The project was approved by the Ethics Committee of the Universidad de La Frontera (decision 109/22). All participants provided written informed consent, and strict confidentiality protocols were followed, including anonymization prior to analysis.

This cross-sectional exploratory analytic study is based on the integrated analysis of data collected at the conclusion of a community-based oral health intervention conducted in the La Araucanía Region of Chile. The intervention corresponded to the regional implementation of the previously described hybrid dental care model and was delivered through a mobile clinical unit serving rural and urban municipalities. Services included geriatric dental assessment and treatment, preventive and restorative care, prosthetic evaluation, and referral management according to treatment complexity. Clinical information, coordination of care, and specialist consultation were supported through the TEGO digital platform following standardized protocols described in previous publications [29,45,47,48].

The La Araucanía Region is distinguished by its high proportion of Indigenous Mapuche population (34.2%) [21]. The region continues to experience persistent socioeconomic disadvantages, including elevated poverty rates, a predominantly rural context, and pronounced health inequities, particularly reflected in limited access to preventive and routine oral health care services [49,50].

In response to these challenges, a community-based oral health intervention program was implemented with funding from a national Presidential Grant (exempt resolution number 472). Initial recruitment was conducted by telephone using databases provided by the regional branch of the National Senior Citizen Service, in collaboration with local community organizations. Municipalities were selected based on logistical feasibility and implementation criteria rather than probabilistic sampling. Program eligibility included adults aged 60 years or older, residing in the region, and physically and cognitively able to attend dental appointments.

Between January and September 2023, a total of 480 older adults received care through this program, which delivered services through scheduled outreach visits using a mobile dental clinic in 4 municipalities: 2 rural (Cholchol and Los Sauces) and 2 urban (Temuco and Angol). The intervention was implemented in coordination with municipalities and community organizations, including the Araucanía Regional Federation of Older Adults, which brings together municipal-level unions of older adult clubs, as well as local community groups such as church organizations.

This study was based on clinical and assessment data collected during program implementation and recorded through the TEGO digital platform. For the present study, eligibility criteria included being a community-dwelling, functionally independent older adult, defined as a Barthel Index score ≥90. Consequently, 9 institutionalized individuals were excluded, resulting in a final sample of 471 participants. Figure 1 presents the detailed participant flow, including eligibility assessment, application of inclusion and exclusion criteria, and the final analytic sample.

The WHO-CSDH framework informed the research question, variable selection and classification, and interpretation of findings by conceptualizing oral health as shaped by processes of social stratification and the unequal distribution of power, resources, and opportunities [6]. This framework distinguishes between structural determinants—such as socioeconomic and political context and axes of social stratification, including education, ethnicity, and place of residence—and intermediary determinants, which encompass material conditions, health-related behaviors, psychosocial factors, and access to health systems. Within this framework, ethnicity was conceptualized as a structural determinant reflecting social position, guiding comparisons of oral health outcomes across groups.

The use of the WHO-CSDH framework is particularly relevant in this study context, as it allows oral health inequalities to be examined beyond individual-level risk factors by situating them within broader social and systemic conditions. This is especially important in Indigenous populations, where health disparities are shaped by historical, social, and territorial inequities that are not fully captured by conventional epidemiological approaches focused on proximal determinants [9]. By organizing variables into structural determinants, intermediary determinants, and oral health outcomes, this framework supports a more comprehensive assessment of how health inequalities are patterned across social groups and how different levels of determinants coexist.

Consistent with the WHO-CSDH model, structural and intermediary determinants are theorized to be hierarchically related through pathways linking social position to health outcomes. However, given the cross-sectional design, the present study examines the distribution of these determinants and their associations with oral health outcomes without inferring causal relationships.

Clinical data were collected by 2 final-year dental interns under direct supervision of 2 experienced dentists (CAA and FM-S), using a standardized clinical protocol under uniform examination conditions. Examinations were conducted by the same team that implemented the initial pilot intervention in the Mapuche community, and the regional phase began 1 month later, maintaining the same examiners, instruments, and diagnostic criteria.

Prior to implementation, examiners underwent structured theoretical and supervised clinical training to apply the complete TEGO clinical record, which includes medical-geriatric, odontogeriatric, behavioral, and prosthodontic assessment modules. Training covered caries detection, periodontal examination, prosthetic evaluation, and standardized collection of self-reported variables (eg, oral hygiene practices and service use).

