Notes
Factors Affecting Patients’ Acceptance of Switching to Biosimilars Are Disease-Dependent: A Cross-Sectional Study
2023 Volume 46 Issue 1 Pages 128-132
Details
2023 Volume 46 Issue 1 Pages 128-132
Biosimilars (BS) are promoted worldwide because of the high cost of biologics. However, patients are apprehensive about switching to BS. For some diseases, several factors, which may be disease-dependent, influence patients’ acceptance of switching to BS. Herein, we evaluated whether factors influencing acceptance for switching were disease-dependent among Japanese patients with different diseases. This cross-sectional study involved pharmacists’ interviews with patients who used or planned to use biologics. Demographic and clinical characteristics were retrospectively investigated using the patients’ medical records. Multivariate logistic regression showed that switch refusal was associated with a history of adverse reactions to biologics (odds ratio [95% confidence interval (CI)] = 3.38 [1.35–8.44]), history of complaints related to disease activity (3.57 [1.53–8.32]), and unacceptability of generic drugs (7.62 [2.70–21.60]). Subgroup analyses suggested that the unacceptability of generic drugs was a common factor, regardless of the disease. Concomitantly, histories of adverse reactions to biologics and complaints related to disease activity were disease-dependent factors. Healthcare professionals should help patients in selecting BS, considering these factors according to the disease.
Biologics are widely used to treat various diseases, and governments of several countries have promoted biosimilars (BS) because of the high cost of bio-originators.1,2) BS should be similar to the reference drug in terms of quality, safety, and efficacy; this is one of the principles for assessing the utility of BS in several countries.3) However, in recent years, patients’ negative perceptions regarding switching to BS have reportedly led to nocebo effects, including new or worsening symptoms and adverse events.4) Gasteiger et al.5) showed that patients’ negative safety perceptions are associated with their sex (female), short-term bio-originator use, illness beliefs, seeking health information online, high perceived sensitivity, and negative beliefs about medicines. Scherlinger et al.6) reported that patients’ refusal to switch to BS is correlated with a negative opinion regarding generic drugs, older age, and longer disease duration. However, these studies only included patients with rheumatic diseases; therefore, little is known about patient-associated factors in other diseases.
While educating patients about BS and encouraging patients to switch to BS, we observed that factors influencing patients’ acceptance regarding switching differ depending on the disease. Disease symptoms, efficacy and adverse effects of drugs, and support provided for treatment expenses vary with the disease. Therefore, the factors affecting patients’ acceptance of BS may differ depending on the disease, and thus warrants further research.
In this study, we aimed to investigate whether the factors affecting patients’ acceptance to switch to BS were disease-dependent in Japanese patients with malignant lymphoma, inflammatory bowel disease (IBD), rheumatic disease, and osteoporosis.
A cross-sectional study to examine the knowledge of BS, acceptance of switching to BS, and its factors in Kameda Medical Center, Chiba Prefecture (with 912 beds and approximately 3000 outpatients per day) was designed; we collected information from patients’ medical records following validation by two people.
Ethical ConsiderationsThe study complied with the standards of the Declaration of Helsinki and current Ethical Guidelines for Medical and Health Research Involving Human Subjects (https://www.mhlw.go.jp/file/06-Seisakujouhou-10600000-Daijinkanboukouseikagakuka/0000080278.pdf). The study design and methodology, including the opt-out method of consent made available to all patients, were approved by the Kameda General Hospital Clinical Research Review Committee (Approval No. 21-002), and they adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Patient consent was obtained before the study.
