Effect of operator experience level on clinical outcomes of indirect- and direct-bonding techniques: A single-center, open-label, quasi-randomized controlled clinical trial Authors: Kana Kono, Saori Tanizaki, Atsuro Fujisawa, Takashi Murakami, Noriaki Kawanabe, Masahiro Nakamura, Mitsuhiro Hoshijima, Satoru Hayano, Takashi Izawa, Hiroshi Kamioka Published: 2026 ABSTRACT ObjectiveThis study aimed to compare the clinical outcomes of indirect- and direct-bonding techniques for bracket placement in orthodontic treatment, focusing on the influence of operator experience.MethodsA total of 153 patients were quasi-randomly assigned to the indirect- or direct-bonding group, of whom 124 were analyzed. Clinical outcomes in the two groups were compared and further stratified by operator experience (1–3 years vs. 4–6 years). The evaluated parameters included total treatment duration, chair time, discomfort during bonding (visual analogue scale scores), plaque control record, bracket failures, intentional reattachments, and post-treatment occlusal index (Cast-Radiograph Evaluation [CR-Eval] scores).ResultsIn procedures performed by operators with 1–3 years of experience, the indirect-bonding group demonstrated significantly shorter treatment duration, shorter chair time, fewer intentional reattachments, and better post-treatment occlusal outcomes than the direct-bonding group. However, the number of bracket failures was significantly higher in the indirect-bonding group, regardless of operator experience. In procedures performed by operators with 4–6 years of experience, most parameters showed no significant intergroup differences.ConclusionsThe indirect-bonding technique improved treatment efficiency and quality, with potential advantages for less-experienced operators. However, attention to technique is essential to minimize bracket failures. These findings support the incorporation of indirect bonding in clinical training programs. INTRODUCTION Accurate and efficient bracket placement is essential for achieving ideal tooth alignment and occlusion in fixed orthodontic treatment. Improper bracket positioning can hinder the desired tooth movement, prolong treatment duration, and necessitate additional chairside adjustments.1 Two primary techniques are currently used for bracket bonding: direct bonding, wherein brackets are placed intraorally on each tooth, and indirect bonding, wherein brackets are first positioned on a dental model and subsequently transferred to the patient’s dentition using a customized transfer tray.2-4 Although indirect bonding can improve bracket-placement accuracy and reduce clinical chair time,1,3,5-7 concerns regarding its clinical reliability—such as higher bracket failure rates and procedural complexity—persist, particularly among less-experienced clinicians.8-10 Most previous studies have described the advantages and disadvantages of each technique separately,5,6,11,12 focusing on individual factors rather than comprehensively evaluating all relevant variables in a single model. Operator proficiency also appears to influence orthodontic bonding outcomes, but its modifying effect on direct versus indirect bonding has rarely been examined. Less-experienced practitioners are more likely to misplace brackets; therefore, they may benefit from the standardized workflow provided by indirect-bonding systems (IDBSs).13,14 To address these gaps, we had previously performed a clinical evaluation comparing direct-bonding systems (DBSs) and IDBSs within a single institutional setting.14 In the present study, we further investigated how operator experience affects the choice and clinical performance of these bonding techniques. This quasi-randomized controlled trial aimed to provide guidance for selection of the appropriate bonding technique in clinical and educational contexts. MATERIALS AND METHODS Eligibility criteria Patients who met the predefined inclusion criteria were screened for participation. The inclusion and exclusion criteria are listed in detail in Supplementary Table 1. Written informed consent was obtained from the patients, and patients who declined were excluded. This single-center, open-label trial adhered to the Declaration of Helsinki. The study protocol received approval from the Institutional Review Board of Okayama University Hospital (Approval No. d10001) and was previously published.13 The trial was retrospectively registered with UMIN (UMIN000022182) on May 2, 2016. Endpoints This study compared the clinical outcomes of indirect- and direct-bonding techniques. The primary endpoint was the