A survey of the research practice in general medicine departments of Japanese universities: A cross‐sectional study

Abstract Background Few studies have focused on research practice in Japanese university general medicine (GM) departments. Methods This is a questionnaire‐based cross‐sectional study to clarify the research achievement and associated factors of Japanese university GM department. Univariate analysis was performed to compare the number of English‐language research publications and explanatory variables. Results Forty‐seven universities responded. Over a 3 years period, the median number of English‐language research publications was 6. Perceived degree of research necessity, staff numbers, collaborative research, conference presentations, and obtaining research grants were significantly associated with a higher number of English‐language research publications. Conclusions While GM research output was found to be limited, numerous associated factors can potentially change Japanese GM departments' research environments.

on the public mailing list of the Council of Japanese University Hospitals for General Medicine. The council covers GM departments in university hospitals throughout Japan, and its annual meeting aims to promote communication and information sharing among Japanese university hospitals. Responses were collected using Google Forms. The department chairperson was responsible for completing the questionnaire, which required department name and chairperson position to be stated. All authors collaboratively developed the questionnaire; it comprised five-point Likert-scale, yes-no, descriptive, and numeric questions. The questions and definitions appear in the Appendix S1.
We set the primary outcome of this study as the number of English-language research publications (ELRP) in the 3 years from 2017 to 2019 because it is an internationally assessable research level. The Spearman correlation coefficient (r) was calculated to test correlations with a range of factors. Completed questionnaires were divided into two groups based on the median number of ELRP, and missing values were excluded from the comparative analysis.
Continuous variables were expressed using median values and interquartile range and were compared using the Mann-Whitney U-test.
Categorical variables were expressed as percentages and compared using chi-square tests or Fisher's exact test. Statistical significance was set at p < 0.05. IBM SPSS version 25 (IBM Corp., Armonk, NY) was used for statistical analyses.
All subjects gave informed consent on the questionnaire Web site. The Ethics Committee of Saga University Hospital waivered this study because it was not conducted on humans, and did not include personal information, and university names were anonymized in the analysis.
other factors, and results of univariate analysis are shown in Table 1.  Table 1).
In the univariate analysis, the group with the higher number of This study showed that only on average (median) six ELRP were published by GM university departments over this 3 year period. A previous report showed a lower rate of ELRP in major international journals associated with primary care 4,5 ; therefore, the growth of academic practice in GM in Japan has been strongly expected. The

| CON CLUS IONS
Research practice in GM departments of Japanese universities was found to be limited. Research achievements were associated with the perceived degree of research necessity, staff and postgraduate student numbers, collaborations with other facilities, presentations in academic conferences, and obtaining research grants. Focusing on these factors could help to establish systems for research education, research collaboration, and research sharing within and between universities and more broadly in the field of GM.

ACK N OWLED G EM ENTS
We thank Prof. Taro Shimizu, Dokkyo Medical University, Dr. Yosuke Sasaki, Toho University School of Medicine, and Dr. Hiromizu Takahashi, Juntendo University for their support to this work. We thank Edanz Group (https://en-autho r-servi ces.edanz.com/ac) for editing a draft of this manuscript.

CO N FLI C T O F I NTE R E S T
The authors have stated explicitly that there are no conflicts of interest in connection with this article.