The COVID-19 pandemic catalyzed unprecedented global expansion of telehealth services, with utilization increasing exponentially from pre-pandemic baselines. While digital health technologies promise to democratize healthcare access, they have simultaneously created and amplified new forms of healthcare disparity. This review distinguishes between health equity - the absence of avoidable, unjust differences in health outcomes across population groups, and health equality- the provision of identical services regardless of differing needs. These concepts, while related, require distinct policy responses. Our aim was to synthesize global evidence on equity challenges in telehealth implementation, identify key barriers to equitable access, and analyze strategies for promoting inclusive digital health delivery, with Israel examined as an in-depth case study illustrating broader themes. A narrative review methodology was employed, searching PubMed/MEDLINE and Google Scholar (January 2020-June 2025) using Boolean search strings combining telehealth-related terms with equity, access, and barrier keywords. Initial searches yielded 847 articles; 623 underwent title/abstract screening after deduplication; 156 met criteria for full-text review; 49 high-quality studies formed the final evidence base. Israel was integrated as an in-depth case study within this global synthesis. Significant telehealth utilization disparities persist globally across age, race/ethnicity, geography, and language groups. Technology access alone is insufficient to ensure equity: cultural barriers, digital literacy gaps, provider readiness, and systemic policy failures produce compound digital disadvantage. Israel's experience demonstrates that even in technologically advanced universal-coverage systems, Arab-Jewish and center-periphery disparities persist despite high smartphone ownership and internet access. Achieving telehealth equity requires multi-layered interventions addressing digital divides, cultural and linguistic barriers, and systemic inequalities. Success depends on equity-first design, multi-modal service delivery, provider training encompassing all available modalities, infrastructure investment, and continuous monitoring of disparate impacts.
Explore modern metastatic breast cancer care: longer survival, smarter symptom control, caregiver support, and steps to close racial treatment gaps.
Trust in public institutions is low, while the systems that shape health and wellbeing are often difficult to understand, navigate, and influence. This commentary argues that organizational health literacy (OHL) offers a practical and equity-oriented framework for redesigning those systems, not only in healthcare but across sectors such as education, housing, environmental planning, social services, and across local government. OHL shifts attention from individuals' ability to understand complex information to organizations' responsibility to make information, services, processes, and decisions clear, accessible, culturally responsive, and accountable. Building on public health frameworks that recognize the conditions that produce health, we propose that OHL can help operationalize health equity by reducing administrative burden, improving communication, strengthening navigation support, and creating feedback loops with the people most affected by institutional decisions. We trace the evolution of health literacy from an individual-focused concept to an organizational responsibility, identify theoretical frameworks that can support adoption across diverse sectors, and describe mechanisms such as accreditation, grant requirements, workforce training, equity audits, community accountability structures, and performance measurement. Examples illustrate how unclear communication, fragmented workflows, inaccessible public processes, and nonresponsive documentation can deepen inequities, and how OHL strategies could reduce these harms. Future research should test implementation strategies, measurement approaches, costs, and equity outcomes across sectors.