total treatment duration (months). Secondary endpoints included patient-reported discomfort during bonding (visual analogue scale [VAS] scores), oral hygiene status after bonding (plaque control record [PCR], %),15 post-treatment occlusal index (Cast-Radiograph Evaluation [CR-Eval] scores),16 chairside bonding time (minutes), number of bracket failures, and number of intentional bracket reattachments. The null hypothesis posited that the two techniques had no significant difference in efficacy. The 124 analyzed patients were followed from bracket placement to removal. Outcomes were further analyzed on the basis of the operators’ clinical experience level, which was categorized into two groups (1–3 years and 4–6 years) on the basis of the time required to obtain certification by the Japanese Orthodontic Association. Treatment methods Sample-size calculation Sample-size determination was based on a previously described method.14 Calculations were based on detecting differences in total treatment duration between groups, assuming a significance level of 0.05, 80% power, and standard deviation of 5 months. A minimum of 50 patients per group was considered to be adequate. Intervention This study was conducted from January 2015 to March 2019. Eligible patients were quasi-randomly allocated to undergo either the conventional direct-bonding technique or the indirect-bonding technique. Twenty-one orthodontists, all full-time staff members at the Department of Orthodontics, Okayama University Hospital, who had received prior training in both bonding methods, participated in the study. Most operators performed both techniques. The operators’ clinical experience ranged from 1 to 6 years. Bracket-bonding procedures and materials were employed using previously described protocols.14 In this study, both the direct- and indirect-bonding procedures—including tray fabrication, transfer media, adhesive and primer selection, and curing protocols—were performed exactly as described in our previous publication,14 which provides a detailed description of the complete workflow. Baseline cephalometric and cast analyses were performed at the initial visit. Age was not used as a stratification factor during quasi-randomization, nor was it included as a covariate in the statistical analysis. Although the age distribution varied among participants, skeletal and dental characteristics—including arch-length discrepancy—were comparable between the IDBS and DBS groups, indicating similar severity levels of malocclusion at baseline. The indirect-bonding procedure employed the same single-tooth transfer trays described previously. When severe crowding limited access to the anterior teeth and precluded simultaneous bonding of multiple teeth, the trays were seated in multiple steps. In such cases, the total chairside bonding time was calculated as the sum of the time required for all bonding steps to maintain consistency in efficiency evaluation. Tray detachment or loss of adaptation during indirect bonding was documented when present; however, such events were infrequent and did not systematically prolong the bonding time. Follow-up assessments Participants were monitored during active orthodontic treatment. Chair time (minutes) was recorded at bonding. Within one month, patients completed a questionnaire that assessed discomfort during bonding by using a 100-point VAS. Oral hygiene status (%) was assessed by a dental hygienist. At the end of the treatment period, the outcomes measured included total treatment duration, post-treatment occlusal index, and the number of bracket failures. The occlusal index was assessed using CR-Eval scores. Both post-treatment dental casts and panoramic radiographs were evaluated by a certified orthodontist blinded to group allocation. Both were reevaluated one week later, and the mean value was used. Quasi-randomization After confirming eligibility, patients were quasi-randomly allocated to either group. The project lead (KK), who was blinded to clinical data, allocated eligible patients alternately to the two groups according to the order of their medical record numbers. Operator assignment Operators were stratified into two predefined experience categories (1–3 years and 4–6 years), and those within each category had comparable levels of clinical exposure to fixed-appliance therapy. Case allocation was conducted in accordance with the quasi-randomized clinical workflow of our department: patients were assigned sequentially on the basis of daily scheduling and operator availability, and operators could not choose