Clinical notes are the written backbone of modern healthcare. Every consultation, ward round, procedure, and handover generates documentation that serves multiple simultaneous purposes: recording what happened, guiding what happens next, protecting the clinician legally, and communicating across care teams who may never meet in person. Despite their centrality to clinical practice, the standards, formats, and expectations around clinical notes vary considerably, by setting, by specialty, and increasingly, by the technology available to produce them.

What are clinical notes?

Clinical notes are the written or structured records that clinicians create to document patient encounters, observations, clinical decisions, and care plans. They serve two distinct functions: as a legal record of the care provided, and as a communication tool that supports continuity across the wider care team.

A clinical note may be as brief as a two-line progress update during a ward round, or as detailed as a comprehensive discharge summary covering a complex inpatient admission. What unites them is their role as the authoritative account of clinical activity. Other clinicians rely on them when taking over a patient's care, commissioners and insurers use them to verify activity, and courts may examine them if care is disputed.

Why clinical notes matter in modern healthcare

Accurate clinical documentation supports more than record-keeping. Good documentation directly supports continuity of care, informed clinical decision-making, medico-legal protection, and, in many systems, reimbursement and coding accuracy.

When notes are incomplete, ambiguous, or delayed, the consequences are practical and sometimes serious. Clinicians picking up a patient mid-episode may lack the context to make safe decisions. Referrals may be rejected or delayed. Clinical coding, which drives both resource allocation and payment in many healthcare systems, depends on structured, accurate documentation.

There is also a well-documented tension between documentation quality and the time required to produce it. Documentation burden is a significant contributor to clinician burnout, with high-quality discharge summaries alone described as contributing substantially to administrative load. This tension sits at the centre of most current conversations about clinical documentation reform.

Types of clinical notes

SOAP notes

The SOAP format, which stands for Subjective, Objective, Assessment, Plan, is the most widely used structured note format across primary and secondary care. Each section serves a distinct purpose:

• Subjective: The patient's reported symptoms, history, and concerns in their own words

• Objective: Measurable clinical findings, including vital signs, examination results, and investigation data

• Assessment: The clinician's interpretation, including working diagnosis or differential diagnoses

• Plan: The intended management, including investigations, treatments, referrals, and follow-up

SOAP notes are valued for their logical structure, which makes them easy to scan and act upon. They are particularly prevalent in general practice and outpatient settings, though clinicians use them across virtually all clinical disciplines.

POMR notes (problem-oriented medical records)

Problem-oriented medical records organise documentation around a patient's active problem list rather than chronological encounter records. A clinician reviewing a problem-oriented medical record can immediately identify the current active problems and trace the relevant history for each, rather than reading through a series of dated entries. This format is particularly useful in complex patients with multiple long-term conditions, where a purely chronological record can obscure the clinical picture.

Progress notes

Progress notes are the ongoing record of a patient's status during a care episode. Clinicians use them during ward rounds, follow-up consultations, and throughout inpatient admissions to document changes in condition, responses to treatment, and evolving clinical decisions. Progress notes form a continuous thread through an episode of care, allowing any clinician joining the team to understand where the patient is in their treatment journey.

Discharge summaries

The discharge summary is the structured handover document produced at the end of an inpatient admission. It covers the presenting diagnosis, investigations performed, treatment delivered, and follow-up instructions, and clinicians typically share it with the patient's general practitioner and any relevant outpatient services. High-quality discharge summaries are essential for safe care transitions, yet they are among the most time-consuming documents clinicians produce. Research comparing AI-generated and physician-written discharge summaries in a Dutch academic hospital found that fully automated, medical record system-integrated tools can produce summaries of comparable quality, though robust real-world validation of such systems remains an active area of study.

Referral letters

Referral letters are the clinical communications sent when transferring a patient's care to a specialist or secondary care service. A well-constructed referral summarises the relevant history, current presentation, investigations already performed, and the specific question being put to the receiving clinician. In systems using Advice and Guidance pathways, referral correspondence also serves as a direct channel for clinician-to-clinician consultation without a formal onward referral.

