Clin Rheumatol DOI 10.1007/s10067-013-2482-1
ORIGINAL ARTICLE
Prescription patterns and trends in anti-rheumatic drug use based on a large-scale claims database in Japan Hirotaka Katada & Naoichiro Yukawa & Hisashi Urushihara & Shiro Tanaka & Tsuneyo Mimori & Koji Kawakami
Received: 30 August 2013 / Revised: 25 December 2013 / Accepted: 27 December 2013 # Clinical Rheumatology 2014
Abstract This drug utilization study aimed to investigate prescription patterns and trends for anti-rheumatic drug use in Japanese patients with rheumatoid arthritis (RA), clarifying if patients with RA in Japan are being treated according to EULAR recommendations and ACR guidelines. We used a large-scale claims database consisting of the medical claims of employee health insurance recipients, which included approximately one million insured people. The claims data for incident 5,126 patients with diagnosis codes of RA between January 1, 2005 and October 31, 2011 were analyzed. The number of patients who received disease modifying antirheumatic drugs (DMARDs) including biologics as initial therapy was 629 (12.3 %), while the others received non-DMARD therapy only. During the study period, use of methotrexate (MTX) and biologics as first-line drugs increased from 1.9 to 8.0 % and from 0 to 1.6 %, respectively (p < 0.001 for both), while that of non-steroidal anti-inflammatory drugs (NSAIDs) H. Katada : H. Urushihara : S. Tanaka : K. Kawakami (*) Department of Pharmacoepidemiology, Graduate School of Medicine and Public Health, Kyoto University, Yoshidakonoe-cho, Sakyo-ku, Kyoto 606-8051, Japan e-mail: [email protected] H. Katada e-mail: [email protected] H. Urushihara e-mail: [email protected] S. Tanaka e-mail: [email protected] N. Yukawa : T. Mimori Department of Rheumatology and Clinical Immunology, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto, Japan N. Yukawa e-mail: [email protected] T. Mimori e-mail: [email protected]
decreased (p = 0.004). Time from first RA diagnosis to the start of treatment with DMARDs decreased significantly from 2005 to 2010. These findings suggest that many early RA patients in Japan do not yet receive aggressive treatment, albeit that this prescribing practice has gradually changed to better comply with clinical recommendations. The current, obsolete Japanese RA guidelines require urgent updating to reflect the most recent knowledge and care with effective treatment modalities. Keywords Anti-rheumatic drugs . Claims database . Drug utilization study . Prescription patterns . Rheumatoid arthritis
Introduction In the treatment strategy for rheumatoid arthritis (RA), the aim of initial therapy was relief of pain by non-steroidal antiinflammatory drugs (NSAIDs) [1]. However, recent studies have demonstrated that bone and joint destruction progress rapidly and irreversibly within only a few years after onset [2–4]. Therefore, the initiation of more aggressive therapy using disease-modifying anti-rheumatic drugs (DMARDs), including methotrexate (MTX) and biologics, at the early stage of disease development might effectively delay irreversible progression and achieve low disease activity or remission [5, 6]. Recently published or updated clinical guidelines by the European League Against Rheumatism (EULAR) [7, 8] and the American College of Rheumatology (ACR) [9] recommend the prescription of DMARDs as soon as possible after the establishment of a diagnosis of RA. In particular, MTX is highly recommended as an initial “anchor drug” for RA treatment [10]. In Japan, however, the regulatory authority initially approved MTX for RA treatment at up to 8 mg/week, but only for secondary use following the failure of the initial therapy. In Europe and the USA, in contrast, maximum approved dosage
Clin Rheumatol
is 25–30 mg/week, and initial use in RA is both approved and recommended [11–13]. In February 2011, the Japanese authority belatedly approved dose escalation regimens in initial use of up to 16 mg/week [14]. With regard to biologics, six have become available since 2003, namely infliximab, etanercept, tocilizumab, adalimumab, abatacept, and golimumab. The Japan College of Rheumatology (JCR) issued separate guidelines for the individual use of these biologics from 2008 to 2012 [15–17]. Despite ongoing and continually evolving changes, however, the College has not updated its comprehensive guidelines for RA treatment since 2004 [18]. Despite the increasing medical and social importance of drug utilization studies [19, 20], only a few studies have examined the utilization of anti-rheumatic drugs in Japan [21–25] and it remains unknown whether patients with RA in Japan are being treated according to EULAR recommendations and ACR guidelines. We therefore aimed to reveal the prescription patterns and trends for anti-rheumatic drug use in general outpatient settings rather than specialized settings for RA treatment using of a large-scale claims database [26].
