Abstract
Data regarding the long‐term outcomes of generic antihypertensive drugs are limited. This nationwide retrospective database analysis aimed to evaluate the efficacy and safety of a generic versus brand‐name nifedipine for hypertension treatment. Patients who were prescribed generic or brand‐name nifedipine between January 1, 2008, and December 31, 2013, were identified from the National Health Insurance Research Database of Taiwan. The efficacy outcomes included all‐cause mortality and the composite cardiovascular (CV) outcome, including CV death, non‐fatal myocardial infarction, non‐fatal stroke, coronary revascularization, and hospitalization for heart failure. Safety outcomes included headache, peripheral edema, constipation, acute kidney injury, hypotension, syncope, new diagnosis of cancer, and cancer death. Among the 98 335 patients who were eligible for analysis, 21 087 (21.4%) were prescribed generic nifedipine. Both the generic and the brand‐name groups included 21 087 patients after propensity score matching. At a mean follow‐up of 4.1 years, the generic nifedipine was comparable to the brand‐name drug with regard to all‐cause mortality (7.2% vs. 7.1%; hazard ratio [HR] 1.02, 95% confidence interval [CI] 0.95–1.09) and the composite CV outcomes (11.6% vs. 11.9%; HR 0.97; 95% CI 0.92–1.03). The generic nifedipine was associated with higher rates of headache, peripheral edema, and constipation but a modest reduction in the risk of newly diagnosed cancer (7.1% vs. 7.8%; subdistribution HR 0.90, 95% CI 0.84–0.97). The risks of acute kidney injury, hypotension, syncope, and cancer death were not significantly different between the two groups. In conclusion, the generic nifedipine was comparable to the brand‐name drug with regard to the risks of all‐cause mortality and the composite CV outcome. The finding of cancer risk could be chance and should be interpreted with caution.
Keywords: efficacy, generic drugs, hypertension, nifedipine, safety
1. INTRODUCTION
Hypertension has long been recognized as a global health issue and is a major preventable risk factor for cardiovascular (CV) disease. 1 , 2 , 3 , 4 A survey on the global burden of hypertension estimated that 26.4% of the adult population had hypertension in 2000, and the prevalence is expected to increase to 29% by 2025. 5 Medication‐based treatments for high blood pressure can lower the risk of coronary heart disease and heart failure by 20% to 25% and can lower stroke risk by 35% to 40%. 6 , 7 Given the continuously increasing prevalence of hypertension and the associated disease burden, hypertension management has become a top public health priority.
Generic drugs have been considered to be important components of healthcare systems, worldwide, due to the lower costs of generic drugs compared with their brand‐name counterparts. In the United States, generic drugs accounted for 80% of all prescriptions dispensed in 2011. 8 Generic drugs contain the same active chemical ingredients as brand‐name products and are approved by regulators based on evidence of pharmaceutical equivalence and bioequivalence with branded drugs. However, according to the US Food and Drug Administration (FDA), documentation of bioequivalence with branded drugs requires only 24–36 healthy volunteers. 9 Whether pharmaceutical equivalence and bioequivalence between generic drugs and their branded counterparts translate to equivalent clinical outcomes remains to be investigated. Large randomized controlled trials are rarely conducted to compare the effectiveness of generic and brand‐name drugs because they are not mandatory for generic drug approval.
Long‐acting nifedipine is a commonly used dihydropyridine calcium‐channel blocker (CCB) during hypertension management. The antihypertensive efficacy and safety profile of Nifedipine Gastrointestinal Therapeutic System (GITS) formulation has been demonstrated in large‐scale, randomized, clinical trials. 10 , 11 However, available generic drugs may not be all non‐inferior to their brand‐name counterpart. Among the extended‐release formulations, the osmotic‐controlled release oral delivery system (OROS) uses osmotic pressure as the driving force to deliver drugs through one or several laser‐drilled holes, allowing for a more predictable pharmacokinetic profile. 12 , 13 In this study, we aimed to compare the clinical outcomes of the generic and branded nifedipine OROS formulations for hypertension treatment. Furthermore, cancer risks associated with antihypertensive treatments have been the subject of debate for decades. 14 , 15 , 16 This concern may have been further raised by the recalls of some generic angiotensin II receptor blockers (ARBs) by the US Food and Drug Administration for containing probable human carcinogens. 17 These recalls may strengthen the negative perception that healthcare providers and patients harbor toward generic antihypertensive drugs. Therefore, we also compared the risks of newly diagnosed cancer and cancer death between the generic and the branded nifedipine.
2. METHODS
2.1. Data source
We used the National Health Insurance Research Database (NHIRD) of Taiwan to conduct this retrospective, nationwide, database analysis. The National Health Insurance (NHI) program is a single‐payer system with mandatory enrollment and low co‐pays that is operated by the government. The NHIRD contains detailed computerized claims data, submitted by medical institutions, including patient‐level data regarding demographic information, diagnoses, dates of admission and discharge, prescription drugs, and the use of medical facilities. Previous studies have validated the diagnostic accuracy of this database. 18 , 19 , 20 , 21 All antihypertensive drugs are available only through physician prescriptions in Taiwan. Every prescription of these drugs can be captured in the NHIRD. Both the brand‐name and the generic nifedipine are reimbursed by the insurance program, and the co‐payments of the two drugs are very low. Therefore, prescription of brand‐name or generic nifedipine may not be related to patients' economic status but at physicians' discretion. The NHIRD has encrypted all patients’ identification numbers to protect their privacy and to ensure anonymity; therefore, informed consent was waived. This study was approved by the Institutional Review Board at Linkou Chang Gung Memorial Hospital, Taiwan (IRB No.201901524B0).
