prescriptions filed during the same period.

Early results from the comparison of the data confirm that patients on NSAIDS do have a higher risk for GI bleeding - a 71 percent greater risk.

"This is not something that is totally new, however, it is much more precise from what you see in a lot of the standard literature," Kotzan said.

"That said, even a 71 percent increase is small when you consider that our data also shows very few people are at risk for developing a bleed in the first place," he said. "So, if I am a young Medicaid recipient who has twisted my knee, should Medicaid pay for me to take a Cox II inhibitor over a less expensive NSAIDS? Probably not. However, for an elderly person who is arthritic and for whom a GI bleed can be catastrophic, the new drug seems a better choice."

Kotzan's colleague, associate professor Bradley Martin, taps the large pool of data in his research on Medicaid patients diagnosed with schizophrenia. He has linked the Medicaid files of more than 16,000 schizophrenic patients with Georgia's 10 state-run psychiatric hospitals, creating a study that spans nearly eight years of data. The resulting research has shown that people who lose Medicaid benefits - a frequent occurrence with patients who have mild cases of schizophrenia - are being admitted to state hospitals at a higher rate than patients who, because of the severity of their illness, are continuously eligible for Medicaid benefits.

This runs counter to the results Martin expected. It turns out that a new, highly effective generation of drug therapies - typically administered on an outpatient basis to Medicaid patients - are effective at managing schizophrenia, even in the most ill patients.

The results of Martin's research will help state policymakers in planning treatment of schizophrenic patients.

"Is it more cost effective to keep schizophrenic patients, no matter what their level of illness, on Medicaid assistance because it is shown that, for people on these medications, hospital utilization goes down? Or is it more cost effective to put them in an inpatient hospital without Medicaid benefits? From the perspective of the state of Georgia, this data will be helpful in determining the most rational way of providing care," he said.

Research projects with the Medicaid database typically take Kotzan and Martin two to four months from start to finish. Kotzan and Martin find that they raise as many questions as they answer - such as Kotzan's curious finding that more people suffer gastrointestinal bleeding in North Georgia than South Georgia. Why?

"No one knows," he said. "One of the amazing things about this style of research is that every time you start doing something, you get to a place where you just don't know where you are."

Sometimes the researchers stumble on an answer they weren't even looking for. Take Martin's schizophrenia research. His primary focus was to study the frequency of hospitalization. However, his data also challenges previous studies that suggested people born in February, March and April were more likely to develop schizophrenia because of prenatal exposure to certain bacteria and viral infections that are more prevalent in those months.

"Ours is one of the largest such studies to date on schizophrenics, and we did not find any statistical evidence to affirm anyone's findings that people born in February, March or April are at greater risk of developing the disease due to bacteria and viral infections," Martin said.

"I was born in March, so maybe my research is biased," he quipped.

E-mail jkotzan@rx.uga.edu for more information.


Return to Summer 1999 Index
Research Communications, Office of the VP for Research, UGA
For comments or for information please e-mail the editor: rcomm@uga.edu
To contact the webmaster please email: ovprweb@uga.edu