Title: Differences in preventing new-onset cardiovascular events with statin therapy in seniors aged 75

years and over: a cohort study in the South Korean National Health Insurance Database
Accepted Article
Running title: Statin for preventing new-onset cardiovascular events in seniors aged ≥ 75 years

Authors’ full names: Kyungim Kim, Ph.D.1,2*, Arim Kwak, M.S.1, Cheol Ung Choi, MD, Ph.D.3, Jae

Hyun Kim, Ph.D.4, Myeong Gyu Kim, Ph.D.5, Jung Mi Oh, Pharm.D. 4, Eunhee Ji, Ph.D.6

Affiliations of authors:
1
College of Pharmacy, Korea University, Sejong, South Korea; 2Biomedical Research Center, Korea

University Guro Hospital, Seoul, South Korea; 3Cardiovascular Center, Korea University Guro Hospital,

Seoul, Korea; 4College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National

University, Seoul, South Korea; 5Graduate School of Clinical Pharmacy, CHA University, Pocheon,

South Korea; 6College of Pharmacy, Gachon University, Incheon, South Korea

(Received 3 December 2018; Accepted 19 February 2019)

Correspondence:

Kyung Im Kim, Ph.D.

College of Pharmacy, Korea University

251l Sejong-ro, Sejong, 30019, Republic of Korea

Telephone: +82-44-860-1624

Fax: +82-44-860-1606

E-mail: kim_ki@korea.ac.kr

Keywords: Cardiovascular, Older people, Primary prevention, Rational pharmacotherapy, Statin, South

Korea

This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/bcpt.13229
This article is protected by copyright. All rights reserved.
 Abstract
The aim of this cohort study was to compare the effectiveness of statin regimens for primary prevention
Accepted Article
among seniors aged ≥ 75 years. Seniors aged 75-100 years for whom statin therapies for primary

prevention was newly initiated between 1 January 2009 and 31 December 2011, and who continued the

same statin regimen during the first year after the index date were identified using the claims data from

the South Korean National Health Insurance Database. A propensity score matching and multivariable

Cox proportional hazards model was developed to evaluate adjusted ischaemic cardiovascular-

cerebrovascular event (CCE) risk and all-cause mortality risk for all patients, as well as for subgroups. A

total of 5,629 older patients aged 75-100 years were included in the study population. Compared to

moderate-intensity statin therapy, low-intensity statin therapy was significantly associated with increased

risk of ischaemic CCEs, while high-intensity statin therapy was associated with reduced risk of ischaemic

CCEs; however, compared to moderate-intensity statin therapy, both low-intensity and high-intensity

statin therapies were associated with increased risk of all-cause mortality. For the 4,689 older patients

who regularly received moderate-intensity statin therapy including 10 mg atorvastatin, 20 mg atorvastatin,

10 mg rosuvastatin, or 20 mg simvastatin for primary prevention, multivariable regression adjusting for

potential covariates revealed no significant difference in ischaemic CCEs or all-cause mortality between

the moderate-intensity statin users and 10 mg atorvastatin users both before and after propensity scoring

matching. No significant heterogeneity was detected in the patient subgroups. The results of this study

based on real-world data can supply evidence-based reasons for choice of statin regimen for the primary

prevention of CCEs in older people aged ≥ 75 years.

INTRODUCTION

The elderly population is increasing steadily worldwide. The percentages of people aged older

than 65 years and older than 80 years are expected to nearly double and triple, respectively, between 2015

and 2050.[1] As cardiovascular disease (CVD) is more common among the elderly than among younger

individuals and is responsible for approximately 40% of all deaths of individuals over 75 [2], statin use

for cardiovascular prevention has focused on this age group.

This article is protected by copyright. All rights reserved.

 Although there is still debate concerning the effectiveness of statins for preventing new-onset

cardiovascular events in seniors aged 75 years or over, some large clinical studies have supported statin
Accepted Article
use to reduce CVD morbidity in elderly populations without previous cardiovascular events, that is, in the

context of primary prevention.[3-5] A recent meta-analysis study also revealed that primary prevention

with statin therapy was associated with significant reduction in the risk of atherosclerotic cardiovascular

events including non-fatal myocardial infarctions, non-fatal strokes or cardiovascular death in seniors

aged ≥ 70 years.[6] Additionally, Odden et al. determined the cost-effectiveness of statins for primary

prevention in people aged ≥ 75 years.[7] In fact, a significant number of seniors take statins for primary

cardiovascular prevention. In a study at the Geisinger Health System in Pennsylvania, 28% of patients

aged 75 to 79 years and 22% of patients aged ≥ 80 years were found to take statins for primary

prevention.[8] Another study that analysed the Medical Expenditure Panel Survey reported that the rate of

statin use for primary prevention in adults older than 79 years increased from 8.8% in 1999-2000 to 34.1%

in 2011-2012.[9]

Owing to the practical difficulty of enrolling seniors aged 75 years and over in clinical trials,

however, there has been no clinical trial in this age group directly comparing the effectiveness of different

statin therapies for primary prevention. The direct application of efficacy data associated with patients

under 75 years of age to seniors aged 75 and over should be undertaken cautiously considering the altered

relationship between cholesterol levels and cardiovascular outcomes caused by the biological changes

associated with aging, polypharmacy, and limited life expectancy.[10, 11] To address this, the objective of

this study was to compare the effectiveness of different statin therapies for seniors aged 75 years and over

for primary prevention using real-world data.

