BACKGROUND: A suitable definition of primary care to capture the variety of prevailing international organisation and service-delivery models is lacking. AIM: Evaluation of strength of primary care in Europe. DESIGN AND SETTING: International comparative cross-sectional study performed in 2009-2010, involving 27 EU member states, plus Iceland, Norway, Switzerland, and Turkey. METHOD: Outcome measures covered three dimensions of primary care structure: primary care governance, economic conditions of primary care, and primary care workforce development; and four dimensions of primary care service-delivery process: accessibility, comprehensiveness, continuity, and coordination of primary care. The primary care dimensions were operationalised by a total of 77 indicators for which data were collected in 31 countries. Data sources included national and international literature, governmental publications, statistical databases, and experts' consultations. RESULTS: Countries with relatively strong primary care are Belgium, Denmark, Estonia, Finland, Lithuania, the Netherlands, Portugal, Slovenia, Spain, and the UK. Countries either have many primary care policies and regulations in place, combined with good financial coverage and resources, and adequate primary care workforce conditions, or have consistently only few of these primary care structures in place. There is no correlation between the access, continuity, coordination, and comprehensiveness of primary care of countries. CONCLUSION: Variation is shown in the strength of primary care across Europe, indicating a discrepancy in the responsibility given to primary care in national and international policy initiatives and the needed investments in primary care to solve, for example, future shortages of workforce. Countries are consistent in their primary care focus on all important structure dimensions. Countries need to improve their primary care information infrastructure to facilitate primary care performance management.
Primary care serves as the cornerstone in a strong healthcare system. However, it has long been overlooked in the United States (USA), and an imbalance between specialty and primary care exists. The objective of this focused review paper is to identify research evidence on the value of primary care both in the USA and internationally, focusing on the importance of effective primary care services in delivering quality healthcare, improving health outcomes, and reducing disparities. Literature searches were performed in PubMed as well as "snowballing" based on the bibliographies of the retrieved articles. The areas reviewed included primary care definitions, primary care measurement, primary care practice, primary care and health, primary care and quality, primary care and cost, primary care and equity, primary care and health centers, and primary care and healthcare reform. In both developed and developing countries, primary care has been demonstrated to be associated with enhanced access to healthcare services, better health outcomes, and a decrease in hospitalization and use of emergency department visits. Primary care can also help counteract the negative impact of poor economic conditions on health.
BACKGROUND: Even though there is general agreement that primary care is the linchpin of effective health care delivery, to date no efforts have been made to systematically review the scientific evidence supporting this supposition. The aim of this study was to examine the breadth of primary care by identifying its core dimensions and to assess the evidence for their interrelations and their relevance to outcomes at (primary) health system level. METHODS: A systematic review of the primary care literature was carried out, restricted to English language journals reporting original research or systematic reviews. Studies published between 2003 and July 2008 were searched in MEDLINE, Embase, Cochrane Library, CINAHL, King's Fund Database, IDEAS Database, and EconLit. RESULTS: Eighty-five studies were identified. This review was able to provide insight in the complexity of primary care as a multidimensional system, by identifying ten core dimensions that constitute a primary care system. The structure of a primary care system consists of three dimensions: 1. governance; 2. economic conditions; and 3. workforce development. The primary care process is determined by four dimensions: 4. access; 5. continuity of care; 6. coordination of care; and 7. comprehensiveness of care. The outcome of a primary care system includes three dimensions: 8. quality of care; 9. efficiency care; and 10. equity in health. There is a considerable evidence base showing that primary care contributes through its dimensions to overall health system performance and health. CONCLUSIONS: A primary care system can be defined and approached as a multidimensional system contributing to overall health system performance and health.
