Why Patients Stop Taking Statins and What Can Be Done About It

Linda Brookes, MSc

February 14, 2008

Introduction

Discontinuation rates in long-term use of statins remains high. Major clinical trials of statins found 5-year cumulative discontinuation rates of between 6% and 30%.[1-6] Retrospective analyses of prescription databases have reported much higher rates, especially during the first 6 months of treatment, and have also reported that 50% of patients receiving statins discontinue therapy after 1 year of treatment.[7-10] The reasons for discontinuation are complex and, to date, poorly understood, but it is important to identify those that are modifiable so that interventions can be successful. As well as side effects, factors implicated as important in determining adherence to lipid-lowering therapy include gender, age, ethnicity, and socioeconomic status. Studies, such as some presented at the American Heart Association (AHA) 2007 Annual Scientific Sessions, continue to determine which factors are most important and which measures can improve compliance.

Side Effects Not the Main Reason for Stopping Statin Therapy

Presented at the 2007 AHA, a new retrospective analysis reported by Michael H. Davidson, MD (Radiant Research, The University of Chicago Pritzker School of Medicine, Chicago, Illinois), and supported by Walgreens Health Services (Deerfield, Illinois) and Merck & Co (West Point, Pennsylvania) appears to confirm that statin therapy discontinuation rates remain high, with >50% of patients stopping therapy within 12 months.[11] Risk factors associated with discontinuation included increased side effects at higher doses, socioeconomic issues, and high cost, and language barriers presented a particular barrier to persistence with treatment.

The study consisted of 2 parts. The first was a retrospective cohort analysis of prescription data from September 1, 2004 to December 31, 2006, obtained from the pharmacy claims database of Walgreens, a large US national retail pharmacy chain. This component included a 6-month baseline period and a 12-month treatment follow-up period. Data were used from 768,184 patients aged ≥18 years who were taking a statin during the study period, continuously refilled other medications at the pharmacy stores during the follow-up period, but had no refills of the statin during this time.

Discontinuation rates, defined as the proportion of patients who discontinued the study medication over the 12-month follow-up period, were similar for all the different statins used (atorvastatin, fluvastatin, lovastatin, pravastatin, rosuvastatin, and simvastatin alone or combined with ezetimibe). Overall discontinuation rates increased over time, from 27.87% at 3 months to 41.37% at 6 months, and 58.73% at 12 months.

From 3 months onward, patients who were younger, women, Spanish speakers, paid high co-payment, and those in the South were significantly more likely to discontinue statin therapy ( ). The highest rate of discontinuation in the study occurred in Spanish speaking patients (hazard ratio 1.7, 1.9, and 2.1 at 3, 6, and 12 months, respectively). Patients who used the internet to refill their prescriptions had much lower rates of discontinuation, which Dr. Davidson suggested was because they represented a higher socioeconomic group. It could not be related to age, he noted, since younger patients were more likely than older patients to stop treatment. Patients with hypertension were also significantly less likely to discontinue. Using clopidogrel as an index drug, patients with coronary heart disease were less likely to discontinue, but those who stopped clopidogrel within 3 months had a much higher rate of noncompliance with statin treatment.

  Factors Implicated in Discontinuation of Statin Therapy at 12 Months

Variable 12 Months
Odds Ratio 95% CI
Age (y) 0.989 0.988-0.989
Gender, men (women as reference) 0.820 0.812-0.828
Language (English as reference)
Spanish 2.086 2.031-2.146
Other 0.990 0.934-1.049
Census region (South as reference)
Northeast 0.772 0.759-0.785
Midwest 0.799 0.791-0.808
West 0.846 0.835-0.857
Use Internet (no Internet as reference) 0.709 0.701-0.717
Copayment level ($) 1.011 1.010-1.011
Number of chronic diseases 1.027 1.024-1.029
Diabetes 1.073 1.060-1.086
Hypertension 0.877 0.868-0.886
Coronary heart disease 0.350 0.340-0.361
Stopping clopidogrel 3.935 3.808-4.066
High dose 1.040 1.015-1.065

CI = confidence interval

The second part of the study was a mailed survey among a subset of patients to determine why patients discontinued their lipid-lowering therapy. Preliminary results of the survey revealed that only 19% of subjects stopped taking their medications because of side effects, a gratifying result, Dr. Davidson commented. The most common side effects experienced were muscle ache (33%) and weakness (16%). Other side effects experienced by >4% of subjects included stomach complaints (9%), headache (8%), memory loss (7%), flushing (6%), and liver problems (5%).

