Introduction
As the global population ages, particularly in nations such as Thailand, the prevalence of osteoporosis is increasing rapidly. Consequently, antiresorptive medications, including bisphosphonates and denosumab, are widely prescribed to mitigate the risk of fragility fractures [1], [2]. Although these medications are crucial for reducing osteoporosis-related morbidity and mortality, their use is frequently constrained by concerns regarding medication-related osteonecrosis of the jaw (MRONJ).
While MRONJ is a rare adverse event, with an estimated cumulative incidence of less than 0.5% among patients with osteoporosis [3], [4], multiple clinical factors including drug type, duration of therapy, and invasive dental procedures, may further modulate individual risk [5], [6], [7], [8], [9]. Despite the low absolute risk, MRONJ remains a feared complication among patients receiving antiresorptive therapy [10]. Qualitative evidence indicates that limited understanding of MRONJ risk and preventive strategies may coexist with substantial psychological and social concerns [11].
More broadly, uncertainty and concern regarding rare adverse effects may undermine the acceptability of and adherence to osteoporosis treatment [12].
This anxiety may be further amplified in the digital era. Patients often receive insufficient information about MRONJ risk from healthcare providers [11] and may consequently seek health-related information from online platforms. Online content pertaining to dental conditions frequently contains misleading information, which may amplify anxiety and reinforce misconceptions [13].
Furthermore, exposure to health information on social media can trigger information overload, increase anxiety, and diminish health self-efficacy [14]. These platforms often expose individuals to graphic or emotionally charged MRONJ-related content presented without adequate clinical context, fostering catastrophic thinking and mistrust [13]. In addition, a substantial communication gap exists in routine clinical practice.
There is often limited communication between prescribing clinicians and dental teams regarding MRONJ risk [11], [15], and dentists may feel hesitant to treat patients receiving antiresorptive therapy due to uncertainty regarding established management guidelines [16], [17], [18]. As a result, patients are frequently confronted with conflicting information, exacerbating distress and hesitation toward dental care [19].
Currently, the standard care typically consists of brief verbal consultations, which are inherently inconsistent due to variations in clinician knowledge and time constraints [11], [15]. Accordingly, there is a need for a standardized tool capable of translating complex risk information into personalized reassurance that is readily understandable to patients. Although patient-oriented tools for osteoporosis treatment are available [20], [21], to our knowledge, we identified no previously published patient-facing digital tool specifically designed to address anticipatory dental anxiety related to MRONJ concerns [22], [23].
Digital health interventions may help bridge this gap by delivering accessible, evidence-based information to support informed and confident patient decision-making. Importantly, anxiety related to MRONJ risk may be conceptually distinct from generalized dental anxiety. This conceptual pathway is consistent with the Common-Sense Model of Self-Regulation, which proposes that cognitive and emotional representations of a health threat, including its perceived consequences, controllability, and coherence, influence coping responses and health behavior [24].
In this context, uncertainty, overestimation of risk, and insufficient contextual information regarding a rare but serious adverse outcome may contribute to anticipatory anxiety related to MRONJ. Perceived susceptibility and worry are related but distinct constructs and may respond differently to risk communication interventions [25]. Within this framework, structured risk communication may influence emotional responses by improving the coherence and perceived controllability of the health threat, while perceived knowledge and emotional reassurance need not change in parallel.
Accordingly, anticipatory anxiety may change through reduced uncertainty or more contextualized threat appraisal without requiring a corresponding increase in perceived knowledge. Together, these concepts provide a theoretical rationale for evaluating whether structured, scenario-specific risk communication is associated with anticipatory anxiety, risk understanding, and communication-related outcomes in patients concerned about MRONJ.
To address this challenge, we developed Dental Confidence, a web-based patient education and risk communication tool informed by the 2022 position paper of the American Association of Oral and Maxillofacial Surgeons (AAOMS) [4]. Rather than evaluating clinical outcomes, this study focused on the psychosocial and communicative dimensions of MRONJ-related concern in dental care. We hypothesized that use of Dental Confidence would be associated with lower immediate anticipatory anxiety scores on the scenario-based VAS in the context of MRONJ risk.
Secondary objectives were to evaluate feasibility, acceptability, self-reported knowledge and communication confidence, and general dental anxiety, and to explore how participants perceived the role of personalized risk communication and reassurance in their emotional responses and communication about dental care.