MATERIALS AND METHODS

The study was conducted in accordance with the Basic & Clinical Pharmacology & Toxicology

policy for experimental and clinical studies.[12]

This article is protected by copyright. All rights reserved.

 Data source

In this retrospective cohort study, the claims data from the South Korean National Health
Accepted Article
Information Database (NHID) was reviewed for eligible patients. The NHID is a public database on

health care utilization, health screening, sociodemographic variables and mortality for the whole

population of South Korea, formed by the National Health Insurance Service (NHIS).[13] The database

contains longitudinal patient information regarding patient demographics, diagnoses [International

Classification of Disease, Tenth Revision (ICD-10)], procedures, prescription drugs (brand name, generic

name, prescription date, days of supply, dose and route of administration), and type of medical utilization

(outpatient, inpatient or emergency department). As health insurance in Korea is mandatory for all ages,

and all statins were fully reimbursed during the study period, the current study was not biased by

differing types of health insurance or differing financial incentives on the part of patients.

Study population

All the older people aged 75 to 100 years who were prescribed at least one statin between 1

January 2009 and 31 December 2011 were identified and their claim data from 1 January 2006 through

31 December 2014 were obtained from the NHID. A new statin user was defined as a patient who had not

been prescribed any statin from 2006 to the date of the first statin prescription (index date). Patients were

excluded if during the year prior to the index date they had a history of cancer diagnosis, lipid lowering

drug use other than statins, or used multiple statins concurrently on one prescription. A one-year interval

(selection period) between the index date and the start of follow-up was chosen because the benefits in

terms of vascular events are expected to start approximately one year after the initiation of statin

therapy.[14] Patients who switched to another dose of the initial statin or to another statin or who

discontinued statin therapy during the selection period were excluded from the study population, as were

those who died within the selection period. Discontinuation of the statin therapy was defined as 12

consecutive weeks without any statin prescription. The grace period of 12 weeks was defined considering

the half-lives of statins and the prescription intervals of statin therapy in Korea. For the analysis of the

primary prevention effects of statin therapy, the patients with a history of ischaemic heart disease (ICD-

10 codes 120-25), cerebrovascular disease (I60-69), diseases of arteries, arterioles or capillaries (I70-73)

This article is protected by copyright. All rights reserved.

 or any thoracic procedure (such as angioplasty, coronary artery bypass grafting, or percutaneous coronary

intervention) from 3 years before the index date to the end of the selection period were also excluded.
Accepted Article
After that, the study population was grouped into high-, moderate-, or low-intensity statin user groups

according to their statin therapy regimen. The intensity of statin therapy was defined according to the

average expected low-density lipoprotein cholesterol response to a specific statin and dose as given in the

American College of Cardiology (ACC)/American Heart Association (AHA) guideline.[15]

Study outcomes & follow-up

The study outcomes were as follows: (1) ischaemic cerebrovascular-cardiovascular event (CCE)

and (2) all-cause mortality. An ischaemic CCE was defined as any hospitalization with a primary

diagnosis of acute ischaemic heart disease (ICD-10 codes: I21-24), ischaemic stroke (I63, I65, I66), or

any procedure involving coronary artery bypass grafts or percutaneous coronary intervention, whichever

occurred earliest. The follow-up started after the one-year selection period and continued until the

development of the study outcomes, discontinuation of the initial statin therapy, or by December 2014 at

the latest, whichever occurred first. The discontinuation date of the statin therapies was considered to be

the anticipated final date of the last statin prescription. The patient was also considered to have

discontinued statin therapy if they switched to a different statin therapy from the initial regimen.

Statistical analysis

Statistical analyses were performed separately between 1) differing lipid-lowering intensities of

statins (moderate-intensity as the control group) and 2) differing types of moderate-intensity statins with

10 mg atorvastatin as the control group. Comparisons of baseline characteristics (categorical covariates)

were performed using Fisher’s exact test between the comparative groups.