The converse was that 77% of the patients said they didn't stop taking their medications because of side effects; rather 10% of these patients said they stopped mainly because of cost and 8% did so following their doctor's advice. However, a much larger percentage, 36%, failed to refill their prescriptions because they were getting their medications elsewhere and 7% because they were given samples. This means that 43% of patients who apparently discontinued statin therapy according to the database were actually still taking their medications, Dr. Davidson pointed out, reconfirming that determination of statin nonadherence from analyses of pharmacy records alone tends to overestimate the percentage of patients who are nonadherent.

Dr. Davidson believes that targeting the risk factors revealed by this study may improve long-term statin compliance. However he noted that each reason has a relatively modest effect, so it is a cumulative issue that probably requires a lot of attention to multiple factors. The most intriguing part of the information gained from this study, he said, is that there is a lack of understanding about the reason for the need for the drug. For instance, more than half of subjects (52%) in the survey did not know their low-density lipoprotein cholesterol (LDL-C) value and 48% did not know whether they had low high-density lipoprotein cholesterol (HDL-C) before they started taking their medication. "Education sounds simplistic," Dr. Davidson admitted, "but it appears to be the main factor associated with noncompliance."

Several delegates pointed out that 1 factor that may have improved since the study was carried out 6 months ago is the cost of statin therapy, now that so many pharmacy chains have reduced the cost of generic statins.

Black Men and Women Less Likely to Succeed in Lowering Low-Density Lipoprotein Cholesterol

Non-black women are less likely to receive lipid-lowering therapy than non-black men or black men and women, and black and non-black women and black men are 20%-40% less likely to achieve LDL-C control than non-black men, according to a study presented at the AHA by Barbara Turner, MD, MSED (University of Pennsylvania, Philadelphia).[12]

Dr. Turner explained that results of previous studies of gender and racial difference in the management and control of LCL-C have been conflicting. As noted in a cross-sectional study, the Quality Assurance Program -- a national program sponsored by Merck & Company conducted during the late 1990s that looked at physician practice patterns in patients with cardiovascular disease seen in the outpatient setting -- female and black patients were less likely to have their LDL-C tested, treated, and controlled than white men.[13] In contrast, a multi-site study, the Multi-Ethnic Study of Atherosclerosis, found that women were more likely than men to be treated and controlled, and, after adjustment, racial differences were not significant.[14]

The objectives of the latest study, which was supported by Pfizer (New York, NY), was to identify differences in time to LDL-C control, defined according to National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) guidelines (<130 mg/dL or <100 mg/dL in patients at high risk with coronary artery disease/coronary heart disease or risk equivalent)[15,16] for 4 gender-race groups: black women, black men, non-black women, and non-black men, and to examine whether demographic, clinical, or healthcare factors account for observed differences in LDL-C control.

The study involved patients registered at 6 urban academic primary care practices from January 1, 2003 to February 8, 2005. Administrative and electronic medical records, pharmacy prescriptions, and laboratory data were used to identify a study cohort of 3484 older patients (>45 years if male and ≥55 years if female) at increased risk of coronary artery disease due to hypertension who had ≥1 LDL-C test, and for whom the last LDL-C test was high, ie, >130 mg/dL or >100 mg/dL in high risk patients. The cohort comprised 1440 black women, 666 black men, 661 non-black women (>90% white), and 717 non-black men (>90% white). Overall, the women were older than the men, the non-blacks had higher income, blacks had a higher incidence of diabetes and vascular disease, and women had higher mean baseline LDL-C.

In terms of care delivery, black patients had more office visits scheduled but kept them less frequently than non-black patients ( ). For non-black patients, the healthcare provider's race was more often white. Similar proportions of black men, black women, and non-black men were prescribed lipid-lowering treatment monotherapy, but non-black women were less likely to receive this treatment. Black women were more likely, and non-black women least likely, to be on a statin >50% of time.

