Abstract
Study Objective
To evaluate the association between statin therapy and risk of psychological disorders including schizophrenia, psychosis, major depression, and bipolar disorder in a military population.
Design
Retrospective, observational, population-based, propensity score-matched, cohort study.
Setting
Database of a patient population enrolled in the San Antonio Military Multi-Market Area as Tricare Prime or Plus.
Patients
Medical records were reviewed from 46249 patients aged 30 to 85 years who were continuously enrolled in the San Antonio Military Multi-Market Area as Tricare Prime or Plus from October 1, 2003 to March 1, 2010. Data were obtained from the Military Health System Management Analysis and Reporting Tool (M2). Based on medication fills during fiscal year 2005, patients were stratified as statin users (N=13626 who received at least 90-days supply of statin) or non-users (N=32623 who never received a statin during the study period). A propensity score-matched cohort of 6972 statin users and 6972 non-users from this population was created.
Measurements and Main Results
The occurrence of psychological disorders between October 1, 2005 and March 1, 2010 was determined using pre-specified groups of International Classification of Diseases, 9th Revision, Clinical Modification codes: 1) Psych1: schizophrenia, schizoaffective disorders, and other psychosis; 2) Psych2: major depression and bipolar disorder; 3) Psych3: all psychological disorders as identified by the Agency for Health Research and Quality-Clinical Classifications (except for categories of childhood or developmental psychiatric disorders). Among matched pairs of statin users and non-users, the odds ratios and 95% confidence intervals (OR, 95%CI) were as follows: Psych1 (0.9, 0.75–1.05), Psych2 (1.02, 0.94–1.11), and Psych3 (1.02, 0.96–1.1).
Conclusion
The risk of developing psychological disorders was similar in this cohort of propensity score-matched statin users and non-users.
Keywords: Statin, schizophrenia, depression, adverse events, psychological injury, observational study
Introduction
Despite their effectiveness in lowering cardiovascular morbidity and mortality, hydroxyl methyl glutaryl coenzyme A reductase inhibitors (statins) have been associated with an extremely wide variety of adverse events.1–6 Pharmacovigilance groups from different parts of the world have implicated statins as a potential cause of severe psychological adverse events. The New Zealand Center for Adverse Reaction Monitoring (CARM) reported the association of 285 cases of psychological adverse events with statins.5 Another review of spontaneous adverse events in an Italian database identified 71 psychological adverse events.7 Adverse events included insomnia, somnolence, agitation, confusion, hallucinations, mood disorders, cognitive disorders, and perception disorders.5, 7 Case reports, case series, and small randomized controlled studies have linked statin use to the increased risk of psychological disorders, such as violence or depression.8–11 The recent FDA warning of cognitive disorders associated with statins has raised additional concerns.12
The objective of this study was to determine if statins were associated with an increased risk of psychological disorders, including schizophrenia, psychosis, major depression, and bipolar disorder in a military population that requires a high degree of psychological preparedness.
Methods
This study was approved by the Institutional Review Board at the Brooke Army Medical Center (BAMC), San Antonio, Texas and the University of Texas Health Science Center, San Antonio, Texas.
This was a retrospective cohort study of patients aged 30 to 85 years who were enrolled in the San Antonio Military Multi-Market Area as Tricare Prime or Plus from October 1, 2003 to March 1, 2010. Using the Military Health System Management Analysis and Reporting Tool (M2), we retrieved all inpatient and outpatient medical encounters, diagnoses, and medication fill histories. M2 is a powerful tool that has been used in administrative and leadership decision making as well as in research.13–17
The M2 database encompasses the full spectrum of clinical care: outpatient electronic medical records, inpatient electronic medical records, medical benefits claims data, laboratory data, and pharmacy data. Outpatient medical records contain all outpatient service activities. The inpatient electronic medical record is used to document all inpatient service activities. Medical benefit claims data contain services and medications from providers outside the military facilities. Laboratory data include all laboratory investigation results performed within the military system. Finally, pharmacy data include the medication issue date, product strength, quantity dispensed, and days supply for all medications dispensed at military facilities.
Patient selection
The study duration was divided into a baseline period and a follow-up period. The baseline period started on October 1, 2003 and concluded on September 30, 2005 and was used to identify patients’ baseline characteristics. The follow-up period began on October 1, 2005 and ended on March 1, 2010 and was used to identify the occurrence of outcome events. All patients were continuously enrolled in the system throughout the study period. We identified two groups of patients. Statin users were patients who received and filled a prescription for a statin medication for at least 90 days between October 1, 2004 and September 30, 2005. Non-users were patients who did not receive a statin at any time from October 1, 2003 to March 1, 2010.
Inclusion criteria
All patients aged 30 to 85 years in the M2 database enrolled in Tricare Prime or Plus in the San Antonio Multi-Market Area were eligible. Tricare is the health care program serving active duty service members, National Guard and Reserve members, retirees, and their families. Eligible patients had at least one outpatient medical encounter during the baseline period (from October 1, 2003 to September 30, 2005), one outpatient medical encounter during the follow-up period (October 1, 2005 to March 1, 2010), and had received at least one prescription medication during the study baseline period.
Exclusion criteria
Burn patients as identified by the Agency for Health Research (AHRQ) Quality-Clinical Classifications Software (category 240) 18 and trauma patients as identified by International Classification of Diseases, 9th Revision, Clinical Modification [ICD-9-CM] codes compiled from the ICD-9 manual and previous publications were not included.19, 20 Due to the mechanism of injury, trauma and burn patients may have comorbidities and outcomes different from the general population,21, 22 and the value of statins is unclear for these patients.23–26 Patients who were newly started on statins after September 30, 2005 (end of baseline period) were not eligible because the follow-up period would fall short of 4.5 years.
Outcome measures
An event was defined as the occurrence of an ICD-9-CM code during the follow-up period, as defined below, in either the inpatient or outpatient setting. We utilized pre-specified diagnosis groups to define psychological disorders (Appendix 1). Briefly, these diagnostic groups were Psych1 (schizophrenia/other psychosis), Psych2 (depression/bipolar disorder/post-traumatic stress disorder [PTSD]), and Psych3 (all psychological disorders). The Psych1 group included patients with schizophrenia, schizoaffective disorders, and other psychosis as identified in VACS (Veterans Aging Cohort Study)27, except for the codes of “developmental diseases and psychosis with origin specific to childhood” because this group is unlikely to be related to statin use. The Psych2 group included patients with major depression, bipolar disorder, and post-traumatic stress disorder (PTSD) as identified in VACS.27 The Psych3 group consisted of patients with psychological disorders as identified by AHRQ-Clinical Classifications Software, except for categories of childhood or developmental psychiatric disorders (categories 654 and 655). Patients in the Psych3 group had diagnoses of adjustment disorder (category 650), anxiety disorder (category 651), attention-deficit, conduct, or disruptive behavior disorder (category 652), impulse control disorder (category 656), mood disorder (category 657), personality disorder (category 658), schizophrenia or other psychotic disorders (category 659), and miscellaneous disorders (category 670).
Data Analysis
Comorbidities were characterized using the Charlson comorbidity score (CCS) described by Deyo et al.28 We created a propensity score-matched cohort of statin users and non-users using age, gender, comorbidities (as defined by the Deyo method28), total CCS score, and histories of obesity, alcohol dependence/abuse, illicit drug use, cigarette smoking, glaucoma, vision defects/blindness, healthcare utilization (number of outpatient and inpatient medical encounters during the baseline and follow-up periods), and medication use (e.g., beta-blockers, diuretics, calcium channel blockers, non-statin lipid lowering drugs, angiotensin-receptor blockers/angiotensin converting enzyme inhibitors, oral hypoglycemic agents, medications that inhibit the Cytochrome P450 system as identified in a recent FDA warning,29 aspirin, non-steroidal anti-inflammatory drugs, selective serotonin reuptake inhibitors, systemic corticosteroids, antipsychotics, sedatives, and tricyclic antidepressants).