2.2. Study cohort and exposure
From January 1, 2008, to December 31, 2013, a total of 144 983 patients who were diagnosed as hypertensive and prescribed OROS formulations of generic (Nifedipine SRFC, Chunghwa Yuming Healthcare Co., Taiwan) or brand‐name nifedipine (Adalat® OROS, Bayer), at a dose of 30 mg daily, were identified from the NHIRD. In Taiwan, the only generic nifedipine with the same release formation and dosage of the brand‐name nifedipine 30 mg is the aforementioned drug. Therefore, we excluded the patients who were prescribed nifedipine of different release formulations or a dosage of nifedipine other than 30 mg. The date that nifedipine was prescribed was assigned as the index date. The use of nifedipine was ascertained by the filling of at least two outpatient prescriptions or one refilled prescription for chronic illness during the first 90‐day window after the index date.
Figure 1 illustrates the flow chart used for patient enrollment. We excluded any patients who were aged less than 20 years (n = 123), pregnant (n = 194), or diagnosed with malignancy (n = 6864). We further excluded any patients who had concomitant prescriptions of other dihydropyridine CCBs, a dosage of nifedipine other than 30 mg (n = 34 123), who switch between generic and brand‐name nifedipine (n = 1363), and those who experienced CV events (n = 1156) or died during the 90‐day exposure window (n = 333). We also excluded the patients with a follow‐up period shorter than 90 days (n = 2492). Finally, a total of 98 335 patients who were diagnosed as hypertensive and treated with nifedipine were eligible for analysis.
FIGURE 1.
Flow chart for patient inclusion. Abbreviation: DCCB, dihydropyridine calcium‐channel blocker
2.3. Outcomes
The efficacy outcomes were defined as all‐cause mortality and the primary composite CV outcome, which included CV death, non‐fatal myocardial infarction (MI), non‐fatal stroke, coronary revascularization (percutaneous coronary intervention or coronary artery bypass grafting), and hospitalization for heart failure. CV death was defined as sudden cardiac death or deaths due to acute MI, heart failure, stroke, CV procedures, CV hemorrhage, and other CV causes. 22 The occurrence of MI, stroke, and heart failure required principle inpatient diagnosis. The accuracy of these diagnoses in the NHIRD has been validated in previous studies. 18 , 20 , 21 Coronary revascularization was detected using the Taiwan NHI reimbursement codes, from inpatient claims data. Mortality was defined by the withdrawal from the NHI program. 23
The safety outcomes included hypotension, syncope, headache, peripheral edema, constipation, acute kidney injury, new diagnosis of cancer, and cancer death. Hypotension and syncope were required to be diagnosed either in emergency departments or during hospitalization. Acute kidney injury was limited to diagnosis during hospitalization. Headache, peripheral edema, and constipation were detected using at least two outpatient visits or anyone hospitalization record. The diagnosis of cancer was ascertained by the possession of a Catastrophic Illness Certificate. All diseases were diagnosed according to the International Classification of Diseases, 9th Revision, Clinical Modification (ICD‐9‐CM) diagnosis codes (Supplemental Table 1). All patients were followed up until they experienced the occurrence of clinical outcomes, switched between generic and brand‐name nifedipine, were prescribed other dihydropyridine CCBs, or the end of the study period (December 31, 2013), whichever came first.
TABLE 1.