  Physician Visits and Treatment During Follow-up

Variables Black
Women
Black
Men
Non-black
Women
Non-black
Men
Highest quartile of visits (%) 25.7 24.9 20.4 18.5
Kept >60% appointments (%) 80.4 79.3 90.0 89.8
Providers race white (%) 60.7 62.5 90.6 91.2
Lipid-lowering treatment (%) 57.8 56.1 51.8 56.8
Statin treatment >50% of time 26.8 21.5 13.4 18.7

During follow-up, non-black men were consistently more likely to have their LDL-C controlled. The median number of days to achieve LDL-C control was lowest in non-black men and greatest in black women (P <.0001 between cohorts) ( ).

  Time to Achievement of Low-Density Lipoprotein Cholesterol Control During Follow-up

Group No. of Days for 50%
of Patients to Achieve
LDL-C Control
Black women 730
Black men 719
Non-black women 692
Non-black men 497

The results did not change after adjustment for a broad range of independent variables: demographics (age, residence income, insurance), clinical factors (related comorbidities such as renal insufficiency, smoking, body mass index, unrelated comorbidities such asthma), healthcare use (hypertension drugs, visits/year, % visits), provider (gender, race type, annual patient workload), and lipid-lowering treatment (time from high LDL-C to study start, very high LDL-C >30 mg/dL above goal, lipid-lowering therapy, maximum dose of lipid lowering therapy, statin treatment >50% of follow-up).

After adjustment for these factors, black women, black men, and non-black women were still less likely to have their LDL-C controlled than non-black men, black women by 40% and the other 2 groups by 20%-25%. Although baseline data suggested that this might be due to clinical factors or lipid-lowering therapy, this association was found to be unchanged after these adjustments.

This was a small study, and the results might have been affected by unrecorded refusal of treatment or adherence failure, Dr. Turner acknowledged. Access to drugs might have been limited by insurance plan. Side effects and allergies were not recorded. Determination of LDL-C control might have been delayed in patients with less frequent visits, although non-black men achieved control more quickly than the other race/gender groups despite having fewer visits scheduled.

The results highlight the need to focus on supporting medication acceptance/adherence in women and black men, Dr. Turner concluded. "I think women don't buy into the construct of cardiovascular disease and needing to take these drugs most of the time," she suggested. She also noted that based on her experience, many African American patients are biased against taking statins. "There are a lot of patients coming in talking about how it 'kills your liver,' 'we have heard it on television,' 'people have muscle problems' -- it has really gotten through the community," she said. "So I think we have a lot of work to do to try to get patients more engaged and over the hump of the fear of the side effects of statins. Although they appeared not to be very high in this study, they may be one of the big deterrents."

Simvastatin But Not Pravastatin Associated With Sleep Impairment

Another reason why patients might stop taking lipid-lowering therapy may be sleep disturbance, which has been associated with statins. The results of the largest randomized clinical trial to date of the effects of statins on sleep appear to corroborate previous evidence suggesting that lipophilic statins may affect sleep in some individuals.[17] Presenting her findings at the 2007 AHA meeting, lead investigator Beatrice A. Golomb, MD, PhD (University of California, San Diego School of Medicine), stressed the importance of sleep to health (eg, insulin resistance), function (including cognition, memory consolidation), mood (including irritability), well-being (including tiredness), and safety (eg, accidents). She explained that lipophilic statins, such as lovastatin and simvastatin, have been implicated in sleep disturbance in a range of studies.

Case reports and series have suggested possible effects of statins on sleep in some subjects and insomnia has been reported as an adverse effect in clinical trials of statins. Most randomized studies that directly evaluated the impact of statins on sleep, however, found no effect,[18,19] and only 1 showed a significant difference between sleep patterns in subjects on simvastatin or pravastatin, although neither statin differed from placebo.[20] Since most of these studies used a small sample size and were of short duration (≤6 weeks), the question of whether statins on average affect sleep favorably or adversely remained unclear.