ProcedureFigure 1 shows a flow chart of the study design. Patients who visited the hematology (n = 924), gastroenterological medicine (n = 152), or rheumatology (n = 4122) departments of Kameda Medical Center between April 2017 and December 2020 were screened for participation in the study. We included patients who 1) had primary or relapsed malignant lymphoma, IBD, rheumatic disease, or osteoporosis; 2) were taking or were scheduled for bio-originator treatments, and 3) were aged ≥20 years. We excluded patients who were not interviewed by pharmacists and those who did not understand the pharmacists’ explanation of BS. Pharmacists, who were approved by the Japanese Society of Hospital Pharmacists (https://www.jshp.or.jp/gaiyou/pamphlet/pamphlet4.pdf), asked patients if they were familiar with BS and provided verbal and written information on 1) manufacturing biologics; 2) similarities between bio-originators and BS in terms of quality, safety, and efficacy, and 3) the potential for cost savings. The pharmacists then ensured patients’ understanding and asked whether they accepted the switch to generics and BS. If patients were reluctant to switch, the pharmacists asked for the reason. Based on their responses, we assigned the patients who accepted the switch to BS to the “acceptance group” and patients who refused to switch to the “refusal group.”
BS, biosimilars; IBD, inflammatory bowel disease.
We retrospectively investigated patients’ demographic (age, sex, and living with family or not) and clinical (diseases, duration of the disease in non-cancer patients, newly diagnosed or relapsed in malignant lymphoma patients, medication history of biologic use, history of adverse reactions to the drug, and complaints related to disease activity shown in Table 1) characteristics from medical records. Additionally, we checked whether patients used public medical expense support programs beyond the universal health care system: “Treatment and Research Enterprise for Specified Diseases” or “High-Cost Medical Treatment System.”7,8)
| Disease | Complaint related to disease activity |
|---|---|
| Malignant lymphoma | B symptoms (fever > 38 °C, weight loss > 5%, or night sweats) |
| Inflammatory bowel disease | Liquid or soft stools; Abdominal pain |
| Rheumatoid arthritis | Tender joints; Swollen joints |
| Osteoporosis | Fracture |
| Behçet’s disease | Genital and mouth ulcers; Red, painful eyes and blurred vision; Acne-like spots; Headaches; Painful, stiff, and swollen joints |
| Systemic lupus erythematosus | Recent onset seizure; Psychosis; Organic brain syndrome; Visual disturbance; New-onset sensory or motor neuropathy involving cranial nerves, Lupus headache; Vasculitis; Arthritis; Myositis |
| Psoriatic arthritis | Tender Joints; Swollen Joints |
| Ankylosing spondylitis | Fatigue and tiredness; Neck, back, or hip pain; Pain or swelling in other joints; Discomfort from tender areas; Morning stiffness |
| Palmoplanter pustulosis | Erythema; Pustules; Desquamation |
| Adult-onset Still’s disease | Spiking fever ≥39 °C; Arthralgia; Transient erythema |
We used Fisher’s exact test and Mann–Whitney U test to analyze categorical and continuous data, respectively. Multivariate logistic regression models were developed to assess the factors affecting patients’ acceptance of switching to BS. According to the data from previous studies,5,6) age, sex, and acceptance of switching to generics were included in multivariate logistic regression. In addition, we included the history of adverse reactions to biologics, history of complaints related to disease activity, use of public payment systems of medical expense support programs beyond the universal health care system, and use of self-injectable biologics in multivariate logistic regression based on our clinical experience. The patients were categorized by age as < 75 or ≥75 years owing to the patient co-payment system in Japan (<75 years old: 20–30% co-payment, ≥75 years old: 10% co-payment). Furthermore, to evaluate the effects of a disease, we stratified the patients into patients with malignant lymphoma (cancer group) and others (non-cancer group). To eliminate bias in the medication history of biologic use, we performed a subgroup analysis of the prevalent users of bio-originators.
All statistical analyses were performed using EZR software (version 1.53; Saitama Medical Center, Jichi Medical University, Saitama, Japan), a modified version of R commander designed to add statistical functions frequently used in biostatistics.9) The significance level was set at p < 0.05.
In total, 210 patients were included in this study. After excluding six patients (two were not interviewed by the pharmacists and four did not understand the pharmacists’ explanation of BS), data from 204 patients were analyzed (Fig. 1). There were no missing data from the included patients.
