Abstract
The use of lipid-modifying agents (LMAs) other than statins has rarely been reported in real clinical settings. We aimed to compare the initiation and subsequent use of LMA classes for prevention of cardiovascular diseases. Using the national claims database, this retrospective cohort study was conducted on patients aged ≥55 years who initiated to use statins, ezetimibe, or fibrates between Fiscal Years (FYs) 2014 and 2017 as the first pharmacotherapy for dyslipidemia in Japan. A permissible gap for defining persistence was set as the median days of supply of a class to an individual. Kaplan–Meier estimates were calculated for rates. Cohorts for primary prevention without/with risk and secondary prevention comprised 1307438, 908378, and 503059 initiators for statins; 44116, 34206, and 11373 for ezetimibe; and 124511, 96380, and 27751 for fibrates. The persistence rates declined shortly after the therapy initiation regardless of the classes, which was approximately 50% at 1 year for any class for primary prevention without risk. A notable sex difference in terms of persistence rates was observed only for statins of secondary prevention. The restarting rates were similar between prevention settings: approximately 50–60% for statins and 30–40% for ezetimibe and fibrates 1 year after first discontinuation. For ezetimibe and fibrates, approximately 10% of initiators were added or switched to statins within 1 year of initiation. Collectively, any class tended to be discontinued early and some restarted; however, there were some unique classes. The findings are useful for improvement of dyslipidemia therapy.
INTRODUCTION
Dyslipidemia is an important risk factor for cardiovascular diseases (CVDs).1,2) Approximately 17.9 million people died from CVDs in 2019, accounting for 32% of all global deaths,3) and since 1985, heart disease has been the second leading cause of death in Japan.4) The use of lipid-modifying agents (LMAs), which improve lipid levels and prevent cardiovascular events,5–8) continues to grow globally.9) Older people are vulnerable to CVDs and commonly use LMAs, but its use among older people aged ≥75 years for the primary prevention of CVDs remains controversial,10–14) mainly due to a paucity of evidence in this group. Thus, the real-life use of LMAs among older people by prevention settings is of interest.
Persistence (i.e., duration that a patient continues medications) and adherence (i.e., degree to which a patient takes medications as prescribed) to LMAs influence the occurrence of cardiovascular events.15–17) The actual use of LMA classes other than statins has rarely been reported.18–20) Additionally, to the best of our knowledge, no study has analyzed the use of other classes among many older people by primary and secondary prevention of CVDs.
The recently revised guidelines of Japan,21) Europe,22) and U.S.A.23) elucidate the positioning of ezetimibe and proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors in addition to statins. In addition, the 2022 revision of the guidelines of Japan21) specifies the triglyceride (TG) reference level in the nonfasting state. Comparing the treatment patterns of recent medications that lower low-density lipoprotein cholesterol (LDL-C) levels in addition to statins and medications that lower TG levels would provide useful information for healthcare providers and healthcare policymakers.
This study of people aged ≥55 years who initiated to use LMA classes of statins, ezetimibe, and fibrates as the first pharmacotherapy for dyslipidemia in Japan aimed to compare the initiation and subsequent use of the classes. Comparison was performed separately for primary prevention without/with risk and secondary prevention. Furthermore, this study focused on sex differences because our previous study24) found that prevention settings had a significant impact on the persistence and adherence of statins and that there were notable sex differences in the persistence rates of statin therapy only for the secondary prevention. Our previous study that targeted statin initiators24) included patients who initiated to use statins together with ezetimibe and/or fibrates, who initiated to use statins in addition to other LMAs, and who switched to statins from other LMAs. However, owing to such population, it was unclear whether the findings of sex differences in terms of the persistence of statins in the previous study represented the true characteristics of statin initiators. This study targeted patients who initiated to use three of each class as the first LMA (i.e., not using other LMAs before initiation) and singly without the combination of the classes so as to contrast the characteristics of each class.
MATERIALS AND METHODS
Study Design and Data SourceThis retrospective cohort study with a new user design was conducted using a claims dataset extracted from the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB Japan). The dataset included the claims data of the NDB Japan from Fiscal Year (FY) 2012 to FY2019, and its details were described elsewhere.24) This study was part of the Anti-dyslipidemic Agent and Fracture in the Japanese Older People Study. It complied with the Ethical Guidelines for Medical and Health Research Involving Human Subjects and was approved by the ethics committee of Tokushima Bunri University in February 2020 (No. R1–39).
Study Population and Definition of PreventionThe study population comprised individuals aged ≥55 years who newly initiated three classes of LMAs, namely, statins, ezetimibe, and fibrates, between FY2014 and FY2017. New initiation was defined as the absence of a dispensing record of the medications for at least 2 years before the initiation. To depict the difference in the three classes through this study, initiators who met the following criteria were analyzed: patients who initiated to use these three classes as the first LMA (i.e., not using other LMAs before initiation) and singly without the combination of the classes. Individuals were observed from the initiation to the death date or study end date (March 2020), whichever came first. Although older people are generally defined as those aged ≥65 years, this study focused on those aged ≥55 years as LMAs were thought to be often initiated at the age of <65 years; furthermore, analyses on younger individuals could help identify the characteristics of LMA use among older initiators.