Patients, materials, and methods Study design and data source This drug utilization study was based on a large-scale claims database constructed by Japan Medical Data Center (JMDC) Co. Ltd, Tokyo, Japan, which accumulates medical claims for inpatient, outpatient, and pharmacy benefits collected from multiple insurance societies. Total coverage of this database has reached approximately one million insured people since the start of data collection in January 1, 2005. The data contained were encrypted with regard to personal identifier, patient age and sex, medical procedures, diagnostic codes according to the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10), and name, dose, and number of days of supply of prescribed drugs. Claims data between January 1, 2005 and October 31, 2011 were analyzed. Study drugs For the purpose of this study, anti-rheumatic drugs were classified into five drug classes using the drug codes of the Japan Standard Commodity Classification (JSCC) [27]: glucocorticoids (oral or internal glucocorticoids only); NSAIDs; MTX; biologics; and other DMARDs (not including MTX and biologics), with an approved indication for RA treatment in Japan. “Non-DMARD therapy” was defined as the therapy using glucocorticoids or NSAIDs without MTX, biologics and other
DMARDs. ‘DMARD therapy’ was defined as the therapy using MTX, biologics, or other DMARDs. Initial drugs were defined as the initial anti-rheumatic drugs prescribed on or after the date of first RA diagnosis. Secondary drugs were defined as anti-rheumatic drugs prescribed when any change or addition to the anti-rheumatic regimen occurred after the prescription date of initial drugs. Definition of disease The study population was defined as outpatients with an established diagnosis of RA identified according to incident user design within the database, who satisfied all the following inclusion criteria [28]: (1) continuous enrollment in the database for at least 6 months between January 1, 2005 and October 31, 2011; (2) no ICD-10 codes for RA for 6 months from initial enrollment in the database; and (3) prescription for anti-rheumatic drugs after the first date of an ICD-10 code for RA. Diagnosis of RA was established with the first record of the following ICD-10 codes, M05 (rheumatoid arthritis with rheumatoid factor) and M06 (other rheumatoid arthritis), on its earliest recorded date. We excluded the following diagnostic codes because these represent a different clinical entity to RA: rheumatoid myopathy with rheumatoid arthritis (M0530), RS3PE syndrome (M0600), Adult Still’s disease (M0610), inflammatory polyarthropathy (M0640), and arthritis mutilans (M0680). Statistical analysis Annual distribution of the age and sex of patients prescribed initial and secondary anti-rheumatic drugs were summarized. Further, prescription trends by drug class, time, and type of medical facility and clinical department were examined for each initial and secondary drug. One-year survival curves for the time from the first diagnosis of RA to the start of DMARD therapy were estimated for patients diagnosed in each of the years from 2005 to 2010 using the Kaplan–Meier method, and compared with the Tarone–Ware test [29, 30]. For survival analysis, patients who were removed from the database before the start of DMARD therapy were censored. Additionally, the proportions of patients receiving DMARD therapy within the month of the first RA diagnosis and those prescribed biologics as initial drugs were estimated annually. The proportions of patients prescribed infliximab without concomitant MTX in initial regimens to third regimens, which is listed as required in the package insert of infliximab [31], was also estimated annually. Difference between groups was examined using the t-test and χ2 tests. The Cochran–Armitage test was used to examine one-way time trends in the proportions of prescriptions by drug class. For hypothesis testing, a two-tailed significance level of 0.05 was used. Statistical analyses were performed
Clin Rheumatol
with IBM SPSS Statistics Version 20 and R Version 2.15.0 for Windows.
Table 2 Type of medical facility prescribing initial or secondary drugs Initial drugs
Secondary drugs
5,126
1,803
265 (5.2) 342 (6.7)
127 (7.0) 178 (9.9)
3,724 (72.6) 873 (17.0) 14 (0.3)
1,220 (67.7) 381 (21.1) 3 (0.2)
Ethics statement Because the data investigated in this study were de-identified by the database provider, JMDC, before their provision to us in an unlinked manner, the study was exempt from obtaining informed consent from individual patients according to local ethical guidelines for epidemiological research. This study and the waiver of informed consent were approved by Kyoto University Graduate School and Faculty of Medicine, Ethics Committee (E1420).