Methods
Study design and setting
We conducted a prospective explanatory sequential mixed methods study to evaluate the feasibility, acceptability, and immediate patient-reported outcomes of Dental Confidence and to explore plausible explanations for the observed quantitative response patterns [26].
The study comprised two sequential phases:1.A quantitative single-arm evaluation with assessments before and immediately after intervention use to examine feasibility, acceptability, and changes in anxiety and self-reported knowledge outcomes, followed by2.A qualitative explanatory phase designed to explore participants' experiences and provide contextual explanations for the observed quantitative response patterns.
The study sequence was quantitative pretest (T0), application use, quantitative posttest (T1), purposive selection of participants based on quantitative response profiles, qualitative interviews and analysis, and final integration of the quantitative and qualitative findings.
Feasibility was assessed descriptively using intervention and assessment completion, withdrawals, and missing outcome data; acceptability was assessed using postintervention satisfaction. Quantitative response profiles informed purposive sampling for the qualitative phase, and quantitative and qualitative findings were integrated during interpretation through narrative integration and a joint display [26], [27].
The study was conducted at a single tertiary orthopedic center in Bangkok, Thailand, accredited by the Chulabhorn Royal Academy, which serves a high volume of older adults receiving long-term antiresorptive therapy. Reporting of the quantitative component was informed by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement, and reporting of the qualitative component was guided by the Standards for Reporting Qualitative Research (SRQR), with completed checklists provided as supplementary material [28], [29]. Because of the visible nature of the intervention, blinding of participants and investigators was not feasible.
Ethical considerations
The study protocol was reviewed and approved by the Human Research Ethics Committee of the Chulabhorn Research Institute (IRB No. 107/2568). All procedures were conducted in accordance with the Declaration of Helsinki.
All participants provided written informed consent prior to enrollment. The study was not prospectively registered because, at study initiation, the investigators considered it a minimal-risk, feasibility-oriented evaluation of a patient education and risk-communication intervention. Following editorial review and reconsideration of the ICMJE clinical trial definition, the study was retrospectively registered with the Thai Clinical Trials Registry (TCTR20260911017) on September 11, 2026.
Participants
Participants were recruited consecutively from the outpatient osteoporosis clinic. Inclusion criteria were age 50 years or older, a diagnosis of osteoporosis, and current or planned treatment with antiresorptive medications, including oral or injectable formulations. Exclusion criteria were cognitive impairment that precluded meaningful engagement with the digital tool or inability to use a smartphone independently.
All participants provided written informed consent prior to enrollment.
For the qualitative explanatory phase, five participants were purposively selected from the quantitative cohort to capture variation in medication type, baseline anxiety, and key response profiles, including participants with larger decreases in VAS scores, smaller decreases, little or no change in VAS scores, and low baseline VAS scores. The focused aim, participant specificity, and targeted interview and analysis strategy were consistent with the concept of information power, whereby smaller samples may be appropriate when participants provide highly relevant information [30]. The qualitative phase was designed to contextualize contrasting quantitative response patterns.
The intervention
Dental Confidence was implemented as a bilingual web-based application accessible without installation across common smartphone and tablet devices (Fig. 1a). It was developed to support patient education and risk communication regarding MRONJ in older adults receiving antiresorptive therapy.
The intervention provided scenario-specific, guideline-informed information in a format designed to be accessible, emotionally reassuring, and supportive of communication with clinicians.Fig. 1User interface of the “Dental Confidence” web application on a mobile device.(a) Welcome screen with bilingual language options and simplified navigation intended for older adults.(b) Medication selection screen using product images to support recognition and reduce recall burden.(c) Scenario-specific output presenting a guideline-informed fixed risk range, concise explanatory text, and a patient summary to support discussion with clinicians.Fig. 1
The content and decision rules were informed by the 2022 position paper of the AAOMS and the supporting evidence cited in the application [4]. The rule-based workflow first distinguished noninvasive from invasive dental procedures and then branched according to medication class. Treatment duration and selected additional risk modifiers were considered where applicable.
For example, for an invasive dental procedure in a participant receiving denosumab, the pathway displayed guidance to schedule the procedure approximately 3 to 4 months after the most recent dose and to resume denosumab 6 to 8 weeks after the procedure, together with a fixed MRONJ risk range of approximately 0.04% to 0.3%, rather than an extraction-specific risk estimate. Each pathway generated a guideline-informed recommendation, a fixed numerical risk range when available, and a brief summary to support discussion with clinicians (Fig. 2).