There are challenges that must be overcome to make recommender systems useful in healthcare settings. The reasons are varied: the lack of publicly available clinical data, the difficulty that users may have in understanding the reasons why a recommendation was made, the risks that may be involved in following that recommendation, and the uncertainty about its effectiveness. In this work, we address these challenges with a recommendation model that leverages the structure of psychometric data to provide visual explanations that are faithful to the model and interpretable by care professionals. We focus on a narrow healthcare niche, gerontological primary care, to show that the proposed recommendation model can assist the attending professional in the creation of personalised care plans. We report results of a comparative offline performance evaluation of the proposed model on healthcare datasets that were collected by research partners in Brazil, as well as the results of a user study that evaluates the interpretability of the visual explanations the model generates. The results suggest that the proposed model can advance the application of recommender systems in this healthcare nic
INTRODUCTION: This article presents the scientific evidence for the merits of telemedicine interventions in primary care. Although there is no uniform and consistent definition of primary care, most agree that it occupies a central role in the healthcare system as first contact for patients seeking care, as well as gatekeeper and coordinator of care. It enables and supports patient-centered care, the medical home, managed care, accountable care, and population health. Increasing concerns about sustainability and the anticipated shortages of primary care physicians have sparked interest in exploring the potential of telemedicine in addressing many of the challenges facing primary care in the United States and the world. MATERIALS AND METHODS: The findings are based on a systematic review of scientific studies published from 2005 through 2015. The initial search yielded 2,308 articles, with 86 meeting the inclusion criteria. Evidence is organized and evaluated according to feasibility/acceptance, intermediate outcomes, health outcomes, and cost. RESULTS: The majority of studies support the feasibility/acceptance of telemedicine for use in primary care, although it varies significantly by demographic variables, such as gender, age, and socioeconomic status, and telemedicine has often been found more acceptable by patients than healthcare providers. Outcomes data are limited but overall suggest that telemedicine interventions are generally at least as effective as traditional care. Cost analyses vary, but telemedicine in primary care is increasingly demonstrated to be cost-effective. CONCLUSIONS: Telemedicine has significant potential to address many of the challenges facing primary care in today's healthcare environment. Challenges still remain in validating its impact on clinical outcomes with scientific rigor, as well as in standardizing methods to assess cost, but patient and provider acceptance is increasingly making telemedicine a viable and integral component of primary care around the world.
Depression is underdiagnosed in primary care, yet timely identification remains critical. Recorded clinical encounters, increasingly common with digital scribing technologies, present an opportunity to detect depression from naturalistic dialogue. We investigated automated depression detection from 1,108 audio-recorded primary care encounters in the Establishing Focus study, with depression defined by PHQ-9 (n=253 depressed, n=855 non-depressed). We compared three supervised approaches, Sentence-BERT + Logistic Regression (LR), LIWC+LR and ModernBERT, against a zero-shot GPT-OSS. GPT-OSS achieved the strongest performance (AUPRC=0.510, AUROC=0.774), with LIWC+LR competitive among supervised models (AUPRC=0.500, AUROC=0.742). Combined dyadic transcripts outperformed single-speaker configurations, with providers linguistically mirroring patients in depression encounters, an additive signal not captured by either speaker alone. Meaningful detection is achievable from the first 128 patient tokens (AUPRC=0.356, AUROC=0.675), supporting in-the-moment clinical decision support. These findings argue for passively collected clinical audio as a low-burden complement to existing screening wor
Demand for health care is constantly increasing due to the ongoing demographic change, while at the same time health service providers face difficulties in finding skilled personnel. This creates pressure on health care systems around the world, such that the efficient, nationwide provision of primary health care has become one of society's greatest challenges. Due to the complexity of health care systems, unforeseen future events, and a frequent lack of data, analyzing and optimizing the performance of health care systems means tackling a wicked problem. To support this task for primary care, this paper introduces the hybrid agent-based simulation model SiM-Care. SiM-Care models the interactions of patients and primary care physicians on an individual level. By tracking agent interactions, it enables modelers to assess multiple key indicators such as patient waiting times and physician utilization. Based on these indicators, primary care systems can be assessed and compared. Moreover, changes in the infrastructure, patient behavior, and service design can be directly evaluated. To showcase the opportunities offered by SiM-Care and aid model validation, we present a case study for
Millions of people now use non-clinical Large Language Model (LLM) tools like ChatGPT for mental well-being support. This paper investigates what it means to design such tools responsibly, and how to operationalize that responsibility in their design and evaluation. By interviewing experts and analyzing related regulations, we found that designing an LLM tool responsibly involves: (1) Articulating the specific benefits it guarantees and for whom. Does it guarantee specific, proven relief, like an over-the-counter drug, or offer minimal guarantees, like a nutritional supplement? (2) Specifying the LLM tool's "active ingredients" for improving well-being and whether it guarantees their effective delivery (like a primary care provider) or not (like a yoga instructor). These specifications outline an LLM tool's pertinent risks, appropriate evaluation metrics, and the respective responsibilities of LLM developers, tool designers, and users. These analogies - LLM tools as supplements, drugs, yoga instructors, and primary care providers - can scaffold further conversations about their responsible design.