Formal calibration with assessment of interexaminer and intraexaminer reliability was conducted for dental caries using the International Caries Detection and Assessment System (ICDAS). Interexaminer and intraexaminer κ values were 0.66 and 0.78-0.85, respectively [29]. Periodontal and prosthetic assessments were recorded using standardized objective diagnostic criteria under direct supervision to ensure procedural consistency.

All findings were recorded directly into the TEGO [45,47], ensuring standardized data capture and quality control procedures.

Guided by the WHO-CSDH framework, variables were classified as structural determinants, intermediary determinants, or oral health outcomes.

Structural determinants were defined as variables reflecting the broader socioeconomic context that shapes health equity through social stratification. These included: sex (male or female, self-reported); age, recorded as a continuous variable in years and categorized as 60-69 years, 70-79 years, or ≥80 years; and ethnic self-identification, classified as Mapuche (based on self-reported Indigenous affiliation) or non-Mapuche. Educational attainment was dichotomized as no formal or only primary education (≤8 years) vs secondary or higher education (>8 years).

Socioeconomic status was approximated using Chile’s Social Registry of Households (RSH), a national administrative classification used to target social programs. The RSH assigns households to 7 socioeconomic percentiles based on an algorithm that integrates income sources, household composition, and indicators of vulnerability (eg, age, disability, and access to goods and services) [51]. For this study, households were dichotomized into the most vulnerable group (0%-40%) vs higher percentiles (>40%).

Additional structural variables included place of residence, classified as urban or rural according to municipality-level administrative definitions [52], and health insurance coverage, categorized according to the Chilean health system as the National Health Fund (FONASA), the health systems for the armed forces and police (National Defense Social Security Fund [CAPREDENA] and Carabineros de Chile Social Security Directorate [DIPRECA]), the Comprehensive Health Care and Reparation Program for Victims of Human Rights Violations (PRAIS), private health insurance institutions (ISAPREs), or no health coverage.

Descriptive analyses were conducted to characterize the study population. Categorical variables were summarized as frequencies and percentages, and continuous variables as means with SDs or 95% CIs, as appropriate. Bivariate analyses compared structural determinants, intermediary determinants, and oral health indicators by ethnicity (Mapuche vs non-Mapuche). Associations between categorical variables were assessed using the Pearson chi-square test, with P values of <.05 considered statistically significant.

Guided by the WHO-CSDH framework, bivariate and multivariable logistic regression analyses were then conducted to examine the associations between structural and intermediary determinants and selected outcomes. Three groups of outcomes were evaluated: clinical oral health outcomes (edentulism, nonfunctional dentition, periodontitis, and caries), access-related outcomes (time since the last dental visit, presence of fixed dental rehabilitation, and clinically unmet denture need), and treatment needs (restorative and prosthetic). Separate multivariable logistic regression models were fitted for each binary outcome (9 models in total) to estimate adjusted associations between structural and intermediary determinants and the outcomes.

Missing data were present in a small proportion of observations for several variables (<10%), and complete-case analysis was therefore considered appropriate given the overall low level of missingness. The proportion of missing data for periodontitis was higher (51/386, 13%). For this outcome, missingness was not significantly associated with most evaluated covariates (all P>.05), although excluded participants were older, suggesting a potential age-related component. For all other variables, no substantial differences were observed between included and excluded participants across the evaluated sociodemographic characteristics. Missing periodontal data most likely reflects incomplete clinical examinations rather than item nonresponse, as periodontal assessment requires full-mouth probing in dentate participants. Such incomplete examinations may occur due to participant refusal, logistical constraints during fieldwork, or medical conditions preventing the assessment.

Because some outcomes were only applicable to specific subgroups (eg, dentate individuals or denture users) and some covariates had incomplete observations, the number of observations varied across models; therefore, logistic regression analyses were conducted using complete-case data for each model. Covariates were selected a priori based on the conceptual framework of social determinants of health, rather than through data-driven selection procedures. Indicators of health system access were conceptualized as intermediary determinants; however, in specific models they were also analyzed as outcomes to explore inequalities in access to care. In models examining oral health status outcomes, time since the last dental visit was included as a covariate to account for differential exposure to dental care. Models were also adjusted for the number of natural teeth present, when applicable, given its potential role as a confounder. Results are presented as odds ratios (ORs) with 95% CI, and all analyses were conducted using Stata (version 18; StataCorp LLC).