The primary analysis compared the occurrence of psychological disorders in the statin user group to the non-user group in the propensity score-matched cohort. Two sets of secondary analyses were conducted using different patient cohorts and different durations of statin use. For the cohort analysis, we adjusted for potential confounders and assessed the risk of psychological disorders for the all-patients cohort (all patients that met the study criteria), the primary prevention cohort (patients with no CCS comorbidity [CCS = 0]; statins in this subgroup were presumably used for primary prevention), and the no baseline psychological comorbidity cohort (patients who did not have one of the psychological disorders of interest during the baseline period). For the duration of statin use analysis, we assessed the risk of psychological disorders in statin users who continued using statins for ≥1 year and statin users who continued using statins for ≥2 years in comparison to non-users.
Statistical analysis
Baseline characteristics of statin users and non-users were compared using appropriate two-way tests (i.e., chi-square for categorical variables and Student’s t test for continuous variables). Continuous variables were summarized by mean and standard deviation. Comparisons were considered statistically significant if the calculated p-value was less than an alpha level of 0.05.
Propensity score matching
Logistic regression was used to create the propensity scores and test the balance of covariates in our models using the routines developed by Becker and Ichino.30 Using the technique developed by Leuven and Sianesi, we conducted nearest number matching with a caliper of 0.001.31
Finally, outcomes among statin users and non-users were compared using logistic regression analysis. Covariates for the secondary analyses are listed in Table 4. Statistical analyses were performed using STATA 12 (StataCorp. 2011. Stata Statistical Software: Release 12. College Station, TX: StataCorp.) and SPSS statistical software version 19 (IBM Corp. Released 2010. IBM SPSS Statistics for Windows. Version 19.0. Armonk, NY: IBM Corp).
Table 4.
Baseline characteristics of statin users and non-users in patients with no psychological disorders* during the baseline period
| Statin-users (n =11458) | Non-users (n =28861) | p-value | |
|---|---|---|---|
| Age in years: mean ± SD | 60.3 ±12.2 | 44.7 ±11.2 | < 0.0001 |
| Male gender: n (%) | 7077 (61.8) | 13419 (46.5) | < 0.0001 |
| Comorbidities in baseline period: | |||
| Acute myocardial infarction*: n (%) | 616 (5.4) | 90 (0.3) | < 0.0001 |
| Congestive heart failure*: n (%) | 560 (4.9) | 119 (0.4) | < 0.0001 |
| Peripheral vascular disease*: n (%) | 723 (6.3) | 151 (0.5) | < 0.0001 |
| Cerebrovascular disease*: n (%) | 390 (3.4) | 148 (0.5) | < 0.0001 |
| Dementia*: n (%) | 12 (0.1) | 10 (<0.01) | < 0.009 |
| Chronic obstructive lung disease*: n (%) | 1583 (13.8) | 2026 (7) | < 0.0001 |
| Rheumatologic diseases*: n (%) | 233 (2) | 362 (1.3) | < 0.0001 |
| Peptic ulcer disease*: n (%) | 163 (0.4) | 211 (0.7) | < 0.0001 |
| Mild liver disease*: n (%) | 34 (0.3) | 81 (0.2) | < 0.4 |
| Diabetes mellitus*: n (%) | 3641 (31.8) | 690 (2.4) | < 0.0001 |
| Diabetes mellitus with complications*: n (%) | 1374 (12) | 147 (0.5) | < 0.0001 |
| Hemiplegia/paraplegia*: n (%) | 27 (0.2) | 15 (0.1) | < 0.0001 |
| Renal disease*: n (%) | 388 (3.4) | 88 (0.3) | < 0.0001 |
| Malignancy*: n (%) | 829 (7.2) | 917 (3.2) | < 0.0001 |
| Liver disease (moderate/severe)*: n (%) | 7 (0.1) | 28 (0.1) | 0.2 |
| Metastatic neoplasm*: n (%) | 32 (0.3) | 75 (0.3) | 0.4 |
| HIV *: n (%) | 11 (0.1) | 31 (0.1) | 0.4 |
| Charlson Comorbidity Total Score: mean ± SD | 1.18 ± 1.58 | 0.24 ± 0.76 | < 0.0001 |
| Illicit drug use: n (%) | 17 (0.1) | 5 (<0.01) | 0.4 |
| Alcohol abuse/dependence: n (%) | 96 (0.8) | 142 (0.5) | < 0.0001 |
| Smoking: n (%) | 966 (8.4) | 1554 (5.4) | < 0.0001 |
| Obesity: n (%) | 1919 (16.7) | 2641 (9.2) | < 0.0001 |
| Vision defects/blindness | 5198 (45.4) | 11725 (40.6) | 0.001 |
| Number of outpatient medical encounters during baseline period: mean ± SD | 36.9 ± 39.7 | 20.4 ± 23.3 | < 0.0001 |
| Number of inpatient admissions during baseline period: mean ± SD | 0.37 ± 0.89 | 0.15 ± 0.49 | < 0.0001 |
| Medications: | |||
| Beta-blocker: n (%) | 3241 (28.3) | 1766 (6.1) | < 0.0001 |
| Diuretic: n (%) | 4238 (37) | 2880 (10) | < 0.0001 |
| Calcium channel Blocker: n (%) | 2892 (25.2) | 1387 (4.8) | < 0.0001 |
| Non-statin lipid lowering drugs: n (%) | 1952 (17) | 485 (1.7) | < 0.0001 |
| ACE/ARB: n (%) | 6682 (58.3) | 2943 (10.2) | < 0.0001 |
| Oral Hypoglycemic: n (%) | 2317 20.2) | 299 (1) | < 0.0001 |
| Cytochrome p 450: n (%) | 1205 (10.5) | 1216 (4.2) | < 0.0001 |
| Aspirin: n (%) | 6091 (53.2) | 2244 (7.8) | < 0.0001 |
| NSAID: n (%) | 6220 (54.3 5) | 17642 (61.1) | < 0.0001 |
| SSRI: n (%) | 934 (8.2) | 1722 (6) | < 0.0001 |
| Systemic corticosteroid: n (%) | 409 (3.6) | 1100 (3.8) | < 0.1 |
| Antipsychotic: n (%) | 29 (0.3) | 44 (0.2) | 0.02 |
| Sedatives: n (%) | 1890 (16.5) | 3853 (13.4) | < 0.0001 |
| Tricyclic antidepressants: n (%) | 26 (0.2) | 39 (0.1) | 0.02 |
ACE/ARB: Angiotensin-receptor blockers & angiotensin converting enzyme inhibitors; NSAID: non-steroidal anti-inflammatory drugs; Cytochrome p 450: medications that inhibit the Cytochrome p450 system as identified in a recent FDA warning; 29 SSRI: selective serotonin reuptake inhibitors
Schizophrenia/other psychosis, mood disorders, depression, and bipolar disorder as detailed in Appendix 1
Results
Of the 60891 patients who met the inclusion criteria, 14642 were excluded (2124 were burn or trauma patients, 10476 received statins after September 30, 2005, and 2042 did not receive a medication during the baseline period). The study population consisted of 46249 patients who were statin users (n=13626) and 32623 non-users (n=32623).
The mean ± standard deviation (SD) cumulative duration of statin use among statin users was 1695 ± 662 days, with a median (interquartile range) of 1860 (1260–2220) days. Approximately, 34% of statin users had been prescribed intensive statin therapy, defined as simvastatin 80 mg, atorvastatin 80 mg, pravastatin 80 mg, or rosuvastatin 40 mg. During the study period, 73.5% of statin prescriptions were simvastatin, 17.4% atorvastatin, 7% pravastatin, 1.7% rosuvastatin, and 0.24% fluvastatin or lovastatin.
Table 1 describes characteristics of the all-patients cohort. Statin users were older, more likely to be male, had more comorbidities, were prescribed more medications, and used the healthcare system more frequently than non-users. Importantly, there were no significant differences in baseline characteristics between statin users and non-users after the propensity score matching (primary analysis) (Table 2).
Table 1.
