A provision of the Affordable Care Act that allows insurers to charge smokers higher premiums may have discouraged smokers from signing up for insurance, undercutting a major goal of the law, according to a study published this month.

The surcharges, of up to 50 percent over nonsmokers' premiums, also showed no sign of encouraging people to quit.

The Affordable Care Act eliminated insurers' ability to charge higher premiums based on whether a person was sick. But it does allow them to vary premiums with age, geography, family size and smoking status.

"Some people wanted smokers to bear responsibility for the added cost of smoking, and the other thought was that some people would quit," said Abigail S. Friedman, a health economist at the Yale School of Public Health, who led the study published in Health Affairs this month. Her analysis of data from the Centers for Disease Control and Prevention from 2011 to 2014 suggests that neither of the expected outcomes happened.

To explain how the surcharge works, Friedman and colleagues provided this example: Take the case of a 49-year-old with an income at 150 percent of the poverty level. If that person were a nonsmoker, he would pay only $60 per month for a particular health plan, because the tax credit would pay the rest. But the premium surcharge that smokers face are not eligible for tax credits. So a smoker would — based on an analysis of 43 states that allowed surcharges in 2014 — pay a median $70 surcharge on top of that premium, doubling their out-of-pocket costs of insurance.

After the marketplaces made insurance available in 2014, the number of smokers with insurance grew from 70 percent to 76 percent of the total, the researchers found. They estimated that if the surcharges did not exist, that number would have risen to 80 percent.

The effects were especially noticeable for smokers under 40 years old. In 2014, 65 percent of the smokers in the sample had health insurance, but 75 percent would have enrolled if the surcharges were eliminated, the researchers estimated.

Friedman said this subgroup of younger smokers is important for several reasons: First, younger smokers tend to have health-care costs similar to other young nonsmokers, so their participation in the marketplaces helps balance the risk and makes insurance work. Second, stopping smoking when people are younger can have big, long-term health benefits for individuals and in helping to prevent costly chronic conditions.

Aaron Albright, a spokesman for the Centers for Medicare and Medicaid Services pointed out that the law mandated free smoking cessation programs and medications for plans sold on the marketplace.

"And, because of the Affordable Care Act, you can no longer be denied coverage for being a smoker. These are important steps forward for the health of our country, as is the fact that 20 million people have gained health insurance because of the law," Albright said.

The number of smokers has declined in the United States, from nearly 21 percent of adults in 2005, to 17 percent in 2014, according to the CDC.

But the study found no big differences in the likelihood of quitting smoking among those who were subject to an insurance premium surcharge and those who were not — except, strangely, for those who faced very low surcharges, who were less likely to quit than people with no surcharge.

The authors posit possible reasons for this, but have no clear answer. Perhaps, they say, the low surcharge could have had the perverse result of alleviating any guilt about the behavior.

To do the study, the authors analyzed the responses of about 200,000 non-elderly people who qualified for health insurance tax credits, as reported on a CDC behavioral risk survey. A quarter of the sample were smokers.

In a written statement, the authors said there are two possible policy implications from their data. First, states that want to increase coverage or are concerned about the sustainability of their marketplaces could lower or eliminate the allowable smoking surcharges, because these appear to reduce enrollment. Another option would be to waive the surcharge if smokers enroll in smoking cessation programs.

A separate concern that Friedman raised was that the rules could lead people to lie about whether they smoke, to avoid the surcharge. If that also led to patients lying to their doctors, it could have real public-health consequences.