Back pain results in tremendous disability and cost; therefore, understanding predisposing or protective factors deserves attention. Statins are widely used, but there is no consensus on whether statins are protective1 of or deleterious2–4 to musculoskeletal conditions. A previously published propensity score (PS)–matched study of statin users and nonusers found an association between statin use and increased risk of use-related injury and arthropathies.2 Scarce data exist on the association of statins with back pain.3,4 Because statins may increase vulnerability to myalgias and contribute to the myopathic component often experienced with back pain, the objective of this study was to examine the association of statin use with the risk of back disorder diagnoses. Our a priori hypothesis was that statin use would be associated with back disorders, including spondylosis and intervertebral disc disorders.
Methods
We retrieved health care data for patients enrolled in TRICARE, the health insurance system of the US Department of Defense, in the San Antonio military area from October 1, 2003, to March 1, 2012.5 Inclusion criteria were being older than 30 years and having at least 1 medical encounter during the baseline period (2 years before the index date) and the follow-up period (starting 90 days after the index date through March 1, 2012, except for nonusers who were subsequently prescribed statins and for whom follow-up was censored at the date they filled their statin prescription).5
Two treatment groups were identified: (1) statin users who recently received a first-time prescription for a statin and continued its use for 120 days or more and (2) statin nonusers who never used statins as well as statin users prior to being prescribed statins.5 Prevalent statin users were excluded. This study was approved by the institutional review boards of Brooke Army Medical Center and the VA North Texas Health System. Patient informed consent was waived by both institutional review boards because the data were deidentified before being forwarded to the investigators. Data analyses were performed between February 23, 2016, and August 8, 2016.
Our outcome was an occurrence of disease category 205 for back disorders (eg, spondylosis, intervertebral disc disorders, and other back problems), as listed in the Agency for Healthcare Research and Quality’s Clinical Classifications Software (based on the International Classification of Diseases, Ninth Revision, Clinical Modification codes).5
We created a PS using 115 baseline characteristics and matched treatment groups in a 1:1 ratio for the nearest neighbor (caliper of 0.01).5
Primary analyses used conditional logistic regression analysis to examine the odds ratio (OR) of outcomes in the PS-matched cohort. We also calculated the number needed to harm.6
Secondary analyses examined the OR of outcomes in the overall cohort (all patients identified before PS matching) and in several prespecified cohorts. Baseline characteristics for treatment groups were examined using χ2 test to compare categorical variables and 2-tailed t test for continuous variables. Standardized differences were also calculated. Comparisons of outcomes, using conditional logistic regression, were considered statistically significant at P < .05. Statistical analyses were performed using SPSS software version 23 (IBM).
Results
The overall cohort included 60 455 patients, of whom 28 831 (47.7%)were men and 31 624 (52.3%)were women with a mean (SD) age of 46.6 (12.2) years. TRICARE beneficiaries include approximately 17% active duty military, their families, and veterans. We matched 6728 statin users with 6728 nonusers, and selected baseline characteristics are listed in Table 1. Statin users took statins for a median (interquartile range [IQR]) of 3.7 (1.9–4.9) years. Among statin users, 425 935 prescriptions (72%) were for simvastatin.
Table 1.
Selected Baseline Characteristics of Propensity Score–Matched Cohort of Statin Nonusers and Usersa
| Baseline Characteristic | Statin Nonusers (n = 6728) | Statin Users (n = 6728) |
|---|---|---|
| Age, mean (SD), y | 52 (14) | 52 (14) |
| Women, No. (%) | 3168 (47.1) | 3154 (46.9) |
| Health care use | ||
| Follow-up duration, mean (SD), y | 4.0 (2.1) | 4.0 (1.6) |
| FY 2004–2006 | 5181 (77.0) | 5166 (76.8) |
| FY 2007–2009 | 1492 (22.2) | 1504 (22.4) |
| FY 2009-March 1, 2012 | 55 (0.8) | 58 (0.9) |