Baseline demographic features, disease characteristics, and outcomes
| Statin users (n = 13626) | Non-users (n = 32623) | p-value | |
|---|---|---|---|
| Age in years: mean ± SD | 60.4 ±12.3 | 44.8 ± 11.3 | < 0.0001 |
| Male gender: n (%) | 7947 (58.3) | 14263 (43.7) | < 0.0001 |
| Comorbidities in baseline period: | |||
| Acute myocardial infarction*: n (%) | 797 (5.8) | 120 (0.4) | < 0.0001 |
| Congestive heart failure*: n (%) | 746 (5.5) | 163 (0.5) | < 0.0001 |
| Peripheral vascular disease*: n (%) | 859 (6.3) | 189 (0.6) | < 0.0001 |
| Cerebrovascular disease*: n (%) | 553 (4) | 226 (0.7) | < 0.0001 |
| Dementia*: n (%) | 58 (0.4) | 44 (0.1) | < 0.0001 |
| Chronic obstructive lung disease*: n (%) | 2065 (15.2) | 2464 (7.6) | < 0.0001 |
| Rheumatologic diseases*: n (%) | 290 (2.1) | 471 (1.4) | < 0.0001 |
| Peptic ulcer disease*: n (%) | 220 (1.6) | 263 (0.8) | < 0.0001 |
| Mild liver disease*: n (%) | 48 (0.4) | 117 (0.4) | 0.5 |
| Diabetes mellitus*: n (%) | 4386 (32.2) | 858 (2.6) | < 0.0001 |
| Diabetes mellitus with complications*: n (%) | 1663 (12.2) | 178 (0.5) | < 0.0001 |
| Hemiplegia/paraplegia*: n (%) | 49 (0.4) | 27 (0.1) | < 0.0001 |
| Renal disease*: n (%) | 471 (3.5) | 117 (0.4) | < 0.0001 |
| Malignancy*: n (%) | 1010 (7.4) | 1101 (3.4) | < 0.0001 |
| Liver disease (moderate/severe)*: n (%) | 8 (0.1) | 41 (0.1) | 0.03 |
| Metastatic neoplasm*: n (%) | 48 (0.4) | 95 (0.3) | 0.17 |
| HIV *: n (%) | 13 (0.1) | 39 (0.1) | 0.3 |
| Charlson Comorbidity total score: mean ±SD | 1.2 ±1.6 | 0.3 ±0.8 | < 0.0001 |
| Psychological disorders at baseline **: n (%) | 2168 (15.9) | 3762 (11.5) | < 0.0001 |
| Illicit drug use: n (%) | 20 (0.1) | 62 (0.2) | 0.35 |
| Alcohol abuse/dependence: n (%) | 133 (1) | 237 (0.7) | .007 |
| Smoking: n (%) | 1226 (9.0) | 1903 (5.8) | < 0.0001 |
| Obesity: n (%) | 2396 (17.6) | 3259 (10.0) | < 0.0001 |
| Vision defects/blindness | 6247 (45.8) | 13318 (40.8) | < 0.0001 |
| Mean HDL-C in baseline period: mean ± SD† | 52.5 ±14.5 | 58.5 ±17.8 | < 0.0001 |
| Mean HDL-C in follow-up period : mean ± SD† | 51.1 ±13.6 | 57 ±16.5 | < 0.0001 |
| Mean LDL-C in baseline period: mean ± SD† | 105.1 ±33.9 | 111 ±27.9 | < 0.0001 |
| Mean LDL-C in follow-up period: mean ± SD† | 98 ±31.3 | 112.5 ±27.4 | < 0.0001 |
| Mean TG in baseline period: mean ± SD† | 156.2 ±97.6 | 120.1±84.2 | < 0.0001 |
| Mean TG in follow-up period: mean ± SD† | 143.3 ±78.7 | 114.7 ±74 | < 0.0001 |
| Number of outpatient medical encounters during baseline period: mean ± SD | 40.9 ±44.7 | 23 ±31.8 | < 0.0001 |
| Number of inpatient admissions during baseline period: mean ± SD | 0.44 ±1.0 | 0.18 ±0.6 | < 0.0001 |
| Number of outpatient medical encounters during follow-up period: mean ± SD | 118.8 ±149 | 63.5 ±79 | < 0.0001 |
| Number of inpatient admissions during follow-up period: mean ± SD | 2.87 ±3.06 | 1.87 ± 1.94 | < 0.0001 |
| Medications: | |||
| Beta-blocker: n (%) | 3911 (28.7) | 2167 (6.6) | < 0.0001 |
| Diuretic: n (%) | 5117 (37.6) | 3405 (10.4) | < 0.0001 |
| Calcium channel Blocker: n (%) | 3514 (25.8) | 1643 (5.0) | < 0.0001 |
| Non-statin lipid lowering drugs: n (%) | 2321 (17.0) | 573 (1.8) | < 0.0001 |
| ACE/ARB: n (%) | 7982 (58.6) | 3466 (10.6) | < 0.0001 |
| Oral Hypoglycemic: n (%) | 2820 (20.7) | 383 (1.2) | < 0.0001 |
| Cytochrome p 450: n (%) | 1464 (10.7) | 1404 (4.3) | < 0.0001 |
| Aspirin: n (%) | 7274 (53.4) | 2659 (8.2) | < 0.0001 |
| NSAID: n (%) | 7568 (55.5) | 20136(61.7) | < 0.0001 |
| SSRI: n (%) | 2513 (18.4) | 4301 (13.2) | < 0.0001 |
| Systemic corticosteroid: n (%) | 532 (3.9) | 1369 (4.2) | 0.08 |
| Antipsychotic: n (%) | 180 (1.3) | 322 (1.0) | 0.002 |
| Sedatives: n (%) | 2862 (21.0) | 5430(16.6) | < 0.0001 |
| Tricyclic antidepressants: n (%) | 35 (0.3) | 58 (0.2) | 0.09 |
| Outcomes | |||
|
| |||
| Psych1 (schizophrenia/other psychosis): n (%) | 767 (1.7) | 533 (1.2) | < 0.0001 |
|
| |||
| Psych2 (depression / bipolar disorder): n (%) | 3105 (6.7) | 6375 (13.8) | < 0.0001 |
|
| |||
| Psych3 (all psychological disorders): n (%) | 11933 (25.8) | 5691 (12.3) | < 0.0001 |
ACE/ARB: Angiotensin-receptor blockers & angiotensin converting enzyme inhibitors; Cytochrome p 450: medications that inhibit the Cytochrome p450 system as identified in a recent FDA warning;29 HDL-C: High-density lipoprotein cholesterol in mg/dL; LDL-C: Low-density lipoprotein cholesterol in mg/dL; NSAID: non-steroidal anti-inflammatory drugs; SSRI: Selective serotonin reuptake inhibitors; SSRI: selective serotonin reuptake inhibitors; TG: Triglycerides in mg/dL.
Diagnosis is based on ICD-9-CM codes as identified in Deyo method for applying the Charlson comorbidity score.28
Definition is detailed in Appendix 1.
Lipid measurements represent the mean value for each patient throughout the baseline or follow-up periods; these laboratory measurements were missing in 8647–7520 patients in statin-users and 26546 – 18619 patients in the non-users.
Table 2.
