論文ID: CJ-20-0966
Background: Non-compliance with angiotensin receptor blockers (ARB) or statin is one of the major hurdles to optimal medical treatment. This study investigated whether fixed-dose combination (FDC) improved compliance to medication compared with traditional free combination (FC).
Methods and Results: In this retrospective nationwide cohort study, medication persistency, medication adherence measured by proportion of days covered (PDC), and all-cause death of 123,992 patients who started ARB and stain were investigated for 540 days. Patients had a mean age of 63 years and 48% were male. Persistency, PDC, and proportion of PDC ≥80% of FDC (N=34,776) were higher than those for FC (N=89,216) in both unadjusted analysis (54.5% vs. 27.8%; 84.1% vs. 63.1%; 75.5% vs. 48.1%) and propensity-score matched analysis (P<0.001, all). Death risk for the investigation period (0–540 days) was lower in FDC in unadjusted (1.8% vs. 2.6%, P<0.001) and adjusted cohort (P<0.05). In landmark analyses at days 180 and 360, there was no significant difference of death risk between FDC and FC (P>0.05).
Conclusions: In this real-world data analysis, patients taking FDC of ARB and statin showed higher medication persistence and adherence compared to patients taking FC of ARB and statin up to 540 days. The risk of all-cause death was not different between FDC and FC despite better medication compliance in the FDC patients.
Hypertension and hypercholesterolemia are major modifiable cardiovascular risk factors that frequently co-exist.1 Despite advancements in medicine, opportunities for improving concomitant hypertension and hypercholesterolemia control persist. In a US survey, 69% of hypertensive hypercholesterolemic patients failed to be controlled.2
Compliance with medication is a cornerstone of the successful management of hypertension and hyperlipidemia. Unfortunately, such compliance is often difficult to achieve and maintain. Only 50–60% of patients with hypertension or hyperlipidemia could adhere to their prescribed medication,3,4 and patients non-compliant to antihypertensive or statin treatment were at a greater risk of morbidity and mortality.5
The use of a fixed-dose combination (FDC), which integrates 2 drugs or more into 1 pill, has a much simpler dose schedule and less pill burden compared to the traditional free combination (FC). FDC is potentially more convenient for patients requiring treatment with multiple drugs.6,7 FDC is recommended for hypertension by guidelines from the American College of Cardiology and American Heart Association (2019), as well as the European Society of Cardiology and European Society of Hypertension (2018).8,9
FDC may be also advantageous for patients with both hypertension and hyperlipidemia, given that these are 2 frequently coexisting diseases with unsatisfactory medication compliance. However, the clinical evidence for the advantage of FDC over FC of antihypertensive and statin medications is limited. We analyzed a large population database and investigated whether FDC of antihypertensive and statin improved compliance compared to FC, and whether it resulted in the better clinical outcome in the FDC patient group.
The study design was a retrospective longitudinal nationwide cohort analysis using a Korean national healthcare administrative database. The target population was the entire Korean population who were beneficiaries of the National Healthcare Insurance Service, a single-payer compulsory healthcare insurance of Korea. During the study period, FDC of angiotensin receptor blocker (ARB) with either atorvastatin or rosuvastatin with the following combinations, rosuvastatin+olmesartan, rosuvastatin+valsartan, rosuvastatin+telmisartan, atorvastatin+irbesartan, were available in the Korean healthcare market. Data were retrieved on 18 November 2018.
The Samsung Medical Center Institutional Review Board approved this study and waived informed consent given that the study reported aggregated results derived from de-identified data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.10 The study was conducted in accordance with the Declaration of Helsinki.
Patient SelectionPatients who were prescribed ARB and 1 of 2 statins (atorvastatin or rosuvastatin) during the 12-month selection period from 1 July 2014 to 30 June 2015 were screened using the de-identified pharmacy claims. Patients on ARB or statin monotherapy without combination, those with missing information, age <20 years, were taking statins other than rosuvastatin or atorvastatin, and who had a medication combination duration of <30 days were excluded. Patients with fewer than 2 claims in prescriptions within 360 days were considered inadequate for the assessment of medication compliance and were excluded. The remaining patients were classified into FDC or FC cohorts according to the medications. Medication patterns using ARB and statin were classified into naïve (no prior use of antihypertensive, lipid-lowering, or cardiovascular medications), add-on (addition of ARB to the prior statin or addition of statin to the prior ARB) and switched (medication was switched from the other regimen).
