Copyright 2014 American Medical Association. All rights reserved. Conflict of Interest Disclosures: None reported. 1. Ross MA, Aurora T, Graff L, et al. State of the art: emergency department observation units. Crit Pathw Cardiol. 2012;11(3):128-138. 2. Sheehy AM, Graf B, Gangireddy S, et al. Hospitalized but not admitted: characteristics of patients with “observation status” at an academic medical center [published online July 8, 2013]. JAMA Intern Med. doi:10.1001 /jamainternmed.2013.8185. 3. Ross MA, Compton S, Medado P, Fitzgerald M, Kilanowski P, O’Neil BJ. An emergency department diagnostic protocol for patients with transient ischemic attack: a randomized controlled trial. Ann Emerg Med. 2007;50(2):109-119. 4. Decker WW, Smars PA, Vaidyanathan L, et al. A prospective, randomized trial of an emergency department observation unit for acute onset atrial fibrillation. Ann Emerg Med. 2008;52(4):322-328. 5. Department of Health and Human Services Office of Inspector General. Hospitals’ use of observation stays and short inpatient stays for medicare beneficiaries (report OEI-02-12-00040). July 29, 2013. http://oig.hhs.gov/oei /reports/oei-02-12-00040.asp. Accessed November 9, 2013. Statins and Musculoskeletal Adverse Events To the Editor Mansi and colleagues 1 have published an analy- sis of musculoskeletal conditions, arthropathies, and injuries among users and nonusers of statins. They reported higher ad- justed odds ratios (ORs) for statin users in all outcome groups. Are their findings clinically significant? The OR is (P s /1 - P s )/(P n /1 - P n ), where P s is the probabil- ity of adverse outcomes in statin users and P n is the probabil- ity in nonusers. The OR is not easy to interpret. More desir- able for interpretative purposes is the relative risk (RR), P s / P n . It may not be widely known to nonepidemiologists that the OR is a good approximation to the RR when the outcome is “rare” (<10%), but that the OR greatly overestimates the RR when the outcome is common. 2 Table 4 in the article by Mansi et al 1 showed that 86.9% of statin users and 84.8% of nonus- ers had an adverse event (there is an error in the table, the num- ber among nonusers should be 5905). This difference in inci- dence of 2% translates into an OR of 1.19, which “looks” much larger. Zhang and Yu 3 have published a method to convert lo- gistic regression ORs into RRs. I applied this method and cal- culated that the RR for “all musculoskeletal diseases” among statin users, compared with nonusers was 1.02 (95% CI, 1.01- 1.04). The risk of adverse events among statin users was thus 2% higher than among nonusers. When presented in this manner, one wonders whether this small risk difference might be attributable to residual confounding not eliminated by the propensity score matching and whether this differ- ence, if real, is clinically significant. It is true that, in the large population of statin users, small differences in risk translate into large numbers of affected individuals, but I think that stronger evidence is required before public health policy can be modified. To their credit, the authors have called for additional re- search. Regrettably, this study was widely covered in the me- dia and the impression was given that the presence of a high risk of adverse events was established. An important maxim in medicine is “first, do no harm.” An unintended conse- quence of this publication might be that statin users will stop their medication, placing them at risk of serious circulatory events. I think that it would be appropriate for the Journal’s editor to advance the appropriate cautions. Murray M. Finkelstein, PhD, MD Author Affiliation: Department of Family and Community Medicine, University of Toronto, Toronto, Ontario, Canada. Corresponding Author: Murray Finkelstein, PhD, MD, Granovsky-Gluskin Family Medicine Centre, Mt Sinai Hospital, Department of Family and Community Medicine, University of Toronto, 60 Murray St, Fourth Floor, Toronto, ON M5T 3L9, Canada (murray.finkelstein@utoronto.ca). Conflict of Interest Disclosures: None reported. 1. Mansi I, Frei CR, Pugh MJ, Makris U, Mortensen EM. Statins and musculoskeletal conditions, arthropathies, and injuries. JAMA Intern Med. 2013;173(14):1318-1326. 2. Knol MJ, Le Cessie S, Algra A, Vandenbroucke JP, Groenwold RH. Overestimation of risk ratios by odds ratios in trials and cohort studies: alternatives to logistic regression. CMAJ. 2012;184(8):895-899. 3. Zhang J, Yu KF. What’s the relative risk? a method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280(19):1690-1691. To the Editor We read with interest the study of Mansi et al 1 that reported a higher likelihood of musculoskeletal adverse events among statin users. 1 Strengths of this retrospective study are the propensity-scored and subsequent analysis, the com- plete follow-up, and the access to the same health care sys- tem. Surprisingly, the authors reported a high rate of muscu- loskeletal diseases during follow-up (both groups >80%), with a small but significant difference toward statin users. These unusual rates could be related to the type of the population (mainly active-duty soldiers and veterans) and/or the use of International Classification of Diseases, Ninth Revision, Clini- cal Modification codes that lack specificity and do not inform on the severity of the symptoms nor the need to stop therapy. The authors state that these findings are concerning ow- ing to the use of statins in primary prevention in young and/or physically active patients. However, additional data analyses are needed to support such a general statement. For ex- ample, the reported adverse effects were obtained by simply grouping subjects into statin users or nonusers. This classifi- cation, correct from a mechanism of action perspective, is in- sensitive to differences of the safety profile of the 6 statins ana- lyzed. This information would be needed to determine what effect statin type, dose, or other pharmacological properties has on the incidence of adverse effects, as we learned with cerivastatin. Increasing evidence has shown that musculoskeletal ad- verse effects depend on the dose, potency, and lipophilicity of the statin. 2,3 Recently, the Food and Drug Administration rec- ommended, after analyzing the Study of the Effectiveness of Additional Reductions in Cholesterol and Homocysteine (SEARCH) trial and others, avoiding high-dose simvastatin treatment and required an addition to its label new contrain- dications (gemfibrozil) and interactions (amlodipine). 4 Since 70% used simvastatin and 34% used maximal dose in this co- hort, it would be important to test if the difference rates of ad- verse events are driven by patients on high-dose simvastatin, which is probably overrepresented according to current clini- cal practice. Letters 302 JAMA Internal Medicine February 2014 Volume 174, Number 2 jamainternalmedicine.com Copyright 2014 American Medical Association. All rights reserved. Downloaded From: http://archinte.jamanetwork.com/ by a Mount Sinai Hospital User on 05/06/2014