Baseline characteristics and outcomes of propensity score-matched statin users and non-users.
| Statin Users (n =6972) | Non-users (n =6972) | p | |
|---|---|---|---|
| Age in years: mean ±SD | 56.5 ±12.5 | 56.8 ±12.2 | 0.2 |
| Male gender: n (%) | 3769 (54.1) | 3778 (54.2) | 0.9 |
| Comorbidities in baseline period: | |||
| Acute myocardial infarction*: n (%) | 59 (0.8) | 66 (0.9) | 0.5 |
| Congestive heart failure*: n (%) | 130 (1.9) | 130 (1.9) | 1.0 |
| Peripheral vascular disease*: n (%) | 136 (2) | 142 (2) | 0.8 |
| Cerebrovascular disease*: n (%) | 153 (2.2) | 146 (2.1) | 0.7 |
| Dementia*: n (%) | 29 (0.4) | 29 (0.4) | 1.0 |
| Chronic obstructive lung disease*: n (%) | 817 (11.7) | 835 (12) | 0.7 |
| Rheumatologic diseases*: n (%) | 157 (2.3) | 170 (2.4) | 0.8 |
| Peptic ulcer disease*: n (%) | 99 (1.4) | 102 (1.5) | 0.9 |
| Mild liver disease*: n (%) | 30 (0.4) | 30 (0.4) | 1.0 |
| Diabetes mellitus*: n (%) | 699 (10) | 680 (9.8) | 0.6 |
| Diabetes mellitus with complications*: n (%) | 182 (2.6) | 162 (2.3) | 0.3 |
| Hemiplegia/paraplegia*: n (%) | 9 (0.1) | 9 (0.1) | 1.0 |
| Renal disease*: n (%) | 96 (1.4) | 91 (1.32) | 0.8 |
| Malignancy*: n (%) | 438 (6.3) | 435 (6.2) | 0.9 |
| Liver disease (moderate/severe)*: n (%) | 4 (0.1) | 4 (0.1) | 1.0 |
| Metastatic neoplasm*: n (%) | 24 (0.3) | 20 (0.3) | 0.6 |
| HIV *: n (%) | 7 (0.2) | 9 (0.1) | 0.6 |
| Charlson Comorbidity total score: mean ± SD | 0.59 ± 1.1 | 0.57 ± 1.15 | 0.4 |
| Psychological disorders at baseline **: n (%) | 980 (7) | 1010 (7.2) | 0.5 |
| Illicit drug use: n (%) | 11 (0.2) | 10 (0.1) | 0.8 |
| Alcohol abuse/dependence: n (%) | 72 (1) | 60 (0.9) | 0.3 |
| Smoking: n (%) | 565 (8.1) | 581 (8.3) | 0.6 |
| Obesity | 1066 (15.3) | 1059 (15.2) | 0.9 |
| Vision defects/blindness | 3073 (44.1) | 3096 (44.4) | 0.7 |
| Number of outpatient medical encounters during baseline period: mean ± SD | 32.3 ±31.8 | 32.2 ± 51.3 | 0.7 |
| Number of inpatient admission during baseline period: mean ± SD | 0.27 ± 0.8 | 0.27 ± 0.7 | 0.9 |
| Number of outpatient medical encounters during follow-up period: mean ± SD | 89.5 ± 84.2 | 88.4 ± 115 | 0.5 |
| Number of inpatient admission during follow- up period: mean ± SD | 0.8 ± 1.8 | 0.8 ± 1.9 | 0.8 |
| Medications: | |||
| Beta-blocker: n (%) | 1274 (18.3) | 1275 (18.3) | 1.0 |
| Diuretic: n (%) | 1934 (27.7) | 1950 (28) | 0.8 |
| Calcium channel Blocker: n (%) | 1105 (15.8) | 1086 (15.6) | 0.7 |
| Non-statin lipid lowering drugs: n (%) | 520 (7.5) | 483 (6.9) | 0.2 |
| ACE/ARB: n (%) | 2444 (35.1) | 2427 (34.8) | 0.8 |
| Oral hypoglycemic: n (%) | 321 (4.6) | 297 (4.3) | 0.3 |
| Cytochrome p 450: n (%) | 465 (6.7) | 461 (6.6) | 0.9 |
| Aspirin: n (%) | 2169 (31.1) | 2147 (30.8) | 0.7 |
| NSAID: n (%) | 3998 (57.3) | 3993 (57.3) | 0.9 |
| SSRI: n (%) | 1167 (16.7) | 1191 (17.1) | 0.6 |
| Systemic corticosteroid: n (%) | 272 (3.9) | 270 (3.9) | 1.0 |
| Antipsychotic: n (%) | 89 (1.3) | 94 (1.3) | 0.7 |
| Sedatives: n (%) | 1377 (19.8) | 1370 (19.7) | 0.9 |
| Tricyclic antidepressants: n (%) | 13 (0.2) | 20 (0.3) | 0.2 |
| Outcomes | |||
| Psych1 (schizophrenia/other psychosis): n (%) | 265 (1.9) | 297 (2.1) | 0.2 |
| Psych2 (depression / bipolar disorder): n (%) | 1490 (21.4) | 1462 (21) | 0.6 |
| Psych3 (all psychological disorders): n (%) | 2828 (40.6) | 2784 (39.9) | 0.5 |
ACE/ARB: Angiotensin-receptor blockers & angiotensin converting enzyme inhibitors;
NSAID: non-steroidal anti-inflammatory drugs; Cytochrome p 450: medications that inhibit the Cytochrome p450 system as identified in a recent FDA warning; 29 SSRI: selective serotonin reuptake inhibitors
Diagnosis is based on ICD-9-CM codes as identified in Deyo method for applying the Charlson comorbidity score28
Definition is detailed in Appendix 1
For the secondary analysis, the primary prevention cohort included 33513 patients (6113 statin users and 27400 non-users). Statin users in the primary prevention cohort (Table 3) were older, more likely to be male, had higher rates of obesity, smoking, and alcohol abuse/dependence, were prescribed more medications, and used the healthcare system more frequently than non-users. Lastly, patients in the no baseline psychological comorbidity cohort (Table 4) were similar to the all-patients cohort in terms of differences between statin users and non-users.
Table 3.
Baseline characteristics of statin users and non-users in patients with no comorbidity (based on Charlson Comorbidity Score28)
| Statinusers (n =6113) | Non-users (n =27400) | p-value | |
|---|---|---|---|
| Age in years: mean ± SD | 56.6 ± 12.1 | 43.6 ± 10.3 | < 0.0001 |
| Male gender: n (%) | 3707 (60.6) | 12246 (44.7) | < 0.0001 |
| Charlson Comorbidity total score: mean ± SD | 0 ± 0 | 0 ± 0 | |
| Psychological disorders at baseline *: n (%) | 789 (12.9) | 2802 (10.2) | < 0.0001 |
| Illicit drug use: n (%) | 6 (0.1) | 40 (0.1) | 0.45 |
| Alcohol abuse/dependence: n (%) | 48 (0.8) | 153 (0.6) | 0.04 |
| Smoking: n (%) | 477 (7.8) | 1401 (5.1) | < 0.0001 |
| Obesity: n (%) | 901 (14.7) | 2520 (9.2) | < 0.0001 |
| Vision defects/blindness | 2613 (42.7) | 11048 (40.3) | 0.001 |
| Number of outpatient medical encounters during baseline period: mean ± SD | 27.2 ± 25.4 | 19.4 ± 21 | < 0.0001 |
| Number of inpatient admissions during baseline period: mean ± SD | 0.18 ± 0.5 | 0.13 ± 0.4 | < 0.0001 |
| Number of outpatient medical encounters during follow-up period: mean ± SD | 81.1 ± 78.3 | 55.6 ± 59.6 | < 0.0001 |
| Number of inpatient admissions during follow-up period: mean ± SD | 2.1 ± 1.9 | 1.7 ± 1.5 | < 0.0001 |
| Medications: | |||
| Beta-blocker: n (%) | 1385 (22.7) | 1500 (5.5) | < 0.0001 |
| Diuretic: n (%) | 1748 (28.6) | 2305 (8.4) | < 0.0001 |
| Calcium channel Blocker: n (%) | 1092 (17.9) | 1064 (3.9) | < 0.0001 |
| Non-statin lipid lowering drugs: n (%) | 755(12.4) | 360 (1.3) | < 0.0001 |
| ACE/ARB: n (%) | 2371 (38.8) | 2265 (8.3) | < 0.0001 |
| Oral hypoglycemic: n (%) | 32 (0.5) | 89 (0.3) | 0.02 |
| Cytochrome p 450: n (%) | 465 (7.6) | 1113 (4.1) | < 0.0001 |
| Aspirin: n (%) | 2434 (39.8) | 1761 (6.4) | < 0.0001 |
| NSAID: n (%) | 3550 (58.1) | 16860 (61.5) | < 0.0001 |
| SSRI: n (%) | 944 (15.4) | 3256 (11.9) | < 0.0001 |
| Systemic corticosteroid: n (%) | 127 (2.1) | 873 (3.2) | < 0.0001 |
| Antipsychotic: n (%) | 62 (1.0) | 232 (0.8) | 0.2 |
| Sedatives: n (%) | 1094 (17.9) | 4182 (15.3) | < 0.0001 |
| Tricyclic antidepressants: n (%) | 11 (0.2) | 42 (0.2) | 0.4 |
ACE/ARB: Angiotensin-receptor blockers & angiotensin converting enzyme inhibitors; NSAID: non-steroidal anti-inflammatory drugs; Cytochrome p 450: medications that inhibit the Cytochrome p450 system as identified in a recent FDA warning; 29 SSRI: selective serotonin reuptake inhibitors
Schizophrenia/other psychosis, mood disorders, depression, and bipolar disorder as detailed in Appendix 1
Outcome measures
Table 5 compares the psychological diagnoses made during the follow-up period for statin users and non-users. In the propensity score-matched cohort (primary analysis), statin users and non-users had similar risks of Psych1, Psych2, and Psych3 diagnoses.