Number of patients Medical facility Public hospital University hospital Clinic Othera Unknown
Data are presented as numbers with percentages in parentheses Duplicate counts between types of medical facility were allowed a
Hospitals with more than 20 beds which were not public or university hospitals
Results A total of 5,126 patients (male, 1,689 [32.9 %]; female, 3,437 [67.1 %]; Table 1) aged 45.6 ± 12.7 years (mean ± SD; male, 45.1 ± 13.0; female, 45.8 ± 12.6) were eligible for analysis. Average age and sex ratio of RA patients were not significantly different across 2005 to 2011. Of the total, 1,803 (male, 503 [27.9 %]; female, 1,300 [72.1 %]) were prescribed secondary drugs. A majority of the patients was prescribed initial and secondary drugs in clinics (72.6 and 67.7 %, respectively, Table 2). Note that a physician in clinics can be certified as a specialist in rheumatology, since specialists and generalists are not officially differentiated in the health care system in Japan. Non-DMARD therapy accounted for 87.7 % of initial and 68.6 % of secondary prescriptions (Figs. 1 and 2). Of 4,497 patients prescribed glucocorticoids or NSAIDs as initial drugs, 1,520 (33.8 %) had a prescription record of secondary drugs. The distribution of drug classes of secondary drugs in these 1,520 patients was
Table 1 Characteristics of Japanese patients with RA treated with antirheumatic drugs Year
Male
Female na (%)
na (%)
Age (years)
Total
1,689 (32.9)
45.1±13.0
3,437 (67.1)
45.8±12.6
2005 2006 2007 2008 2009 2010 2011
78 (35.3) 154 (38.0) 143 (33.6) 240 (33.5) 313 (31.1) 382 (31.4) 379 (33.5)
44.8±11.8 45.9±12.3 44.1±12.8 44.9±14.5 44.7±12.4 46.0±13.2 44.6±12.8
143 (64.7) 251 (62.0) 283 (66.4) 476 (66.5) 695 (68.9) 836 (68.6) 753 (66.5)
46.2±12.8 48.1±13.1 46.2±13.3 44.4±13.0 45.6±12.5 45.3±12.1 46.3±12.5
Age (years)
Data are presented as means±SD or numbers with percentages in parentheses a
Number of patients diagnosed with RA for the first time in each year
almost the same as that for the initial therapy, namely combination of glucocorticoids and NSAIDs, 521 (34.3 %); glucocorticoids only, 435 (28.6 %); NSAIDs only, 231 (15.2 %); other DMARDs only, 193 (12.7 %); MTX only, 123 (8.1 %); combination of MTX and other DMARDs, 11 (0.7 %); and biologics, 6 (0.4 %). NSAIDs were prescribed most frequently as initial drugs, while the prescription ratios of MTX, other DMARDs, and biologics were less than 10 % in all clinical departments (Fig. 3) and this result did not change in secondary drugs (data not shown). Prescribing pattern was similar across the different types of medical facility (data not shown). Changes in the proportions of patients prescribed glucocorticoids and patients prescribed other DMARDs as initial drugs were insignificant during the course of study period from 2005 to 2011 (p = 0.35, p = 0.28, respectively, Cochran–Armitage test, Fig. 4). However, the proportion of patients prescribed NSAIDs decreased significantly (p = 0.004), while significant increases were observed for MTX (1.9 to 8.0 %, p < 0.001) and biologics (0.0 to 1.8 %, p < 0.001). Time from first RA diagnosis to the start of DMARD therapy decreased significantly from 2005 to 2010 (p < 0.001, Tarone–Ware test, Fig. 5), while the annual proportion of patients receiving DMARD therapy as the initial therapy within the month of RA diagnosis increased significantly (p < 0.001, Cochran–Armitage test): 5.4 % (12/221) in 2005, 6.9 % (28/ 405) in 2006, 4.9 % (21/426) in 2007, 9.1 % (65/716) in 2008, 11.4 % (115/1,008) in 2009, 9.6 % (117/1,218) in 2010, and 13.8 % (156/1,132) in 2011. The total number of patients prescribed biologics as the initial drugs was 50. This is shown as 1.0 % in Fig. 1. Of the 47 patients treated with infliximab in the initial to third regimens, 13 (28 %) were not prescribed concomitant MTX (one in 2007, two in 2009, four in 2010, and six in 2011).