Referring to the 2017 edition of the Atherosclerosis Prevention Guidelines of Japan, three cohorts, primary prevention without risk, primary prevention with risk, and secondary prevention, were set in this study. The primary prevention cohort without risk included patients who did not have codes of high-risk factors for CVDs in addition to CVD-related codes in the 2 years prior to the initiation. The primary prevention cohort with risk comprised patients who had the former codes but did not have the latter within the period. The secondary prevention cohort consisted of patients who had CVD-related codes within the period. The high-risk factors for CVDs included diabetes, cerebral infarction, chronic kidney disease, and dialysis. Details of codes for CVD as well as high-risk factors are described in our previous study.24)
MedicationsStatins (Anatomical Therapeutic Chemical classification, C10AA) included atorvastatin, fluvastatin, pitavastatin, pravastatin, rosuvastatin, and simvastatin. Fibrates included bezafibrate (C10AB), clofibrate (C10AB), fenofibrate (C10AB), and pemafibrate. Ezetimibe (C10AX09) is a single drug. Other LMA classes included anion-exchange resins (C10AC01 and V03AE06), probucol (C10AX02), omega-3 fatty acids (C10AX06), niacin (C10AD and nicomol), PCSK9 inhibitors (C10AX13, C10AX14), microsomal TG transfer protein inhibitors (C10AX12), and other LMAs (A11HA32, B05AA05, elastase ES, gamma oryzanol, soysterol, and polyenephosphatidyl choline). Pemafibrate was released on the market on June 2018 in Japan, which was outside the patient registration period. Thus, patients who initiated to use pemafibrate as the first LMA were not included among the fibrate initiators in this study; however, patients who were added or switched to pemafibrate were included in the statin or ezetimibe initiators for the analyses.
The permissible gap method25) was used to define persistence and discontinuation (Supplementary Fig. S1). For main analyses, the permissible gap for each class was defined on individual basis, i.e., the median days that the class was supplied per prescription were calculated individually and adopted as the permissible gap for that individual. Sensitivity analyses were conducted, changing the permissible gap to 90 d. Switching within each class and changing the medication dose were considered persistent.
Statistical AnalysesThe cumulative persistence rate and proportion of patients with poor adherence during the first persistence, and cumulative restarting rate after the first discontinuation were analyzed by each class cohort for each prevention setting. In addition, different LMA classes that were added or switched to by the end of the observation period after the initiation of each class were analyzed.
Poor adherence was defined as the proportion of days covered (PDC) of <0.8.25) PDC was calculated as the PDC during persistence (Supplementary Fig. S1), which is called the compliance rate.26) Restarting occurred if the same classes were dispensed again by the end of the observation period after the first discontinuation. The persistence rate, restarting rate, and different classes that were added or switched to were described using Kaplan–Meier curves, and the curves were analyzed overall and by age subgroups. The log-rank test was used to compare the Kaplan–Meier curves between sex groups, and the results were used as reference, considering that there may have been negligible or no significant differences due to the big sample size even if low p-values were shown. Analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, U.S.A.).
RESULTS
In total, 3675949, 564868, and 565011 patients initiated to use statins, ezetimibe, and fibrates at the age of ≥55 years between FY2014 and FY2017 in Japan, respectively (Fig. 1). The initiators of each class identified by each prevention setting were as follows: primary without risk vs. primary with risk vs. secondary: 1450169 vs. 1155203 vs. 648450 for statins; 170034 vs. 216023 vs. 155895 for ezetimibe; and 217769 vs. 229645 vs. 87749 for fibrates. The proportion of patients who used any of them as the first LMA was the highest for statin initiators (primary without risk vs. primary with risk vs. secondary: 90.7 vs. 78.9 vs. 78.0%) but lower for ezetimibe (29.6 vs. 17.3 vs. 9.1%) and fibrates (59.4 vs. 42.9 vs. 32.5%) initiators. Many of the initiators of ezetimibe and fibrate had statins dispensed before the initiation. Excluding those who used any other LMAs before the initiation and those who initiated to use any combination of the three classes, the analyzed cohorts consequently comprised initiators who singly used the class as the first LMA: primary without risk vs. primary with risk vs. secondary: 1307438 vs. 908378 vs. 503059 for statins; 44116 vs. 34206 vs. 11373 for ezetimibe; and 124511 vs. 96380 vs. 27751 for fibrates.
The characteristics of the analyzed cohorts are shown in Table 1. The male proportion was high for the fibrate cohorts for all prevention settings and the statin cohort for the secondary prevention. The primary prevention cohorts without risk included approximately 10–20% of patients aged ≥75 years, whereas the secondary prevention cohorts included approximately 30–50% patients. Patients with essential hypertension accounted for approximately 35–45% of the patients in the primary prevention cohorts without risk and approximately 70–80% in the secondary prevention cohorts. The median of the observation period was approximately 4 years for all cohorts, and the permissible gap was approximately 30 d for all cohorts.
