There are few more potent combinations than lawyers and journalists. Together they can demonize, loot, and even bankrupt the largest industry. And do so based on the flimsiest evidence. But the tide is turning, as evidenced by the decision of U.S. District Court Judge Louis Bechtle to bar junk testimony in the fen-phen diet drug litigation.

Three years ago, studies linked the appetite suppressant fenfluramine to heart valve damage. Six million people had taken fenfluramine (or the related dexfenfluramine), often along with phentermine, a separate product that increases the calorie burn rate, creating a combination known as fen-phen. The Food and Drug Administration quickly ordered fenfluramine’s withdrawal from the market.

Bad publicity followed, abetted by trial attorneys, who began running magazine and newspaper ads, hitting TV talk shows, creating Web sites and organizing lawyer networks. As defense attorney David Bernick put it, the trial bar was ”investing in litigation rather than pursuing it.”

A legal deluge followed: thousands of individual suits backed by massive class actions. Caught in the backwash was phentermine, even though research fingered only fenfluramine. Sue anyone with a deep pocket is the trial bar’s chief operating principle.

Phen litigants have had an increasingly difficult time, however.

Two years ago, judges in Arkansas, New Jersey and Washington dismissed claims against phentermine.

Courts in Massachusetts, Mississippi and Texas did the same last year.

For instance, Mississippi’s County Circuit Judge Lamar Pickard directed a verdict for the defendant, explaining: ”that no evidence was admitted that phentermine may cause any injury alleged by plaintiff, there is no substantial evidence, and indeed not a scintilla of evidence, of proximate cause as to phentermine.”

Some plaintiffs have attempted to adduce such evidence, but with little success.

Last December, a Massachusetts Superior Court rejected testimony linking phentermine and primary pulmonary hypertension by Drs. Paul Wellman and Alan Woolf, ruling that it did not reflect acceptable scientific methodology.

Even more significant was the recent ruling by Judge Bechtle, who is overseeing federal fen-phen class action litigation, including the $3.75 billion settlement by fenfluramine-maker American Home Products Corp.

There will be no similar settlement involving phentermine. Earlier this year, phentermine defendants moved to exclude testimony from Dr. Wellman, the behavioral psychologist who lost out in the Massachusetts case, and Dr. Timothy Maher, a pharmacologist, tying the drug to heart disease.

Bechtle held a two-day hearing, based on the Supreme Court’s Daubert decision, which set standards for evaluating scientific evidence. He ruled that ”the opinions of Doctors Maher and Wellman do not amount to scientific knowledge” and had ”not demonstrated reliable data to support their hypothesis.”

Indeed, Bechtle tore the guts out of the plaintiffs’ case, which he debunked in a closely reasoned 35-page opinion. After reciting their claims at length, he pointed to ”the unreliability of Doctor Maher’s and Doctor Wellman’s opinions.” Their views are ”based more on personal opinion than on” scientific knowledge. ”Doctors Maher and Wellman have not demonstrated reliable data to support their hypothesis.”

Their views, stated Bechtle, ”are not generally accepted in the scientific community.” He found problems with not only the potential witnesses’ data and methodology. There was ”an analytical gap.” Concluded the judge: ”Even if their data … and the methodology … are reliable, the conclusions that Doctors Maher and Wellman have reached … do not reliably flow from this data and methodology.”

Bechtle’s ruling demonstrates the value of the role of judges as an evidentiary gatekeeper. In Daubert, the Supreme Court stated that the trial judge was to ”ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.” Expert testimony must reflect ”scientific knowledge.”

Stated the court: ”The adjective ‘scientific’ implies a grounding in the methods and procedures of science,” and ”the word ‘knowledge’ connotes more than subjective belief or unsupported speculation.”

In fairness to plaintiffs, judges must be flexible. New theories can be proved correct and established orthodoxies often seek to squelch new thinking. New theories with a reliable basis deserve their chance in court, subject to cross-examination and criticism by other experts.

But, jurors have difficulty sorting among flawed but often superficially plausible theories. The task is not easy for judges either, of course. However, they can take specialized technical courses, avail themselves of aid offered by groups like the American Association for the Advancement of Science, and hire independent experts.

Moreover, the Daubert hearing concentrates on the critical reliability issue.

Liability rules are important, as a mechanism to make whole the injured and punish the irresponsible. But a properly functioning tort system must also distinguish between the innocent and guilty as Bechtle has done between fen and phen.