Prescription Opioids: Risk Management and Strategies for Safe Use _ _________________________________
ARGUMENTS USED TO SUPPORT ERRONEOUS CONCLUSIONS IN BIASED RESEARCH REPORTING
Form of Argument
Definition
Explanation or Example
False conclusions of causation based on correlation Non causa pro causa (no cause for cause)
One or more events suggested as causing another event Causation based on an association between two or more event trends or outcomes that occur together in time
Even when data show a statistically significant correlation, assumption of cause and effect is erroneous. 1) The correlation may be significant, but correlation is not causation, and more research is needed to rule out other explanations for the association. 2) The direction of causation may be the reverse of the false conclusion. This is common in observational and open-label studies, because factors that actually influence outcome are not controlled for. This “regression to the mean” can falsely be attributed to treatment effect. Common in data-mining studies and erroneous due to: 1) The data cluster may be the result of chance. 2) Even if not random, the cause may differ from what is stated by the researchers. Often seen in pain medicine, as when the lack of long- term controlled studies on opioid safety and efficacy in chronic pain is stated as evidence against long-term opioid use in chronic pain In an argument with weak factual support, this is used to mislead the reader into not questioning the accuracy, reliability, or validity of the data the argument is based on. Argues that a popular treatment (e.g., homeopathic pain remedies) would not be so widely used if it did not work. Avoids the need to show credible evidence. This fallacy has been used when infrequent patient outcomes stand out and are generalized to represent all patient outcomes. Can occur when data from a small, highly select group of patients with pain, or even data of individual patients by anecdote, is used to characterize an entire population of patients. Reflected by statements such as “no true patient in pain would abuse their medication” Epidemiologic studies may record the rate of opioid abuse by the number persons who either did or did not ingest a non-prescribed opioid analgesic in the past year. This neglects any detailed analysis, such as motivation by untreated pain, inadequately treated pain, or desire to get high. This appeal to altruism and the presumption of good intentions may be used to deter examination of possibly deficient or biased reasoning or harmful unintended consequences.
Cum hoc, ergo propter hoc (with this, therefore because of this)
Conclusion of causality based solely on the sequence of events
Post hoc, ergo propter hoc (after this, therefore because of this)
Regression fallacy
Pain severity declines over time to a lower average level during the natural course. Certain variables showing a close association are selected from a vast array of data, and a cause-effect relationship is concluded.
Texas sharpshooter fallacy
False arguments used in support of a conclusion Argumentum ad ignoratum (appeal to ignorance)
Missing evidence is itself evidence for lack of an effect.
The high-status source of a publication is used to affirm the results.
Argumentum ad verecundiam (appeal to authority)
The widespread use and acceptance of a practice prove its validity.
Argumentum ad populum (appeal to the people or popularity)
Illusory correlation
An expected relationship between data, observations, or events is found when a true causal relationship is absent. A large, complex phenomenon is oversimplified by reducing it to a smaller, simpler component. Used as an ad hoc rescue of a reductionist argument that comes under criticism Forces simple answers to complex questions with an argument in which only two choices are offered
Reductionism
The “no true Scotsman” fallacy
False dichotomy
Myths of beneficence
Programs or policies are argued as beneficial to patients or the public and thus should be accepted.
Source: [217; 218]
Table 12
54
MDMS1526
Powered by FlippingBook