Table 1. Characteristics of Initiators Who Singly Initiated to Use Statins, Ezetimibe, and Fibrates as the First LMAs
| Characteristics | Primary prevention without risk | Primary prevention with risk | Secondary prevention |
|---|
| Statin initiators | Ezetimibe initiators | Fibrate initiators | Statin initiators | Ezetimibe initiators | Fibrate initiators | Statin initiators | Ezetimibe initiators | Fibrate initiators |
|---|
| n = 1307438 | n = 44116 | n = 124511 | n = 908378 | n = 34206 | n = 96380 | n = 503059 | n = 11373 | n = 27751 |
|---|
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) |
|---|
| Male | 436582 (33.4) | 14459 (32.8) | 79547 (63.9) | 419126 (46.1) | 14942 (43.7) | 65810 (68.3) | 297100 (59.1) | 5118 (45.0) | 17763 (64.0) |
| Female | 870856 (66.6) | 29657 (67.2) | 44964 (36.1) | 489252 (53.9) | 19264 (56.3) | 30570 (31.7) | 205959 (40.9) | 6255 (55.0) | 9988 (36.0) |
| Male |
| 55–64 yearsa) | 197130 (45.2) | 6778 (46.9) | 42721 (53.7) | 129676 (30.9) | 4969 (33.3) | 27470 (41.7) | 68263 (23.0) | 1199 (23.4) | 5652 (31.8) |
| 65–74 yearsa) | 173881 (39.8) | 5463 (37.8) | 28290 (35.6) | 178674 (42.6) | 6158 (41.2) | 26670 (40.5) | 118494 (39.9) | 1995 (39.0) | 7230 (40.7) |
| 75–84 yearsa) | 57029 (13.1) | 1973 (13.6) | 7752 (9.7) | 93840 (22.4) | 3249 (21.7) | 10429 (15.8) | 87830 (29.6) | 1566 (30.6) | 4180 (23.5) |
| ≥85 yearsa) | 8542 (2.0) | 245 (1.7) | 784 (1.0) | 16936 (4.0) | 566 (3.8) | 1241 (1.9) | 22513 (7.6) | 358 (7.0) | 701 (3.9) |
| Female |
| 55–64 yearsb) | 367152 (42.2) | 12321 (41.5) | 18353 (40.8) | 139963 (28.6) | 5323 (27.6) | 9080 (29.7) | 35977 (17.5) | 1151 (18.4) | 1926 (19.3) |
| 65–74 yearsb) | 349794 (40.2) | 11774 (39.7) | 17634 (39.2) | 198115 (40.5) | 7787 (40.4) | 12006 (39.3) | 71276 (34.6) | 2236 (35.7) | 3459 (34.6) |
| 75–84 yearsb) | 126213 (14.5) | 4638 (15.6) | 7407 (16.5) | 117102 (23.9) | 4817 (25.0) | 7532 (24.6) | 67512 (32.8) | 2080 (33.3) | 3298 (33.0) |
| ≥85 yearsb) | 27697 (3.2) | 924 (3.1) | 1570 (3.5) | 34072 (7.0) | 1337 (6.9) | 1952 (6.4) | 31194 (15.1) | 788 (12.6) | 1305 (13.1) |
| Comorbiditiesc): ICD-10 code |
| Malignant neoplasm: C00–97 | 78981 (6.0) | 3047 (6.9) | 8150 (6.5) | 92841 (10.2) | 3678 (10.8) | 9972 (10.3) | 64154 (12.8) | 1600 (14.1) | 3663 (13.2) |
| Diabetes mellitus: E10–14 | NA | NA | NA | 549132 (60.5) | 21363 (62.5) | 66681 (69.2) | 211178 (42.0) | 5283 (46.5) | 12800 (46.1) |
| Essential hypertension: I10 | 440781 (33.7) | 16267 (36.9) | 57394 (46.1) | 558457 (61.5) | 21201 (62.0) | 64987 (67.4) | 358254 (71.2) | 8808 (77.4) | 22603 (81.4) |
| Angina: I20 | NA | NA | NA | NA | NA | NA | 325940 (64.8) | 8646 (76.0) | 20722 (74.7) |
| Myocardial infarction: I21–23 | NA | NA | NA | NA | NA | NA | 28646 (5.7) | 460 (4.0) | 1103 (4.0) |
| Ischemic heart diseaseb): I24–25 | NA | NA | NA | NA | NA | NA | 77334 (15.4) | 1908 (16.8) | 4512 (16.3) |
| Heart failure: I50 | 40375 (3.1) | 1434 (3.3) | 3850 (3.1) | 99340 (10.9) | 3791 (11.1) | 8953 (9.3) | 149641 (29.7) | 3549 (31.2) | 7540 (27.2) |
| Cerebrovascular disease: I60–69 | NA | NA | NA | 270713 (29.8) | 9292 (27.2) | 21535 (23.3) | 121783 (24.2) | 2964 (26.1) | 6186 (22.3) |
| Observation period, yearsd) | 4.2 (3.1–5.1) | 4.2 (3.1–5.2) | 4.2 (3.1–5.1) | 4.0 (3.0–5.0) | 4.1 (3.0–5.1) | 4.1 (3.0–5.1) | 3.9 (2.8–5.0) | 4.1 (3.0–5.1) | 4.1 (3.0–5.1) |
| Permissible gap, dd) | 30 (26–40) | 30 (28–42) | 30 (28–35) | 30 (25–42) | 30 (28–49) | 30 (28–42) | 29 (20–40) | 30 (28–45) | 30 (26–40) |
LMA, lipid-modifying agent; NA, not applicable. a) Percentage depicts the proportion of males. b) Percentage depicts the proportion of females. c) Comorbidities during the year before the initiation of statins, ezetimibe, and fibrates. d) Medians (25–75th percentiles).
Figure 2 shows the cumulative persistence rates of the first pharmacotherapy for dyslipidemia. The cumulative persistence rates shortly declined after the initiation irrespective of the classes and tended to be the lowest for the primary prevention without risk for all classes, followed by primary prevention with risk and secondary prevention for all classes. The rates were only nearly 50% at 1 year for the primary prevention without risk for any class and nearly 70% for statins and 60% for ezetimibe and fibrates at 1 year for secondary prevention. There was no or almost no sex difference in the cumulative persistence rate for any class in the primary prevention cohorts without/with risk. However, a notable sex difference was observed only for statins in the secondary prevention cohort, with a lower persistence rate in females. On the contrary, almost no sex difference was observed for ezetimibe and fibrates in the secondary prevention cohorts. With older age strata, the cumulative persistence rates tended to be higher for ezetimibe in any prevention setting but almost the same for statins and fibrates in the primary prevention with risk and in secondary prevention (Supplementary Fig. S2). The notable sex difference only for statins in the secondary prevention setting was observed in all age strata.
