Abstract
Objective: To assess financial conflicts of interest (COI) and nonfinancial COI among psoriatic arthritis (PsA) clinical practice guideline (CPG) authors in Japan and the US, and to evaluate the quality of evidence and strength of recommendations of PsA CPGs. Methods: We performed a retrospective analysis using payment data from major Japanese pharmaceutical companies and the US Open Payments Database from 2016 to 2018. All authors of PsA CPGs issued by the Japanese Dermatological Association (JDA) and American College of Rheumatology (ACR) were included. Results: Of 23 CPG authors in Japan, 21 (91.3%) received at least 1 payment, with a combined total of $3,335,413 between 2016 and 2018. Regarding 25 US authors, 21 (84.0%) received at least 1 payment, with a combined total of $4,081,629 during the same period. The 3-year combined mean ± SD payment per author was $145,018 ± $114,302 in Japan and $162,825 ± $259,670 in the US. A total of 18 authors (78.3%) of the JDA PsA CPG and 12 authors (48.0%) of the ACR PsA CPG had undisclosed financial COI worth $474,663 and $218,501, respectively. The percentage of citations with at least 1 CPG author relative to total citations was 3.4% in Japan and 33.6% in the US. In sum, 71.4% and 88.8% of recommendations for PsA in the JDA and ACR were supported by low or very low quality of evidence. Conclusion: More rigorous cross-checking of information disclosed by pharmaceutical companies and self-reported by physicians and more stringent and transparent COI policies are necessary.
| Original language | English |
|---|---|
| Pages (from-to) | 1278-1286 |
| Number of pages | 9 |
| Journal | Arthritis Care and Research |
| Volume | 75 |
| Issue number | 6 |
| DOIs | |
| State | Published - Jun 2023 |
Fingerprint
Dive into the research topics of 'Evaluation of Financial and Nonfinancial Conflicts of Interest and Quality of Evidence Underlying Psoriatic Arthritis Clinical Practice Guidelines: Analysis of Personal Payments From Pharmaceutical Companies and Authors’ Self-Citation Rate in Japan and the United States'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver
}
In: Arthritis Care and Research, Vol. 75, No. 6, 06.2023, p. 1278-1286.
Research output: Contribution to journal › Article › peer-review
TY - JOUR
T1 - Evaluation of Financial and Nonfinancial Conflicts of Interest and Quality of Evidence Underlying Psoriatic Arthritis Clinical Practice Guidelines
T2 - Analysis of Personal Payments From Pharmaceutical Companies and Authors’ Self-Citation Rate in Japan and the United States
AU - Mamada, Hanano
AU - Murayama, Anju
AU - Kamamoto, Sae
AU - Kaneda, Yudai
AU - Yoshida, Makoto
AU - Sugiura, Sota
AU - Yamashita, Erika
AU - Kusumi, Eiji
AU - Saito, Hiroaki
AU - Sawano, Toyoaki
AU - Tanimoto, Tetsuya
AU - Vassar, Matt
AU - Ozieranski, Piotr
AU - Ozaki, Akihiko
N1 - Funding Information: We reviewed and compared COI policies developed by the JDA (32), ACR (33), US National Academy of Medicine (1), Japanese Association of Medical Sciences (34), and Guidelines International Network (5). The COI policy by the Japanese Association of Medical Sciences (34), National Academy of Medicine (1), and Guidelines International Network (5) were considered as international and domestic standard COI policies as of October 28, 2021. Additionally, descriptive analysis of demographic, self-citation, and payment data was conducted. To evaluate nonfinancial COI, medians and interquartile ranges (IQRs) of self-citations to total citations in the CPG were calculated, following our previous study (10). To assess the accuracy of self-reported financial COI, we compared authors’ self-reported financial COI with database-based COI and calculated the number of cases, payments, and undisclosed authors for each of the disclosed and undisclosed financial COI. As indicated by each CPG, the period of collation was set to 2 years between 2017 and 2018 for the JDA PsA CPG and 1 year between November 30, 2017 and November 29, 2018 for the ACR PsA CPG. Furthermore, since the authors of the JDA PsA CPG stated in the CPG that they were following the COI policy of the Japanese Association of Medical Sciences, payments exceeding 500,000 yen (¥) (US $4,682) per year per company were reported as financial COI (see Supplementary Table 1, available on the Arthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.1002/acr.25032). The calculations were based on payments in Japanese yen for the JDA PsA CPG and in US dollars for the ACR PsA CPG. Japanese yen were converted into US dollars using 2016, 2017, and 2018 average monthly exchange rates