Health Care Decisions Research Articles

Health Diagnostic Assistant using LLMs

The Health Diagnostic Assistant leverages advanced Large Language Models (LLMs) and Natural Language Processing (NLP) techniques to enhance patient diagnosis and healthcare decision-making. This innovative system employs Retrieval-Augmented Generation (RAG) to combine the strengths of pre-trained language models with a dynamic retrieval mechanism, allowing it to access and synthesize real-time medical knowledge from a wide array of databases. By analyzing patient symptoms, medical histories, and contextual data, the assistant generates accurate, context-aware recommendations and insights. The project aims to streamline the diagnostic process, reduce the burden on healthcare professionals, and improve patient outcomes by providing evidence-based suggestions tailored to individual cases. Through continuous learning and integration of user feedback, the Health Diagnostic Assistant aspires to evolve into a reliable tool for both patients and clinicians, fostering informed decision-making in the healthcare landscape.

Hospital Length-of-Stay Prediction Using Machine Learning Algorithms—A Literature Review

Predicting hospital length of stay is critical for efficient hospital management, enabling proactive resource allocation, the optimization of bed availability, and optimal patient care. This paper explores the potential of machine learning algorithms to revolutionize hospital length-of-stay predictions, contributing to healthcare efficiency and patient care. The main objective is to identify the most effective machine learning algorithm for building a predictive model capable of predicting hospital length of stay. The bibliographic search of the existing literature on machine learning algorithms applied to hospital length of stay predictions highlighted the most relevant papers within this area of research. The papers were analyzed in terms of model types and metrics that contributed to the considerable impact on healthcare decision making. We also discuss the challenges and limitations of machine learning algorithms for predicting length of stay, and the importance of data quality and ethical considerations.

Challenges associated with access to recently developed hemophilia treatments in routine care: perspectives of healthcare professionals.

The treatment landscape for haemophilia continues to rapidly develop, and expectations for future treatment success are high. There is limited information on the challenges to accessing new and innovative therapies. The aim of this study was to explore challenges with accessing haemophilia treatment from the perspective of healthcare professionals (HCPs). A crosssectional study design was used. A pilot-tested, online survey was distributed to haemophilia treatment centres in Australia, Canada, France, Italy, New Zealand, Republic of Ireland, Turkey, the United States, and the United Kingdom. The questionnaire covered questions on product access, economic considerations, health technology assessment requirements, and patient organization involvement. The results were analyzed descriptively using SPSS. A total of 154 HCPs completed the questionnaire. There was heterogeneity across countries, regions, and centres regarding HCPs' knowledge of access to novel recently developed treatments. Notable limitations to access were reported such as differences in access based on age of patient and type of product, economic considerations, and the growing influence of HTA bodies. Many countries have a hemophilia patient organization that does not have a vote at the decision-making table. There is a need to empower HCPs to better understand national healthcare structures and decisions that lead to access limitations. Requirements from HTA bodies must be understood to optimally design clinical studies and value generation of treatment options. This may strengthen the haemophilia treatment centre's voice to collectively mandate for exchange with key involved individuals, such as the payers and politicians for the provision of optimal therapy.

Widely Integrated Services in Home (WISH) for homebound older adults: a study protocol for a randomized encouragement trial.

Home-based primary care (HBPC) is an emerging patient-centered, interprofessional healthcare service model that can address unmet medical needs and care burdens for homebound older adults. In December 2022, the Ministry of Health and Welfare in South Korea launched the Home-Based Medical Center Demonstration project to provide a new bundle payment for physician home visits. In this study, we seek to determine whether the recently introduced HBPC services in South Korea have been associated with a reduction in long-term care (LTC) facility admissions and acute hospitalizations among homebound older adults. The study is a community-based, multicenter, two-arm, randomized encouragement design trial with a 12-month follow-up period (n = 600). Eligible study participants are community-dwelling LTC recipients with multimorbidity and functional deterioration. Study participants are recruited from five HBPC centers located in urban areas (Northeast Seoul, West Seoul, Daejeon, Wonju, and Paju). The study participants are randomly assigned to either the HBPC group or the usual care group with a 1:1 allocation ratio. Those assigned to the HBPC group receive longitudinal home visits at least once a month by an interprofessional HBPC team according to the Widely Integrated Services in Home (WISH) intervention protocol. This protocol adheres to the Integrated Care for Older People principles, which call for a person-centered assessment and broader integration of health and LTC services at the micro-, meso- and macro-levels. Primary outcomes of the trial are 1) between-group community survival days and 2) between-group potentially avoidable hospitalizations. Results of the treatment are estimated by both modified intention-to-treat and complier average causal effect analytic methods. This study aims to investigate the real-world effectiveness of HBPC on the reduction of LTC facility admissions and acute care hospitalizations in the community setting. The findings may inform healthcare policy decisions to expand HBPC services in South Korea and other countries. CRIS KCT0007921.