The poor adherence proportion during the first persistence was the highest in the primary prevention without risk, followed by primary prevention with risk and secondary prevention, regardless of the classes: approximately 10 and 5% for the primary prevention without risk and the secondary prevention, respectively, if all patients were analyzed (Table 2). Poor adherence gradually improved in the older age strata of almost all cohorts. The sex difference in the poor adherence proportion was likely to be observed in the statin initiators for the secondary prevention, with a higher poor adherence proportion in female patients.
Table 2. Poor Adherence during the First Persistence
a)| Sex, age | Statin initiatorsb) | Ezetimibe initiatorsb) | Fibrate initiatorsb) |
|---|
| Poor adherencec) | | Poor adherencec) | | Poor adherencec) |
|---|
| n | n (%) | n | n (%) | n | n (%) |
|---|
| Primary prevention without risk |
| All patients | 1114702 | 125001 (11.2) | 36275 | 3776 (10.4) | 103387 | 11892 (11.5) |
| Male |
| 55–64 years | 165331 | 20810 (12.6) | 5423 | 649 (12.0) | 35121 | 4967 (14.1) |
| 65–74 years | 149216 | 14209 (9.5) | 4521 | 422 (9.3) | 23840 | 2547 (10.7) |
| 75–84 years | 49081 | 3616 (7.4) | 1641 | 141 (7.7)d) | 6530 | 576 (8.8) |
| ≥85 years | 7316 | 449 (6.1) | 196 | 643 | 45 (7.0) |
| Female |
| 55–64 years | 310909 | 42996 (13.8) | 10035 | 1230 (12.3) | 14926 | 1791 (12.0) |
| 65–74 years | 299892 | 32702 (10.9) | 9737 | 971 (10.0) | 14736 | 1382 (9.4) |
| 75–84 years | 108693 | 8871 (8.2) | 3903 | 323 (8.3) | 6247 | 509 (8.1) |
| ≥85 years | 24264 | 1348 (5.6) | 819 | 40 (4.9) | 1344 | 75 (5.6) |
| Primary prevention with risk |
| All patients | 815995 | 56710 (6.9) | 29792 | 2045 (6.9) | 84579 | 6814 (8.1) |
| Male |
| 55–64 years | 117091 | 9675 (8.3) | 4325 | 348 (8.0) | 24217 | 2438 (10.1) |
| 65–74 years | 162139 | 9626 (5.9) | 5397 | 332 (6.2) | 23610 | 1728 (7.3) |
| 75–84 years | 85400 | 3879 (4.5) | 2887 | 149 (5.2) | 9171 | 583 (6.4) |
| ≥85 years | 15405 | 564 (3.7) | 487 | 23 (4.7) | 1066 | 59 (5.5) |
| Female |
| 55–64 years | 124451 | 11720 (9.4) | 4617 | 426 (9.2) | 7876 | 721 (9.2) |
| 65–74 years | 176183 | 13721 (7.8) | 6711 | 473 (7.0) | 10409 | 798 (7.7) |
| 75–84 years | 104595 | 6299 (6.0) | 4160 | 219 (5.3) | 6517 | 408 (6.3) |
| ≥85 years | 30731 | 1226 (4.0) | 1208 | 75 (6.2) | 1713 | 79 (4.6) |
| Secondary prevention |
| All patients | 465839 | 17953 (3.9) | 9918 | 530 (5.3) | 24191 | 1626 (6.7) |
| Male |
| 55–64 years | 64215 | 2582 (4.0) | 1058 | 67 (6.3) | 4967 | 460 (9.3) |
| 65–74 years | 111639 | 3330 (3.0) | 1748 | 86 (4.9) | 6360 | 415 (6.5) |
| 75–84 years | 82467 | 1899 (2.3) | 1423 | 57 (4.0) | 3675 | 175 (4.8) |
| ≥85 years | 20985 | 398 (1.9) | 305 | 11 (3.6) | 605 | 28 (4.6) |
| Female |
| 55–64 years | 32062 | 2633 (8.2) | 992 | 80 (8.1) | 1662 | 154 (9.3) |
| 65–74 years | 64322 | 3791 (5.9) | 1891 | 112 (5.9) | 2970 | 186 (6.3) |
| 75–84 years | 61520 | 2533 (4.1) | 1790 | 92 (5.1) | 2812 | 161 (5.7) |
| ≥85 years | 28629 | 767 (2.7) | 711 | 25 (3.5) | 1140 | 47 (4.1) |
PDC, proportion of days covered. a) The permissible gap was defined as the median days that each LMA class was supplied per prescription to an individual. b) Initiators who had dispensing records of different dispensing dates during the persistent period. c) Of the initiators who had dispensing records of different dispensing dates during persistence, the number of initiators whose PDC was <0.8. d) Values are combined because the number was <10, according to the guidelines of NDB Japan.
The cumulative restarting rates after the first discontinuation were almost comparable regardless of the prevention settings and comparatively high for all classes, particularly for statins (Fig. 3). The cumulative restarting rates 1 year after the first discontinuation were approximately 50–60% for statins and 30–40% for ezetimibe and fibrates. The cumulative restarting rates 4 years after the discontinuation were approximately 70% for statins and 40–50% for ezetimibe and fibrates. Sex differences were nearly not observed in all cohorts. With older age strata, the cumulative restarting rates tended to be lower for statins and fibrates in any prevention setting (Supplementary Fig. S3). There was almost no sex difference regardless of the age subgroups for any class of any prevention setting.