Table 2 shows the patients’ characteristics. In total, 74 (36.3%) patients had malignant lymphoma (cancer group), and 130 (63.7%) had one of the following diseases: IBD, rheumatoid arthritis (RA), other rheumatic diseases, and osteoporosis (non-cancer group). The cancer group showed better acceptance of BS than the non-cancer group.
| Acceptance group (n = 138) | Refusal group (n = 66) | p | |
|---|---|---|---|
| Age, years, median [range] | 69 [57–77] | 58 [46–71] | 0.001** |
| ≥75 years old (10% co-payment) | 40 (29.0%) | 11 (16.7%) | 0.060 |
| Sex, male | 59 (42.8%) | 26 (39.4%) | 0.762 |
| Living with family, yes | 126 (91.3%) | 54 (81.8%) | 0.063 |
| Diseasesa) | |||
| Malignant lymphoma | 67 (48.6%) | 7 (10.6%) | <0.001*** |
| Rheumatoid arthritis | 36 (26.1%) | 29 (43.9%) | 0.016* |
| Inflammatory bowel disease | 21 (15.2%) | 22 (33.3%) | 0.005** |
| Other rheumatic diseasesb) | 17 (12.3%) | 17 (25.8%) | 0.026* |
| Osteoporosis | 7 (5.1%) | 3 (4.5%) | 1.000 |
| Duration of the disease (years), median [range]c) | 6 [3–11] | 10 [4–15] | 0.015* |
| History of complaint(s) related to disease activity, yes | 68 (49.3%) | 56 (84.8%) | <0.001*** |
| Acceptance of switching to generics, yes | 132 (95.7%) | 44 (66.7%) | <0.001*** |
| Knowledge of biosimilars, yes | 2 (1.4%) | 2 (3.0%) | 0.596 |
| Medication history of biologic use | |||
| Number of biologicsd) | |||
| 0 | 46 (33.3%) | 4 (6.1%) | <0.001*** |
| 1 | 69 (50.0%) | 40 (60.6%) | |
| 2 | 18 (13.0%) | 15 (22.7%) | |
| 3 | 5 (3.6%) | 5 (7.6%) | |
| 4 | 0 (0.0%) | 2 (3.0%) | |
| Duration of biologic use (years), median [range] | 0 [0–3] | 3 [0.3–5.8] | <0.001*** |
| Route of biologic administration | |||
| Intravenous infusion | 92 (66.7%) | 35 (53.0%) | 0.066 |
| Subcutaneous injection (not self-injection) | 18 (13.0%) | 5 (7.6%) | 0.345 |
| Subcutaneous injection (self-injection) | 29 (21.0%) | 27 (40.9%) | 0.004** |
| History of adverse reactions to the drug, yes | 49 (35.5%) | 30 (45.5%) | 0.219 |
| History of adverse reactions to biologics, yes | 12 (8.7%) | 22 (33.3%) | <0.001*** |
| Use of public payment systems for medical expenses support programs beyond the universal health care system | 54 (39.1%) | 37 (56.1%) | 0.025* |
| High-Cost Medical Treatment System | 19 (13.8%) | 2 (3.0%) | 0.024* |
| Treatment and Research Enterprise for Specified Diseases | 35 (25.4%) | 35 (53.0%) | <0.001*** |
*: p < 0.05; **: p < 0.01; ***: p < 0.001. a) Non-cancer group: rheumatoid arthritis (RA), inflammatory bowel disease (IBD), other rheumatic diseases, osteoporosis. Some patients in the non-cancer group had multiple diseases. b) Other rheumatic diseases: Behçet’s disease, systemic lupus erythematosus, psoriatic arthritis, ankylosing spondylitis, palmoplanter pustulosis, adult-onset Still’s disease, relapsing polychondritis, cardiac sarcoidosis, dermatomyositis, and granulomatosis with polyangiitis. c) Duration of the disease: data were collected in the non-cancer group. In the cancer group, the duration of the disease was analyzed as missing values. d) Number of biologics: number of biologics the patients used when the pharmacists interviewed them. Biologics included abatacept, adalimumab, belimumab, certolizumab pegol, denosumab, epoetin beta pegol, etanercept, golimumab, infliximab, rituximab, sarilumab, teriparatide, tocilizumab, ustekinumab, and vedolizumab.