Selected baseline characteristics of the hypertensive patients with use of the generic or brand‐name nifedipine
| Variable | Before matching | After matching | ||||
|---|---|---|---|---|---|---|
| Generic (n = 21 087) | Brand‐name (n = 77 248) | STD | Generic (n = 21 087) | Brand‐name (n = 21 087) | STD | |
| Age (years) | 62.3 ± 13.2 | 63.1 ± 13.5 | −0.06 | 62.3 ± 13.2 | 62.3 ± 13.7 | <.01 |
| Male sex | 10 834 (51.4) | 41 126 (53.2) | −0.04 | 10 834 (51.4) | 10 852 (51.5) | <.01 |
| Comorbidity | ||||||
| Diabetes mellitus | 4311 (20.4) | 20 666 (26.8) | −0.15 | 4311 (20.4) | 4421 (21.0) | −.01 |
| Dyslipidemia | 3760 (17.8) | 18 738 (24.3) | −0.16 | 3760 (17.8) | 3871 (18.4) | −.01 |
| Prior stroke | 1197 (5.7) | 7753 (10.0) | −0.16 | 1197 (5.7) | 1207 (5.7) | <.01 |
| Heart failure | 465 (2.2) | 2930 (3.8) | −0.09 | 465 (2.2) | 492 (2.3) | −.01 |
| PAOD | 334 (1.6) | 1439 (1.9) | −0.02 | 334 (1.6) | 327 (1.6) | <.01 |
| Chronic kidney disease | 1449 (6.9) | 9358 (12.1) | −0.18 | 1449 (6.9) | 1,473 (7.0) | <.01 |
| COPD | 791 (3.8) | 3208 (4.2) | −0.02 | 791 (3.8) | 819 (3.9) | −.01 |
| Myocardial infarction | 101 (0.5) | 697 (0.9) | −0.05 | 101 (0.5) | 114 (0.5) | −.01 |
| Coronary artery disease | 2327 (11.0) | 11 871 (15.4) | −0.13 | 2327 (11.0) | 2420 (11.5) | −.01 |
| Obstructive sleep apnea | 32 (0.2) | 254 (0.3) | −0.04 | 32 (0.2) | 37 (0.2) | −.01 |
| Atrial fibrillation | 152 (0.7) | 1183 (1.5) | −0.08 | 152 (0.7) | 173 (0.8) | −.01 |
| Liver disease | 1271 (6.0) | 5067 (6.6) | −0.02 | 1271 (6.0) | 1298 (6.2) | −.01 |
| Charlson Comorbidity Index score | 0.69 ± 1.15 | 0.99 ± 1.39 | −0.24 | 0.69 ± 1.15 | 0.70 ± 1.12 | −.02 |
| Number of other antihypertensive agents | 1.2 ± 1.1 | 1.6 ± 1.1 | −0.37 | 1.2 ± 1.1 | 1.2 ± 1.0 | <.01 |
| Follow−up (years) | 4.1 ± 1.7 | 4.1 ± 1.8 | −0.02 | 4.1 ± 1.7 | 4.1 ± 1.8 | .01 |
Data were presented as frequency (percentage) or mean ± standard deviation.
COPD, chronic obstructive pulmonary disease; PAOD, peripheral arterial occlusive disease; STD, standardized difference.
2.4. Covariates
The covariates included sex, age, twelve comorbidities, Charlson Comorbidity Index (CCI) score, previous coronary intervention, nine classes of antihypertensive agents, other than nifedipine, and five classes of other medication. Comorbidities were defined by at least two outpatient diagnoses or one inpatient diagnosis during the previous year. The ICD‐9‐CM diagnostic codes for included comorbidities are listed in the supplemental information (Table S1). All information regarding medications during the 90‐day window after the index date was extracted from the claims data for outpatient visits or refills for chronic illnesses submitted to pharmacies, using the Anatomical Therapeutic Chemical codes or the Taiwan NHI reimbursement codes.
2.5. Statistical analysis
To reduce possible confounding factors due to treatment selection bias, a propensity score matching (PSM) method was used in this study. The propensity score was the predicted probability to be included in the generic nifedipine group, given the values of covariates, determined using multivariable logistic regression, without considering interaction effects. The variables used to calculate the propensity score are listed in Table 1, where the follow‐up year was replaced by the index date. Each patient in the generic nifedipine group was matched with one counterpart in the brand‐name nifedipine group. The matching was processed using a greedy, nearest‐neighbor algorithm, with a caliper of 0.2‐times the standard deviation of the logit of the propensity score, with random matching order and without replacement. The quality of matching was verified using the absolute value of the standardized difference (STD) between the groups, where a value of less than 0.1 was considered to be a negligible difference.
The risks of fatal time‐to‐event outcomes (ie, all‐cause mortality, primary composite outcome) between the groups were compared by the Cox proportional hazard model. The incidences of non‐fatal time‐to‐event outcomes (ie, stroke) were compared between groups by the Fine and Gray subdistribution hazard model, which considered all‐cause mortality a competing risk. The study group was the only explanatory variable in the survival models. The within‐pair clustering of outcomes after PSM was accounted for by using a robust standard error, known as a marginal model. 24
Subgroup analyses for the primary composite outcome were conducted on 12 pre‐specified subgroup variables, including sex, age (dichotomized between younger and older than 65 years), diabetes, dyslipidemia, prior stroke, chronic kidney disease, coronary artery disease, liver disease, number of other antihypertensive agents, antiplatelet agents, statins, and oral anti‐diabetic drugs. The proportions of newly diagnosed cancer subtypes were compared between groups using the chi‐squared test. A two‐sided p value <.05 was considered to be statistically significant, and no adjustments for multiple testing (multiplicity) were performed in this study. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC), including the “psmatch” procedure for PSM.
3. RESULTS
3.1. Patient characteristics
Among the 98 335 patients who were prescribed nifedipine for hypertension treatment, 21 087 were in the generic group and 77 248 were in the brand‐name group. Among the 21 087 patients in the generic group, 4288 (20.3%) switched to the brand‐name nifedipine from the 91th day after the index date to the end of follow‐up. Among the 77 248 patients in the brand‐name group, 7373 (9.5%) switched to the generic drug from the 91th day after the index date to the end of follow‐up. The patient characteristics and the use of medications are shown in Table 1 (brief) and Table S2 (complete). Before matching, the generic group had a lower prevalence of diabetes, dyslipidemia, prior stroke, chronic kidney disease, and coronary artery disease. Both groups had low CCI scores (<1). The generic group was less frequently prescribed alpha‐blockers, angiotensin receptor blockers, beta‐blockers, loop diuretics, nitrates, vasodilators, antiplatelet agents, statins, oral anti‐diabetic drugs, or insulin compared with the brand‐name group. Similar to previous studies, 25 about 70% of the patients enrolled in our study required at least two agents for blood pressure control. The average numbers of other prescribed antihypertensive drugs were 1.2 and 1.6 for the generic and brand‐name groups before propensity score matching, respectively. The mean follow‐up duration was 4.1 years for both groups. After matching, all covariates were balanced between the two groups.