A total of 471 community-dwelling older adults were included in the analysis (Figure 1). Of these, 24.2% (114/471) self-identified as Mapuche. The mean age of the sample was 72.8 (SD 7.7) years, with Mapuche participants being significantly younger than non-Mapuche counterparts (P=.006).

Univariate and bivariate descriptive statistics for the study variables are presented in Table 1. Overall, 59.9% (282/471) of participants were female, with no significant differences between Mapuche and non-Mapuche participants (P=.62). A total of 126 participants were not registered in RSH, as enrollment in this administrative system is voluntary in Chile. Among those registered (n=345 of 471), most (301/345, 87.2%) were classified within the ≤40% bracket, indicating high socioeconomic vulnerability. No significant differences were observed between Mapuche and non-Mapuche participants (P=.10). Regarding health coverage, most participants (405/471, 85.9%) were enrolled in FONASA, 9.8% (46/471) had no health insurance, and 4.2% (20/471) were covered by other schemes (ISAPRE, CAPREDENA/DIPRECA, and PRAIS). Coverage differed by ethnicity; nearly all Mapuche participants (111/114, 97.4%) had FONASA, while non-Mapuche participants showed more variation, including a higher proportion without insurance (45/357, 12.6%; χ²₂=16.6; P<.001). Compared with their non-Indigenous counterparts, Mapuche individuals had lower educational attainment and resided predominantly in rural municipalities.

After conducting bivariate analyses for all selected outcomes and covariates, multivariable logistic regression models were fitted to estimate the adjusted effect of each independent variable. Statistically significant bivariate associations are reported in the corresponding table. Given the exploratory nature of the analysis, all predictor variables were included in the multivariable models regardless of their bivariate significance in order to control for potential confounding and obtain adjusted estimates.

In the multivariable logistic regression models, structural determinants exhibited the most consistent associations with oral health outcomes. Age and educational attainment were significantly associated with tooth loss: participants aged 80 years or older had higher odds of nonfunctional dentition (OR 5.55, 95% CI 2.34-13.08; P<.001) and complete edentulism (OR 3.75, 95% CI 1.62-8.68; P=.002) compared with those aged 60 to 69 years, while ≤8 years of education was linked to increased odds of nonfunctional dentition (OR 4.28, 95% CI 2.34-7.88; P<.001) and complete edentulism (OR 3.04, 95% CI 1.52-6.09; P=.002). Women had higher odds of edentulism (OR 2.38, 95% CI 1.24-4.58; P=.009) but lower odds of periodontitis (OR 0.53, 95% CI 0.31-0.90; P=.02). Mapuche ethnicity was associated with higher odds of cavitated caries (OR 2.40, 95% CI 1.25-4.63; P=.009) and periodontitis (OR 2.43, 95% CI 1.21-4.88; P=.01). Rurality showed no significant associations (Table 2).

Among intermediary determinants, behavioral factors appeared to be the most relevant. Specifically, lack of interproximal hygiene was linked to tooth loss outcomes, while consumption of sugary snacks was associated with increased odds of periodontitis (OR 1.86, 95% CI 1.11-3.12; P=.01; Table 2). Psychosocial factors showed no significant associations, and biological factors had limited associations, with dry mouth linked to nonfunctional dentition (OR 2.08, 95% CI 1.20-3.63; P=.009) and cavitated caries (OR 1.78, 95% CI 1.02-3.13; P=.04). Regarding dental care access, having received care in the past 2 years was associated with lower odds of total edentulism (OR 0.53, 95% CI 0.28-0.99; P=.048). These multivariable results, including the 4 oral health outcomes, are summarized in Figure 2.

In the multivariable analysis, variables related to dental care were examined (full results are provided in Multimedia Appendix 1). Time since the last dental visit was associated with toothbrushing frequency (OR 1.92, 95% CI 1.02-3.62; P=.04) and interproximal hygiene practices (OR 1.90, 95% CI 1.13-3.19; P=.01). Rural residence was associated with higher odds of unmet denture need (OR 2.86, 95% CI 1.31-6.24; P=.008), whereas brushing frequency (OR 0.25, 95% CI 0.11-0.57; P<.001) was associated with lower odds.