| During baseline period, mean (SD), No. | ||
| Outpatient medical encounters | 66 (105) | 65 (86) |
| Outpatient procedures | 40 (93) | 39 (70) |
| Social history, No. (%) | ||
| Smokingb | 1816 (27.0) | 1821 (27.1) |
| Alcohol abuse/dependence | 99 (1.5) | 92 (1.4) |
| Comorbid condition/disease, No. (%)c | ||
| Charlson comorbidity score, mean (SD) | 0.74 (1.33) | 0.73 (1.31) |
| Obesity or overweight | 1732 (25.7) | 1735 (25.8) |
| Diabetes | 1321 (19.6) | 1354 (20.1) |
| Hypertension | 3556 (52.9) | 3510 (52.2) |
| Coronary artery disease | 418 (6.2) | 439 (6.5) |
| Chronic kidney disease | 113 (1.7) | 121 (1.8) |
| Rheumatoid arthritis | 104 (1.5) | 110 (1.6) |
| Osteoarthritis and other nontraumatic joint disorder | 2714 (40.3) | 2653 (39.4) |
| Spondylosis, disc disorders, and other back problems | 2020 (30.0) | 1992 (29.6) |
| Use-related joint disorders (dislocations, sprains, and strains) | 1363 (20.3) | 1368 (20.3) |
| Osteoporosis | 299 (4.4) | 285 (4.2) |
| Rehabilitation care, fitting of prostheses, and adjustment of devices | 1193 (17.7) | 1179 (17.5) |
| Malignant neoplasm | 386 (5.7) | 389 (5.8) |
| Medications during baseline period, No. (%) | ||
| Oral hypoglycemics | 503 (7.5) | 489 (7.3) |
| Insulins | 166 (2.5) | 173 (2.6) |
| NSAIDs | 3941 (58.6) | 3902 (58.0) |
| Bisphosphonate | 483 (7.2) | 452 (6.7) |
| Systemic corticosteroid | 671 (10.0) | 671 (10.0) |
Abbreviations: FY, fiscal year; NSAIDs, nonsteroidal anti-inflammatory drugs.
Complete description of all baseline characteristics, including standardized mean differences, was previously published.5 Patients were matched on all baseline characteristics included in the propensity score.
Smoking as defined by the International Classification of Diseases, Ninth Revision, Clinical Modification codes 3051 and V1582.
Diagnoses as defined by the Agency for Healthcare Research and Quality’s Clinical Classifications Software disease categories (based on the International Classification of Diseases, Ninth Revision, Clinical Modification codes).
Statin users in the PS-matched cohort had a higher likelihood of back disorders (OR, 1.27; 95%CI, 1.19–1.36). The number needed to be exposed for an additional harm was 17.6 All secondary analyses showed similar results, including analyses of longer use and higher intensity of statin (Table 2).
Table 2.
Risk of Back Disorder Diagnoses Among Statin Users and Nonusers
| Analysis | No. of Users/Nonusers in Cohort | Diagnosed With Back Disorder, No. (%)
|
Odds Ratio (95% CI) | |
|---|---|---|---|---|
| Statin Users | Statin Nonusers | |||
| Primary analyses | ||||
| Propensity score-matched cohort | 6728/6728 | 3318 (49.3) | 2913 (43.3) | 1.27 (1.19–1.36)a |
| Secondary analyses | ||||
| Overall cohortb | 10 910/49 545 | 5022 (46.0) | 23 061 (46.5) | 1.30 (1.23–1.38)c |
| 1.26 (1.19–1.34)d | ||||
| Overall cohort restricted to statin users | ||||
| ≥2 y vs nonusers | 7007/49 545 | 3613 (51.6) | 23 061 (46.5) | 1.47 (1.39–1.56)c |
| ≥4 y vs nonusers | 3427/49 545 | 1898 (55.4) | 23 061 (46.5) | 1.59 (1.47–1.71)c |
| Overall cohort restricted to high-intensity statin users vs nonuserse | 2470/49 545 | 1216 (49.2) | 23 061 (46.5) | 1.47 (1.34–1.62)c |
| Nonobese cohortf | 5340/29 821 | 2163 (40.5) | 12 445 (41.7) | 1.28 (1.18–1.38)c |
| Healthy cohortg | 4563/35 783 | 2032 (44.5) | 15 895 (44.4) | 1.54 (1.42–1.66)c |
| Musculoskeletal diseases incident cohorth | 4592/25 802 | 1419 (30.9) | 9236 (35.8) | 1.29 (1.18–1.41)c |
Unadjusted odds ratio because patients were matched on all baseline characteristics.
Overall cohort included all patients who met study inclusion and exclusion criteria before propensity score matching.
Odds ratio with adjustment for propensity score.
Overall cohort with adjustment for propensity score, medications use, and undergoing revascularization procedures during the follow-up period. Medications were included that may be associated with increased body weight or musculoskeletal symptoms (eg, systemic corticosteroids, selective serotonin reuptake inhibitors, antipsychotic, bisphosphonate, hormone replacement therapy, testosterone, warfarin, and cytochrome P450 inhibitors). Invasive revascularization procedures (percutaneous coronary intervention, coronary artery bypass surgery, or peripheral revascularization procedure) were assumed to result in prolonged bed rest and hospitalization; hence, it may contribute to musculoskeletal symptoms.