To investigate the risk of ischaemic CCE or all-cause mortality, the Cox proportional hazard

model was used to adjust the covariates of age, gender, the Charlson comorbidity index (CCI) score,

comorbidities at baseline, and adherence to statin therapy and concomitant medication use during the

selection and follow-up period. The covariates were identified based on literature, expert opinion and the

availability of covariates within the data. The CCI score was estimated from the disease record using

This article is protected by copyright. All rights reserved.

 previously validated algorithms.[16] Adherence to statin therapy was estimated by calculating the

medication possession ratio (MPR, ratio of sum of the days of supply divided by the total number of days
Accepted Article
of the selection and follow-up period), and a ratio ≥ 80% was considered to represent adherence. A cutoff

of 80% for the MPR is chosen to correspond to the statin adherence determined by previous studies as

sufficient for the treatment to achieve a satisfactory clinical effect.[17-19] Medication use included

antihypertensive drugs (angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists, β-

adrenergic antagonists, calcium channel blockers or thiazide diuretics), antidiabetic drugs (metformin,

sulfonylureas, dipeptidyl peptidase-4 inhibitors, thiazolidinediones or other hypoglycemic agents), aspirin,

other antiplatelet drugs, and oral anticoagulants (warfarin and other anticoagulants). Kaplan–Meier

survival curves were plotted to compare the event-free survival trend between groups. Subgroup analyses

were also performed to determine whether the results were consistent across subgroups differing by age,

gender, CCI score, comorbidities at baseline, adherence to statin therapy or comedication.

For the patients on moderate-intensity statin therapies, propensity score matching was also used

to balance observed covariates between the comparative groups in the proportion 1:1. The clinical risk

factors, age, gender, CCI score and comorbidities at baseline were added into a non-parsimonious

multivariable logistic regression model. The predicted probability derived from the logistic equation was

used for the propensity score matching for each individual. Users on 10 mg atorvastatin therapy were

randomly matched to other statin regimen users (patients from 20 mg atorvastatin, 10 mg rosuvastatin,

and 20 mg simvastatin use groups, respectively). If no appropriate propensity score match could be found

within the comparative groups for an individual subject, that subject was excluded from further analysis.

All the two-sided p-values are reported for each analysis, and the results are considered

statistically significant for p < 0.05. All statistical analyses were performed using SAS software version

9.4 (SAS Institute, Inc., Cary, NC, USA).

This article is protected by copyright. All rights reserved.

 Ethics approval

This study was approved by the institutional review board of Korea University (KU-IRB-16-29-A-1), and
Accepted Article
admitted by the Korea NHIS Medical Information Disclosure Committee (NHIS-2018-1-234).

RESULTS

Study population characteristics

A total of 177,505 older patients aged 75-100 years who were prescribed at least one statin

therapy during 2009-2011 were identified from the NHID. The 5,629 older patients chosen by the

selection process who were aged 75-100 years and regularly received initial statin therapy for the primary

prevention of CCE during the 1-year selection period were included in the study population (Fig. 1).

From among the moderate-intensity group, patients on 5 mg rosuvastation, 40 mg simvastatin, 80 mg

fluvastatin, 2 mg pitavastatin and 40 mg pravastatin were excluded from the further analysis because the

number of patients under the same statin regimen was too small. Thus, the remaining 4,689 older patients

who regularly received 10 mg atorvastatin, 20 mg atorvastatin, 10 mg rosuvastatin or 20 mg simvastatin

were included in the analysis comparing different moderate-intensity statin regimens. The basic

demographics of the study population before propensity score matching are summarized in Appendix

Table 1. The entire study population received mostly moderate-intensity statins and the mean age was

79.5 years (±4.0 years). Among the moderate-intensity statin regimens, 10 mg atorvastatin (60.1%) was

the most commonly prescribed statin before propensity score matching. After propensity score matching,

the total analytical sample for the comparison between moderate-intensity statin regimens numbered

1,292 seniors with 323 in each of the moderate-intensity statin groups (Table 1). Most of the baseline

characteristics were well-balanced between the comparative groups.

Comparisons of preventive effect between statin intensity groups

For comparative analysis of statin intensities, propensity score matching was not performed

because the number of patients with low- or high-intensity statin regimens was too small (5,143 patients

with a moderate-intensity statin regimen, 288 patients with a low-intensity statin regimen and 198

patients with a high-intensity statin regimen). During the follow-up period, a total of 556 patients (5.8%)

This article is protected by copyright. All rights reserved.

 reported ischaemic CCE, of which ischaemic stroke was the most common for all statin groups: 23

patients (8.0%) in the low-intensity statin group, 302 (5.9%) in the moderate-intensity statin group and 3
Accepted Article
patients (1.5%) in the high-intensity statin group. Cox proportional hazard regression analysis indicated

that, compared to moderate-intensity statin therapy, low-intensity statin therapy was significantly

associated with increased risk of ischaemic CCEs, while high-intensity statin therapy was associated with

reduced risk of ischaemic CCEs (HR 1.41 [1.02-1.95], p = 0.040 in low-intensity statin group; HR 0.36

[0.17-0.76], p = 0.007 in high-intensity statin group). However, compared to moderate-intensity statin use,

both low- and high-intensity statin use significantly increased the risk of all-cause mortality (HR 1.54

[1.18-2.01], p = 0.001; HR 1.43 [1.02-2.00], p = 0.038, respectively) (Table 2, Fig. 2).