Recounting the untold stories of breast cancer patient experiences: lessons learned from a patient-public involvement and engagement storytelling event.

Breast cancer remains a prevalent disease in women worldwide. Though advancements in breast cancer care have improved patient survival, a breast cancer diagnosis, and subsequent interventions have a lasting impact on patients' lived experiences during the pandemic. We present the collaborative learning process from this patient engagement workshop series as a community-academic partnership. Narrative medicine tools were used to recount patients' lived experiences following diagnosis, where both patients and researchers shared their cancer research activities in each workshop, and the role of the multidisciplinary healthcare team was discussed. We used an iterative approach to cohort building, narrative development, and the use of multiple media formats to capture stories. Over 20 patients with breast cancer shared their stories for the first time since their diagnosis with a wider audience. Here, we present the learning process and considerations from this event. Understanding patients' lived experiences can support researchers and healthcare professionals in developing an empathetic approach to shared healthcare decision making. Moreover, understanding the lived experiences of patients is critical to addressing disparities in healthcare.

Indirect Treatment Comparisons in Healthcare Decision Making: A Targeted Review of Regulatory Approval, Reimbursement, and Pricing Recommendations Globally for Oncology Drugs in 2021-2023.

Indirect treatment comparisons (ITCs) evaluate novel treatments compared to appropriate comparators when direct evidence is unavailable or infeasible. The objective of this study was to highlight the prevalence of different ITC methods in oncology drug submissions and to provide insights into how ITCs have been used in recent regulatory approval, reimbursement recommendations, or pricing decisions across various regions and diverse assessment frameworks. A targeted literature review was conducted to identify assessment documents for oncology drug submissions that included ITCs. This included hand searches of the websites of four regulatory bodies and four health technology assessment (HTA) agencies with varying assessment frameworks across North America, Europe, and Asia-Pacific. A total of 185 documents were included for synthesis. Documents were retrieved from all four HTA agencies and the European Medicines Agency (EMA), the only regulatory body with eligible records. Within these, 188 unique submissions included a total of 306 supporting ITCs of various methods. Authorities more frequently favored anchored or population-adjusted ITC techniques for their effectiveness in data adjustment and bias mitigation. Furthermore, ITCs in orphan drug submissions more frequently led to positive decisions compared to non-orphan submissions. This review highlights the crucial role and widespread use of ITCs in global healthcare decision-making, particularly when direct evidence is lacking, and in the discernment of market-specific clinical benefits. This work contributes to bolstering the credibility and recognition of ITCs across regulatory and HTA agencies of diverse regions and assessment frameworks.

Identifying and prioritizing inefficiency causes in Iran's health system.

