The Concepts used in the GNUmed EMR

Help translate this into Deutsch Español Français Polska Português

In brief, the concepts and key references are…

  • encounter-oriented chronology
  • problem-oriented documentation (POMR)
  • SOAP structured progress notes
  • optional aggregation of encounters into episodes of care
  • optional aggregation of episodes into health issues
  • classification of diagnostic certainty
  • problem list
  • Lawrence L. Weed

If you know what this is all about, you can stop reading here, and jump to GmManualStartingGnumed.

What they all mean…

The concepts used in the GNUmed EMR align (un?)surprisingly well with those written up by the UK NHS for the CEN/ISO Concepts for Continuity of Care.

The Encounter

In health care, an encounter is commonly understood as a distinct contact of a patient with the health care system. In a GP setting, most encounters start when the patient enters the building and end when the patient leaves the building. Contact with a doctor may or may not have taken place. An encounter need not end on the day it started, e.g. when care is given over midnight the encounter will span a date boundary. Likewise there may well be two or even three encounters in one day (think of patients you have seen in and sent home from the practice in the morning, only to have to admit into hospital later in the day).

Technically, in the GNUmed EMR, an encounter need not always include a patient visit, or even doctor (or other human) interaction. They are a fusion of a conceptual participant <==> health care related interaction and a technical EMR (database) session.

In addition to the usual patient-provider physical visits, then, the following situations are considered encounters, too:

  • provider accesses the EMR without the patient being physically present, as in the recording of a phone interaction
  • office assistant accesses the EMR, to take care of a task
  • automated processes, such as an importer scripts for lab data adds information to the medical record

Inanspruchnahme, Arzt-Patienten-Kontakt

Each encounter can span a time interval and thereby conveniently group together (interconnect) multiple pieces in a single patient's record. This does not prevent later reassignment of any of these pieces (which keep their date and time stamps) to a different encounter, where it would make sense to do so.

Reason for Encounter

RFE, Reason for Visit, Reason for Consultation.

This may evolve from the original purpose at the time that the encounter was planned, such as a chief complaint or planned followup, to include additional purposes or tasks that may come up in a visit as needing also to be looked after.

Beratungsursache (nicht -anlaß !)
"keifende Nachbarin" (Nachbarin: "Du nervst mich mit Deinem ewigen Gejammer über den Rücken ! Geh doch mal zum Doktor !")
"Rückenschmerz" (Patient: "Herr Doktor, mir kneift es schon 3 Wochen im Rücken !")

Assessment of Encounter


This offers the ability to capture a larger insight or formulation for the encounter as a whole and where multiple problems may have been touched upon, for example "Malignancy now explains all' or "General decompensation due to renal failure" or "Increasing frailty, close to failing at home".


Problem-Oriented Documentation of Care

In a problem-oriented medical record (POMR), all stored clinical data has been associated with an explicitly-stated problem suffered by a patient. Problems need not be diagnoses. They need not be hard scientific facts. They can be syndromes, they can be findings, they can be history items. Over time, they are likely to merge and consolidate into well-founded diagnoses. Or they may not. That is the beauty of GP-level health care.

Note that during one single encounter, several problems with the patient's health can be dealt with.

The SOAP Schema

In 1964 Lawrence L Weed introduced the SOAP structuring of progress notes in medical records. This concept roughly says that all clinical data associated with giving care to a patient is to be grouped into the categories Subjective, Objective, Assessment, and Plan. Various criticisms have been put forth as to where this classification lacks sophistication or falls short of properly capturing clinical information. However, setting aside academically-proper validation and evaluation, most clinical data at the GP level can be grouped into one of:

what the patient narrates (Deutsch: Anamnese)

what findings were elicited at the encounter in question (Deutsch: Befunde)

what clinical meaning is given to the Subjective and Objective data, as to causes and consequences for the patient's health (Deutsch: Bewertung)

what do the patient and clinician intend or agree to do about the patient's health (Deutsch: Procedere)

Each problem will, in GNUmed, have its own, dedicated, SOAP-structured data.