Patient letters

Patient letters are written communications addressed directly to the patient, summarising consultation outcomes, next steps, or test results in accessible, non-technical language. The distinction between a clinical note and a patient letter matters: the former is written for clinicians, the latter for patients. Patients increasingly have direct access to their clinical records, which has prompted greater attention to the language and tone used in clinical documentation more broadly.

Nursing notes

Nursing notes capture observations, interventions, and care delivery from a nursing perspective. They form a distinct but complementary layer of the clinical record, documenting the continuous monitoring and care that nursing staff provide between physician assessments. In inpatient settings, nursing documentation is often the most frequent and granular record of a patient's condition over time.

Procedure and operative notes

Procedure and operative notes are the structured records created immediately following a clinical procedure or surgery. They document the technique used, intraoperative findings, any complications, and post-procedure instructions. These notes carry significant medico-legal weight and clinicians are typically expected to complete them immediately after the procedure, while details are fresh and accurate.

Common formats for clinical documentation

Free-text notes

Free-text notes allow clinicians to write in natural language without a fixed structure. They offer flexibility, particularly valuable when a clinical situation does not map neatly onto a predefined template, but carry well-documented limitations for data extraction, interoperability, and consistency. A free-text note may be perfectly legible to the clinician who wrote it while being difficult for others to parse, search, or code.

Structured and templated notes

Structured notes use predefined fields and templates to ensure consistency across clinicians and encounters. A multicentre retrospective study comparing 144 unstructured versus 144 structured medical record system notes found that structured documentation significantly improved note quality, with mean scores rising from 6.83 to 7.52 out of 10, as well as conciseness and clarity. A 2025 audit study aligned with National Institute for Health and Care Excellence clinical documentation guidelines found similar improvements in completeness and consistency following the introduction of standardised templates, and noted that hybrid models combining structured fields with narrative sections best supported clinical workflow flexibility.

Structured notes also support clinical coding, which depends on consistent, extractable data fields. The trade-off is that rigid templates can feel constraining in complex or atypical cases, and there is a documented risk of note bloat, where templates encourage over-documentation of irrelevant detail while genuine clinical reasoning is crowded out.

Voice-dictated notes

Voice dictation has been a feature of clinical documentation for decades, particularly in secondary care settings where clinicians dictate notes for transcription by medical secretaries. The traditional model introduces a delay between the encounter and the completed note, and depends on transcription accuracy. It remains common in some specialties, notably radiology and pathology, where structured reporting adoption varies by setting and jurisdiction, with dictation workflows persisting in some institutions.

AI-generated notes

AI-assisted documentation is a more recent development in which an AI medical assistant listens to a consultation in real time and drafts a structured clinical note from the conversation. This approach, sometimes described as ambient voice technology, differs from traditional dictation in that it captures the natural flow of a consultation rather than requiring the clinician to narrate directly to a recording device. Figures cited in industry reporting suggest AI documentation tools can achieve 95 to 98 per cent accuracy and reduce documentation time by 50 to 70 per cent, though these figures should be interpreted with reference to the specific tools and settings evaluated. Expert evaluation of AI-generated discharge letter summaries has found that automated metrics often inadequately capture clinical relevance and safety, underscoring the importance of clinician review of AI-generated content.

Clinical notes across care settings

Primary care (GP practices)

General practice operates under significant documentation pressure. General practitioners (GPs) document high volumes of short encounters, often ten to fifteen minutes each, while managing complex, longitudinal patient histories. SOAP notes and structured templates are the dominant formats, and documentation burden in primary care is widely cited as a driver of clinician burnout. The combination of high volume, time constraint, and the need for accurate clinical coding makes general practice one of the settings where structured documentation tools and AI-assisted approaches have attracted the most interest.

