In the article by Shalev et al,1 the authors improperly draw causal inferences from observational connections between statin therapy compliance and survival. Historically, associations of drug use and compliance with favorable health outcomes can depart from causal findings, not only in magnitude but also in direction (recall the hormone therapy "data" on cardiac events and dementia).

Lower cognitive function (whether or not promoted by worse health)2-3 as well as lower conscientiousness3-4 portend both lower statin therapy compliance3 and greater mortality.2, 4 In addition, unfavorable health factors predict risk of statin adverse effects,5 which are in turn linked to noncompliance,6 and do so disproportionately with high-potency statins.5 This forges a further link between lower compliance and higher mortality that is statin specific—and greater with high-dose statin use—but not attributable to statin therapy benefits.5 Creatine phosphokinase and liver function test results are inadequate to exclude statin adverse effects associated with noncompliance, which are commonly gastroenterological and neurological.5

Levothyroxine compliance is not a suitable control. Reasons to discontinue thyroid medication vs statins may differ radically. Those using thyroid medications may consider themselves to be treating an organ dysfunction rather than a risk factor and often wish to avoid feeling worse by discontinuing use of the drug (while those using statins may wish to avoid feeling worse by continuing treatment).

Most crucially, prior higher-quality, placebo-controlled trials have also shown a marked mortality advantage in patients compliant with lipid-lowering medications (approximately 40%), similar to that reported by Shalev and colleagues.1 However, an equal or greater mortality advantage (47%) was seen in those compliant with placebo7 (a reminder that not merely the size but direction of the inferred statin advantage is open to question). Thus, a relation of statin compliance to survival of the magnitude reported by Shalev and colleagues1 is expected, based on explanations wholly independent of statin benefits. Once again, association need not imply causation.

References

1. Shalev V, Chodick G, Silber H, Kokia E, Jan J, Heymann AD. Continuation of statin treatment and all-cause mortality: a population-based cohort study. Arch Intern Med. 2009;169(3):260-268.
2. Smits CH, Deeg DJ, Kriegsman DM, Schmand B. Cognitive functioning and health as determinants of mortality in an older population. Am J Epidemiol. 1999;150(9):978-986.
3. Stilley CS, Sereika S, Muldoon MF, Ryan CM, Dunbar-Jacob J. Psychological and cognitive function: predictors of adherence with cholesterol lowering treatment. Ann Behav Med. 2004;27(2):117-124.
4. Friedman HS, Tucker JS, Schwartz JE; et al. Childhood conscientiousness and longevity: health behaviors and cause of death. J Pers Soc Psychol. 1995;68(4):696-703.
5. Golomb BA, Evans MA. Statin adverse effects: a review of the literature and evidence for a mitochondrial mechanism. Am J Cardiovasc Drugs. 2008;8(6):373-418.
6. Bruckert E, Simonetta C, Giral P, CREOLE Study Team. Compliance with fluvastatin treatment characterization of the noncompliant population within a population of 3845 patients with hyperlipidemia. J Clin Epidemiol. 1999;52(6):589-594.
7. Coronary Drug Project Research Group. Influence of adherence to treatment and response of cholesterol on mortality in the coronary drug project. N Engl J Med. 1980;303(18):1038-1041.