Enhancing efficiency is crucial in addressing the escalating scarcity of healthcare resources. It plays a pivotal role in achieving Universal Health Coverage (UHC), with the ultimate goal of ensuring health equity for all. A fundamental strategy to bolster efficiency involves pinpointing the underlying causes of inefficiency within the healthcare system through empirical research. This study aimed to determine and prioritize the causes of inefficiency in Iran's health system. This mixed-method study comprised three phases. The initial phase involved identifying the causes of inefficiency through a comprehensive literature review of relevant studies published between January 1, 2010, to January 1, 2021. The causes were then aligned and prioritized using criteria derived from the literature and expert opinion. Finally, the identified causes were ranked based on their significance using Multiple-Criteria Decision Analysis (MCDA). From an initial pool of 307 causes of inefficiency, they were reduced to 121 causes in the first round of screening which were categorized into 13 thematic topics. The second screening process further narrowed the list to 48 causes. Among these, the leading causes of inefficiency included the inadequate supply and unequal distribution of hospital beds, the overuse of health services, and the mismanagement of the health workforce. In contrast, the use of traditional treatment methods was determined to be the least significant factor contributing to inefficiency. This study identified key inefficiencies in Iran's health system, such as resource misallocation, overuse of services, and workforce mismanagement. Addressing these issues is essential for optimizing resource utilization, enhancing service delivery, and achieving UHC. The findings suggest that policymakers should prioritize reforms in hospital bed distribution, implement strategies to reduce unnecessary health service use, and strengthen human resource management. Additionally, targeted policies that focus on decentralizing healthcare decision-making and enhancing primary care could significantly improve system-wide efficiency. Future research should evaluate the effectiveness of these interventions and explore the role of digital health solutions in mitigating identified inefficiencies.

What are the priorities of consumers and carers regarding measurement for evaluation in mental healthcare? Results from a Q-methodology study.

The purpose of this study was to identify and describe common views of people with lived experience of mental health challenges - consumers and carers, families and supporters - of what they consider the most important measures to include in health economic evaluations which assess the incremental value of competing options in mental health care. Participants (n = 111) were people living in the state of Victoria, Australia, who identified as consumers of mental healthcare (n = 38); carers, family members and/or supporters (n = 43); or both (n = 30). Factor analysis based on Q-Methodology was used to identify clusters of people who hold similar viewpoints. Common viewpoints were described in terms of the characteristics of the group, and a qualitative interpretation was conducted on the basis of distinguishing statements and quotes provided in participants' own words. We identified four common views: (1) safety before all else, prioritizing physical, sexual and psychological safety; (2) hope and partnership in processes of care; (3) physical and emotional health and wellbeing; and (4) care access, continuity and partnership with families. Although different priorities were identified for each viewpoint, key priority areas that were common to all views were having an environment in the health service that fosters respect and dignity, and that consumers feel heard and listened to. In sub-group and qualitative analyses, differences were observed regarding the likelihood of consumers and carers holding each of the views, as well as by age group. While some differences were noted between the views of consumers and carers and different age groups, there was also common ground regarding what outcomes are of most importance to measure. Including these measures in evaluation frameworks would provide a way of focussing mental healthcare decisions on the aspects of mental healthcare that are of most value to consumers and carers, thereby addressing an important shortcoming of current approaches to decision-making in mental healthcare.

Ensuring AI Algorithm Fairness in Healthcare Decision-Making

This paper provides a comprehensive analysis of artificial intelligence (AI) implementation in Ghana's healthcare system, synthesizing findings from recent Ghanaian studies. It examines legal and regulatory frameworks, ethical considerations, bias and fairness, transparency and accountability, interdisciplinary approaches, patient rights and data protection, healthcare access and equity, and implementation and oversight. The analysis reveals both significant opportunities for AI to improve healthcare delivery in Ghana and substantial challenges in ensuring its ethical, equitable, and effective implementation. Key findings highlight the need for AI-specific healthcare regulations, investment in local capacity building, culturally sensitive AI design, robust fairness and accountability mechanisms, and ongoing stakeholder engagement. This paper contributes to the discourse on AI in healthcare in developing countries, offering insights into the unique challenges faced by lower-middle-income countries. It aims to inform evidence-based policymaking and enrich academic understanding of AI implementation in diverse cultural and economic contexts.

