"It helps the patient understand that we are serious about helping them get treated," said Dr. Deb Sodt, who leads the Minnesota Department of Health Tuberculosis Prevention & Control Program and was not involved in the new work.

Recent studies suggest that fewer than half of people don't follow doctors' medication orders, even for life-threatening conditions. Around 130,000 people die each year in the United States because they don't take their medication to prevent heart disease, for example. (See Reuters Health story of July 25, 2012:.)

For the new study, Dr. Nancy Petry and her team at the University of Connecticut looked at 21 studies that offered incentives to people taking medication for tuberculosis, substance abuse, HIV, hepatitis, schizophrenia and stroke prevention.

One-third of the studies offered people with tuberculosis (TB) weekly rewards of $5 to $10 to stick to their treatment schedules.

People who took their drugs on time - and researchers either watched them as they did so, or used electronic bottle caps that tracked when bottles were opened - were rewarded with grocery vouchers, bus tokens, food and clothing, or money.

Overall, the researchers found that people were most likely to stick to their medication if they could earn $50 or more per week by doing so. However, only five of the studies offered this level of incentive, to people taking medication for HIV or substance abuse - rewards of up to $96 a week, on average, were available.

Sodt's Minnesota program has been offering $5 grocery vouchers to people who take their TB medication for years.

But it's not automatic, Sodt said. Incentives are only used if they will help a person complete their therapy.

NOT THE ONLY OPTION

In one study Petry's team looked at, people could earn up to $70 a week - at $2 to $10 per dose - for taking their HIV meds correctly over four weeks. Ninety percent of people who were offered a reward took their medication, compared to 69 percent who didn't receive any financial incentive.

Among people with TB in a 1990 study, 64 percent of people who received a $5 reward for each clinic visit finished their treatment, compared to 27 percent of those who didn't receive a reward. Patients were treated for 9 to 12 months.

Rewards offered at least once a week were also more likely to encourage people to take medication, and the longer the rewards were available, the better. Studies lasted 24 weeks, on average.

Petry noted that using incentives gets more complicated for long term medications, like HIV and hypertension drugs, that people have to take for the rest of their lives.

But there may be substantial public health consequences if people don't complete their treatment.

With HIV for example, if a patient doesn't continue the drug regimen they develop drug resistance and can pass the resistant virus on to others, who would then have fewer options for treatment, Petry told Reuters Health.

According to Petry, offering people financial incentives for taking medication could be cost-effective for health insurers, particularly in the case of blood pressure and cholesterol medications.

Dr. Mark Pauly, who studies healthcare costs at the University of Pennsylvania, disagrees. He told Reuters Health that doctors would most likely end up paying the people who would have taken their medication anyway.

Financial rewards aren't the only option. Recent research on diabetes suggests that support from a buddy or mentor can be as effective as money when it comes to helping people stick to their medication.

For Pauly, there are still too many unanswered questions. "Did these incentives make people do what doctors wanted to them to do? It doesn't necessarily translate into better health."

SOURCE: bit.ly/NJEfRh American Journal of Medicine, online July 16, 2012.

• A Permanent Economic Emergency