Dr. Golomb and her colleagues at the University of California, San Diego (UCSD) examined the effects of simvastatin and pravastatin on sleep as a prespecified secondary endpoint of the UCSD Statin Study, a double-blind, randomized, placebo-controlled trial set up to ascertain the beneficial or adverse effects of statins on a set of noncardiac endpoints, including cognition, behavior, and serotonin biochemistry.[21] The study enrolled 1016 subjects aged ≥20 years (postmenopausal, if female), without known cardiovascular disease or diabetes, and with LDL-C between 115 and 190 mg/dL and fasting blood glucose <140 mg/dL. Subjects were randomized to placebo, simvastatin 20 mg, or pravastatin 40 mg (equipotent LDL-C-lowering doses for drug arms with simvastatin and pravastatin chosen to represent the extremes of the lipophilicity spectrum/the most hydrophilic and lipophilic statins) for 6 months of treatment followed by 2 months' post cessation follow-up. The study is funded by the National Heart, Lung, and Blood Institute (NHLBI) and UCSD.

Sleep quality was measured at baseline and on treatment with a modification of the Leeds Sleep Evaluation Questionnaire (using a visual analog scale 0-30), a subjective sleep measure that was used in the study that previously demonstrated differences between simvastatin and pravastatin.[20] Sleep problems were measured using a rating scale from -2 (much worse) to +2 (much better). Tiredness, irritability, and cognitive impairment, all of which are associated with sleep impairment, were also assessed on scales of 0-10.

Groups were comparable at baseline on all variables, including both sleep measures. At 6 months, both the statin treatment groups showed the expected reductions in LDL-C compared with placebo, although reductions in total cholesterol, LDL-C, and triglycerides were greater with simvastatin than with pravastatin ( ).

  Changes in Cholesterol at 6 Months

Agent TC LDL-C Triglycerides HDL-C
Placebo -2.1 -1.1 -2.1 -1.8
Pravastatin -47** -36** -13* -1.8
Simvastatin -57** -49** -26** -1.9

HDL-C = high-density lipoprotein cholesterol; LDL-C = low-density lipoprotein cholesterol; TC = total cholesterol;
*P<.05
**P <.0001 vs placebo

Simvastatin use was associated with worse sleep quality, and greater reported sleep problems than either pravastatin or placebo ( ). Although the differences on the rating scale were small, subjects on pravastatin actually perceived their sleep quality as being better than average (>15 on the rating scale) whereas those on simvastatin perceived theirs to be slightly worse. Similar results were seen for sleep problems.

  Effects of Statins on Sleep (means on rating scales)

Agent Sleep Quality
(0 to 30)
Sleep Problems
(-2 to 2)
Placebo 15.50 0.074
Pravastatin 15.79 0.098
Simvastatin 14.67 0.213

These differences between simvastatin and pravastatin or placebo were significant when analyzed by t-test of mean on-treatment sleep scores across randomization groups by t-test and regression analyses adjusted for baseline values of the respective sleep assessment ( ). Pravastatin did not differ significantly from placebo on any sleep outcome. In a subset of patients who reported much worse or better sleep in terms of sleep problems, only half of the patients on simvastatin as on placebo reported a "much better" rating, and 2.7 times as many reported that their sleep problems were much worse.

  Comparison of Sleep Scores by t-Test and Regression Analyses (P values)

Agent Comparison Sleep Quality Sleep Problems
t-test Regression t-test Regression
Simvastatin vs placebo .026 .036 .012 .007
Simvastatin vs pravastatin .003 .004 .037 .038
Pravastatin vs placebo .43 .37 .055 .048

Sleep quality and problems were all found to be associated with tiredness, irritability, and cognition, with stronger relationships between sleep quality and tiredness and between sleep problems and irritability and cognitive impairment.

Dr. Golomb acknowledged the study limitations: only 1 lipophilic and 1 hydrophilic statin were investigated, dose response was not examined; and it remains unknown whether the effects are modified in groups excluded from the study, such as subjects with diabetes or heart disease. However, she suggested that the difference observed between simvastatin and pravastatin in this study may due to their respective lipophilic or hydrophilic properties, although the differences in lipid reduction or other differences may be involved in the differences in sleep disturbance.

Asked about the mechanisms that might be involved, Dr. Golomb noted that in primates cholesterol lowering, although not by statins, has been shown to reduce levels of central serotonin, a precursor of melatonin, the sleep-regulating hormone; statins have also been shown to alter the ratio of omega-3 to omega-6 fatty acids (omega-3 fatty acids are believed to be important for sleep); or, as observed by Dr. Golomb's group, in some patients statins lead to mitochondrial dysfunction, which is associated with sleep-disordered breathing.