The optimal approach for retrieving systematic reviews was achieved when searching MEDLINE and Epistemonikos in addition to reference checking: a methodological validation study

BackgroundSystematic reviews (SRs) are used to inform clinical practice guidelines and healthcare decision making by synthesising the results of primary studies. Efficiently retrieving as many relevant SRs as possible is challenging with a minimum number of databases, as there is currently no guidance on how to do this optimally. In a previous study, we determined which individual databases contain the most SRs, and which combination of databases retrieved the most SRs. In this study, we aimed to validate those previous results by using a different, larger, and more recent set of SRs.MethodsWe obtained a set of 100 Overviews of Reviews that included a total of 2276 SRs. SR inclusion was assessed in MEDLINE, Embase, and Epistemonikos. The mean inclusion rates (% of included SRs) and corresponding 95% confidence intervals were calculated for each database individually, as well as for combinations of MEDLINE with each other database and reference checking. Features of SRs not identified by the best database combination were reviewed qualitatively.ResultsInclusion rates of SRs were similar in all three databases (mean inclusion rates in % with 95% confidence intervals: 94.3 [93.9–94.8] for MEDLINE, 94.4 [94.0-94.9] for Embase, and 94.4 [93.9–94.9] for Epistemonikos). Adding reference checking to MEDLINE increased the inclusion rate to 95.5 [95.1–96.0]. The best combination of two databases plus reference checking consisted of MEDLINE and Epistemonikos (98.1 [97.7–98.5]). Among the 44/2276 SRs not identified by this combination, 34 were published in journals from China, four were other journal publications, three were health agency reports, two were dissertations, and one was a preprint. When discounting the journal publications from China, the SR inclusion rate in the recommended combination (MEDLINE, Epistemonikos and reference checking) was even higher than in the previous study (99.6 vs. 99.2%).ConclusionsA combination of databases and reference checking was the best approach to searching for biomedical SRs. MEDLINE and Epistemonikos, complemented by checking the references of the included studies, was the most efficient and produced the highest recall. However, our results point to the presence of geographical bias, because some publications in journals from China were not identified.Study registrationhttps://doi.org/10.17605/OSF.IO/R5EAS (Open Science Framework).

Potential of ChatGPT Schematics to Enhance Patient Understanding and Healthcare Decision-Making

Potential of ChatGPT Schematics to Enhance Patient Understanding and Healthcare Decision-Making

Cost-effectiveness of radiofrequency echographic multi-spectrometry (REMS) for the diagnosis of osteoporosis in the United States

Abstract Radiofrequency echographic multi spectrometry (REMS) is an innovative, non-ionizing diagnostic technique that has shown high accuracy and precision, making it a promising alternative to dual-energy X-ray absorptiometry (DXA) for osteoporosis diagnosis in clinical settings. With economic considerations playing an increasingly crucial role in healthcare decisions, this study aims to evaluate the cost-effectiveness and economic impact of improved osteoporosis diagnosis using REMS followed by treatment in the United States (US). A microsimulation-based Markov model was constructed to estimate the cost per quality-adjusted life year (QALY) gained (in US$2022) for REMS followed by treatment versus no diagnosis and treatment in US women aged 50 years and older with osteoporosis. Women were categorized as high risk (receiving alendronate monotherapy for 5 years) or very high risk (receiving an 18-month course of anabolic treatment, abaloparatide, followed by 5 years of alendronate). The study evaluated two medication adherence scenarios: one assuming full adherence to treatment and the other reflecting real-world adherence. The results indicate that REMS followed by treatment is associated with improved health outcomes, including more QALYs and fewer fractures, and reduced fracture-related costs compared to no diagnosis and treatment. The incremental cost-effectiveness ratio of REMS was estimated at $33 891 and $49 198 per QALY gained, under the full adherence and real-world adherence scenarios, respectively. These values are below the US cost-effectiveness thresholds of $100 000 to $150 000 per QALY. Moreover, a 5% increase in the diagnosis and treatment of women over 50 years at high and very high risk of fractures using REMS is projected to save approximately 30 000 life years, 43 500 QALYs, and prevent 100 000 fractures over a lifetime under real-world medication adherence. In conclusion, this study suggests that REMS is a cost-effective strategy for the diagnosis and management of osteoporosis in US women, offering substantial potential economic benefits and improved health outcomes.