Clin Rheumatol Fig. 1 Distribution of initial drugs used for the treatment of Japanese RA patients. Dagger, includes concomitant MTX or other DMARDs
3,000
58.4%
2,500
Patients (n)
2,000
1,500
1,000
16.8% 12.5%
500 6.5% 3.9% 0.9%
1.0%
combination of MTX and other DMARDs
biologics†
0 NSAIDs only
combination of glucocorticoids other DMARDs only only glucocorticoids and NSAIDs
Discussion NSAIDs dominated all prescriptions of anti-rheumatic drugs in both the initial and secondary regimens regardless of the type of clinical department or medical facility. Common prescribing practice among non-specialized settings in Japan
MTX only
such as general practitioners, where most of our data were derived, appears to remain the initiation of non-DMARD therapy as the initial therapy. A number of studies have reported that NSAIDs or glucocorticoids mono-therapy cannot delay or prevent the bone and joint destruction of RA or improve bodily function [32–35]. According to EULAR
600
Fig. 2 Distribution of secondary drugs for the treatment of Japanese RA patients. Dagger, includes concomitant MTX or other DMARDs
29.8% 500 24.7%
Patients (n)
400
300 14.1%
14.6%
10.3%
200
100
3.9% 2.6%
0 combination of glucocorticoids NSAIDs only other DMARDs glucocorticoids only only and NSAIDs
MTX only
combination of MTX and other DMARDs
biologics†
Clin Rheumatol 100
Fig. 3 Initial drugs used for RA treatment by clinical department
glucocorticoids NSAIDs
90
other DMARDs MTX
80
biologics
Patients (%)
70
60 50 40 30 20 10 0 Surgical department
(n = 2,281)
(n = 235)
recommendations 2013, though glucocorticoids at low dose used in combination with DMARDs can be considered as the initial therapy for up to 6 months, glucocorticoids monotherapy should not be used unless all DMARDs have contraindications because of safety issues [8]. Many large clinical trials have recently demonstrated that earlier aggressive treatment with agents having disease-modifying effects, such as MTX or other DMARDs, provides better outcomes, such as low disease activity or remission [5, 6]. These findings indicate that not only RA specialists but also general practitioners should prescribe MTX or other DMARDs as initial or secondary drugs as soon as possible after the establishment of a diagnosis of RA. In recent years, it was demonstrated that higher doses of MTX were more effective for RA treatment [36]. The Ministry of Health, Labour and Welfare’s approval of up to 16 mg MTX in February, 2011 indicates that the use 100
Proportion of each initial drug (%)
Fig. 4 Time trend of initial drugs used for treating Japanese RA patients by drug class. Parentheses indicate the total number of patients treated with initial drugs per year. *p = 0.004, **p < 0.001, Cochran–Armitage test. A, 32.9 % in 2005 to 34.5 % in 2011 (p = 0.35). C, 7.5 % in 2005 to 7.7 % in 2011 (p = 0.28)
Internal department
Orthopedic department
Other (n = 729)
(n = 2,028)
of high-dose MTX for RA is set to increase, thereby an increased treatment efficiency is expected. Recommendations for the management of RA published by EULAR in 2010 and 2013 stated that DMARDs therapy should be started as soon as an RA diagnosis is made even in primary care settings, and that MTX should be an integral part of the initial therapy strategy in patients with active RA [7, 8]. In contrast, the ACR’s updated 2012 recommendations for DMARDs and biologics were targeted at low disease activity or remission, and provided a practical and detailed algorithm on the use of MTX, biologics, and other DMARDs aimed at helping more doctors start early aggressive treatment after the establishment of an RA diagnosis [9]. These recommendations might lead to a shortening of the time to the start of treatment with MTX and increase the proportion of patients treated with biologics [37, 38]. They also highlight the need
B: 87.0 %
B*: 80.9 %
80 Total N = 5,126 60
A: glucocorticoids B: NSAIDs C: other DMARDs
40
D: MTX 20
E: 0.0 %
E: biologics D: 1.9 %
D**: 8.0 % E**: 1.6 %
0 Year 2005 n (161)
2006
2007
2008
2009
2010
2011
(346)
(400)
(679)
(979)
(1,227)
(1,334)
Clin Rheumatol Fig. 5 Kaplan–Meier curves for the time of first RA diagnosis to the start of treatment with MTX or other DMARDs within 1 year. Parentheses show the numbers of patients per year. Survival time was significantly different from 2005 to 2010 (p
ORIGINAL ARTICLE
Prescription patterns and trends in anti-rheumatic drug use based on a large-scale claims database in Japan Hirotaka Katada & Naoichiro Yukawa & Hisashi Urushihara & Shiro Tanaka & Tsuneyo Mimori & Koji Kawakami