This study analysed the relationship between menstrual health and self-reported depression and anxiety in women and people who menstruate (PWM) (≥18 years) in Spain, from a critical gender perspective. An online survey was used to collect data in 32 healthcare centres in 2023. Descriptive and ordinal logistic regression models were performed with 1,404 participants. Menstrual bleeding of over 80 ml was associated with higher odds of more severe depression (aOR=1.61; 95%CI: 1.45-1.79; p < 0.001) and anxiety (aOR=1.67; 95%CI: 1.35-2.06; p < 0.001). Higher menstrual pain intensity was also linked to higher odds of more severe depression (aOR =  2.05; 95%CI: 1.82-2.31; p < 0.001) and anxiety (aOR=1.82; 95% CI: 1.43-2.30; p < 0.001). A higher frequency of premenstrual symptoms was associated with increased odds for more severe depression (aOR=14.54; 95% CI: 10.73-19.70; p < 0.001) and anxiety (aOR=12.95; 95%CI: 9.10-18.44; p < 0.001). In addition, higher reports of menstrual taboo, stigma and discrimination were associated with higher odds of more severe depression (aOR=2.23; 95%CI: 1.81-2.76; p < 0.001) and anxiety (aOR=2.34; 95%CI: 1.91-2.87; p < 0.001). Reports of more severe depression (aOR=2.50; 95%CI: 1.65-3.78; p < 0.001) and anxiety (aOR =  2.55; 95%CI: 1.70-3.83; p < 0.001) increased with the severity of menstrual poverty. Greater impact on social participation during menstruation was strongly associated with higher odds of more severe depression (aOR=7.41; 95%CI: 4.90-11.191; p < 0.001) and anxiety (aOR=6.69; 95%CI: 5.22-8.59; p < 0.001). Sociodemographic factors may mediate the menstrual-mental health relationship. Our study sheds light on the interplay between menstrual and mental health in our context, from a gender and social health inequities perspective. There is little research on how menstrual health is linked to other health outcomes, especially with emotional and mental health. This study analysed whether menstrual health is associated with depression and anxiety in women and people who menstruate (PWM) aged 18 years or older. This study was conducted in 32 healthcare centres across five Spanish regions (Andalusia, Basque Country, Canary Islands, Catalonia, and Galicia). We developed an online survey and collected data during 2023. The data showed depression and anxiety were more likely in the following cases: 1) When there was heavy menstrual bleeding, 2) When there was intense menstrual pain, 3) Where there were high levels of menstrual stigma, taboo, or discrimination, 4) When women and PWM frequently lacked access to menstrual products and 5) When menstruation made it harder for women and PWM to take part in social activities. Other factors (e.g., age, financial hardship) also seemed to affect menstrual health and depression or anxiety. To sum up, this study shows a connection between poorer menstrual health and depression and anxiety.
Critical Health Literacy (CHL) emphasizes that health-related action is shaped by power, inequality and structural conditions, not by individual skills alone. This study examines a more specific and under-theorized link in that process: how negative and ambivalent social ties affect whether health-related competencies are enacted as concrete action in everyday healthcare. Drawing on 12 semi-structured interviews with adults living with multimorbidity in contexts of social vulnerability, the study explores how health-related knowledge, intentions, and professional advice are converted, delayed, redirected, or blocked in concrete health-related episodes. Analysis was abductive and used Compensatory Network Capital (CNC) as a sensitizing framework. Participants often described knowledge, motivation, and awareness of recommended care, but action appeared to be constrained when support was difficult to mobilize, poorly matched to the task, insufficiently recognized in institutional encounters, or when no helper was available to act on the person's behalf. These patterns were analysed through four tie-level mechanisms: Activation, Function-specific Help, Recognition and Substitution. Negative and ambivalent ties appeared to operate not only as psychosocial stressors but also as practical barriers affecting help-seeking, follow-up, care coordination and digital navigation. The study contributes to CHL research by specifying a micro-meso relational layer through which health-related competencies may become actionable, partial or blocked in practice. It suggests that equity-oriented health promotion may need to address not only individual skills, but also the relational and organizational conditions that enable, constrain or block health-related action.