cases, preventing systematic differences among operators. The distribution of cases employing direct and indirect bonding per operator is presented in Supplementary Table 2. Blinding Due to the inherent visual and technical differences between direct- and indirect-bonding methods, blinding of operators and patients was not feasible. Statistical considerations Data were analyzed using either a two-sample t test or Mann–Whitney U test, depending on the data distribution. Normality was assessed using the Shapiro–Wilk test. Parametric and non-parametric tests were applied accordingly. Statistical significance was set at P < 0.05 with a 95% confidence interval. Analyses were conducted using SPSS 28.0 (IBM Corp., Armonk, NY, USA). Additional statistical analyses To account for the hierarchical clustering of patients within operators, a linear mixed-effects model was applied with operator as the random intercept and the bonding technique (direct vs. indirect) and operator experience (1–3 years vs. 4–6 years) as fixed effects. The model was used to assess total treatment duration, chairside bonding time, bracket failures, and intentional reattachments, with statistical significance set at P < 0.05. All analyses were performed using SPSS. RESULTS Participant flow and treatment abandonment The participant flow was previously reported.14 Of the 1,059 patients who underwent screening, 153 met the inclusion criteria and were quasi-randomly assigned to the IDBS (n = 72) or DBS (n = 81) group. Among these, 29 patients were lost to follow-up (IDBS group, 14; DBS group, 15), and the remaining 124 patients were included in the analysis: 58 in the IDBS group (20 male, 38 female; mean age 20.63 ± 5.69 years) and 66 in the DBS group (14 male, 52 female; mean age 23.17 ± 8.83 years). Operator experience was defined by postgraduate year at the time of bonding (Figure 1). Among the 21 orthodontists who were initially scheduled to participate in the study, one operator was excluded from the analysis because the patient assigned to this operator transferred to another clinic before debonding. Thus, 20 operators contributed cases in this study. Each operator was involved in 0–8 direct-bonding cases and 0–7 indirect-bonding cases (Supplementary Table 2). Baseline characteristics No significant differences in patient skeletal or dental characteristics, including arch-length discrepancy, were observed between the DBS and IDBS groups, as previously reported.14 No significant differences were observed between the 1–3- and 4–6-year experience groups within either bonding method (all P > 0.05; Supplementary Table 3). Extraction patterns (non-extraction, maxillary extraction, bimaxillary extraction) also did not differ significantly in relation to the bonding methods or experience levels (all P > 0.05; Supplementary Table 4). A subsequent between-subjects analysis of variance (ANOVA) confirmed that the extraction pattern showed no significant main effect on the total treatment duration (P = 0.068) and no significant interactions with the bonding method (P = 0.743) or operator experience level (P = 0.212). Total treatment duration (months) The total treatment duration was shorter in the IDBS group than in the DBS group (median, 30.52 vs. 33.85), with a lower interquartile range, indicating less variability in treatment duration (Supplementary Figure 1). In procedures performed by operators with 1–3 years of experience, the treatment duration was significantly shorter with indirect bonding than with direct bonding (30.61 ± 7.55 vs. 35.43 ± 8.50, P < 0.01; Figure 2A). In contrast, in procedures performed by operators with 4–6 years of experience, the IDBS and DBS groups showed no significant difference in the total treatment duration (30.01 ± 5.44 vs. 28.17 ± 8.21; Figure 2B). Subgroup distributions are shown in Supplementary Figure 2. Chair time (minutes) Chair time was defined as the total time required for bonding and banding, including surface cleaning, trial fitting, curing, and removal of excess resin or cement. For patients who required multiple bonding visits, chair time was the sum of the time required for all sessions. In procedures performed by operators with 1–3 years of clinical experience, the time required to bond brackets from the mandibular incisors to the first molars was significantly shorter in the IDBS group (DBS group: 61.63 ± 24.98, IDBS group: 36.25 ± 16.55, P < 0.001) (Figure 3A). However, this parameter did not differ significantly between the IDBS and DBS groups in procedures performed by operators with 4–6 years of clinical experience (DBS group: 42.31 ± 19.80, IDBS group: 33.84 ± 6.00) (Figure 3B). Discomfort at bonding (VAS scores, mm) Discomfort during bonding was evaluated for the mandibular incisors through premolars and mandibular first molars (Table 1). In procedures performed by operators with 1–3 years of experience, discomfort during mandibular first molar bonding was significantly lower with indirect bonding than with direct bonding (30.22 ± 21.88 vs. 41.57 ± 24.08, P < 0.05). In contrast, in procedures performed by operators with 4–6 years of experience, the perceived duration of bracket placement on the mandibular incisors through premolars was significantly longer with indirect bonding (39.38 ± 20.56 vs. 13.05 ± 13.81, P < 0.05), and discomfort after bracket placement on the mandibular incisors through premolars was also significantly higher with indirect bonding (63.50 ± 8.69 vs. 35.20 ± 21.51, P < 0.05; Table 1). Oral hygiene (PCR, %) The DBS and IDBS groups showed no significant differences in PCR scores at any time point and in either experience group (all P > 0.05; Figure 4). Number of bracket failures The number of bracket failures was significantly higher in the IDBS group across both experience levels: 1–3 years: DBS group, 2.99 ± 3.21 vs. IDBS group, 7.21 ± 5.11; P < 0.001 (Figure 5A) 1–3 years: DBS group, 2.99 ± 3.21 vs. IDBS group, 7.21 ± 5.11; P < 0.001 (Figure 5A) 4–6 years: DBS group, 2.54 ± 2.10 vs. IDBS group, 7.69 ± 4.18; P < 0.01 (Figure 5B) 4–6 years: DBS group, 2.54 ± 2.10 vs. IDBS group, 7.69 ± 4.18; P < 0.01 (Figure 5B) Number of intentional reattachments Intentional reattachments were significantly fewer in the IDBS group than in the DBS group at both experience levels: 1–3 years: DBS group, 9.15 ± 8.14 vs. IDBS group, 3.48 ± 3.74; P < 0.001 (Figure 6A) 1–3 years: DBS group, 9.15 ± 8.14 vs. IDBS group, 3.48 ± 3.74; P < 0.001 (Figure 6A) 4–6 years: DBS group, 7.54 ± 5.65 vs. IDBS group, 2.05 ± 2.24; P < 0.01 (Figure 6B) 4–6 years: DBS group, 7.54 ± 5.65 vs. IDBS group, 2.05 ± 2.24; P < 0.01 (Figure 6B) Post-treatment occlusal index (CR-Eval scores) In procedures performed by operators with 1–3 years of experience, total CR-Eval scores were significantly lower with indirect bonding than with direct bonding (47.74 ± 10.21 vs. 53.11 ± 9.67, P < 0.05), indicating better occlusal outcomes with indirect bonding (Figure 7A). However, CR-Eval scores showed no significant differences in relation to the bonding method in procedures performed by operators with 4–6 years of experience (43.90 ± 10.73 vs. 54.44 ± 8.21; Figure 8A). No significant method-related differences were observed in alignment, marginal ridge level, buccolingual inclination, overjet, occlusal contacts, or root angulation in either experience group (all P > 0.05; Figures 7B–F, I and 8B–F, I). Occlusal relationship scores were significantly better with indirect bonding in procedures performed by operators with 1–3 years of experience (5.98 ± 3.75 vs. 7.89 ± 4.34, P < 0.05; Figure 7G). Similarly, interproximal contact scores were significantly better with indirect bonding in both experience groups (1–3 years: 2.99 ± 2.18 vs. 5.93 ± 2.36, P < 0.001; 4–6 years: 1.90 ± 1.16 vs. 6.44 ± 1.98, P < 0.01; Figures 7H and 8H). Intra-examiner reliability was excellent, with ICC (1,1) values of 0.953 for the IDBS group and 0.857 for the DBS group, both exceeding 0.75 and indicating sufficient consistency and reproducibility of the measurements. Additional mixed-effects analysis Given the hierarchical structure of the data (patients clustered within operators), a linear mixed-effects model was employed with bonding technique (direct vs. indirect) and operator experience (1–3 years vs. 4–6 years) as fixed effects and operator identity as a random intercept. The model showed that indirect bonding significantly reduced the total treatment duration (F = 4.827, P < 0.05; mean reduction, 3.40 months) and chair time (F = 35.019, P < 0.001; mean reduction, 22.26 minutes) while showing significantly more bracket failures (F = 28.492, P < 0.001) and significantly fewer intentional reattachments (F = 28.100, P < 0.001) than with direct bonding. Neither the main effect of operator experience nor the method × experience interaction was significant in any of the mixed-effects models (all P > 0.05), suggesting that potential experience-related differences should be interpreted cautiously after accounting for clustering within operators (Supplementary Table 5). Harms No harms were observed during the study. DISCUSSION This study evaluated the differences between indirect and direct bonding