The displayed numerical estimates were fixed ranges encoded from the cited clinical sources and were not calculated as individualized probabilities using a predictive model. They were presented in language designed for patients to support understanding and discussion with clinicians. The tool was intended as a communication and decision aid rather than a formal clinical risk stratification system.Fig.
2Simplified guideline-informed communication workflow of the “Dental Confidence”.The schematic summarizes the principal information flow from patient and dental inputs through procedure and medication-specific pathways to the patient-facing output. To preserve readability, detailed branching rules and complete patient-facing output text are not shown. Within the invasive procedure pathways, bisphosphonate guidance considered treatment duration longer than 4 years and concomitant steroid use; denosumab guidance addressed timing relative to the last dose; romosozumab guidance considered concomitant steroid use; and teriparatide guidance provided routine continuation recommendations.
Each pathway generated plain language recommendations and risk information, including a fixed numerical range when available. These ranges were encoded from the cited clinical sources rather than calculated as individualized probabilities. The application was intended to support clinician discussion and not to replace clinical assessment.Fig.
2
The application and its rule-based algorithm were developed collaboratively by an orthopedic physician with expertise in osteoporosis care and two dentists. The clinical content, branching logic, and patient-facing language underwent iterative internal review against the cited clinical guidance. An external clinical review was subsequently conducted by a senior orthopedic specialist in osteoporosis care and a geriatrician, and the application was revised in response to their recommendations.
Before data collection, the application underwent informal user testing over approximately one month with a small group of healthcare personnel (physicians, dentists, nurses, and nursing assistants) and older lay users. Feedback informed refinements to medication identification, response options, navigation, numerical risk presentation, and visual design. Fewer than 10 individuals participated in this preparatory testing, and none were included in the subsequent study cohort (N = 51).
This preparatory testing was not a formal usability evaluation. The interface was designed with geriatric-friendly principles, including sequential task progression, simple language, and visual support to reduce cognitive burden. To facilitate accurate medication identification, the input phase included photographic images of commonly used antiresorptive medications (Fig.
1b). To support emotional reassurance, the results screen presented scenario-specific risk information using non-threatening visual elements and concise explanatory text. The tool also generated a brief patient summary card intended to support discussions with clinicians (Fig.
1c). To promote safe use, all result screens included a disclaimer stating that the output was a preliminary, guideline-informed aid and not a substitute for professional clinical judgment. Depending on the medication pathway, participants viewed five to eight screens, including the welcome and personalized result screens.
Participants accessed the intervention using their own smartphones or study-provided tablets. A brief face-to-face orientation was provided before use, and technical assistance was available on request. The tool was evaluated as a fixed-version intervention during the study period.
Application use time was not recorded separately from the complete quantitative session, which lasted approximately 15–20 min.
Phase 1: quantitative outcomes and measures
Quantitative outcomes were assessed immediately before application use (T0) and immediately after use (T1) (Supplementary File 3).
Primary Outcome: Anticipatory Anxiety (Scenario-Based VAS):
Anticipatory anxiety was assessed using a single-item, scenario-based Visual Analog Scale (VAS) ranging from 0 (no anxiety) to 100 (extreme anxiety). Participants responded to the standardized scenario, “Assume you require a tooth extraction tomorrow.” A single-item VAS was selected because it permits rapid assessment of transient anxiety with minimal respondent burden and has demonstrated associations with established measures of state and dental anxiety in oral surgery and dental settings [31], [32]. Because no validated instrument specifically measuring MRONJ-related anxiety was available, the VAS was used as a pragmatic measure of immediate anticipatory anxiety elicited by the prospect of tooth extraction in the context of MRONJ risk.
It was not interpreted as a validated measure of MRONJ-specific anxiety. General dental anxiety was assessed separately using the Modified Dental Anxiety Scale (MDAS).
Secondary Outcomes:-General Dental Anxiety: General dental anxiety was assessed using the MDAS [33]. At study initiation, no validated Thai-language version of the MDAS was available [34]. Therefore, the scale underwent forward and backward translation by two bilingual dental experts and was pilot tested in 10 patients for clarity and face validity.
Internal consistency was assessed separately using Cronbach's alpha, which was 0.81 in this pilot sample. Given the small pilot sample, this estimate was considered preliminary and was not interpreted as evidence of full psychometric validation.-Self-Reported Knowledge and Communication Confidence: A six-item questionnaire developed for this study assessed perceived knowledge of MRONJ-related dental precautions and confidence in communicating with clinicians. Five items addressed relevant dental, medication, and treatment factors, while one item assessed confidence in asking physicians or dentists about dental care.