Edge computing-based ensemble learning model for health care decision systems

A growing number of humans have suffered severe chronic illnesses, which has caused a boost in the requirement for diagnostic and medical treatment procedures that are both accurate and fast. Improved patient conditions and enhanced Decision-Making Systems (DMS) for healthcare professionals are the primary objectives of the Clinical Decision Support System (CDSS) recommended in this research article. The main drawback of traditional Machine Learning (ML) techniques is their failure to predict reliably. To solve this problem, the proposed model creates an Ensemble Extreme Learning Machine (EN-ELM) algorithm that combines predictors trained on several different data sets. This lowers the chance of overfitting. The suggested CDSS uses many different data processing methods, including Adaptive Synthetic (ADASYN) and isolation Forest (iForest), which fix problems like outliers and class imbalance. This approach significantly enhances the framework’s classification performance. Also, the CDSS is compatible with an EC model, which enables real-time computation while minimizing the requirement for integrated systems. The recommended CDSS applies iForest and ADASYN to execute large-scale trials validating high standards of accuracy across numerous datasets. Researchers concluded that a suitable ELM classification threshold of 85% is the most effective, which substantially boosts the accuracy of the predictive model. When applied to various medical datasets, such as Hepatocellular Carcinoma (HCC), Cervical Cancer, Chronic Kidney Disease (CKD), Heart Disease, and Arrhythmia, the EN-ELM achieved accuracy rates of 99.36%, 98.15%, 97.85%, 97.06%, and 96.72%, respectively. By measuring this progress, the CDSS could dramatically improve the accuracy of chronic illness diagnosis and treatment, which similarly affects clinicians.

A thematic analysis of shared decision-making in consultations with patients with a presumed brain tumor and neurosurgeons

Abstract Background Patients with a brain tumor, family members and healthcare professionals face complex healthcare decisions that can significantly impact everyday life. Engaging in shared decision-making (SDM) can help ensure treatment decisions align with patients’ preferences and goals. Little is known about the decision-making process in neurosurgical consultations. This study aims to qualitatively explore and describe the SDM process during neurosurgical consultations for patients with a presumed brain tumor. Methods The first consultation between patients with a presumed glioma or meningioma and their neurosurgeons was audio-recorded and transcribed. Decision-making models were used as sensitization concepts, with coding carried out inductively. We employed a reflexive thematic analysis to develop themes and sub-themes until saturation was beginning to emerge. Results Analyzing eleven consultations, we developed three main themes and twelve sub-themes. The results depict decision-making guided by neurosurgeons (theme 1), with varying degrees of engagement from patients and family members (theme 2) and personalization of the decision-making process based on the medical situation and everyday life priorities (theme 3). The consultations lacked specific attention to everyday life priorities of the patients. Conclusions The decision-making processes seemed to focus primarily on medically informing patients with a presumed brain tumor, in line with IDM. However, less emphasis was placed on learning more about the patient’s everyday life and preferences to integrate into the decision-making process, which is key to SDM. Incorporating patient goals may lead to greater personalization and decisions that align with both the patient’s medical situation and their goals.

Mental healthcare and pragmatic shared decision-making in general practice: An interview study.

Healthcare providers in general practice are expected to deliver mental healthcare to patients through shared decision-making (SDM). It is unclear if they perceive their SDM to be affected by challenging circumstances in mental healthcare, eg, how waiting time for therapy plays a role. To explore how healthcare providers and patients in general practice engage in SDM, given the challenging circumstances in mental healthcare. A qualitative interview study in seven Dutch general practices. Semi-structured interviews were conducted with general practitioners (GPs; N=9), practice nurses mental health (PNMHs; N=8) and patients that sought mental healthcare (N=18). The interviews were deductively and inductively thematically analyzed. The participants mainly reported on SDM regarding treatment in secondary mental healthcare. The PNMHs explained they lack an overview of available treatments and waiting times in facilities. The PNMHs therefore instruct patient to also search for options themselves. Most patients found this approach burdensome, especially those new to mental healthcare. These patients were said to often express no strong treatment preferences and rely on advice by their healthcare providers. The GPs and PNMHs explained that in such cases, they often adopt a pragmatic approach and for example refer indecisive patients to facilities with little waiting time. The healthcare providers and patients in general practice experience they adapt their approach to SDM in reaction to the circumstances in mental healthcare. Further exploration of how SDM is implemented and shaped by challenging circumstances across different healthcare settings is needed.