Approximately 40% of ezetimibe initiators and approximately 30% of fibrate initiators were added or switched to different LMA classes by the end of the observation period after the initiation for any prevention setting (Table 3). Most of them were added or switched to statins (approximately 30% in ezetimibe initiators and approximately 20% in fibrate initiators for any prevention setting). The cumulative rate of the addition of or switching to statins was approximately 10% 1 year after the initiation of ezetimibe and fibrates for any prevention setting (Fig. 4). Meanwhile, for statin initiators, only approximately 15% was added or switched to different LMA classes by the end of the observation period for any prevention setting (Table 3). The cumulative rate of the addition of or switching to ezetimibe was 2–3% at 1 year for any prevention setting (Fig. 4). With older age strata, the added or switched rates to statin tended to be lower in ezetimibe and fibrate initiators for any prevention setting (Supplementary Fig. S4).
Table 3. LMA Class That Was Added or Switched to after Therapy Initiation
| LMA class | Primary prevention without risk | Primary prevention with risk | Secondary prevention |
|---|
| Statin initiators | Ezetimibe initiators | Fibrate initiators | Statin initiators | Ezetimibe initiators | Fibrate initiators | Statin initiators | Ezetimibe initiators | Fibrate initiators |
|---|
| n = 1307438 | n = 44116 | n = 124511 | n = 908378 | n = 34206 | n = 96380 | n = 503059 | n = 11373 | n = 27751 |
|---|
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) |
|---|
| Any LMAs | 162206 (12.4) | 17673 (40.1) | 37018 (29.7) | 120921 (13.3) | 12960 (37.9) | 28813 (29.9) | 79841 (15.9) | 4255 (37.4) | 8601 (31.0) |
| Statinsa) | NA | 13207 (29.9) | 24361 (19.6) | NA | 9068 (26.5) | 18494 (19.2) | NA | 2988 (26.3) | 5534 (19.9) |
| Rosuvastatinb) | NA | 6083 (13.8) | 10853 (8.7) | NA | 4053 (11.8) | 8380 (8.7) | NA | 1361 (12.0) | 2640 (9.5) |
| Atorvastatinb) | NA | 4084 (9.3) | 7752 (6.2) | NA | 2826 (8.3) | 5580 (5.8) | NA | 961 (8.4) | 1655 (6.0) |
| Pitavastatinb) | NA | 2952 (6.7) | 5224 (4.2) | NA | 2170 (6.3) | 4385 (4.5) | NA | 727 (6.4) | 1278 (4.6) |
| Fluvastatinb) | NA | 221 (0.5) | 309 (0.2) | NA | 177 (0.5) | 204 (0.2) | NA | 42 (0.4) | 63 (0.2) |
| Simvastatinb) | NA | 342 (0.8) | 693 (0.6) | NA | 198 (0.6) | 437 (0.5) | NA | 49 (0.4) | 133 (0.5) |
| Pravastatinb) | NA | 1922 (4.4) | 3581 (2.9) | NA | 1293 (3.8) | 2448 (2.5) | NA | 455 (4.0) | 681 (2.5) |
| Fibratesa) | 32398 (2.5) | 2830 (6.4) | NA | 19765 (2.2) | 2234 (6.5) | NA | 7686 (1.5) | 642 (5.6) | NA |
| Ezetimibea) | 59850 (4.6) | NA | 6254 (5.0) | 39060 (4.3) | NA | 5034 (5.2) | 32180 (6.4) | NA | 1371 (4.9) |
| Anion-exchange resinsa) | 1583 (0.1) | 160 (0.4) | 204 (0.2) | 966 (0.1) | 129 (0.4) | 114 (0.1) | 356 (0.1) | 49 (0.4) | 34 (0.1) |
| Probucola) | 1826 (0.1) | 116 (0.3) | 241 (0.2) | 1017 (0.1) | 92 (0.3) | 132 (0.1) | 410 (0.1) | 31 (0.3) | 45 (0.2) |
| Omega-3 fatty acidsa) | 51062 (3.9) | 3103 (7.0) | 10277 (8.3) | 43861 (4.8) | 2641 (7.7) | 7909 (8.2) | 29205 (5.8) | 915 (8.0) | 2344 (8.4) |
| Niacina) | 24447 (1.9) | 1118 (2.5) | 2874 (2.3) | 19240 (2.1) | 977 (2.9) | 2431 (2.5) | 10027 (2.0) | 311 (2.7) | 790 (2.8) |
| PCSK9 inhibitorsa) | 346 (0.0) | 13 (0.0) | 11 (0.0) | 361 (0.0) | 13 (0.0) | 19 (0.0) | 686 (0.1) | 15 (0.1) | 10 (0.0) |
| Othersa) | 16337 (1.2) | 653 (1.5) | 2073 (1.7) | 14364 (1.6) | 565 (1.7) | 1720 (1.8) | 10438 (2.1) | 212 (1.9) | 595 (2.1) |
LMA, lipid-modifying agent; NA, not applicable; PCSK9, proprotein convertase subtilisin/kexin type 9. a) The total does not meet the number of patients taking any LMAs because of duplicate counts. b) The total does not meet the number of patients taking statins because of duplicate counts.
