Those who seek to defend the legacy of Irving Selikoff rightly point to his publications on hazards of asbestos. These papers were mostly epidemiologic studies. Selikoff was not trained as an epidemiologist, and most of his papers on asbestos had co-authors who had a reasonable pretense to expertise in epidemiology. Presumably, Selikoff’s contributions to his papers were as a clinician. Putting aside Professor Bartrip’s concerns over the quality of Selikoff’s clinical medical education, what is known about Selikoff’s ability as a clinician?
My first experience with an asbestos case, in which Selikoff had written a report, was the Leddy case. Mr. Leddy worked had worked for the Reading Railroad, in Reading, Pennsylvania. Selikoff, with the union, organized a screening of the railroad workers, and Selikoff read the chest films. He read Mr. Leddy’s films as showing 3/3 on the ILO scale — and he interpreted the films as showing severe asbestosis. The Motley firm filed a case on behalf of Leddy in the Eastern District of Pennsylvania, around 1982, and the case landed on my desk to defend.
Shortly after being diagnosed with advanced asbestosis by Selikoff, Mr. Leddy developed signs and symptoms of lung cancer. Leddy went to a local hospital, where he was diagnosed with lymphangitic bilateral spread of lung cancer. He died shortly after diagnosis. The hospital pathologist, who was aware that Mr. Leddy had a lawsuit pending, and who was aware of Selikoff’s reputation as a pioneer in the health effects of asbestos, conducted an autopsy. The post-mortem findings were astounding. Mr. Leddy, who had been a heavy smoker, indeed had, and died of, lung cancer. In over 90 lung tissue sections, however, with appropriate Prussian-Blue staining, from all five lobes, the pathologist could find no evidence ofnasbestosis. (Later fiber counting confirmed that the plaintiff had less asbestos in his lungs than the average resident of New York City.)
These were the good ole days when defense counsel were permitted to conduct ex parte interviews of treating physicians, and so I telephoned the hospital pathologist to discuss the case. The pathologist was eager to talk about the case, because of the Selikoff’s fame and the apparent error of Selikoff’s diagnosis. The pathologist had another motive — he was a bit put out by the widow’s reaction to the news that her late husband did not have asbestosis. When he called Mrs. Leddy with the results of the autopsy, she accused him of being in cahoots with the railroad company! The pathologist willingly agreed to make himself available for a videotaped deposition on his findings, after which the case, notwithstanding the Selikoff diagnosis, settled for a nominal amount.
Of course, the Leddy case is just an anecdote, and anyone could mistake lymphangitic pulmonary carcinomatosis for asbestosis. Still, it led me to wonder about exactly how skilled Selikoff was in the clinic. Later I heard rumors that Selikoff had taken and failed the B-Reader examination, given by NIOSH, for certifying proficiency in applying the ILO system to classify chest radiographs for the pneumoconioses. (I would be interested to hear from anyone who has information that confirms or refutes these rumors.)
Perhaps more interesting than Selikoff’s score on the B-Reader examination is how he measured up against some of the outstanding pulmonary clinicians who had studied asbestosis, both in Europe and in North America. In 1972, C.E. Rossiter, of the U.K.’s Medical Research Council, published an important study on the reliability of a system of classifying chest radiographs for pneumoconiotic changes. Using 12 X-ray readers, including some of the leading radiologists and pulmonary physicians of the day, Rossiter assessed whether the classifications assigned were repeatable between and among readers, and by individual readers themselves. Among the readers were Drs. Irving Selikoff, Leonard Bristol, Benjamin Felson, Eugene Pendergrass, Corbett McDonald, and Sluis-Cremer. Today, such a study would be accompanied by much more statistical apparatus (Kappa statistics, etc.), but one does not need any statistical analysis to see that Selikoff was an outlier, who read films as showing abnormal profusion of small, irregular densities up to twice as often as the most reliable readers in the study. Rossiter, “Initial repeatability trials of the UICC/ Cincinnati classification of the radiographic appearances of pneumoconioses.” 29 Brit. J. Indus. Med. 407 (1972).
Perhaps when Selikoff’s contributions to the health effects science of asbestos are honestly assessed, a disinterested observer will be able to evaluate whether Selikoff overstated the health effects of asbestos, by design, by the undue unfluence, or by lack of ability.
