Andrew Richards remembers that he had just sat down in front of the TV when the lightning bolt struck. "It was almost like it went through my ear because I could hear it. It was kind of like, 'fwomp,' right through my head," he says.

Stunned and disoriented, Richards tried to get up, he says, but spasms rumbled through his torso and his back; his muscles kept clenching and relaxing. He doesn't recall falling. "It was almost like a little piece of time was missing, almost like a record skipped ahead a little bit," he says, "like I was sitting on my couch and time had moved forward and I was on the floor."

The next thing Richards remembers was calling out to his wife, who scooped him up and drove him to the emergency room. He would soon discover that the lightning bolt that struck him down was an electrical storm taking place entirely in his brain: a seizure. Less than a year later he was in court, contending that his seizure -- and the spasms that he says still plague him -- were caused by a faulty generic medication.

Generic medications are generally safe and effective, because the Food and Drug Administration mandates tests to show that they are equivalent to the name-brand originals. But in 2011, the Food and Drug Administration determined that a major laboratory had committed such “egregious” research violations that years’ worth of its tests — many comparing generics to name-brand drugs — were potentially worthless. The FDA has required drug manufacturers to redo many of the tests from that lab, owned by the now-defunct Cetero Research. But a ProPublica investigation shows that the agency took little action to prevent consumers from taking medicines the FDA no longer knew for sure were safe or effective.

Richards’ suit involved a drug that was not tested by Cetero. But his case spotlights the possible dangers of a generic medication that isn’t actually equivalent to the name-brand original.  

The drug Richards took is Teva Pharmaceutical USA's 300 milligram version of extended-release buproprion -- an antidepressant better known by its brand name, Wellbutrin XL. Shortly after Teva's generic Wellbutrin was approved by the FDA in 2006, patients began to complain that the drug wasn't working as well as the name-brand version. Some contended that it was causing more severe side effects.

Richards says he suffered his seizure in March 2008 after his pharmacy switched him to the Teva generic. Within a few days of starting the new pills, he says, he started to experience jolts and jerks. "Sometimes when you're falling asleep, you get what's called a sleep start," Richards says. "Well, I would get that but I was awake." Then the seizure hit.

Despite having stopped taking Teva's drug, Richards says he continues to have spasms several times a day. Richards, who was a lead plaintiff in a lawsuit against Teva, contends that it was a direct result of his taking Teva's generic Wellbutrin. The suit ended in a settlement in which Teva did not admit fault. Teva and Richards won’t discuss the financial terms, but a judge’s ruling approving the settlement suggests that it was for $4.5 million, the vast majority of which went to legal fees and costs.

Neither Teva, nor Impax Laboratories, which manufactures the pill, have admitted that there was anything wrong with the pills. "The product was approved on the basis of the bioequivalency testing that we were required to perform at the time," said Marc Donohue, Impax's Senior Director for Corporate Communications. "There had never been a safety issue with the product." Teva did not answer questions about the Wellbutrin case.

"Just after [Teva's generic] went on the market, within 6 to 8 weeks, first there was a trickle, then half a dozen, then a dozen letters," says Joe Graedon, a consumer health advocate and author of "The People's Pharmacy," a syndicated drug advice column. After writing about the potential problems, the complaints became so numerous that Graedon commissioned his own study to test Teva's pill side by side with the original. The finding: As the pills dissolved, the generic version wasn't acting the same way as the name-brand. "The Teva generic releases much faster -- much, much faster -- than the name brand," says Tod Cooperman, president of ConsumerLab, the company that performed the test.  "There's this huge burst that occurs right away."

Faced with doubts from consumers, the FDA first defended the drug. In April 2008, the agency stated that Teva's generic was "bioequivalent and therapeutically equivalent to... Wellbutrin XL." The FDA dismissed complaints of ineffectiveness and side-effects as being consistent with the ups and downs of depression.

Still, the FDA asked Teva and Impax to conduct a study to clear up any doubts. Teva terminated the study after recruiting only eight patients. "It was because of slow enrollment," says Impax's Donohue, who said that the protocol required subjects to stay in a clinic for 24 days, which likely scared off potential participants.

The FDA picked up where Teva left off, and by August 2012, the agency’s own study showed that the two drugs weren't equivalent. This forced FDA to reverse its position; in October 2012, the agency declared that Teva's generic is "not therapeutically equivalent" to the original.

The FDA declined to comment.

The announcement heartens Richards, but one thing he regrets, ironically, is that he'll never again be able to take even the brand-name version of Wellbutrin, which he says was tremendously effective in treating his depression. "Everything clicked into place. Everything felt normal again," he says. But, he says, his doctors now warn that taking any form of Wellbutrin is out of the question. "I actually asked them if I could go back on the brand, but the damage was done."