3.2. Clinical outcomes
Table 2 shows the efficacy and the safety outcomes. At a mean follow‐up of 4.1 years, generic and brand‐name nifedipine had comparable efficacy outcomes with regard to the risk of all‐cause mortality (7.2% vs. 7.1%; hazard ratio [HR] 1.02; 95% confidence interval [CI] 0.95–1.09) and the primary composite CV outcome (11.6% vs. 11.9%; HR 0.97; 95% CI 0.92–1.03). No significant differences were observed for the risks of the individual components of the composite CV outcome between the two groups. Figure 2A and B illustrate the cumulative event rates for all‐cause mortality and the primary composite CV outcome, respectively.
TABLE 2.
Follow‐up outcomes of the hypertensive patients with use of the generic or brand‐name nifedipine
| Outcome | Generic (n = 21 087) | Brand‐name (n = 21 087) | HR or SHR of Generic (95% CI) | p |
|---|---|---|---|---|
| Efficacy outcomes | ||||
| All‐cause mortality | 1527 (7.2) | 1496 (7.1) | 1.02 (0.95–1.09) | .597 |
| Primary composite cardiovascular events | 2444 (11.6) | 2516 (11.9) | 0.97 (0.92–1.03) | .354 |
| CV death | 786 (3.7) | 808 (3.8) | 0.97 (0.88–1.07) | .562 |
| Non‐fatal myocardial infarction | 288 (1.37) | 286 (1.36) | 1.003 (0.85–1.18) | .972 |
| Non‐fatal stroke | 1139 (5.4) | 1102 (5.2) | 1.05 (0.97–1.15) | .211 |
| Coronary revascularization | 613 (2.9) | 673 (3.2) | 0.91 (0.81–1.01) | .072 |
| Heart failure | 350 (1.66) | 353 (1.67) | 0.99 (0.85–1.14) | .865 |
| Safety outcomes | ||||
| Hypotension | 210 (1.0) | 235 (1.1) | 0.89 (0.74–1.07) | .218 |
| Syncope | 410 (1.9) | 451 (2.1) | 0.91 (0.79–1.03) | .143 |
| Headache | 5008 (23.7) | 4185 (19.8) | 1.22 (1.17–1.27) | <.001 |
| Peripheral edema | 2023 (9.6) | 1619 (7.7) | 1.26 (1.18–1.35) | <.001 |
| Constipation | 4012 (19.0) | 3831 (18.2) | 1.045 (1.001–1.090) | .048 |
| Acute kidney injury | 732 (3.5) | 777 (3.7) | 0.94 (0.85–1.04) | .198 |
| New diagnosis of cancer | 1506 (7.1) | 1655 (7.8) | 0.90 (0.84–0.97) | .003 |
| Cancer death | 387 (1.84) | 374 (1.77) | 1.03 (0.90–1.19) | .660 |
Data were presented as frequency (percentage).
CI, confidence interval; CV, cardiovascular; HR, hazard ratio; SHR, subdistribution hazard ratio.
FIGURE 2.
Cumulative event rates for the primary composite outcome (A) and all‐cause mortality (B) for the brand‐name and generic nifedipine groups, in the propensity‐score‐matched cohort
With respect to the safety outcomes, the risks of hypotension, syncope, and acute kidney injury were not significantly different between the two groups (Table 2). However, the generic nifedipine was associated with a higher risk of headache (23.7% vs. 19.8%; subdistribution HR [SHR] 1.22; 95% CI 1.17–1.27), peripheral edema (9.6% vs. 7.7%; SHR 1.26; 95% CI 1.18–1.35), and constipation (19.0% vs. 18.2%; SHR 1.045; 95% CI 1.001–1.09). The generic nifedipine was associated with a modest reduction in the risk of newly diagnosed cancer (7.1% vs. 7.8%; SHR 0.90, 95% CI 0.84–0.97), but the risk of cancer death was comparable between the two groups (1.84% vs. 1.77%; HR 1.03; 95% CI 0.9–1.19). Detailed results regarding the subtypes of newly diagnosed cancers are shown in Table 3. The proportions of head and neck (7.2% vs. 4.4%; p = .001) and pancreatic cancers (2.9% vs. 1.7%; p = .021) were higher in the generic group than in the brand‐name group. However, the proportions of breast (5.4% vs. 8.6%; p = .001), kidney (2.7% vs. 4.2%; p = .027), and thyroid cancers (1.3% vs. 2.4%; p = .025) were lower in the generic group than the brand‐name group.
TABLE 3.