The growing worldwide challenges of climate change, environmental degradation with biodiversity loss, and systemic inequities call for accessible, transformative, and impactful planetary health education that embraces a global perspective. However, traditional international learning experiences favor those with financial advantages, limiting equitable access to (planetary health) knowledge, intercultural collaboration competencies, and leadership opportunities. To address this challenge, the University Medical Center (UMC) Utrecht in the Netherlands and St. Luke's Medical Center College of Medicine's Planetary and Global Health Program (PGHP) in the Philippines developed a joint online course on Planetary Health and Climate-Resilient Health Systems, combining synchronous and asynchronous learning activities. This novel approach allowed for collaboration between Dutch and Filipino medical students, integrating international perspectives while providing an opportunity to solve local problems and offering equitable access to education. The course incorporated online modules, workshops, and mentorshipblending expertise from educators from the Netherlands and the Philippines. Through the course, participants engaged in intercultural teamwork, focusing on student-identified planetary health priority issues such as eco-anxiety, fast fashion, and healthcare waste management. This initiative illustrates the importance and feasibility of collaborative, glocal approaches to planetary health education, with learners gaining insights into localized solutions for global issues. It also showed how the values of decolonizing global health education and bidirectional knowledge exchange can turn intercultural learning into educational practice. Future iterations will expand the course to involve more countries, refine the use of interactive tools, address time-zone challenges, and incorporate education research to further study the learning process and outcomes. Institutional support is essential to scale this educational model, ensuring planetary health education remains accessible and impactful.
Black, non-Hispanic, and Indigenous perinatal patients in the United States experience disproportionately high rates of maternal mortality and morbidity compared with other racial and ethnic groups. While multiple social and institutional factors contribute to these inequities, structural racism, including implicit bias, remains key drivers shaping patient-health care professional interactions and the quality and equity of care. To evaluate, using a health equity lens, changes in perinatal health professionals' knowledge, attitudes, awareness, and self-reported preparedness and actions to mitigate racism before and after participation in the Institute for Perinatal Quality Improvement's SPEAK UP Champion course on implicit and explicit racial bias. We analyzed 1,303 pre-course surveys, 772 post-course surveys, and 48 follow-up implementation surveys collected from 28 SPEAK UP Champion courses delivered between 2021 and 2023. Surveys assessed knowledge, attitudes, awareness, and self-reported planned or implemented bias-mitigation strategies. Univariate analyses summarized outcomes, and chi-square tests examined racial differences in self-reported implicit bias. Post-course surveys showed improvements in understanding race as a social construct, increased recognition of racism as a driver of perinatal disparities, and heightened awareness of disrespect toward Black and Indigenous patients. Participants reported greater preparedness to identify biased behaviors and develop equity-focused quality improvement (QI) goals, including self-reported planned or early implementation of bias-mitigation strategies. Implementation survey findings reflected respondent-reported practice changes among a small, self-selected subset of participants. These findings suggest that (1) the SPEAK UP Champion course was associated with increased racial bias awareness and enhanced participants' reported capacity to develop bias-mitigation QI action plans; and (2) integrating equity-focused education on implicit and explicit racial bias with QI methods may represent a promising, systems-level approach to advancing equity across the perinatal continuum that warrants broader implementation and further evaluation using rigorous outcome measures.
AI has become an essential component of modern health care delivery in Epic (Epic Systems Corporation) electronic medical record (EMR) systems, supporting predictive analytics, diagnostic decision-making, and population health management. Despite these advancements, evidence reveals that AI algorithms can perpetuate or even amplify existing health inequities through biased training data and flawed model design. Such algorithmic bias poses ethical challenges for health care leadership, regulatory compliance, and executive communication, especially in ensuring patient equity, transparency, and public accountability. This conceptual paper examines how algorithmic bias in Epic's AI modules influences executive decision-making, organizational communication, and trust within health care systems. It integrates organizational communication theory and public health informatics research to propose a framework for ethical, transparent, and equitable communication in AI-integrated health care settings. Drawing upon the ethical communication and algorithmic trust framework (ECATF), this paper synthesizes interdisciplinary literature on AI bias, data governance, and leadership communication. The framework explains how transparent executive communication creates stakeholder trust in the context of bias identification and regulatory oversight. This conceptual analysis suggests that algorithmic bias influences leadership communication, equity framing, and governance strategies in AI-integrated health care systems. Incorporating AI bias auditing alongside regulatory monitoring and public education initiatives may support fairness, accountability, and health literacy across communities. As AI continues to shape health care leadership and policy, ongoing evaluation, ethical foresight, and regulatory vigilance are essential. Transparency, collaborative governance, and adaptive education will be necessary to ensure that AI supports equitable innovation rather than reinforcing unintended harm.