Norwegian and Swedish value sets for the EORTC QLU-C10D utility instrument.

This study aimed to develop utility weights for the European Organization for Research and Treatment of Cancer (EORTC) QLU-C10D, a cancer-specific utility instrument, tailored to the Norwegian and Swedish populations. The utility weights are intended for use in the specific welfare contexts of Norway and Sweden to support more precise healthcare decision-making in cancer treatment and care. This cross-sectional study included 1019 Norwegian and 1048 Swedish participants representative in age and gender of the two general populations. Participants completed a discrete choice experiment involving 960 choice sets, each consisting of two EORTC QLU-C10D health states described by the instrument's domains and the duration of each state. Utility weights were calculated using generalized estimation equation models, and non-monotonic levels were merged to ensure consistent valuation. In the Norwegian participants, the largest utility decrements were seen for the domain of physical functioning (decrement of -0.263 for highest level "very much"), followed by pain (decrement -0.205 for level "very much") and role functioning (-0.139). Among the cancer-specific domains, nausea had the largest utility decrement (-0.124). In the Swedish participants, the largest utility decrements were also observed for physical functioning (-0.207 for the response "very much"), followed by pain (-0.139), role functioning (-0.133), and nausea (-0.119). Emotional functioning also exhibited a sizable utility decrement (-0.115). This study provides the first set of utility weights for the EORTC QLU-C10D specific to Norway and Sweden, reflecting the unique health preferences of these populations. The generated utility decrements can inform cost-utility analyses and optimize resource allocation in cancer care within the Norwegian and Swedish healthcare systems.

Developing and Evaluating SEE-Diabetes: A Patient-Centered Educational Decision Support System for Diabetes Car.

This feasibility study evaluated the effectiveness of Support-Engage-Empower-Diabetes (SEE-Diabetes), a patient-centered educational tool designed to promote shared decision-making of diabetes management in older adults. We aimed to assess SEE-Diabetes's ability to facilitate patient engagement and collaborative goal setting, as measured by the Observational Patient Involvement (OPTION) scale and Shared Decision-Making Questionnaire (SDM-Q-Doc). We hypothesized that these instruments would effectively differentiate between healthcare providers who actively leveraged SEE-Diabetes to guide patient-centric conversations and set goals compared to those who did not. SEE-Diabetes, developed through a 4-year user-centered design process, was employed in simulated clinical encounters at the University of Missouri Health Care. We conducted an analysis of 12 clinical encounters using video recordings. This analysis involved three simulated patients and four providers, two internals and two externals, utilizing a mixed-methods approach. We assessed the decision-making process using SEE-Diabetes by SDM-Q-Doc, OPTION scale, and conversation analysis. The average scores for the SDM-Q-Doc and the OPTION scale, out of a possible 100, were 52.6 and 75.9, respectively. Our findings revealed that active provider engagement with SEE-Diabetes during patient interactions served as an effective medium to facilitate shared decision-making and to set patient-centered goals. Providers who actively utilized SEE-Diabetes to guide conversations, ask open-ended questions, and incorporate patient input into goal setting demonstrated significantly higher OPTION and SDM-Q-Doc scores compared to those who used the tool less frequently or primarily for documentation purposes. Providers expressed positive feedback, highlighting its conciseness, patient-centricity, and optimism about integrating SEE-Diabetes into their future practices. SEE-Diabetes showed considerable promise in improving interactions between patients and providers, presenting an innovative approach to diabetes management for older adults. This tool has the potential to not only close communication gaps but also enable patients to take a more active role in their healthcare decisions.

Accuracy of Prospective Assessments of 4 Large Language Model Chatbot Responses to Patient Questions About Emergency Care: Experimental Comparative Study.