Received: 30 August 2013 / Revised: 25 December 2013 / Accepted: 27 December 2013 # Clinical Rheumatology 2014
Abstract This drug utilization study aimed to investigate prescription patterns and trends for anti-rheumatic drug use in Japanese patients with rheumatoid arthritis (RA), clarifying if patients with RA in Japan are being treated according to EULAR recommendations and ACR guidelines. We used a large-scale claims database consisting of the medical claims of employee health insurance recipients, which included approximately one million insured people. The claims data for incident 5,126 patients with diagnosis codes of RA between January 1, 2005 and October 31, 2011 were analyzed. The number of patients who received disease modifying antirheumatic drugs (DMARDs) including biologics as initial therapy was 629 (12.3 %), while the others received non-DMARD therapy only. During the study period, use of methotrexate (MTX) and biologics as first-line drugs increased from 1.9 to 8.0 % and from 0 to 1.6 %, respectively (p < 0.001 for both), while that of non-steroidal anti-inflammatory drugs (NSAIDs) H. Katada : H. Urushihara : S. Tanaka : K. Kawakami (*) Department of Pharmacoepidemiology, Graduate School of Medicine and Public Health, Kyoto University, Yoshidakonoe-cho, Sakyo-ku, Kyoto 606-8051, Japan e-mail: [email protected] H. Katada e-mail: [email protected] H. Urushihara e-mail: [email protected] S. Tanaka e-mail: [email protected] N. Yukawa : T. Mimori Department of Rheumatology and Clinical Immunology, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto, Japan N. Yukawa e-mail: [email protected] T. Mimori e-mail: [email protected]
decreased (p = 0.004). Time from first RA diagnosis to the start of treatment with DMARDs decreased significantly from 2005 to 2010. These findings suggest that many early RA patients in Japan do not yet receive aggressive treatment, albeit that this prescribing practice has gradually changed to better comply with clinical recommendations. The current, obsolete Japanese RA guidelines require urgent updating to reflect the most recent knowledge and care with effective treatment modalities. Keywords Anti-rheumatic drugs . Claims database . Drug utilization study . Prescription patterns . Rheumatoid arthritis
Introduction In the treatment strategy for rheumatoid arthritis (RA), the aim of initial therapy was relief of pain by non-steroidal antiinflammatory drugs (NSAIDs) [1]. However, recent studies have demonstrated that bone and joint destruction progress rapidly and irreversibly within only a few years after onset [2–4]. Therefore, the initiation of more aggressive therapy using disease-modifying anti-rheumatic drugs (DMARDs), including methotrexate (MTX) and biologics, at the early stage of disease development might effectively delay irreversible progression and achieve low disease activity or remission [5, 6]. Recently published or updated clinical guidelines by the European League Against Rheumatism (EULAR) [7, 8] and the American College of Rheumatology (ACR) [9] recommend the prescription of DMARDs as soon as possible after the establishment of a diagnosis of RA. In particular, MTX is highly recommended as an initial “anchor drug” for RA treatment [10]. In Japan, however, the regulatory authority initially approved MTX for RA treatment at up to 8 mg/week, but only for secondary use following the failure of the initial therapy. In Europe and the USA, in contrast, maximum approved dosage
Clin Rheumatol
is 25–30 mg/week, and initial use in RA is both approved and recommended [11–13]. In February 2011, the Japanese authority belatedly approved dose escalation regimens in initial use of up to 16 mg/week [14]. With regard to biologics, six have become available since 2003, namely infliximab, etanercept, tocilizumab, adalimumab, abatacept, and golimumab. The Japan College of Rheumatology (JCR) issued separate guidelines for the individual use of these biologics from 2008 to 2012 [15–17]. Despite ongoing and continually evolving changes, however, the College has not updated its comprehensive guidelines for RA treatment since 2004 [18]. Despite the increasing medical and social importance of drug utilization studies [19, 20], only a few studies have examined the utilization of anti-rheumatic drugs in Japan [21–25] and it remains unknown whether patients with RA in Japan are being treated according to EULAR recommendations and ACR guidelines. We therefore aimed to reveal the prescription patterns and trends for anti-rheumatic drug use in general outpatient settings rather than specialized settings for RA treatment using of a large-scale claims database [26].