In the adjusted model, women were more likely to have received fixed prostheses (OR 3.46, 95% CI 1.51-7.94; P=.003), as were participants aged 80 years or older (OR 5.05, 95% CI 1.68-15.17; P=.004). Conversely, participants with ≤8 years of education were less likely to have received fixed prostheses (OR 0.37, 95% CI 0.15-0.91; P=.03). Consumption of sugary snacks was also associated with higher odds of having received fixed prosthetic rehabilitation (OR 3.48, 95% CI 1.56-7.78; P=.002). Other determinants, including ethnicity, did not show significant associations with this outcome (Figure 3).

When treatment needs were analyzed, no structural determinants were significantly associated with these outcomes. However, higher consumption of sugary snacks between meals was linked to increased odds of requiring restorative treatment (OR 1.69, 95% CI 1.03-2.77; P=.03). In addition, current smoking was associated with higher odds of normative prosthetic treatment need (OR 3.49, 95% CI 1.3-9.31; P=.01; Figure 4). Full adjusted model results are presented in Multimedia Appendix 2.

This study highlights differences in both structural and intermediary social determinants of health between Indigenous (Mapuche) and non-Indigenous older adults in a sample from the La Araucanía Region of Chile. Compared with non-Indigenous participants, Mapuche individuals had lower educational attainment and were more likely to reside in rural areas. They also showed a higher prevalence of oral diseases, alongside a higher number of risk factors, including less frequent toothbrushing and higher sugar consumption. In addition, they had greater restorative treatment needs and reduced access to complex dental care services.

When examining the independent influence of structural and intermediary determinants on various oral health outcomes, structural factors showed more consistent associations than intermediary ones, in line with the WHO-CSDH framework. The associations observed for these variables may be consistent with the cumulative influence of structural disadvantage across the life course, whereby intermediary determinants in older age could reflect the long-term manifestation of social inequalities. Although our cross-sectional design does not allow direct assessment of life-course processes, longitudinal cohort research has demonstrated that early socioeconomic conditions shape oral health beliefs and behaviors, which subsequently influence adult clinical outcomes and quality of life, supporting the plausibility of cumulative pathways [73].

Notably, the pattern of associations varied depending on the outcome assessed. For instance, female sex was significantly associated with a higher prevalence of edentulism while showing lower odds of periodontitis. Similarly, lower educational attainment was associated with increased prevalence of nonfunctional dentition and edentulism; however, it showed no significant association with caries or periodontitis. For these latter outcomes, ethnicity showed the most consistent association. In particular, Mapuche participants had more than twice the odds of periodontitis compared with non-Mapuche counterparts (OR 2.43, 95% CI 1.21-4.88; P=.01). These findings are in line with international evidence reporting a 35% higher prevalence of periodontitis among Indigenous populations compared with non-Indigenous groups [2].

Regarding caries, Chilean adults aged 65 to 74 years exhibit one of the highest burdens worldwide (57,729 per 100,000) [74]. In our sample, all participants had a history of caries experience, and Mapuche participants showed a significantly higher prevalence of untreated (cavitated) caries compared with their non-Indigenous peers. This association remained significant in the multivariable model.

When examining variables related to access to dental care, no significant differences were observed in the frequency of the last dental visit between ethnic groups. However, the lower prevalence of complex treatments, such as dental implants and fixed prosthetic rehabilitations (eg, crowns and bridges), along with higher restorative treatment needs, including active caries and defective restorations, among Indigenous participants suggests a potential disparity in the quality of dental care received.

Ethnicity was significantly associated with 2 of the 9 outcomes in the multivariable models, a pattern similar to that observed for education, sex, and age. Within the WHO-CSDH framework, ethnicity is conceptualized as a structural determinant that interacts with socioeconomic, territorial, and cultural conditions, and is linked to cumulative disadvantages in health [6,9,75]. Historical and structural inequities have been documented as shaping access to services, use of dental prostheses, and the persistence of restorative treatment needs [4,5,76]. These findings highlight that ethnicity, alongside sex, age, and educational attainment, functions as a key axis of social stratification associated with differences in disease distribution and patterns of access and quality of treatment, rather than merely representing background characteristics.

