Health Articles

 

The international classification of primary care (icpc)

USERS GUIDE
ICPC-2 PLUS
This guide consists of five Sections outlining the advantages of coding andclassifying your data, giving guidelines to the use of ICPC-2 PLUS and anumber of Appendices which are updated with each release.
You may print parts of the guide, the whole document (≈ 54 pages) or use itas an "on line" version on your computer screen.
To navigate through the guide click the to jump to thatSection or Appendix.
The beginning and bottom of most pages has a link to bring you back here.
Family Medicine Research Centre
Department of General Practice
University of Sydney
in co-operation with World Organisation of Family Doctors
ICPC – 2 PLUS Users' Guide ICPC-2 PLUS
Family Medicine
Research Centre
ICPC-2
 World Organisation of National Colleges, Academics and Academic Associations of General Practitioners/Family
ICPC-2 PLUS and ICPC-2 PLUS User's Guide
 University of Sydney, 1998
Family Medicine Research Centre All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the Family Medicine Research Centre, University of ICPC – 2 PLUS Users' Guide The International Classification of Primary Care (ICPC-2)
Copyright 1987 WONCA
ICPC-2 PLUS
Copyright 1994 by the Family Medicine Research Centre, University of Sydney
All rights reserved. No part of this publication may be reproduced, stored on a retrieval system, ortransmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwisewithout prior written permission from the Family Medicine Research Centre, University of Sydney. Forinformation, address the Family Medicine Research Centre, Department of General Practice, Universityof Sydney; Acacia House, Westmead Hospital; WESTMEAD NSW AUSTRALIA 2145 LICENSING FEES FOR GENERAL PRACTICE (AUSTRALIA AND NEW ZEALAND):
WONCA SITE LICENSE COSTS (once only payment):
The licence fee is payable to the FMRU, who pass payment on to WONCA at regularintervals. The monies received by WONCA will assist the Classification Committee incontinuing its work in further developing classification systems (including ICPC-2) forgeneral practice. ANNUAL SITE LICENCE FEE FOR ICPC-2 PLUS FOR GENERAL
PRACTICE (AUSTRALIA AND NEW ZEALAND):
This fee will cover the provision of regular database updates. If you feed back terms youfind difficult or impossible to find we will add them in time for the next upgrade.
The University of Sydney is a non-profit organisation and any surplus will be utilised forresearch and development of classification systems for general practice.
OUTSIDE AUSTRALIA AND NEW ZEALAND:
BY QUOTATION
ICPC – 2 PLUS Users' Guide TABLE OF CONTENTS THE INTERNATIONAL CLASSIFICATION OF PRIMARY CARE –STRUCTURE AND

ICPC – 2 PLUS Users' Guide SECTION 1: As published in Australian Family Physician Vol 24. No 4. April 1995. pages 612-615
General Practice Medical Records:
Why code? Why classify?
Helena Britt BA, PhD, Director, Family Medicine Research Centre, Department of General Practice, University of Sydney Neil Beaton MBBS, MRCGP, DA. Medical Advisor, The Aboriginal Primary Health Care Project, Far North Queensland Division of General Practice. Graeme Miller MBBS, FRACGP. Clinical Senior Lecturer, Department of General Practice, University of Sydney. Recently the Information Management Steering Group (IMSG), [a
RACGP-AMA-Commonwealth Government committee responsible
for the planning of information management in general practice],
held a Coding Workshop at which available coding systems and
their application in general practice computerised medical records
were reviewed. As there has been in the past some discussion as
to the value of coding, the workshop participants agreed that a
paper outlining the reasons for coding and classifying clinical data
should be prepared and disseminated to all general practitioners.
Over 95 million general practice service rebates are claimed through Medicare every year. These services are provided by over 20,000 medical practitioners including 16,000 recognised general practitioners. Eighty two percent of the population visit a general practitioner at least once a year and each patient visits a GP an average five times per year (data obtained for the General Practice Branch, Department of Human Services and Health). However the management of the information recorded in the course of these services is generally poor. As a result very little of this large data ICPC – 2 PLUS Users' Guide source provides information which is useful to general practitioners for comprehensive and continuing patient and community care. Clinicians need reliable clinical data: to provide quality patient management; for patient audit; for quality assurance and for practice management. Group information for the practice or the community also aids the management of specific groups of patients.
Further, data collected for the primary purpose of patient care have secondary benefits.
When collected and aggregated it can be used for clinical, health services and health economics research. The use of a computer system will improve the quality of patient medical records2 as they will be always be legible and will usually have greater structure than paper based systems. However the collection of information alone is not sufficient to provide meaningful data. In order to make information useful to individual practitioners, the practice or to epidemiologists, it must be easily accessed.
Why code the data?
Having decided to introduce computerised medical records into your practice you will be
faced with the question of whether or not to code the data. Without codes computers will allow you to access information stored in free text by use of word search mechanisms. These allow you to sort your patients into groups, such as "all patients who have the word hypertension in their record". However the margin for error is great for you must ask the computer to search for every term you may have used to describe this diagnosis. The search will miss words which you have misspelt; and you will miss abbreviations (now forgotten) which you entered in a hurry during a rushed surgery.
For analysis across the practice you must also be aware of all terms which may have been used by your partner or locum. A code is a shorthand for a concept. It accurately compresses the data for storage. In the computer records environment most agree that it is unsatisfactory to store information only in free text, if you wish to retrieve the data and collate it at a later date.
ICPC – 2 PLUS Users' Guide At a workshop held by the RACGP during 1993 specific reasons for coding data in medical records were identified. They include: to provide an audit trail; practice audit; memory prompts; continuity of care; assisting other carers in the practice (eg: partners and locums); quality assurance; better record keeping; decision support and protocols; periodic and incidental health checks; checking for disease-medication interactions; dynamic structured patient records; cost savings. The use of codes does not preclude the use of synonyms, acronyms and key words to describe a concept in a medical record but ensures that the concept represented by the code is uniform for all practitioners. This results in greater consistency of data input and reliability of reporting. Clarity of communication in a common language, whether between partners or between primary and secondary care providers will increase quality of care. Just as importantly it facilitates accurate and speedy data retrieval. If a code has been attached to the label at the time of the encounter, a single search for one code will provide a list of all patients with the disease of interest, no matter how you described it in the medical record. In addition to aiding in the selection of records for groups of patients, coding aids the linkage of events over time within a patient record. Using a single code to represent the disease or problem, during a consultation you can ask the computer to show you all encounters for this patient at which code X was managed, and view them consecutively on the screen. This is far easier than searching through pages and pages of records, (whether on paper or computer) when trying to build a historical view of the progress of a disease or of its management.
ICPC – 2 PLUS Users' Guide Coding may also have some disadvantages which include the impact it has on the style of practice of the doctor. It requires the practitioner to think more specifically about the term selected to describe a problem. Coding systems must therefore be sufficiently flexible to allow individual preferences in terminology whilst maintaining accurate interpretation by others. Like computerised records, the introduction of coding systems requires a new form of discipline and a new way of thinking about records by Some people suggest that coding the problems and their management is an extra job for the clinician.2 They inevitably describe the time consuming task of looking up a list, selecting the code and then entering it on the record manually. Think again. Things are fast changing. In most software systems available overseas, and in an increasing number in Australia the concept of "terming" has been introduced. Terming refers to the entry of a few key letters (eg: osteo), an acronym (OA) or a brief key term to access a list of terms which should include the one you are looking for. You select the term
from the pick list by a single key stroke (ie terming). The selected label and its code are automatically entered and stored by the computer. In the future, as hand written recognition systems are refined and become less expensive, you will be able to use the same process with the tip of your pen. Coding is regarded by some as only useful in the diagnostic area. However, increasingly coding systems will include codes for all sorts of information in your medical records including the details of the drugs you prescribe, therapeutic procedures, pathology results, family history, risk factors etc. This will allow you to undertake more complex record reviews. eg: to investigate the number of patients who are on risk levels of a selected medication for a selected disease; develop reliable recall systems for preventive or follow-up care; check that none of your patients on NSAIDs also suffer from asthma. It is hoped that codes which facilitate the use of warning systems for allergies, adverse effects, possible contraindications and drug interactions, will also soon be available. This cannot help but assist general practitioners in the provision of quality care.
ICPC – 2 PLUS Users' Guide Classification
The next question is "why classify?"
A classification is merely a method of placing the codes in a sorted and meaningful manner. A good structure allows you to manage the data within the practice in terms of groups of codes rather than just a specific individual code.
We use structured coding systems in many areas of our lives. Using an example within a student registration system, a typical hierarchical system may be: Student ID number is 9435267: 94. . . = year of entry; 3. . = faculty of science; 5. . = school of physics; 267 = student number. With this structured coding system you can easily identify: the number of students who first enrolled in 1994 at the university; (sort on 94 only) the number who first enrolled in the faculty of science in 1994; (943) the number specifically studying physics in the faculty in 1994.(9435) the individual student Applying the same thinking to general practice where D = digestive disease; D1 = digestive infection; D116 = specifically gastroenteritis, you can select records at all levels: all patients who have at least one digestive disease; all who have had a digestive infection; and all who have suffered specifically from gastroenteritis. If the codes are not structured in a hierarchical manner the search at the gastroenteritis level is easy but those above become increasingly difficult. Without a hierarchy a search for digestive problems would require you to list all the codes which you feel fall into that group.
Without a code you would have to list and search for all the possible terms you may have used to describe a digestive problem of any type. Codes which are classified make any data retrieval easier.
ICPC – 2 PLUS Users' Guide One classification system or many?
There are many levels at which data retrieved from medical records are useful:
• individual patient information in general practice• . information about groups and families in a general practice• . community or practice population information• . regional health information (eg: Divisions)• . state and national information• . international comparisons The level of specificity required by each of these participants in the health care system varies considerably. Ideally one universal coding system would meet the needs of all these competing interests, but this is not usually the case. For example a coding system which classifies diseases in terms of broad group headings only (eg: digestive; cardiovascular etc.), is of little use in the GP surgery when you wish to record, code and store minute details about the health event which has occurred to your patient. On the other hand, the fine detail of a patient record is of little use to a regional health planner who may wish to allocate resources on the basis of broad parameters such as a body system heading. The decision of which coding system must be determined by the
level of detail required at the first entry point, in this case the patient medical record. Where the chosen classification does not suit all needs, multiple systems can be utilised through a process of "mapping". This is the process of working out the relationship between individual codes in each of the systems. Usually the more detailed system is "mapped" to the less detailed. Thus a group of codes, or multiple individual codes from a variety of sections in one system may be placed together under one code in another.