Patients, materials, and methods Study design and data source This drug utilization study was based on a large-scale claims database constructed by Japan Medical Data Center (JMDC) Co. Ltd, Tokyo, Japan, which accumulates medical claims for inpatient, outpatient, and pharmacy benefits collected from multiple insurance societies. Total coverage of this database has reached approximately one million insured people since the start of data collection in January 1, 2005. The data contained were encrypted with regard to personal identifier, patient age and sex, medical procedures, diagnostic codes according to the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10), and name, dose, and number of days of supply of prescribed drugs. Claims data between January 1, 2005 and October 31, 2011 were analyzed. Study drugs For the purpose of this study, anti-rheumatic drugs were classified into five drug classes using the drug codes of the Japan Standard Commodity Classification (JSCC) [27]: glucocorticoids (oral or internal glucocorticoids only); NSAIDs; MTX; biologics; and other DMARDs (not including MTX and biologics), with an approved indication for RA treatment in Japan. “Non-DMARD therapy” was defined as the therapy using glucocorticoids or NSAIDs without MTX, biologics and other
DMARDs. ‘DMARD therapy’ was defined as the therapy using MTX, biologics, or other DMARDs. Initial drugs were defined as the initial anti-rheumatic drugs prescribed on or after the date of first RA diagnosis. Secondary drugs were defined as anti-rheumatic drugs prescribed when any change or addition to the anti-rheumatic regimen occurred after the prescription date of initial drugs. Definition of disease The study population was defined as outpatients with an established diagnosis of RA identified according to incident user design within the database, who satisfied all the following inclusion criteria [28]: (1) continuous enrollment in the database for at least 6 months between January 1, 2005 and October 31, 2011; (2) no ICD-10 codes for RA for 6 months from initial enrollment in the database; and (3) prescription for anti-rheumatic drugs after the first date of an ICD-10 code for RA. Diagnosis of RA was established with the first record of the following ICD-10 codes, M05 (rheumatoid arthritis with rheumatoid factor) and M06 (other rheumatoid arthritis), on its earliest recorded date. We excluded the following diagnostic codes because these represent a different clinical entity to RA: rheumatoid myopathy with rheumatoid arthritis (M0530), RS3PE syndrome (M0600), Adult Still’s disease (M0610), inflammatory polyarthropathy (M0640), and arthritis mutilans (M0680). Statistical analysis Annual distribution of the age and sex of patients prescribed initial and secondary anti-rheumatic drugs were summarized. Further, prescription trends by drug class, time, and type of medical facility and clinical department were examined for each initial and secondary drug. One-year survival curves for the time from the first diagnosis of RA to the start of DMARD therapy were estimated for patients diagnosed in each of the years from 2005 to 2010 using the Kaplan–Meier method, and compared with the Tarone–Ware test [29, 30]. For survival analysis, patients who were removed from the database before the start of DMARD therapy were censored. Additionally, the proportions of patients receiving DMARD therapy within the month of the first RA diagnosis and those prescribed biologics as initial drugs were estimated annually. The proportions of patients prescribed infliximab without concomitant MTX in initial regimens to third regimens, which is listed as required in the package insert of infliximab [31], was also estimated annually. Difference between groups was examined using the t-test and χ2 tests. The Cochran–Armitage test was used to examine one-way time trends in the proportions of prescriptions by drug class. For hypothesis testing, a two-tailed significance level of 0.05 was used. Statistical analyses were performed
Clin Rheumatol
with IBM SPSS Statistics Version 20 and R Version 2.15.0 for Windows.
Table 2 Type of medical facility prescribing initial or secondary drugs Initial drugs
Secondary drugs
5,126
1,803
265 (5.2) 342 (6.7)
127 (7.0) 178 (9.9)
3,724 (72.6) 873 (17.0) 14 (0.3)
1,220 (67.7) 381 (21.1) 3 (0.2)
Ethics statement Because the data investigated in this study were de-identified by the database provider, JMDC, before their provision to us in an unlinked manner, the study was exempt from obtaining informed consent from individual patients according to local ethical guidelines for epidemiological research. This study and the waiver of informed consent were approved by Kyoto University Graduate School and Faculty of Medicine, Ethics Committee (E1420).