Recent surveys indicate that 48% of consumers actively use generative artificial intelligence (AI) for health-related inquiries. Despite widespread adoption and the potential to improve health care access, scant research examines the performance of AI chatbot responses regarding emergency care advice. We assessed the quality of AI chatbot responses to common emergency care questions. We sought to determine qualitative differences in responses from 4 free-access AI chatbots, for 10 different serious and benign emergency conditions. We created 10 emergency care questions that we fed into the free-access versions of ChatGPT 3.5 (OpenAI), Google Bard, Bing AI Chat (Microsoft), and Claude AI (Anthropic) on November 26, 2023. Each response was graded by 5 board-certified emergency medicine (EM) faculty for 8 domains of percentage accuracy, presence of dangerous information, factual accuracy, clarity, completeness, understandability, source reliability, and source relevancy. We determined the correct, complete response to the 10 questions from reputable and scholarly emergency medical references. These were compiled by an EM resident physician. For the readability of the chatbot responses, we used the Flesch-Kincaid Grade Level of each response from readability statistics embedded in Microsoft Word. Differences between chatbots were determined by the chi-square test. Each of the 4 chatbots' responses to the 10 clinical questions were scored across 8 domains by 5 EM faculty, for 400 assessments for each chatbot. Together, the 4 chatbots had the best performance in clarity and understandability (both 85%), intermediate performance in accuracy and completeness (both 50%), and poor performance (10%) for source relevance and reliability (mostly unreported). Chatbots contained dangerous information in 5% to 35% of responses, with no statistical difference between chatbots on this metric (P=.24). ChatGPT, Google Bard, and Claud AI had similar performances across 6 out of 8 domains. Only Bing AI performed better with more identified or relevant sources (40%; the others had 0%-10%). Flesch-Kincaid Reading level was 7.7-8.9 grade for all chatbots, except ChatGPT at 10.8, which were all too advanced for average emergency patients. Responses included both dangerous (eg, starting cardiopulmonary resuscitation with no pulse check) and generally inappropriate advice (eg, loosening the collar to improve breathing without evidence of airway compromise). AI chatbots, though ubiquitous, have significant deficiencies in EM patient advice, despite relatively consistent performance. Information for when to seek urgent or emergent care is frequently incomplete and inaccurate, and patients may be unaware of misinformation. Sources are not generally provided. Patients who use AI to guide health care decisions assume potential risks. AI chatbots for health should be subject to further research, refinement, and regulation. We strongly recommend proper medical consultation to prevent potential adverse outcomes.

Investigation of reporting bias in interrupted time series (ITS) studies: a study protocol

Background Interrupted time-series (ITS) studies are commonly used to examine the effects of interventions targeted at populations. Suppression of ITS studies or results within these studies, known as reporting bias, has the potential to bias the evidence-base on a particular topic, with potential consequences for healthcare decision-making. Therefore, we aim to determine whether there is evidence of reporting bias among ITS studies. Methods We will conduct a search for published protocols of ITS studies and reports of their results in PubMed, MEDLINE, and Embase up to December 31, 2022. We contact the authors of the ITS studies to seek information about their study, including submission status, data for unpublished results, and reasons for non-publication or non-reporting of certain outcomes. We will examine if there is evidence of publication bias by examining whether time-to-publication is influenced by the statistical significance of the study’s results for the primary research question using Cox proportional hazards regression. We will examine whether there is evidence of discrepancies in outcomes by comparing those specified in the protocols with those in the reports of results, and we will examine whether the statistical significance of an outcome’s result is associated with how completely that result is reported using multivariable logistic regression. Finally, we will examine discrepancies between protocols and reports of results in the methods by examining the data collection processes, model characteristics, and statistical analysis methods. Discrepancies will be summarized using descriptive statistics. Discussion These findings will inform systematic reviewers and policymakers about the extent of reporting biases and may inform the development of mechanisms to reduce such biases.

The human rights act and the community nurse.

Iwan Dowie talks about the Human Rights Act, a controversial piece of statutory legislation. The article emphasises the critical role of community nurses in upholding human rights within their practice and outlines the historical context of human rights protections in the UK. The article also addresses the legal implications of resource constraints in healthcare decisions.