Digital health technologies are increasingly used across oncology services, but the available evidence remains concentrated on patient-facing and clinical applications. Organisational perspectives and the involvement of non-clinical professionals remain poorly represented. This scoping review mapped digital health technologies used in adult oncology care, examined their functions for healthcare professionals (HCPs), and assessed the extent and nature of evidence involving non-clinical professionals (NCPs). A scoping review was conducted to identify studies reporting the use of digital health technologies in adult cancer care. PubMed, Embase, CINAHL, and Web of Science were searched for studies published between 2015 and June 2025, and reference lists were hand-searched. Qualitative, quantitative, mixed-methods, pilot, feasibility, and implementation studies were eligible. Findings were synthesised narratively according to the primary function of each technology and the professional groups involved. Forty-one studies were included. The most frequently reported technologies were electronic patient-reported outcome systems, telehealth and telemonitoring platforms, mobile applications, and decision-support or AI-enabled systems. These technologies were used mainly for symptom monitoring, communication, follow-up, survivorship support, clinical decision support, and care coordination. Reported benefits included earlier symptom detection, improved communication, greater patient reassurance, and more structured clinical workflows. Common barriers included limited digital literacy, declining engagement over time, alert burden, workflow misalignment, poor interoperability, and inadequate technical infrastructure. Only three studies explicitly included non-clinical professionals, indicating that organisational and managerial applications remain substantially under-researched. Interoperability and workflow integration emerged as the most consistently supported implementation priorities. Equity, scalability, sustainability, and economic impact require further evaluation. Because evidence involving non-clinical professionals and AI-enabled implementation remains limited, recommendations concerning organisational adoption, AI governance, and workforce preparation should be treated as priorities for implementation research rather than established policy requirements.
暂无摘要（点击查看详情）
Robot-assisted surgery (RAS) may expand minimally invasive capacity, but its adoption in middle-income countries is shaped by constraints not captured by platform-level comparisons, including foreign-currency exposure, imported consumables and technical support, limited reimbursement, scarce simulators and proctors, and geographic concentration of specialized care. This narrative review examined literature on surgical innovation, health technology assessment, implementation, economics, training, governance and equity. A targeted PubMed/MEDLINE literature search last updated on 15 July 2026 was supplemented by reference chaining and searches of official institutional and professional sources; the Scale for the Assessment of Narrative Review Articles (SANRA) informed reporting transparency. The synthesis positions the proposed framework alongside existing consensus guidance, national governance models and disease-specific standardized outcome sets. It translates these sources into three linked components: responsible-adoption domains; a minimum institutional dataset covering clinical, technical, economic, training, access and governance variables; and a decision matrix for pilot adoption, expansion, correction, pause, restriction, reallocation or discontinuation. The framework treats RAS as a complex health-system intervention and links procedure-specific incremental value to the real local comparator, lifecycle costs, team competence, technical reliability, patient-centered consent and equity effects of centralization. It proposes locally prespecified review triggers rather than universal thresholds. This author-developed synthesis is not a validated instrument, consensus guideline, formal health technology assessment or cost-effectiveness model. Its next step is content review, feasibility piloting, structured consensus and multicenter evaluation. Used with local regulation and procedure-specific evidence, it could support transparent, accountable decisions about where RAS may be introduced, expanded, limited or stopped.