Number of patients Medical facility Public hospital University hospital Clinic Othera Unknown
Data are presented as numbers with percentages in parentheses Duplicate counts between types of medical facility were allowed a
Hospitals with more than 20 beds which were not public or university hospitals
Results A total of 5,126 patients (male, 1,689 [32.9 %]; female, 3,437 [67.1 %]; Table 1) aged 45.6 ± 12.7 years (mean ± SD; male, 45.1 ± 13.0; female, 45.8 ± 12.6) were eligible for analysis. Average age and sex ratio of RA patients were not significantly different across 2005 to 2011. Of the total, 1,803 (male, 503 [27.9 %]; female, 1,300 [72.1 %]) were prescribed secondary drugs. A majority of the patients was prescribed initial and secondary drugs in clinics (72.6 and 67.7 %, respectively, Table 2). Note that a physician in clinics can be certified as a specialist in rheumatology, since specialists and generalists are not officially differentiated in the health care system in Japan. Non-DMARD therapy accounted for 87.7 % of initial and 68.6 % of secondary prescriptions (Figs. 1 and 2). Of 4,497 patients prescribed glucocorticoids or NSAIDs as initial drugs, 1,520 (33.8 %) had a prescription record of secondary drugs. The distribution of drug classes of secondary drugs in these 1,520 patients was
Table 1 Characteristics of Japanese patients with RA treated with antirheumatic drugs Year
Male
Female na (%)
na (%)
Age (years)
Total
1,689 (32.9)
45.1±13.0
3,437 (67.1)
45.8±12.6
2005 2006 2007 2008 2009 2010 2011
78 (35.3) 154 (38.0) 143 (33.6) 240 (33.5) 313 (31.1) 382 (31.4) 379 (33.5)
44.8±11.8 45.9±12.3 44.1±12.8 44.9±14.5 44.7±12.4 46.0±13.2 44.6±12.8
143 (64.7) 251 (62.0) 283 (66.4) 476 (66.5) 695 (68.9) 836 (68.6) 753 (66.5)
46.2±12.8 48.1±13.1 46.2±13.3 44.4±13.0 45.6±12.5 45.3±12.1 46.3±12.5
Age (years)
Data are presented as means±SD or numbers with percentages in parentheses a
Number of patients diagnosed with RA for the first time in each year
almost the same as that for the initial therapy, namely combination of glucocorticoids and NSAIDs, 521 (34.3 %); glucocorticoids only, 435 (28.6 %); NSAIDs only, 231 (15.2 %); other DMARDs only, 193 (12.7 %); MTX only, 123 (8.1 %); combination of MTX and other DMARDs, 11 (0.7 %); and biologics, 6 (0.4 %). NSAIDs were prescribed most frequently as initial drugs, while the prescription ratios of MTX, other DMARDs, and biologics were less than 10 % in all clinical departments (Fig. 3) and this result did not change in secondary drugs (data not shown). Prescribing pattern was similar across the different types of medical facility (data not shown). Changes in the proportions of patients prescribed glucocorticoids and patients prescribed other DMARDs as initial drugs were insignificant during the course of study period from 2005 to 2011 (p = 0.35, p = 0.28, respectively, Cochran–Armitage test, Fig. 4). However, the proportion of patients prescribed NSAIDs decreased significantly (p = 0.004), while significant increases were observed for MTX (1.9 to 8.0 %, p < 0.001) and biologics (0.0 to 1.8 %, p < 0.001). Time from first RA diagnosis to the start of DMARD therapy decreased significantly from 2005 to 2010 (p < 0.001, Tarone–Ware test, Fig. 5), while the annual proportion of patients receiving DMARD therapy as the initial therapy within the month of RA diagnosis increased significantly (p < 0.001, Cochran–Armitage test): 5.4 % (12/221) in 2005, 6.9 % (28/ 405) in 2006, 4.9 % (21/426) in 2007, 9.1 % (65/716) in 2008, 11.4 % (115/1,008) in 2009, 9.6 % (117/1,218) in 2010, and 13.8 % (156/1,132) in 2011. The total number of patients prescribed biologics as the initial drugs was 50. This is shown as 1.0 % in Fig. 1. Of the 47 patients treated with infliximab in the initial to third regimens, 13 (28 %) were not prescribed concomitant MTX (one in 2007, two in 2009, four in 2010, and six in 2011).