Comparisons of the preventive effect of different statin types with moderate-intensity regimens

After propensity score matching of the 1,452 patients, a total of 119 (9.2%) reported ischaemic

CCE during the follow-up period, with ischaemic stroke being the most common event for all statin

groups: 17 patients (5.3%) in the 10 mg atorvastatin group, 25 (7.7%) in the 20 mg atorvastatin group, 9

(2.8%) in the 10 mg rosuvastatin group and 15 (4.6%) in the 20 mg simvastatin group. Due to all-cause

mortality, 136 patients in total (10.5%) died during the follow-up period. Multivariable regression

adjusting potential covariates showed that any of the moderate-intensity statin regimens were not

significantly different from 10 mg atorvastatin for preventing ischaemic CCE or all-cause mortality

(Table 3). The Kaplan–Meier survival curve also showed that there was no significant difference between

the groups in terms of ischaemic CCE-free survival and overall survival (p = 0.240 and 0.827,

respectively; Fig. 2). The sub-group analysis showed that the protective effects of the differing statin

types against ischaemic CCE or all-cause mortality were not significantly different from those of 10 mg

atorvastatin (Appendix Table 2).

DISCUSSION

Using the national real-world data, this study provided evidence that (1) compared to moderate-

intensity statin therapy, low-intensity statin therapy was significantly associated with increased risk of

developing ischaemic CCE and all-cause mortality, while high-intensity statin therapy was significantly

This article is protected by copyright. All rights reserved.

 associated with reduced risk of ischaemic CCEs but increased risk of all-cause mortality (2). The

moderate-intensity regimens of all the statin-types studied were as effective as atorvastatin therapy of the
Accepted Article
same intensity for preventing the development of ischaemic CCE or all-cause mortality in seniors aged ≥

75 years.

The results of the statin intensity comparison study show that for seniors aged 75 years and over,

while there is an intensity-dependent association between statin use and prevention of new-onset

ischaemic CCEs, both low- and high-intensity statin therapy significantly increased all-cause mortality

compared to moderate-intensity statin therapy. This intensity-independent result for all-cause mortality

could be explained by ‘a reverse J-shaped’ or ‘an inverted U-shaped’ association which has been reported

by many observational studies, where low cholesterol levels were found to be associated with high all-

cause mortality in populations 80 years old and above.[20] A recent post-hoc analysis of the

Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT)

randomized clinical trial also reported a trend toward increased all-cause mortality with pravastatin

compared to the usual care for seniors 75 years and older, although the result was not statistically

significant.[21] As yet, however, the optimal level of cholesterol necessary to prevent CCEs or all-cause

mortality in this age group has not been clearly identified. Therefore, a “risk-benefit” balance should be

considered before starting statin therapies for primary prevention, particularly for seniors older than 75

years.

The real-world-data-based finding that the protective effects of the moderate-intensity statin

regimens are equivalent is notable since no clinical trials have been conducted directly comparing the

effect of statin regimens on seniors aged 75 years and over. Although several published studies have

compared different statin regimens including rosuvastatin, atorvastatin or simvastatin, they were not

restricted to seniors ≥ 75 years old. A recent meta-analysis, which included 14 randomized, controlled

trials, found that the effectiveness of atorvastatin (mean daily dose, 16.7 mg) and rosuvastatin (mean

daily dose, 14.8 mg) did not differ for the primary prevention of major coronary events (OR = 1.44, 95%

CI = 0.73–2.83) and all-cause mortality (OR = 1.07, 95% CI = 0.51–1.87).[22] Another retrospective

observational cohort study on Dutch patients with no prior cardiovascular disease also found that, after

adjusting for effect modifiers, rosuvastatin users (mean daily dose, 10.8±3.3 mg) had similar incidence

This article is protected by copyright. All rights reserved.

 rates of cardiovascular disease (HR = 0.83, 95% CI = 0.63–1.10) or myocardial infarction (HR = 0.77, 95%

CI = 0.41–1.43) to atorvastatin users (mean daily dose, 17.3±10.8 mg).[23] Other studies comparing the
Accepted Article
efficacies of atorvastatin and simvastatin for the primary prevention of cardiovascular and

cerebrovascular diseases reported inconsistent results. For example, two previous studies by Foody et al.

and Jacobson et al. reported that moderate-intensity atorvastatin therapy was associated with significantly

fewer cardiovascular events than moderate- or high-intensity simvastatin therapy.[24, 25] In contrast, a

more recent retrospective analysis by Swindle et al. found results similar to our study. Using pharmacy

and medical claims from a US health plan database, they observed no statistically significant difference in

the risk of cardiovascular events between atorvastatin users and simvastatin users aged ≥ 65 years (HR =

1.078; 95% CI = 0.976–1.192).[26] The inconsistencies between the studies may be partly caused by the

endpoints examined, the characteristics of the patients enrolled and the methods used. They may also be

possibly explained by the availability of generics. Generic simvastatin became available in the US in June

2006, after the patient identification period in Foody et al. and shortly before the end of the patient

identification period in the Jacobson et al. study. In contrast, the patient identification period of our study

and the Swindle et al. study, which produced similar results, began after generic simvastatin had become

available in the US and Korea, respectively.

