PU570 Unit 4 Causal Reasoning Paper Example

Nine Considerations, Two of Them Weak and One Not Applicable: A Unit 4 Causal Reasoning Paper Testing Long-Term Fine Particulate Exposure and Cardiovascular Death Against Hill's 1965 List Without Treating It as a Checklist

[Student Name]

Purdue University Global

PU570: Chronic Disease Epidemiology

Unit 4 Causal Reasoning Paper

[Instructor Name]

[Date]

The paper is the course assignment's own; the exposure, the cohorts and every study cited are real.

The Association Being Tested

Long-term residence in areas with higher concentrations of fine particulate matter is associated with higher cardiovascular mortality. The cohort that opened this field followed residents of six cities and found mortality rising with long-term particulate concentration, with the most polluted city carrying roughly a 26 percent higher adjusted mortality rate than the least (Dockery et al., 1993). Later analysis in a far larger population found the association persisting below current regulatory limits (Di et al., 2017).

The Nine Considerations, One at a Time

1. Strength. Moderate rather than strong. Hazard ratios cluster near 1.05 to 1.15 per ten micrograms. Hill was explicit that a weak association does not rule out causation, and here the weakness is partly a feature of the exposure contrast, since nobody in these cohorts is unexposed.

2. Consistency. Strong. The association appears in cohorts in different countries, with different pollution sources, different analytic approaches and different confounding structures, which is the consideration doing most of the work in this case.

3. Specificity. Not applicable here, and the paper records that instead of bending the consideration to fit. Particulate exposure is associated with several outcomes, which is what a general inflammatory and vascular mechanism predicts. Hill himself treated specificity as the weakest of his considerations.

4. Temporality. Satisfied by design in the cohort studies, where exposure is assigned from residence before death. The honest qualification is that long-term exposure estimates are often assigned from recent monitoring and extrapolated backward.

5. Biological gradient. Strong. Risk rises across exposure categories, and the extension of the gradient below regulatory limits is the most consequential recent finding (Di et al., 2017).

6. Plausibility. Strong. Inhaled fine particles promote systemic inflammation, endothelial dysfunction, autonomic imbalance and thrombosis, mechanisms set out in the cardiology literature's own statement on the subject (Brook et al., 2010).

7. Coherence. Strong. The association fits the short-term literature, where daily increases in particulate levels are followed by excess cardiovascular events within days, and fits the declines in mortality observed where concentrations fell.

8. Experimental evidence. Partial. No trial randomizes people to polluted air. Controlled exposure studies in humans show acute vascular changes, and intervention evidence comes from natural experiments where a plant closed or fuel changed, which is weaker than a trial and stronger than nothing.

9. Analogy. Weak as evidence, available as reasoning: tobacco smoke is a particulate exposure with established cardiovascular effects, which makes the claim unsurprising without supporting it.

Specificity does not apply here, and saying so is better than converting a general mechanism into a point in favor.

Why Two of the Nine Carry the Argument

Not every consideration deserves equal weight in a given case, and in this one consistency and the biological gradient do most of the work. Consistency matters here because the confounding structure differs between settings: the people exposed to higher concentrations in one country are not socially similar to those exposed in another, so an association that survives across them is harder to explain by social position alone. The gradient matters because it was extended downward after the original studies, into concentrations below the regulatory limits, and a bias that produced a spurious association at high exposure would have no obvious reason to reproduce it at low exposure in a different population (Di et al., 2017). The remaining considerations are either satisfied by design, as temporality is, or supportive rather than probative, as plausibility and analogy are. A paper that gave all nine equal weight would conclude the same thing for weaker reasons.

What Hill Actually Proposed, and the Misuse of It

Hill offered nine viewpoints to help decide whether an association is more likely causal than not, and warned explicitly against treating them as rules or requirements (Hill, 1965). Two misuses are common in student papers and in published ones. The first is scoring: counting how many considerations are met and declaring a majority. The second is requiring all nine, which would rule out causation for most environmental exposures including smoking. The useful question is which considerations bear on the particular doubt a skeptic would raise.

The Doubt a Skeptic Would Raise Here

Confounding by place. People living in higher-pollution areas differ in income, occupation, smoking, access to care and housing quality, and every one of those predicts cardiovascular death. This is the real argument against causation and no number of satisfied considerations answers it directly. What answers it partly: the association persists after adjustment for individual smoking and income in cohorts that measured them; it appears within cities as well as between them; and it appears in analyses using exposure changes over time within the same population, where fixed characteristics of place cannot explain it.

Where the Evidence Is Weakest

Two places. Exposure assignment is ecological: everybody at an address gets the modeled concentration, which ignores time indoors, commuting and occupational exposure, and that error is likely to attenuate rather than create the association. And the composition question is open, since fine particulate matter from traffic, coal combustion and wildfire smoke differ chemically, and the cohorts estimate an average effect of a mixture that varies by place and over time.

What a Reader Should Not Take From This Paper

Two conclusions do not follow. First, that any individual's heart disease was caused by air pollution; the evidence is about rates in populations, and an attributable fraction at population level says nothing identifiable about one person's case. Second, that the exposure is the largest cardiovascular risk factor in any population; it is a modest relative effect applied to an enormous number of exposed people, which is what makes it important for policy and unimportant for clinical advice to a given patient. Keeping those two statements separate is the practical value of having worked through the considerations rather than reading the headline figure.

The Conclusion the Evidence Supports

The association is more likely causal than not, carried mainly by consistency across settings, a gradient that extends below regulatory limits, and a mechanism with independent support. The claim should be stated at that strength rather than as established, and the paper separates two things a reader will otherwise merge: that long-term particulate exposure causes cardiovascular deaths is well supported, while the number of deaths attributable to a specific reduction in a specific place rests on modeling assumptions that deserve their own scrutiny. Reasoning about causation is a judgment under uncertainty rather than a verdict, which is the point the methodological literature keeps making and the one Hill's paper opened with (Rothman & Greenland, 2005).

References

Brook, R. D., Rajagopalan, S., Pope, C. A., Brook, J. R., Bhatnagar, A., Diez-Roux, A. V., Holguin, F., Hong, Y., Luepker, R. V., Mittleman, M. A., Peters, A., Siscovick, D., Smith, S. C., Whitsel, L., & Kaufman, J. D. (2010). Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association. Circulation, 121(21), 2331-2378. https://doi.org/10.1161/CIR.0b013e3181dbece1

Di, Q., Wang, Y., Zanobetti, A., Wang, Y., Koutrakis, P., Choirat, C., Dominici, F., & Schwartz, J. D. (2017). Air pollution and mortality in the Medicare population. The New England Journal of Medicine, 376(26), 2513-2522. https://doi.org/10.1056/NEJMoa1702747

Dockery, D. W., Pope, C. A., Xu, X., Spengler, J. D., Ware, J. H., Fay, M. E., Ferris, B. G., & Speizer, F. E. (1993). An association between air pollution and mortality in six U.S. cities. The New England Journal of Medicine, 329(24), 1753-1759. https://doi.org/10.1056/NEJM199312093292401

Hill, A. B. (1965). The environment and disease: Association or causation? Proceedings of the Royal Society of Medicine, 58(5), 295-300. https://doi.org/10.1177/003591576505800503

Rothman, K. J., & Greenland, S. (2005). Causation and causal inference in epidemiology. American Journal of Public Health, 95(S1), S144-S150. https://doi.org/10.2105/AJPH.2004.059204