Study Flow and DefinitionsThe overall study flow is shown in Figure 1. The index date was the first prescription day of both ARB and statin during the selection period from 1 July 2014 to 30 June 2015. The 10th revision of International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes and administrative, medical services, and pharmacy claims issued in the preceding 180-day look-back period were used to investigate baseline clinical characteristics and medical history (Supplementary File). Death was verified by the National Healthcare Insurance Service database linked to the Statistics of Korea. Medication compliance and all-cause death up to 540 days were assessed until 31 December 2016. Follow up was determined to be completed if mortality was confirmed or if more than 2 medication claims were issued within 360 days. No patient was lost to follow up with respect to death.
Study follow-up scheme. FDC, fixed-dose combination; FC, free combination; PDC, proportion of days covered.
Medication compliance was quantitatively measured according to the recommendation of the Medication Compliance and Persistence Special Interest Group from the International Society for Pharmacoeconomics and Outcomes Research (ISPOR).11 Medication persistence and adherence was defined by continuing medication for the prescribed duration and the extent of conformity to the daily treatment, respectively.11,12 Discontinuation, which is also known as non-persistence, was defined as the number of days from the index day to the first refill gap day exceeding 30 days. Patients who remained on medications were reported every 180 days. The proportion of days covered (PDC) was defined by the proportion of days with medication supplies in the pre-defined period, which was set to be 180 days. Good medication adherence was defined by PDC ≥80% and was reported every 180 days.
Outcomes of InterestThe primary outcome of interest was medication persistence assessed by days on medication until the first prescription gap of >30 days, if it existed. The cumulative risk of discontinuation was plotted on Kaplan-Meier curves. Secondary outcomes of interest included medication adherence assessed by the PDC, the proportion of PDC ≥80% which represents good medication adherence dichotomously, and all-cause death.13
Statistical AnalysisConfounding is a major concern in non-experimental studies and can lead to biased estimates of effects of treatment. Therefore, a 1 : 3 propensity score for FDC was generated based on the following potential confounders: demographics including age and male gender; clinical characteristics including hypertension, dyslipidemia, diabetes complicated, diabetes non-complicated, chronic kidney disease, cardiovascular disease, myocardial infarction, heart failure, chronic respiratory disease, liver disease, rheumatologic disease, malignancy, stroke with hospitalization record, dementia, paralysis, human immunodeficiency virus infection, and the Charlson comorbidity index, which categorizes comorbidities of patients based on the ICD codes in administrative data;14 pattern of medication including naïve, add-on, or switched; prior medications including antiplatelet agents, anticoagulants, calcium channel blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, β-blockers, diuretics, diuretics combined with antihypertensive, lipid-lowering agents, diabetic medications, and the number of concomitant medications (0 to ≥3). The discriminative performance of the propensity model was good (c-statistics=0.74) (Supplementary Figure 1).
Analytic results of both the unadjusted cohorts and the propensity score-matched cohorts are shown in parallel. The imbalance was assessed by standardized mean difference (SMD), which is advantageous over the P value in that it estimates the magnitude of the difference and is insensitive to sample size.15 The differences among clinical characteristics of cohorts were compared by using a t-test or chi-squared test, as appropriate. The cumulative risk of discontinuation was plotted on Kaplan-Meier curves and was compared by using a log-rank test. One minus the discontinuation rate was used to calculate the proportion of patients remaining. SAS version 9.4 (SAS Institute Inc.) and R version 3.6 (R Foundation for Statistical Computing) were used. Statistical significance was defined by a 2-tailed P<0.05.
A total of 2,547,603 patients met the inclusion criteria. After exclusion of 2,423,661 patients who were mostly treated with monotherapy or with statins other than atorvastatin or rosuvastatin, 123,992 patients were enrolled and classified into FDC (N=34,776, 28.1%) or FC groups (N=89,216, 71.9%) (Figure 2).
Patient enrollment process. Unadjusted analysis, and 1 : 3 propensity score-matched analysis details.