Classification of Diagnostic Certainty

Based upon Subjective and Objective data, a provider will formulate an Assessment of the patient's state of health. This assessment is often captured as a short catchy phrase which, in GP-level care – despite being referred-to as a diagnosis – may start out as a summary of the state of health, and may only reach a professionally-formed, scientifically-founded working diagnosis, and not one that attains a scientific or gold-standard proof. GNUmed has (as of 0.6) a provision for labelling assessments at the level of Episode and Health Issue with a classification of the diagnostic certainty of that assessment. The literature describes four levels of certainty:

A - sign
a single symptom such as back pain, or a single sign such as elevated BP, which implies and may even be denoted "Not Yet Diagnosed" (NYD). (Deutsch: Symptom)
B - cluster of signs
a group of signs often seeming to go together but not yet safely recognizable as a particular disease as to be diagnosable, such as rash and fever and still implying and more often denoted NYD. (Deutsch: Symptomgruppe)
C - syndromic diagnosis
a group of presenting signs and findings are specific enough to allow for a reasonably likely presumptive diagnosis, even without scientific proof, as yet eg. Strep throat in the absence of throat swab results. (Deutsch: Bild der Krankheit)
D - scientific diagnosis
the exact ailment has been adequately scientifically proven by microbiological culture, lab results, autopsy, histology etc. for example Scarlet Fever in ASL +ve Strep throat. (Deutsch: Diagnose)

Such a codification might have value to flag patients who may have symptoms or abnormalities that have recurred or persisted without explanation, but which might (if taken together) point more clearly to a diagnosis.

The Episode of Care

Encounters are often clustered in time. In the course of possibly several encounters, a number of health problems will be worked on, incorporated into one (or more) episode(s) of care. It is entirely at the discretion of the clinician how long the episode lasts. Usually an episode will only be assumed to be "case closed" after the patient did not report back for an extended period of time, or when some knowledge of resolution has somehow come to the clinician.

Each episode of care may comprise one or several encounters. While GNUmed does not yet model this graphically, an analogy based on an example at Dipity, may inform. Encounters would distributed from left to right along the patient's time line. Episodes would each form a horizontal stripe, with the various stripes "stacked" vertically, with each episode positioned at an _arbitrary height, with a left edge determined by the episode start date, and a right edge defined by the episode's end date, except if the episode remains open / active.

An episode may be associated with a health issue. Each episode can also be labelled with a diagnostic certainty.

Note how a (clinically defined) episode is different from a (administrative) period of care, an example of which might be the duration of a hospital stay within the course of one episode, or the duration of a given treatment regimen.

The Health Issue

At times the clinician will recognize a cluster or subset of distinct episodes of care as looking suspiciously related. In such cases it may make sense to group them under one health issue. Thus, health issues may be at the fundamental or foundational level of a patient's health. They may be active or inactive. Post-MI state is likely to be clinically relevant for the rest of the patient's life, despite that it may not be an active problem at a given time. On the other hand, a traumatically amputated finger will always be both clinically relevant and active if it confers continuous disability. Health issues will more often be understood as diagnoses than will "problems" or "episodes". Therefore, health issues can be further associated with a diagnostic certainty. In GNUmed, past medical history items will mostly be stored as health issues.

Setting aside simple "past history" items, each health issue will aggregate one or several episodes of care.

The Problem List

The problem list (the list of active problems) consists of items being worked on, or kept in mind, while trying to improve the health of the patient. This list includes:
  • the clinically relevant health issues (whether or not these would have an open episode)
  • the open episodes (whether these would be unattributed, or attributed to a health issue, even if the health issue was not clinically important or "relevant")
  • by contrast, closed (inactive) episodes are not included i.e. are not shown in GNUmed as "problems", nor are those health issues that – while historically undeniable – have gone unmarked as clinically relevant

Putting Things Together

The structure of a patient's EMR can be seen as a tree:

* each health issue will aggregate, within it, clinically-related episodes of care
  • episodes will have data added to them during one or several encounters
    • such data is grouped into the SOAP schema

* an early diagram of this is hosted at the internet archive, here.

