FreshRSS

🔒
❌ Acerca de FreshRSS
Hay nuevos artículos disponibles. Pincha para refrescar la página.
AnteayerTus fuentes RSS

Use of AI to enhance written information in paediatric settings--stochastic parrot or clinical tool?

Por: OSullivan · C. · Gaddum · C. · Lee · A. J.

Commentary on: Brewster RCL, Gonzalez P, Khazanchi R, et al. (2024) Performance of ChatGPT and Google Translate for Pediatric Discharge Instruction Translation. Pediatrics. 2024;154(1).

Implications for practice and research

  • Artificial intelligence (AI) has significant potential to impact healthcare, but further research is required to evaluate validity of translation for communication.

  • Enhanced clinician communication is needed for non-native speaking patient groups, and AI translations may be useful. However, content must reflect true clinical ‘meaning’.

  • Context

    Today’s world is increasingly augmented by AI, and use and application of natural language process-based translation services could enhance clinical communications. They have the potential to support confidential and cost-effective communication mechanisms for non-native language patients. This study1 evaluates quality, cultural sensitivity, terminology, context, consistency and risks associated with use of AI translators for Spanish, Portuguese and Haitian paediatric patients, illustrating shortcomings in quality, accuracy and patient...

    Osteoporosis treatment gap and prescribing patterns in Ireland: a cross-sectional analysis of the DXA HIP project

    Por: Carey · J. J. · Brennan · A. · Armstrong · C. · Heaney · F. · Dempsey · A. · Egan · R. · Gorham · K. · Yang · L. · DXA MAP · Erjiang · Wang · Whelan · OSullivan · Silke · Yu · Dempsey · Fitzgerald
    Objectives

    Previous studies suggest Ireland has the smallest osteoporosis treatment in Europe and very little inappropriate prescribing, in contrast to our experience. In this study, we examine the osteoporosis treatment gap in Ireland by assessing the prevalence of appropriate and inappropriate prescribing in 2 subgroups of the Irish dual-energy X-ray absorptiometry (DXA) Health Informatics Prediction (HIP) Project. Treatment eligibility was defined using established intervention thresholds, including prior fracture, femoral-neck T-score ≤–2.5, glucocorticoid use, or Fracture Risk Assessment Tool (FRAX) major osteoporotic fracture risk ≥20% or hip fracture risk ≥3%.

    Design

    Secondary cross-sectional analysis of a subgroup of the DXA HIP Project Cohort.

    Setting

    3 hospitals in the West of Ireland. DXA referrals come from primary care providers, hospital consultants and the osteoporosis service.

    Participants

    5564 participants of a previously described convenience cohort including: (i) 3474 subjects referred for a DXA scan, and (ii) 2090 patients who completed a DXA scan.

    Results

    82.4% were female with a mean age of 66.6 years, 59.6% of whom had a prior fracture. Prescribing data of calcium and vitamin D were available for 3738 (67.2%) subjects, and osteoporosis medication for 4157 (74.7%) subjects. Prescribing information was available for more than 99% of the DXA group, but just over 50% of the referral group. When examined in aggregate, the treatment gap is 6% for calcium and vitamin D and 38% for osteoporosis medication, in line with prior publications. However, among those with prescribing information and at least one indication for treatment, only 58.3% were prescribed calcium and vitamin D and 39.1% an osteoporosis medication. Furthermore, among patients without a clear indication for treatment, 50.6% were prescribed calcium and vitamin D, and 32.5% an osteoporosis medication.

    Conclusions

    These data suggest the majority of patients with osteoporosis or at high risk of fracture in Ireland today do not receive appropriate osteoporosis treatment, while inappropriate prescribing is substantial. These findings suggest that the true treatment gap in Ireland is substantially larger than aggregate estimates imply.

    Nuts For Babies Study: protocol for a randomised controlled trial in Australia investigating if the risk of developing peanut and cashew nut allergies during infancy can be reduced by a high peanut and cashew nut maternal diet for the first 6 months of la

    Por: Sullivan · T. R. · McWilliam · V. · OSullivan · M. · Netting · M. · Perrella · S. · Geddes · D. · Tang · M. · Campbell · D. E. · Perrett · K. P. · Palmer · D. J.
    Introduction

    The predisposition to food allergy development and the induction of allergen-specific immune responses appears to be initiated early in infancy. Early exposure to food allergens, such as peanut and cashew nut, via human milk is likely important in initiating oral tolerance and reducing risk of food allergy development. This trial aims to determine if the risk of developing peanut and cashew nut allergy during infancy can be reduced by a high peanut and cashew nut maternal diet during lactation.

