I cut off other drivers in traffic, then flip them the bird. I hang up on customer-service representatives telling me why they can’t replace my broken refrigerator. And in the line for the playground swing set, don’t even think about letting your kid get in front of mine.

I’m irritable. But it may not be my fault, at least not entirely.

I was given a diagnosis of epilepsy in 2001, when I was 24. Luckily, my seizures are easily controlled by a drug called levetiracetam, also known as Keppra. Its side effects, according to the manufacturer’s Web site, include “psychotic symptoms, suicidal ideation, irritability, and aggressive behavior.”

I’ve been spared the worst of these. But irritability? Well, I once screamed at a man on the street who saw me carrying a heavy air conditioner and dared to suggest that it might be smarter to put the unit in my car.

But irritability is a vague condition. It’s not a specific malady such as a rash or vomiting, but a state of mind. It even appears in the Diagnostic and Statistical Manual of Mental Disorders.

I could blame a pharmaceutical for my anger-management failures. But before I took Keppra, I was an impatient child, a moody teenager and a testy young adult. You could say I’ve been irritable since birth.

I’m not sure if I’m more or less irritable than I was before Keppra. The older I get, the more stressed I get. Sure, I get steamed when my toddler throws broccoli to get my attention or when my dog goes after the mailman. How can I compare my state of mind today to that of a decade ago, when I didn’t have a daughter or a dog?

How much of my current irritability is Keppra, and how much is just me?

“There’s one end of the bell curve where it’s obvious,” says David Spencer, a professor of neurology at Oregon Health and Science University in Portland. “The behavior is way out of range.”

Spencer describes two of his patients who may have been driven to abnormal behavior by levetiracetam. One, a “mild-mannered” academic, collared a student who asked about a grade. Another started displaying psychotic behavior, left his family and moved to Mexico.

“Those are exceptions. The common one is ‘He’s a little snappier than he used to be,’ ” Spencer says. But it’s hard to parse the chemical and the psychological. “Was it just something that happened to him, or was it the drug?”

No one can say for sure.

“If you find somebody [who] can, I’d like to talk to them,” says Steve White, the scientific director of a program at the University of Utah that aims to develop anticonvulsant drugs for people who are prone to seizures. White points out that 30 percent of epilepsy patients are depressed and don’t necessarily need Keppra to be irritable.

“Seizures themselves are associated with neurochemical changes,” he says. As for the irritability displayed by people on levetiracetam, “it would be difficult for anyone to put their finger on it and say, ‘This is the real reason.’ ”

White has special authority in thinking about levetiracetam: After having a seizure related to a brain tumor, he started taking the drug.

“I noticed in myself a ‘quick to flash,’ if you will,” White says, describing moments of road rage while he was on the drug. “That’s not my normal temperament.”

White was taking what he calls a “pretty high dose” of levetiracetam — 2,000 milligrams per day. (I take 2,500 per day.) Still, he was reluctant to blame the medicine for his troubles.

“There was a lot of anxiety,” he says. “I made my career doing epilepsy research and then — to have a seizure? . . . I was angry.” White switched to another medication.

The ‘nocebo’ effect

Figuring out whether an outburst is chemically induced gets more complicated when you factor in suggestibility: If a person thinks a drug may make him drowsy or dizzy, there’s a good chance he’ll feel drowsy, dizzy or both. This is called the “nocebo effect,” a harmful consequence that results from the expectation of a harmful consequence.

“The principle of informed consent obligates physicians to explain possible side effects when prescribing medications,” reads a paper published last year in the American Journal of Bioethics that discussed how nocebo effects could be minimized. “This disclosure may itself induce adverse effects.”

“We want to warn people of potential side effects, but not suggest,” Spencer says.

It’s a fine line. Doctors need a way to discuss side effects without patient examinations deteriorating into a graduate seminar on deconstruction. Is “irritability” the same as moodiness? What about anxiety?

“It’s a fuzzy term,” Daniel Safer, an associate professor of psychiatry and behavior sciences at Johns Hopkins Medicine, says of irritability. In 2009, Safer wrote a paper that recommended limiting use of the term. The problem? It has too many meanings.

“There is the frequent non-specific use of the term in psychiatry and then the dictionary meaning of the word,” Safer wrote in an e-mail. “Next is the fact that it is a subjective term that is self reported.”

Finding the right terms

Another issue: How to compare irritability in people taking a certain drug with that of naturally irritable people not taking the drug? As Safer says, “Predisposition hasn’t been studied much.”

To fight this fuzziness, the Food and Drug Administration uses a standardized vocabulary to describe side effects and “adverse events.” This lexicon is drawn from the Medical Dictionary for Regulatory Activities (MedDRA).

“It’s like a thesaurus a physician or researcher can use to label symptoms of a patient taking a drug,” says Kelly Simontacchi, a medical director at UCB, the company that invented Keppra. “This makes for consistency across clinical trials.”

During patient visits and clinical trials, doctors and researchers can use MedDRA’s 70,000-plus terms to describe symptoms and side effects. Still, something can be lost in translation. What I call “irritability,” another patient or a doctor might call “dysfunctional anger.” For a company trying to get a drug approved, the latter sounds a lot worse than the former, but patients and doctors may use more colloquial language.

“Why did you choose that term when other terms might also have been available?” asks Patrick Revelle, director of MedDRA’s maintenance and support services organization, of the word choices made by researchers — many of whom receive funding from drug manufacturers. “Was it a conscious effort to obscure something? These are questions that regulatory authorities ask.”

When Keppra was developed in the 1990s, MedDRA wasn’t yet in use. In a previous classification system, Simontacchi says, “irritability” wasn’t an option. “Nervousness” was.

You don’t need a medical degree to understand that these words are different. Woody Allen, for example, seems nervous — indecisive and neurotic, but not threatening. The gun-toting cast of a Martin Scorcese movie, however, seems irritable.

“When patients in clinical studies are queried for side effects, their reports have to be fit into preestablished categories,” Spencer says. “There’s a little disconnect between clinical studies and what we see in practice.”

‘Tolerable’ side effects

Part of a second generation of anticonvulsant drugs developed since 1989, Keppra offers consistent seizure control with side effects deemed more tolerable than those of medications of the first generation. My previous medication — valproic acid, also known as Depakote — can cause liver damage. According to a National Institutes of Health study, it’s also associated with a loss of bone density, and my neurologist told me it would increase my chances of osteoporosis. Compared with such side affects, irritability seems minor.

Baycol, a cholesterol drug, was taken off the market in 2001 after it was linked to fatal cases of kidney failure. The anti-obesity drug fenfluramine/phentermine, also called fen-phen, was pulled from the U.S. market because it may cause heart problems.

Simontacchi describes Keppra as a “blockbuster success” for UCB and for those living with epilepsy. Safer called it “a fairly smooth drug” — the irritability issue aside, I had trouble finding anyone who would criticize the drug.

Still, when I become enraged that the guy in front of me at the post office is in line just to buy stamps that he could buy from a machine, I wonder how much my medicine contributes to my prickliness. For me, life without levetiracetam would be difficult. But as it protects me, it may be making me a jerk.

“I feel your pain,” White says. “When the side effects begin to outweigh what one judges is the benefit, it’s a tough situation for any patient to be put in.”