of ¥108.8, ¥112.1, and ¥110.4 per $1, respectively. All statistical analyses were performed using Microsoft Excel, version 16.0, and Stata, version 15, by HM and AM. In evaluating the CPG recommendations, the quality of evidence was initially presented on a 3-point scale (A = high, B = low, and C = very low) in the JDA PsA CPG and on a 4-point scale (high, moderate, low, and very low) in the ACR PsA CPG following the Grading of Recommendations Assessment, Development and Evaluation system (see Supplementary Table 2, available on the Arthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.1002/acr.25032) (30,35,36). Both the JDA and ACR PsA CPGs classified strength of recommendation into 2 groups: class 1 (strong) and class 2 (weak or conditional). The Ethics Committee of the Medical Governance Research Institute approved this study. Informed consent from study participants was not required as all data were publicly available. The data underlying this article will be shared on reasonable request to the corresponding author. We conducted a cross-sectional analysis to examine the financial and nonfinancial COI of PsA CPG authors in Japan and the US. All authors of the Japanese PsA CPG published by the Japanese Dermatological Association (JDA) on December 20, 2019 (26) and the PsA CPG published by the American College of Rheumatology (ACR) on November 30, 2018 (27,28) were considered. The JDA had 12,080 members as of March 2019 and is considered the leading medical society for clinical dermatology in Japan. The ACR is the most authoritative rheumatology society in the US, founded in 1934, with more than 7,700 health care professionals, and has developed many CPGs for arthritis and rheumatic diseases. The CPGs issued by the JDA are influential in Japan, and the CPGs issued by the ACR are influential in the US and several other countries. Data of author names, sex, affiliations, positions, individual COI statements, cited publications, quality of evidence, and the strength of recommendations were extracted from the CPGs. The quality of CPGs may be devalued when there are many self-citations by its authors (10), and the self-citation rate has been used as an objective evaluable item for nonfinancial COI in recent years. To assess the extent of nonfinancial COI from self-citation in PsA CPGs from the JDA and ACR, all citations were extracted. Authors’ names were extracted from PubMed and by manual Google searches. To analyze financial COI of all JDA PsA CPG authors, we used payment data published between 2016 and 2018 for all 83 companies belonging to the Japan Pharmaceutical Manufacturers Association. This research focused on personal payments and excluded research payments, since in Japan, the name, institution, and position of the author or researcher who received the research payment are not disclosed, which makes assessing research payments difficult. The period of observation was the fiscal years of 2016 to 2018, with variations among the companies (27), as the JDA PsA CPG was published on December 20, 2019, and the JDA requires CPG authors to report COI for the past 3 years. We extracted data on individual payments for lecturing, writing, and consultancy, as in our previous studies (10,13,28). The extracted data included recipient names, monetary amount, payment category, and pharmaceutical company name. For each person named in the database, we checked to find and remove all duplicates, as described previously (13). Similarly, to evaluate financial COI among the ACR PsA CPG authors, the US Open Payments Database was used. Since only US physicians were covered, ACR PsA CPG authors with an MD and affiliated with US organizations were included in the payment search, as described previously (11,29). The extraction period was set at 3 years from before the first online publication on November 30, 2018 (30). Payment data for all categories of general payments such as speaking, consulting, meals, and travel expenses from pharmaceutical companies between November 30, 2015, and November 29, 2018, were extracted from this database. Personal payments such as lecturing, writing, and consulting were paid directly to CPG authors by pharmaceutical companies, and considering the nature of payments, personal payments are likely to have a greater impact on authors’ work in developing CPGs and making CPG recommendations than research payments. Thus, this study focused on the general payments. Further payments made by medical device companies were excluded from the Open Payments data to ensure consistency with the Japanese data, which only cover pharmaceutical