The majority of internationally recognised classifications are mapped to multiple other classifications. For example, the International Classification of Primary care (ICPC)3 has a body system based structure and is ideal for population based general practice data analysis. The International Classification of Diseases4 has a disease based structure, far more specific codes and is designed for hospital data systems. The latter is mapped ICPC – 2 PLUS Users' Guide to the former so that data collected in terms of ICD can be analysed in terms of ICPC.
Another example is the Read Clinical codes which are mapped to both ICD-9 and to The mapping process is the responsibility of the classification designer and allows people using different classification systems, possibly at different levels of care, to transfer information in a common language. It is therefore to your advantage to select a coding system which is internationally recognised, mapped to other systems, and designed for use in general practice.
For meaningful data retrieval there is one more facet of computerised medical records which needs to be considered in parallel to the selection of a coding system or systems.
The medical record must be structured in a manner which allows the computer to recognise the true meaning of the code or term. For example, the code for "Ca-breast" has a very different meaning in the family history section of the record to its meaning in the patient problem list. Any data retrieval requires consideration of the sector of the record in which the term or code is noted. The RACGP manual record system has three structural elements based on the work of Lawrence Weed5 and both can be effectively applied to computerised systems. Firstly the problem orientation allows linkage of a problem over time and tracks changing problem definition. Secondly the SOAP structure of the data within the encounter - subjective data, objective data, assessment and plan- which assures differentiation between lable meanings. Thirdly, the patient summary which includes an up-to-date list of all important morbidity and its ICPC – 2 PLUS Users' Guide Summary
Classification systems are being increasingly adopted by general practitioners
throughout the world, particularly in the fast growing number of practices utilising computerised medical records. The advantages of coding data at the point of entry have been outlined in this paper and general practitioners in Australia should consider these points when thinking of using computers, whether for full medical records, recall systems, direct printing of prescriptions or age-sex disease registers. The implications for later meaningful data retrieval to assist practitioners in the provision of quality care for their patients are enormous. Dick SR, Steen EB. (ed). The computer-based patient record, an essential technology for health care. Institute of Medicine. National Academy Press.
Washingtom DC 1991.
2. Regan B, Ireland M. Inappropriateness of coding systems for recording clinical data. Informatics in Healthcare Australia July 1994;3(2):101-014.
3. Wood M, Lamberts H, Meijer JS, Hofmans-Okkes IM. The conversion between ICPC and ICD-10. Requirements for a family of classification systems in the next decade. Fam Pract 1992; 9: 340-348. 4. WHO.(ed) International classification of diseases (9th revision). Geneva: World Health Organisation, 1977. 5. Weed LL, Hertzberg R. Clinical application of medical software for problem solving in ambulatory care J Ambulatory Care Manage 1985; 8: 66-83. ICPC – 2 PLUS Users' Guide AS PUBLISHED IN INFORMATICS IN HEALTH CARE AUSTRALIA.
September/October 1996 Vol 5, No 4.
Which code? Which classification?
In the past few years, there has been considerable discussion about whether or not coding data in computerised clinical systems is necessary or even desirable1,2,3.
More recently, in parallel with wider adoption of computerised clinical systems in primary care such discussion has subsided, suggesting a broader acceptance of the need to code and classify. However, as it appears that the Commonwealth Government is unlikely to make a directive decision about a set of "preferred classifications" for primary care in the foreseeable future, it is probably time to review the advantages and disadvantages of those that are available. While this paper concentrates on morbidity classifications we should not lose sight of the need to code other fields in the patient's record. In the recently completed NSW data modeling exercise for Community Health, over 400 data items which could be classified were identified. These included some fields that clinicians automatically "classify". For example: • the type of consultation - in general practice this is classified with the Commonwealth Medical Benefits Schedule (CMBS); • patient characteristics such as age and sex are automatically classified (through Many other patient characteristics such as country of birth, language spoken at home etc. can be classified according to standardised systems provided by the Australian Bureau of Statistics (ABS ).
However, the area presenting the greatest classification problems to primary care providers is clinical information. Thinking in terms of the SOAP structure recommended by the RACGP, we have a wide range of information which ideally
would be classified according to accepted national and international standards: ICPC – 2 PLUS Users' Guide S = Subjective
• Patient reasons for encounter (the patient expressed reasons for attending on that occasion). These may be in terms of the need for a service (check up, referral, etc.), symptoms (headache), or a diagnostic label (about my • presenting symptoms (information collected by the GP in the diagnostic O = Objective findings
A = Assessment (problem label or diagnosis)
P = Plan (including prescribing, therapies and other treatments).
While all of these fields should be able to be coded and classified, the majority of recent interest has centred on the classification of the patient reasons for encounter, the presenting systems, and the problem labels or diagnoses. When coding, one should always keep in mind the purpose. Coding should be used to facilitate logical and meaningful data retrieval, to find groups of patients for practice audit, follow up and recall systems, or to gain a view of your practice population and its morbidity. The classification you choose should therefore have sufficient specificity to allow you to select and save the term you want but not be so specific that it cannot be used with consistency and reliability.
THE INTERNATIONAL CLASSIFICATION OF DISEASES is the oldest and most
widely recognised diagnostic classification available4. The 10th edition is used widely
in the Australian hospital system (ICD 10AM) where trained coders receive paper records from clinicians and secondarily classify the diagnostic data. ICPC – 2 PLUS Users' Guide ICD began its development in the late 19th century as an international list of causes of death. It therefore has an emphasis on disease in terms of its aetiology, pathology and morphology. In primary care many of the problems presented and managed remain ill-defined at the end of the consultation and it is difficult to classify them in such a disease oriented system. In a comparative study of classification systems it was estimated that almost half the problems dealt with in primary care could be not classified with ICD6. In the late 1980's when New Zealand family practitioners were told to code using ICD 9, they simply stopped coding because it was far too difficult.
All looked forward to the new version (ICD 10) which was said to overcome this problem. Alas, the experts agree it fails to do so.
Advantages:
ICD is an international classification widely used in tertiary institutions in Australia
and elsewhere.
Disadvantages:
• The rubric (i.e. the ICD description of the medical concept) is often a false
terminology with little relationship to the natural language of clinicians, having been designed for secondary, rather than clinician coding.
• ICD is a static classification. It is only revised every 10-15 years so it does not keep up to date with changing medical terminology and the discovery of new • It lacks sufficient rubrics for the many ill-defined conditions managed in primary ICPC – 2 PLUS Users' Guide THE READ CLINICAL CODES:
The Read codes have been described in more detail elsewhere7 but in summary:
they are a comprehensive nomenclature of medical terms derived from international classifications such as ICD 94 and OPCS 4 (a classification of surgical operations and procedures, similar to our CMBS).8 The version presently used in the UK (READ 5 byte set) includes over 100,000 preferred terms and 150,000 synonyms. The original developer, a GP named James Read, started with a list of terms he used in his practice and asked GPs to try and use it. As they requested the addition of other specific terms they were added to RCC. It is regularly updated each three months in response to GP needs.
Read codes suffer from confusing version terminology. A brief history is tabulated No of Codes
brief problem list GP record summary Hospital record summary Full medical record Full health record The 5 byte set had two versions of data file structure, version 1 and 2. Read Version 3 represents a term set, code set and file structure. Future versions will be labelled Advantages: (Read 5 byte set)
• It was designed in general practice, and therefore includes terms used by GPs.
• It was the first system designed for computers rather than for paper based secondary coding; • It is a hierarchical system with five levels, each level being more specific.
• Its quarterly updates make it dynamic rather than static ICPC – 2 PLUS Users' Guide Disadvantages:
• The five level structure means you can go down one path looking for the term you
want and not realise there is a more appropriate term somewhere else in the hierarchy. This can lead to poor inter-practitioner reliability in the coding of the same medical concept. • It is sometimes far too specific: For example:- you can code "hit simultaneously by two trains moving in opposite directions".
• There is therefore a high signal to noise ratio which often makes it difficult to locate the term you want. • In other areas it lacks sufficient specificity for clinical purposes (particularly in the psychological and social areas).
• Analysing data using Read can also be difficult because of its size and its ICD From a practical viewpoint there are other things about Read that Australian primary care providers should consider. The Aus Read Trial9 demonstrated that:- • many of the Read "preferred terms" are not suitable in the Australian environment — the synonymous terms are more appropriate • the hierarchy is not always suitable (e.g.: asthma is classified as a specific type of • many of the key words, (i.e. words that the GP enters in order to find the term they want), need to be Australianised Since the trial the Australian Government has not bought a National license for the codes so they will remain fully controlled from the UK. Individuals who wish to buy a license for Read Clinical Codes can do so by contacting the distributors direct. There will be no Australian back-up nor production of an Australian version unless Australia negotiates a licence. Those considering buying New Zealand medical record software which uses Read should be aware of this.
ICPC – 2 PLUS Users' Guide Read Version 3 is even larger than its predecessor. It has been broadened to cover terms used in the entire health record. It includes many specialty codes which have been added after a $40 million project in which 40 professional colleges designed a term set for its specialty. It has 250,000 terms organised in a hierarchical structure one could describe as a "tree". Classifications can be "laid over" the top of the terms, so you can analyse data almost any way you wish. However it is not yet in widespread practical use. Software systems using Read 3 are still under development. Some feel that general practice terms have been somewhat lost in the THE INTERNATIONAL CLASSIFICATION OF PRIMARY CARE (ICPC)
In the 50's and 60's a number of countries (including Canada and the UK) attempted
to develop new classifications specifically for primary care. In the early seventies, the newly formed Classification Committee of the World Organisation of Family Doctors (WONCA) decided it was time to develop a new international classification for primary care. This resulted in the International Classification of Health Problems in Primary Care (ICHPPC), the second version having the added advantage of inclusion and exclusion criteria10. However, it still lacked sufficient codes for ill- defined conditions and patient expressed reasons for encounter largely because it retained the ICD structure which confined its flexibility. In 1978 WHO set up a working party to develop a classification system for patient reasons for encounter. Most of its members were also members of WONCA. This work resulted in the International Classification of Primary Care (ICPC)11 which incorporated codes for patient reasons for encounter, symptoms and ill-defined conditions, with the addition of the morbidity codes from ICHPPC2 (Defined).10 ICPC – 2 PLUS Users' Guide The structure of ICPC differs from that of ICD and Read. It has a biaxial structure with 17 chapters on one axis , mainly based on body systems with an additional chapter for broad, ill defined conditions (e.g.: feeling tired; general ill feeling etc.), another for psychological problems and one for social problems. On the other axis are 7 components. Component 1 covers symptoms and complaints. Component 7 covers diagnosis/disease and components 2-6 are process codes (e.g. check, immunisation, test results) which apply equally in all chapters. It was designed for paper based data collection with the primary care provider selecting the code at the time of the encounter.