Structural determinants are understood to shape differential exposure to intermediary determinants, including behavioral and biological risk factors [6,75]. In our study, unhealthy lifestyle behaviors, particularly the consumption of sugary snacks between meals, were more prevalent among Mapuche participants and were also linked to periodontitis and restorative treatment needs. These patterns are consistent with broader dietary changes reported in Indigenous populations, where colonization and globalization have contributed to the loss of traditional food systems, with a shift from subsistence diets toward store-bought and processed foods rich in refined sugars [77]. Structural factors, including corporate determinants of health such as the marketing of sugary products, may further reinforce this trend [78]. These dietary changes have been consistently linked with poorer oral health outcomes, including caries and periodontal disease [77]. The higher prevalence of periodontitis and caries in Mapuche participants may be related to a combination of factors, including ethnicity, dietary practices, and a lower brushing frequency, even though brushing frequency was not a significant predictor in the multivariable model. This finding supports the WHO-CSDH framework, which emphasizes the complex and not yet fully understood pathways through which structural determinants translate into biological outcomes [75].

Intermediary determinants played a more limited but still relevant role. Adverse outcomes were associated with dental visits occurring more than 2 years ago, frequent consumption of sugary snacks, and reduced toothbrushing frequency, while dry mouth, a biological factor linked with polypharmacy and aging [79], was also linked to both caries and nonfunctional dentition. Interestingly, better oral hygiene practices were associated with more recent dental visits. Addressing this requires expanding preventive interventions, particularly dietary counseling and management of xerostomia, as well as integrating tailored oral health education at the time of prosthetic delivery to improve long-term outcomes and avoid premature failures.

The Chilean health system, organized under a mixed public-private model, provides coverage to approximately 80% of the national population through its main government-funded scheme, FONASA, which primarily focuses on essential curative oral health care [3-5,80]. In this study, health insurance type was not included as a covariate since most participants were enrolled in FONASA, resulting in limited variability in coverage. Nevertheless, despite formal insurance coverage, a high burden of oral health problems alongside important gaps in service provision was observed. Although dental visit rates within the past 2 years were comparable across groups, 54% (206/383) of participants required restorative treatment, and a substantial proportion of existing restorations were found to be defective. Access to prosthetic care also showed notable gaps. While the national health guarantee (Explicit Health Guarantees [GES]) provides removable dentures for individuals at age 60 years, and other programs extend this benefit to vulnerable adults aged older than 20 years, important unmet needs persist [65,81-84]. In this study, 2 prosthetic indicators were considered. Unmet prosthetic need referred to individuals requiring a prosthesis but not currently wearing one, serving as a proxy for limited previous access to prosthetic care. In contrast, prosthetic treatment needs included both individuals requiring a new prosthesis and those with existing prostheses requiring repair or replacement, reflecting the overall demand for prosthetic treatment and providing an indicator relevant for service planning. Nearly one-fifth of participants presented unmet denture needs, and among those who already had dentures, almost 80% required repair or replacement. These findings underscore the importance of ensuring the quality of removable dentures provided in both public and private settings, particularly within publicly financed dental care, where more than 300,000 dentures are delivered annually in Chile [65]. A considerable proportion of these prostheses remain unused by patients due to poor fit, discomfort, or insufficient follow-up and adaptation protocols, which are critical to achieving functional rehabilitation and patient adherence [65]. The limited provision of complex treatments, coupled with gender-, education-, and ethnicity-based disparities in the use of fixed prostheses, is indicative of persistent inequalities. Within our sample, access to oral rehabilitation and preventive care appears more limited among older adults characterized by structural disadvantages, including advanced age, lower educational attainment, rural residence, and Indigenous ethnicity. Taken together, these patterns suggest that insurance coverage alone does not guarantee equity in oral health.

An innovative aspect of this study was the implementation of a teledentistry ecosystem (TEGO) supported by a mobile clinic. Although TEGO was designed for the broader older population, its pilot implementation in a Mapuche community demonstrated feasibility, user satisfaction, and a culturally respectful approach through the incorporation of intercultural facilitators, suggesting potential for supporting access to care in underserved communities [29]. Within the WHO-CSDH framework, strategies to reduce health inequities must address both structural and intermediary determinants of health [6,7]. In this context, technology-supported models of care may help mitigate structural barriers related to geographic isolation and limited service availability through remote consultation, screening, and triage, particularly in rural and underserved populations [44,85-88]. At the same time, digital health interventions may also influence intermediary determinants by strengthening oral health literacy, preventive behaviors, and patient engagement [85,89]. In this regard, the TEGO model illustrates a potentially equity-oriented strategy that combines digital tools with mobile service delivery to address barriers across the continuum of care. By facilitating access to primary oral health services (eg, restorative care and repair of removable prostheses) and, in some cases, enabling referral to secondary care for complex treatments such as implants, this approach may help reduce gaps in service use among populations facing structural disadvantages. Such strategies are particularly relevant for older adults living in rural or historically underserved settings and align with broader calls to action aimed at reducing health inequities and expanding access to care, including the objectives of United Nations Sustainable Development Goals, particularly goal 3 (good health and well-being) and goal 10 (reduced inequalities), while also contributing to the growing agenda of digital health equity [35,90-94].