Referral workflow inefficiencies, including misaligned referrals and delays, contribute to suboptimal patient outcomes and higher healthcare costs. In this study, we investigated the possibility of predicting procedural needs based on primary care diagnostic entries, thereby improving referral accuracy, streamlining workflows, and providing better care to patients. A de-identified dataset of 2,086 orthopedic referrals from the University of Texas Health at Tyler was analyzed using machine learning models built on Base General Embeddings (BGE) for semantic extraction. To ensure real-world applicability, noise tolerance experiments were conducted, and oversampling techniques were employed to mitigate class imbalance. The selected optimum and parsimonious embedding model demonstrated high predictive accuracy (ROC-AUC: 0.874, Matthews Correlation Coefficient (MCC): 0.540), effectively distinguishing patients requiring surgical intervention. Dimensionality reduction techniques confirmed the model's ability to capture meaningful clinical relationships. A threshold sensitivity analysis identified an optimal decision threshold (0.30) to balance precision and recall, maximizing referral eff
Large Language Models(LLMs) hold promise for improving healthcare access in low-resource settings, but their effectiveness in African primary care remains underexplored. We present a methodology for creating a benchmark dataset and evaluation framework focused on Kenyan Level 2 and 3 clinical care. Our approach uses retrieval augmented generation (RAG) to ground clinical questions in Kenya's national guidelines, ensuring alignment with local standards. These guidelines were digitized, chunked, and indexed for semantic retrieval. Gemini Flash 2.0 Lite was then prompted with guideline excerpts to generate realistic clinical scenarios, multiple-choice questions, and rationale based answers in English and Swahili. Kenyan physicians co-created and refined the dataset, and a blinded expert review process ensured clinical accuracy, clarity, and cultural appropriateness. The resulting Alama Health QA dataset includes thousands of regulator-aligned question answer pairs across common outpatient conditions. Beyond accuracy, we introduce evaluation metrics that test clinical reasoning, safety, and adaptability such as rare case detection (Needle in the Haystack), stepwise logic (Decision Poin
INTRODUCTION: Multimorbidity is common among the heterogeneous primary care population, but little data exist on its association with health care utilization or cost. OBJECTIVE: The aim of this observational study was to examine the prevalence and associated health care utilization and cost of patients with multimorbidity. METHODS: All patients >50 years of age were eligible for the study which took place in three primary care practices in the West of Ireland. Chronic medical conditions and associated health care utilization in primary and secondary care were identified through patient record review. RESULTS: In a sample of 3309 patients in the community, the prevalence of multimorbidity was 66.2% (95% CI: 64.5-67.8) in those >50 years of age. Health care utilization and cost was significantly increased among patients with multimorbidity (P < 0.001). After multivariate adjustment for age, gender and free medical care eligibility, the addition of each chronic condition led to an associated increase in primary care consultations (P = 0.001) (11.9 versus 3.7 for >4 conditions versus 0 conditions); hospital out-patient visits (P = 0.001) (3.6 versus 0.6 for >4 conditions versus 0 conditions); hospital admissions (P = 0.01) [adjusted odds ratio (OR) of 4.51 for >4 conditions versus 0 conditions] and total health care costs (P < 0.001) (€4,096.86 versus €760.20 for >4 conditions versus 0 conditions) over the previous 12 months. CONCLUSIONS: Multimorbidity is very common in primary care and in a system with strong gatekeeping is associated with high health care utilization and cost across the health care system. Interventions to address quality and cost associated with multimorbidity must focus on primary as well as secondary care.