Advantages:
• Its structure follows the natural process of primary care and facilitates access to
meaningful morbidity groups (e.g.: all cardiovascular disease, all respiratory symptoms; all skin infections; all injuries; all preventive care; all immunisations).
• It is small enough to handle, having only 1300 rubrics Disadvantages
• Because it was designed as an epidemiological tool, it only includes a specific
rubric for the most common problems managed in general practice. The less common problems are placed into "rag-bag" codes such as ‘other diseases of the respiratory system' or "other digestive symptom".
• The published version of ICPC (now out of print) has a poor index. Therefore, when a practitioner cannot find the term in the index s/he has to make a decision about where it should best be classified. This leads to a lack of coding reliability which has repercussions when you later wish to later compare data from multiple • For computerised clinical systems the disk copy of the book lacks sufficient specificity for medical records or even for disease registers and recall systems. A prime example is: all types of diabetes are grouped in one rubric (diabetes) - yet for quality care, for legal reasons and for sheer convenience you need to have recorded IDDM or NIDDM in the patient's record, not just diabetes. Another example is the combination of HIV+ and AIDS together in one code.
• Like ICD, it is a static system which is only reviewed each decade. ICPC – 2 PLUS Users' Guide ICPC PLUS:
Recognising the advantages of ICPC as an analytical tool and its disadvantages in
terms of its lack of specificity for clinical systems the Family Medicine Research Centre, University of Sydney developed an extended version of the ICPC. The extension was based on data recorded by GPs about more than 800,000 GP-patient encounters, collected during the Australian Morbidity and Treatment Survey (AMTS)12, the Country-Metropolitan Comparison Study13 and a popular quality assurance option the Morbidity and Therapeutic index. All the terms recorded by GPs to describe patient reasons for encounter and the problems managed were classified according to the ICPC. However, each term was given its own extension code. As with Read Clinical Codes, multiple key words were attached to each term to facilitate easy access to terms. On entry of a key word (e.g. OA) a pick list is offered to the user who highlights the term required and "clicks" or hits "return" to select it.
This is called "terming" rather than "coding". The clinician should hardly be aware that a "code" is attached - that is the computer's job. Used in suitable computerised clinical system ICPC PLUS allows the clinician to save the term as selected - it does not replace that term with a "higher" description of the concept. Like Read, ICPC PLUS is updated quarterly with additions made in response to users' requests.
USAGE OF THESE CLASSIFICATIONS:
ICD9 CM (A) is being used throughout the hospital system in Australia and many
other countries. Some primary care software relies on selected sub-groups of its available codes. It is available through the National Coding Centre. Read Clinical Codes are being used throughout general practice in the UK and are
being trialed in New Zealand in both primary and secondary care. About 15 practices in Australia use them, most having adopted them as part of the Demonstration Practice Trials, funded by the Commonwealth Department of Human Services and Health. However, since the trials are finished these practices will now require a license for Read which can be obtained from their UK distributor, Computer Aided Medical Services.
ICPC – 2 PLUS Users' Guide ICPC: is being used by between 20 and 40 practices in Australia, usually having
been provided with a disk copy of the codes only . Without the book which described the philosophy of ICPC the users may have difficulties when needing to choose a code for an unlisted concept. Throughout Scandinavia and in parts of North America ICPC is being used in combination with ICD 10. While ICPC is not available in book form at present , a disk copy of the book can be obtained from the Family Medicine Research Centre, University of Sydney. The revised version, ICPC 2, will be published in 1997 through Oxford University Press.
ICPC PLUS: is being used by 45 general practices in Australia and one in Fiji; the
Department of Veterans' Affairs for the Health Care Plans; and the national hypertension study (ANBP2). A feasibility study of its application in Community Health Centres is under way, funded by a consortium of NSW, SA, ACT and QLD State Health Departments. It is being considered by the RFDS, Aboriginal Health, Northern Territory and Victorian Community Health. Under Federal Government funding it is presently being mapped to ICD 9 CM(A), so that data from different sources can be compared. ICPC PLUS is distributed by the Family Medicine Research Centre.
OTHER SYSTEMS
There are some other coding systems which should be mentioned for completeness:
SNOMED is an internationally designed classification with its origins in pathology. It
works on a combination of pathophysiology, histopathology and anatomical site. It is a constructionist model which allows the development of highly specific codes which may be extremely useful in pathology. The level of specificity it allows goes far beyond the interests of primary care. Further, the system allows you to "build" nonsensical constructs such as "broken heart", "fractured eye". Its structure may be suitable to the reductionist theory of specialist practice but is the very antithesis of holistic care. Further, while SNOMED has enthusiastic supporters throughout the world, it is not widely used in general practice .
ICPC – 2 PLUS Users' Guide UMLS (Universal Medical Language System) is strictly speaking a database
access system, not a classification. The database includes both medical and scientific terms. It originated with the keywords from Medline and recently the terms used in many classification systems have been added. It is a search engine which was designed to contain sufficient terms (including words such as "aluminium") to allow access to information from multiple sources using different medical terminology. It is, however, being used experimentally in several hospitals in the United States for the coding of medical problems.
In addition to these international systems, there are a multitude of Australian home- grown coding systems which do not have an international basis. The most well known of these is Docle which is based on the Linnean model, invented some 200
years ago for the classification of species. Unlike other systems it does not utilise numeric codes. It was described at a recent conference14 as made up of two core concepts: " The first core concept of Docle is an algorithm that converts a piece of real world medical vernacular into a standard abbreviation. For example 'diabetes mellitus' is repackaged by the Docle algorithm as 'diabetesMellitus' before it maps to the Docle word 'diabm'. The second core concept is that of operators. Docle words can be combined together to form any number of complicated expressions by combining Docle terms with operators. For example a fracture of the radius can be expressed as 'frac.radi' - the dot operator translates to 'located at'." Docle is being used by one Australian GP software vendor.
ICPC – 2 PLUS Users' Guide CONCLUSION
When selecting a coding system you need to consider the issues raised above. All
classifications have some disadvantages but the most important issues are: • breadth of coverage of primary care• ease of access to the required term• national and international acceptance and comparability• ease of "sorting" the information for analysis and identification of patient groups.
Note that the classification you choose should be "mapped" (with authorisation) to ICD 10 if transfer of information across the health care system is ever to come to The computerisation of primary care is at the beginning of an exponential curve. Any acceptable classification system will cost you money. Think before you buy.
About the Author
Helena Britt is an Associate Professor and the Director of the Family MedicineResearch Centre, Department of General Practice University of Sydney. She is aclinical psychologist with over twenty years research experience, the majority inhealth services research in general practice. She has a PhD in general practice.
Her interests centre in improved information management and analyticaltechniques in primary care and the classification of clinical data.
ICPC – 2 PLUS Users' Guide References:
1. Regan B, Ireland M. Inappropriateness of coding systems for recording clinical
data. Informaitcs in Healthcare - Australia. 1994; 3(3):101-4.
2. Britt H. Recording clinical data - the advantages of coding - a response.
Informatics in Healthcare - Australia 1994; 3(4): 149-152.
3. Britt H, Beaton N, Miller G. Coding and classification in computerised general practice medical records: Why code? Why classify? Aust Fam Physician 1995;24: 612-615.
4. World Health Organisation. International classification of diseases (9th revision) Geneva: World Health Organisation, 1978.
5. World Health Organisation: International classification of diseases and related health problems, 10th revision. Geneva, World Health Organisation 1992.
6. Westbury RC, Tarrant M. Classification of disease in general practice: A comparative study. Can Med Assoc J 1969; 101:82.
7. Miller G, Britt H. Data collection and changing health care systems. 1. United Kingdom. Med J Aust 1993; 159; 471-475.
8. Classification of Surgical Operations and Procedures, Version 4. 9. London: Office of Population Census and Survey; Operations, 1987.
9. Family Medicine Research Centre, University of Sydney. The Aus-Read Trial.
Report to the Evaluation Steering Group, General Practice Branch, Canberra:Department of Human Services and Health. August 1994.
10. Classification Committee of WONCA. ICHPPC-2 defined. (International classification of health problems in primary care). Oxford: Oxford UniversityPress, 1983.
11. Lamberts H and Woods M (eds). ICPC - the International classification of primary care. Oxford: Oxford University Press, 1987.
12. Bridges-Webb C, Britt H, Miles DA, Neary S, Charles J, Traynor V. Morbidity and treatment in general practice in Australia 1990-1991. Med J Aust 1992; 157Suppl: S1-S56 13. Britt H, Miles DA, Bridges-Webb C, Neary S, Charles J, Traynor V. A comparison of country and metropolitan general practice. Med J Aust 1993, 159 Suppl: S9-S64.
14. Oon YK. The Linnean Model of Medical Classification. In the Proceeedings of the Fourth National Health Informatics Conference, pp153-9. Melbourne: HealthInformatics Society of Australia, August 1996.
ICPC – 2 PLUS Users' Guide SECTION 2: FOR PRIMARY CARE
Charles Bridges-Webb,
Emeritus Professor of General Practice,
University of Sydney and
Chairman, WONCA Classification Committee
The International Classication of Disease (ICD) is the most widely recognisedclassification of diseases. The advent of yet another revision, the tenth (ICD-10)1 in1992 can cause confusion for those familiar with the currently used ICD-92. This maybe even worse for primary health care physicians who may also need to use otherclassifications such as the International Classification of Health Problems in PrimaryCare (ICHPPC)3 or the International Classification of Primary Care (ICPC)4/(ICPC-2)5. The difficulties might be lessened by an account of their historical development,role, relationships, and relative merits.