Secondary care and hospitals

Hospital documentation spans a wider range of note types: ward round progress notes, multidisciplinary team records, operative notes, and discharge summaries, often recorded across multiple legacy systems that do not communicate with each other. The complexity of inpatient documentation, and the number of clinicians contributing to a single patient record, makes consistency and legibility particularly important, and particularly difficult to achieve.

Mental health services

Clinical documentation in mental health carries specific considerations beyond those in other specialties. Notes may include risk assessments, therapeutic observations, and sensitive personal disclosures that require careful handling. The format of mental health notes varies: SOAP is used in some services, while formats such as BIRP (Behaviour, Intervention, Response, Plan) are more common in behavioural health settings. The sensitivity of mental health records also raises particular questions about access, sharing, and the language used. Patients increasingly have access to their own records, and clinical language that is appropriate in a professional context may be experienced differently by the patient reading it.

Private and specialist care

In private and specialist settings, patient letters and detailed referral correspondence tend to carry more weight than in National Health Service primary care. Patients in private care often expect comprehensive written summaries of their consultations, and the standard of documentation in referral letters, both outgoing and incoming, reflects on the quality of the clinical service. Detailed, well-structured documentation also supports the insurance and billing processes that are more central to private care workflows.

Best practices for writing clinical notes

Write contemporaneously

The clinical and legal importance of documenting as close to the encounter as possible is well established. Delayed notes introduce risk: memory fades, details are lost, and a note written hours or days after an encounter may not accurately reflect what was observed or decided at the time. Timely documentation is a core standard in most clinical governance frameworks, and retrospective reconstruction of events from memory is considered poor practice.

Be accurate, objective, and specific

Clinical notes should use precise, factual language, documenting what was observed, measured, or reported rather than what the clinician assumed or inferred without clinical grounding. Ambiguous language, subjective commentary not rooted in clinical observation, and vague descriptors such as "seems unwell" or "doing better" reduce the utility of a note for other clinicians and create medico-legal exposure. Using standardised terminology, including Systematised Nomenclature of Medicine (SNOMED) codes and International Classification of Diseases (ICD) classifications where applicable, supports both clarity and data reuse.

Use consistent structure and templates

Standardised templates reduce variability across clinicians and encounters, support clinical coding, and make records easier to act on. Research on standardised note templates in medical education found that structured templates with embedded guidance improved both confidence and efficiency in note-writing, suggesting that the benefits of structure extend to clinicians at all career stages. The key is selecting templates that match the clinical workflow rather than imposing structures that clinicians work around.

Avoid excessive abbreviations

Abbreviations are common in clinical practice and can speed up documentation, but non-standard shorthand creates real risk in handovers and cross-specialty communication. An abbreviation that is self-evident within one specialty may be ambiguous or misleading in another context. Clinical documentation guidelines generally recommend limiting abbreviations to those on an approved, organisation-wide list.

Ensure notes are accessible to the whole care team

A clinical note that only makes sense to the clinician who wrote it has limited value. Notes should be written with the assumption that colleagues outside the original specialty will read them, including GPs receiving discharge summaries, nurses acting on care plans, and clinicians covering out of hours. Framing documentation as a shared structured language for care continuity is a useful lens for evaluating whether a note is fit for purpose.

Review and amend correctly

When a clinical note requires correction, the appropriate approach is to add a dated amendment rather than overwrite or delete the original entry. This preserves the integrity of the record and makes the timeline of clinical decisions transparent. Overwriting original entries is considered poor practice in most clinical governance frameworks and can create significant medico-legal problems if the record is later scrutinised.

How AI is changing clinical documentation

Ambient voice technology and AI medical assistants are beginning to shift the model of clinical documentation in meaningful ways. Rather than writing notes retrospectively after the patient has left, often under time pressure, clinicians using AI-assisted tools can capture documentation in real time during the consultation itself, with the AI generating a structured draft note from the conversation.