companies (31). We reviewed and compared COI policies developed by the JDA (32), ACR (33), US National Academy of Medicine (1), Japanese Association of Medical Sciences (34), and Guidelines International Network (5). The COI policy by the Japanese Association of Medical Sciences (34), National Academy of Medicine (1), and Guidelines International Network (5) were considered as international and domestic standard COI policies as of October 28, 2021. Additionally, descriptive analysis of demographic, self-citation, and payment data was conducted. To evaluate nonfinancial COI, medians and interquartile ranges (IQRs) of self-citations to total citations in the CPG were calculated, following our previous study (10). To assess the accuracy of self-reported financial COI, we compared authors’ self-reported financial COI with database-based COI and calculated the number of cases, payments, and undisclosed authors for each of the disclosed and undisclosed financial COI. As indicated by each CPG, the period of collation was set to 2 years between 2017 and 2018 for the JDA PsA CPG and 1 year between November 30, 2017 and November 29, 2018 for the ACR PsA CPG. Furthermore, since the authors of the JDA PsA CPG stated in the CPG that they were following the COI policy of the Japanese Association of Medical Sciences, payments exceeding 500,000 yen (¥) (US $4,682) per year per company were reported as financial COI (see Supplementary Table 1, available on the Arthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.1002/acr.25032). The calculations were based on payments in Japanese yen for the JDA PsA CPG and in US dollars for the ACR PsA CPG. Japanese yen were converted into US dollars using 2016, 2017, and 2018 average monthly exchange rates of ¥108.8, ¥112.1, and ¥110.4 per $1, respectively. All statistical analyses were performed using Microsoft Excel, version 16.0, and Stata, version 15, by HM and AM. In evaluating the CPG recommendations, the quality of evidence was initially presented on a 3-point scale (A = high, B = low, and C = very low) in the JDA PsA CPG and on a 4-point scale (high, moderate, low, and very low) in the ACR PsA CPG following the Grading of Recommendations Assessment, Development and Evaluation system (see Supplementary Table 2, available on the Arthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.1002/acr.25032) (30,35,36). Both the JDA and ACR PsA CPGs classified strength of recommendation into 2 groups: class 1 (strong) and class 2 (weak or conditional). The Ethics Committee of the Medical Governance Research Institute approved this study. Informed consent from study participants was not required as all data were publicly available. The data underlying this article will be shared on reasonable request to the corresponding author. We conducted a cross-sectional analysis to examine the financial and nonfinancial COI of PsA CPG authors in Japan and the US. All authors of the Japanese PsA CPG published by the Japanese Dermatological Association (JDA) on December 20, 2019 (26) and the PsA CPG published by the American College of Rheumatology (ACR) on November 30, 2018 (27,28) were considered. The JDA had 12,080 members as of March 2019 and is considered the leading medical society for clinical dermatology in Japan. The ACR is the most authoritative rheumatology society in the US, founded in 1934, with more than 7,700 health care professionals, and has developed many CPGs for arthritis and rheumatic diseases. The CPGs issued by the JDA are influential in Japan, and the CPGs issued by the ACR are influential in the US and several other countries. Data of author names, sex, affiliations, positions, individual COI statements, cited publications, quality of evidence, and the strength of recommendations were extracted from the CPGs. The quality of CPGs may be devalued when there are many self-citations by its authors (10), and the self-citation rate has been used as an objective evaluable item for nonfinancial COI in recent years. To assess the extent of nonfinancial COI from self-citation in PsA CPGs from the JDA and ACR, all citations were extracted. Authors’ names were extracted from PubMed and by manual Google searches. To analyze financial COI of all JDA PsA CPG authors, we used payment data published between 2016 and 2018 for all 83 companies belonging to the Japan Pharmaceutical Manufacturers Association. This research focused on personal payments and excluded research payments, since in Japan, the name, institution, and position of the author or researcher who received the research payment are not disclosed, which makes