Classification, nomenclature and thesaurus
Firstly however, it is important to appreciate what a classification is, and how it differs
from a nomenclature or coding system. A medical nomenclature is a list or
catalogue of approved terms for describing and recording clinical and pathological
observations. It should be extensive so that any morbid condition that can be
separately described has a specific designation. Classification is a method of
generalisation to obtain data about groups of cases rather than individual
occurrences. The categories should be chosen so that they will facilitate the
statistical study of disease phenomena, grouping like with like. A specific disease
entity should have a separate title in the classification only when its separation is
warranted because of the frequency of its occurrence, or its importance as a morbid
condition. Many titles in the classification will refer to groups of separate but related
morbid conditions.

ICPC – 2 PLUS Users' Guide It is the grouping of like with like which is the essential feature of a classification. If
this is done in a hierarchical way, with varying levels of specificity (eg. diseases of
the respiratory tract, diseases of the lung, pneumonia, lobar pneumonia) continuing
into great detail, then a comprehensive classification such as ICD can be used in its
most detailed form as a nomenclature. However an alphabetical list of conditions in a
nomenclature cannot act as a classification.
Labelling aspects of general/family practice, such as reasons for encounter and
health problems, requires that the available labels reflect the characteristics of the
domain: general practice/family medicine. Labels should be derived from a
nomenclature or thesaurus. A nomenclature contains all the terms and professional
jargon of medicine, and a thesaurus is a storehouse of terms like an encyclopedia
or computer tape with a large index and synonyms6.
Classification systems provide a structure to order named objects in classesaccording to established criteria. They do not necessarily contain all terms, anddifficulties arise when they are used as a nomenclature and terms are not foundwithin them. Often many terms are included within one rubric, so that the use ofcoding based on a classification does not provide adequate specificity6.
ICPC is a classification which reflects the characteristic distribution and content ofaspects of primary care. It is not a nomenclature. The richness of medicine at thelevel of the individual patient needs a nomenclature and thesaurus much moreextensive than ICPC, particularly for recording the specific detail required in anindividual patient record. The use of ICPC together with ICD-10 and otherclassification systems, such as the Anatomical-Therapeutic-Chemical classificationof medications (ATC), can provide the basis of an adequate nomenclature andthesaurus, but if full coding is required these must be supplemented by even morespecific coding systems. However unless such coding systems are based upon asuitable classification, such as ICPC is for general/family practice, it is not possible toextract coherent data about populations rather than just individuals6.
For clinical and medical record purposes a comprehensive nomenclature is needed,since the greatest possible level of specificity is required (eg. lobar pneumonia notjust pneumonia or lung disease) without grouping even rare conditions together.
However for statistical purposes this leads to so many categories that the data areunmanageable. Grouping of similar rare or less important conditions is essential.
ICPC – 2 PLUS Users' Guide Historical background of Classification
The statistical study of disease began with work on the London Bills of Mortality inthe seventeenth century2. In the eighteenth and the early nineteenth century therewas much interest in the relative incidence of diseases and especially in the changein incidence caused by the disappearance of plague and the control of smallpox.
Information, however, was based for the most part on mortality. Scanty references tomorbidity were derived from general impressions only.
In the early part of the twentieth century James McKenzie appreciated theimportance of morbidity data in his work at St Andrews and in the 1930's WilliamPickles reported important epidemiological investigations in country practice withinformation based on recording of morbidity. Since then there has been increasingrecognition of the importance of morbidity studies, particularly from general practice,to describe and detect changes in the community's health status, and to predicthealth trends that may affect the need for medical services. Morbidity data arenecessary for the study of causation of disease and factors influencing the incidenceand natural history of disease, and in the evaluation of the effect of preventiveprocedures and medical care on the prevalence and severity of disease and thedisability resulting from disease.
The International Classification of Diseases began its career in 1893 as aninternational list of causes of death. Listing for both mortality and morbidity purposes,and change of name, did not occur until the sixth revision in 1948. The emphasis hastherefore been on diseases in terms of their aetiology, pathology and morphology.
Historical Background of ICPC
In primary care many of the conditions treated are vague and ill-defined and they canbe classed only under broad general headings. In 1963 the Royal College of GeneralPractitioners estimated that only fifty-five per cent of diseases in general practicecould be diagnosed accurately in terms of aetiology, pathology and morphology.
Others can only be diagnosed in terms of symptoms or complaints, and someconsultations such as those for immunisation or medical examination do not relate toan underlying condition. ICPC – 2 PLUS Users' Guide Until the mid 1970's most morbidity data collected in primary care research wasclassified using the International Classification of Diseases (ICD)7,8.This had theimportant advantage of international recognition, aiding comparability of data fromdifferent countries. However there was the disadvantage that the many symptomsand non-disease conditions that present in primary care were difficult to code withthis classification, originally designed for application to mortality statistics and with adisease-based structure.
The Classification Committee of the World Organisation of National Colleges,Academies and Academic Associations of General Practitioners/Family Doctors(WONCA) first met in 1972 in Melbourne at the time of the inauguration of WONCA.
Many of its members had already been corresponding for some years aboutmorbidity classifications for general practice. The Committee agreed that it wastime to design a classification specifically for primary care. Recognising the problems of the ICD, and the need for an internationally recognisedclassification for general practice, the WONCA Classification Committee designedthe International Classification of Health Problems in Primary Care (ICHPPC), firstpublished in 19759 , with a second edition in 197910 related to the 9th revision of ICD.
Although this provided a section for the classification of some undiagnosedsymptoms, it was still based on the ICD structure and was still inadequate. A thirdedition (ICHPPC-2-Defined) in 1983 had added to it criteria for the use of most of therubrics11 greatly adding to the reliability with which it could be used, but notovercoming its deficiencies for primary care. A new classification was needed forboth the patient's reason for encounter and the provider's record of the patientsproblems.
At the 1978 World Health Organisation (WHO) Conference on Primary Health Carein Alma Ata12, adequate primary health care was recognised as the key to the goal of"health for all by the year 2000". Subsequently both WHO and WONCA recognisedthat the building of appropriate primary care systems to allow the assessment andimplementation of health care priorities was only possible if the right information wasavailable to health care planners. This led to the development of new classificationsystems.
ICPC – 2 PLUS Users' Guide Later in 1978 WHO appointed what became the WHO Working Party for
Development of an International Classification of Reasons for Encounter in Primary
Care13. This group, a majority of whose members were also members of the
WONCA Classification Committee, developed a Reason for Encounter Classification
(RFEC) 13,14,15 which later became ICPC.
Reasons for Encounter
Reasons for encounter (RFEs) are the agreed statement of the reason(s) why apatient enters the health care system, representing the demand for care by thatperson. They may be symptoms or complaints (headache or fear of cancer), knowndiseases (flu or diabetes), requests for preventive or diagnostic services (a bloodpressure check or an ECG), a request for treatment (repeat prescription), to get testresults, or administrative (a medical certificate). These reasons are usually related toone or more underlying problems which the doctor formulates at the end of theencounter as the conditions that have been treated, which may or may not be thesame as the reason for the encounter.
Disease classifications are designed to allow the health care providers' interpretationof a patient's health care problem to be coded in the form of an illness, disease, orinjury. In contrast, a Reason for Encounter classification focuses on data elementsfrom the patient's perspective13,16,17. In this respect, it is patient-oriented rather thandisease- or provider -oriented. The reason for encounter, or demand for care, givenby the patient has to be clarified by the physician or other health worker before thereis an attempt to interpret and assess the patient's health problem in terms of adiagnosis, or to make any decision about the process of management and care.
The working group developing the RFE classification tested its several versions infield trials. The first field trial to test the completeness and reliability of the RFEC wasa pilot study carried out in the Netherlands in 198014. The results obtained from thispilot study prompted further feasibility testing in 1983. This was carried out in ninecountries, namely, Australia, Brazil, Barbados, Hungary, Malaysia, The Netherlands,Norway, the Phillipines and the United States15,18,19. The entire classification wastranslated from English into several languages, including French, Hungarian,Norwegian, Portuguese and Russian. The analysis of more than 90,000 reasons forencounter recorded during over 75,000 individual encounters and the collectiveexperience of the participants resulted in the development of a more comprehensiveclassification 15,18,19. ICPC – 2 PLUS Users' Guide In the course of this feasibility testing it was noted that the RFEC could easily beused to classify simultaneously the reasons for encounter and two other elements ofproblem-oriented care, namely the process of care and the health problemsdiagnosed. Thus this conceptual framework allowed the evolution of the Reason forEncounter Classification into the International Classification of Primary Care (ICPC). Problems in relation to the concurrent development of ICD-10 prevented WHO frompublishing the RFEC. However WONCA was able to develop ICPC from it andpublish the first edition in 1987. While ICPC-1 was much more appropriate forprimary care than previous classifications based on the ICD framework, it did notinclude inclusion criteria for the rubrics, or any cross referencing. It was thus in thisrespect less useful than the previous publication, ICHPPC-2-defined, though itreferred to it as a source of inclusion criteria which could e used.
In 1985 a project began in a number of European countries to use the newclassification system to produce morbidity data from general practice for nationalhealth information systems. This involved translations of the classification andcomparative studies across countries. The results were published in 1993 in a bookincluding an update of ICPC20.
In 1980 WONCA became a Non-Government Organisation (NGO) in official relationswith WHO, and joint work together since has led to a better understanding of therequirements of primary care for its own information systems and classificationswithin an overall framework encompassing all health services.