The educational and ethnic inequalities observed in oral disease are consistent with prior evidence and support the need to account for differential disease burden in the design of public health policies [30,67,95]. Ongoing monitoring of these indicators may help identify service gaps and inform more equitable interventions. Even though a “one-size-fits-all” approach focused on individual treatment has done little to improve oral health outcomes at the population level, priority actions should focus on reducing structural disadvantages that influence access to and the cost of dental services, with a strong emphasis on equity [78]. This may include prioritizing socially disadvantaged groups and ensuring that treatments, such as dentures, are provided appropriately to individuals whose anatomical and functional characteristics allow effective use, while simultaneously establishing a subsidized system to improve access to more complex dental treatments for those who require them. Efforts should focus on strengthening the quality and follow-up of dental care, as well as implementing culturally tailored health promotion strategies for Indigenous and rural populations. Additionally, community-based delivery models, including mobile and digitally supported services, may contribute to reducing geographic and social inequities in access to care.

Certain limitations should be considered when interpreting these findings. First, the cross-sectional design precludes causal inference. The observed associations reflect patterns of co-occurrence rather than directional relationships, and reverse causation remains plausible. For example, poorer oral health may influence psychosocial domains such as depressive symptoms or social participation in older adults [96]. Accordingly, findings should be interpreted as evidence of social patterning rather than causal pathways.

Second, the sampling strategy and implementation context may have introduced selection bias and limited external validity. Participants were community-dwelling, functionally independent older adults recruited from a publicly funded, mobile, digitally supported care program implemented in municipalities selected for logistical feasibility. Participation required the capacity and willingness to attend scheduled dental visits, which likely underrepresents individuals with greater dependency, mobility limitations, or who are harder to reach—groups that may occupy more disadvantaged positions within the social gradient. This may have resulted in a healthier or more health-engaged sample, potentially underestimating the magnitude of oral health inequalities. At the same time, as an access-oriented intervention, the program may have attracted individuals with heterogeneous levels of perceived need, including both those seeking care for existing conditions and those engaging in preventive checkups, which may further influence the composition of the analytic sample.

Third, measurement limitations may have affected internal validity. Several intermediary determinants were self-reported, including dental visits, oral hygiene practices, current tobacco use, and chronic conditions used to define multimorbidity. These measures are susceptible to recall and social desirability bias, and the absence of detailed exposure assessment (eg, pack-years of smoking) or verification through clinical records may have introduced measurement error. Clinical outcomes were assessed by trained examiners using standardized protocols and direct supervision; however, periodontal and prosthetic assessments were more susceptible to field variability than caries, which was formally calibrated. These factors may have affected the precision and consistency of associations, particularly for periodontitis and for behavioral and biological variables.

Fourth, residual confounding cannot be excluded. Although the multivariable models were adjusted for the number of natural teeth as a key clinical factor, other potentially relevant confounders may not have been fully captured. In particular, the dataset lacked more granular measures of socioeconomic position (eg, wealth and income) and detailed health system characteristics, limiting the full operationalization of the structural dimension of the WHO-CSDH framework.

Fifth, missing data may have introduced bias. Although missingness was generally low, it was higher for periodontitis and polypharmacy. Incomplete periodontal examinations may occur under field conditions when full assessment is not feasible (eg, due to pain, tooth mobility, or time constraints during clinical evaluation). Missingness was not consistently associated with most measured characteristics, including sex, education, ethnicity, and rurality. However, for periodontitis, excluded participants were older than those included in the analysis, suggesting that examination feasibility may have differed by age. Given the overall low level of missingness and the limited evidence of systematic differences across most covariates, complete-case analysis was considered appropriate. Nonetheless, some degree of bias cannot be entirely excluded.