assessing research payments difficult. The period of observation was the fiscal years of 2016 to 2018, with variations among the companies (27), as the JDA PsA CPG was published on December 20, 2019, and the JDA requires CPG authors to report COI for the past 3 years. We extracted data on individual payments for lecturing, writing, and consultancy, as in our previous studies (10,13,28). The extracted data included recipient names, monetary amount, payment category, and pharmaceutical company name. For each person named in the database, we checked to find and remove all duplicates, as described previously (13). Similarly, to evaluate financial COI among the ACR PsA CPG authors, the US Open Payments Database was used. Since only US physicians were covered, ACR PsA CPG authors with an MD and affiliated with US organizations were included in the payment search, as described previously (11,29). The extraction period was set at 3 years from before the first online publication on November 30, 2018 (30). Payment data for all categories of general payments such as speaking, consulting, meals, and travel expenses from pharmaceutical companies between November 30, 2015, and November 29, 2018, were extracted from this database. Personal payments such as lecturing, writing, and consulting were paid directly to CPG authors by pharmaceutical companies, and considering the nature of payments, personal payments are likely to have a greater impact on authors’ work in developing CPGs and making CPG recommendations than research payments. Thus, this study focused on the general payments. Further payments made by medical device companies were excluded from the Open Payments data to ensure consistency with the Japanese data, which only cover pharmaceutical companies (31). We reviewed and compared COI policies developed by the JDA (32), ACR (33), US National Academy of Medicine (1), Japanese Association of Medical Sciences (34), and Guidelines International Network (5). The COI policy by the Japanese Association of Medical Sciences (34), National Academy of Medicine (1), and Guidelines International Network (5) were considered as international and domestic standard COI policies as of October 28, 2021. Additionally, descriptive analysis of demographic, self-citation, and payment data was conducted. To evaluate nonfinancial COI, medians and interquartile ranges (IQRs) of self-citations to total citations in the CPG were calculated, following our previous study (10). To assess the accuracy of self-reported financial COI, we compared authors’ self-reported financial COI with database-based COI and calculated the number of cases, payments, and undisclosed authors for each of the disclosed and undisclosed financial COI. As indicated by each CPG, the period of collation was set to 2 years between 2017 and 2018 for the JDA PsA CPG and 1 year between November 30, 2017 and November 29, 2018 for the ACR PsA CPG. Furthermore, since the authors of the JDA PsA CPG stated in the CPG that they were following the COI policy of the Japanese Association of Medical Sciences, payments exceeding 500,000 yen (¥) (US $4,682) per year per company were reported as financial COI (see Supplementary Table 1, available on the Arthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.1002/acr.25032). The calculations were based on payments in Japanese yen for the JDA PsA CPG and in US dollars for the ACR PsA CPG. Japanese yen were converted into US dollars using 2016, 2017, and 2018 average monthly exchange rates of ¥108.8, ¥112.1, and ¥110.4 per $1, respectively. All statistical analyses were performed using Microsoft Excel, version 16.0, and Stata, version 15, by HM and AM. In evaluating the CPG recommendations, the quality of evidence was initially presented on a 3-point scale (A = high, B = low, and C = very low) in the JDA PsA CPG and on a 4-point scale (high, moderate, low, and very low) in the ACR PsA CPG following the Grading of Recommendations Assessment, Development and Evaluation system (see Supplementary Table 2, available on the Arthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.1002/acr.25032) (30,35,36). Both the JDA and ACR PsA CPGs classified strength of recommendation into 2 groups: class 1 (strong) and class 2 (weak or conditional). The Ethics Committee of the Medical Governance Research Institute approved this study. Informed consent from study participants was not required as all data were publicly available. The data underlying this article will be shared on reasonable request to the corresponding author. Publisher Copyright: © 2022 American College of Rheumatology.