The International Classification of Primary Care (ICPC)
The International Classification of Primary Care (ICPC*)21 broke new ground in theworld of classification when it was published in 1987 by WONCA, the WorldOrganisation of National Colleges, Academies, and Academic Associations ofGeneral Practitioners/Family Physicians, now known more briefly as the WorldOrganisation of Family Doctors. For the first time health care providers couldclassify, using a single classification, three important elements of the health careencounter; reasons for encounter (RFE), diagnoses or problems, and process ofcare. Linkage of elements permits categorisation from the beginning of the encounterReturn to Instructions ICPC – 2 PLUS Users' Guide The new classification departed from the traditional International Classification of
Disease (ICD) chapter format where the axes of its several chapters vary, from body
systems *(Chapters III, IV, V, VI, VII, VIII, IX, X, XI, XIII and XIV) to *aetiology
(Chapters I,II,XVII,XIX,XX) and to others (Chapters *XV,XVI,XVIII,XXI). This mixture
of axes creates confusion, since diagnostic entities can with equal logic be classified
in more than one chapter, for example influenza in either the infections chapter or
the respiratory chapter, or both. Instead of conforming to this format, the ICPC
chapters are all based on body systems, following the principle that localization
has precedence over aetiology . The components that are part of each chapter
permit considerable specificity for all three elements of the encounter, yet their
symmetrical structure and frequently uniform numbering across all chapters facilitate
usage even in manual recording systems. The rational and comprehensive structure
of ICPC is a compelling reason to consider the classification a model for future
international classifications.
Since publication ICPC has gradually received increasing world recognition as anappropriate classification for general/family practice and primary care, and has beenused extensively in some parts of the world, notably in Europe 20 and Australia 22. * ICPC was first published in 198720. This is now referred to as ICPC-1. In 1993 itwas included in a publication about its use in Europe19. This is referred to as ICPC-E. This 1998 publication is referred to as ICPC-2. ICPC is used when referring tothe generic classification.
Classifications for primary care
Classifications for primary care have a number of requirements which differ fromthose of other branches of medicine because of the different spectrum of conditionsseen and the different diagnostic and management processes involved.
The classification must cover the full spectrum of conditions treated, includingundifferentiated complaints and symptoms, health promotion and prevention, as wellas a full range of specified diseases. Abdominal pain should be reportable as"abdominal pain" and not, for example, as "?appendicitis" simply because there isnowhere else to include it.
ICPC – 2 PLUS Users' Guide Conditions must be able to be determined on clinical grounds, without requiringdistinctions to be made by inappropriate sophisticated investigations, or, even worse,by requiring knowledge of the underlying pathogy.
The classification should be logical and based on recognised criteria such as bodysystems, so that every condition has only one logical place. This is not the case withICD, where the main groupings (chapters) include body systems, aetiology andpatient age, with the result that influenza could as logically be in the infection chapteras in the respiratory chapter, and all perinatal conditions in any body system aregrouped in one chapter.
The classification should be hierarchical, allowing entry of data at a specific levelwhen indicated (eg. adeno-carcinoma of the colon) or at a much less specific level(bowel cancer) when either the clinical condition is not yet clear or the purpose of thedata recording does not require detail. The more specificity required, the less reliableis the data. Finally the classification should have clear outlines and rules, so that usersappreciate how conditions are related within it. This means that the basic structureshould not be too extensive, and the classification not too large.
No one classification meets all these requirements. Even if one did, it would notnecessarily be ideal for all purposes. It is however important in the interests ofcomparability of data, particularly but not only on an international basis, to useclassifications which have a defined relationship to others, especially to the mostused, ICD. The narrow pathological basis of ICD has been considerably widened inthe 9th and 10th revisions, the latter now including in its title "diseases and relatedhealth problems". However it is far from easy to use for primary care purposes, andthe classifications specially developed for that purpose are to be preferred.
ICPC – 2 PLUS Users' Guide World Health Organisation, International Statistical Classification of Diseases and Related Health Problems, 10th Revision, ICD-10, Geneva, WHO, 1992 World Health Organisation, International Classification of Diseases, 9th Revision. Geneva, WHO 1977 Classification Committee of WONCA, ICHPPC-2-Defined; International Classification of Health Problems in Primary Care, Oxford, Oxford University Press, 1983 Lamberts H and Wood M (eds) ICPC: International Classification of Primary Care, Oxford, Oxford University Press, 1987 5 World Organisation of National Colleges, Academic & Academic Associations of General Practitioners/Family Physicians. The International Classification of Primary Care, Oxford University Press. 1998 (ISBN 0-10-262802-X) Hofmans-Okkes IM, Lamberts H. The International Classification of Primary Care (ICPC): new applications in research and computer based patient records in family practice.Family Practice 1996; 13: 294-302 *International classification of diseases (9th revision). *Geneva, World Health Organization, 1977. *International Statistical Classification of Diseases and Related Health Problems (10th revision).Geneva,World Health International Classification of Health Problems in Primary Care (ICHPPC). Chicago, World Organization National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians (WONCA)/American Hospital Association (AHA), 1975. 10 ICHPPC-2 (International Classification of Health Problems in Primary Care). Oxford, Oxford University Press, 1979. 11 ICHPPC-2-Defined: International Classification of Health Problems in Primary Care, 3rd edition. Oxford, Oxford University Press, 1983. 12 Report of the International Conference on Primary Care, Alma Ata, USSR, 6-12, September 1978; WHO/Alma Ata/78.10. 13 Meads, S. The WHO Reason for Encounter classification. WHO Chronicle, 1983; 37 (5): 159-162. 14 *.Lamberts H, Meads S, and Wood M. Classification of reasons why persons seek primary care: pilot study of a new system. Public Health Reports, 1984; 99: 597-605. 15 *Lamberts H, Meads S, and Wood M. Results of the international field trial with the Reason for Encounter Classification (RFEC). Med Sociale Preventive,1985; 30: 80-87. 16 Working Party to develop a classification of the 'Reasons for Contact with Primary Health Care Services'. Report to the World Health Organization, Geneva, Switzerland, 1981. 17 *Wood M. Family medicine classification systems in evolution. J Fam Pract, 1981; 12: 199-200. 18 *Lamberts H, Meads S, and Wood M. Results of the field trial with the Reason for Encounter Classification (RFEC). In: Cote RA, Protti AJ, and Scherner JR eds. Role of Informatics in Health Data Coding and Classification Systems. Amsterdam, Elsevier Sci Publ/JFIP-JMIA, 1985. 19 Bentsen BG. International Classification of Primary Care. Scandinavian J Primary Hlth Care 1986;4:43-56 20 *Lamberts H, Wood M, Hofmans-Okkes I, eds. The International Classification of Primary Care in the European Community: with Multi-Language Layer. Oxford, Oxford University Press, 1993. 21 Lamberts H, Wood M eds. ICPC: International Classification of Primary Care. Oxford, Oxford University Press, 1987 22 Bridges-Webb C, Britt H, Miles DA, Neary S, Charles J, Traynor V. Morbidity and treatment in general practice in Australia 1990-1991. Med J Aust 1992; 157, Supp.19 Oct :S1-S56. ICPC – 2 PLUS Users' Guide SECTION 3: CLASSIFICATION OF PRIMARY CARE
–STRUCTURE AND DERIVATIVES
This section has largely been drawn from: Lamberts H, Wood M. (eds). ICPC International Classification of Primary Care. Oxford. Oxford University Press 1987.1 Though ICPC was primarily designed for the classification of patient reasons for encounter (RFEs)* by the primary care provider at the time of the consultation, it can also be applied to the provider's assessment of the problem (diagnoses) and to the diagnostic and therapeutic interventions utilised at the encounter.
Patient reasons for encounter (RFEs) are very different from the diagnoses orproblems managed. For an adequate description of the content of generalpractice, the whole process of care needs to be considered, including thereasons patients seek care. While the concept of recording patients' reasonsfor attendance is relatively new in Australia, in many countries increasedinterest in the patient-centred approach to medical care has led GPs to recordthe patients RFE in their medical records as a matter of course. In Australiangeneral practice it is more likely that the GP will record presenting symptoms,which may represent only part of the patient's RFE. ICPC is ideal for recordingboth RFEs and presenting symptoms, and it is up to the individual generalpractitioner to decide whether either or both of these elements are to be coded. SOAP
Therefore three of the four parts of a patient's problem-oriented clinical record, which
reflect the essential elements of each patient/provider encounter, can be coded using ICPC – 2 PLUS Users' Guide Subjective:the patient's reason for encounter (or presenting symptoms) O = Objective:this element cannot be classified using ICPC
A = Assessment:the provider's interpretation of the problem in the form of a diagnosis
P = Plan:the process of care intervention undertaken by the provider
ICPC STRUCTURE
It has a biaxial structure with 17 chapters on one axis and seven components on the
other.
Chapters are based on body systems with an additional chapter for psychological
problems and one for social problems. Each chapter is identified by a single alpha
code which is the first character of all rubrics belonging in the chapter (Figure 3.1)
Each chapter is divided into seven components, identified by a range of two digit
numeric codes which are not always uniform across chapters.
Figure 3.1: Structure of ICPC
1. Symptoms, complaints 2. Diagnostic,screening prevention 3. Treatment, procedures medication 5. Administrative 7. Diagnoses,disease Blood, blood forming N W Pregnancy, family planning ICPC – 2 PLUS Users' Guide Component 1 provides rubrics for symptoms and complaints. It drew on the National
Ambulatory Medical Care Survey/Reason for Visit Classification (NAMCS/RFV) 2,3 and
on the RFE-C developed by the WHO working party.2,4,5. Rubrics in this component
can be used to describe presenting symptoms, and are valuable for describing the
problem under management (in a problem list in the medical record) when the condition
is as yet ill-defined (eg: general ill-feeling; feeling tired).
Component 7 is the diagnoses/disease component in each chapter. This component
will be the one most often used when you have sufficient information to arrive at a
diagnosis in the medical record or problem list. It is based on the ICHPPC-2 and most
rubrics are directly comparable. However the psychological and social chapters of ICPC
are drawn from problem lists developed by the WHO sponsored Triaxial Classification
Group 6,7.
Within this diagnostic component are five sub groups which are not numerically uniform
across chapters:
. infectious diseases; . neoplasms; . injuries; . congenital anomalies; . other diseases. Components 1 and 7 in ICPC function independently in each chapter and either can
be used to code patient RFEs, presenting symptoms, or problems managed.
Components 2-6 are common throughout all chapters, each rubric being equally
applied to any body system.
Component 2 covers diagnostic screening, prevention. It is useful when there is no
underlying pathology for the problem under management eg: immunisation, check up
(partial or full); advice and health instruction.
Component 3, treatment, procedures and medication. This component should rarely if
ever be used to describe a problem under management as it covers the processes
involved in patient care. However for those who wish to code procedures as well as
problems, these codes will prove very useful.
