Introduction
Antimicrobial resistance (AMR) is an urgent global and domestic health threat. Reference Murray, Ikuta and Sharara1,Reference Tsuzuki, Matsunaga and Yahara2 Furthermore, the COVID-19 pandemic has demonstrated that infectious disease outbreaks are not merely a health problem, but can also have devastating social and economic consequences. 3 Although antimicrobials are necessary for the prevention and treatment of bacterial infections, indiscriminate use contributes to the emergence of drug-resistant strains. Reference O’Neill4 Therefore, there is a need to systematically improve the rational and judicious use of antimicrobials based on proper indications and diagnoses. 5
In addition to medical practice, antimicrobials are also prescribed in dental practice. Accordingly, it is important to ensure that dental prescribing also conforms to current guidelines. Reference Marra, George, Chong, Sutherland and Patrick6 Studies from France, Norway, and Japan have reported that approximately 10% of antimicrobials are prescribed by dentists. Reference Simon, Pereira, Constant, Guillet-Thibault and Pulcini7–Reference Ono, Ishikane and Kusama9 The total use of antimicrobials in Japan in 2019 (ie, before the COVID-19 pandemic) was estimated to be 13.1 defined daily doses per 1,000 inhabitants per day (DID), 10 which was lower than that of the European Union’s median consumption of 19.3 DID (interquartile range: 14.4–21.1 DID). 11 However, dentists in Japan are reported to preferentially prescribe broad-spectrum cephalosporins (accounting for 65.1% of cases), and are less likely to choose narrow-spectrum penicillins (accounting for only 5.5% of cases). 10 These prescribing patterns are markedly different from other countries, where penicillins account for the majority of antimicrobial prescriptions (57–79.5%) in dental clinics. Reference Tousi, Haroni, Lie and Lund8,Reference Durkin, Hsueh and Sallah12–Reference Teoh, Stewart, Marino and McCullough13 Furthermore, penicillins are recommended as first-choice antimicrobials in guidelines issued by the Japanese and American societies for infectious diseases, cardiology, and dentistry. 14–17 In this context, dental antimicrobial prescribing in Japan is not in accordance with these guidelines.
In a previous survey on antimicrobial prescribing trends in small dental practices in Japan, we reported that cephalosporins accounted for 56.8% of prescribed antimicrobials, and that four in five dental clinics administered prophylactic antimicrobials for three days after surgery. Reference Koizumi, Kusama and Ishikane18 Another Japanese study analyzed the effectiveness of a pharmacist-led intervention at a university hospital’s outpatient dental clinic to provide direct feedback on antimicrobial selection and dosage to the prescribing dentists; the intervention was found to increase the mean monthly proportions of penicillin prescriptions (45.6 to 84.1 per 100 oral antimicrobial prescriptions) while reducing third-generation cephalosporin prescriptions (40.3 to 7.3 per 100 oral antimicrobial prescriptions) over a five-year period. Reference Okihata, Michi, Sunakawa and Tagashira19 While such initiatives may be improving antimicrobial stewardship in hospitals, prescribing habits in dental clinics—where the majority of dental care in Japan is provided—remain unclear.
To facilitate the improvement of antimicrobial use in dental clinics, this cross-sectional questionnaire study was conducted in cooperation with the Japan Dental Association to examine the antimicrobial prescribing practices in dental clinics throughout Japan. Specifically, this study aimed to identify the factors associated with the use of penicillins as the first-choice antimicrobial in dental practice.
Methods
Study design and period
We conducted a nationwide cross-sectional study using a paper-based questionnaire survey of dentists working in dental practice in Japan. Questionnaires were sent by post in July 2020 with self-addressed prepaid return envelopes, and responses were collected during the two-month period from July to September 2020. During this period, we sent a reminder to respond by post one month after the questionnaires were distributed. No incentives were offered to improve the response rate.