Table 5.
Comparison of psychological disorders during follow-up in statin users and non-users
| OR | 95%CI | p-value | ||
|---|---|---|---|---|
| Propensity-matched patients | ||||
| Psych1 (schizophrenia/other psychosis) | 0.9 | 0.75 | 1.05 | 0.2 |
| Psych2 (depression / bipolar disorder) | 1.02 | 0.94 | 1.11 | 0.6 |
| Psych3 (all psychological disorders) | 1.02 | 0.96 | 1.1 | 0.5 |
| All patients cohort* | ||||
| Psych1 (schizophrenia/other psychosis) | 0.9 | 0.78 | 1.06 | 0.2 |
| Psych2 (depression / bipolar disorder) | 1.03 | 0.95 | 1.11 | 0.5 |
| Psych3 (all psychological disorders) | 1.05 | 0.99 | 1.12 | 0.1 |
| Patients with no Charlson comorbidity† | ||||
| Psych1 (schizophrenia/other psychosis) | 0.95 | 0.76 | 1.19 | 0.7 |
| Psych2 (depression / bipolar disorder) | 1.01 | 0.9 | 1.12 | 0.8 |
| Psych3 (all psychological disorders) | 1.03 | 0.95 | 1.05 | 0.5 |
| Patients with no psychological disorders at baseline‡ | ||||
| Psych1 (schizophrenia/other psychosis) | 0.88 | 0.72 | 1.08 | 0.2 |
| Psych2 (depression / bipolar disorder) | 0.97 | 0.89 | 1.07 | 0.6 |
| Psych3 (all psychological disorders) | 1.02 | 0.96 | 1.09 | 0.6 |
Adjusting for age, gender, statin use, all comorbid conditions as in table 2, total Charlson comorbidity score, obesity, smoking, alcohol use, illicit drug use, Psychological disorders at baseline, vision defects/blindness, number of all admissions in baseline period, number of all outpatients medical encounters in the baseline period, and use of different classes of medications as listed in table 1.
adjusting for the same factors mentioned earlier except for Charlson comorbidity total score and its components.
adjusting for the same factors mentioned in “all patients cohort” except for “psychological disorders at baseline” covariate.
In the secondary analyses, after adjusting for potential confounders in the all-patients cohort, the primary prevention cohort, and the no baseline psychological comorbidity cohort, statin users and non-users had similar risks of Psych1, Psych2, and Psych3 diagnoses. The secondary analyses were repeated for sub-groups of patients on statins for different durations (Tables 6, 7). Once again, statin users and non-users had similar risks of Psych1, Psych2, and Psych3 diagnoses across all 3 cohorts.
Table 6.
Comparison of psychological disorders during follow-up in statin users (≥ = 1 year) and non-users
| OR | 95%CI | p-value | ||
|---|---|---|---|---|
| All patients cohort* (statin users = 12883 patients, non-users = 32623 patients) | ||||
| Psych1 (schizophrenia/other psychosis) | 0.89 | 0.76 | 1.04 | 0.1 |
| Psych2 (depression / bipolar disorder) | 1.05 | 0.96 | 1.14 | 0.3 |
| Psych3 (all psychological disorders) | 1.06 | 0.998 | 1.132 | 0.06 |
| Patients with no Charlson comorbidity† (statin users = 5646 patients, non-users = 27400 patients) | ||||
| Psych1 (schizophrenia/other psychosis) | 0.94 | 0.74 | 1.18 | 0.6 |
| Psych2 (depression / bipolar disorder) | 1.03 | 0.93 | 1.14 | 0.6 |
| Psych3 (all psychological disorders) | 1.05 | 0.97 | 1.14 | 0.2 |
| Patients with no psychological disorders at baseline‡ (statin users = 10826 patients, non- users = 28861 patients) | ||||
| Psych1 (schizophrenia/other psychosis) | 0.89 | 0.73 | 1.08 | 0.2 |
| Psych2 (depression / bipolar disorder) | 1.0 | 0.9 | 1.1 | 0.9 |
| Psych3 (all psychological disorders) | 1.06 | 0.99 | 1.13 | 0.1 |
Adjusting for age, gender, statin use, all comorbid conditions as in table 2, total Charlson comorbidity score, obesity, smoking, alcohol use, illicit drug use, psychological disorders at baseline, vision defects/blindness, number of all admissions in baseline period, number of all outpatients medical encounters in the baseline period, and use of different classes of medications as listed in table 1.
adjusting for the same factors mentioned earlier except for Charlson comorbidity total score and its components.
adjusting for the same factors mentioned in “all patients cohort” except for “psychological disorders at baseline” covariate.
Table 7.
Comparison of psychological disorders during follow-up in statin users (≥ = 2 years) and non-users
| OR | 95%CI | p-value | ||
|---|---|---|---|---|
| All patients cohort* (statin users = 12006 patients, non-users = 32623 patients) | ||||
| Psych1 (schizophrenia/other psychosis) | 0.87 | 0.76 | 1.02 | 0.09 |
| Psych2 (depression / bipolar disorder) | 1.15 | 1.05 | 0.97 | 0.2 |
| Psych3 (all psychological disorders) | 1.06 | 0.99 | 1.13 | 0.1 |
| Patients with no Charlson comorbidity† (statin users = 5172 patients, non-users = 27400 patients) | ||||
| Psych1 (schizophrenia/other psychosis) | 0.88 | 0.69 | 1.12 | 0.3 |
| Psych2 (depression / bipolar disorder) | 1.03 | 0.93 | 1.15 | 0.5 |
| Psych3 (all psychological disorders) | 1.04 | 0.96 | 1.13 | 0.4 |
| Patients with no psychological disorders at baseline‡ (statin users = 10095 patients, non- users = 28861 patients) | ||||
| Psych1 (schizophrenia/other psychosis) | 0.85 | 0.7 | 1.03 | 0.1 |
| Psych2 (depression / bipolar disorder) | 1.01 | 0.9 | 1.11 | 0.9 |
| Psych3 (all psychological disorders) | 1.05 | 0.98 | 1.12 | 0.2 |
Adjusting for age, gender, statin use, all comorbid conditions as in table 2, total Charlson comorbidity score, obesity, smoking, alcohol use, illicit drug use, psychological disorders at baseline, vision defects/blindness, number of all admissions in baseline period, number of all outpatients medical encounters in the baseline period, and use of different classes of medications as listed in table 1.
adjusting for the same factors mentioned earlier except for Charlson comorbidity total score and its components.
adjusting for the same factors mentioned in “all patients cohort” except for “psychological disorders at baseline” covariate.
