The findings, reported in the Annals of Internal Medicine, sound logical. But they lend some hard numbers to the idea that lower drug costs should improve people's adherence to their medication regimens.

"It was striking to us," said lead researcher Meera Viswanathan, of RTI International, a Durham, North Carolina-based research institute.

"If you help people with costs, even a little, it seems to improve adherence," Viswanathan said in an interview.

She and her colleagues reviewed several dozen U.S. studies on various efforts to improve people's ability to stick with their prescriptions. A few of those studies focused on insurance coverage - either giving people drug coverage or lowering their out-of-pocket costs for prescriptions.

Some looked at what happened after Medicare prescription coverage took effect in 2006; others looked at cutting out-of-pocket payments for people with private insurance.

Overall, better coverage seemed to help. In a study of nearly 6,000 heart attack patients, for example, those given full drug coverage through their insurer got more prescriptions filled over about a year.

Of patients who were on their usual insurance, 36 percent to 49 percent filled their prescriptions, depending on the medication. Those rates were four to six percentage points higher among people with full drug coverage.

Patients with full coverage also suffered a new complication, like a stroke or second heart attack, at a slightly lower rate: 11 percent, versus just under 13 percent.

But while there is some evidence of actual health benefits, not many studies have followed people long-term to see if the better drug adherence translates into a longer or healthier life.

"There were some encouraging findings," Viswanathan said. But more research is needed to know what the long-term health effects are, she and her colleagues write.

The results do not mean that better drug coverage is the only way to get people to fill their prescriptions, according to Viswanathan.

The studies in the review found some other tactics to work, too. Education plus "behavioral support" was one.

That goes beyond telling patients about their health problem, and why a particular medication is needed, Viswanathan said. "You would also try to get through the barriers that may keep a patient from taking it," she said.

If a patient was afraid of side effects, for example, a nurse might discuss that with him or her.

Another measure that seemed effective was "case management." That means the health provider would try to identify patients at high risk of not using their prescriptions, then follow-up with them - with phone calls, for instance.

It's not clear, Viswanathan said, how programs like that could be "scaled up" to be widely used in everyday practice, and not just clinical trials.

And the specific fixes might differ depending on the health problem. With high blood pressure, a fairly simple move seemed to boost patients' adherence to their medication: Giving prescriptions in blister packs rather than bottles, so people could more easily keep track of whether they'd taken their daily dose.

With more complex measures, the question of how to work them into the real world remains. "We need to know, what does it take to implement them into clinical practice?" Viswanathan said.

Figuring out how to get people to stick with their medications is considered a key part of improving healthcare. Studies show that 20 percent to 30 percent of prescriptions are never filled, and half of medications people take for chronic ills are not taken correctly.

All of that is thought to contribute to 125,000 deaths a year, and to cost the U.S. healthcare system as much as $289 billion annually.

SOURCE: bit.ly/QKnvoZ Annals of Internal Medicine, online September 10, 2012.

• A Permanent Economic Emergency