In lieu of an abstract, here is a brief excerpt of the content:

16 On Drugs, Disability, and Death Whether statisticians like it or not, their results are used to decide between hypotheses, and it is elementary that if p entails q, q does not necessarily entail p. We cannot get from “the data are unlikely given the hypothesis” to “the hypothesis is unlikely given the data” without some additional rule of thought. Those that reject inverse probability have to replace it by some circumlocution, which leaves it to the student to spot where the change of data has been slipped in[, in] the hope that it will not be noticed. harold jeffreys 1963, 409 By 1988 the International Committee of Medical Journal Editors had been sufficiently pressured by the Rothmans and Altmans to revise their “uniform requirements for manuscripts submitted to biomedical journals.” “When possible,” the committee wrote, “quantify findings and present them with appropriate indicators of measurement error or uncertainty (such as confidence intervals). Avoid sole reliance on statistical hypothesis testing, such as the use of p values, which fail to convey important quantitative information”(in Fidler et al. 2004b, 120). The formulation is not ideal. The “error” in question is tacitly understood to be sampling error alone when after all a good deal of error does not arise from the smallness of samples. “Avoid sole reliance” on the significance error should be “Don’t commit” the significance error. The “important quantitative information” is effect size, which should have been mentioned explicitly . Still, it was a good first step and in 1988 among the sizeless sciences was amazing. The requirements—on which at a formative stage Rothman, among others, had contributed an opinion—were widely published. They ap176 peared, for instance, in the Annals of Internal Medicine—where later the Vioxx study was published—and in the British Medical Journal. More than three hundred medical and biomedical journals, including the American Journal of Public Health, notified the international committee of their willingness to comply with the manuscript guidelines (Fidler et al. 2004b, 120). But the requirements have not helped. “Significant” Temptations to Use Drugs The essence of the problem of reform—and the proof that we need to change academic and institutional incentives, including criteria for winning grants—is well illustrated in a study we have already mentioned of “temptation to use drugs” published in the Journal of Drug Issues. The study was financed by the Centers for Disease Control. It was authored by two professors of public health at Emory University (one of them an associate dean for research) and a third professor, a medical sociologist at Georgia State University. The study was conducted in Atlanta between August 1997 and August 2000. Its subjects were African American women—mothers and their daughters—living in low-income neighborhoods of Atlanta.1 The dependent variable was “frequency-of-[drug] use and times-per-day” multiplied for each drug type and summed by month. In the 125 women studied the value of the dependent variable ranged from zero to 910, that is, from zero to an appalling thirty drug doses a day. Statistical significance decides everything. Initially, each of the temptations-to-use drugs variables was entered into simple regression equations, to determine if they were statistically significant predictors of the outcome measure. Next, those found to be related to amount of drug use reported were entered simultaneously into a stepwise multiple regression equation. . . . Next, the bivariate relationships between the other predictor variables listed earlier were examined one by one, using Student’s t tests whenever the independent variable was dichotomous. . . . Items that were found to be marginallyor statistically-significant predictors in these bivariate analyses were selected for entry into the multivariate equation. (Klein, Elifson, and Sterk 2003, 169, 170) The authors do at least report mean values of the temptations to use drugs—a first step in determining substantive significance. For example, they report that women were “least tempted to use drugs when they were talking and relaxing (74.0%), experiencing withdrawal symptoms On Drugs, Disability, and Death ⱐ 177 (73.3%), [and] waking up and facing a difficult day (70.7%). And they would be tempted “quite a bit” or “a lot” when they were “with a partner or close friend who was using drugs (38.5%)” or when “seeing another person using and enjoying drugs (36.1%)” [170]. Recall how the authors presented their findings. while with friends at a party (p ⬍ .001), while talking and relaxing (p ⬍ .001), while with a partner or close friend who is using drugs (p...

Share