Components 2 and 3 are based broadly on the ICD-9 Procedures in Medicine 8 and
are heavily influenced by the International Classification of Process in Primary Care (IC-
Process-PC).9
Component 4, Test results and Component 5, Administrative, provide somewhere to
put those difficult problem labels which frequently have no pathology (eg: completing a
patient's application for a passport would fall into Component 5).

ICPC – 2 PLUS Users' Guide The structure of ICPC represents a move away from the combined anatomical andaetiology based structure of ICD. For example, where ICD includes a separate chapterfor neoplasms, one for infections and infestations, and another for injuries, suchproblems are distributed among chapters in ICPC, depending on the body system towhich they belong. Regrouping of the rubrics (eg for all neoplasms in all body systems)can still be undertaken across chapters if analysis of totals is required. Grouping isfurther discussed in Section 5.
1. Lamberts H, Woods M (eds). ICPC. The international classification of primary care. Oxford: Oxford University Press, 1987.
2. A reason for visit classification for ambulatory care. Hyattsville,MD: U.S. Public Health Service, National Centre for Health Statistics (DHEW Pub. 79-1352), 1979. 3. Patients' reasons for visiting physicians: National Ambulatory Medical Care Survey, United States, 1977-1978. Hyattsville,MD: Series 13,56 (DHS Publications 82-1717), 1981. 4. Lamberts H, Meads S, Wood M. Results of the international field trial with the reason for encounter classification (RFEC). Role of informatics in health data coding and classification systems.
Amsterdam Elsevier Sci. Publications, 1985. 5. Meads S. The WHO reason-for-encounter classification. WHO Chronicle 1983; 37: 159-162. 6. Lipkin M and Kupka K.(eds) Psycho-social factors affecting health. New York: Praeger, 1982. 7. WHO. Psychological factors affecting health assessment, classification and utilisation. Report of the World Health Organisation on the Bellagio Conference. Geneva: WHO, 1980. 8. WHO. International classification of diseases (9th revision). Geneva: World Health Organisation, 1977. 9. WONCA Classification Committee. IC-Process-PC (International classification of processin primary care ) Oxford: Oxford University Press, 1986 ICPC – 2 PLUS Users' Guide SECTION 4: ABOUT ICPC-2 PLUS
This section has been based on the following paper:
Development of a database for the International Classification of Primary Care, for direct
entry. H Britt.
Presented to the 14th WONCA WORLD CONFERENCE, Hong Kong. June 1995.
Background to ICPC PLUS
The International Classification of Primary Care, or ICPC, was designed by the WONCAClassification Committee, primarily for use by the health care provider at the time of theconsultation on paper based records.
The Family Medicine Research Centre (FMRC), University of Sydney has however used it inthe centralised coding of patient reasons for encounter (RFEs) and problems managed inthe Australian Morbidity and Treatment Survey (AMTS)1 and in a popular Quality Assuranceoption, the Morbidity and Therapeutic Index (MTI). The FMRU have secondarily coded over3/4 of a million patient RFEs and equivalent numbers of problems managed from paperbased encounter records completed by general practitioners throughout Australia.
In the early stages of this paper based coding, it was evident that high inter and intra coder reliability was difficult to attain with ICPC. Errors in the index were noted and the classification's layout was sometimes confusing. But more importantly the ICPC's alphabetical index was found to be inadequate for reliable coding of many terms commonly
recorded by GPs. From 1991-97 thousands of terms were added by the FMRU to the ICPC index to facilitateaccess to the correct code. When unsure of where to place a term in ICPC, the Chairman ofthe WONCA Classification Committee was referred to, who by reference to ICD-9/10,selected the correct code. Utilising this mechanism the index continued its expansion on thebasis of terminology used by Australian GPs, to describe patient RFEs and diagnoses. ICPC – 2 PLUS Users' Guide Development of the computerised database
The FMRU found that using ICPC with a widely expanded index worked well in a centralisedpaper based coding system and provided sufficient specificity for epidemiological dataanalysis. However, in the absence of any other acceptable classification system, the Centrecame under increasing pressure from GPs to adapt ICPC for computer usage in the clinicalprimary care setting.
One of the major problems with using ICPC for computerised medical systems, was its lackof specificity in some areas. For epidemiologists it may be adequate to count the number of patients or encounters
with an "other viral illness" (i.e. the number of A77's that arise). However, in a clinical
setting the practitioner must be able to differentiate between the 33 viral illnesses
which fall into this rubric:
A77 - Other viral illness
For Example, if you practice in one of our sub-tropical areas, it may be very important to be
able to identify the patients who are suffering specifically from Ross River Fever, not just
"other viral illness".
There are two ways around the problem, neither ideal: Select all records including A77 then use a word search engine to find all recordsinvolving Ross River Fever. and hope you have never mis-spelt it. Create another level in the hierarchy. When developing ICPC PLUS (a computerised classification system derived from ICPC)some users of ICPC suggested that further hierarchical layers should be added forincreased specificity. However ICPC was not designed for such extension. Using hypertension as an example, a possible hierarchy may have one code at the upperlevel for hypertension (not otherwise stated). The next level may be defined as "primary" or"secondary". At the third level it may be "benign" or "malignant" and at the fourth level"with/without target organ involvement" (see Figure 4.1). Specific target organ involvementcould be differentiated at the next level. ICPC – 2 PLUS Users' Guide In ICPC there are already two codes for hypertension, one without complications and
one with involvement of target organs. Both of these rubrics cross multiple levels of
the hierarchy.
FIGURE 4.1: A possible hierarchy for the diagnosis of hypertension.
(not otherwise stated) PRIMARY
Benign Malignant Malignant Non-malignant
No + + No
involved target involved target involved target involved target
organs organs organs organs
Extension of the hierarchy is further complicated by lack of agreement about definitions and
synonymous terms, even within one country. Using Diabetes as an example (see Figure
4.2) we could probably gain clinician consensus that: Type I diabetes is synonymous with
insulin dependent diabetes, and that IDDM is an acceptable acronym; that Type II diabetes
is synonymous with non-insulin dependent diabetes and NIDDM is an acceptable acronym.
However, the extent to which the remaining terms listed in Figure 4.2 are synonymous with either of these labels is questionable. Some people would state that "juvenile onset diabetes" equates with Type I. Others would disagree, stating that Type I diabetes can also be adult onset. In the long term, agreement between practitioners, (both nationally and internationally), regarding definitions and synonyms may be reached. Until such time, creating a hierarchical structure under the upper level of (e.g.) "diabetes" is not possible. ICPC – 2 PLUS Users' Guide Figure 4.2 Possible terms in the Diabetes Mellitus group
Diabetic coma
Diabetes Mellitus (NOS)
Diabetes; complicated
Non Insulin dependent diabetes
Insulin dependent diabetes
Juvenile onset diabetes
Adult onset diabetes
Diabetes:Type I
Diabetes:Type II
Figure 4.3 - ICPC CODE A77 - Other viral illness
ICPC Code
Term Code
Disease;hand foot & mouth Disease;hand foot & mouth ICPC – 2 PLUS Users' Guide ICPC PLUS
The extended medical term index developed from the GP encounter forms was entered in a
SQL relational database.
ICPC PLUS term and code creation
Terms codes: As each term was entered into the database, the computer automatically
allocated a term code. (eg: in Fig. 4.3: Viral illness = term code 005 within ICPC code A77).
Thus each term has a unique identifier (A77023) which allows storage of the more specific
term in the medical record.
Keywords: For each term, one or more keywords were allocated.
eg: in Fig 4.3: Term code 005 has three keywords by which access can be gained to that
term: vira, illness and disease. Therefore, while each term within any one ICPC rubric has a
unique code, a GP may, on different occasions, use different keys to access that term.
The GP may choose to describe a single medical concept in different terms on specificoccasions. (e.g. "Diabetes Type I" and "Insulin dependent diabetes"), and his/her termselection should be saved in the medical record - not changed to the description of the ICPCrubric. Note that the terms are listed in term code order which is derived purely from the order ofentry. There is only one term code per term, irrespective of the number of keywords. Newkeywords and additional codes are being added and will continue to be added on request. In ICPC PLUS, as in any other indexed system, a single keyword can lead to multipleconcepts and at times, to multiple ICPC rubrics, and these should be offered as a pick list.
For example, if you enter the term diab or diabetes, the picklist (all the terms attached todiabetes - in the second column in Figure 4.2) should appear together with others (such asdiabetes in pregnancy) which do not belong with code T90.
The concept follows that used in the Read Clinical Codes. For the clinician, the exercise is
one of terming, rather than coding. What secondary coders do (i.e. medical records coding
clerks in hospitals) is find the recorded medical term and allocate the correct code to the
term. In contrast, this system allows the practitioner to select the TERM most suitable to
his/her needs, from the pick list. The computer transparently attaches the correct ICPC
code for the concept and the code number of the selected term, i.e. "terming" rather than
"coding". To all intents and purposes, you should be relatively unaware of the coding
process.
Distribution of ICPC/ICPC-2
Permission was sought from WONCA to offer the database to software houses for wider
application in general practice. While some GPs were already using ICPC in non-
commercial GP data systems, in the main they were using unlicensed copies of ICPC.
Further, in many cases, they used idiosyncratic adaptations which resulted in non-
comparable data.
In an effort to improve the standardised use of the classification, and to ensure copyright ofthe ICPC/ICPC-2 was upheld, WONCA provided the FMRU with the exclusive license for thedistribution of ICPC/ICPC-2 in electronic form in Australia and the Pacific Basin.
ICPC – 2 PLUS Users' Guide ICPC-2 PLUS
ICPC PLUS broke new ground when it was first released in 1991. It was the first
computerised classification system that had been specifically designed for Australian
general practice and primary care. It has been implemented by a number of software
suppliers in clinical systems across the nation and has an End User base that is rapidly
growing and diversifying.
With the advent of the World Organisation of Family Doctors release of a revised edition ofICPC the Family Medicine Research Centre has created a second version of ICPC PLUS,namely ICPC-2 PLUS. This new revised version was developed with continued consultation with the Chairman ofthe WONCA classification committee. It maintains its predecessors structure of terms andprocess of keyword allocation approach to classification as a terming rather than codingprocess Development of ICPC-2 PLUS
The conversion process from ICPC PLUS to ICPC-2 PLUS was a 3 stage process due to the
significant changes in the structure of ICPC-2. A number of ICPC rubrics were deleted in
ICPC-2 and a number of new rubrics were added to the classification.