There was no significant difference in the comparison of demographics or clinical characteristics between FDC and FC groups, except for a higher frequency of dyslipidemia in the FDC compared with the FC (97.2% vs. 90.3%) group. The medication pattern at the time of study enrollment was different between the FDC and FC groups. “Switched” from other medications was most common in patients on a FDC (67.2%), whereas “add-on” of ARB or statin was most common in patients on a FC (55.9%). Patients in the FDC group were also prescribed a greater number of concomitant medications including calcium channel blocker (38.1% vs. 30.5%) and ARB (37.8% vs. 19.7%) (Table 1).
| Unmatched cohort (N=123,992) |
Propensity score-matched cohort (N=58,780) |
|||||
|---|---|---|---|---|---|---|
| FDC (N=34,776) |
FC (N=89,216) |
SMD | FDC (N=14,695) |
FC (N=44,085) |
SMD | |
| Demographics | ||||||
| Age group (years) | ||||||
| Age (mean±SD) | 62.50±11.64 | 63.16±11.67 | 0.057 | 62.61±11.73 | 62.55±11.82 | 0.005 |
| Male gender | 16,853 (48.5) | 42,679 (47.8) | 0.012 | 7,122 (48.5) | 21,316 (48.4) | 0.002 |
| History of comorbidities | ||||||
| Hypertension | 33,732 (97.0) | 86,523 (97.0) | 0.001 | 14,291 (97.3) | 42,701 (96.9) | 0.023 |
| Dyslipidemia | 33,818 (97.2) | 80,545 (90.3) | 0.291 | 14,224 (96.8) | 42,791 (97.1) | 0.016 |
| Diabetes mellitus | 17,141 (49.3) | 44,558 (49.9) | 0.013 | 7,342 (50.0) | 22,337 (50.7) | 0.014 |
| Non complicated† | 14,636 (42.1) | 37,438 (42.0) | 0.002 | 6,263 (42.6) | 19,118 (43.4) | 0.015 |
| Complicated† | 6,029 (17.3) | 17,132 (19.2) | 0.048 | 2,604 (17.7) | 8,193 (18.6) | 0.022 |
| Chronic kidney disease | 363 (1.0) | 1,772 (2.0) | 0.077 | 183 (1.2) | 672 (1.5) | 0.024 |
| Cardiovascular disease | 8,180 (23.5) | 24,433 (27.4) | 0.089 | 3,568 (24.3) | 10,836 (24.6) | 0.007 |
| Myocardial infarction | 1,416 (4.1) | 4,968 (5.6) | 0.070 | 615 (4.2) | 1,910 (4.3) | 0.007 |
| Congestive heart failure | 1,474 (4.2) | 4,622 (5.2) | 0.044 | 656 (4.5) | 2,063 (4.7) | 0.010 |
| Chronic respiratory disease | 3,422 (9.8) | 8,981 (10.1) | 0.008 | 1,469 (10.0) | 4,531 (10.3) | 0.009 |
| Liver disease | 8,963 (25.8) | 23,501 (26.3) | 0.013 | 3,897 (26.5) | 12,620 (28.6) | 0.047 |
| Rheumatological disease | 903 (2.6) | 2,839 (3.2) | 0.035 | 399 (2.7) | 1,312 (3.0) | 0.016 |
| Cancer | 1,041 (3.0) | 3,054 (3.4) | 0.024 | 463 (3.2) | 1,412 (3.2) | 0.003 |
| Stroke | 504 (1.4) | 1,624 (1.8) | 0.029 | 220 (1.5) | 762 (1.7) | 0.018 |
| Dementia | 786 (2.3) | 2,137 (2.4) | 0.009 | 353 (2.4) | 1,071 (2.4) | 0.002 |
| Paralysis | 381 (1.1) | 1,173 (1.3) | 0.020 | 183 (1.2) | 553 (1.3) | 0.001 |
| Charlson comorbidity index | ||||||
| Median (interquartile range) | 0 (0–2) | 1 (0–2) | 0.064 | 1 (0–2) | 1 (0–2) | 0.051 |
| 0 | 17,510 (50.4) | 42,432 (47.6) | 0.067 | 7,210 (49.1) | 20,602 (46.7) | 0.051 |
| 1 | 5,189 (14.9) | 13,998 (15.7) | 2,217 (15.1) | 6,661 (15.1) | ||
| 2 | 7,580 (21.8) | 19,841 (22.2) | 3,270 (22.3) | 10,361 (23.5) | ||
| 3 | 2,970 (8.5) | 8,149 (9.1) | 1,298 (8.8) | 4,173 (9.5) | ||
| ≥4 | 1,527 (4.4) | 4,796 (5.4) | 700 (4.8) | 2,288 (5.2) | ||
| Medication pattern | ||||||
| Naïve | 2,064 (5.9) | 3,755 (4.2) | 0.617 | 867 (5.9) | 3,000 (6.8) | 0.080 |
