Viper's flesh versus sweet almonds? If only it were that easy. Doctors researching diabetes treatments in the 17th century might have studied which treatment worked better: broken red coral or sweet almonds? Or maybe viper's flesh versus the fresh flowers of blind nettles? Some creative types may have even studied combinations of treatments. Unfortunately, these exotic treatments didn't work. For centuries, diabetes was a death sentence. Today this once-rare condition is a chronic disease that can be treated with a wide range of medications. To discover which diabetes medications work best for which patients, researchers supported by the Agency for Healthcare Research and Quality (AHRQ) are doing patient-centered outcomes research, also known as comparative effectiveness research. They are finding out which medications work best in certain circumstances and which ones are more likely to cause weight gain, reduce cholesterol, or cause side effects such as hypoglycemia (low blood sugar).
A complex, common, and costly condition
Type 2 diabetes is a complex metabolic disorder that affects how the body stores and uses sugar from food. In this condition, either the pancreas does not produce enough insulin to metabolize the sugar for energy or the body is unable to recognize insulin and use it properly (insulin resistance). Uncontrolled high blood-sugar levels in diabetes are associated with heart disease, eye and kidney problems, poor circulation, and nerve damage. Patients with diabetes must struggle daily to control their blood-sugar levels with exercise, proper diet, and medication. The disease that was once considered a medical curiosity now affects millions of Americans. According to AHRQ, approximately 19 million U.S. adults reported receiving treatment for diabetes in 2007, more than double the 9 million who said they received care in 1996. Diabetes is also expensive. Outpatient costs doubled to about $10 billion in 2007 from $5 billion in 1996, according to AHRQ. Complex, common, and costly, diabetes remains tough to treat, despite the new classes of medications that have become available and new choices for patients that may include combinations of treatments. Complicating treatment even more, patients with diabetes often suffer from other conditions such as obesity, which further increase their risk for heart attack and stroke, and that clinicians must consider when prescribing medication.
Determining the best treatment for each patient
Clinicians and patients want to know: Which diabetes treatment is best for an individual patient? AHRQ is taking a lead in the quest for answers through its Effective Health Care Program. Initiated in 2005, the Effective Health Care Program provides valid evidence about the comparative effectiveness of different medical interventions. Diabetes is one of 14 priority conditions established by the Secretary of Health and Human Services for the Effective Health Care Program. Since 2007, the Program has generated two comprehensive comparative effectiveness reviews of diabetes treatments. The reviews support systematic appraisals of existing scientific evidence on treatment of many common and chronic conditions. Both reviews about diabetes treatments were prepared by the Johns Hopkins Evidence-based Practice Center, one of the Effective Health Care Program's research partners. These pioneering reviews compare the effectiveness of different types of medications typically used to treat type 2 diabetes. Some medications stimulate the pancreas to produce more insulin, others increase the body's sensitivity to insulin, while others alter the body's metabolism in other ways.
The medications vary in their effect on weight, their reduction of blood-glucose levels, their risk of dropping blood-glucose levels too low (hypoglycemia), and in their side effects—outcomes that are important to patients and their families. They also differ in how often and how they are taken (with or without food, by mouth or injection), how often patients have to check their blood-sugar levels, and how much they cost—factors also important to patients and their families. Patients and their doctors can now rely on this research to decide which of a confusing array of medications is right for each patient. The first review on diabetes, "Comparative Effectiveness and Safety of Oral Diabetes Medications for Adults with Type 2 Diabetes," was published in 2007 and will be updated in 2011. The initial review established that most oral medications prescribed for type 2 diabetes are similarly effective for reducing blood glucose, but the drug metformin is less likely to cause weight gain and may be more likely than other treatments to decrease LDL ("bad") cholesterol.