Participants
Questionnaires were sent to 1,700 (2.5%) of the 68,404 dental care facilities in Japan. 20 The distribution was weighted by the number of dental care facilities per prefecture as of December 2019 using information from the Regional Bureaus of Health and Welfare. Facilities were randomly selected using the RAND function in Microsoft Excel®, which generates random numbers to produce an unbiased selection. A questionnaire was mailed to each facility, and the target respondents were either the directors of the facility or dentists who were familiar with practice at that facility. In order to raise awareness of the survey and encourage responses, the Japan Dental Association was asked to cooperate with the study. The Japan Dental Association is a professional organization with a membership that includes 64,544 (61.5%) of the 104,908 dentists in Japan. 21 One week prior to the distribution of the questionnaires, the Japan Dental Association informed all members about the survey’s purpose and contents, and encouraged them to respond. The respondents’ personal information was anonymized before analysis.
Questionnaire
To examine prescribing habits and their underlying factors, the following sections were included in the questionnaire (Supplementary material) based on previous studies: (i) Basic Characteristics, (ii) Regarding AMR, (iii) Regarding the Prevention of Infective Endocarditis (IE), and (iv) Regarding the Prescription of Antimicrobials in Outpatient Dentistry.
Outcome measures
From the questionnaire responses, we determined the following study outcome measures in each respondent: (i) provision of on-site dispensing of penicillins (yes or no) and (ii) use of penicillins as the first-choice antimicrobial (yes or no) among dental clinics that stock penicillins for on-site dispensing.
Candidate factors
This study examined the associations of the following candidate factors with the outcome measures: age, years in practice, type of university from which the respondent graduated (private or public), use of journals/websites as information sources, clinic with ≥2 full-time dentists, specialty certification, clinic performs implant placements, awareness of the term “AMR,” awareness of the “National Action Plan (NAP) on AMR,” implementation of AMR countermeasures, consideration of IE risk in daily consultations, adherence to IE guidelines, attended at least one seminar in the past six months, and clinic stocks ≥2 antimicrobials. Age and years in practice were analyzed as continuous variables, while the other factors were analyzed as categorical variables.
Statistical analysis
First, we descriptively analyzed the questionnaire responses. Categorical variables were expressed as numbers and percentages, and continuous variables were expressed as median values and interquartile range. We described the basic characteristics of the respondents and their clinics, as well as their awareness of AMR, implementation of AMR countermeasures, and awareness of IE. In addition, we examined the timing and duration of prophylactic antimicrobial use for various procedures, including tooth extraction (non-impacted and impacted), implant placement, and periodontal surgery.
Next, we examined the associations between the individual candidate factors and on-site dispensing of penicillins (yes or no) using univariable analyses. The associations between the continuous variables (age and years in practice) and on-site dispensing of penicillins were analyzed using simple logistic regression models, while the associations between the categorical variables and on-site dispensing of penicillins were analyzed using Fisher’s exact test. Thereafter, we constructed a multivariable logistic regression model using independent variables that were hypothesized a priori to be clinically or epidemiologically important. For variables representing similar constructs, only the more relevant variable was selected for inclusion (eg, years in practice vs age, implementation of AMR countermeasures vs awareness of the term “AMR”). The following factors were included in the multivariable logistic regression model: years in practice, clinic with ≥2 full-time dentists, implementation of AMR countermeasures, adherence to IE guidelines, attended at least one seminar in the past six months, and clinic stocks ≥2 antimicrobials. The odds ratios (OR) and 95% confidence intervals (CI) of the independent variables were calculated.
Using the same univariable analyses (simple logistic regression and Fisher’s exact test), we examined the associations between the individual candidate factors and the use of penicillins as the first-choice antimicrobial (yes or no) among dental clinics that stock penicillins for on-site dispensing. We also constructed a multivariable logistic regression model using the same factors listed above.
Statistical significance was set as a two-sided P value of <.05. All statistical analyses were performed using R ver. 4..0 (R Foundation for Statistical Computing, Vienna, Austria).
Ethical approval
This study was approved by the Institutional Review Board of the National Center for Global Health and Medicine (Approval number: NCGM-G-003345-00).