Each item was rated from 1 to 5, yielding a total score from 6 to 30, with higher scores indicating greater perceived knowledge and communication confidence. The questionnaire underwent multidisciplinary review within the research team, external clinical review by an orthopedic specialist, review for clarity by a nurse, and pilot testing in 10 patients; formal psychometric validation and reliability testing were not performed.-User Satisfaction: Overall satisfaction with the application in providing information and supporting dental care planning was assessed immediately after use using a single item rated from 1 (not satisfied at all) to 5 (very satisfied).
Phase 2: qualitative explanatory phase
Semi-structured interviews were conducted in Thai by a nurse familiar with the clinical setting but outside routine clinical interactions, guided by core questions and optional follow-up probes (Supplementary File 4). Questions began with open stems before more targeted probes were used for clarification. Interviews were conducted in a private room, audio recorded with consent, lasted approximately 15 to 20 min, and were transcribed verbatim from the recordings.
Two orthopedic physician authors collaboratively reviewed the interview material using a descriptive qualitative content analysis approach. Relevant statements were identified and organized into explanatory categories reflecting participants' experiences with risk information, emotional responses, and communication with clinicians. The categories were discussed and refined through consensus and subsequently considered alongside individual quantitative response patterns during mixed-methods integration.
The analysts' involvement in intervention development and the potential influence of the clinical interview setting were considered during interpretation.
Study procedure
Upon arrival at the clinic, eligible participants were approached by a trained research assistant. After providing informed consent, participants completed baseline demographic data and T0 assessments (VAS, MDAS, and the self-reported knowledge and communication confidence questionnaire). Participants then accessed the Dental Confidence application via QR code on their personal smartphones or study-provided tablets.
A brief face-to-face orientation was provided to explain navigation, after which participants used the tool independently, with technical assistance available upon request.
Immediately after tool use, participants completed the T1 assessments and satisfaction item. Participants purposively selected for the qualitative phase were interviewed either on the same day or at a later scheduled time. The complete quantitative session lasted approximately 15 to 20 min; application use time was not recorded separately.
To reduce procedural variability, all participants completed the same assessment sequence, received the same brief orientation, and used a fixed-version intervention under similar clinic conditions.
Statistical analysis
No formal a priori sample size calculation was performed, and all consecutive eligible participants recruited during the study period were included. For the primary outcome, a sensitivity analysis based on the achieved sample indicated a minimum detectable standardized within-participant change of approximately Cohen's dᶻ = 0.40, assuming 80% power and a two-sided alpha of 0.05. This sensitivity analysis characterized the statistical sensitivity of the achieved sample and was not intended to establish clinical importance.
Descriptive statistics were used to summarize participant characteristics. Normality of the paired differences was assessed using Shapiro–Wilk tests and quantile–quantile (Q–Q) plots. Because the primary analysis focused on the mean within-participant change, paired t-tests were used to estimate mean paired differences and corresponding 95% confidence intervals.
Given the observed departures from normality in the paired differences, Wilcoxon signed-rank tests were additionally performed as nonparametric sensitivity analyses. Effect sizes were calculated as Cohen's dᶻ by dividing the mean paired difference by the standard deviation of the paired differences. All tests were two-sided, and statistical significance was defined as P < 0.05.
Results
Phase 1: quantitative results
Participant characteristics
A total of 51 eligible patients were recruited between December 2025 and January 2026. All participants completed the intervention and both assessments, with no withdrawals or missing outcome data. Intervention and assessment completion rates were therefore 100%.
The mean age was 70.3 years (SD 7.7), and 49 participants (96.1%) were female. Most participants were receiving injectable antiresorptive therapy only (36/51, 70.6%), while 8 (15.7%) were receiving oral therapy only, 4 (7.8%) had received both oral and injectable therapy, and 3 (5.9%) were receiving other osteoporosis medications. Participant characteristics are summarized in Table 1.Table 1Baseline characteristics of study participants (N = 51).CharacteristicValueAge (years) Mean (SD)70.3 (7.7) Range (Min-Max)50–88Gender, n (%) Female49 (96.1) Male2 (3.9)Osteoporosis Medication Type, n (%)a Injectable Therapy Only36 (70.6) Oral Therapy Only8 (15.7) Both Oral and Injectable4 (7.8) Other (e.g., Anabolic)3 (5.9)aCategories were mutually exclusive; participants receiving both oral and injectable therapy were classified in a separate category.