| Add-on | 9,352 (26.9) | 49,858 (55.9) | 4,571 (31.1) | 12,184 (27.6) | ||
| Switched | 23,360 (67.2) | 35,603 (39.9) | 9,257 (63.0) | 28,901 (65.6) | ||
| Prior medications | ||||||
| Lipid-lowering agent | 26,940 (77.5) | 67,437 (75.6) | 0.044 | 11,102 (75.5) | 31,386 (71.2) | 0.099 |
| Anti-hypertensive agents | ||||||
| Calcium channel blockers | 13,250 (38.1) | 27,211 (30.5) | 0.161 | 5,396 (36.7) | 15,712 (35.6) | 0.022 |
| ACE inhibitors | 1,361 (3.9) | 2,279 (2.6) | 0.077 | 538 (3.7) | 1,558 (3.5) | 0.007 |
| Angiotensin receptor blockers | 13,131 (37.8) | 17,532 (19.7) | 0.409 | 5,134 (34.9) | 14,722 (33.4) | 0.033 |
| β-blockers | 4,287 (12.3) | 12,274 (13.8) | 0.042 | 1,860 (12.7) | 5,497 (12.5) | 0.006 |
| Diuretics | 9,535 (27.4) | 21,835 (24.5) | 0.067 | 4,026 (27.4) | 12,284 (27.9) | 0.010 |
| Diuretics+anti-hypertensive agents‡ | 4,358 (12.5) | 14,551 (16.3) | 0.108 | 1,982 (13.5) | 6,284 (14.3) | 0.022 |
| Antiplatelet agent | 15,943 (45.8) | 42,681 (47.8) | 0.040 | 6,759 (46.0) | 19,845 (45.0) | 0.020 |
| Anticoagulants | 1,597 (4.6) | 4,474 (5.0) | 0.020 | 711 (4.8) | 2,141 (4.9) | 0.001 |
| Diabetic medications | 12,289 (35.3) | 32,323 (36.2) | 0.019 | 5,220 (35.5) | 15,707 (35.6) | 0.002 |
| Number of concomitant medications | ||||||
| 0 | 8,354 (24.0) | 16,758 (18.8) | 0.181 | 3,226 (22.0) | 9,344 (21.2) | 0.041 |
| 1 | 8,345 (24.0) | 20,085 (22.5) | 3,497 (23.8) | 10,323 (23.4) | ||
| 2 | 7,494 (21.5) | 18,214 (20.4) | 3,185 (21.7) | 9,226 (20.9) | ||
| ≥3 | 10,583 (30.4) | 34,159 (38.3) | 4,787 (32.6) | 15,192 (34.5) | ||
Data are presented as n (%). Tabulated comparison of patients on fixed-dose combinations (FDC) and patients on free combinations (FC). Propensity score matching process was conducted so that all parameters have a standardized mean difference (SMD) <0.10 and a P value >0.05 after matching. †Any diabetes without complication and diabetes with complications were counted. ‡Fixed combination of diuretics with anti-hypertensive agents.
Medication compliance decreased over time in both FDC and FC groups, as shown in the persistence rate curves and PDC bar plots. Compared to the FC group, the persistence rate of the FDC group was higher in both unadjusted analysis (54.5% vs. 40.1) and propensity score-matched analysis (53.6% vs. 35.2%; P<0.001, all) (Figure 3).
Medication persistence rate among patients in the fixed-dose combination (FDC) and free combination (FC) groups. (A) Unadjusted Kaplan-Meier curves for days to medication discontinuation. (B) Propensity score-adjusted Kaplan-Meier curves for days to medication discontinuation.
Compared to the FC group, the PDC for days 0–540 of the FDC group was higher in both unadjusted (84.1±23.1% vs. 63.1±34.9%) and propensity score-matched analyses (83.9±23.3% vs. 57.8±35.4%). The PDC for days 0–180, days 181–360, and days 361–540 were also higher in the FDC group (P<0.001, all) (Figure 4A,B). The proportion of good medication compliance defined by a PDC ≥80% for days 0-540 was higher in the FDC group in both unadjusted (75.5±43.0% vs. 48.1±50.0%) and propensity score-matched analyses (75.2±43.2% vs. 41.4±49.3%). The PDC ≥80% for days 0–180, days 181–360, and days 361–540 was also higher in the FDC group (P<0.001, all) (Figure 4C,D) (Supplementary Table 1).