Results
Respondent characteristics and AMR-related responses
Of the 1,700 dental care facilities that were sent questionnaires, 342 (20.1%) dentists from dental clinics responded. The respondent characteristics and AMR-related responses are summarized in Table 1. The median age of the respondents was 57 years (interquartile range: 49–65 yr), and 87.6% were men. Approximately 74.9% were graduates of private universities, and 76.3% were working in clinics with one full-time dentist. Implant placements were performed in 47.8% of the clinics. The majority of respondents (62.7%) attended 1–5 seminars within the past six months. The most prevalent information source was the journal and/or website of the Japan Dental Association (74.6%). While 93.9% of respondents were aware of the term “AMR,” only 20.8% were aware of Japan’s NAP. Most clinics (79.4%) had implemented AMR countermeasures, and the main countermeasures were as follows: consideration of indications for antimicrobial prophylaxis (71.1%), provision of medication guides to patients (43.3%), and shorter prescription durations (31.1%). Although 72.9% of respondents indicated that they considered the risk of IE in daily consultations, 59.5% were not familiar with the contents of IE guidelines.
Table 1. Respondent characteristics and AMR-related responses (n = 342)
Values are presented as n (%) unless otherwise stated.
AMR, antimicrobial resistance; IE, infective endocarditis.
Timing and duration of prophylactic antimicrobial use for dental procedures
Prophylactic antimicrobials were most commonly administered postoperatively for three days, regardless of procedure (Table 2). In clinics with on-site dispensing, the most common numbers of stocked antimicrobials were two (27.6%) and three (27.6%), followed by four (22.0%), five or more (11.8%), and one (10.8%). The most commonly prescribed antimicrobials were third-generation cephalosporins (76.0%), macrolides (73.4%), and penicillins (48.7%); and the most commonly stocked combination antimicrobial was third-generation cephalosporins with macrolides (56.6%).
Table 2. Timing and duration of prophylactic antimicrobial use for dental procedures (n = 331)
Values are presented as n (%). a Includes multiple responses.
Factors associated with penicillin use among dentists
We first investigated the factors associated with the provision of on-site dispensing of penicillins (Table 3). ORs greater than 1 indicate a higher probability of penicillin being dispensed at a clinic. In the univariable analyses, we found that “consideration of IE risk in daily consultations” (OR: 1.24, 95% CI: 1.24–3.83, P = .004), “adherence to IE guidelines” (OR: 2.63, 95% CI: 1.38–5.07, P = .001), and “clinic stocks ≥2 antimicrobials” (OR: 4.93, 95% CI: 1.91–15.08, P<.001) were positively associated with on-site dispensing of penicillins. The multivariable analysis showed that “adherence to IE guidelines” (OR: 2.56, 95% CI: 1.48–4.50, P = .001) and “clinic stocks ≥2 antimicrobials” (OR: 5.02, 95% CI: 1.91–15.08, P = .002) were positively associated with on-site dispensing of penicillins.
Table 3. Analysis of factors associated with on-site dispensing of penicillins (n = 304)
Values are presented as n (%) unless otherwise stated. The multivariable analysis excluded responses with missing values.
AMR, antimicrobial resistance; CI, confidence interval; IE, infective endocarditis.
Next, we analyzed the factors associated with the use of penicillins as the first-choice antimicrobial among dental clinics that stock penicillins for on-site dispensing (Table 4). ORs greater than 1 indicate a higher probability of penicillin being used as the first choice. In the univariable analyses, none of the factors were significantly associated with the use of penicillins as the first-choice antimicrobial. The multivariable analysis showed that “clinic with ≥2 full-time dentists” (OR: .45, 95% CI: .22–.93, P = .033) was negatively associated with the use of penicillins as the first-choice antimicrobial.
Table 4. Analysis of factors associated with using penicillins as the first-choice antimicrobial (n = 148)
Values are presented as n (%) unless otherwise stated. The multivariable analysis excluded responses with missing values.