stratified by operator experience. Mixed-effects analysis accounted for clustering within operators. Baseline cephalometric and cast analyses showed comparable skeletal and dental characteristics between the groups, and no statistically significant effect of extraction pattern was detected. Bracket types and bonding materials were standardized according to the published protocol.14 Together, these findings suggested that the shorter treatment duration observed with indirect bonding in the less-experienced subgroup was primarily attributable to the bonding method rather than differences in baseline characteristics or extraction patterns; however, this subgroup finding should be interpreted as exploratory. Among operators with 1–3 years of experience, total treatment duration was significantly shorter with indirect bonding than with direct bonding. In contrast, the total treatment durations with indirect and direct bonding showed no significant differences in operators with 4–6 years of experience, consistent with the findings of previous reports.10 This subgroup pattern suggests a potential experience-related difference in the efficiency of indirect bonding and should be interpreted as exploratory. The similar treatment duration across experience levels in the IDBS group may reflect reduced operator influence associated with the standardized indirect-bonding workflow. Another key finding was the significant reduction in chair time with indirect bonding, particularly among operators with 1–3 years of experience. This is consistent with the findings of previous reports showing that indirect bonding improves efficiency by using pre-positioned transfer trays and reducing intraoral working time.17-19 However, no time-saving effect was seen among operators with 4–6 years of experience, suggesting that growing familiarity with direct bonding compensates for its procedural inefficiencies. Although banding may require more time than bonding, the lack of a difference between DBS and IDBS among operators with 4–6 years of experience suggests that the influence of banding time was minimal. Regarding patient-reported experience during bonding, significantly lower discomfort specifically during mandibular first molar bonding was observed with indirect bonding in procedures performed by less-experienced operators. Although this contrasts with the findings of a previous report,20 it may reflect an interaction between operator experience and the bonded region: transfer trays are used to stabilize posterior brackets, where visibility and access are limited; however, experienced operators can perform anterior bonding more efficiently with direct bonding, and tray seating in indirect bonding may increase pressure on anterior teeth. Because the questionnaire was administered within one month rather than immediately after bonding, the possibility of a response shift due to psychological adaptation cannot be excluded. The two groups showed no significant differences in PCR values at any time point, consistent with the findings of several previous studies.8,21,22 Although previous studies suggested improved cleanliness with indirect bonding,23,24 the absence of differences under standardized baseline conditions (≤ 20%) indicates that plaque accumulation did not confound the comparison. The number of bracket failures was significantly higher in the IDBS group regardless of operator experience, consistent with previous concerns outlining minor deviations in bracket position or adhesive contact during transfer, particularly in curved or moisture-sensitive regions.4,8 Nevertheless, the lower number of intentional reattachments indicates superior initial bracket positioning when bonding was successful. Consistent with the findings of our previous analysis,14 failures in the IDBS group were diffusely distributed across both arches rather than being localized to specific teeth, indicating technique-dependent factors such as reduced pressure transmission during tray transfer, limited moisture control, and increased occlusal interference in the mandibular dentition. Reduced pressure transmission may be related to transfer tray properties, although excessive rigidity may reduce transfer accuracy,25 indicating that optimization of tray characteristics requires further investigation. These factors may explain the greater number of bracket failures despite fewer reattachments and align with reports describing higher failure rates with indirect bonding (13–35%) than with direct bonding (~2%).8 In our cohort, plastic brackets in the premolar regions were