Proportion of days covered (PDC) and the PDC ≥80% among patients in the fixed-dose combination (FDC) and free combination (FC) groups. (A,B) The proportion of good medication compliance defined by PDC ≥80% (C,D) was higher in the FDC compared to FC groups, in both unadjusted and propensity score-matched cohorts. Numerical results are described in Supplementary Table 1.
There was a total of 2,934 all-cause deaths (2.4%) during the study period. Compared to the FC group, the cumulative risk of death for the FDC group for days 0–540 was lower in both unadjusted analysis (1.8% vs. 2.6%, P<0.001) and propensity score matched analysis (2.0% vs. 2.5%, P<0.05) (Figure 5A,B).
Cumulative risk of all-cause death among patients in the fixed-dose combination (FDC) and free combination (FC) groups. (A,B) The cumulative risk of death for days 0–540 was lower in the FDC group compared to the FC group in both unadjusted analysis and propensity score-matched analyses. (C–F) In the landmark analyses at day 180 and 360, the cumulative risk of death was lower in unadjusted analyses, but not in propensity score-matched analyses. Numerical results are described in Supplementary Table 2.
The effect of FDC or FC on the risk of death was investigated further in the landmark analyses at day 180 and 360. In unadjusted analyses, the cumulative risk of death for days 181–540 and days 361–540 were lower in the FDC group (1.4% vs. 1.8%; 0.7% vs. 1.0%, respectively, P<0.001, all) (Figure 5C,E). In propensity score-matched analyses, the cumulative risk of death for days 181–540 and days 361–540 were not different between FDC and FC groups (1.6% vs. 1.8%; 0.8% vs. 0.9%, respectively, P>0.05) (Figure 5D,F) (Supplementary Table 2).
Subgroup AnalysesMedication compliance and all-cause death risk for the FDC and FC groups were also compared in subgroups classified by medication patterns. The decrease of all-cause death risk was significant in the switched groups but not in naïve or add-on groups (Table 2).
| Unadjusted (N=123,992) |
Propensity score matched (N=58,780) |
|
|---|---|---|
| P value | P value | |
| All medication patterns | ||
| PDC ≥80%, day 0–540* | <0.001 | <0.001 |
| Persistence, day 0–540 | <0.001 | <0.001 |
| All-cause death, day 0–540 | <0.001 | <0.01 |
| All-cause death, day 181–540 | <0.001 | >0.05 |
| All-cause death, day 361–540 | <0.001 | >0.05 |
| Unadjusted (N=5,819) |
Propensity score matched (N=3,867) |
|
| P value | P value | |
| Medication pattern: Naïve | ||
| PDC ≥80%, day 0–540* | <0.001 | <0.001 |
| Persistence, day 0–540 | <0.001 | <0.001 |
| All-cause death, day 0–540 | >0.05 | >0.05 |
| All-cause death, day 181–540 | >0.05 | >0.05 |
| All-cause death, day 361–540 | >0.05 | – |
| Unadjusted (N=59,210) |
Propensity score matched (N=16,755) |
|
| P value | P value | |
| Medication pattern: Add-on | ||
| PDC ≥80%, day 0–540* | <0.001 | <0.001 |
| Persistence, day 0–540 | <0.001 | <0.001 |
| All-cause death, day 0–540 | >0.05 | >0.05 |
| All-cause death, day 181–540 | >0.05 | >0.05 |
| All-cause death, day 361–540 | >0.05 | >0.05 |
| Unadjusted (N=58,963) |
Propensity score matched (N=38,158) |
|
| P value | P value | |
| Medication pattern: Switched | ||
| PDC ≥80%, day 0–540* | <0.001 | <0.001 |
| Persistence, day 0–540 | <0.001 | <0.001 |
| All-cause death, day 0–540 | <0.001 | <0.001 |
| All-cause death, day 181–540 | <0.001 | >0.05 |
| All-cause death, day 361–540 | 0.001 | >0.05 |
P value from the log-rank test or odds ratio are shown.
In this study, we showed that a 540-day medication compliance of a FDC was higher than FC among patients who were prescribed either atorvastatin or rosuvastatin in addition to ARB. Importantly, the results were derived from a nationwide healthcare database, which could reduce sampling or selection bias in retrospective large database studies.16 FDC of antihypertensive medications or antihypertensive with statin improved medication adherence and increased therapeutic efficacy.17–19 However, the comparative study of FDC and FC regarding ARB and statin are limited. To our knowledge, this study is the largest of its kind that compared medication compliance of FDC with those of FC for combination of ARB and statin.