Finally, the absence of an a priori sample size calculation and the low frequency of some outcomes (eg, implants or complex rehabilitations) resulted in limited statistical precision for certain estimates, as reflected in wide CIs. Also, the study may not have been specifically powered to detect all potential differences between Mapuche and non-Mapuche participants; therefore, subgroup analyses should be interpreted cautiously.

Taken together, these findings are primarily applicable to community-dwelling, functionally independent older adults in rural and urban municipalities of La Araucanía, including both Mapuche and non-Indigenous individuals. Extrapolation to populations with higher levels of dependency or barriers to access, other regions of Chile, larger metropolitan areas, or Indigenous populations with different social or health care contexts should be undertaken cautiously. Moreover, as data were collected within a mobile, digitally supported care model specific to this program, generalizability to older adults receiving standard primary care services may be limited. This program-based implementation reflects real-world conditions of service delivery within its operational context, which may support ecological validity in terms of care processes and service pathways.

Despite these limitations, this study contributes to the literature by examining oral health outcomes among older adults through a structured social determinants framework that simultaneously models structural determinants, including ethnicity, and intermediary factors. Ethnicity was associated only with cavitated caries and periodontitis. This analytical approach allows observed ethnic differences to be interpreted within broader social and territorial conditions, rather than attributing disparities to ethnic group membership alone. It does not eliminate ethnic gradients but situates them within patterned social conditions that may shape exposure, risk, and access over time [9]. In the Chilean context, national health surveys include self-identified Indigenous affiliation and have examined several social determinants, such as education, income, and insurance status. However, oral health outcomes have rarely been analyzed using multivariable frameworks that incorporate ethnicity alongside other determinants.

The findings further indicate that oral health inequalities in older adults may operate through multiple and outcome-specific pathways. Structural determinants—particularly age and educational attainment—were strongly associated with cumulative indicators of tooth loss, such as nonfunctional dentition and complete edentulism. In contrast, behavioral factors showed closer associations with selected access to dental treatment and treatment needs. This pattern suggests that different levels of determination may be more relevant depending on whether the outcome reflects long-term accumulation of disadvantage (eg, tooth loss) or more proximal conditions and service use patterns.

The inclusion of both rural and urban participants highlights the contextual dimension of these inequalities. While rural residence was not consistently associated with all clinical outcomes, it was linked to selected treatment-related indicators, suggesting that territorial factors may influence patterns of rehabilitation and unmet need even when not uniformly affecting disease prevalence.

Future longitudinal studies incorporating more detailed measures of lifetime exposures, including behavioral factors such as smoking history, as well as health service trajectories and contextual factors, would help clarify the pathways linking structural determinants to oral health outcomes in later life. Research may also benefit from intersectional approaches capable of examining how structural determinants interact to shape oral health disadvantages. Additionally, evaluating the long-term effectiveness and equity of innovative service delivery models, such as teledentistry and mobile dental clinics, may provide valuable evidence to inform strategies aimed at reducing disparities. Mixed methods studies incorporating qualitative perspectives on social, cultural, and system-level barriers may further contribute to a more comprehensive understanding of the mechanisms underlying the observed patterns.

Oral health inequalities differed between Mapuche and non-Indigenous older adults in La Araucanía and were associated with both structural and behavioral determinants. Structural determinants, particularly lower education, female sex, older age, and ethnicity, showed consistent associations with oral health status, whereas dental care access reflected a combination of structural and behavioral factors. In contrast, current treatment needs were mainly associated with behavioral factors, including sugar consumption and tobacco use. Although Mapuche participants exhibited poorer oral health, higher restorative needs, and reduced access to complex dental care in bivariate analyses, ethnicity remained associated only with cavitated caries and periodontitis in adjusted models.

These findings underscore the importance of incorporating social determinants into oral health planning for underserved older adults, particularly in Indigenous and rural populations experiencing persistent social and territorial inequities. By applying the WHO-CSDH framework within a hybrid teledentistry and mobile clinic model, this study illustrates how digitally supported territorial care programs can generate clinically grounded population health data that help identify inequities in oral health status, access to care, and treatment needs. Integrating these approaches into oral health information systems and service planning may support more equity-oriented prioritization, resource allocation, and delivery of preventive, restorative, and prosthetic care for older adults in underserved territories.