Prospective evaluation of AI-generated hospital course summaries has assessed both the safety and the burden-reduction potential of these tools in real clinical settings. Early findings suggest that AI-generated summaries can be of comparable quality to physician-written ones, though the evidence base is still developing and expert evaluation frameworks are being established to assess clinical relevance and safety in ways that automated metrics cannot capture. Research has also demonstrated that large language models can provide quality improvement feedback on clinical notes, identifying gaps in documentation that might otherwise go unaddressed.

The shift from retrospective note-writing to in-consultation capture has implications for note quality as well as clinician time. Notes drafted in real time are less likely to suffer from recall errors, and the clinician's attention can remain on the patient rather than on the screen. AI-generated notes require clinician review and sign-off, as they are drafts rather than final records, and the quality of the output depends significantly on the accuracy of the underlying transcription and the appropriateness of the template being populated.

The evidence on AI documentation tools, while growing, is not uniformly positive. Study populations, clinical settings, and the specific tools evaluated vary considerably, and most prior evaluations have been limited to drafts, small cohorts, or non-integrated settings. Clinicians and organisations adopting these tools should seek evidence from real-world, medical record system-integrated deployments rather than relying solely on vendor-reported accuracy figures.

Data security, privacy, and compliance in clinical notes

Clinical notes contain some of the most sensitive personal data that exists. The obligations around their creation, storage, and sharing are substantial, and clinicians adopting new documentation tools, including AI-assisted ones, need to understand what those obligations require.

In the UK and European Union, clinical documentation falls under the General Data Protection Regulation (GDPR), which requires that patient data is processed lawfully, stored securely, and not transferred outside approved jurisdictions without appropriate safeguards. Data residency, meaning where data is processed and stored, is a particular consideration when using cloud-based AI documentation tools. Clinicians and organisations should verify that any AI tool used in clinical documentation processes data within the required geographic boundaries and under appropriate data processing agreements.

Key considerations when evaluating AI documentation tools from a compliance perspective include:

• Whether the tool holds relevant certifications, such as ISO 27001 for information security management

• Whether it meets the requirements for a medical device under applicable regulation, such as the UK Medical Device Regulation or EU Medical Device Regulation, depending on its intended function

• What access controls are in place to ensure that clinical notes are accessible only to those with a legitimate clinical need

• How data is retained, deleted, and audited

Medical record system documentation standards also recommend periodic documentation audits to ensure that records meet organisational and regulatory requirements, not just at the point of adoption of a new tool, but as an ongoing governance process.

What good clinical documentation looks like

Good clinical notes share a consistent set of characteristics regardless of the format, setting, or technology used to produce them. They are:

• Timely: written as close to the encounter as possible, with amendments dated and transparent

• Accurate and objective: grounded in clinical observation and measurement, not assumption or subjective commentary

• Structured: using consistent formats and templates that support readability, coding, and data reuse

• Accessible: written to be understood and acted upon by the whole care team, not just the originating clinician

• Secure: stored and shared in compliance with GDPR and relevant national frameworks, with appropriate access controls

These principles apply whether a note is handwritten, typed, dictated, or drafted by an AI assistant. Ambient voice technology and AI-assisted documentation are increasingly part of clinical practice, but the standards that define a good clinical note remain grounded in the same fundamentals: accuracy, clarity, timeliness, and a genuine commitment to the patient's safety and continuity of care.

Frequently asked questions

▶ What are clinical notes and what are they used for?

Clinical notes are the written or structured records that clinicians create to document patient encounters, observations, clinical decisions, and care plans. They serve two distinct functions: as a legal record of the care provided, and as a communication tool that supports continuity across the wider care team. Other clinicians rely on them when taking over a patient's care, commissioners and insurers use them to verify activity, and courts may examine them if care is disputed.

▶ What are the main types of clinical notes?

The main types include SOAP notes (Subjective, Objective, Assessment, Plan), problem-oriented medical records, progress notes, discharge summaries, referral letters, patient letters, nursing notes, and procedure or operative notes. Each type serves a distinct purpose, from documenting a brief ward round update to providing a comprehensive handover at the end of an inpatient admission.