particularly prone to failure under indirect bonding, likely due to lower shear bond strength, lower durability, and greater susceptibility to occlusal interference. First-debonding events occurred predominantly during the early treatment phase, supporting the view that the procedural limitations of indirect bonding predispose brackets to early detachment. In a previous CR-Eval study, 54.9% of the cases showed scores exceeding 30 points,26 supporting the appropriateness of the scores observed here, although CR-Eval reliability may be limited by subjectivity and our slightly higher scores may reflect stricter assessment criteria.27 Significantly better final occlusal outcomes with indirect bonding were observed among less-experienced operators, supporting the benefits of standardized bracket positioning in reducing technique variability. Interproximal contact scores were consistently better with indirect bonding regardless of operator experience, whereas poorer scores with direct bonding may be related to molar band use. No significant between-group differences were found in alignment, marginal ridge level, buccolingual inclination, or overjet, indicating that both techniques can achieve acceptable occlusal outcomes with appropriate finishing. Although this study used the same cohort, methodology, and clinical outcomes as our previous work,14 it addressed a distinct topic, i.e., the influence of operator experience on the outcomes of direct and indirect bonding. The operator-experience subgroup and mixed-effects analyses are new to the present study. Because bonding is mainly performed by early-stage residents, the number of operators with 4–6 years of experience was limited, limiting the interpretation of comparisons between experience groups. Despite this limitation, the subgroup findings suggested a potential experience-related pattern in bonding efficiency, which should be interpreted as exploratory. Indirect bonding may help standardize treatment quality across different proficiency levels. This study had several limitations. The greater number of bracket failures in the IDBS group may have been related to adhesive handling or tray-fabrication quality. In addition, discomfort assessments are inherently subjective and may be influenced by psychological or communication factors. Moreover, blinding of operators and patients was not feasible because the direct- and indirect-bonding procedures were visually distinct. In addition, multiple outcomes and operator subgroups were analyzed without formal correction for multiple testing, increasing the risk of type I error. Furthermore, the sample-size calculation was performed at the patient level and did not account for operator clustering. Because each operator treated multiple patients, within-operator correlations may have reduced the effective sample size, and the study was not powered to detect interaction effects between operator experience and bonding technique. Accordingly, the findings of these subgroup analyses should be interpreted as exploratory. Finally, operator experience was classified into two groups based on national certification criteria. Although this approach was practical, it may have oversimplified skill development and limited assessment precision. Years of experience may not fully reflect individual proficiency within each category. Future studies using more refined operator-level metrics may better capture experience-related effects. In conclusion, indirect bonding may improve efficiency and occlusal outcomes, particularly for less-experienced clinicians. However, careful tray design and strict bonding protocols remain essential to minimize bracket failures. Early incorporation of IDBS may support clinical training without hindering direct-bonding skill development. IDBS should complement, not replace, direct-bonding proficiency in structured training programs. CONCLUSIONS Indirect bonding offers meaningful clinical advantages over direct bonding, with potential advantages for less-experienced operators. It significantly reduced treatment duration, chair time, and intentional reattachments while improving occlusal outcomes. However, the greater number of bracket failures highlights the need for careful tray design, moisture control, and strict adherence to bonding protocols. Experience-related subgroup patterns should be interpreted as exploratory. Overall, indirect bonding may improve treatment predictability, efficiency, and quality in clinical training while complementing, rather than replacing, direct-bonding proficiency. FUNDING None to declare.