Clin Rheumatol Fig. 1 Distribution of initial drugs used for the treatment of Japanese RA patients. Dagger, includes concomitant MTX or other DMARDs
3,000
58.4%
2,500
Patients (n)
2,000
1,500
1,000
16.8% 12.5%
500 6.5% 3.9% 0.9%
1.0%
combination of MTX and other DMARDs
biologics†
0 NSAIDs only
combination of glucocorticoids other DMARDs only only glucocorticoids and NSAIDs
Discussion NSAIDs dominated all prescriptions of anti-rheumatic drugs in both the initial and secondary regimens regardless of the type of clinical department or medical facility. Common prescribing practice among non-specialized settings in Japan
MTX only
such as general practitioners, where most of our data were derived, appears to remain the initiation of non-DMARD therapy as the initial therapy. A number of studies have reported that NSAIDs or glucocorticoids mono-therapy cannot delay or prevent the bone and joint destruction of RA or improve bodily function [32–35]. According to EULAR
600
Fig. 2 Distribution of secondary drugs for the treatment of Japanese RA patients. Dagger, includes concomitant MTX or other DMARDs
29.8% 500 24.7%
Patients (n)
400
300 14.1%
14.6%
10.3%
200
100
3.9% 2.6%
0 combination of glucocorticoids NSAIDs only other DMARDs glucocorticoids only only and NSAIDs
MTX only
combination of MTX and other DMARDs
biologics†
Clin Rheumatol 100
Fig. 3 Initial drugs used for RA treatment by clinical department
glucocorticoids NSAIDs
90
other DMARDs MTX
80
biologics
Patients (%)
70
60 50 40 30 20 10 0 Surgical department
(n = 2,281)
(n = 235)
recommendations 2013, though glucocorticoids at low dose used in combination with DMARDs can be considered as the initial therapy for up to 6 months, glucocorticoids monotherapy should not be used unless all DMARDs have contraindications because of safety issues [8]. Many large clinical trials have recently demonstrated that earlier aggressive treatment with agents having disease-modifying effects, such as MTX or other DMARDs, provides better outcomes, such as low disease activity or remission [5, 6]. These findings indicate that not only RA specialists but also general practitioners should prescribe MTX or other DMARDs as initial or secondary drugs as soon as possible after the establishment of a diagnosis of RA. In recent years, it was demonstrated that higher doses of MTX were more effective for RA treatment [36]. The Ministry of Health, Labour and Welfare’s approval of up to 16 mg MTX in February, 2011 indicates that the use 100
Proportion of each initial drug (%)
Fig. 4 Time trend of initial drugs used for treating Japanese RA patients by drug class. Parentheses indicate the total number of patients treated with initial drugs per year. *p = 0.004, **p < 0.001, Cochran–Armitage test. A, 32.9 % in 2005 to 34.5 % in 2011 (p = 0.35). C, 7.5 % in 2005 to 7.7 % in 2011 (p = 0.28)
Internal department
Orthopedic department
Other (n = 729)
(n = 2,028)
of high-dose MTX for RA is set to increase, thereby an increased treatment efficiency is expected. Recommendations for the management of RA published by EULAR in 2010 and 2013 stated that DMARDs therapy should be started as soon as an RA diagnosis is made even in primary care settings, and that MTX should be an integral part of the initial therapy strategy in patients with active RA [7, 8]. In contrast, the ACR’s updated 2012 recommendations for DMARDs and biologics were targeted at low disease activity or remission, and provided a practical and detailed algorithm on the use of MTX, biologics, and other DMARDs aimed at helping more doctors start early aggressive treatment after the establishment of an RA diagnosis [9]. These recommendations might lead to a shortening of the time to the start of treatment with MTX and increase the proportion of patients treated with biologics [37, 38]. They also highlight the need
B: 87.0 %
B*: 80.9 %
80 Total N = 5,126 60
A: glucocorticoids B: NSAIDs C: other DMARDs
40
D: MTX 20
E: 0.0 %
E: biologics D: 1.9 %
D**: 8.0 % E**: 1.6 %
0 Year 2005 n (161)
2006
2007
2008
2009
2010
2011
(346)
(400)
(679)
(979)
(1,227)
(1,334)
Clin Rheumatol Fig. 5 Kaplan–Meier curves for the time of first RA diagnosis to the start of treatment with MTX or other DMARDs within 1 year. Parentheses show the numbers of patients per year. Survival time was significantly different from 2005 to 2010 (p