Unlike most primary headaches, secondary headaches need specialized care and can have devastating consequences if not treated promptly. Clinical guidelines highlight several 'red flag' features, such as thunderclap onset, meningismus, papilledema, focal neurologic deficits, signs of temporal arteritis, systemic illness, and the 'worst headache of their life' presentation. Despite these guidelines, determining which patients require urgent evaluation remains challenging in primary care settings. Clinicians often work with limited time, incomplete information, and diverse symptom presentations, which can lead to under-recognition and inappropriate care. We present a large language model (LLM)-based multi-agent clinical decision support system built on an orchestrator-specialist architecture, designed to perform explicit and interpretable secondary headache diagnosis from free-text clinical vignettes. The multi-agent system decomposes diagnosis into seven domain-specialized agents, each producing a structured and evidence-grounded rationale, while a central orchestrator performs task decomposition and coordinates agent routing. We evaluated the multi-agent system using 90 expert-valid
Relationship-centred care (RCC) recognises that healthcare quality depends not only on outcomes, but on how voice, responsibility, and emotional labour are negotiated among patients, caregivers, and providers. As AI systems enter sensitive care contexts, they introduce a new participant into these negotiations. Drawing on empirical work in Advance Care Planning (ACP) and peer support, we argue that AI's primary impact in high-subjectivity domains is not optimisation but redistribution: it reorganises who speaks, who decides, and who bears moral responsibility. Across both settings, participants were less concerned with technical accuracy than with relational consequences: whether AI would appropriately represent their decision, reduce burden, or blur accountability, scaffold connection, or subtly displace it. We identify three relational dimensions: authority, temporality, and visibility, through which AI reshapes care relationships, and propose design provocations centred on relational legibility, bounded agency, responsibility traceability, and non-substitutive scaffolding.
Current deep learning models are mostly task specific and lack a user-friendly interface to operate. We present Meta-EyeFM, a multi-function foundation model that integrates a large language model (LLM) with vision foundation models (VFMs) for ocular disease assessment. Meta-EyeFM leverages a routing mechanism to enable accurate task-specific analysis based on text queries. Using Low Rank Adaptation, we fine-tuned our VFMs to detect ocular and systemic diseases, differentiate ocular disease severity, and identify common ocular signs. The model achieved 100% accuracy in routing fundus images to appropriate VFMs, which achieved $\ge$ 82.2% accuracy in disease detection, $\ge$ 89% in severity differentiation, $\ge$ 76% in sign identification. Meta-EyeFM was 11% to 43% more accurate than Gemini-1.5-flash and ChatGPT-4o LMMs in detecting various eye diseases and comparable to an ophthalmologist. This system offers enhanced usability and diagnostic performance, making it a valuable decision support tool for primary eye care or an online LLM for fundus evaluation.
Large language models (LLMs) show promise for tailored healthcare communication but face challenges in interpretability and multi-task integration particularly for domain-specific needs like myopia, and their real-world effectiveness as patient education tools has yet to be demonstrated. Here, we introduce ChatMyopia, an LLM-based AI agent designed to address text and image-based inquiries related to myopia. To achieve this, ChatMyopia integrates an image classification tool and a retrieval-augmented knowledge base built from literature, expert consensus, and clinical guidelines. Myopic maculopathy grading task, single question examination and human evaluations validated its ability to deliver personalized, accurate, and safe responses to myopia-related inquiries with high scalability and interpretability. In a randomized controlled trial (n=70, NCT06607822), ChatMyopia significantly improved patient satisfaction compared to traditional leaflets, enhancing patient education in accuracy, empathy, disease awareness, and patient-eyecare practitioner communication. These findings highlight ChatMyopia's potential as a valuable supplement to enhance patient education and improve satisfac
Large language models (LLMs) often match or exceed clinician-level performance on medical benchmarks, yet very few are evaluated on real clinical data or examined beyond headline metrics. We present, to our knowledge, the first evaluation of an LLM-based medication safety review system on real NHS primary care data, with detailed characterisation of key failure behaviours across varying levels of clinical complexity. In a retrospective study using a population-scale EHR spanning 2,125,549 adults in NHS Cheshire and Merseyside, we strategically sampled patients to capture a broad range of clinical complexity and medication safety risk, yielding 277 patients after data-quality exclusions. An expert clinician reviewed these patients and graded system-identified issues and proposed interventions. Our primary LLM system showed strong performance in recognising when a clinical issue is present (sensitivity 100\% [95\% CI 98.2--100], specificity 83.1\% [95\% CI 72.7--90.1]), yet correctly identified all issues and interventions in only 46.9\% [95\% CI 41.1--52.8] of patients. Failure analysis reveals that, in this setting, the dominant failure mechanism is contextual reasoning rather than
Missing data, inaccuracies in medication lists, and recording delays in electronic health records (EHR) are major limitations for target trial emulation (TTE), the process by which EHR data are used to retrospectively emulate a randomized control trial. EHR TTE relies on recorded data that proxy true drug exposures and outcomes. We investigate the under-utilized criterion that a patient has indications of primary care provider (PCP) encounters within the EHR. Such patients tend to have more records overall and a greater proportion of the types of encounters that materialize comprehensive and up-to-date records. We examine the impact of including a PCP feature in the TTE model or as an eligibility criterion for cohort selection, contrasted with ignoring it altogether. To that end, we compare the estimated effects of two first line antidiabetic drug classes on the onset of Alzheimer's Disease and Related Dementias (ADRD). We find that the estimated treatment effect is sensitive to the consideration of a PCP feature, particularly when used as an eligibility criterion. Our work suggests that this PCP feature should be further researched.