Cancer type of the hypertensive patients with use of the generic or brand‐name nifedipine in the propensity score matched cohort
| Cancer type (ICD‐9‐CM diagnostic code) | Total (n = 3161) | Generic (n = 1506) | Brand‐name (n = 1655) | p value |
|---|---|---|---|---|
| Head and neck (140–149) | 180 (5.7) | 108 (7.2) | 72 (4.4) | .001 |
| Digestive system | ||||
| Esophagus (150) | 43 (1.4) | 24 (1.6) | 19 (1.1) | .280 |
| Stomach (151) | 141 (4.5) | 61 (4.1) | 80 (4.8) | .287 |
| Colon and rectum (153–154) | 513 (16.2) | 253 (16.8) | 260 (15.7) | .407 |
| Liver and biliary tract (155–156)) | 470 (14.9) | 233 (15.5) | 237 (14.3) | .364 |
| Pancreas (157) | 72 (2.3) | 44 (2.9) | 28 (1.7) | .021 |
| Lung and mediastinum (160–165) | 434 (13.7) | 214 (14.2) | 220 (13.3) | .454 |
| Bone and soft tissue (170–171) | 20 (0.63) | 12 (0.80) | 8 (0.48) | .267 |
| Skin (172–173) | 112 (3.5) | 53 (3.5) | 59 (3.6) | .945 |
| Breast (174–175) | 224 (7.1) | 82 (5.4) | 142 (8.6) | .001 |
| Genitourinary system | ||||
| Uterus (179, 182) | 48 (1.52) | 23 (1.53) | 25 (1.51) | .970 |
| Cervix (180) | 72 (2.3) | 35 (2.3) | 37 (2.2) | .868 |
| Ovary (183) | 22 (0.70) | 10 (0.66) | 12 (0.73) | .837 |
| Prostate (185) | 208 (6.6) | 97 (6.4) | 111 (6.7) | .763 |
| Bladder (188) | 125 (4.0) | 55 (3.7) | 70 (4.2) | .405 |
| Kidney (189) | 110 (3.5) | 41 (2.7) | 69 (4.2) | .027 |
| Central nervous system (191–192) | 42 (1.329) | 20 (1.3) | 22 (1.3) | .997 |
| Thyroid (193) | 60 (1.9) | 20 (1.3) | 40 (2.4) | .025 |
| Hematologic and lymphatic (200–208) | 110 (3.5) | 48 (3.2) | 62 (3.7) | .392 |
| Others | 155 (4.9) | 73 (4.8) | 82 (5.0) | .889 |
ICD‐9 CM, International Classification of Diseases, Ninth Revision, Clinical Modification.Bold values are statistic significance
3.3. Subgroup analysis for the primary composite outcome
In the pre‐specified subgroup analysis performed for the primary composite outcome, the results demonstrated that the brand‐name nifedipine was associated with more favorable outcomes in the subgroup of patients with dyslipidemia (p for interaction = .029). On the other hand, compared with the branded drug, the generic nifedipine was associated better outcomes in the subgroups of patients with concomitant antiplatelet agents (p for interaction = .04) and without other concomitant antihypertensive drugs (p for interaction = .025) (Figure 3).
FIGURE 3.
Pre‐specified subgroup analyses of the primary composite outcome in the propensity‐score‐matched cohort. Abbreviation: CI, confidence interval; HR, hazard ratio
4. DISCUSSION
This nationwide database study compared the clinical outcomes of hypertensive patients who were treated with the OROS formulations of either generic or brand‐name nifedipine. We found that the generic nifedipine was comparable to the brand‐name nifedipine with regard to all‐cause mortality and the primary composite CV outcome, including CV death, non‐fatal MI, non‐fatal stroke, coronary revascularization, and hospitalization for heart failure. The generic nifedipine was associated with higher rates of headache, peripheral edema, and constipation and a lower risk of newly diagnosed cancer compared with its brand‐name counterpart. The finding of newly diagnosed cancer should be interpreted with caution because of the lack of mechanistic explanations. In addition, the risk of cancer death was comparable between the two groups.