Discussion
The risk of developing psychological disorders was similar after 4.5 years of follow-up in this cohort of propensity score-matched statin users and non-users. These results were consistent for all pre-specified psychological disorder groups and all patient cohorts. To our knowledge, this is the largest observational study to investigate the relationship between statin therapy and the risk of psychological disorders. In a prospective observational study, Young-Xu et al. used a validated questionnaire to assess for anxiety, depression, and hostility in 590 patients recruited from a cardiology clinic.32 The patients were followed for an average of 4 years. The authors created a propensity score for statin use and used that score in their final regression model. It is noteworthy that the investigators initially screened 2598 patients, and identified 761 patients who met study inclusion criteria, yet they only included 590 patients in their final analysis.
In a secondary analysis of the LIPID (Long-term Intervention with Pravastatin in Ischemic Disease study) trial, a representative sample of 1222 patients were followed using a standard self-administered questionnaire at baseline and on repeated follow-up intervals over a 4-year period.33 The LIPID trial randomized patients to pravastatin 40 mg or placebo. A total of 1130 participants returned the questionnaire. The questionnaire assessed anxiety, anger, depression, impulsiveness, alcohol consumption, and adverse life events. This study found no significant association between pravastatin and psychological adverse events. In another prospective observational analysis from the Singapore Longitudinal Aging Study, 1083 individuals (out of 2084 eligible individuals) responded to several validated questionnaires; medical comorbidities and functional status were self-reported.34 This study found no association between statin use and depression. In another cross-sectional analysis of the same patient cohort, statin users had fewer depressive symptoms than non-users (OR: 0.71; 95%CI: 0.52–0.97).35 Finally, a recent prospective observational study of 965 patients with coronary artery disease found that statin use was associated with a 48% reduction in the odds of developing depression during follow up.36
It is important to note that each of these studies collected data using patient surveys or questionnaires, which are subject to selection bias. Compared to patients who elect to complete the surveys or questionnaires, non-responders are more likely to experience significant physical illness, be unmarried, have lower education, perform more poorly on cognitive testing, and have a history of psychological illness.37 Therefore, our use of ICD-9 codes as a method of identifying psychological disorders avoids potential selection bias and adds to the existing literature.
Contrary to the studies discussed above, several small, randomized controlled trials (RCT) have reported adverse outcomes with statin use. In a RCT of 80 elderly volunteers, simvastatin was associated with more depressive symptoms and a decrease in positive affect.8 In another crossover RCT (placebo vs. simvastatin) of 120 patients, simvastatin resulted in a statistically significant increase in depression and somatization but no change in anxiety, hostility, or aggression scores.9 Statins were also associated with minimal decline in testosterone levels. Data from a small RCT of 12 patients randomized to atorvastatin 20 mg (9 patients) or lovastatin 20 mg (3 patients) and followed by standardized psychological tests at 4 weeks and 52 weeks demonstrated a statistically significant increase in impulsivity at 4 weeks. However, treatment-emergent impulsivity resolved over time, and depression rating scale scores improved at 52 weeks.11
The largest body of evidence supporting an association between statins and psychological disorders comes from case reports, case series, and pharmacovigilance reviews.38–40 Golomb et al. reported 6 cases of severe irritability associated with statins that resolved after stopping the drug, but recurred upon re-challenge in 4 patients.10 A review of reports from the New Zealand Center for Adverse Reaction Monitoring (CARM) showed that 285 out of 364 psychological adverse drug reactions were associated with statins.5 The psychological conditions included mood disorders, cognitive disorders, sleep disorders, perception, and psychotic disorders. In a review of an Italian database of spontaneous adverse drug reaction reporting, the investigators compared statin adverse events with other drugs.7 Out of 35314 adverse event reports, 71 psychological adverse events in 60 reports were associated with statin use; 14 reports noted a positive re-challenge. The five most frequent adverse events were insomnia, somnolence, agitation, confusion, and hallucination. Adjusted and unadjusted ORs for statin-associated psychological adverse events were similar to other medications. Only insomnia was higher with statins as compared with all other drugs (OR: 3.3; 95%CI: 1.9–5.7).7
Despite the value of pharmacovigilance reports and case series, their passive surveillance approaches are limited because there is no control group; therefore, we cannot calculate a relative risk or odds ratio. Furthermore, those types of data lack generalizability, because of the patients self-select for participation. Hence, our study offers a unique opportunity to assess the risk of psychological adverse events in a large, well-described patient population. The lack of an association between statin use and psychological disorders in our study is reassuring; however, it does not preclude the possibility of rare adverse effects.
Changes in the serotonin system have been suggested as a mechanism underlying psychological adverse events related to statins. One study measured serotonin transporter (SERT) activity in 17 hyperlipidemic patients treated with simvastatin.41 In this group of patients, SERT activity increased significantly and cell membrane cholesterol content decreased following 1 month of therapy, but both showed a tendency to normalize after about 1 year of therapy. Increased SERT activity may be linked to impulsiveness,41 and may lead to a decrease in serotonin neurotransmitter activity, which has been associated with increased impulsivity.42 Data from an animal model of serotonin1A receptor activity suggested that chronic cholesterol depletion by mevastatin reduces the level of specific ligand binding and G-protein coupling to serotonin1A receptors. 43
Coenzyme Q10 (CoQ10) may provide another plausible mechanism for a relationship between statins and depression. Statins lower plasma CoQ10 levels, which were significantly lower in 35 patients with depression in comparison to control subjects.44 In addition, studies on hypercholesterolemic, but otherwise healthy subjects, demonstrated that higher levels of ω-3 polyunsatutrated fatty acids were associated with lower scores on a standardized depression scales.45
There are some limitations to our study. First, we cannot exclude the presence of infrequent adverse events, such as those reported in case series and pharmacovigilance reports. Future observational studies may focus on studying patients who discontinued statins, not only those who continued statins. Additionally, as in any retrospective observational study, unaccounted for confounders may still be present even after propensity score matching. The use of an administrative database has its own inherent weaknesses, because diagnosis codes may lack accuracy when compared to actual patient medical charts.46 Some psychological disorders may not be adequately diagnosed by healthcare professionals. We also did not account for ethnicity, which may affect diagnosis rates of psychological disorders.47 Some of the current non-users may represent a group of patients who experienced statin-related severe adverse events prior to the date of our study (2003). Our study did not address the risk of suicide, which has been a concern in lipid-lowering interventions, nor did it account for mortality as a cause of competing outcome. However, it is unlikely that such events occur in the absence of any ICD-9-CM code suggestive of psychological illness. We also did not study the effect of statin use on drug dependence; however, we are not aware of any study that has identified an association between statin use and drug or alcohol dependence.
In conclusion, in this cohort of propensity score-matched statin users and non-users, the risk of developing psychological disorders was similar after 4.5 years of follow-up. In contrast, other reports have suggested the presence of such an association; therefore, this topic warrants further study.
Acknowledgments
Funding source: No funding was provided for the conduct of this study or the preparation of this paper. CRF is supported by the U.S. National Institutes of Health (NIH) in the form of a NIH/KL2 career development award (RR025766). In addition, CRF has received research grants and/or served as a scientific consultant/advisor for AstraZeneca, Bristol Myers Squibb, Elan, Forest, Ortho-McNeil Janssen, and Pfizer.