We evaluate the impact of large language model-based clinical decision support in live care. In partnership with Penda Health, a network of primary care clinics in Nairobi, Kenya, we studied AI Consult, a tool that serves as a safety net for clinicians by identifying potential documentation and clinical decision-making errors. AI Consult integrates into clinician workflows, activating only when needed and preserving clinician autonomy. We conducted a quality improvement study, comparing outcomes for 39,849 patient visits performed by clinicians with or without access to AI Consult across 15 clinics. Visits were rated by independent physicians to identify clinical errors. Clinicians with access to AI Consult made relatively fewer errors: 16% fewer diagnostic errors and 13% fewer treatment errors. In absolute terms, the introduction of AI Consult would avert diagnostic errors in 22,000 visits and treatment errors in 29,000 visits annually at Penda alone. In a survey of clinicians with AI Consult, all clinicians said that AI Consult improved the quality of care they delivered, with 75% saying the effect was "substantial". These results required a clinical workflow-aligned AI Consult i
This review underscores the vital role of interoperability in digital health, advocating for a standardized framework. It focuses on implementing a Fast Healthcare Interoperability Resources (FHIR) server, addressing technical, semantic, and process challenges. FHIR's adaptability ensures uniformity within Primary Care Health Information Systems, fostering interoperability. Patient data management complexities highlight the pivotal role of semantic interoperability in seamless patient care. FHIR standards enhance these efforts, offering multiple pathways for data search. The ADR-guided FHIR server implementation systematically addresses challenges related to patient identity, biometrics, and data security. The detailed development phases emphasize architecture, API integration, and security. The concluding stages incorporate forward-looking approaches, including HHIMS Synthetic Dataset testing. Envisioning FHIR integration as transformative, it anticipates a responsive healthcare environment aligned with the evolving digital health landscape, ensuring comprehensive, dynamic, and interconnected systems for efficient data exchange and access.
Family-focused prevention programs have been shown to effectively reduce a range of negative behavioral health outcomes but have had limited reach. Three key barriers must be overcome to expand the reach of family-focused prevention programs and thereby achieve a significant public health impact. These barriers are (1) current social norms and perceptions of parenting programs; (2) concerns about the expertise and legitimacy of sponsoring organizations to offer parenting advice; and (3) a paucity of stable, sustainable funding mechanisms. Primary healthcare settings are well positioned to overcome these barriers. Recent changes within health care make primary care settings an increasingly favorable home for family-focused prevention and suggest possibilities for sustainable funding of family-focused prevention programs. This paper discusses the existing advantages of primary care settings and lays out a plan to move toward realizing the potential public health impact of family-focused prevention through widespread implementation in primary healthcare settings. Family-focused prevention programs have been shown to effectively reduce a range of negative behavioral health outcomes but have had limited reach. Three key barriers must be overcome to expand the reach of family-focused prevention programs and thereby achieve a significant public health impact. These barriers are (1) current social norms and perceptions of parenting programs; (2) concerns about the expertise and legitimacy of sponsoring organizations to offer parenting advice; and (3) a paucity of stable, sustainable funding mechanisms. Primary healthcare settings are well positioned to overcome these barriers. Recent changes within health care make primary care settings an increasingly favorable home for family-focused prevention and suggest possibilities for sustainable funding of family-focused prevention programs. This paper discusses the existing advantages of primary care settings and lays out a plan to move toward realizing the potential public health impact of family-focused prevention through widespread implementation in primary healthcare settings.