Posted on 12/15/2002 2:21:13 PM PST by theFIRMbss
"Mystery Illness" or Mysterious Reporting?
What’s wrong with science and health reporting? For a quick inventory of some common journalistic shortcomings and scientific carelessness, you could turn to a special report that appeared in The Nashville Tennessean in late September. In fact, many readers had that opportunity. The three part series, entitled "Mystery Illnesses Found Nationwide; Sickness Surrounds Nation’s Nuclear Weapons Complexes," generated national attention, with reports from USA Today ("News Report Traces Pattern of Illnesses Near Nuclear Plants," September 29) and the Associated Press ("Widespread Illnesses Hits 11 States," September 30).
The central charge of the Tennessean’s report is that "Men, women and children who work at or live near the nation’s nuclear weapons complexes ... are suffering from a pattern of baffling health problems their doctors cannot explain." Specifically, the Tennessean interviewed 410 people "with unexplained health problems around 13 sites in 11 states."
Unfortunately, the 410 sick individuals did not represent a proper sample, but rather a "convenience sample" of self-selected or referred individuals. The Tennessean notes that reporters Susan Thomas and Laura Frank, "In the course of reporting a pattern of unexplained illnesses in Oak Ridge last year ... heard similar symptoms were occurring around other weapons sites. Frank and Thomas went to those sites, asking questions. In some cases, environmental activists offered the names of people they believed to be sick. In most cases, the reporters found one or two sick people to tell their stories, these people then directed the reporters to others in the community with similar symptoms." Such selection methods would make any epidemiologist wince.
Second, what Thomas and Frank encountered were self-reports of symptoms, largely taken at face value without the cross-check of a physician’s diagnosis. Moreover, the symptoms represent no recognized diagnostic classification. Instead, they are a polyglot miscellany. Among the ailments mentioned are: fatigue, tumors, memory loss, blood disorders, headaches, reproductive abnormalities, dizziness, liver problems, sleeplessness, rashes, panic attacks, hair loss, cleft palate, tremors, numbness, vision loss, immune system deficiencies, depression, hearing loss, asthma, acute muscle and joint pain, chemical sensitivity, rapid heartbeats, thyroid malfunctions, nervous disorders, and intestinal and digestive disorders.
Third, there was no effort to establish a control group, and no investigation of possible epidemiological confounders. No comparative numbers were offered to provide a sense of whether illness rates in the sample were higher or lower than any other population in any other setting. George Lucier, director of environmental toxicology at the National Institute of Environmental Health Sciences, is quoted saying, "Four hundred people is a lot of people." Indeed it is, being more than three hundred but not as many as five hundred. But since the population within a 50 mile radius of the respective facilities totals over 4 million (as the AP report mentioned), how many illnesses should we expect?
Especially troublesome is the logic of the report regarding "unexplained" illnesses. Perversely, the very fact that the illnesses are of unknown causation sometimes signals to reporters the presence of a "master cause" — toxic emanations from nuclear sites. That is, the more mysterious the illness, the more compelling they regard the evidence that some unknown agent must have brought it on.
Finally, the report is organized around the fallacy that "correlation equals causation" — in this case, that a person with an unexplained illness who lives near a church has simply got an unexplained illness, whereas the same illness in the vicinity of a nuclear facility must be caused by the facility.
The Tennessean does note editorially, "What (Thomas and Frank) found was anecdotal evidence of a problem, not scientific proof ... they were acting as journalists, not scientists." But it is difficult to produce good journalism from flawed scientific thinking.
At some point it will
get reasonable to ask
if it's just numbers...
Becareful! Drudge sez:
"Marine recruit dies after
rash spreads" and, of course,
these are the symptoms
the early victims display
in Crichton's book Prey!
Right now I still think
the press figures sickness sells
and headlines this stuff.
But you never know.
There could be a whole new kind
of war going on.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.