Assumption checks and sensitivity analyses
Shapiro–Wilk tests indicated departures from normality for the paired differences in VAS, knowledge and communication confidence, and MDAS scores (all P < 0.001). Wilcoxon signed-rank sensitivity analyses yielded P < 0.001 for VAS, P = 0.923 for self-reported knowledge and communication confidence, and P = 0.016 for MDAS. For MDAS, the Wilcoxon sensitivity result (P = 0.016) differed from the paired t-test result (P = 0.05).
Primary outcome: anticipatory anxiety (scenario-based VAS)
Immediately after application use, the mean VAS score was 44.80 (SD 33.03), compared with 53.24 (SD 32.62) at baseline. The mean paired change was −8.43 points (95% CI −13.16 to −3.70; P < 0.001), with a standardized paired effect size of Cohen's dᶻ = 0.50 (Table 2). The distribution of VAS scores before and after application use is shown in Fig.
3.Table 2Summary of Primary and Secondary Outcomes (N = 51).Outcome MeasurePre-Intervention Mean (SD)Post-Intervention Mean (SD)Mean Difference (95% CI)P-valueEffect size (Cohen's dᶻ)Primary OutcomeScenario-Based Anticipatory Anxiety in the Context of MRONJ Risk (VASb, 0–100)53.24 (32.62)44.80 (33.03)−8.43 (−13.16 to −3.70)< 0.001a0.50Secondary OutcomesSelf-Reported Knowledge and Communication Confidence Score (6–30)25.49 (4.12)25.43 (4.56)−0.06 (−1.02 to 0.90)0.900.02MDASd Score (5–25)11.63 (4.99)10.51 (5.10)−1.12 (−2.24 to 0.00)0.050.28User Satisfaction (1–5)N/A4.45 (0.81)N/AN/AN/AcMean differences were calculated as values immediately after application use minus values before application use; negative values indicate lower scores after use. Effect sizes are reported as absolute Cohen's dᶻ values, calculated by dividing the mean paired difference by the standard deviation of the paired differences.aP-values calculated using paired t-tests.bVAS: Visual Analog Scale.cN/A: not applicable.dMDAS: Modified Dental Anxiety Scale.Fig. 3Scenario-based anticipatory anxiety in the context of MRONJ risk before and immediately after application use (N = 51).Boxplots show the distribution of Visual Analog Scale scores from 0 to 100.
The central line represents the median, the black square represents the mean, the box represents the interquartile range, and the whiskers represent the minimum and maximum values. Individual observations are overlaid as gray dots. Mean scores were 53.24 before use and 44.80 immediately after use, with a mean paired difference of −8.43 points (95% CI −13.16 to −3.70; paired t-test, P < 0.001).Fig.
3
Secondary outcomes
Secondary outcomes are presented in Table 2.–General Dental Anxiety (MDAS): Mean MDAS scores changed from 11.63 (SD 4.99) at baseline to 10.51 (SD 5.10) after application use, with a mean paired change of −1.12 points (95% CI −2.24 to 0.00; P = 0.05).–Self-Reported Knowledge and Communication Confidence: Mean self-reported knowledge and communication confidence scores were similar before and after application use (25.49 [SD 4.12] vs 25.43 [SD 4.56]; mean paired change −0.06, 95% CI −1.02 to 0.90; P = 0.90).–User Satisfaction: Mean satisfaction was 4.45 (SD 0.81) on the 5-point scale, and 46 of 51 participants (90.2%) rated their satisfaction as 4 or 5.
Phase 2: qualitative results and explanatory categories
Five participants were interviewed to contextualize contrasting quantitative response patterns. Descriptive qualitative content analysis identified four explanatory categories: reassurance through clear risk framing, emotional comfort from visual design, contextualization of numerical risk, and support for communication with clinicians. One participant provided a contrasting account in which persistent anxiety centered primarily on anticipated pain and the dental procedure rather than concern about MRONJ risk.