AMR, antimicrobial resistance; CI, confidence interval; IE, infective endocarditis.
Discussion
This cross-sectional study was conducted to examine antimicrobial prescriptions in dental clinics after the NAP 2016–2020 was published in Japan. 22 We obtained responses from 20.1% of the 1,700 target facilities that were initially sent questionnaires. The respondents were predominantly male dentists in their fifties, which was fairly representative of the demographics of dentists in 2020. 23 Although the vast majority of respondents (93.9%) were aware of the term “AMR,” only 20.8% were familiar with the NAP, which describes the Japanese government’s specific policies designed to address AMR. This suggests that dentists may not have been sufficiently informed about the NAP compared to physicians, and highlights the need for further targeted educational interventions. On the other hand, almost 80% of respondents indicated that their clinics had implemented AMR countermeasures. The most prevalent countermeasure was the consideration of indications for antimicrobial prophylaxis, followed by the provision of medication guides and shorter prescription durations. Despite the low awareness of the NAP, the majority of respondents appeared to recognize the importance of AMR countermeasures. This could reflect the outcomes of multiple educational interventions and other measures designed to improve antimicrobial stewardship in dentistry based on the NAP. 22 For example, seminars and workshops on infectious diseases, prevention of postoperative infections, and drug resistance have been conducted by regional dental associations, academic societies, and government-related organizations throughout Japan to educate dentists on appropriate antimicrobial use. 24–27
Our survey revealed that dentists in dental clinics do not commonly use penicillins as a standard first-choice antimicrobial, and that many are not familiar with or adhere to IE guidelines. IE is a potentially fatal infection that can develop after dental procedures, and its prevention is a major objective for antimicrobial use in dentistry. Penicillins are the recommended choice for prophylaxis against IE, with macrolides as the subsequent option; cephems are recommended for patients who are allergic to the first two choices. The Japanese Circulation Society’s Guidelines for the Prevention and Treatment of IE recommend a single dose of penicillin immediately before oral surgery. 15 Penicillins are preferentially used for prophylaxis in other countries. For example, amoxicillin/clavulanic acid is the most commonly prescribed prophylactic antimicrobial (57%) for implant placements in Portugal, Reference Correia, Ribeiro-Vidal, Gouveia and Faria Almeida28 and 95.7% of dentists indicated that amoxicillin is their first choice of prophylactic antimicrobial in a Brazilian study. Reference Sinhorelli and Oliveira29 However, third-generation cephalosporins were the most frequently used (50.8%) first-choice antimicrobial among our respondents. Moreover, prophylactic antimicrobials were most frequently administered for three days after surgery (69.3%) for all procedures. These patterns deviate from the recommendations of Japanese guidelines. 15 Our findings support those of a large-scale Japanese study that analyzed insurance claims data from 2015 to 2019, and reported that third-generation cephalosporins were the most commonly used prophylactic antimicrobials for tooth extraction (48.5%) in 2019, and that over 80% of cases were prescribed prophylactic antimicrobials for three or more days. Reference Hirayama, Kanda and Hashimoto30 Similarly, a nationwide survey of antimicrobial use among dentists in Japan conducted in 2011 (ie, prior to the release of NAP 2016–2020) also found that third-generation cephalosporins were the most common first choice for the prevention and treatment of periodontal disease (51.2%), and that prophylactic antimicrobials tended to be administered for three or four days (90.4%) after surgery (67.3%). Reference Amari, Uehara and Watanabe31 These results suggest that the antimicrobial type, timing, and duration of prophylaxis in dental practice has remained relatively unchanged even after the NAP was published.