The findings of this study should be interpreted in light of its limitations. Firstly, as a claim data-

based analysis, our analysis shares the limitations of the nationwide database. Although we controlled for

the most important risk factors for ischaemic CCE, too few patients had available data on body mass

index, smoking, family disposition and other cardiovascular risk factors such as diet, physical activity and

baseline high density lipoprotein cholesterol, low density lipoprotein cholesterol, or C-reactive protein

levels, so these factors could not be included in model adjustment. Secondly, some important adverse

effects of statins could not be analysed since insufficient data were available for the study population. It is

important to consider statin-induced adverse effects before choosing statin regimens for seniors due to the

increased risk of comorbidity, polypharmacy and the associated risk of drug-drug interactions. Even small

adverse effects could offset the cardiovascular benefit of statin therapy and harm seniors due to age-

related factors. Thus, utilizing the results of this study for public health recommendation in practice

requires further analyses that can compensate for the lack of specific data on adverse effects or

This article is protected by copyright. All rights reserved.

 laboratory-derived data.

In spite of these limitations, this study provides important contributions worthy of mention.
Accepted Article
Firstly, it used a population-based, nationwide database and included nearly 100% of the older patients

and their medical information, which had been properly recorded under the regulations of national health

insurance. Secondly, the mean observation period, including the selection and follow-up, was 2.8 years

(±0.3 years). This period can be considered sufficiently long to identify the potential differences in the

clinical efficacy of statins. Thirdly, the adherence rate to statin therapy was also considered as a covariate

in the multivariable analysis of this study. It is well known that there is a close relationship between poor

adherence to statin therapy and worse clinical outcomes[27, 28], and individuals aged ≥ 75 years are

more likely to be more troublesome with respect to statin regimen adherence than those < 75 years

because of a number of sociodemographic, economic and other treatment-related factors.[29]

In conclusion, the results of this real-world-data-based study suggest that, with seniors aged 75

years and over, to prevent new-onset ischaemic CCE events and all-cause mortality, moderate-intensity

statins should be considered and that moderate-intensity statin regimens have a similar effectiveness

when compared to 10 mg atorvastatin. Further studies are required to determine whether the differing

outcomes of the statin regimens for this age group are due to their safety or other treatments and

demographic characteristics.

ACKNOWLEDGEMENTS

The authors would like to appreciate the South Korean National Health Insurance Service for the

provision of the claims data from the South Korean National Health Information Database.

FUNDING

This study was supported by the National Research Foundation, Korea (NRF-2017R1C1B5018232). The

funder had no role in the study design, data collection and analysis, decision to publish or preparation of

the manuscript.

This article is protected by copyright. All rights reserved.

 CONFLICT OF INTEREST

The authors have declared no conflicts of interest.

Accepted Article
REFERENCES

1. Wan He DG, Paul Kowal. International Population Reports, P95/16-1, An Aging World: 2015. In:

Bureau USC, editor. Washington, DC: U.S. Government Publishing Office; 2016.

2. National Heart, Lung, and Blood Institute. Motbidity and mortality: 2012 chart book on

cardiovascular, lung, and blood diseases: National Institute of Health; 2012.

https://www.nhlbi.nih.gov/files/docs/research/2012_ChartBook_508.pdf

3. Yusuf S, Bosch J, Dagenais G, Zhu J, Xavier D, Liu L, et al. Cholesterol Lowering in

Intermediate-Risk Persons without Cardiovascular Disease. N Engl J Med. 2016;374(21):2021-

31.

4. Ridker PM, Macfadyen JG, Nordestgaard BG, Koenig W, Kastelein JJ, Genest J, et al.

Rosuvastatin for primary prevention among individuals with elevated high-sensitivity c-reactive

protein and 5% to 10% and 10% to 20% 10-year risk. Implications of the Justification for Use of

Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial for

"intermediate risk". Circ Cardiovasc Qual Outcomes. 2010;3(5):447-52.

5. Glynn RJ, Koenig W, Nordestgaard BG, Shepherd J, Ridker PM. Rosuvastatin for primary

prevention in older persons with elevated C-reactive protein and low to average low-density

lipoprotein cholesterol levels: exploratory analysis of a randomized trial. Ann Intern Med.

2010;152(8):488-96, W174.