High-intensity statin was defined by the 2013 guidelines of the American College of Cardiology/American Heart Association for cholesterol management, with modification to include simvastatin, 80mg, as high-intensity statin. High-intensity statin users used these medications for 120 days or more.
Nonobese cohort excluded patients diagnosed with obesity at baseline or at follow-up.
Healthy cohort excluded patients with any component of the Charlson comorbidity index or other severe comorbidities, as previously described.5
Musculoskeletal diseases incident cohort excluded patients who were diagnosed with back disorder, nontraumatic arthropathy, and use-related injury during the baseline period.
Discussion
In this study, statin use was associated with increased likelihood of back disorder diagnoses and a dose response to both dosage and duration. To our knowledge, this study is the first to report greater odds of back disorders among statin users compared with the odds of nonusers in a population with equal access to and the same cost of health care.
Few studies have examined the association of statins with back pain. In a cross-sectional study of statin users vs nonusers, statin users had higher adjusted OR of lower back pain (OR, 1.59; 95%CI, 1.04–2.44).4 In another study, among those without arthritis, statin users had a higher adjusted prevalence ratio of lower back pain.3
Limitations of this study include its retrospective nature and the use of Clinical Classifications Software codes. Because the study participants were TRICARE enrollees, these results may not be generalizable to a more sedentary population or to those without exposure to military or veteran experience.
Our results provide additional motivation to further investigate the overall influence of statin therapy on musculoskeletal health, specifically if prescribed for primary prevention in physically active individuals.
Acknowledgments
Funding/Support: This research was supported in part by the US Department of Veterans Affairs, Veterans Health Administration; by National Institutes of Health/National Center for Advancing Translational Sciences grants KL2TR001103 and UL1TR001105 from the UT Southwestern Center for Translational Medicine as well as by VA Health Services Research and Development Career Development Award 14-425 (Dr Makris); by grant K08 DK101602 from the National Institutes of Health (Dr Alvarez); and by grant R24 HS022418 from the Agency for Healthcare Research and Quality and the University of Texas Southwestern Center for Patient-Centered Outcomes Research (Dr Mortensen).
Footnotes
Conflict of Interest Disclosures: None reported.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed herein are those of the authors and do not reflect the official policy or position of the US Department of the Army, US Department of Defense, US Department of Veterans Affairs, or the US government. The authors are employees of the US government. This study was prepared as part of the authors’ official duties and, as such, there is no copyright to be transferred.
Author Contributions: Drs Makris and Mansi had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Alvarez, Mortensen, Makris, Mansi.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: Makris, Alvarez, Mortensen, Mansi.
Statistical analysis: Alvarez, Mortensen, Mansi.
Administrative, technical, or material support: Makris, Mortensen.
Study supervision: Mansi.
References
- 1.Kadam UT, Blagojevic M, Belcher J. Statin use and clinical osteoarthritis in the general population: a longitudinal study. J Gen Intern Med. 2013;28(7):943–949. doi: 10.1007/s11606-013-2382-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mansi I, Frei CR, Pugh MJ, Makris U, Mortensen EM. Statins and musculoskeletal conditions, arthropathies, and injuries [published correction appears in JAMA Intern Med. 2013;173(15):1477] JAMA Intern Med. 2013;173(14):1–10. doi: 10.1001/jamainternmed.2013.6184. [DOI] [PubMed] [Google Scholar]
- 3.Buettner C, Rippberger MJ, Smith JK, Leveille SG, Davis RB, Mittleman MA. Statin use and musculoskeletal pain among adults with and without arthritis. Am J Med. 2012;125(2):176–182. doi: 10.1016/j.amjmed.2011.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Buettner C, Davis RB, Leveille SG, Mittleman MA, Mukamal KJ. Prevalence of musculoskeletal pain and statin use. J Gen Intern Med. 2008;23(8):1182–1186. doi: 10.1007/s11606-008-0636-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mansi IA, Frei CR, Halm EA, Mortensen EM. Association of statins with diabetes mellitus and diabetic complications: role of confounders during follow-up. J Investig Med. 2017;65(1):32–42. doi: 10.1136/jim-2016-000218. [DOI] [PubMed] [Google Scholar]
- 6.Bender R, Blettner M. Calculating the “number needed to be exposed” with adjustment for confounding variables in epidemiological studies. J Clin Epidemiol. 2002;55(5):525–530. doi: 10.1016/s0895-4356(01)00510-8. [DOI] [PubMed] [Google Scholar]