Appendix 1. Groups of psychiatric comorbidity and their definitions
| Group Name (Reference) | Definitions/ICD-9-CM codes |
|---|---|
| Psychiatric diseases at Baseline (Skanderson) 21 | Schizophrenia |
| Other psychosis | |
| Schizoaffective disorders | |
| Bipolar disorder | |
| Major depression | |
| Other depression (formerly mild depression) | |
| PTSD (post-traumatic stress disorder) | |
| Developmental diseases and psychosis with origin specific to childhood | |
| Psych1 (Skanderson) 21* | Schizophrenia |
| Other psychosis | |
| Schizoaffective disorders | |
| Psych2 (Skanderson) 21† | Bipolar disorder |
| Major depression | |
| Other depression (formerly mild depression) | |
| PTSD (post-traumatic stress disorder) | |
| Psych 3 (AHRQ)18 | Anxiety disorders: 29384 30000 30001 30002 30009 30010 30020 30021 30022 30023 30029 3003 3005 30089 3009 3080 3081 3082 3083 3084 3089 30981 3130 3131 31321 31322 3133 31382 31383 |
| Adjustment disorders: 3090 3091 30922 30923 30924 30928 30929 3093 3094 30982 30983 30989 3099 | |
| Attention-deficit, conduct, and disruptive behavior disorders: 31200 31201 31202 31203 31210 31211 31212 31213 31220 31221 31222 31223 3124 3128 31281 31282 31289 3129 31381 31400 31401 3141 3142 3148 3149 | |
| Impulse control disorders, NEC: 31230 31231 31232 31233 31234 31235 31239 | |
| Mood disorders: 29383 29600 29601 29602 29603 29604 29605 29606 29610 29611 29612 29613 29614 29615 29616 29620 29621 29622 29623 29624 29625 29626 29630 29631 29632 29633 29634 29635 29636 29640 29641 29642 29643 29644 29645 29646 29650 29651 29652 29653 29654 29655 29656 29660 29661 29662 29663 29664 29665 29666 2967 29680 29681 29682 29689 29690 29699 3004 311 | |
| Personality disorders: 3010 30110 30111 30112 30113 30120 30121 30122 3013 3014 30150 30151 30159 3016 3017 30181 30182 30183 30184 30189 3019 | |
| Schizophrenia and other psychotic disorders: 29381 29382 29500 29501 29502 29503 29504 29505 29510 29511 29512 29513 29514 29515 29520 29521 29522 29523 29524 29525 29530 29531 29532 29533 29534 29535 29540 29541 29542 29543 29544 29545 29550 29551 29552 29553 29554 29555 29560 29561 29562 29563 29564 29565 29570 29571 29572 29573 29574 29575 29580 29581 29582 29583 29584 29585 29590 29591 29592 29593 29594 29595 2970 2971 2972 2973 2978 2979 2980 2981 2982 2983 2984 2988 2989 | |
| Miscellaneous disorders (670): 293.89, 293.9, 3001.1, 3001.2, 3001.3, 300.14, 300.15, 300.16, 300.19, 300.6, 300.7, 300.81, 300.82, 302.1, 302.2, 302.3, 302.4, 302.50, 302.51,, 302.52, 302.53, 302.6, 302.70, 302.71, 302.72, 302.73, 302.74, 302.75, 302.76, 302.79, 302.81, 302.82, 302.83, 302.84, 302.85, 302.89, 302.9, 306.0, 306.1, 306.2, 306.3, 306.4, 306.50, 306.51, 306.52, 306.53, 306.59, 306.6, 306.7, 306.8, 306.9, 307.1, 307.40, 307.41, 307.42, 307.43, 307.44, 307.45, 307.46, 307.47, 307.48, 307.49, 307.50, 307.51, 307.52, 307.53, 3075.4, 307.59, 307..80, 307.81, 307.89, 310.1, 316, 648.40, 648.41, 648.42, 648.43, 648.44, V402, V403, V403.1, V403.9, V409, V673 |
In difference with the VACS, we did not include codes 299.x (pervasive developmental disorder), and code V11.0 (personal history of schizophrenia), which are not in consideration as a complication for statin-associated side effects.
In difference with the VACS, we did not include codes code V11.1 (personal history of manic-depressive psychosis), which is not in consideration as a complication for statin-associated side effects.
Footnotes
Conflict of interest: none.
Disclaimer: The views expressed herein are those of the authors and do not reflect the official policy or position of the Department of the Army, Department of Defense, or the US Government. The authors are employees of the US government. This work was prepared as part of their official duties and, as such, there is no copyright to be transferred.
References
- 1.Golomb BA, Evans MA. Statin adverse effects : a review of the literature and evidence for a mitochondrial mechanism. Am J Cardiovasc Drugs. 2008;8:373–418. doi: 10.2165/0129784-200808060-00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Evans MA, Golomb BA. Statin-associated adverse cognitive effects: survey results from 171 patients. Pharmacotherapy. 2009;29:800–11. doi: 10.1592/phco.29.7.800. [DOI] [PubMed] [Google Scholar]
- 3.Fernandez G, Spatz ES, Jablecki C, et al. Statin myopathy: a common dilemma not reflected in clinical trials. Cleve Clin J Med. 2011;78:393–403. doi: 10.3949/ccjm.78a.10073. [DOI] [PubMed] [Google Scholar]
- 4.Goldstein MR, Mascitelli L, Pezzetta F. The double-edged sword of statin immunomodulation. Int J Cardiol. 2009;135:128–30. doi: 10.1016/j.ijcard.2008.01.023. [DOI] [PubMed] [Google Scholar]
- 5.Tatley M, Savage R. Psychiatric adverse reactions with statins, fibrates and ezetimibe: implications for the use of lipid-lowering agents. Drug Saf. 2007;30:195–201. doi: 10.2165/00002018-200730030-00003. [DOI] [PubMed] [Google Scholar]
- 6.Wagstaff LR, Mitton MW, Arvik BM, et al. Statin-associated memory loss: analysis of 60 case reports and review of the literature. Pharmacotherapy. 2003;23:871–80. doi: 10.1592/phco.23.7.871.32720. [DOI] [PubMed] [Google Scholar]
- 7.Tuccori M, Lapi F, Testi A, et al. Statin-associated psychiatric adverse events: a case/non-case evaluation of an Italian database of spontaneous adverse drug reaction reporting. Drug Saf. 2008;31:1115–23. doi: 10.2165/0002018-200831120-00007. [DOI] [PubMed] [Google Scholar]
- 8.Morales K, Wittink M, Datto C, et al. Simvastatin causes changes in affective processes in elderly volunteers. J Am Geriatr Soc. 2006;54:70–6. doi: 10.1111/j.1532-5415.2005.00542.x. [DOI] [PubMed] [Google Scholar]
- 9.Hyyppa MT, Kronholm E, Virtanen A, et al. Does simvastatin affect mood and steroid hormone levels in hypercholesterolemic men? A randomized double-blind trial. Psychoneuroendocrinology. 2003;28:181–94. doi: 10.1016/s0306-4530(02)00014-8. [DOI] [PubMed] [Google Scholar]
- 10.Golomb BA, Kane T, Dimsdale JE. Severe irritability associated with statin cholesterol-lowering drugs. QJM. 2004;97:229–35. doi: 10.1093/qjmed/hch035. [DOI] [PubMed] [Google Scholar]
- 11.Ormiston T, Wolkowitz OM, Reus VI, et al. Behavioral implications of lowering cholesterol levels: a double-blind pilot study. Psychosomatics. 2003;44:412–4. doi: 10.1176/appi.psy.44.5.412. [DOI] [PubMed] [Google Scholar]
- 12.FDA. Drug Saftey Communication: Important safety label changes to cholesterol-lowering statin drugs. The U.S. Food and Drug Administration; 2012. [Google Scholar]
- 13.Gantt CJ, Neely JA, Villafana IA, et al. Analysis of weight and associated health consequences of the active duty staff at a major Naval medical center. Mil Med. 2008;173:434–40. doi: 10.7205/milmed.173.5.434. [DOI] [PubMed] [Google Scholar]
- 14.George SZ, Childs JD, Teyhen DS, et al. Brief psychosocial education, not core stabilization, reduced incidence of low back pain: results from the Prevention of Low Back Pain in the Military (POLM) cluster randomized trial. BMC Med. 2011;9:128. doi: 10.1186/1741-7015-9-128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Enewold L, Brinton LA, McGlynn KA, et al. Oral contraceptive use among women in the military and the general U.S. population. J Womens Health (Larchmt) 2010;19:839–45. doi: 10.1089/jwh.2009.1706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Moniz C. Graduate Management Project. Baltimore, MD: US Army Medical Department Center and School; 2008. Outpatient Workload (RVU) Predictors: Age, Gender & Beneficiary Category. 15-06-2008. [Google Scholar]
- 17.Kugler J. Military Health System Patient Centered Medical Home Guide: Tricare. 2011 Jun [Google Scholar]
- 18.Elixhauser A, Steiner C, Palmer L. Clinical Classifications Software (CCS) for ICD-9-CM. Databases and Related Tools from the Healthcare Cost and Utilization Project (HCUP) U.S. Agency for Healthcare Research and Quality; 2012. p. Appendix A. [Google Scholar]
- 19.Elixhauser A, Steiner C, Harris DR, et al. Comorbidity measures for use with administrative data. Med Care. 1998;36:8–27. doi: 10.1097/00005650-199801000-00004. [DOI] [PubMed] [Google Scholar]
- 20.Selim AJ, Fincke G, Ren XS, et al. Comorbidity assessments based on patient report: results from the Veterans Health Study. J Ambul Care Manage. 2004;27:281–95. doi: 10.1097/00004479-200407000-00011. [DOI] [PubMed] [Google Scholar]
- 21.Gibson D, Helmick K, Jaffe M, et al. Traumatic brain injury care in the Department of Defense. the Department of Defense. 2009 available at: http://www.dcoe.health.mil/Content/Navigation/Documents/Traumatic%20Brain%20Injury%20Care%20in%20the%20Department%20of%20Defense.pdf.