* a further treatment of how this works, but which you may rather skip for the moment, is initiated at EncounterEpisodeIssue

Next: Starting GNUmed


(sorted by lastnames of authors)

  • Bayegan E, Nytrø O. A problem-oriented, knowledge-based patient record system. Stud Health Technol Inform. 2002;90:272-6. (PMID: 15460701)
  • Bayegan E, Nytrø Ø, Grimsmo A. Ranking of information in the computerized problem-oriented patient record. Medinfo. 2001;10(Pt 1):594-8. (PMID: 11604806)
  • Bayegan E, Tu S. The helpful patient record system: problem oriented and knowledge based. Proc AMIA Symp. 2002;:36-40. (PMID: 12463782)
  • Blumenthal-Barby K, Fichtner N. [EDV (computers) within the scope of health- and social care and their utilization for solving of tasks in non-numerical information processing. Notes on paper by J. Peil (Zschr. ärztl. Fortbild. 67, 1973, 261] Z Arztl Fortbild (Jena). 1973 Nov 1;67(21):1071-5. German. (PMID: 4590770)
  • Claus PL, Carpenter PC, Chute CG, Mohr DN, Gibbons PS. Clinical care management and workflow by episodes. Proc AMIA Annu Fall Symp. 1997;:91-5. (PMID: 9357595)
  • Crouch MA, Thiedke CC. Documentation of family health history in the outpatient medical record. J Fam Pract. 1986 Feb;22(2):169-74. (PMID: 3484779)
  • Donnelly WJ, Brauner DJ. Why SOAP is bad for the medical record. Arch Intern Med. 1992 Mar;152(3):481-4. No abstract available. (PMID: 1546910)
  • Feinstein AR. The problems of the "problem-oriented medical record". Ann Intern Med. 1973 May;78(5):751-62. (PMID: 4711779)
  • Goldfinger SE. The problem-oriented record: a critique from a believer. N Engl J Med. 1973 Mar 22;288(12):606-8. (PMID: 4568844)
  • Haber P, Röggla G, Braun RN. Classification of reasons for consultation and results of consultation in a selected sample from specialized pulmonary outpatient care * Wien Klin Wochenschr. 1989 Nov 24;101(22):767-8.* (PMID: 2609657)
  • Hartmann F. Beschreibung und/oder Benennung krankhafter Vorgänge. Der Allgemeinarzt 1983;5:362–70 und 498–507.
  • Hofmans-Okkes IM, Lamberts H. The International Classification of Primary Care (ICPC): new applications in research and computer-based patient records in family practice. Fam Pract. 1996 Jun;13(3):294-302. Review. (PMID: 8671139) fulltext
  • Holmes C. The Problem List beyond Meaningful Use. Part 1: The problems with problem lists. J AHIMA 2011 Feb;82(2):30-3; quiz 34. fulltext (PMID: 21337850)
  • Holmes C. The Problem List beyond Meaningful Use. Part 2: Fixing the problem list. J AHIMA. 2011 Mar;82(3):32-5; quiz 36. fulltext (PMID: 21413516)
  • Huibert J. Tange, MD, PhD?, Harry C. Schouten, MD, PhD?, Arnold D. M. Kester, PhD? and Arie Hasman, PhD?. The Granularity of Medical Narratives and Its Effect on the Speed and Completeness of Information Retrieval. Journal of the American Medical Informatics Association 5:571-582 (1998) fulltext
  • Hurst JW, Walker HK, Hall WD. More reasons why Weed is right. N Engl J Med. 1973 Mar 22;288(12):629-30. (PMID: 4568846)
  • Klinkman MS, Green LA. Using ICPC in a computer-based primary care information system. Fam Med. 1995 Jul-Aug;27(7):449-56. (PMID: 7557010)