Statin and ARB or angiotensin-converting enzyme (ACE) inhibitor are class I recommended medications in both European and American guidelines.9,20 Atorvastatin and rosuvastatin are potent statins and have been extensively studied for their anti-atherosclerotic effects. ARBs effectively lower blood pressure, improve endothelial dysfunction and insulin resistance caused by statin, and are expected to have synergistic anti-atherosclerotic effects with a statin.21,22 ARB is at least as effective as the ACE inhibitor at reducing cardiovascular risk.23 The combination of ARB with a statin is ideal for a single dosage form, given the fact that they act by a different mechanism, do not have widely different pharmacokinetics, and that there is no supra-additive toxicity of the ingredients.24
Our study showed lower all-cause death in FDC in day 0–540, but it is insufficient to confirm the clinical superiority of FDC given the inherent limitations of retrospective analysis. Despite better medication compliance in FDC, there was no difference of clinical outcome defined by risk of all-cause death between FDC and FC in landmark analyses of propensity score-matched cohorts. This might be explained by the following. First, the clinical effect of changing a medication regimen would be evident only after a sufficient lapse of time, but the study period of 540 days might be not long enough to test the effect of FDC or FC on the clinical outcome. The reduction of all-cause death was evident in the medication pattern of the switched group, but not of the naïve or add-on groups. Second, as the medication compliance constantly decreased during the 540-day study period, the difference in the effect of FDC over FC might be diluted over time. Third, the different pharmacokinetics between FDC and FC might affect the clinical outcome. FC allows patients to take the pills with a flexible schedule; the pills can be taken at different times or dropped as the patient chooses.25 FDC has less flexibility in modifying the doses of individual components and may expose patients to unnecessary therapy. The blood pressure may also be lowered too much. When these several substances are stopped at once, the risk of rebound hypertension may be higher in the FDC group compared to the FC group.26 Fourth, physicians can selectively prescribe FDC to patients who need flexible dosage adjustments or are at higher risk for non-adherence. This ‘confounding by indication’ artifact is difficult to exclude or adjust in a retrospective setting.
With the introduction of real-world big medical data analysis, real-world evidence can be generated and may complement the traditional randomized clinical trials or pragmatic trial. Further large-scaled studies of real-world data analysis or target trial emulation investigating patient-specific medication requirements and drug-taking behaviors may validate the benefit of FDC on the clinical outcome, which is challenging in the setting of traditional clinical trials (Supplementary Figure 2).27
Study LimitationsThe results of this study should to be interpreted with the following precautions. The source data consisted of administrative claims, which lack codes for specific conditions. The retrospective observational nature of the data limits the mitigation of unmeasured confounding variables, even with the use of multivariable and propensity modeling analyses.
Medication adherence was indirectly measured from electronic administrative claims data, assuming that patients take all dispensed medications. The real drug-taking behaviors were not investigated. The discrepancy for adherence measured by self-report, interview, or medication event monitoring systems is not uncommon.28,29 Besides, it is difficult to apply direct measures of medication adherence to a large population. Calculated medication adherence does not represent clinically decided prescriptions or cessation of medications.
There is a concern with FDC in terms of mismatching individual dosage requirements and identifying a drug in case of an adverse reaction. In contrast, FDC has the potential advantages of greater therapeutic efficacy, less risk of adverse events that may be evoked by high dose monotherapy, and less cost.25,26 These potential patient-specific clinical or economic issues were not investigated. Improved control of hypertension and hyperlipidemia might be a very useful endpoint, but such data were not available from the National Healthcare Insurance Service database. As the main benefits of antihypertensive treatment are largely independent of the drugs used, treatment persistence might be more important than medication persistence in the assessment of clinical outcome.8
In this real-world population study of patients who started both ARB and statin, FDC resulted in a better 540-day medication persistence and adherence compared to FC. However, the superiority of FDC regarding the reduction of all-cause death was not evident, which may warrant validation in a further investigation of real-world data or a large-scaled study.
The authors declare that they have no conflicts of interest.
This study was supported by the research grants from Hanmi Pharmaceutical Co., Ltd. and National Research Foundation of Korea (grant number 2017R1A2B310918).
The Samsung Medical Center Institutional Review Board (#2018-02-130) approved this study.
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-20-0966
