Heterogeneity within populations and subgroups
Idea: How people respond to treatment may vary from one subgroup to another--When is this a matter of chance or of undetected additional variables? How do we delineate the boundaries between subgroups?
Lagakos provides a statistician's cautions about the significance of results derived from subgroups of the whole population, especially if the subgroups were only defined after exploring the data.
The opposite caution is that treating everyone as if they were from the same population (for good statistical reasons) distracts our attention from the clues that might lead us to seeing that the population is not one uniform whole, but is a mixture of types. This can have significant health care implications -- see case studies about different kinds of breast cancer (Regan) and aspirin resistance.
Davey-Smith provides an epidemiologist's warning against paying much attention to heterogeneity. (My paper for session 12 pushes back against D-S's use of heritability studies to support his position and, more generally, against his "gloomy prospect.")
Fazel's 2012 review of "commonly used instruments" for "making decisions about sentencing, release or preventative detention" showed that low-risk and high-risk offenders needed to be separated and the predictive value of the instruments was very poor for the high-risk offenders.
Steinbach et al. 2014 examine the finding that "living in more affluent areas protects children from injury [from pedestrian accidents, but] this is not true for those in some minority ethnic groups."
The other supplementary readings address heterogeneity among subgroups with respect to aspirin and a purportedly race-specific medicine (BiDil).
(There are a heterogeneity of heterogeneities, as is conveyed in this discussion paper: HeterogeneityIowaPaper.pdf
. If you are interested to explore the significance of heterogeneity further, read the introduction at http://wp.me/pPWGi-6e
, listen to the lecture and follow along on the pdf, "Troubled by Heterogeneity? Opportunities for Fresh Views on Long-standing and Recent Issues in Biology and Biomedicine." Some of the issues raised arise in weeks beyond this one specifically on heterogeneity, but they might be worth previewing here to help get a sense of what kinds of heterogeneity are and are not addressed in this week's readings.)
Notes and annotations from 2007 course
Common readings and cases: Regan 2005 (Forms of breast cancer), Lagakos 2006 (Statistical concerns)
Supplementary Reading: Davey-Smith 2011, Eikelboom 2003, Fazel 2013, Gum 2003, Kahn 2007, Nelson 2005, Steinbach 2014
- Davey Smith, G. (2011). "Epidemiology, epigenetics and the 'Gloomy Prospect': embracing randomness in population health research and practice " International Journal of Epidemiology 40: 537-562.
- Eikelboom, J. W. and G. J. Hankey (2003). "Aspirin resistance: a new independent predictor of vascular events?" Journal of the American College of Cardiology 41: 966-968.
- Fazel, S. (2013). "Coin-flip judgement of psychopathic prisoners' risk." New Scientist.
- Gum, P. A., K. Kottke-Marchant, et al. (2003). "A prospective, blinded determination of the natural history of aspirin resistance among stable patients with cardiovascular disease." Journal of the American College of Cardiology 41: 961-965.
- Kahn, J. (2007). "Race in a Bottle." Scientific American(July 15).
- Lagakos, S. W. (2006). "The challenge of subgroup analysis--Reporting without distorting." New England Journal of Medicine 354: 1667-1669.
- Nelson, M. R., D. Liew, et al. (2005). "Epidemiological modelling of routine use of low dose aspirin for the primary prevention of coronary heart disease and stroke in those aged >=70." British Medical Journal 330: 1306-1311.
- Regan, M. M. and R. D. Gelber (2005). "Predicting response to systematic treatments: Learning from the past to plan for the future." The Breast 14: 582-593.
- Steinbach, R., J. Green, et al. (2014). "Is ethnic density associated with risk of child pedestrian injury? A comparison of inter-census changes in ethnic populations and injury rates." Ethnicity & Health.
Annotations on common readings
Annotated additions by students
(In alphabetical order by author's name with contributor's initials and date at the end.)
Scientific American, July 15, 2007
Race in a Bottle
Drugmakers are eager to develop medicines targeted at ethnic groups, but so
far they have made poor choices based on unsound science
By Jonathan Kahn
In this article, Jonathan Kahn explores the growing appeal of race-specific drugs; unmasking a tale of noble motives being sullied by the lure of lucrative incentives. Khan contends that since the birth of Human Genome Projects the scientific community has labored to prevent the inappropriate use of biological knowledge garnered from advances in genetic research. Nonetheless, misconstrued findings continue to avow racial categorization as being a biologically endowed phenomenon (and not merely a social construct); thus purporting it to be pharmacogenomically exploitable at the population level. However, to the contrary, it is known that at the genetic level more variation is observed within populations than between populations.
Kahn acknowledges that science and commerce have been mutually responsible for the advancement of medicine. However, he posits that there have been striking instances of commerce bypassing science to influence policies favorable to its self-interest. Kahn proffers that FDA approval BiDil, dubbed as being the first “ethnic” drug; as it was intended to treat congestive heart failure in African Americans only, augured a new age of medicine. BiDil’s approval was touted as significant progress towards the actualization of personalized medicine (pharmaceuticals engineered to work with an individual's particular genetic makeup in the safest and most effective manner).
It is Kahn’s conjecture that the history of BiDil’s genesis exposes its ethnic drug designation as being pretense. The author argues that BiDil’s formulation and mechanism, at most, only speculatively represents clinical application of pharmacogenomics. Moreover, there is insufficient evidence that BiDil exhibits differential efficacy in African Americans compared to the general population. Kahn does not dispute the legitimacy of BiDil’s procured affects in the unrestricted patient population. However, censures the use of race as a means to surmounting regulatory hurdles, which in effect allowed BiDil to swindle commercial advantage in the pharmaceutical market place.
Kahn surmises, “most problematically, the patent award and FDA approval of BiDil have given the imprimatur of the federal government to using race as a genetic category (Kahn).” The author asserts that approval of this drug relayed to society the scientifically unsupported message that a subject population's race was a substantial biological variable in assessing safety and efficacy. Kahn reasons that the FDA's working assumption, albeit unstated, is that the white population is an appropriate default when assaying the effectiveness of drugs. The author proposes that, “the same assumption should apply when the trial population happens to be black. Otherwise, the FDA is implying that African-Americans are somehow less fully representative of humanity than whites are (Kahn).”
Kahn’s concludes that the BiDil fiasco illustrates the failure of researchers and regulators alike to value the biological limitation of racial categories. Kahn asserts that the imposition of race-specific labels on drugs in absentia of convincing scientific evidence of a genetic or biological basis for any observed racial differences in safety or efficacy engenders haphazard regulation, substandard medical treatment and a litany of other, arguably unintended, consequences. (SY)