Exam Ruling Derails Foreign Doctor’s Dream : Medicine: Man licensed to practice in Mexico has failed the U.S. test repeatedly. He has been told to retake it and believes that he is erroneously being accused of cheating. Others are in similar straits.

After 10 years of struggle and failure, Luis Garcia-Higgins finally believed that his dream of becoming a doctor in the United States was within his grasp.

Garcia, 42, a naturalized U.S. citizen who is licensed as a doctor in Mexico, was beaming with confidence when he finished the two-day, 1,000-question Foreign Medical Graduate Examination last January in San Diego.

He had taken and failed the exam a dozen times since 1983, but this time he had taken a course in test-taking and spent extra time reviewing his books. There is no limit on the number of times a foreign graduate can take the test, and Garcia, the son of a leathergoods shopkeeper in Colombia, had been determined to keep trying.

When a letter arrived April 2 from the Philadelphia-based commission that administers the test, Garcia and his wife, Claudia, were buoyant. The letter said his exam had been reviewed but his results were being withheld “pending completion of statistical analyses.”

“We told all our friends: We’re going to get it this time!” said Claudia Garcia, a recent graduate of UC San Diego. “Everyone knew how hard he’d been preparing for the test.”

Ten days later a second letter arrived. A statistical analysis had “revealed a strong relationship of joint wrong answers between you and one or more examinees, which suggests that the examination may not be a valid measure of your knowledge.”

Although the letter did not use the word, Garcia believed that he was being accused of cheating, of somehow conspiring with others to share answers. His exam results were being scrapped; he would have to take a makeup exam.

Garcia and his wife were crushed.

“I started crying,” Garcia said. “I’m a soldier. I don’t cry on a normal basis. But this is my life, this is what I’ve been fighting for for years and years, and they’re trying to take it away from me.”

Stunned and angry, Garcia has decided to fight. A high-profile UC San Diego professor has taken up his cause and they are deciding whether to appeal or sue.

He is not alone. Other graduates of foreign medical schools believe that they are similarly abused. After the test, more than 100 were sent letters similar to Garcia’s.

An attorney in Rhode Island, Bernard Ferguson, who was formerly counsel to several medical schools in the Caribbean, said the commission’s zeal to root out cheaters has gone too far.

He is considering a class-action lawsuit on behalf of the growing number of foreign medical graduates who, like Garcia, are having their test scores invalidated, they say, without any proof of cheating other than statistical analyses.

Officials at the Educational Commission for Foreign Medical Graduates, which administers the test, say the system of statistical analyses that pointed an accusing finger at Garcia is necessary to ensure the integrity of the examination. The commission is a private group organized and supported by the nation’s major medical associations.

“You have to remember that these people are going into hospitals and provide treatment for you and your family,” said Marie Shafron, commission vice president for operations. “We want to be absolutely certain of their abilities before we let them do that.”

Although the commission has administered the exam since the mid-1950s, concern about cheating and purloined tests has heightened considerably in the past decade.

In 1983, when the commission got information that a test had leaked out, one of the two testing sessions scheduled for that year was canceled. A grand jury in Miami was convened three years ago to check rumors of stolen tests, although there were no indictments.

Two years ago in New York City, two graduates of medical schools in India were arrested in the parking lot outside a test site trying to buy a test. Investigators have been particularly worried about reports of cheating from graduates from the Philippines.

In Garcia’s case, the commission’s statistical analyses found the “strong relationship of joint wrong answers” with five people who took the test in New York, one in Philadelphia and one in New Orleans.

After the January exam, 109 people who had taken the test at various centers got letters saying their scores would not be recorded because of the “strong relationship.”

Lest any of the 109 claim that the similarity was coincidental, the letters used an intimidating set of numbers. The letter to Garcia said that the “approximate probability” that he arrived at the answers innocently was “less than one in a billion.”

The letter offered a kind of plea bargain: Take a makeup examination and no report of the “strong relationship” will be forwarded to the state where you are seeking a license.

Peter Irons, a professor of political science at UC San Diego and director of the university’s Earl Warren Bill of Rights Project, is championing Garcia’s case. He says the commission’s appeals process, which asks graduates to prove that they did not cheat, is “something right out of Kafka” because it presumes someone is guilty until proven innocent.

Garcia, who was born in Colombia, was a sergeant in the U.S. Army and, after ending his Army hitch in Panama, attended medical school in Guadalajara.

In 1983 he took the Educational Commission for Foreign Medical Graduates test, one of two major hurdles for foreign graduates seeking a medical license in California. He failed the English proficiency and science parts of the test.

He has worked as a janitor, a physician in a Tijuana clinic, and, for the past two years, as a $40,000-a-year physician’s assistant in the federal prison at Lompoc. He kept taking the test, finally passing the English proficiency section and edging closer to passing the medical and clinical sections.

The test is administered twice a year to up to 20,000 people at 175 sites worldwide. The passage rate, by design, is low: About one-third pass each time the test is given.

Garcia finds preposterous the notion that he is being accused of somehow conspiring with people he has never met and who took the test thousands of miles away from San Diego.

“In the electronic age, anything is possible,” said commission attorney Bruce Hubbard of Boulder, Colo. He notes that the test is given in various time zones, with plenty of time for examinees to share information through faxes and phone calls.

According to Hubbard, Garcia apparently fell into the commission’s statistical snare because he fit the profile of those whose exams the commission has decided to scrutinize: He was a repeater and his score showed improvement.

A commission document stamped “confidential” says the commission has “compelling evidence” that an illicit copy of the January, 1993, test had been obtained “by some examinees.” For that reason, different exams were substituted at the last minute at four testing sites.

Shafron, the commission vice president, said a criminal investigation is under way although she declines to say whether Garcia and the seven people in New York, Philadelphia and New Orleans are among its targets.

Convinced that Garcia’s rights are being trampled, Irons is prepared to present the commission with testimony from faculty colleagues with expertise in statistics who say that the commission’s method of statistical reasoning is bogus.

One of those asked by Irons to review Garcia’s case was J. Laurie Snell, one of the nation’s leading authorities in probability statistics and author of numerous texts on the subject.

Snell calls the commission’s finding “crazy” and does not understand how it can draw such conclusions based on the shared wrong answers theory. In brief, he says that in an exam with 1,000 questions, five possible answers per question, and 20,000 people taking the exam, he cannot see how a similarity of 100 or so wrong answers between two or more people can be construed as even circumstantial evidence of foul play.

Commission attorney Hubbard said the statistical model was devised by experts at the National Board of Medical Examiners. He is confident that their methodology will withstand professional review.

The commission’s organizational members are the American Medical Assn., the American Board of Medical Specialties, the American Hospital Assn., the Assn. of American Medical Colleges, the Assn. for Hospital Medical Education, the Federation of State Medical Boards of the U.S. and the National Medical Assn. The groups retain membership on the commission’s board of directors.

In the face of such austere groups, Claudia Garcia, a recent UC San Diego graduate, is not optimistic. But she shares her husband’s feeling that taking a makeup exam is tantamount to admitting that he cheated.

“How are we going to fight someone who is backed up by the AMA?” she asks. “Who are we? You strive all your life to be someone, and then they say: ‘Forget it. We’re not going to let you practice medicine.’ ”