References

  1. The Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383-1389. Abstract

  2. Shepherd J, Cobbe SM, Ford I, et al; the West of Scotland Coronary Prevention Study Group. Prevention of coronary heart disease in men with hypercholesterolemia. N Engl J Med. 1995;333:1301-1307. Abstract

  3. Sacks FM, Pfeffer MA, Moye LA, et al; the Cholesterol and Recurrent Events Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol. N Engl J Med. 1996;335:1001-1009. Abstract

  4. The Long-term intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med. 1998;339:13490-1357.

  5. Downs JR., Clearfield M, Weis S, et al; the AFCAPS/TexCAPS Research Group. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS. JAMA. 1998;279:1615-1622. Abstract

  6. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20 536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360:7-22. Abstract

  7. Avorn J, Monette J, Lacour A, et al. Persistence of use of lipid-lowering medications.A cross-sectional study. JAMA. 1998;279:1458-1462. Abstract

  8. Benner JS, Glynn RJ, Mogun H, et al. Long-term persistence in use of statin therapy in elderly patients. JAMA. 2002;288:455-461. Abstract

  9. Jackevicius CA, Mamdani M, Tu JV. Adherence with statin therapy in elderly patients with and without acute coronary syndromes. JAMA. 2002;288:462-467. Abstract

  10. Simons LA, Simons LA, Levis G, Simons J. Apparent discontinuation rates in patients prescribed lipid-lowering drugs. Med J Aust. 1996;164:208-211. Abstract

  11. Davidson MH, Sun SX, Nag SS, et al. Risk factors for discontinuation of statin medications: evidence from a large national pharmacy claims database. Circulation. 2007;116(16 suppl):II-847. Abstract 3726.

  12. Turner BJ, Hollenbeak CS, Weiner MG, Roberts CS. Meeting quality of care goals for hypercholesterolemia: Understanding deficiencies for women and black men compared with non-black men. Circulation. 2007;116(16 suppl):II-847. Abstract 3727

  13. Massing MW, Foley KA, Carter-Edwards L, et al. Disparities in lipid management for African Americans and Caucasians with coronary artery disease: a national cross-sectional study. BMC Cardiovasc Disord. 2004;4:15.

  14. Goff DC Jr, Bertoni AG, Kramer H, et al. Dyslipidemia prevalence, treatment, and control in the Multi-Ethnic Study of Atherosclerosis (MESA): gender, ethnicity, and coronary artery calcium. Circulation. 2006;113:647-656. Abstract

  15. National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III): Final report. Circulation. 2002;106;3143-3421. Abstract

  16. Grundy SM, Cleeman JI, Merz CN, et al; American College of Cardiology Foundation; American Heart Association. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation. 2004;110:227-239. Abstract

  17. Golomb BA, Kwon EK, Criqui MK, Dimsdale JE. Simvastatin but not pravastatin affects sleep: findings from the UCSD Statin Study. Circulation. 2007;116(16 suppl):II-847. Abstract 3725

  18. Keech AC, Armitage JM, Wallendszus KR, et al; Oxford Cholesterol Study Group. Absence of effects of prolonged simvastatin therapy on nocturnal sleep in a large randomized placebo-controlled study. Br J Clin Pharmacol. 1996;42:483-490. Abstract

  19. Santanello NC, Barber BL, Applegate WB, et al. Effect of pharmacologic lipid lowering on health-related quality of life in older persons: results from the Cholesterol Reduction in Seniors Program (CRISP) Pilot Study. J Am Geriatr Soc. 1997;45:8-14. Abstract

  20. Harrison RW, Ashton CH. Do cholesterol-lowering agents affect brain activity? A comparison of simvastatin, pravastatin, and placebo in healthy volunteers. Br J Clin Pharmacol. 1994;37:231-236. Abstract

  21. Golomb BA, Criqui MH, White HL, Dimsdale JE. The UCSD Statin Study: a randomized controlled trial assessing the impact of statins on selected noncardiac outcomes. Control Clin Trials. 2004;25:178-202. Abstract