Table 3 shows the results of the multivariate logistic regression analyses. Patients who developed adverse reactions to biologics were more reluctant to switch to BS than patients who had not suffered any adverse effects (odds ratio [95% confidence interval (CI)] = 3.38 [1.35–8.44]). Additionally, patients who had complaints related to disease activity were less willing to switch to BS than those who had no complaints (3.57 [1.53–8.32]). Similarly, patients who refused generic drugs were more reluctant to switch than those who accepted generic drugs (7.62 [2.70–21.60]).
| All | Cancer group | Non-cancer group | ||||
|---|---|---|---|---|---|---|
| Univariatea) | Multivariateb) | Univariatea) | Multivariateb) | Univariatea) | Multivariateb) | |
| Age, <75 years | 2.04 [0.97–4.30] | 0.81 [0.32–2.07] | 0.48 [0.10–2.30] | 0.334 [0.042–2.67] | 1.82 [0.68–4.87] | 0.884 [0.27–2.86] |
| Sex, male | 0.87 [0.48–1.58] | 1.24 [0.60–2.57] | 2.43 [0.44–13.40] | 1.24 [0.18–8.53] | 1.02 [0.49–2.09] | 1.19 [0.51–2.78] |
| History of adverse reactions to biologics, yes | 5.25 [2.40–11.50] | 3.38 [1.35–8.44] | 3.56 [0.32–39.70] | 7.05 [0.49–101] | 3.81 [1.58–9.17] | 3.09 [1.14–8.40] |
| Acceptance of switching to generics, no | 11.00 [4.19–28.90] | 7.62 [2.70–21.60] | 26.4 [2.03–344] | 36.7 [1.76–763] | 6.77 [2.35–19.50] | 5.00 [1.66–15.10] |
| Use of public payment systems of medical expenses support programs beyond the universal health care system, yes | 1.98 [1.10–3.60] | 1.43 [0.67–3.05] | 0.42 [0.05–3.73] | 0.87 [0.063–11.9] | 1.61 [0.80–3.24] | 1.52 [0.63–3.64] |
| History of a complaint related to disease activity, yes | 5.76 [2.72–12.20] | 3.57 [1.53–8.32] | — | —c) | 4.58 [1.25–16.80] | 3.52 [0.89–13.90] |
| Use of self-injectable biologics | 2.60 [1.37–4.93] | 1.43 [0.66–3.12] | — | —d) | 1.22 [0.61–2.45] | 1.08 [0.49–2.37] |
*: p < 0.05; **: p < 0.01; ***: p < 0.001. a) Univariate analysis: crude odds ratio of univariate logistic regression. b) Multivariate analysis: adjusted odds ratio of multivariate logistic regression. Adjusted covariates were age (<75 years), sex, history of adverse reactions to biologics, acceptance of switching to generics, use of public payment systems of medical expenses support programs beyond the universal health care system, history of a complaint related to disease activity, and use of self-injectable biologics. c) All patients who had experienced a complaint related to disease activity accepted the switch to BS in the cancer group. d) Bio-originator therapy was rituximab; therefore, there were no self-injectable biologics in the cancer group.
The patient characteristics are provided in Supplementary Data 1. Patients who refused generic drugs were more likely to switch than patients who accepted generic drugs (36.7 [1.76–763]).
Subgroup Analysis: Non-cancer GroupThe detailed patient characteristics are provided in Supplementary Data 2. In the non-cancer group, multivariate logistic regression showed that patients who had suffered adverse reactions to biologics were more reluctant to switch to BS than those who did not (odds ratio: 3.09 [1.14–8.40]). Additionally, patients who refused generic drugs were more likely to switch to BS than those who accepted generic drugs (odds ratio: 5.00 [1.66–15.10]).