Sensitivity analyses with a permissible gap of 90 d revealed similar trends to the results of the main analyses, such as the declined persistence rate soon after initiation regardless of the classes and the sex difference that was notable only for statins of the secondary prevention (Supplementary Fig. S5), poorest adherence for the primary prevention without risk for all classes and improvement in the older age strata of almost all cohorts (Supplementary Table S1), and almost comparable restarting rates regardless of the prevention settings, higher restarting rate for statins, and almost no sex differences in all cohorts (Supplementary Fig. S6). However, the rates and proportions differed from those in the main analyses, which revealed higher persistence rate, higher proportion of poor adherence, and lower restarting rate.
DISCUSSION
This study presents novel findings on the use of statins, ezetimibe, and fibrates in real clinical settings by comparing these classes among patients aged ≥55 years who initiated to use them as the first pharmacotherapy for dyslipidemia in Japan. It is valuable due to the limited reports on LMAs other than statins worldwide and also the lack of decent reports among older people in Japan. Previous reports on the use of LMAs in Japan19,27–29) included a small number of older patients and insufficiently tracked the patients’ medication record because databases containing the medication history of older people did not exist until the establishment of the NDB Japan. The findings of the present study using claims data of the NDB Japan were brought from almost completely tracked medication record of nearly all initiators of the relevant age.
First, regardless of the class in which the therapy was initiated, persistence rates declined shortly after the initiation and were lower for primary prevention than for secondary prevention. The persistence rates for the main analyses were only approximately 50% at 1 year for the primary prevention without risk for any class and approximately 70% for statins and approximately 60% for ezetimibe and fibrates at 1 year for the secondary prevention. Conversely, the restarting rates after the discontinuation were comparatively high for all classes with almost comparable rates between prevention settings, and among those who discontinued each class, approximately 50–60% restarted statins and 30–40% restarted ezetimibe and fibrates within 1 year after discontinuation for main analyses.
Early discontinuation of statin therapy has often been reported for both the primary and secondary prevention in Japan24) and overseas,26,30,31) although the authors could not find reports on ezetimibe and fibrates by prevention. May et al.15) reported interesting findings regarding statin therapy among patients with atherosclerotic CVDs. They found that the group that received adequate statin therapy for 5 years had a lower rate of major cardiovascular events than the group that received inadequate therapy in the first year but adequate therapy in the second through fifth years. Healthcare providers should pay attention to the reasons for LMA discontinuation and ensure that the therapy continues.
Second, except for statins for the secondary prevention, sex difference in terms of persistence rates was negligible for the three classes in any prevention setting. The findings were consistent for any age stratum. A similar trend of sex difference was likely to be observed for poor adherence. Sex differences in persistence and adherence regarding statin therapy have been previously reported.24,30–33) Interestingly, previous studies of statin therapy presented different results in Japan than from those obtained in other countries. Overseas reports of statin therapy indicated sex differences for both the primary30,31) and secondary30–33) prevention settings, with female patients having lower persistence or adherence. However, in the previous report from Japan,24) no sex difference was observed in the primary prevention, but a notable difference was observed for the secondary prevention. Confirming this previous finding,24) the present study analyzed “pure” initiators by excluding patients who used other LMAs before statin initiation and who initiated statins in combination with ezetimibe or fibrates. The results of the present study were the same as those of the previous studies in Japan.24) Additionally, the present study found that no sex differences were observed in the restarting rates after discontinuing statins for any prevention setting, even by stratified age. The reasons for the sex differences regarding statin use are unclear; particularly, why the findings of sex differences were different between prevention settings among Japanese patients and why the findings for the primary prevention was different between reports of Japan and overseas. Age would probably have a negligible effect on sex differences from the results of the age-stratified analyses in the present study. It might be influenced by (1) female predominance, occurrence and exacerbation of adverse events of statins,34,35) (2) differences in medication responsiveness by Japanese and Western people (lower dose of statins in Japan), and (3) differences in the onset time of CVD by sex (later onset in female patients).36) Considering the lack of a sex difference for ezetimibe and fibrates, (1) and (2) may be more contributing factors. Further studies are warranted to determine the reasons for the sex differences in terms of statin use.
Third, this study showed that statins had been used as the mainstay LMA in clinical practice in Japanese patients aged ≥55 years, backed by the following findings of statins: high initiation proportion as the first LMA, high restarting rates, and high addition or switching rates from ezetimibe and fibrates. On the contrary, this study found that a low percentage of statin initiators were added or switched to ezetimibe within 1 year after the initiation. Some of them would intensify LDL-C-lowering therapy by combining statins and ezetimibe,6) and others would be statin-intolerant.37)
Fourth, the results of fibrates were likely to be similar to those of ezetimibe in terms of persistence rates, poor adherence proportion, and restarting rates, but change in the persistence and restarting rates according to older age strata slightly differed between classes. Additionally, fibrates were used more often in male patients, whereas ezetimibe was used more often in female patients. Fibrates lower TG levels, which are considered important as a residual risk factor of CVD, especially for the secondary prevention cohort.38) In October 2018, which was midway through the observation period in this study, restrictions on contraindications between statins and fibrates were removed in Japan, enabling physicians to prescribe the combination of both. The removal may have influenced the results of this study regarding the addition or switching between statins and fibrates.
This study had some limitations. First, the reasons for discontinuing medications were unclear as insurance claims data were used in the analyses. The possible reasons were the occurrence of adverse events, patient intention, and switch to other LMA classes. Furthermore, owing it to be the claims data, whether the patients were actually taking the medications as per the claims data was unclear. Second, studies to validate the codes used to define the prevention cohorts are lacking, leading to potential effects on the accuracy of findings in the present study. It was difficult to identify disease codes related to the prevention cohorts from the International Statistical Classification of Diseases and Related Health Problems 10th Revision. Additionally, the disease codes of claims data are less accurate. Especially, the definition of high-risk factors for CVDs was not sufficient, because risk factors such as smoking states and family medical history are not available in claims data.