Participant-level quantitative patterns, illustrative quotations, and integrated interpretations are presented in Table 3.Table 3Joint display integrating quantitative VAS response patterns and qualitative findings following use of the Dental Confidence.Quantitative Pattern (Phase 1)Qualitative Finding (Phase 2)Illustrative QuoteIntegrated InterpretationLarge decrease in VASa score (90 to 20; Δ = −70 points)Reassurance through clear risk framing“I just needed to know where I stood. Once the app explained it clearly, I felt relieved.” (Participant #23)Reassurance and clearer interpretation of risk may help contextualize the observed VAS change, rather than additional information acquisition alone.Decrease in VAS score with high satisfactionEmotional comfort from non-threatening visual design“The cartoon doctor made me feel calm. It felt like the doctor was telling me it's okay.” (Participant #20)Friendly visual elements were described as providing emotional comfort and supporting reassurance.No change in VAS despite high satisfactionContrasting case: Persistent procedure-related anxiety“I understand the risk is low, but I'm still afraid of the pain and the needle.” (Participant #28)Persistent anxiety appeared to relate mainly to pain and procedural fear rather than MRONJb risk.Moderate decrease in VAS scoreQuantitative risk contextualization“Seeing the number 0.05% made me realize how rare it really is.” (Participant #06)Participants described numerical risk information as helping place concerns shaped by social media in context.Low baseline anxiety, high satisfactionSupport for communication with clinicians“I would use this when talking with the dentist so I can explain my situation more clearly.” (Participant #41)For some participants, the tool was described primarily as a communication aid rather than as a tool for reducing anxiety.This joint display integrates participant-level quantitative response patterns from Phase 1 with qualitative findings from Phase 2, comprising four explanatory categories and one contrasting case.
The qualitative findings provide contextual explanations for variation in the observed VAS response patterns and should not be interpreted as evidence of causality. Illustrative quotations were translated from Thai into English for reporting.aVAS: Visual Analog Scale.bMRONJ: Medication-related osteonecrosis of the jaw.
Category 1: reassurance through clear risk framing
Participants with larger decreases in VAS scores emphasized the importance of receiving clear, individualized information about MRONJ-related risk in relation to dental care. Rather than seeking extensive technical detail, these individuals valued concise and structured guidance, which they described as reducing uncertainty and helping them interpret their situation more confidently. Participants described this form of reassurance as helping them move from ambiguity toward greater emotional clarity, which they associated with reduced anxiety.
Category 2: emotional comfort from non-threatening visual design
Several participants described the application's visual elements, such as friendly illustrations and non-alarmist presentation of risk, as contributing to emotional comfort. These design features were perceived as reducing fear and cognitive burden, particularly among older adults with visual limitations or difficulty processing dense textual information. Participants described the calming visual environment as facilitating engagement and reinforcing feelings of safety.
Category 3: quantitative risk contextualization and correction of misperceptions
For some participants, numerical representation of MRONJ helped place concerns shaped by social media or anecdotal information in context. Participants described the displayed risk ranges, presented alongside information relevant to their medication and dental procedure, as helping them place MRONJ risk in context, reconsider catastrophic interpretations, and feel less inclined to avoid dental care because of fear.
Category 4: support for communication with clinicians
Participants with low baseline anxiety described the application as a useful support for conversations with dental providers. They described the patient summary card as enhancing communication confidence by helping them organize and convey relevant information during clinical encounters. In this context, participants viewed the tool less as a source of independent decision making and more as a communication aid that supported discussion between patients and clinicians.
Contrasting case: persistent procedure-related anxiety
One participant showed little change in VAS score after application use despite reporting awareness that MRONJ risk was low. The participant described persistent anxiety centered on anticipated pain, needles, and the extraction procedure itself. This contrasting account illustrates a response pattern in which procedural fear remained prominent, unlike accounts emphasizing reassurance from MRONJ risk information.
Integration of quantitative and qualitative findings
Integration of participant-level quantitative patterns with the qualitative findings showed that larger decreases in VAS scores were accompanied by descriptions of reassurance, clearer risk understanding, and supportive visual presentation. Participants with low baseline anxiety emphasized communication support rather than changes in anxiety, whereas the participant with persistent anxiety described concerns centered on pain and the dental procedure. Together, these findings offered contextual explanations for variation in the quantitative response patterns without establishing causality, as shown in Table 3.
Discussion and conclusion
Discussion
This study examined the psychosocial and communicative dimensions of MRONJ-related concern in older adults receiving antiresorptive therapy. Complete intervention and assessment data, together with high participant satisfaction, supported the feasibility and acceptability of Dental Confidence under supported outpatient conditions. Use of the application was associated with lower immediate VAS scores for anticipatory anxiety in the context of MRONJ risk, with a mean paired change of −8.43 points on the 100-point scale.