6. Ridker PM, Lonn E, Paynter NP, Glynn R, Yusuf S. Primary Prevention With Statin Therapy in

the Elderly: New Meta-Analyses From the Contemporary JUPITER and HOPE-3 Randomized

Trials. Circulation. 2017;135(20):1979-81.

7. Odden MC, Pletcher MJ, Coxson PG, Thekkethala D, Guzman D, Heller D, et al. Cost-

effectiveness and population impact of statins for primary prevention in adults aged 75 years or

older in the United States. Ann Intern Med. 2015;162(8):533-41.

This article is protected by copyright. All rights reserved.

 8. Chokshi NP, Messerli FH, Sutin D, Supariwala AA, Shah NR. Appropriateness of statins in

patients aged >/=80 years and comparison to other age groups. Am J Cardiol.
Accepted Article
2012;110(10):1477-81.

9. Johansen ME, Green LA. Statin Use in Very Elderly Individuals, 1999-2012. JAMA Intern Med.

2015;175(10):1715-6.

10. Mortensen MB, Falk E. Primary Prevention With Statins in the Elderly. J Am Coll Cardiol.

2018;71(1):85-94.

11. Liguori I, Aran L, Bulli G, Russo G, Curcio F, Sasso G, et al. Statins in cardiovascular

prevention in the oldest-old. A black hole. Journal of gerontology and geriatrics 2017;65:263-70.

12. Tveden-Nyborg P, Bergmann TK and Lykkesfeldt J. Basic & Clinical Pharmacology &

Toxicology Policy for Experimental and Clinical studies. Basic Clin Pharmacol Toxicol.

2018;123:233-25.

13. Seong SC, Kim YY, Khang YH, Heon Park J, Kang HJ, Lee H, et al. Data Resource Profile: The

National Health Information Database of the National Health Insurance Service in South Korea.

Int J Epidemiol. 2017;46(3):799-800.

14. Baigent C, Keech A, Kearney PM, Blackwell L, Buck G, Pollicino C, et al. Efficacy and safety

of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in

14 randomised trials of statins. Lancet. 2005;366(9493):1267-78.

15. Neil J. Stone, Jennifer Robinson, Alice H. Lichtenstein, C. Noel Bairey Merz, Conrad B. Blum,

Robert H. Eckel, et al. 2013 ACC/AHA Guideline on the Treatment of Blood Cholesterol to

Reduce Atherosclerotic Cardiovascular Risk in Adults: A Report of the American College of

Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation.

2014;129(25 Suppl 2):S1–45.

16. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for

defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care.

2005;43(11):1130-9.

17. Chi MD, Vansomphone SS, Liu IL, Cheetham C, Green KR, Scott RD, et al. Adherence to

statins and LDL-cholesterol goal attainment. Am J Manag Care. 2014;20(4):e105-12.

This article is protected by copyright. All rights reserved.

 18. Cramer JA, Roy A, Burrell A, Fairchild CJ, Fuldeore MJ, Ollendorf DA, et al. Medication

compliance and persistence: terminology and definitions. Value Health. 2008;11(1):44-7.

Accepted Article
19. Wei L, Wang J, Thompson P, Wong S, Struthers AD, MacDonald TM. Adherence to statin

treatment and readmission of patients after myocardial infarction: a six year follow up study.

Heart. 2002;88(3):229-33.

20. Petersen LK, Christensen K, Kragstrup J. Lipid-lowering treatment to the end? A review of

observational studies and RCTs on cholesterol and mortality in 80+-year olds. Age Ageing.

2010;39(6):674-80.

21. Han BH, Sutin D, Williamson JD, Davis BR, Piller LB, Pervin H, et al. Effect of Statin

Treatment vs Usual Care on Primary Cardiovascular Prevention Among Older Adults: The

ALLHAT-LLT Randomized Clinical Trial. JAMA Intern Med. 2017;177(7):955-65.

22. Naci H, Brugts JJ, Fleurence R, Tsoi B, Toor H, Ades AE. Comparative benefits of statins in the

primary and secondary prevention of major coronary events and all-cause mortality: a network

meta-analysis of placebo-controlled and active-comparator trials. European journal of preventive

cardiology. 2013;20(4):641-57.

23. Heintjes EM, Penning-van Beest FJ, Johansson S, Stalenhoef AF, Herings RM. Comparison of

incidences of cardiovascular events among new users of different statins: a retrospective

observational cohort study. Current medical research and opinion. 2009;25(11):2621-9.

24. Foody JM, Joyce AT, Rudolph AE, Liu LZ, Benner JS. Cardiovascular outcomes among patients

newly initiating atorvastatin or simvastatin therapy: a large database analysis of managed care

plans in the United States. Clinical therapeutics. 2008;30(1):195-205.

25. Jacobson TA, Wertz DA, Hoy T, Kuznik A, Grochulski D, Cziraky M. Comparison of

cardiovascular event rates in patients without cardiovascular disease in whom atorvastatin or

simvastatin was newly initiated. Mayo Clinic proceedings. 2008;83(12):1316-25.