    Methods and analysis

    This is a multisite, parallel, two-arm (1:1 allocation), single-blinded (outcome assessors, statistical analyst and investigators), randomised controlled trial. Target sample size is 4412 participants (2206 per group). Women (aged 18–50 years) with a singleton pregnancy, who are planning to breastfeed and do not have peanut and/or cashew nut allergies are eligible to participate. After obtaining written informed consent, participants are randomised to either a high peanut and cashew nut diet (at least 60 peanuts and 40 cashew nuts per week) or a low peanut and cashew nut diet (no more than 20 peanuts and 12 cashew nuts per week). Participants are asked to follow their allocated diet from birth to 6 months postnatal. Individual lactation consultant advice and support is provided as required. The study’s primary outcome is food challenge proven IgE-mediated peanut and/or cashew nut allergy during infancy (0–18 months). Key secondary outcomes include infant sensitisation to peanut and/or cashew nut. Analyses will be performed on an intention-to-treat basis according to a prespecified statistical analysis plan.

    Ethics and dissemination

    Ethical approval has been granted from the Western Australian Child and Adolescent Health Service Human Research Ethics Committee (approval number RGS0000006685). Trial results will be presented at scientific conferences and published in peer-reviewed journals.

    Trial registration number

    Australian New Zealand Clinical Trials Registry (ACTRN ACTRN12624000134527)

    Clinical practice guidelines for the care of people experiencing chronic primary pain: protocol for a systematic review with interpretation against an established chronic pain care priority framework

    Por: Briggs · A. M. · Siegfried · N. · Waller · R. · OConnell · N. E. · Romero · L. · Klem · N.-R. · Ampiah · P. K. · Belton · J. L. · Blyth · F. M. · De Morgan · S. · Lord · S. M. · Nicholas · M. · OSullivan · P. B. · Shakya · A. · Smith · A. J. · Slater · H.
    Introduction

    Most clinical practice guidelines (CPGs) for assessing and managing people’s chronic pain focus on specific pain conditions, body sites or life course stages. This creates complexity for clinicians making care choices in the absence of a diagnosis and/or where a person experiences more than one pain condition. Specific to this context is the ICD-11 classification of chronic primary pain where an experience of pain cannot be better accounted for by another condition. CPGs for chronic primary pain, agnostic to condition or body part, may support clinicians towards best pain care since many of the principles of person-centred chronic pain care are transdiagnostic. The two aims of this systematic review are to (1) identify and appraise CPGs for chronic primary pain, relevant across the life course and (2) map the CPG content against a pain care priority framework to evaluate the extent to which the CPG content aligns with the priorities of people with lived chronic pain experience.

    Methods and analysis

    We will systematically search nine scholarly databases, the Epistemonikos database and international and national guidelines clearinghouses. CPGs published within 2015–2025, in any language, that offer recommendations about assessment and/or management of chronic primary pain for people of any age, excluding hospitalised inpatients or institutionalised populations, will be included. Pairs of reviewers will independently screen citations for eligibility and appraise CPG quality and implementation potential using the Appraisal of Guidelines for Research and Evaluation (AGREE)-II and the AGREE-Recommendations Excellence tools, respectively. Data extraction will include the citation and scope characteristics of each CPG, methods used to develop recommendations, verbatim recommendations, guiding principles or practice information and narrative excerpts related to the GRADE Evidence-to-Decision (EtD) considerations (or equivalent). We will use the PROGRESS-PLUS framework as a checklist to identify whether determinants of health equity were considered by guideline developers. CPG recommendations will be organised according to common topics and categorised in a matrix according to strength and direction. Qualitative content analysis will be used to synthesise excerpts relating to GRADE EtD considerations (or equivalent), and we will map extracted data against an established chronic pain care priority framework to determine the extent to which the CPGs align with values and preferences of people with lived experience. Interpretation will be informed by an interdisciplinary Advisory Group, including lived experience partners.

    Ethics and dissemination

    Ethical approval is not required for this systematic review. Results will be disseminated through publication in an open-access peer-reviewed journal, through professional societies, and integrated into education curricula and public-facing resources. Reporting will be consistent with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement.

    PROSPERO registration number

    CRD420251000482.

    ❌