However, no validated minimal clinically important difference is available for this scenario-based VAS measure. The statistical significance and standardized effect size, therefore, should not be interpreted as establishing clinical importance, and a clinically meaningful responder threshold was not applied. The integrated qualitative findings offered plausible contextual explanations for the observed response patterns, particularly reassurance, risk contextualization, visual comfort, and support for communication with clinicians.
A notable finding was that mean VAS scores were lower immediately after application use, while self-reported knowledge and communication confidence scores remained stable. In this cohort, perceived knowledge regarding MRONJ-related risk and prevention appeared relatively high at baseline, which may have limited the potential for measurable improvement. The mixed methods findings helped contextualize this pattern: rather than indicating additional factual knowledge, participants described reassurance, reduced uncertainty, and clearer interpretation of risk information, consistent with the distinction between deliberative and affective responses to health risk [25].
This pattern is also consistent with the Common-Sense Model of Self-Regulation, in which cognitive and emotional representations of a health threat may influence coping responses [24]. As summarized in Table 3, participants with the largest decreases in VAS scores described the value of clear, individualized information that helped them understand the relevance of MRONJ-related risk to their own dental situation. Other explanatory categories included emotional comfort from nonthreatening visual design and numerical risk contextualization, which participants described as helping them reconsider concerns shaped by social media exposure [13], [14].
For some participants with low baseline anxiety, the tool was described primarily as a support for communication with clinicians rather than as a tool for reducing anxiety.
Interpretation of the VAS findings requires caution because the standardized tooth extraction scenario may have captured fear of pain, needles, or dental procedures in addition to MRONJ-related concern. Consistent with this possibility, general dental anxiety measured using the MDAS [33] showed only a modest change, while one participant with persistent anxiety attributed it mainly to longstanding procedural fear. This contrasting account highlights that anticipatory anxiety in the tooth-extraction scenario may involve procedural concerns that are distinct from uncertainty about MRONJ risk.
At study initiation, no validated Thai-language version of the MDAS was available to our knowledge. A validated Thai-MDAS has since been published [34]. Accordingly, the forward-backward translation and preliminary pilot testing used in the present study should be considered study-specific, and future studies should use a formally validated Thai version.
Our findings align with previous patient-oriented communication and decision tools in osteoporosis care [20], [21], [23], while extending this approach to anxiety surrounding a feared treatment-related adverse event and dental care. The interactive and scenario-specific nature of the intervention appeared to support reassurance and communication in settings where routine verbal counseling is often brief or inconsistent [11], [15]. Broader research on digital oral health interventions for older adults has identified potential benefits for access, education, motivation, and self-care, together with barriers including limited digital literacy, technical difficulties, and reliance on caregivers [35].
Research on digital interventions addressing anxiety among older adults similarly emphasizes ease of use, tailoring to individual needs, and the availability of human support [36]. The bilingual visual presentation, brief orientation, and availability of technical assistance in the present study may have supported engagement. Nevertheless, the high completion and satisfaction observed in a supported tertiary clinic, among participants eligible for independent smartphone use and with technical assistance available, should not be assumed to generalize to independent use or to older adults with limited access to or familiarity with digital technology.
Alternative explanations for the observed pre-post difference in VAS scores cannot be excluded. Repeated assessment, reassurance arising from interaction with research or clinical staff, Hawthorne effects, regression to the mean, and the structured clinical setting may have contributed to the observed change. Interviews conducted within the clinical setting may also have encouraged socially desirable responses.
The qualitative categories should therefore be interpreted as plausible contextual explanations rather than evidence of causality.
This study has several strengths. It evaluated a patient-centered digital intervention designed specifically for MRONJ-related concern in older adults, and the explanatory sequential mixed methods design linked quantitative response patterns with participants' experiences. Several limitations should also be acknowledged.
The study was retrospectively rather than prospectively registered. Therefore, a public registry record was not available before participant enrollment, although the complete mixed-methods protocol had received IRB approval before enrollment began. The single-arm design and absence of a control group preclude causal inference, while the modest sample and absence of a formal a priori sample size calculation require cautious interpretation.
Immediate assessment after application use precludes conclusions about durability, subsequent dental attendance, or effects on clinical decisions. The primary VAS was a single-item, scenario-based measure, and the study-specific knowledge and communication confidence questionnaire had not undergone formal psychometric validation. Eligibility requiring independent smartphone use, together with the lack of formal assessment of digital literacy, education, treatment duration, and previous dental experience, may have introduced selection bias and limited contextual interpretation.