26. Swindle JP, Potash J, Kulakodlu M, Kuznik A, Buikema A. Drug utilization patterns and

cardiovascular outcomes in elderly patients newly initiated on atorvastatin or simvastatin. The

American journal of geriatric pharmacotherapy. 2011;9(6):471-82.

27. Chowdhury R, Khan H, Heydon E, Shroufi A, Fahimi S, Moore C, et al. Adherence to

This article is protected by copyright. All rights reserved.

 cardiovascular therapy: a meta-analysis of prevalence and clinical consequences. European heart

journal. 2013;34(38):2940-8.
Accepted Article
28. Ho PM, Magid DJ, Shetterly SM, Olson KL, Maddox TM, Peterson PN, et al. Medication

nonadherence is associated with a broad range of adverse outcomes in patients with coronary

artery disease. Am Heart J. 2008;155(4):772-9.

29. Ofori-Asenso R, Jakhu A, Curtis AJ, Zomer E, Gambhir M, Jaana Korhonen M, et al. A

Systematic Review and Meta-analysis of the Factors Associated With Nonadherence and

Discontinuation of Statins Among People Aged >/=65 Years. The journals of gerontology Series

A, Biological sciences and medical sciences. 2018.

Figure legend

Figure 1. Study population flow chart. ASCVD, atherosclerotic cardiovascular disease

Figures 2. Kaplan-Meier survival analysis of the protective effect of statin regimens on the

development of ischaemic CCE and all-cause mortality. Comparison of (a) ischaemic CCE-free

survival for regimens of differing statin intensity (moderate-intensity as the control group); (b) overall

survival for regimens of differing statin intensity (moderate-intensity as the control group); (c) ischaemic

CCE free-survival for moderate-intensity regimens of differing statin type (10 mg atorvastatin as the

control group); (d) overall survival for moderate-intensity regimens of differing statin type (10 mg

atorvastatin as the control group)

This article is protected by copyright. All rights reserved.

 Table 1 Characteristics of moderate-intensity statin users after propensity score matching

Atorvastatin Atorvastatin Rosuvastatin Simvastatin

Accepted Article
Variable, n (%) 10 mg 20 mg 10 mg 20 mg p-value
(n=323) (n=323) (n=323) (n=323)

Age, years* 0.998

75-79 206 (63.8) 206 (63.8) 204 (63.2) 206 (63.8)

≥ 80 117 (36.2) 117 (36.2) 119 (36.8) 117 (36.2)

Gender* 1.000

Male 85 (26.3) 85 (26.3) 84 (26.0) 85 (26.3)

Female 238 (73.7) 238 (73.7) 239 (74.0) 238 (73.7)

CCI score* 1.000

0 56 (17.3) 55 (17.0) 58 (18.0) 56 (17.3)

1 103 (31.9) 103 (31.9) 101 (31.3) 102 (31.6)

2 60 (18.6) 60 (18.6) 61 (18.9) 62 (19.2)

≥3 104 (32.2) 105 (32.5) 103 (31.9) 103 (31.9)

Comorbidities*

Hypertension 288 (89.2) 288 (89.2) 286 (88.5) 288 (89.2) 0.992

Diabetes Mellitus 181 (56.0) 182 (56.3) 182 (56.3) 182 (56.3) 1.000

Atrial fibrillation 16 (5.0) 16 (5.0) 17 (5.3) 16 (5.0) 0.997

AntiHTN drug use†

ACEI 41 (12.7) 42 (13.0) 42 (13.0) 40 (12.4) 0.995

ARB 129 (39.9) 121 (37.5) 137 (42.4) 114 (35.3) 0.276

ARB+ Diuretics§ 73 (22.6) 69 (21.4) 76 (23.5) 63 (19.5) 0.632

BB 100 (31.0) 91 (28.2) 108 (33.4) 75 (23.2) 0.029

CCB 173 (53.6) 162 (50.2) 145 (44.9) 195 (60.4) 0.001

CCB+ARB§ 43 (13.3) 41 (12.7) 29 (9.0) 34 (10.5) 0.280

Diuretics 143 (44.3) 146 (45.2) 145 (44.9) 126 (39.0) 0.340

AntiDM drug use†

This article is protected by copyright. All rights reserved.