"It was one of our first research reviews on this important topic," said Jean Slutsky, director of AHRQ's Center for Outcomes and Evidence. "It summarized scientific evidence on the benefits and risks of all approved oral medications commonly used in the U.S. for type 2 diabetes in 2007." The Program's second review, "Comparative Effectiveness, Safety, and Indications of Insulin Analogues in Premixed Formulations for Adults with Type 2 Diabetes," was published in September 2008. According to the review, pre-mixed insulin analogues (genetically altered insulin) are more effective than long-acting analogues for controlling high blood-sugar levels after meals in patients with type 2 diabetes. Yet, conventional pre-mixed insulin appears to be as effective as pre-mixed insulin analogues for lowering blood-sugar levels when patients go 8 or more hours without eating.
Getting medication evidence to clinicians and patients
At AHRQ, every comparative effectiveness review becomes more than a published report. Clinicians access printed and online guides with charts that rate the evidence, faculty slide sets, and continuing education modules. Health organizations, including the American Academy of Nurse Practitioners, have tapped into the Program's diabetes research for their own online continuing education program. Free, plain-language guides on diabetes are printed in both English and Spanish for consumers. "Pills for Diabetes" covers the effectiveness, side effects, and costs of diabetes pills and is the Effective Health Care Program's most requested guide. "Premixed Insulin for Type 2 Diabetes" compares the benefits, side effects, and costs of newer premixed insulin with other kinds of insulin and pills for diabetes. Both guides for consumers, which include price comparison charts, are available at many pharmacy stores, including Safeway.
AHRQ audio spots in stores announce that the brochures are available. In addition, AARP (formerly the American Association of Retired Persons) recently printed the diabetes guides. "We're excited to bring our members and all Americans access to unbiased, evidence-based information about the price and effectiveness of the diabetes drugs they take," Margaret Hawkins, AARP manager of health promotion, told Research Activities. "Giving people clear, concise information about cost, safety, and quality can help them start an informed conversation with their doctor about the best choice for them."
Using the evidence to make medication decisions
The Mayo Clinic's Knowledge and Evaluation Research (KER) Unit is taking a different approach to getting conversations started. "A lot of recent data suggests about 20 percent primary nonadherence related to medications for diabetes," said Nilay Shah, Ph.D., an investigator in the unit. "Our hope is to get a conversation started in the clinical encounter that involves the patient in decisionmaking and hopefully leads to higher adherence." Using AHRQ's comparative effectiveness research on oral medications for adults with type 2 diabetes and funding from the American Diabetes Association, the Center developed a deck of diabetes cards for clinicians and patients to review together. Dr. Shah and his colleagues called the first deck "baseball cards." Each card was devoted to an individual drug and statistical information on concerns, such as effectiveness and side effects.
"We realized these were great," said Dr. Shah, "Everyone liked them, but we wanted to create more of a conversation between the clinician and patient." So, they cut the deck, so to speak. They developed "narrative cards" with written descriptions of each drug. But patients and clinicians still weren't talking enough. Dr. Shah and his colleagues did more research. They "shuffled" the deck and came up with "issue cards" depicted here. (Select for an image of the Mayo Clinic issue cards). Each card deals with a different medication issue or concern for patients. The initial six issue cards covered patient's concerns in plain language: weight change, low blood sugar, improved blood-sugar control, side effects, daily routine, and monitoring. The most popular? "Weight change was most important to the patients," said Dr. Shah. "But then they'd see the daily routine for that: 'Oh, it's an injection twice a day?'"
After more discussion, the Mayo group added another issue: cost. "We want patients to be able to make decisions more aligned with their values and preferences," said Dr. Shah. With AHRQ-funding, the Mayo Clinic's KER Unit is testing how the cards are used and their effectiveness in 20 different clinical sites in Wisconsin, Iowa, and Minnesota. "It's a fascinating way to get patients involved in having their say." Today, patient-centered outcomes research for diabetes is helping clinicians and patients find the best available treatments for an individual patient. But more research is still needed and will be for the foreseeable future—at least until diabetes becomes rare and today's treatments seem as odd as viper's flesh and blind nettles.
Editor's Note: AHRQ's Effective Health Care Program and its many comparative effectiveness reviews and related publications can be viewed at
Effective Health Care Program