  • Kühlein Th, Maibaum Th, Gensichen J, Engeser P. Dokumentation und Kodierung in der Hausarztpraxis - ein Lösungsvorschlag nach Wegfall der neuen Kodierrichtlinien. Z Allg Med. 2011; 87 (10) fulltext (DOI 10.3238/zfa.2011.0400)
  • Lamberts H, Hofmans-Okkes I. Episode of care: a core concept in family practice. J Fam Pract. 1996 Feb;42(2):161-9. (PMID: 8606306)
  • Lamberts H, Hofmans-Okkes I. The core of computer based patient records in family practice: episodes of care classified with ICPC. Int J Biomed Comput. 1996 Jul;42(1-2):35-41. (PMID: 8880267)
  • Linnarsson R, Nordgren K. A shared computer-based problem-oriented patient record for the primary care team. Medinfo. 1995;8 Pt 2:1663. (PMID: 8591533)
  • Meyer RL. Zum 90. Geburtstag von Robert Nikolaus Braun, dem grossen Pionier der wissenschaftlichen Allgemeinmedizin. PrimaryCare 2004;4:20–3. (fulltext)
  • Nygren E, Henriksson P. Reading the medical record. I. Analysis of physicians' ways of reading the medical record. Comput Methods Programs Biomed. 1992 Sep-Oct;39(1-2):1-12. (PMID: 1302665)
  • Rakel RE. The problem-oriented medical record (POMR). Am Fam Physician. 1974 Sep;10(3):100-11. (PMID: 4416438)
  • Schmidt EC, Schall DW, Morrison CC. Computerized problem-oriented medical record for ambulatory practice. Med Care. 1974 Apr;12(4):316-27. (PMID: 4823649)
  • Singer EP. Why SOAP is good for the medical record?: another view. Arch Intern Med. 1992 Dec;152(12):2511; author reply 2511, 2514. (PMID: 1456869)
  • Soler JK. Sick leave certification: a unique perspective on frequency and duration of episodes - a complete record of sickness certification in a defined population of employees in Malta. BMC Fam Pract. 2003 Mar 27;4:2. Epub 2003 Mar 27. (PMID: 12697050 )
  • Son RY, Taira RK, Bui AA, Kangarloo H, Cardenas AF. A context-sensitive methodology for automatic episode creation. Proc AMIA Symp. 2002;:707-11. (PMID 12463916)
  • Verbeke M, Schrans D, Deroose S, De Maeseneer J. The International Classification of Primary Care (ICPC-2): an essential tool in the EPR of the GP. Stud Health Technol Inform. 2006;124:809-14. (PMID: 17108613)
  • Wall EM. Continuity of care and family medicine: definition, determinants, and relationship to outcome. J Fam Pract. 1981 Oct;13(5):655-64. (PMID: 7024464)
  • Weed LL. Medical records that guide and teach. N Engl J Med. 1968 Mar 14;278(11):593-600. (PMID: 5637758)
  • Weed LL. Medical records that guide and teach. 1968. MD Comput. 1993 Mar-Apr;10(2):100-14. (PMID: 8469091)
  • Wingert TD, Kralewski JE, Lindquist TJ, Knutson DJ. Constructing episodes of care from encounter and claims data: some methodological issues. Inquiry. 1995-1996 Winter;32(4):430-43. (PMID: 8567080)
  • Wright et al. Clinician attitudes toward and use of electronic problem lists: a thematic analysis. BMC Medical Informatics and Decision Making 2011 11:36. fulltext (PMID: 21612639; DOI: 10.1186/1472-6947-11-36)

Topic revision: 04 Sep 2018, KarstenHilbert
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback
Powered by Olark