No large RCTs have ever compared the efficacy and safety of generic and branded drugs for the treatment of hypertension or other CV diseases. Thus, no recommendation has been given for generic drugs in the international guidelines on hypertension management. The primary analysis of the present study showed no significant differences in the risks of all‐cause mortality and the composite CV outcome between patients who were prescribed generic and brand‐name nifedipine. Our results are consistent with the earlier observation that generic antihypertensive drugs may be as effective as brand‐name drugs in preventing CV events. In a meta‐analysis of studies examining the outcomes of generic versus branded drugs, Kesselheim and colleagues identified 47 articles, covering 9 subclasses of CV medications, among which 38 (81%) were RCTs with small sample sizes. 26 The study showed no evidence of brand‐name drug superiority compared with generic drugs for the treatment of CV diseases. Editorials addressing generic substitutions were also reviewed in this study; however, more than half of those (53%) did not support the interchangeability of generic drugs. A more recent meta‐analysis of 53 RCTs examining CV medicines, performed by Manzoli and colleagues, reported no significant differences between the generic and brand‐name drugs for the combined estimate of efficacy or in any stratified analysis of outcomes. 27 Although both meta‐analyses included RCTs that compared generic and brand‐name CCBs, the majority of these were conducted on amlodipine, and none were conducted on nifedipine. Furthermore, many of these trials were bioequivalence studies, which are limited by small patient numbers, short follow‐up periods, and the inclusion of disproportionately young and healthy participants; therefore, these studies are often inadequately powered to detect significant differences in clinical outcomes. 27 In a population‐based, case–control study, patients who were always treated with generic antihypertensive drugs did not experience different CV risks than those who were always treated with brand‐name drugs or who switched between generic and brand‐name drugs for blood pressure control. 28 A recent analysis of two US commercial insurance databases further supported the comparative effectiveness of generic and brand‐name antihypertensive drugs, for the composite endpoint of hospitalization for MI, ischemic stroke, and coronary revascularization procedures. 29
The meta‐analysis by Manzoli and colleagues showed similar risks of adverse events between generic and branded CV medications, for all stratified analyses, including drug class and the severity of adverse events. 27 In contrast, a population‐based time‐series study performed in Canada showed an increase in adverse events (hospitalizations and emergency room consultations) immediately following the commercialization of 3 generic ARBs. 30 However, the major limitations of that study included a lack of risk‐adjustment methodology for the outcome analysis and the lack of details regarding the reasons for more frequent emergency room visits and hospitalizations being associated with generic drugs. Our study examined some of the common side effects during nifedipine treatment. The generic nifedipine was associated with higher rates of headache, peripheral edema, and constipation. However, there was no significant difference between generic and brand‐name nifedipine for the risks of acute kidney injury, syncope, or hypotension, which might require emergency room visits or hospitalizations. Our findings support the notion that using generic instead of brand‐name drugs does not imply a loss in safety.
Although reports have been published raising concerns regarding the relationship between nifedipine and cancer, 31 , 32 , 33 some studies have shown that such a relationship may have been the result of chance. 34 , 35 In the present study, we found a modest reduction in the risk of newly diagnosed cancer in the generic nifedipine group. However, this finding should be interpreted with caution because of the lack of mechanistic explanations. The excessive cancer risk associated with the brand‐name nifedipine could be explained by overdiagnosis. Individuals in the brand‐name group had more medications and more comorbidities, which may have increased their probability of going to the doctor and being diagnosed with cancer. Furthermore, the risk of cancer death was comparable between the two groups. Further studies are warranted to examine the associations between antihypertensive drugs and cancer and the comparative risks of cancer between generic and brand‐name drugs.
4.1. Study limitations
The strength of this health administrative database study is the large sample size and comprehensive patient follow‐up. However, the retrospective study design has several inherent limitations. 36 First, this study may be subject to selection bias since the patients who were prescribed the brand‐name nifedipine were generally sicker and required more antihypertensive agents than those who were prescribed the generic nifedipine before propensity score matching. We attempted to mitigate potential biases by matching for comorbid conditions and proxies of blood pressure and some laboratory data, such as the numbers and types of antihypertensive drugs and the use of statins and oral anti‐diabetic drugs. However, unmeasured variables can still confound data interpretation. The subgroup analysis of the primary composite outcome showed more favorable outcomes with the use of brand‐name nifedipine among the patients with dyslipidemia and more favorable outcomes with the use of generic nifedipine among the patients prescribed antiplatelet agents but without other concomitant antihypertensive drugs. Due to the lack of mechanistic explanations, caution should be taken when interpreting these findings. Although we did rigorous propensity score matching to balance potential differences between the treatment groups, potential selection bias and unmeasured confounding may have influenced the results of subgroup analysis. Second, the clinical outcomes were extracted from the database but were not verified independently. Insurance claims for each health care encounter were all reviewed and inspected by medical reimbursement specialists. Although we cannot exclude the possibility of incorrect coding in daily practice, physicians or institutions that violate clinical practice guidelines should be penalized. The risk of ascertainment bias for mortality, MI, and stroke in this study may be low because these events have been validated in the NHIRD of Taiwan. 18 , 20 , 21 , 23 The diagnosis of cancer was ascertained by the Catastrophic Illness Certificate; therefore, the diagnostic accuracy is high. 37 Third, the safety outcomes did not include edema, which is a common side effect of CCBs. Physicians may not always incorporate edema in patients' diagnoses or take specific actions for this side effect. Therefore, the prevalence of edema could be underestimated in the database. Fourth, this study compared the outcomes of OROS formulations of generic versus brand‐name nifedipine in a relatively low‐risk population. The results cannot be generalized to other formulations of nifedipine, different antihypertensive drugs, or populations with higher comorbidity burdens. Finally, given the retrospective design of the present study, our results are hypothesis‐generating and cannot resolve the controversy surrounding generic and brand‐name interchangeability. Well‐powered, randomized controlled trials, with sufficient follow‐up, are warranted to address this issue.
5. CONCLUSIONS
In this nationwide retrospective cohort study, the generic nifedipine was comparable to its brand‐name counterpart with regard to all‐cause mortality and the composite CV outcome, acute kidney injury, syncope, and hypotension. The rates of headache, peripheral edema, and constipation were higher in the generic group than in the brand‐name group. The generic nifedipine may be associated with a lower risk of newly diagnosed cancer compared with its brand‐name counterpart. However, this finding could be chance and should be interpreted with caution.