Firstly the final version of ICPC PLUS was copied into a new database that contained ICPC-2's structure and revised rubrics. ICPC PLUS terms located in rubrics that had been deletedin ICPC-2 were then moved to applicable ICPC-2 rubrics and their path mapped. Finally allnew ICPC-2 rubrics were identified and terms added to them.
In particular, there were significant additions made in the areas of therapeutic and diagnosticprocedures, psychological counseling and referrals to specialists and allied healthprofessionals.
Map creation
Through this entire process of term movement a map was developed to convert
retrospective data collected by End Users of ICPC PLUS into ICPC-2 PLUS. Such a map
allows users to access all information previously collected in ICPC PLUS in a valid ICPC-2
PLUS term.
Once the conversion from ICPC PLUS to ICPC-2 PLUS was complete the following stepswere undertaken as part of the ongoing development of the extended vocabulary: A number of new terms were added in response to user requests and the Centre's ownreview process:- • Significant revision of keywords and term access pathways was undertaken• Picklist structure and presentation was targeted for refinement• Tests ordered (particularly pathology and imaging) were expanded ICPC – 2 PLUS Users' Guide Analysing stored information with ICPC PLUS
Getting the data in to a medical record system in a classified form is of course only the first
part of the process. There is no point in entering the information if you are unable to draw it
out in a useful manner. It is important that you utilise the conversion map provided to your
software developer to transfer any data collected in ICPC PLUS into ICPC-2 PLUS terms
before you begin any data analysis.
Sorting at ICPC-2 rubric level
This is the easiest method of identifying a specific group of patients in your practice.
You want a list of all your patients who have attended your practice for uncomplicatedhypertension: Enter hypertensi and you will find the normal picklist associated with that keyword; highlight
the one you want ("hypertension uncomplicated") and ask the computer to provide the list
of patients having that entry.
Alternatively you may wish to identify the ICPC-2 code for uncomplicated hypertension (K86)and then ask for the list of all records having that code attached. Sorting at multiple rubric level
While it is useful to be able to sort your stored data at individual rubric level (as above),
sometimes you may want to identify a group of patients who have any one of multiple
diagnoses or symptoms.
Example 1
You want to identify all patients in the practice who have attended for uncomplicated
hypertension (K86); hypertension with complications (K87); and elevated blood pressure
without diagnosis of hypertension (K85). You would search for all records which include any
one of these three codes.
Example 2When viewing the relative frequency of presentations of rash, there are two symptom codesavailable to represent this concept: localised skin rash, coded as S06 and generalised skinrash, coded as S07. If you wanted to identify all the patients who had presented with skinrash (generalised or localised) both codes would need to be searched. For these more general concepts which involve multiple ICPC-2 rubrics, you can also utilise
the list of Code Groupers provided in Section 5 to ensure all cases will be selected. These
code groupers have been provided to software developers so you should be able to view the
grouper list on your computer or select them automatically if your software allows.
ICPC – 2 PLUS Users' Guide Revised components
Analysing your records by CHAPTER-COMPONENT Groupers you can identify the codes
to be included when analysing on the basis of broader medical constructs.
As described in Section 3, the ICPC-2 is divided into seven components which are common
to all chapters, only components 2-6 being made up of rubrics which can be applied equally
in all chapters. Components 1 (symptoms) and 7 (diagnoses) differ in each chapter
according to the most frequent symptoms and diagnoses which occur in each body system
or psychosocial area.
While working with ICPC-2 it became clear to the Family Medicine Research Centre that thedivision between the two components was not always correct. Some diagnostic labels (e.g.
paronychia; warts) had been included in the symptom section. This was particularly so in theskin chapter where there were insufficient codes available in the diagnostic section. In othercases symptoms were included in the diagnostic component (eg: K85; high blood pressurewithout a diagnosis of hypertension). Further, attempts had clearly been made by the Classification Committee to subdivide thediagnostic component in a uniform manner across chapters, into infections, neoplasms,injuries, congenital anomalies, and other diagnoses. However, due to the variable extent towhich each of these types of problems applied in each body system, the range of numericalcodes falling into each sub-group varied between chapters. This meant that analysis of asub-group across chapters (eg: all injuries; all neoplasms) could not be undertaken by asimple selection of common numerals across all chapters.
A detailed review of the component breakdown was undertaken and under the guidance ofthe Chairman of the WONCA Classification Committee rubrics were re-allocated to theircorrect component and a more detailed component breakdown created. The new categorisation includes eleven components, the original diagnostic component
(component 7) having been further broken down into the sub-groups mentioned above. A
chart of the new components by chapter, including lists of all codes which should be
included in each cell is shown in your User Guide. This chart has been supplied to all
software developers utilising ICPC-2 and should therefore be included in your program in
some form.
Some examples of analyses using chapter-component combinations are: • All patients treated for any injury over the past year (All chapters- Component 10); • All patients managed for an infection of the skin (Chapter S - Comp 8) in the last month; • All male patients managed for a male genital disease (Chapter Y - component 7 [made up of components 8-12]).
Analysis using part of an ICPC rubric
When you are only interested in one part of an ICPC rubric: first view the list of terms
available with their varied term keys, select all those which you feel should be included and
select these term codes for inclusion in the analysis.
For example:
You wish to send a recall letter to all patients who require DPT immunisations:- review the list of all
terms for *44 and select those which you feel should be included in this sub-group. If the software
allows, you could select on a variety of other fields as well, e.g. all female patients, aged 0-6 years.
REFERENCES
1.
Bridges-Webb C, Britt H, Miles DA, Neary S, Charles J, Traynor V. Morbidity and treatment in general practice. Med JAust 1992; 157 (19 Oct Spec Supl): S1-S56 ICPC – 2 PLUS Users' Guide SECTION 5: HINTS FOR USING ICPC-2 PLUS
Firstly, to get a "feeling" for the classification you are about to use, read the background(particularly the structure of the coding frame, Sections 3 & 4). "Terming' with ICPC-2 Plus
ICPC-2 PLUS utilises a technique of ‘terming' rather than ‘coding‘ to summarise data.
‘Terming' refers to the entry of a few key letters or a brief keyword to access a picklist of
possible terms which may be used to describe a particular medical concept (see over).
When a term is selected the computer transparently attaches the correct ICPC-2 PLUS code
number. In contrast ‘coding' is a much more laborious task which involves interpretation of
the medical record, looking up, selecting and applying the most appropriate code.
Term ICPC
Unable (to); speak Keywords that can be used to Medical concept you wish to Automatically allocated when access a picklist of related you select a term ICPC – 2 PLUS Users' Guide Term organisation
To facilitate the reading of ‘picklists' the terms in ICPC-2 PLUS are generally organised
using two basic structures:
1. ‘Common usage' expression eg. Irritable bowel syndrome or2. ‘Problem or Procedure; type; site' organisation. eg. Pain;cardiovascular; chest.
The Problem or Procedure is usually the first word to appear in the term which provides a‘generic' description of the issue to be coded eg. pain. These ‘first words' may be used askeywords when a comprehensive list of all the terms under that generic description isrequired. Problems include words such as lesion, inability, infection, anaemia, fracture etc.
Procedures include words such as excision, destruction, test , admin etc. (see Section 6 for
a list of the standard process components of ICPC - components 2 to 6).
Type usually includes words such as acute, chronic, benign, malignant etc. These are often
not designated keywords unless they are particularly relevant to a term. The problem or
procedure ‘type' facilitates terming because it provides a secondary level of organisation of
terms on a picklist.
Site identifies the location of the problem or procedure eg. chest, arm, leg, heart etc.
Selecting a ‘site' as a keyword will often produce a long ‘picklist' of terms. This is useful
when identifying terms related to a specific site.
Acronyms you see in ICPC-2
The following acronyms are generally not used in ICPC-2 terms:
NOS: Not otherwise specified (or in clinical terms - not yet able to be more specific), NEC: Not elsewhere classified (or in clinical terms - not able to be classified more preciselywithin the options Levels of specificity are usually indicated by the type or site part of the term. If this level of
specificity is not included then the term is assumed to include the ‘not otherwise specified'.
eg. the term ‘hypertension' is equivalent to the previously used ‘hypertension;NOS'.
ICPC – 2 PLUS Users' Guide Keywords in ICPC-2 PLUS
In addition to the development of a more comprehensive list of Keywords, a number of
changes have been implemented regarding keywords to provide easier access to terms.
Following are some hints specifically related to keyword usage:
Avoid spaces, dashes or slashes within Keywords: To provide greater uniformity of keywordsyntax all spaces, dashes and slashes have been deleted. S/E replaced with SE POST-OP replaced with POSTOPCHECK-UP replaced with CHECKUPSIDE-EFFECT replaced with SIDEEFFECTX-RAY replaced with XRAY HIGH BLOOD replaced with HIGHBLOODPOST TERM replaced with POSTTERMCIN 1replaced with CIN1PORT WINE replaced with PORTWINE Use singular keywords instead of plural TOE instead of TOESINJURY instead of INJURIESTEST instead of TESTS NB. You will see that the keyword listing provided in ‘Appendix A' uses pleural rather thansingular syntax. In general, the keywords on the list are the longest version of the mostsuitable keyword. The list has been created this way as a precaution for users who mightforget to use singular keywords. It is advised, however, that users adopt generally, a practiceof entering keywords in their singular form. Keyword selection. If you cannot find the term you require consider:
spelling.
(ICPC-2 PLUS generally uses Australian English rather than American eg. Oesophagus
NOT esophagus, immunisation NOT immunization)
entering a shorter version of the keyword. eg. Arteri instead of arteriosus, thrombo
instead of thrombocytic. The shorter the keyword the wider the picklist.
a different form of the word . e.g. aged or aging, allergic or allergy, absent or absence,
depression or depressive or depressed
the term organisation.
ie. ‘Common usage' or ‘problem;type;site' structure.
ICPC – 2 PLUS Users' Guide Analysing Information Stored With ICPC-2 PLUS
Code groupers
When trying to analyse the data you have recorded, the medical concept of interest may not
always be contained within a single term. The example of hypertension was described
earlier. Another example is the concept of depression: there are two codes available, one for
the diagnosis of depression and one for the symptom of "feeling depressed". If searching for
all patients with any form of depression you may wish to search the database for the
occurrence of BOTH codes (P03 and P76). A list of code groupers is provided below in
Figure 5.1 and should be available in your software.