Among our survey respondents, 56.7% replied that they did not stock penicillins for on-site dispensing. Our multivariable analysis showed that adherence to IE guidelines and keeping ≥2 antimicrobials in stock were positively associated with on-site dispensing of penicillins. Among the dental clinics that stock penicillins, the multivariable analysis found that that the presence of ≥2 full-time dentists reduced the probability of using penicillins as the first-choice antimicrobial. This implies that the more dentists in a clinic, the less likely they were to preferentially use penicillin. While it is difficult to determine the reason for this observation from our data, we posit that the presence of multiple dentists may have caused their decision-making process to drift toward a more conservative direction. Specifically, these dentists may have adhered to practices commonly observed among other dentists, which could partially explain the preferential selection of cephalosporins over penicillins in clinics with multiple dentists. The frequent use of cephalosporins in dental practice in Japan has been reported in past surveys conducted in the 1980s and 1990s, Reference Yohkoh, Tanaka and Kimura32,Reference Yohkoh, Hoshino and Ohono33 during which time a large proportion of our survey respondents would have been undergoing training at a dental university hospital. Although there is currently no evidence to support this, it is possible that dentists who completed their training at university hospitals may have learned and continued to use third-generation cephalosporins (even in situations where penicillins should ideally be used) after opening their own practice. Our findings highlight the importance of regular educational activities as opportunities for dentists to consult with one another, thereby increasing their awareness of guidelines and acquiring the latest reliable knowledge on best practices. Although our multivariable analysis found that recent attendance in seminars had higher odds for using penicillins as the first-choice antimicrobial (OR: 2.64, 95% CI: .90–8.63, P = .089), this association was not statistically significant. Further studies are needed to examine this relationship in more depth.
This study has three limitations. First, this was a cross-sectional study using a survey of dental practitioners throughout Japan. However, research funding limited the target respondents to 1,700 dental care facilities, of which only 342 responded. These clinics constitute approximately .5% of all dental practices in Japan, and there is the possibility of sampling bias. Second, prescribing habits were ascertained using a self-administered questionnaire, which raises the risk of response bias. There is therefore a need for additional research using empirical prescription data, such as medical records and insurance claims. Although shortages of antimicrobials may also affect prescribing patterns, no major supply disruptions occurred during the study period. 34 Finally, this survey was conducted in 2020, and it is highly likely that the current situation has changed following the publication of the updated NAP 2023–2027. 35 Therefore, further studies are needed to evaluate how it has changed since the release of the revised plan. For example, educational activities in dentistry include dental infection seminars by pharmacist associations and academic societies, AMR seminars for dentists by government agencies, and AMR education for the public. Antimicrobial stewardship programs, such as feedback on prescriptions, are also being implemented in dental university hospitals. Reference Okihata, Michi, Sunakawa and Tagashira19 Surveillance of antimicrobial use in Japan has detected a decrease in broad-spectrum antimicrobials and an increase in narrow-spectrum penicillins in dental practice, indicating that appropriate use is progressing in this field. 10 Continuous monitoring through similar surveys is necessary to clarify changes in prescribing habits.
Despite these limitations, this is the first survey to identify the antimicrobial prescribing practices in dental clinics after the Japanese government published the NAP 2016–2020. Our results provide important insight that could help to inform the future implementation of AMR control measures in dentistry. The study revealed that Japanese dental clinics do not prioritize keeping penicillins stocked for on-site dispensing, and that awareness of IE guidelines is low. Furthermore, adherence to IE guidelines was found to be associated with stocking penicillins for on-site dispensing. The strategies outlined in the NAP 2023–2027 include the promotion of education and training initiatives for dentists. 35 As a result, many organizations, including dental associations, are conducting seminars and workshops on proper antimicrobial use. Enhancing such educational activities and fostering environments that facilitate collaborative decision-making regarding antimicrobial use may help to improve antimicrobial stewardship in Japan’s dental clinics.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/ash.2025.10065
Acknowledgements
We would like to thank the Japan Dental Association and all survey respondents for their cooperation. We are also grateful to the staff of the AMR Clinical Reference Center, National Center for Global Health and Medicine, for their contributions to this study.
Financial support
This work was supported by a research grant from the Ministry of Health, Labour and Welfare of Japan (Grant number: R2-20HA2003).
Competing interests
All authors report no conflicts of interest relevant to this article.
Open access