 Metformin 49 (15.2) 49 (15.2) 42 (13.0) 59 (18.3) 0.323

Sulfonylurea 46 (14.2) 52 (16.1) 50 (15.5) 64 (19.8) 0.255

Accepted Article
DPP4I 14 (4.3) 18 (5.6) 13 (4.0) 10 (3.1) 0.478

Thiazolidinedione 12 (3.7) 14 (4.3) 8 (2.5) 20 (6.2) 0.122

Other DM drugs 14 (4.3) 11 (3.4) 19 (5.9) 12 (3.7) 0.417

Anticoagulant use† 7 (2.2) 10 (3.1) 3 (0.9) 7 (2.2) 0.290

Aspirin use† 43 (13.3) 60 (18.6) 73 (22.6) 48 (14.9) 0.009

Other antiplatelet use† 12 (3.7) 23 (7.1) 23 (7.1) 10 (3.1) 0.028

Abbreviations: ACEI, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; BB, beta
blocker; CCB, calcium channel blocker; CCI, Charlson comorbidity index; DM, diabetes mellitus; DPP4I,
dipeptidyl peptidase-4 inhibitor; HTN, hypertension
*Clinical risk factors, such as age, gender, CCI score and comorbidities at baseline were added into a non-
parsimonious multivariable logistic regression model for propensity score matching.
†Drug use of study subjects was considered if the MPR of each drug was ≥ 80% during the selection and follow-up
periods.
§
Combination drugs

This article is protected by copyright. All rights reserved.

 Table 2 Hazard ratios of developing ischaemic CCE or all-cause mortality for each intensity statin regimen

group
Accepted Article
Incidence/
Group Number Event (%) HR* (95% CI) p-value
1000 person-years

Ischaemic CCE

Moderate-intensity 5,143 510 (9.9) 48.4 1.00 (reference)

Low-intensity 288 39 (13.5) 35.4 1.41 (1.02-1.95) 0.040

High-intensity 198 7 (3.5) 12.6 0.36 (0.17-0.76) 0.007

All-cause mortality

Moderate-intensity 5,143 581 (11.3) 40.3 1.00 (reference)

Low-intensity 288 51 (17.7) 63.2 1.54 (1.18-2.01) 0.001

High-intensity 198 35 (17.7) 63.1 1.43 (1.02-2.00) 0.038

Abbreviations: CCE, cerebrovascular and cardiovascular events; CI, confidence interval; HR, hazard ratio; PSM,
propensity score matching
*Adjusted hazard ratio calculated using Cox proportional hazard model adjusting for gender, age, Charlson
comorbidity index score, comorbidities at baseline, and adherence to statin and comedication use
(antihypertensive drugs, antidiabetic drugs, aspirin, other antiplatelet and anticoagulants) during the selection and
follow-up periods.

This article is protected by copyright. All rights reserved.

Accepted Article

Table 3 Hazard ratios of developing ischaemic CCE or all-cause mortality by statin-type under moderate-intensity regimens
Before PSM After PSM
Group
Event Incidence/ HR* Event Incidence/ HR*
Number p-value Number p-value
(%) 1000 person-years (95% CI) (%) 1000 person-years (95% CI)
Ischaemic CCE
Atorvastatin 10 mg 2,816 273 (9.7) 34.8 1.00 (reference) 323 24 (7.4) 26.4 1.00 (reference)
Atorvastatin 20 mg 728 75 (10.3) 37.1 1.02 (0.79-1.32) 0.895 323 37 (11.5) 41.6 1.55 (0.92-2.60) 0.097
Rosuvastatin 10 mg 398 32 (8.0) 28.9 0.79 (0.54-1.14) 0.206 323 26 (8.0) 28.9 1.05 (0.60-1.84) 0.872
Simvastatin 20 mg 747 78 (10.4) 37.4 0.99 (0.77-1.28) 0.948 323 32 (9.9) 35.5 1.23 (0.72-2.11) 0.445
All-cause mortality
Atorvastatin 10 mg 2,816 323 (11.5) 41.2 1.00 (reference) 323 33 (10.2) 36.4 1.00 (reference)
Atorvastatin 20 mg 728 85 (11.7) 42.0 0.98 (0.79-1.21) 0.832 323 37 (11.4) 41.6 1.15 (0.75-1.76) 0.531
Rosuvastatin 10 mg 398 45 (11.3) 40.7 0.83 (0.62-1.12) 0.224 323 35 (10.8) 38.9 0.97 (0.62-1.52) 0.898
Simvastatin 20 mg 747 72 (9.6) 34.5 0.80 (0.63-1.01) 0.056 323 31 (9.6) 34.4 0.95 (0.61-1.48) 0.819
Abbreviations: CCE, cerebrovascular and cardiovascular events; CI, confidence interval; HR, hazard ratio; PSM, propensity score matching
*Adjusted hazard ratio calculated using Cox proportional hazard model adjusting for gender, age, Charlson comorbidity index score, comorbidities at baseline and
adherence to statin and comedication use (antihypertensive drugs, antidiabetic drugs, aspirin, other antiplatelet and anticoagulants) during the selection and follow-up
periods.
This article is protected by copyright. All rights reserved.
 Figure 1
Accepted Article

This article is protected by copyright. All rights reserved.

 Figure 2
Accepted Article

This article is protected by copyright. All rights reserved.