A Sept. 14 statement from drug-maker Pfizer said the company believes the FDA should follow an advisory committee’s recommendation to remove a serious Black Box warning from its anti-smoking drug Chantix, which was released in 2006. But a report released by the FDA the same day outlines concerns about the advisory committee’s data accuracy when it comes to the psychological effects of the medicine.

The FDA panel of 19 ultimately voted four to 10 to remove the panel, reported Fox.

The current warning advises patients and prescribers that the drug’s side effects can include changes in behavior, depression and even suicide. Pfizer’s trial data, published in the journal the Lancet and submitted to the FDA in support of removing the Black Box, shows that such worries might be overblown and the drug is safer than originally believed.

“We are pleased with the committee's recommendation to remove the boxed warning and believe this is an important step toward updating the CHANTIX labeling to more accurately reflect its neuropsychiatric safety profile and help patients and prescribers make informed decisions about treatment options,” Pfizer CMO Freda Lewis-Hall, MD, said in the statement.

But some FDA staff members aren’t all the way convinced yet, according to the report.

That’s because the data measured in the trial, which Pfizer conducted in partnership with GlaxoSmithKline and included two other smoking-cessation drugs, were difficult to define and not always consistently recorded.

“The trial was designed in a well-intentioned attempt to capture somewhat ill-defined and complex neuropsychiatric phenomena. However, many problems in the implementation were apparent upon review of the collected data,” said the report.

Some of the measured mood changes could have inadvertently been subjectively recorded with varying severity, Bloomberg reported. And that could be a problem, said the FDA’s report, since some of the doctors working on the trial had previously been paid up to $25,000 to support Chantix. The physicians could have (consciously or subconsciously) allowed that positive relationship to influence their reporting in favor of the drug.

But one Pfizer official told Bloomberg that even if that were true, it wouldn’t skew the results one way or the other—the study was double-blind, so if doctors were influenced one way or the other, the biases should cancel each other out.

According to the Wall Street Journal, Chantix does not represent a large portion of Pfizer’s drug sales (less than a quarter of one percent). But reports of suicides related to the drug depressed sales after the 2009 addition of the Black Box warning, the Journal said.

Pfizer’s statement made a connection between people with existing mental illness and higher rates of smoking, explaining that people who already suffer from psychiatric issues might benefit from an aid to help them stop smoking. And the company pointed out that the effects of smoking kill nearly half a million people in the U.S. every year.

And the FDA panelists who voted in favor of removing the Black Box cited smoking’s preventable death rates as reason enough to potentially expand access to the drug, even given its potentially damaging psychiatric effects, said Fox.