CONFLICT OF INTEREST
None.
AUTHOR CONTRIBUTIONS
Ying‐Chang Tung, Wen‐Jone Chen, and Pao‐Hsien Chu involved in design. Ying‐Chang Tung, Tzyy‐Jer Hsu, Chia‐Pin Lin, Fu‐Chih Hsiao, and You‐Chia Chu involved in data collection and analysis. Ying‐Chang Tung, Tzyy‐Jer Hsu, and Chia‐Pin Lin involved in draft. Wen‐Jone Chen and Pao‐Hsien Chu involved in final approving.
ETHICAL APPROVAL
This study was approved by the Institutional Review Board at Linkou Chang Gung Memorial Hospital, Taiwan (IRB No.201901524B0).
INFORMED CONSENT
The NHIRD has encrypted all patients’ identification numbers to ensure anonymity; therefore, informed consent was waived.
Supporting information
Table S1‐S2
ACKNOWLEDGEMENTS
The authors would like to thank Alfred Hsing‐Fen Lin, MS, Raising Statistics Consultant Inc, for his statistical assistance. Mr Lin received compensation and declared no competing interests between the findings of this study and his company.
Tung Y‐C, Hsu T‐J, Lin C‐P, et al. Efficacy and safety outcomes of one generic nifedipine versus ADALAT long‐acting nifedipine for hypertension management. J Clin Hypertens 2020;22:2296–2305. 10.1111/jch.14070
†Ying‐Chang Tung and Tzyy‐Jer Hsu are equal contribution.
REFERENCES
- 1. Angell SY, De Cock KM, Frieden TR. A public health approach to global management of hypertension. Lancet. 2015;385(9970):825‐827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2224‐2260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lawes CM, Vander Hoorn S, Rodgers A. Global burden of blood‐pressure‐related disease, 2001. Lancet. 2008;371(9623):1513‐1518. [DOI] [PubMed] [Google Scholar]
- 4. Collaborators GRF . Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;386(10010):2287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worldwide data. Lancet. 2005;365(9455):217‐223. [DOI] [PubMed] [Google Scholar]
- 6. Collaboration BPLTT . Blood pressure‐lowering treatment based on cardiovascular risk: a meta‐analysis of individual patient data. Lancet. 2014;384(9943):591‐598. [DOI] [PubMed] [Google Scholar]
- 7. Law M, Morris J, Wald N. Use of blood pressure lowering drugs in the prevention of cardiovascular disease: meta‐analysis of 147 randomised trials in the context of expectations from prospective epidemiological studies. BMJ. 2009;338:b1665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Nash DB. The use of medicines in the United States: a detailed review. Am Health Drug Benefits. 2012;5(7):423. [PMC free article] [PubMed] [Google Scholar]
- 9. Lawrence XY, Li BV. FDA Bioequivalence Standards. Vol 13. Springer; 2014. [Google Scholar]
- 10. Brown MJ, Palmer CR, Castaigne A, et al. Morbidity and mortality in patients randomised to double‐blind treatment with a long‐acting calcium‐channel blocker or diuretic in the International Nifedipine GITS study: Intervention as a Goal in Hypertension Treatment (INSIGHT). Lancet. 2000;356(9227):366‐372. [DOI] [PubMed] [Google Scholar]
- 11. Mancia G, Brown M, Castaigne A, et al. Outcomes with nifedipine GITS or Co‐amilozide in hypertensive diabetics and nondiabetics in Intervention as a Goal in Hypertension (INSIGHT). Hypertension. 2003;41(3):431‐436. [DOI] [PubMed] [Google Scholar]
- 12. Conley R, Gupta SK, Sathyan G. Clinical spectrum of the osmotic‐controlled release oral delivery system (OROS), an advanced oral delivery form. Curr Med Res Opin. 2006;22(10):1879‐1892. [DOI] [PubMed] [Google Scholar]
- 13. Malaterre V, Ogorka J, Loggia N, Gurny R. Oral osmotically driven systems: 30 years of development and clinical use. Eur J Pharm Biopharm. 2009;73(3):311‐323. [DOI] [PubMed] [Google Scholar]
- 14. Sipahi I, Debanne SM, Rowland DY, Simon DI, Fang JC. Angiotensin‐receptor blockade and risk of cancer: meta‐analysis of randomised controlled trials. Lancet Oncol. 2010;11(7):627‐636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Bangalore S, Kumar S, Kjeldsen SE, et al. Antihypertensive drugs and risk of cancer: network meta‐analyses and trial sequential analyses of 324 168 participants from randomised trials. Lancet Oncol. 2011;12(1):65‐82. [DOI] [PubMed] [Google Scholar]
- 16. Hicks BM, Filion KB, Yin H, Sakr L, Udell JA, Azoulay L. Angiotensin converting enzyme inhibitors and risk of lung cancer: population based cohort study. BMJ. 2018;363:k4209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.FDA Updates and Press Announcements on Angiotensin II Receptor Blocker Recalls. Available from: https://www.fda.gov/drugs/drug‐safety‐and‐availability/fda‐updates‐and‐press‐announcements‐angiotensin‐ii‐receptor‐blocker‐arb‐recalls‐valsartan‐losartan