Despite these limitations, the strength of this study was related to its findings derived from the analysis of a national Japanese database, which covers nearly all citizens. Any class tended to be discontinued early and restarted; however, there were some unique classes. The findings are useful for the success of pharmacotherapy for dyslipidemia and promoting health policymaking for people aged ≥55 years in Japan.
Acknowledgments
We sincerely thank the Ministry of Health, Labour and Welfare for generating and making the dataset from the NDB Japan available for our study. This work was supported by JSPS KAKENHI Grant Number: JP19K07237.
Conflict of Interest
The authors declare no conflict of interest.
Supplementary Materials
This article contains supplementary materials.
REFERENCES
- 1) Michos ED, McEvoy JW, Blumenthal RS. Lipid management for the prevention of atherosclerotic cardiovascular disease. N. Engl. J. Med., 381, 1557–1567 (2019).
- 2) Nordestgaard BG, Benn M, Schnohr P, Tybjaerg-Hansen A. Nonfasting triglycerides and risk of myocardial infarction, ischemic heart disease, and death in men and women. JAMA, 298, 299–308 (2007).
- 3) WHO. “Cardiovascular diseases.”: ‹http://www.who.int/mediacentre/factsheets/fs317/en/›, accessed 12 April, 2023.
- 4) The Ministry of Health, Labour and Welfare. “Summary of the 2021 vital statistics monthly report, summary.”: ‹https://www.mhlw.go.jp/toukei/saikin/hw/jinkou/geppo/nengai21/index.html›, accessed 12 April, 2023.
- 5) Gencer B, Marston NA, Im K, Cannon CP, Sever P, Keech A, Braunwald E, Giugliano RP, Sabatine MS. Efficacy and safety of lowering LDL cholesterol in older patients: a systematic review and meta-analysis of randomised controlled trials. Lancet, 396, 1637–1643 (2020).
- 6) Wang N, Fulcher J, Abeysuriya N, Park L, Kumar S, Di Tanna GL, Wilcox I, Keech A, Rodgers A, Lal S. Intensive LDL cholesterol-lowering treatment beyond current recommendations for the prevention of major vascular events: a systematic review and meta-analysis of randomised trials including 327 037 participants. Lancet Diabetes Endocrinol., 8, 36–49 (2020).
- 7) Sando KR, Knight M. Nonstatin therapies for management of dyslipidemia: a review. Clin. Ther., 37, 2153–2179 (2015).
- 8) Jun M, Foote C, Lv J, Neal B, Patel A, Nicholls SJ, Grobbee DE, Cass A, Chalmers J, Perkovic V. Effects of fibrates on cardiovascular outcomes: a systematic review and meta-analysis. Lancet, 375, 1875–1884 (2010).
- 9) Blais JE, Wei Y, Yap KKW, Alwafi H, Ma TT, Brauer R, Lau WCY, Man KKC, Siu CW, Tan KCB, Wong ICK, Wei L, Chan EW. Trends in lipid-modifying agent use in 83 countries. Atherosclerosis, 328, 44–51 (2021).
- 10) Kim S, Choi H, Won CW. Moderate-intensity statin use for primary prevention for more than 5 years is associated with decreased all-cause mortality in 75 years and older. Arch. Gerontol. Geriatr., 100, 104644 (2022).
- 11) Awad K, Mohammed M, Zaki MM, Abushouk AI, Lip GYH, Blaha MJ, Lavie CJ, Toth PP, Jukema JW, Sattar N, Banach M. Association of statin use in older people primary prevention group with risk of cardiovascular events and mortality: a systematic review and meta-analysis of observational studies. BMC Med., 19, 139 (2021).
- 12) Ouchi Y, Sasaki J, Arai H, et al. Ezetimibe lipid-lowering trial on prevention of atherosclerotic cardiovascular disease in 75 or older (EWTOPIA 75): a randomized, controlled trial. Circulation, 140, 992–1003 (2019).
- 13) Armitage J, Baigent C, Barnes E, et al. Efficacy and safety of statin therapy in older people: a meta-analysis of individual participant data from 28 randomised controlled trials. Lancet, 393, 407–415 (2019).
- 14) Han BH, Sutin D, Williamson JD, Davis BR, Piller LB, Pervin H, Pressel SL, Blaum CS. Effect of statin treatment vs. usual care on primary cardiovascular prevention among older adults: The ALLHAT-LLT randomized clinical trial. JAMA Intern. Med., 177, 955–965 (2017).
- 15) May HT, Knowlton KU, Anderson JL, Lappé DL, Bair TL, Muhlestein JB. High-statin adherence over 5 years of follow-up is associated with improved cardiovascular outcomes in patients with atherosclerotic cardiovascular disease: results from the IMPRES study. Eur. Heart J. Qual. Care Clin. Outcomes, 8, 352–360 (2022).
- 16) Svensson MK, Sorio Vilela F, Leósdóttir M, Banefelt J, Lindh M, Dun AR, Norhammar A, Villa G. Effects of lipid-lowering treatment intensity and adherence on cardiovascular outcomes in patients with a recent myocardial infarction: a Swedish register-based study. Ups. J. Med. Sci., 127, e8296 (2022).
- 17) Guglielmi V, Bellia A, Pecchioli S, Della-Morte D, Parretti D, Cricelli I, Medea G, Sbraccia P, Lauro D, Cricelli C, Lapi F. Effectiveness of adherence to lipid lowering therapy on LDL-cholesterol in patients with very high cardiovascular risk: a real-world evidence study in primary care. Atherosclerosis, 263, 36–41 (2017).