- 22.Tan CP, Ng A, Civil I. Co-morbidities in trauma patients: common and significant. N Z Med J. 2004;117:U1044. [PubMed] [Google Scholar]
- 23.Esenkaya I, Sakarya B, Unay K, et al. The influence of atorvastatin on tendon healing: an experimental study on rabbits. Orthopedics. 2010;33:398. doi: 10.3928/01477447-20100429-06. [DOI] [PubMed] [Google Scholar]
- 24.Chauhan NB, Gatto R. Synergistic benefits of erythropoietin and simvastatin after traumatic brain injury. Brain Res. 2010;1360:177–92. doi: 10.1016/j.brainres.2010.09.010. [DOI] [PubMed] [Google Scholar]
- 25.Beziaud T, Ru Chen X, El Shafey N, et al. Simvastatin in traumatic brain injury: effect on brain edema mechanisms. Crit Care Med. 2011;39:2300–7. doi: 10.1097/CCM.0b013e3182227e4a. [DOI] [PubMed] [Google Scholar]
- 26.Werhagen L, Borg K. Post-polio syndrome, spinal cord injury and statin myopathy: double trouble or incorrect diagnosis? Two case reports. J Rehabil Med. 2011;43:734–5. doi: 10.2340/16501977-0839. [DOI] [PubMed] [Google Scholar]
- 27.Skanderson M. Description of ICD-9 comorbidity codes used by VACS (Veterans Aging Cohort Study) West Haven, CT: 2006. pp. 1–29. [Google Scholar]
- 28.Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45:613–9. doi: 10.1016/0895-4356(92)90133-8. [DOI] [PubMed] [Google Scholar]
- 29.FDA Drug Safety Communication; Administration USFaD, editor. New restrictions, contraindications, and dose limitations for Zocor (simvastatin) to reduce the risk of muscle injury. U.S. Department of Health and Human Services; 2011. [Google Scholar]
- 30.Becker S, Ichino A. Estimation of average treatment effects based on propensity scores. The Stata Journal. 2002;2:358–77. [Google Scholar]
- 31.Leuven E, Sianesi B. PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. version 4.0.5. 2003. [Google Scholar]
- 32.Young-Xu Y, Chan KA, Liao JK, et al. Long-term statin use and psychological well-being. J Am Coll Cardiol. 2003;42:690–7. doi: 10.1016/S0735-1097(03)00785-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Stewart RA, Sharples KJ, North FM, et al. Long-term assessment of psychological well-being in a randomized placebo-controlled trial of cholesterol reduction with pravastatin. The LIPID Study Investigators. Arch Intern Med. 2000;160:3144–52. doi: 10.1001/archinte.160.20.3144. [DOI] [PubMed] [Google Scholar]
- 34.Feng L, Yap KB, Kua EH, et al. Statin use and depressive symptoms in a prospective study of community-living older persons. Pharmacoepidemiol Drug Saf. 2010;19:942–8. doi: 10.1002/pds.1993. [DOI] [PubMed] [Google Scholar]
- 35.Feng L, Tan CH, Merchant RA, et al. Association between depressive symptoms and use of HMG-CoA reductase inhibitors (statins), corticosteroids and histamine H(2) receptor antagonists in community-dwelling older persons: cross-sectional analysis of a population-based cohort. Drugs Aging. 2008;25:795–805. doi: 10.2165/00002512-200825090-00005. [DOI] [PubMed] [Google Scholar]
- 36.Otte C, Zhao S, Whooley MA. Statin use and risk of depression in patients with coronary heart disease: longitudinal data from the Heart and Soul Study. J Clin Psychiatry. 2012;73:610–5. doi: 10.4088/JCP.11m07038. [DOI] [PubMed] [Google Scholar]
- 37.Launer LJ, Wind AW, Deeg DJ. Nonresponse pattern and bias in a community-based cross-sectional study of cognitive functioning among the elderly. Am J Epidemiol. 1994;139:803–12. doi: 10.1093/oxfordjournals.aje.a117077. [DOI] [PubMed] [Google Scholar]
- 38.Rosenson RS, Goranson NL. Lovastatin-associated sleep and mood disturbances. Am J Med. 1993;95:548–9. doi: 10.1016/0002-9343(93)90343-n. [DOI] [PubMed] [Google Scholar]
- 39.Gregoor PJ. Atorvastatin may cause nightmares. BMJ. 2006;332:950. doi: 10.1136/bmj.332.7547.950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Peters JT, Garwood CL, Lepczyk M. Behavioral changes with paranoia in an elderly woman taking atorvastatin. Am J Geriatr Pharmacother. 2008;6:28–32. doi: 10.1016/j.amjopharm.2008.03.001. [DOI] [PubMed] [Google Scholar]
- 41.Vevera J, Fisar Z, Kvasnicka T, et al. Cholesterol-lowering therapy evokes time-limited changes in serotonergic transmission. Psychiatry Res. 2005;133:197–203. doi: 10.1016/j.psychres.2004.11.005. [DOI] [PubMed] [Google Scholar]
- 42.Golomb BA. Cholesterol and violence: is there a connection? Ann Intern Med. 1998;128:478–87. doi: 10.7326/0003-4819-128-6-199803150-00009. [DOI] [PubMed] [Google Scholar]
- 43.Shrivastava S, Pucadyil TJ, Paila YD, et al. Chronic cholesterol depletion using statin impairs the function and dynamics of human serotonin(1A) receptors. Biochemistry. 2010;49:5426–35. doi: 10.1021/bi100276b. [DOI] [PubMed] [Google Scholar]
- 44.Maes M, Mihaylova I, Kubera M, et al. Lower plasma Coenzyme Q10 in depression: a marker for treatment resistance and chronic fatigue in depression and a risk factor to cardiovascular disorder in that illness. Neuro Endocrinol Lett. 2009;30:462–9. [PubMed] [Google Scholar]
- 45.Conklin SM, Harris JI, Manuck SB, et al. Serum omega-3 fatty acids are associated with variation in mood, personality and behavior in hypercholesterolemic community volunteers. Psychiatry Res. 2007;152:1–10. doi: 10.1016/j.psychres.2006.10.006. [DOI] [PubMed] [Google Scholar]
- 46.Spangler WE, May JH, Strum DP, et al. A data mining approach to characterizing medical code usage patterns. J Med Syst. 2002;26:255–75. doi: 10.1023/a:1015014402846. [DOI] [PubMed] [Google Scholar]
- 47.Lukachko A, Olfson M. Race and the clinical diagnosis of depression in new primary care patients. Gen Hosp Psychiatry. 2012;34:98–100. doi: 10.1016/j.genhosppsych.2011.09.008. [DOI] [PubMed] [Google Scholar]