Major chapter-component groups
These have been discussed earlier. The corrected chapter component groups are presented
below (Figure 5.2) and have been provided to your software supplier for incorporation into
your software.
GPs' Questions Answered
The FMRU has received several inquiries from users of ICPC-2 PLUS, which suggest that
some hints about its application in a clinical setting may be helpful.
Question:
Can appendicectomy, mastectomy and hysterectomy be coded?
Answer: YES
These are all terms which refer to a process rather than to a diagnosis or problem under
management. ICPC-2 PLUS now covers these procedural terms, if, you wish to record
problems in this form in the History section of a record.
This brings up the issue of how the meaning of a medical term may change according towhere it is in the medical record. If you record appendicectomy in the History section itclearly indicates it has been done in the past. If you record it in the Problem list it suggestsyou personally undertook an appendicectomy as part of the management of a problem (i.e.
appendicitis).
ICPC – 2 PLUS Users' Guide Question:
"How do I find a code for "pregnancy results" in my problem list?"
Answer:
Enter "pregnancy", "results", or "test" . Your picklist will offer a choice one of which is "W60" -
Test;results;pregnancy, together with others specifically attached to a body system
However, in a problem based medical record, there should never be a diagnosis/ problemlabelled in this manner. If a patient rings and asks for repeat scripts for multiple problems,each repeat should be recorded within the problem to which it is linked. e.g. Patient requestsrepeat scripts for hypertension and "the pill". You saw the patient recently and you do notneed to see her again at this time. Within the patient's record, the problem list should alreadyinclude problems called oral contraception and hypertension. You need to enter the "Pill"linked to the oral contraception and enter the anti-hypertensive against the hypertension.
The "pregnancy test" should be recorded as part of the process of care.
Using the same argument, if a patient returns for test results, the problem you are managingis still the one for which you requested the tests (eg. diabetes; hypertension; amenorrhoea;general check up). The problem title should not be recorded as "test results". In the aboveexample the problem label may be "question of pregnancy" or "late menses" oramenorrhoea".
Note: ICPC -2 is not designed to code the drug prescribed. However, we are developing adrug classification which will be offered to GPs when available.
Question:
Could you please include severity levels for some diseases (e.g. COAD)
and more detail about site (location) for others?
Answer;
While it is possible to add this more specific information and provide a specific code for
different severity levels or for site, this would enlarge ICPC-2 PLUS by about three or four
fold. One of the advantages of ICPC-2 PLUS over systems such as Read Clinical Codes or
ICD9-CM, is its size. The more codes we introduce the more difficult it becomes to find the
term required as the pick list become longer.
In deciding on the level of detail to be included in ICPC-2 PLUS we have constantly kept inmind that the main use of a hierarchical coding system is for getting the data out in ameaningful manner. E.g.: you may wish to identify all your patients who have presented with an injury. You maywish to further group these patients into sub-groups such as fractures, and still further intofractures of the ulna. Ideally, you may like to differentiate between fractured mid-shaft ulnaversus those with a fractured proximal ulna. However, incorporating this level of detail about site or severity throughout ICPC-2 PLUSwould result in a large database that no longer has its ‘economies of scale'. If you want moredetail about your diagnosis than is provided by ICPC-2 PLUS your software should allow youto add free text to further describe problems. For severity levels we suggest that you speakto your software developer and request the introduction of an additional field for severitywhich you can use when you wish.
ICPC – 2 PLUS Users' Guide In summary: ICPC-2 PLUS will not be expanded to include severity and codes which specifysite will remain at the upper level (i.e. ulna) rather than be expanded (i.e. to mid-shaft versusproximal). Question:
How do I know which keyword is best?
Answer;
Some keywords such as check, disease, pain, cyst, injury, are associated with many terms.
As you get used to the lists provided with these terms, you can experiment with other key
words which may provide a shorter pick list and so lead you to the desired term with less
effort.
For example, for a diagnosis of "whiplash":- if you enter "injury" as your keyword you will bepresented with a long picklist (about 81 options). A shorter list will present if you enter "neck"(13 options) or "whip" (2 options).
If you think that the key word you have entered is appropriate but the "pick list" is too narrow,then consider whether your key word could be made more "generic". For example:"Allerg" will pick up lists for both "allergic" and "allergy""Hypertensi" will pick up a list which includes "hypertension" and "hypertensive""Infecti" will pick up a list which includes "infections" infective" "infectious" etc.
"Degenerat" will pick up "degeneration and "degenerative""Diabet" will pick up "diabetes" and "diabetic" Of course, as you get to know the range of terms in the database, you will learn to enter"infection" in full when you specifically do not want the choice of "infective" options on yourpick list.
Allergies and sensitivities:
If you are using ICPC-2 PLUS to record allergies and sensitivities in a patient medical record
or summary sheet, a variety of key words may be used such as: allerg, allergic; reaction;
adverse; effect; sideeffect; SE.
Furthermore, if the allergy/sensitivity is related to antibiotics the generic drug name may beused as a key word: "Penicillin" for Amoxycillin, Abbocillin etc; "Cephalosporin" for Keflex, Ceclor etc"Macrolide" for Erythromycin, Rulide etc. ICPC – 2 PLUS Users' Guide Question:
"How do I find an ICPC-2 PLUS code for repeat scripts as a problem? "
Answer:
Code -50 (replace the dash with the associated ICPC-2 chapter) does allow the coding of
"repeat script(s)" as a diagnosis. So if you enter "script" as the keyword, you will be offered
a picklist of body systems to which the drug prescribed relates. (e.g. prescription:
respiratory).
Question:
"What do I do when I can't find the term I want?"
Answer:
After considering spelling, keyword length and term organisation if you
still can't find the term you want:-
In each software product, there should be a facility to enter the term in free text and apply a
temporary code - J99. The term you record should be retained in the database and
allocated a term code (so that the first time you use J99, the term you record will be saved
as term number J99-001 and so on.) Your software should allow you to view your list of
temporary terms. You could print out the list and post or fax it to the Family Medicine
Research Centre so that the terms can be allocated to the correct ICPC-2 code and added
to ICPC-2 PLUS with their term keys. The next upgrade you receive will therefore include the
term, the correct code and term code. Your software should allow you to alter your original
record to the correct code (of course always retaining the hidden audit trail of the change for
medico-legal purposes).
Question:
"What do I do when I change a diagnosis?"
Answer:
This question brings up the concept of problem linkage over time
through changes in diagnosis.
For example: A patient presents with a headache, the problem is entered as "headache". At a later date,you decide that these headaches could more specifically be described as Migraineheadaches (which has a different code in ICPC-2). This should not be entered as a newproblem in your problem list. Rather your software should allow a change of label for the oldproblem of headache. Your software will include an audit trail (for legal purposes) whichshould include the old diagnosis and the date of change to the new.
ICPC – 2 PLUS Users' Guide The aim of a problem list in a medical record or summary is to ensure that each problem isonly recorded once. This means that each time you go to enter any information, you shouldview your problem list to check whether the problem is already on the list. Otherwise, doublerecording of the same problem with different labels can occur. For example: A patient arrives for Pap smear. You do a breast check as well and select "checkup;partial;genital;F" (X31 003) from your pick-list. The patient's Pap smear was insufficientand re-test is required. You request the patient's re-attendance, and on this occasion only dothe Pap smear. You may be tempted to enter this problem as "Pap smear" which will beallocated ICPC-2 PLUS code X37 001. If you look at your problem list, you now have twoproblems, both of which cover the same medical concept, one just being slightly morespecific than the other. In fact, the follow-up visit for the Pap smear should have beenentered under the old problem of "female genital check". When the results return they showan abnormality. You could then feel free to record an additional problem called "abnormalPap smear" (X86 001), or if the Pap smear results were more specific:- CIN I (X86 005); CINII (X86 006); CIN III (X86 032) There are a few good examples in ICPC-2 which lend themselves to multiple recording ofthe same problem in different terms. These are usually cases where you have the choice ofbeing more general or more specific. An example is where the patient presents after anaccident, questioning whether his leg is fractured. You suspect it is and you send the patientfor x-ray. The problem at this point should be recorded as injury - leg (L81). As yet, you haven'tsufficient information to label it as a fracture. When the patient returns, x-rays show afracture. It is the same problem to the patient and therefore should not now be recorded as anew problem of "fractured femur" (L75). It should result in a change of diagnosis from injury(L81), to fracture (L75).
The Code Groupers listed later in this section will give you some idea of the more subtlechoices which may present to you in selecting a medical term. E.g. osteoarthritis: there isone general OA code but others are more specifically identified by body part. ICPC – 2 PLUS Users' Guide Question:
What do I do when I want to write a diagnosis with a question mark (e.g.
? anaemia)
Answer;
This form of provisional diagnosis is often useful in your medical notes, as it reminds you of
your thoughts about the symptoms at the previous consultation. However, a problem list
should only contain the diagnosis or problem managed, described at the highest level of
specificity allowed by the information available.
For Example: The patient presents with symptoms of pallor and shortness of breath. After history andexamination , you feel it is likely to be anaemia and send the patient off for blood tests. Therecording of the problem should, at this stage be both pallor and shortness of breath - theseare the problems you are investigating. Only when the results are returned and yourdiagnosis of anaemia is confirmed should the problem be recorded in such specificdiagnostic terminology (linked as earlier suggested). If your software allows, this does notpreclude the additional entry of ? Anaemia in free text in your notes, after that firstconsultation (i.e. not coded). FUTURE PLANS FOR ICPC -2PLUS
Continued improvement of keywords and terms to facilitate easy access.
Extension of the agreement with WONCA, to allow the provision of other classifications as they become available. Possible additions include:- Functional Status Charts and the Duke University Severity Index. The aim will be to include new tools at no additional costs or at an "upgrade cost".
ICPC-2 has inclusion and exclusion criteria for the majority of problems. The possibility ofapplying these criteria in computer systems in the future is being considered by the FMRU.
Your input is extremely useful in the maintenance and ongoing development of ICPC -2 PLUS.
ICPC – 2 PLUS Users' Guide

Source: http://www.fmrc.org.au/ICPC%20User%20Guide.pdf