Generalizability is further constrained by the predominantly female sample, single tertiary center, and small purposively selected qualitative sample. Finally, the informal pre-study user testing did not constitute a formal usability assessment, and the rule-based workflow requires further evaluation as clinical guidance evolves. Future studies should use controlled, preferably multicenter designs with longer follow-up and recruit participants with more diverse sex and digital literacy profiles.
They should incorporate formal usability assessment and validated measures capable of distinguishing MRONJ-related concern from general dental and procedure-related fear, and examine outcomes such as dental attendance, communication with clinicians, medication or treatment decisions, and longer-term anxiety.
Innovation
The innovation of this work can be considered at two levels. At a specific level, this study introduces Dental Confidence as a bilingual, geriatric-friendly, web-based tool designed to deliver structured, scenario-specific MRONJ-related risk communication directly to patients. The tool integrates medication and dental procedure recognition, a rule-based communication workflow, scenario-specific risk contextualization, emotionally nonthreatening visual design, and a patient summary card to support discussions with clinicians.
To our knowledge, we identified no previously published patient-facing digital tool specifically designed to address anticipatory dental anxiety related to MRONJ concerns in older adults receiving antiresorptive therapy [22], [23]. The contribution therefore lies not only in the use of a digital platform, but also in the design of a reassurance-oriented communication intervention tailored to older adults receiving antiresorptive therapy.
At a broader level, this work extends digital health communication methods into an under-addressed clinical and cultural context, specifically anticipatory dental anxiety related to MRONJ concerns among older adults in a Thai outpatient setting. A further dimension of innovation lies in the evaluation approach: the explanatory sequential mixed methods design [26] linked immediate quantitative response patterns with participants' accounts of reassurance, contextualized risk understanding, and communication support. These qualitative explanatory categories may help contextualize variation in the observed VAS response patterns despite little change in perceived knowledge and communication confidence.
This work therefore offers an example of how e-health tools may be adapted to support communication around a specific psychosocial barrier to dental care while remaining closely connected to real-world outpatient practice.
Conclusion
In conclusion, a web-based risk communication tool addressing MRONJ-related concern was feasible and highly acceptable in a supported outpatient setting and was associated with lower immediate scenario-based VAS scores for anticipatory anxiety in the context of MRONJ risk among older adults receiving antiresorptive therapy. The integrated qualitative findings suggested that reassurance, clearer risk contextualization, and communication support may help explain variation in the observed response patterns. These preliminary findings require confirmation in controlled studies with longer follow-up.
Declaration of generative AI and AI-assisted technologies in the manuscript preparation process
During the preparation of this work, the authors used Gemini (Google LLC, Mountain View, CA, USA) to support language editing and improve readability. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.
Clinical trials
The results of this clinical trial and any associated work have been posted in a registry.
This clinical trial was registered with number TCTR20260911017 (Thai Clinical Trials Registry).
CRediT authorship contribution statement
Rathapoom Suwanaratana: Writing – original draft, Visualization, Project administration, Formal analysis, Data curation, Conceptualization. Thanatporn Rujiraphan: Writing – review & editing, Writing – original draft, Investigation, Data curation. Patcharanun Wiphakphongpakorn: Writing – review & editing, Investigation, Data curation.
Nuttawut Chanalithichai: Writing – review & editing, Validation, Supervision, Methodology, Formal analysis, Conceptualization.
Ethics approval
The study protocol was reviewed and approved by the Human Research Ethics Committee of the Chulabhorn Research Institute (IRB No. 107/2568). All procedures were conducted in accordance with the Declaration of Helsinki.
All participants provided written informed consent prior to enrollment. The study was not prospectively registered because, at study initiation, the investigators considered it a minimal-risk, feasibility-oriented evaluation of a patient education and risk-communication intervention. Following editorial review and reconsideration of the International Committee of Medical Journal Editors (ICMJE) clinical trial definition, the study was retrospectively registered with the Thai Clinical Trials Registry (TCTR20260911017) on September 11, 2026.
Ethics declaration
Written informed consent to take part in the study and to publish the article has been obtained from all participants or their legal representatives. The privacy rights of participants have been observed.
This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was conducted in accordance with Declaration of Helsinki (World Medical Association). This study was approved by the Human Research Ethics Committee of the Chulabhorn Research Institute.
(Approval No. 107/2568)
Funding
No external financial support or grants were received from any public, commercial, or not-for-profit entities for the research, authorship, or publication of this article.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.