- 18) Santoleri F, Romagnoli A, Costantini A. Adherence and persistence in the use of statins and ezetimibe over 8 years in a real-life study. Curr. Med. Res. Opin., 37, 2061–2066 (2021).
- 19) Wake M, Onishi Y, Guelfucci F, Oh A, Hiroi S, Shimasaki Y, Teramoto T. Treatment patterns in hyperlipidaemia patients based on administrative claim databases in Japan. Atherosclerosis, 272, 145–152 (2018).
- 20) Yang CC, Jick SS, Testa MA. Discontinuation and switching of therapy after initiation of lipid-lowering drugs: the effects of comorbidities and patient characteristics. Br. J. Clin. Pharmacol., 56, 84–91 (2003).
- 21) Japan Atherosclerosis Society. Japan Atherosclerosis Society guidelines for prevention of atherosclerotic cardiovascular diseases 2022. Japan Atherosclerosis Society, Tokyo (2023).
- 22) Mach F, Baigent C, Catapano AL, et al. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur. Heart J., 41, 111–188 (2020).
- 23) Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation, 139, e1082–e1143 (2019).
- 24) Tomida J, Yoshida T, Senda S, Sato T, Nakatsuma A, Iihara N. Statin persistence and adherence among older initiators: a nationwide cohort study using the national health insurance claims database in Japan. Pharmacoepidemiol. Drug Saf., 32, 873–885 (2023).
- 25) Ofori-Asenso R, Jakhu A, Zomer E, Curtis AJ, Korhonen MJ, Nelson M, Gambhir M, Tonkin A, Liew D, Zoungas S. Adherence and persistence among statin users aged 65 years and over: a systematic review and meta-analysis. J. Gerontol. A Biol. Sci. Med. Sci., 73, 813–819 (2018).
- 26) Rezende Macedo do Nascimento RC, Mueller T, Godman B, MacBride Stewart S, Hurding S, de Assis Acurcio F, Guerra Junior AA, Alvares Teodoro J, Morton A, Bennie M, Kurdi A. Real-world evaluation of the impact of statin intensity on adherence and persistence to therapy: a Scottish population-based study. Br. J. Clin. Pharmacol., 86, 2349–2361 (2020).
- 27) Kajinami K, Ozaki A, Tajima Y, Yamashita S, Arai H, Teramoto T. Real-world data to identify hypercholesterolemia patients on suboptimal statin therapy. J. Atheroscler. Thromb., 26, 408–431 (2019).
- 28) Wake M, Oh A, Onishi Y, Guelfucci F, Shimasaki Y, Teramoto T. Adherence and persistence to hyperlipidemia medications in patients with atherosclerotic cardiovascular disease and those with diabetes mellitus based on administrative claims data in Japan. Atherosclerosis, 282, 19–28 (2019).
- 29) Nagar SP, Rane PP, Fox KM, Meyers J, Davis K, Beaubrun A, Inomata H, Qian Y, Kajinami K. Treatment patterns, statin intolerance, and subsequent cardiovascular events among Japanese patients with high cardiovascular risk initiating statin therapy. Circ. J., 82, 1008–1016 (2018).
- 30) Vinogradova Y, Coupland C, Brindle P, Hippisley-Cox J. Discontinuation and restarting in patients on statin treatment: prospective open cohort study using a primary care database. BMJ, 353, i3305 (2016).
- 31) Chodick G, Shalev V, Gerber Y, Heymann AD, Silber H, Simah V, Kokia E. Long-term persistence with statin treatment in a not-for-profit health maintenance organization: a population-based retrospective cohort study in Israel. Clin. Ther., 30, 2167–2179 (2008).
- 32) Chen ST, Huang ST, Shau WY, Lai CL, Li JZ, Fung S, Tse VC, Lai MS. Long-term statin adherence in patients after hospital discharge for new onset of atherosclerotic cardiovascular disease: a population-based study of real world prescriptions in Taiwan. BMC Cardiovasc. Disord., 19, 62 (2019).
- 33) Daniel H, Christian W, Robin H, Lars S, Thomas M. Statin treatment after acute coronary syndrome: adherence and reasons for non-adherence in a randomized controlled intervention trial. Sci. Rep., 9, 12079 (2019).
- 34) Karalis DG, Wild RA, Maki KC, Gaskins R, Jacobson TA, Sponseller CA, Cohen JD. Gender differences in side effects and attitudes regarding statin use in the Understanding Statin Use in America and Gaps in Patient Education (USAGE) study. J. Clin. Lipidol., 10, 833–841 (2016).
- 35) Goldstein KM, Zullig LL, Bastian LA, Bosworth HB. Statin adherence: does gender matter? Curr. Atheroscler. Rep., 18, 63 (2016).
- 36) Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular disease: the Framingham study. Ann. Intern. Med., 85, 447–452 (1976).
- 37) Kajinami K, Tsukamoto K, Koba S, Inoue I, Yamakawa M, Suzuki S, Hamano T, Saito H, Saito Y, Masuda S, Nakayama T, Okamura T, Yamashita S, Kagawa T, Kaneyama J, Kuriyama A, Tanaka R, Hirata A. Statin intolerance clinical guide 2018. J. Atheroscler. Thromb., 27, 375–396 (2020).
- 38) Vallejo-Vaz AJ, Corral P, Schreier L, Ray KK. Triglycerides and residual risk. Curr. Opin. Endocrinol. Diabetes Obes., 27, 95–103 (2020).