Dental caries (tooth decay) remains the most common noncommunicable disease worldwide and continues to impose significant dental and systemic health and economic burdens despite being largely preventable. Untreated disease contributes to pain, infection, reduced quality of life, and increased demand for restorative dental care. Governments, therefore, face policy decisions regarding population-level prevention strategies. This analysis evaluates policy options related to population-level fluoride exposure, including maintaining existing community water fluoridation (CWF) programs, expanding fluoridation coverage, discontinuing fluoridation in favor of clinical prevention strategies, and implementing alternative delivery mechanisms such as salt fluoridation. Drawing on Bardach's Eightfold Path for policy analysis, options are assessed using public health policy criteria, including health impact, equity implications, economic considerations, safety, and implementation feasibility. Evidence shows that CWF can reduce caries prevalence and treatment needs at the population level, particularly in communities with limited access to preventive dental care. Economic evaluations indicate that fluoridation reduces expenditures associated with dental treatment. However, policy feasibility varies depending on infrastructure, governance systems, and public health capacity. For policymakers, fluoridation represents one component of broader health prevention strategies that combine population-level prevention with improved access to preventive services.
Doula support is increasingly recognized as a promising strategy to address racial inequities in maternal health. The New York City Health Department launched the Citywide Doula Initiative (CDI) in March 2022 to expand access to free, community-based doula support in disinvested neighborhoods. To evaluate the program's first year of implementation, we conducted semi-structured interviews with 44 clients, doulas, doula-organization staff, and Health Department staff between November 2022 and April 2023. We coded interview notes using template analysis (a thematic analysis approach) and grouped barriers and facilitators according to the Practical Robust Implementation and Sustainability Model. Facilitators included high perceived value of the CDI, a shared commitment to community, open communication, and policymaker support for doula work. Barriers included insufficient time for planning, cumbersome bureaucratic processes, and cultural and capacity differences between the Health Department and community partners. Additional factors, such as-variation in hospital treatment of doulas, doula-organization capacity, doula skills and experiences, and client needs and circumstances-had mixed effects on program implementation. Findings indicate that community-based doula models, which have been shown to improve birth outcomes in disinvested communities, can be successfully scaled in an urban setting. Programs seeking to expand access to doula support should prioritize time for planning, allocate sufficient resources for administrative support, and work to establish efficient and responsive payment and contracting processes. Having mission-driven staff at all levels of the program and prioritizing relationships with community-based organizations and doulas are vital for success.
Resident scheduling is a high-dimensional optimization problem with implications for workload, fatigue risk, and equity. Real-world evaluations of AI-based constraint-optimization in health care are limited. This study aimed to evaluate an AI-based constraint-optimization scheduler versus a legacy rule-based scheduler for pediatric residency night calls. This is a single-center retrospective before-after study at a 235-bed tertiary pediatric center. Twenty-four consecutive months of night-call rosters were analyzed: preimplementation (January to December 2024, legacy rule-based autoscheduler) and postimplementation (January to December 2025, AI-based constraint-programming scheduler combining a local-search metaheuristic solver with human-in-the-loop review). The analytic unit was the resident-month. Outcomes were workload distribution, threshold exceedances (>6 total and >2 weekend calls/month), undesirable sequences (consecutive weekend calls; call-rest-call; call-rest-call-rest-call), equity (mean absolute error from equal share [MAE-ES], root mean square error from equal share), publication lead time, as well as pre- and postsurvey experience. We analyzed 6519 shifts across 1530 resident-months (legacy: 803 resident-months/107 physicians; AI: 727/87; weekend share 28.8% [934/3246] vs 28.6% [935/3273]; service mix P=.99). Mean calls/resident-month did not decline (4.04 vs 4.50; P<.001), but within-period SD was approximately halved. Threshold exceedances fell from 133/803 (16.6%) to 28/727 (3.9%) for >6 calls per month (risk ratio [RR] 0.24, 95% CI 0.15-0.34) and 89/803 (11.1%) to 21/727 (2.9%) for >2 weekend calls (RR 0.27, 95% CI 0.16-0.40; both P<.001). Undesirable sequences declined: consecutive weekends 24.4→18.7/100 resident-months (RR 0.77; P=.02); call-rest-call 51.2→23.4 (RR 0.46); call-rest-call-rest-call 5.6→1.0 (RR 0.18; both P<.001). Equity improved overall and within every qualification stratum: MAE-ES -0.26 shifts (95% CI -0.28 to -0.23) and RMSE-ES -0.29 (95% CI -0.32 to -0.26); Senior, Advanced, and Novice strata were all P<.001 after Holm correction. Publication lead time more than doubled (10.7→21.2 d; Δ+10.5, Cohen d=4.78; Cliff δ=1.00; P<.001). Interrupted time-series confirmed immediate level shifts for >6-call exceedances (β=-8.88; P=.004), MAE-ES (β=-0.18; P<.001), call-rest-call (β=-13.17; P=.002), call-rest-call-rest-call (β=-2.35; P=.03), and >2 weekend exceedances (β=-7.32; P<.001), with stable postimplementation fairness slopes. Among survey respondents (n=47 pre; n=38 post), software satisfaction rose 6.77→8.71/10, perceived timeliness 3.28→4.61/5, perceived consecutive-night frequency 3.15→4.24, and perceived equity 2.98→3.61 (all P≤.006). An AI-based constraint-optimization scheduler was associated with significantly more equitable on-call workload across all qualification strata, large reductions in high-risk shift sequences and threshold exceedances, and a doubling of publication lead time, despite no reduction in mean per-physician burden once all physicians were retained. Multisite prospective replication is warranted before generalization.