Subgroup Analysis: Prevalent Users of Bio-OriginatorsMultivariate logistic regression analyses were performed on the prevalent users of bio-originators (Supplementary Data 3). When the patients’ group was restricted to the prevalent users of bio-originators, a history of adverse reactions to biologics (odds ratio: 3.12 [1.25–7.78]), history of complaints related to disease activity (odds ratio: 3.68 [1.33–10.20]), and unacceptability of generic drugs (odds ratio: 4.78 [1.69–13.50]) remained associated with acceptance of the switch.
Reasons for Patients’ Refusal to Switch to BSAmong the 204 patients, 66 (32.4%) refused to switch to BS (Fig. 1). The most common reason for this trend was dependability (45.5%), followed by effectiveness (32.4%), safety (25.8%), economic sufficiency (21.2%), and changes in self-injection devices (12.1%) (Fig. 2). Moreover, some patients refused the switch because of disease remission.
Multiple answers were allowed. BS, Biosimilars. The horizontal axis shows the number of patients.
In this study, we investigated whether the factors affecting patients’ acceptance of switching to BS differed depending on the disease in Japanese patients with malignant lymphoma, IBD, rheumatic disease, and osteoporosis. Our findings showed that a history of adverse reactions to biologics, the unacceptability of generic drugs, and a history of complaints related to disease activity were factors associated with the refusal to switch to BS. These associations did not change even for prevalent users. The subgroup analyses suggested that the unacceptability of generic drugs is a common factor regardless of the disease. In contrast, a history of adverse reactions to biologics was associated with the refusal to switch to BS in the non-cancer group.
Previous studies did not include information regarding the history of adverse reactions to biologics.5,6,10) We found that a history of adverse reactions to biologics was related to patients’ acceptance of switching to BS.
The effect of a history of complaints related to disease activity on patients’ acceptance of switching to BS differed between cancer and non-cancer groups (Supplementary Data 1, 2). Patients in the non-cancer (IBD and RA) groups with complaints may be reluctant to switch medications because the medicine can reduce complaints. The alleviation of complaints directly corresponds to the remission or cure of the disease. However, complaints may not be associated with switching medicine in the cancer group. Fewer complaints do not necessarily mean remission or cure of the disease, such as malignant lymphoma, where disease severity is measured by blood profiling and not physical symptoms.
Our findings that patients refusing generic drugs were reluctant to switch to BS are consistent with the findings of previous studies,6,10) regardless of the disease. In addition, previous studies have shown that good knowledge about generic drugs and the experience of a previous generic drug substitution is associated with a willingness for generic substitution.11,12) Therefore, sufficient knowledge about BS may influence switching to BS similarly.
In this study, most patients did not know about BS, and the reason that the patients refused BS was dependability. First, a pharmaceutical company or related organization should raise awareness of BS and its reliability. Second, physicians and pharmacists should provide sufficient information on BS to patients who doubt the efficacy and safety of BS to help them understand BS, considering that the history of complaints related to disease activity differs depending on the disease. Finally, if needed, physicians and pharmacists should take precautions against the occurrence of any adverse reactions to biologics. These efforts may promote patients’ switching to BS, thereby minimizing nocebo effects. We believe this method may be effective because some prevalent users of bio-originators have managed current treatments by controlling complaints or adverse reactions.
This study had some limitations. First, this was a single-center study; in particular, many patients were geriatric individuals in our hospital and, thus, further research targeting younger patients is needed in other facilities. Second, the cancer group included patients with only malignant lymphoma, not solid tumors. However, patients with solid tumors or diabetes also often use biologics; therefore, additional research is needed to determine whether this study’s findings can be applied to individuals with other diseases. Third, the subgroup analysis involved a small number of patients. Therefore, prospective studies are required to confirm the results of this study.
Factors affecting patients’ acceptance of switching to BS included common factors regardless of the disease, such as the unacceptability of generic drugs, and different factors depending on the disease, such as the history of complaints related to disease activity.
We are grateful to S. Tanaka and S. Yuyama for interviewing patients.
The authors declare no conflict of interest.
This article contains supplementary materials.