- 18. Cheng CL, Kao YHY, Lin SJ, Lee CH, Lai ML. Validation of the National Health Insurance Research Database with ischemic stroke cases in Taiwan. Pharmacoepidemiol Drug Saf. 2011;20(3):236‐242. [DOI] [PubMed] [Google Scholar]
- 19. Wu C‐S, Lai M‐S, Gau SS‐F, Wang S‐C, Tsai H‐J. Concordance between patient self‐reports and claims data on clinical diagnoses, medication use, and health system utilization in Taiwan. PLoS One. 2014;9(12):e112257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Cheng C‐L, Chien H‐C, Lee C‐H, Lin S‐J, Yang Y‐HK. Validity of in‐hospital mortality data among patients with acute myocardial infarction or stroke in National Health Insurance research database in Taiwan. Int J Cardiol. 2015;201:96‐101. [DOI] [PubMed] [Google Scholar]
- 21. Hsieh C‐Y, Chen C‐H, Li C‐Y, Lai M‐L. Validating the diagnosis of acute ischemic stroke in a National Health Insurance claims database. J Formos Med Assoc. 2015;114(3):254‐259. [DOI] [PubMed] [Google Scholar]
- 22. Hicks KA, Mahaffey KW, Mehran R, et al. 2017 Cardiovascular and stroke endpoint definitions for clinical trials. J Am Coll Cardiol. 2018;71(9):1021‐1034. [DOI] [PubMed] [Google Scholar]
- 23. Wu C‐Y, Chen Y‐J, Ho HJ, et al. Association between nucleoside analogues and risk of hepatitis B virus–related hepatocellular carcinoma recurrence following liver resection. JAMA. 2012;308(18):1906‐1913. [DOI] [PubMed] [Google Scholar]
- 24. Austin PC, Fine JP. Propensity‐score matching with competing risks in survival analysis. Stat Med. 2019;38(5):751‐777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Gradman AH, Basile JN, Carter BL, Bakris GL; Group ASoHW . Combination therapy in hypertension. J Am Soc Hypertens. 2010;4(2):90‐98. [DOI] [PubMed] [Google Scholar]
- 26. Kesselheim AS, Misono AS, Lee JL, et al. Clinical equivalence of generic and brand‐name drugs used in cardiovascular disease: a systematic review and meta‐analysis. JAMA. 2008;300(21):2514‐2526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Manzoli L, Flacco ME, Boccia S, et al. Generic versus brand‐name drugs used in cardiovascular diseases. Eur J Epidemiol. 2016;2016(31):351‐368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Corrao G, Soranna D, Merlino L, Mancia G. Similarity between generic and brand‐name antihypertensive drugs for primary prevention of cardiovascular disease: evidence from a large population‐based study. Eur J Clin Invest. 2014;44(10):933‐939. [DOI] [PubMed] [Google Scholar]
- 29. Desai RJ, Sarpatwari A, Dejene S, et al. Comparative effectiveness of generic and brand‐name medication use: a database study of US health insurance claims. PLoS Med. 2019;16(3).e1002763. https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002763 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Leclerc J, Blais C, Rochette L, Hamel D, Guénette L, Poirier P. Impact of the commercialization of three generic angiotensin II receptor blockers on adverse events in Quebec, Canada: a population‐based time series analysis. Circ Cardiovasc Qual Outcomes. 2017;10(10):e003891. [DOI] [PubMed] [Google Scholar]
- 31. Pahor M, Guralnik JM, Ferrucci L, et al. Calcium‐channel blockade and incidence of cancer in aged populations. Lancet. 1996;348(9026):493‐497. [DOI] [PubMed] [Google Scholar]
- 32. Kanamasa K, Kimura A, Miyataka M, Takenaka T, Ishikawa K; Group SP . Incidence of cancer in postmyocardial infarction patients treated with short‐acting nifedipine and diltiazem. Cancer. 1999;85(6):1369‐1374. [DOI] [PubMed] [Google Scholar]
- 33. Guo D‐Q, Zhang H, Tan S‐J, Gu Y‐C. Nifedipine promotes the proliferation and migration of breast cancer cells. PLoS One. 2014;9(12):e113649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Cohen HJ, Pieper CF, Hanlon JT, Wall WE, Burchett BM, Havlik RJ. Calcium channel blockers and cancer. Am J Med. 2000;108(3):210‐215. [DOI] [PubMed] [Google Scholar]
- 35. Poole‐Wilson PA, Kirwan B‐A, Vokó Z, et al. Safety of nifedipine GITS in stable angina: the ACTION trial. Cardiovasc Drugs Ther. 2006;20(1):45‐54. [DOI] [PubMed] [Google Scholar]
- 36. Grimes DA. Epidemiologic research using administrative databases: garbage in, garbage out. Obstet Gynecol. 2010;116(5):1018‐1019. [DOI] [PubMed] [Google Scholar]
- 37. Hsieh C‐Y, Su C‐C, Shao S‐C, et al. Taiwan’s National Health Insurance Research Database: past and future. Clin Epidemiol. 2019;11:349. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1‐S2