This paper aims to address a technical conundrum in accurately costing and delivering assistive products and services in the context of market-shaping strategies designed to promote assistive technology (AT) access. Developed in response to Australian Government policy changes regarding provision of AT to community-dwelling older adults, the AT SERVE (Assistive Technology-Services, Equity, Resources, Valuation, and Enablement) Tool is introduced. This novel interdisciplinary approach informs equitable modelling of AT costs by incorporating both assistive products and essential services that enable the selection, fitting, maintenance, and outcome measurement vital for safe and effective use. The methodology utilised interdisciplinary systems thinking to bring together targeted, policy-focussed reviews of international and Australian statutes and practice guidelines. Key components included (1) assistive product classification; (2) delineation of service provision elements; (3) workforce cadres required for service delivery; (4) healthcare intervention guidance regarding rurality, indigeneity, assistive product complexity, and risk; and (5) applied health economics cost modelling appropriate for a government context. The exemplar presented relates to the Australian context but can be adapted to the intersectional factors relevant in any setting.
Digital health technologies (DHTs) have the potential to improve care delivery and outcomes for patients with noncommunicable diseases. Yet their implementation in rural settings remains uneven, and the factors influencing uptake are not well understood. This mixed methods systematic review aimed to identify barriers and facilitators influencing the implementation and use of DHTs for remote management of noncommunicable diseases in rural areas. We searched Medline, Embase, and CINAHL from inception to February 12, 2026, using terms related to digital health, noncommunicable diseases, and rural settings. Following the Joanna Briggs Institute methodology for mixed-method systematic review, we synthesized quantitative and qualitative studies. Barriers and facilitators were categorized using the Consolidated Framework for Implementation Research, and study quality was appraised using the Mixed Methods Appraisal Tool. From the initial 1491 records, 14 studies met the inclusion criteria, with most conducted in high-income countries (n=11). Key barriers included technical challenges (software instability and hardware issues), poor internet connectivity, financial constraints, and workforce constraints, such as staff shortages and heavy workloads. Key facilitators included user-friendly technology design, strong leadership, effective teamwork, and ongoing communication. Evidence was predominantly qualitative, with only limited quantitative data available. DHTs show promise for improving access and continuity of care for cardiovascular disease, hypertension, and diabetes in rural settings; however, their impact is constrained by structural inequities, including limited broadband access, workforce shortages, and financial fragility. These findings highlight important implications for research, policy, and practice, including the need for rigorous mixed methods evaluations sensitive to rural contexts, long-term equity-oriented financing mechanisms, and strengthened organizational readiness to support effective DHT uptake.