▶ What is the SOAP note format and when is it used?

SOAP stands for Subjective, Objective, Assessment, Plan. It's the most widely used structured note format across primary and secondary care. The Subjective section captures the patient's reported symptoms and concerns, the Objective section records measurable clinical findings, the Assessment section contains the clinician's interpretation and working diagnosis, and the Plan section sets out the intended management. SOAP notes are particularly common in general practice and outpatient settings, though clinicians use them across virtually all clinical disciplines.

▶ What's the difference between free-text notes and structured notes?

Free-text notes allow clinicians to write in natural language without a fixed structure, which offers flexibility but carries well-documented limitations for data extraction, interoperability, and consistency. Structured notes use predefined fields and templates to ensure consistency across clinicians and encounters. Research has found that structured documentation significantly improves note quality, conciseness, and clarity, and better supports clinical coding. The trade-off is that rigid templates can feel constraining in complex or atypical cases, and there's a documented risk of over-documentation crowding out genuine clinical reasoning.

▶ How does AI-assisted documentation work?

An AI medical assistant listens to a consultation in real time and drafts a structured clinical note from the conversation. This approach, sometimes described as ambient voice technology, differs from traditional dictation in that it captures the natural flow of a consultation rather than requiring the clinician to narrate directly to a recording device. AI-generated notes require clinician review and sign-off, as they are drafts rather than final records. The quality of the output depends significantly on the accuracy of the underlying transcription and the appropriateness of the template being populated.

▶ What are the best practices for writing clinical notes?

Good clinical notes should be written as close to the encounter as possible, use precise and factual language grounded in clinical observation, follow consistent structures and templates, avoid non-standard abbreviations, and be written so that colleagues outside the original specialty can understand and act on them. When a note requires correction, the appropriate approach is to add a dated amendment rather than overwrite or delete the original entry.

▶ Why does documentation burden contribute to clinician burnout?

Documentation burden is a significant contributor to clinician burnout because producing high-quality records takes substantial time, often outside of patient-facing hours. General practitioners, for example, document high volumes of short encounters while managing complex, longitudinal patient histories. High-quality discharge summaries alone are described as contributing substantially to administrative load. This tension between documentation quality and the time required to produce it sits at the centre of most current conversations about clinical documentation reform.

▶ What compliance and data security considerations apply to clinical notes?

Clinical notes contain some of the most sensitive personal data that exists. In the UK and European Union, clinical documentation falls under the General Data Protection Regulation, which requires that patient data is processed lawfully, stored securely, and not transferred outside approved jurisdictions without appropriate safeguards. When evaluating AI documentation tools, organisations should check whether the tool holds relevant certifications such as ISO 27001, whether it meets medical device regulation requirements, what access controls are in place, and how data is retained, deleted, and audited.

▶ How do clinical notes differ across care settings?

In primary care, general practitioners document high volumes of short encounters using SOAP notes and structured templates, with documentation burden widely cited as a driver of burnout. In hospitals, documentation spans a wider range of note types across multiple systems, often contributed to by many clinicians. Mental health services carry specific considerations around risk assessments, sensitive disclosures, and patient access to records. In private and specialist care, detailed patient letters and referral correspondence carry more weight, and comprehensive documentation also supports insurance and billing processes.

▶ Is AI-generated clinical documentation reliable enough to use in practice?

The evidence on AI documentation tools is growing but not uniformly positive. Early findings suggest that AI-generated summaries can be of comparable quality to physician-written ones, though study populations, clinical settings, and the specific tools evaluated vary considerably. Expert evaluation frameworks are being established to assess clinical relevance and safety in ways that automated metrics can't capture. Clinicians and organisations adopting these tools should seek evidence from real-world, medical record system-integrated deployments rather than relying solely on vendor-reported accuracy figures.