PY - 2023/6
Y1 - 2023/6
N2 - Objective: To assess financial conflicts of interest (COI) and nonfinancial COI among psoriatic arthritis (PsA) clinical practice guideline (CPG) authors in Japan and the US, and to evaluate the quality of evidence and strength of recommendations of PsA CPGs. Methods: We performed a retrospective analysis using payment data from major Japanese pharmaceutical companies and the US Open Payments Database from 2016 to 2018. All authors of PsA CPGs issued by the Japanese Dermatological Association (JDA) and American College of Rheumatology (ACR) were included. Results: Of 23 CPG authors in Japan, 21 (91.3%) received at least 1 payment, with a combined total of $3,335,413 between 2016 and 2018. Regarding 25 US authors, 21 (84.0%) received at least 1 payment, with a combined total of $4,081,629 during the same period. The 3-year combined mean ± SD payment per author was $145,018 ± $114,302 in Japan and $162,825 ± $259,670 in the US. A total of 18 authors (78.3%) of the JDA PsA CPG and 12 authors (48.0%) of the ACR PsA CPG had undisclosed financial COI worth $474,663 and $218,501, respectively. The percentage of citations with at least 1 CPG author relative to total citations was 3.4% in Japan and 33.6% in the US. In sum, 71.4% and 88.8% of recommendations for PsA in the JDA and ACR were supported by low or very low quality of evidence. Conclusion: More rigorous cross-checking of information disclosed by pharmaceutical companies and self-reported by physicians and more stringent and transparent COI policies are necessary.
AB - Objective: To assess financial conflicts of interest (COI) and nonfinancial COI among psoriatic arthritis (PsA) clinical practice guideline (CPG) authors in Japan and the US, and to evaluate the quality of evidence and strength of recommendations of PsA CPGs. Methods: We performed a retrospective analysis using payment data from major Japanese pharmaceutical companies and the US Open Payments Database from 2016 to 2018. All authors of PsA CPGs issued by the Japanese Dermatological Association (JDA) and American College of Rheumatology (ACR) were included. Results: Of 23 CPG authors in Japan, 21 (91.3%) received at least 1 payment, with a combined total of $3,335,413 between 2016 and 2018. Regarding 25 US authors, 21 (84.0%) received at least 1 payment, with a combined total of $4,081,629 during the same period. The 3-year combined mean ± SD payment per author was $145,018 ± $114,302 in Japan and $162,825 ± $259,670 in the US. A total of 18 authors (78.3%) of the JDA PsA CPG and 12 authors (48.0%) of the ACR PsA CPG had undisclosed financial COI worth $474,663 and $218,501, respectively. The percentage of citations with at least 1 CPG author relative to total citations was 3.4% in Japan and 33.6% in the US. In sum, 71.4% and 88.8% of recommendations for PsA in the JDA and ACR were supported by low or very low quality of evidence. Conclusion: More rigorous cross-checking of information disclosed by pharmaceutical companies and self-reported by physicians and more stringent and transparent COI policies are necessary.
UR - http://www.scopus.com/inward/record.url?scp=85143206464&partnerID=8YFLogxK
U2 - 10.1002/acr.25032
DO - 10.1002/acr.25032
M3 - Article
C2 - 36194077
AN - SCOPUS:85143206464
SN - 2151-464X
VL - 75
SP - 1278
EP - 1286
JO - Arthritis Care and Research
JF - Arthritis Care and Research
IS - 6
ER -