Although atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death in men and women, there are sex differences in its prevalence and burden. There is limited time to address preventive counseling during in-person office visits. We propose an innovative health care delivery strategy with telehealth group counseling to improve care for women with ASCVD. After institutional review board approval, physicians screened our cardiology practice, and eligible patients were contacted. Women of ≥18 years of age were included, and 13 consented. Group counseling was performed via Zoom in one to five sessions over 12 weeks. Each 60-minute session discussed a secondary prevention topic. Surveys were administered before and after each session. Effectiveness was assessed using unpaired t-tests and qualitative feedback. The average participant age was 64.5 years. 69.2% of patients were White, while 30.8% were non-White. Hypertension and hyperlipidemia were present in 53.8% and 100% of participants, respectively. 53.8% had Medicare, 7.7% had Medicaid, and 38.5% had commercial insurance. Sessions averaged 7 participants per session, with an average attendance of 2.7 sessions per person. Pre- and postintervention comparisons showed improvement in ease of receiving answers to questions, feeling rushed, and understanding of cardiac problems, though results did not reach statistical significance. Feedback noted satisfaction with format, quality of information, and accessible communication of topics. Telemedicine group health counseling enhanced patient understanding of cardiac problems. However, significance was limited by size. Telemedicine provides an opportunity to optimize preventive care, highlighting the need for larger studies and correlation with clinical outcomes.
Asian and Latina survivors of international criminal sex trafficking often face profound mental health challenges, yet research on this group remains limited. This study examines mental health outcomes and correlates among 94 survivors in New York City using a Community-Based Participatory Research approach. Multivariable logistic regression identified factors associated with depression, posttraumatic stress disorder (PTSD), and comorbid symptoms. Survivors reported high rates of depression (41.9%), PTSD (41.3%), and co-occurring symptoms (26.6%). Pain and forced abortion were significantly associated with PTSD and comorbidity. Findings highlight the need for trauma-informed, culturally appropriate mental health interventions addressing pain and reproductive coercion to support survivor recovery.
Clinical decision support (CDS) systems can improve care quality, but their implications for equity remain uncertain. We examined whether provider response to CDS alerts differed by patient race and sex in primary care, and whether differences in alert exposure helped explain any observed variation. We conducted a retrospective study using EHR data from a New York City academic health system, focusing on alert-based CDS during outpatient primary care. Logistic regression was used to estimate the likelihood of alert engagement by patient race and sex, while adjusting for encounter and provider factors. We used a generalized structural equation model to assess mediation by alert type, decomposing direct and indirect effects of demographics on response.Direct effects suggest that providers may respond differently to alerts based on patient identity, consistent with interpersonal bias, in which implicit or explicit attitudes shape clinical behavior, and on the context of the visit. Indirect effects highlight disparities in how alerts are assigned across groups, indicating that algorithmic or systemic bias may be embedded within the technology itself. Estimated mediated pathways suggest that even when providers respond uniformly to alerts, unequal exposure can still produce inequitable outcomes. The findings highlight that the type of CDS triggered plays a significant role in differential CDS responses, with provider- and patient-related factors evident in these differences. These findings underscore the need to evaluate not only provider behavior but also the logic and distribution of CDS tools themselves, as both can contribute to disparities in care delivery. Further research should also focus on looking for the potential health impact of the differential response. Digital tools meant to standardize care can unintentionally contribute to which patients receiving care. We investigate whether providers' use of these tools is related to a patient's identity or influenced by the types of tools provided to them in primary care. Using electronic health record data from a large urban health system, we find that providers' responses to alerts are shaped not only by patient identity but also by the nature of the alert itself. Direct effects suggest that providers may engage differently with CDS based on patient demographics, indicating potential interpersonal bias. Indirect effects reveal that certain patient groups are more or less likely to receive specific types of alerts, indicating embedded algorithmic or systemic bias. These findings underscore the importance of evaluating both provider behavior and the design of CDS tools when assessing equity in digital health. Even when providers respond consistently, unequal exposure to alerts can produce inequitable outcomes. Our results underscore the need for more transparent and equity-aware CDS design and implementation strategies that consider both human and technological sources of bias.