CareGraph Point of Care
Point-of-care knowledge support for obstetric counselling and documentation
The problem
Obstetric teams counsel women about complex clinical decisions every day: whether to induce labour for a suspected large baby, planning birth after a previous caesarean, when to escalate monitoring for a growth-restricted fetus. These conversations require clinicians to synthesise current evidence, communicate issues in plain language, and document the discussion in a way that supports continuity of care.
In practice, the quality of these conversations and documentation varies due to a range of factors. The registrar on call at 2 am may not have ready access to an experienced senior. The early career PHO may not want to make the call. Time pressure makes it difficult to recall precise numbers. And the documentation of what was discussed, what the patient understood, and what was agreed upon is sometimes incomplete or even absent.
When counselling is inadequate, women make decisions without the information they need. When documentation is poor, the next clinician may be left without a clear record of the discussion. And the patient safety consequences are not insignificant for patients, clinicians or organisations.
The underlying causes are rarely about the individual clinician. The issue lies in the conditions of care: how units are staffed, how training time is protected, how information is made available when it's needed. Structured, accessible resources reduce clinicians' cognitive load and make it easier to deliver high-quality care in real-world settings.
What this project is
A collection of structured clinical snippets covering the 30 most common counselling and documentation scenarios in obstetric practice. Each snippet has three components:
A concise summary of the clinical question, the key trials, guideline positions, and the numbers a clinician needs to know. Written for the clinician, not the patient, with full references.
Plain-language examples of approaches to explain the issue, present options, and invite the patient's perspective. Not scripts. Starting points that clinicians adapt to their own style and the individual woman's needs.
A structured, copy-pasteable template for the medical record that captures what was discussed, the options presented, the patient's questions and preferences, and the agreed plan.
The snippets are delivered as a mobile-responsive web application with one-tap copy for eMR documentation. Designed for use in clinic, on the ward, or on the birth suite at 2 am.
Why it matters
Across an archive of detailed retrospective analyses of adverse events, documentation adequacy was a contributing factor in 75% of cases. Informed consent failures appear in almost one-half. These are patterns that point to conditions a structured resource can address directly.
Patient safety
The same counselling and documentation gaps appear repeatedly in adverse obstetric outcomes. Structured resources address these patterns directly.
Patient autonomy
Effective counselling is the foundation of informed consent. Each snippet models collaborative, non-directional counselling and links to verified consumer resources.
Clinician support
Front-line clinicians are expected to counsel on a wide range of scenarios, often with limited experience and time. The snippets support clinical judgement rather than replacing it.
Education
Each snippet is grounded in current evidence with full references to primary trials, Cochrane reviews, and college guidelines, including limitations.
How the CareGraph Point of Care is developed
Topic selection is driven by frequency of appearance in detailed retrospective case analysis, depth of coverage in the evidence literature, and clinical importance in routine obstetric practice.
Evidence synthesis draws on a curated library of over 460 obstetric sources (peer-reviewed trials, Cochrane reviews, college guidelines, and clinical governance documents) and an archive of review findings spanning the major obstetric clinical scenarios.
Artificial Intelligence tools accelerate evidence synthesis, generate initial drafts, cross-checks clinical claims against source documents and performs scheduled searches. All content is reviewed, corrected, and refined by the clinical author. The clinical judgement, voice, and final authority rest with the expert clinician.
Research context
This project sits at the intersection of clinical decision support, patient safety, and AI-assisted knowledge tool development. Two related questions are of interest:
How can AI tools, working with a curated clinical evidence base and expert clinician oversight, produce structured knowledge resources that are accurate, current, and useful at the point of care? What is the development workflow, and how does it scale across clinical topics?
Do obstetric teams find the snippets useful? Do they change counselling behaviour or completeness of documentation? Can the impact on documentation quality be measured through an audit? What are the barriers to adoption?
The project is in early development, with the first snippet (suspected fetal macrosomia/LGA) complete and a 30-topic roadmap mapped to the evidence base. Feedback from obstetric clinicians is being sought to refine the format and prioritise subsequent topics.