Only further study will determine whether an anti-malarial drug also used to treat rheumatoid arthritis can prevent or delay the onset of diabetes.

But a study led by the University of Pittsburgh School of Medicine shows early promise.

Published today in the Journal of the American Medical Association, the study shows that hydroxychloroquine (HCQ), sold under the name of Plaquenil, is associated with a progressive decline in the risk of diabetes as patients use the drug for ever-increasing periods.

For those using it for more than four years, the risk of diabetes plummeted by 77 percent, compared with those not using the drug.

The results suggest that HCQ may help to delay the onset or even prevent diabetes in people at high risk for the disease. Although the data could not determine whether the type of diabetes involved was type 1 or type 2, the age of participants would suggest an impact on type 2 diabetes, said Dr. Mary Chester M. Wasko, a rheumatologist and professor of medicine at the University of Pittsburgh School of Medicine and the study's lead author.

People with diabetes have higher than normal levels of glucose in their blood due to problems in producing or using insulin -- the hormone that allows glucose to enter cells and be used as energy.

Stanford University, the University of Cincinnati and the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health participated in the study.

Dr. Wasko said the reduction in diabetes risk in people using HCQ persisted even after researchers adjusted for such other risk factors as body-mass index, degree of disability and use of corticosteroids, which can cause weight gain.

Because people with rheumatoid arthritis tend to be less active and take corticosteroids that can cause weight gain, they often are considered to be at higher risk to develop diabetes.

"Another interesting finding was that the rheumatoid arthritis patients who developed diabetes were less likely to need blood sugar-lowering medication to manage their disease," Dr. Wasko said of patients on HCQ.

Based on study results, she said, HCQ's association with the reduction in diabetes risk was comparable or superior to that of other drugs studied in clinical trials for diabetes prevention and treatment.

Dr. Wasko said she took care of a patient years ago who had symptoms of low blood sugar at night, but never had daytime symptoms.

The low blood sugar levels were thought to be the result of HCQ, which was used briefly in the early 1990s as a type 2 treatment.

In the study, Dr. Wasko and other researchers reviewed questionnaires that 4,905 adults, including 491 from the Pittsburgh area, completed every six months for a period of up to 20 years.

The adults all had rheumatoid arthritis, but may or may not have been treated with HCQ. The study discovered that a larger percentage of people on HCQ remained free of diabetes.

"The mechanism by which this drug may help glucose metabolism is not clear," she said, noting plans to discuss a strategy for follow-up research with endocrinologists at the school of medicine and the University of Pittsburgh Graduate School of Public Health.

In a follow-up study, she said, she would like to track lab testing and diabetes-specific blood measures of people on Plaquenil and patients with rheumatoid arthritis.

If further research confirms the findings, HCQ most likely would be used as a preventative agent rather than as a mainstay treatment for type 2, she said.

Further testing will determine whether it could be used effectively in people at high risk for diabetes, including those who are obese and sedentary with a family history of the disease.

Considered safe, Dr. Wasko said, HCQ can cause nausea, headaches and dizziness among other side effects.

"If the drug is offered early in the evolutionary course, it may have the ability to prevent or delay the onset of diabetes," she said. "This certainly is my hope."